2020-06-01 00:03:03 <doggone> did you find a fix?
2020-06-01 00:03:18 <doggone> or a cause?
2020-06-01 00:22:07 <velix> Does anyone have an example APKBUILD, how to pull from a github repository? or does it work with releases only?
2020-06-01 00:24:55 <velix> ah! https://wiki.alpinelinux.org/wiki/APKBUILD_examples:Git_checkout
2020-06-01 06:20:05 <ikke> velix: note that that only a few people have access to create snapshots
2020-06-01 10:39:48 <indy> hi all, i'm trying to adopt https://wiki.alpinelinux.org/wiki/Creating_an_Alpine_package for customer's need, but struggling with one package which i depend on but customer's partner do not provide release tarball they just send zip file
2020-06-01 10:43:50 <afontain_> it still works
2020-06-01 10:44:13 <afontain_> just add source.zip in sources instead of (e.g.) source.tar.gz
2020-06-01 10:46:55 <indy> afontain_, "source=something.zip" without any http(s)/ftp/whatever?
2020-06-01 10:47:10 <afontain_> it works on http, at least
2020-06-01 10:47:30 <afontain_> I'd say you should just try it
2020-06-01 10:47:41 <afontain_> it's the best way to know if it'll work
2020-06-01 10:48:07 <indy> afontain_, going to try
2020-06-01 10:54:15 <Cogitri> It works with local files as well, yes
2020-06-01 12:10:33 <mat_cz> i needed to install binutils-dev as well @ikke
2020-06-01 12:14:08 <ikke> mat_cz: typically you just need to install binutils-dev
2020-06-01 12:14:13 <ikke> as makedepends
2020-06-01 12:15:43 <mat_cz> i installed binutils only and forogt about -dev :(
2020-06-01 13:53:25 <indy> afontain_, it works :) thanks, is there any example of git-only repository? like upstream projects having debian folder for building debian packages? or spec inside
2020-06-01 13:56:04 <afontain_> if a project is hosted on gitlab or github, you can ask gitlab/hub for a tarball
2020-06-01 13:56:18 <afontain_> it will be generated automatically from a commit hash
2020-06-01 13:56:47 <afontain_> but the proper way is to use a release tarball prepared by upstream
2020-06-01 13:57:06 <ikke> even an automatically generated tarball based on a tag would work
2020-06-01 14:04:01 <indy> how to file bug/issue for package? spdlog-dev has dependency on fmt-dev but in tweakme.h is still commented out "#define SPDLOG_FMT_EXTERNAL"
2020-06-01 14:04:23 <maxice8> Open an issue on gitlab.a.o/alpine/aports
2020-06-01 14:04:35 <maxice8> algitbot: bruh
2020-06-01 14:15:57 <velix> Who do you have to bribe to get a package in `testing`? ;)
2020-06-01 14:19:47 <mps> velix: no one
2020-06-01 14:20:20 <mps> create merge request and wait or create issue with 'package request'
2020-06-01 14:20:44 <velix> okay, thanks
2020-06-01 14:33:01 <velix> Does this channel give support on APKBUILDs? For example, have a quick look over it?
2020-06-01 14:33:20 <Cogitri> Please just make a MR, it's not fun to review things over IRC :)
2020-06-01 14:33:34 <Cogitri> But if you have specific questions we can help with those, of course
2020-06-01 14:34:02 <Cogitri> But TL;DR make a MR, make the CI (so linter&build) happy and then you're pretty good to go usually
2020-06-01 14:37:15 <velix> Cogitri: Sorry, it's my first package contribution. I don't feel fine to publish it as MR directly.
2020-06-01 14:48:09 <Cogitri> velix: There's no problem in making a MR and marking that as WIP
2020-06-01 14:58:23 <Hello71> the whole point of MR is to have review
2020-06-01 15:01:20 <velix> How long does it stay in there normally ?
2020-06-01 15:02:22 <Cogitri> Depends on how quickly you resolve the remarks the reviewer makes
2020-06-01 15:02:37 <Cogitri> I'd say most MRs are merged within a week if there aren't too big problems in the thing
2020-06-01 15:03:12 <Cogitri> And if devs aren't too busy due to other things happening in Alpine (a new release) or smth else
2020-06-01 15:03:27 <Cogitri> Stats on that would be interesting, actually
2020-06-01 16:18:37 <velix> I've got a problem: my source file is redirected on download... how do I do this in APKBUILD correctly?
2020-06-01 16:18:52 <Cogitri> Huh?
2020-06-01 16:19:40 <velix> "https://github.com/mapnik/python-mapnik/archive/7da019c.tar.gz" gets rewritten on server side. Sure, I could wrap around snapshot() and download it from git.
2020-06-01 16:22:53 <velix> brb 10 minute. need a break after 4 hours of APKBUILD
2020-06-01 16:26:46 <Cogitri> Hum, GitHub should have stable checksums for their tarballs, if you mean that
2020-06-01 16:27:37 <Cogitri> Also, since you're a fresh contributor: Feel free to direct feedback about what you had trouble with or what was difficult towards me, hopefully I'll get to rewriting the docs for that in a bit
2020-06-01 16:33:38 <paulw> I have a nodejs package that compiles with cmake on install. This package is failing, and I think its because of musl, so i want to try and compile it using glibc instead. Does anyone know how to change what lib is used in a cmake project?
2020-06-01 16:34:39 <Cogitri> You'd have to use another distro since Alpine doesn't have glibc
2020-06-01 16:34:49 <paulw> i installed glibc on alpine
2020-06-01 16:35:16 <paulw> The node package in question is here: https://github.com/SAP/node-rfc
2020-06-01 16:37:46 <afontain_> alternatively, you can patch out the error, if applying patch is convinient
2020-06-01 16:38:25 <afontain_> If you upstream the patch, you don't have to patch anymore
2020-06-01 16:41:47 <omni> this is very cool https://postmarketos.org/blog/2020/05/31/three-years/
2020-06-01 16:55:14 <velix> Cogitri: The problem is, github doesn't have a stable endpoint for this URLs. If you check the URL I've posted, its name is actually created by a script. Currently, it's render.githubusercontent.com setting the remote name to python-mapnik-7da019cf9eb12af8f8aa88b7d75789dfcd1e901b.tar.gz. I don't know, how abuild doenloads the files from the server.
2020-06-01 16:55:25 <velix> Currently, you cannot link the correct "python-mapnik-7da019cf9eb12af8f8aa88b7d75789dfcd1e901b.tar.gz" directly.
2020-06-01 16:56:40 <Cogitri> Hm, but that definitely works for other APKBUILDs
2020-06-01 16:57:13 <Cogitri> velix: Look at community/telepathy-farstream/APKBUILD which does that
2020-06-01 16:57:22 <Cogitri> It doesn't do anything special
2020-06-01 16:58:21 <mps> and, testing/wiringx
2020-06-01 16:58:30 <mps> it was really simple
2020-06-01 16:58:35 <mps> :)
2020-06-01 16:59:30 <velix> Oh, I thought that wasn't possible.
2020-06-01 16:59:31 <velix> Thanks!
2020-06-01 17:00:20 <velix> Hmm, I don't like the naming scheme: 0.6.2_git20190919 <-- what if there were several commits these day?
2020-06-01 17:00:47 <afontain_> my technique is to wait a day to be sure :D
2020-06-01 17:00:51 <Cogitri> :D
2020-06-01 17:01:19 <velix> What about "git201909-c7299b4" ?
2020-06-01 17:01:28 <velix> Just to be sure..
2020-06-01 17:01:41 <Cogitri> The commit is in the APKBUILD anyway
2020-06-01 17:01:56 <afontain_> yes, but then the second commit will have an inferior version number half of the time
2020-06-01 17:02:10 <velix> ok ok ok ;)
2020-06-01 17:02:11 <Cogitri> Yup. apk wont know what commit is more recent
2020-06-01 17:02:39 <afontain_> we could add the hour in there
2020-06-01 17:02:47 <Cogitri> So you'd have to include seconds, minutes and hours in the pkgver to make really sure it really is that specific commit
2020-06-01 17:02:47 <velix> afontain_: in leapseconds.
2020-06-01 17:03:01 <velix> afontain_: Sorry, I had to make this TBBT joke.
2020-06-01 17:03:04 <Cogitri> But I don't think we often upgrade a package twice a day, reviews aren't quick enough for that :)
2020-06-01 17:03:14 <afontain_> and add the timezone!
2020-06-01 17:03:32 <velix> afontain_: Zulu time rocks.
2020-06-01 17:03:39 <afontain_> by the way, 2020-02-24 looks cleaner and easier to read imo
2020-06-01 17:04:05 <velix> Debian does it 1:0.0~20200224-7da019cf9-1
2020-06-01 17:04:37 <Cogitri> Let's make it RFC 5322 compatible :)
2020-06-01 17:04:45 <velix> 0.0, since it doesn't have a release, then the date, the short commit and the packaging number.
2020-06-01 17:05:25 <velix> afontain_: Perhaps the dashes make problems on autocomplete?
2020-06-01 17:05:41 <velix> 2020<tab><tab>...
2020-06-01 17:05:43 <ikke> Including the hash does not work. What if the new hash sorts before the old hash? No package would be upgraded
2020-06-01 17:06:45 <velix> brb some minutes. I'll put potatoes in the microwafer for my first time. please cross fingers.
2020-06-01 17:08:38 <velix> Okay, it's rotating.
2020-06-01 17:10:24 <afontain_> unrelated, but how do you get libstdc++'s debug syms?
2020-06-01 17:11:30 <Cogitri> Probably not at all since we don't have a -dbg subpackage for gcc unfortunately
2020-06-01 17:14:35 <Hello71> https://pkgs.alpinelinux.org/contents?file=libstdc%2B%2B.so*.debug
2020-06-01 17:15:22 <Hello71> wait
2020-06-01 17:15:54 <afontain_> :-(
2020-06-01 17:16:30 <Hello71> oh yeah, that's right. got confused because musl-dbg is under ld-musl, not libc.musl
2020-06-01 17:17:16 <velix> Sorry for offtopic: Wow, core temperatur of my potatoes is 90 °C after 5 minutes at 900 watts. I bet, they'll explore soon.
2020-06-01 17:17:26 <velix> explode*
2020-06-01 17:24:54 <velix> Does apk auto-recognize the running dependencies from the build dependencies?
2020-06-01 17:25:22 <afontain_> ouch, the gcc APKBUILD is huge
2020-06-01 17:28:24 <hiro> velix: i hope you didn't put any IoT into your potatoes
2020-06-01 17:28:51 <Cogitri> velix: It auto-recognises what packages contain the libraries that a package links against
2020-06-01 17:30:00 <velix> hiro: No, just things. 9 minutes and they're fine. Reached 98.6 °C
2020-06-01 17:30:08 <velix> Cogitri: Wow, nice.
2020-06-01 17:55:05 <Jenkler> Best supported mail client in alpine? See that thunderbird is in testing only
2020-06-01 17:55:34 <c705> mutt?
2020-06-01 17:55:40 <Jenkler> GUI
2020-06-01 17:55:57 <c705> mutt in a terminal emulator
2020-06-01 17:55:58 <Jenkler> that have support for oauth
2020-06-01 17:56:03 <c705> or a web app mail clienty
2020-06-01 17:56:27 <c705> i would be very surprised if mutt didn't support oauth authentication
2020-06-01 17:57:12 <Jenkler> Need a more drag en drop mail client
2020-06-01 17:57:48 <c705> Use thunderbird then. seems like that's what you've already decided
2020-06-01 17:58:32 <Jenkler> is Claws Mail any good?
2020-06-01 17:58:47 <hiro> Jenkler: it has the best looks
2020-06-01 17:58:55 <hiro> thunderbird is ugly and newfangled
2020-06-01 17:59:06 <afontain_> for gtk: claws-mail, geary, evolution
2020-06-01 17:59:06 <afontain_> kmail
2020-06-01 17:59:08 <afontain_> likely others
2020-06-01 17:59:29 <Jenkler> evolution is kinda bloaty right
2020-06-01 17:59:30 <mps> Jenkler: claws is better for normal users, thunderbird is 'corporate people'
2020-06-01 17:59:49 <Jenkler> mps, ahh
2020-06-01 18:00:05 <mps> actually, mutt is best for normal users :)
2020-06-01 18:00:57 <Jenkler> mutt is not that good for  beginners
2020-06-01 18:01:41 <c705> mutt is fine if you are capiable of reading documentation
2020-06-01 18:01:48 <c705> really isn't complicated at all
2020-06-01 18:01:52 <mps> it depends who is beginner, one who want to learn or  .......
2020-06-01 18:02:09 <Jenkler> hehe, yeah for sure ;)
2020-06-01 18:02:25 <Jenkler> But if they is used to thunderbird
2020-06-01 18:02:40 <c705> then use thunderbird
2020-06-01 18:02:54 <mps> my advice will be 'go to ubuntu'
2020-06-01 18:04:33 <Jenkler> is there a special reason why thunderbird exist only in testing?
2020-06-01 18:04:52 <Jenkler> is it unstable with musl?
2020-06-01 18:05:19 <c705> either it was recently added as a package or there isn't anyone actively maintaining the package
2020-06-01 18:06:00 <Jenkler> ok
2020-06-01 18:06:23 <rodrigue7973> hi all
2020-06-01 18:06:29 <Jenkler> hi
2020-06-01 18:06:31 <Cogitri> It's not unstable but no one has stepped up to make sure it gets security upgrades in stable releases
2020-06-01 18:06:34 <mps> c705: this
2020-06-01 18:06:39 <rodrigue7973> i a m a problem repository package
2020-06-01 18:06:55 <Cogitri> I wouldn't say Thunderbird is less "bloaty" than Evolution
2020-06-01 18:07:01 <Cogitri> And Evolution is pretty great
2020-06-01 18:07:09 <c705> mps: am i correct? or is it because thunderbird won't build for arm?
2020-06-01 18:07:10 <u0jQx9gPyrYg> i just had a horrible thunderbird experience, seems like mozilla is gutting not only firefox. they remove all addon support
2020-06-01 18:07:25 <Jenkler> Cogitri, seams that claws is kinda slim
2020-06-01 18:07:28 <mps> c705: no one wants to take maintainership, though I upgraded it few times and Cogitri iirc
2020-06-01 18:07:44 <rodrigue7973> i am error for
2020-06-01 18:07:45 <rodrigue7973> fetch http://dl-5.alpinelinux.org/alpine/3.12/main/x86_64/APKINDEX.tar.gz
2020-06-01 18:07:47 <rodrigue7973> ERROR: http://dl-5.alpinelinux.org/alpine/3.12/main: temporary error (try again later)
2020-06-01 18:08:01 <velix> Did you try dl-cdn ?
2020-06-01 18:08:11 <c705> mps: don't blame them..the apkbuild looks like a freaking nightmare
2020-06-01 18:08:18 <mps> c705: no, not because arm, there are more packages which doesn't build on some arches but they are maintained
2020-06-01 18:08:38 <c705> rodrigue7973: try a different mirror
2020-06-01 18:08:53 <c705> dl-5 is screwy iirc
2020-06-01 18:08:59 <c705> dl-2 is where it's at
2020-06-01 18:09:54 <rodrigue7973> other mirror is even a message !!!
2020-06-01 18:10:05 <ikke> check your dns
2020-06-01 18:10:13 <ikke> temporary issue almost always means your dns is not working
2020-06-01 18:10:18 <velix> I thought dl-cdn.alpine... is better?
2020-06-01 18:10:22 <ikke> temporary error*
2020-06-01 18:10:39 <c705> ikke: sometimes, yes. but some of the mirrors are not healthy as well
2020-06-01 18:11:13 <ikke> c705: we monitor the mirrors, I'm aware of only one mirror currently having issues
2020-06-01 18:11:28 <rodrigue7973> what dns check ?
2020-06-01 18:11:59 <Jenkler> rodrigue7973, ping www.google.com ?
2020-06-01 18:11:59 <ikke> try getent hosts alpinelinux.org
2020-06-01 18:12:40 <rodrigue7973> alpine-srv:/home/rodrigue# ping www.google.com
2020-06-01 18:12:42 <rodrigue7973> PING www.google.com (172.217.168.228): 56 data bytes
2020-06-01 18:12:43 <rodrigue7973> 64 bytes from 172.217.168.228: seq=0 ttl=63 time=15.811 ms
2020-06-01 18:12:45 <rodrigue7973> 64 bytes from 172.217.168.228: seq=1 ttl=63 time=13.928 ms
2020-06-01 18:12:46 <rodrigue7973> 64 bytes from 172.217.168.228: seq=2 ttl=63 time=14.664 ms
2020-06-01 18:12:48 <rodrigue7973> 64 bytes from 172.217.168.228: seq=3 ttl=63 time=13.127 ms
2020-06-01 18:12:49 <rodrigue7973> 64 bytes from 172.217.168.228: seq=4 ttl=63 time=13.990 ms
2020-06-01 18:12:51 <rodrigue7973> 64 bytes from 172.217.168.228: seq=5 ttl=63 time=14.887 ms
2020-06-01 18:12:52 <rodrigue7973> 64 bytes from 172.217.168.228: seq=6 ttl=63 time=12.973 ms
2020-06-01 18:12:54 <rodrigue7973> ^C
2020-06-01 18:12:55 <rodrigue7973> --- www.google.com ping statistics ---
2020-06-01 18:12:57 <rodrigue7973> 7 packets transmitted, 7 packets received, 0% packet loss
2020-06-01 18:12:58 <rodrigue7973> round-trip min/avg/max = 12.973/14.197/15.811 ms
2020-06-01 18:13:00 <rodrigue7973> alpine-srv:/home/rodrigue#
2020-06-01 18:13:03 <maxice8> use a paste service please
2020-06-01 18:13:15 <rodrigue7973> sorry how paste service ?
2020-06-01 18:13:33 <maxice8> Paatebin
2020-06-01 18:13:36 <Jenkler> https://pastebin.com/
2020-06-01 18:13:44 <c705> rodrigue7973: echo whatever | nc termbin.com 9999
2020-06-01 18:16:41 <rodrigue7973> i not say
2020-06-01 18:17:26 <c705> ikke: hmm. I used to randomly select mirrors on boot out of the list on mirrors.alpinelinux.org. some of them were god awfuly slow, but I swear some of them would throw me a "temporary error" even after sucessfully contacting the mirror. might have been temporary issue, i don't know
2020-06-01 18:17:52 <c705> since i've slimmed my list down to 12 or so mirrors that i know to be healthy, i don;t have issues anymore
2020-06-01 18:18:30 <artok> so is evolution now decent to try instead of mutt ?
2020-06-01 18:18:49 <mps> artok: it will never be :)
2020-06-01 18:19:02 <Cogitri> I use it for all of my mailing needs and calendar stuff
2020-06-01 18:19:19 <Cogitri> Integration with EWS is super nice for work
2020-06-01 18:22:02 <artok> oh for the ff
2020-06-01 18:22:32 <artok> I can't login to gmail with password.. it forces to use oauth
2020-06-01 18:23:02 <c705> was someone going to package icecat? that'd be cool
2020-06-01 18:23:43 <mps> someone told few days ago that
2020-06-01 18:24:02 <mps> and I'm waiting for MR
2020-06-01 18:24:07 <mps> amom: ^
2020-06-01 18:24:23 <ikke> artok: I think you should be able to generate an app specific password and use that
2020-06-01 18:25:29 <ikke> c705: It makes sense that some mirrors are not feasible for you, because they are in a different part of the world
2020-06-01 18:25:43 <artok> ikke: I set imap server for gmail and auth method: password on evolution settings, but when it starts to connect, it switches to oauth
2020-06-01 18:26:09 <amom> mps: I looked into it
2020-06-01 18:26:15 <artok> apk del evolution
2020-06-01 18:26:25 <c705> ikke: that makes sense
2020-06-01 18:26:34 <ikke> c705: but in general the mirrors should be at least available and up-to-date
2020-06-01 18:26:35 <amom> the only machines I have that are running alpine are using pmos
2020-06-01 18:26:47 <amom> and Im having trouble with the mesa-git packages installed on them
2020-06-01 18:27:11 <c705> ikke: i'm also connecting over a vpn, so perhaps i'm getting limited or dropped because of that
2020-06-01 18:27:18 <c705> i've always wondered
2020-06-01 18:27:22 <amom> and I dont know how to fix the conflict that arises cause Im new to alpine. Also I dont wanna break my install. Also, I would be compiling on a pinebook pro
2020-06-01 18:27:55 <ikke> c705: but in almost all cases you got a temporary error, it's because the domain for some reason could not be resolved
2020-06-01 18:28:18 <amom> I dont know how to continue until I can read the wiki page on apk. Ill try to figure it out after this assignment
2020-06-01 18:28:37 <amom> I might have to just make an alpine linux virtual machine
2020-06-01 18:29:20 <c705> ikke: i agree, especially when you get no rx from the mirror, but i've seen temporary error in other cases (i imagine its a generic throw when there's network issues in general)
2020-06-01 18:40:39 <rodrigue7973> yes for am virtuabox
2020-06-01 18:44:18 <mps> amom: send me APKBUILD and I will test build on some of our big machines
2020-06-01 18:48:26 <Marquel> ahhhh... the _niiiceeeee_ feeling when one knows how a single command will spool up a bunch of cpus in some remote datacenter, send status leds into hurriedly blinking and the spinning wheel of the powermeters into glowing red from atmospheric friction.... *sigh*
2020-06-01 18:49:33 <maxice8> are you ok ?
2020-06-01 18:53:14 <velix> Hell, I'm going *mad*. The repository has a submodule, which isn't part of the tar.gz archive.
2020-06-01 18:53:25 <maxice8> that is an ongoing issue
2020-06-01 18:53:25 <velix> Now things are getting complicated.
2020-06-01 18:53:43 <maxice8> is it GitHub ?
2020-06-01 18:53:49 <velix> maxice8: yep
2020-06-01 18:54:10 <maxice8> add https://github.com/user/project/commit.tar.gz and in prepare() use `mv` to move it to the correct location
2020-06-01 18:54:22 <Marquel> maxice8: me? no, why?
2020-06-01 18:54:55 <maxice8> @Marquel yes you, hope you get well soon
2020-06-01 18:55:04 <velix> maxice8: I'm having problems to find the commit in github right now. Actually it's not listed on the website. But maybe I can figure it out.
2020-06-01 18:55:20 <maxice8> can you send the link to the repo that has the submodule ?
2020-06-01 18:56:07 <velix> sure: https://github.com/mapnik/python-mapnik - branch v3.0.x - commit 7da019cf9eb12af8f8aa88b7d75789dfcd1e901b
2020-06-01 18:56:47 <Marquel> maxice8: tbh., i'm working for one of the three major server vendors of the planet, am running two of its rather more pricey products out of my own pocket and am the company's linux guru for that matter. whaddaya think? if i were "ok" in any common sense i'd be making wooden furniture or be busy growing food plants on some untampered piece of soil and not hang around in an IRC channel for some obscure linux distribution which just
2020-06-01 18:56:48 <Marquel> by coincidence happens to be the hypervisor for one of the aforementioned pricey x86 servers... ;)
2020-06-01 18:57:20 <amom> mps: I dont even know if itll run
2020-06-01 18:57:25 <amom> do you still want it?
2020-06-01 18:58:38 <maxice8> @velix python-mapnik the submodules are the one under test ?
2020-06-01 18:59:03 <maxice8> @Marquel fascinating
2020-06-01 18:59:33 <velix> maxice8: Uh, I think the problems is different than I thought. *Another* maintainer seems to have missed the file in mapnik-dev (in testing). Now python-mapnik cannot find this file. It's not part of python-manik, but of mapnik-dev as it seems.
2020-06-01 18:59:43 <Marquel> maxice8: *raises eyebrow*
2020-06-01 18:59:50 <mps> amom: well, I thought you want it packaged for alpine, and I was ready to help
2020-06-01 19:00:13 <amom> mps: yes I do. I will boot up the pinebook pro and send it
2020-06-01 19:00:19 <velix> maxice8: No, it's there. Confusing. I'll keep on digging. Thanks for your idea.
2020-06-01 19:00:27 <maxice8> @Velix sorry but i don't understand what you're saying
2020-06-01 19:01:58 <velix> maxice8: Yeah, since I also don't know the problem. python-mapnik (which I'm creating right now) is missing a file, which should have been provided by mapnik-dev.
2020-06-01 19:02:01 <Marquel> maxice8: or, to be a bit more sincere about it: since i do setup datacenters for our customers, spinning up cpus and sending powermeters into heavy rotation is kind of my daily job, sooo.... i'm just exaggerating and putting a bit of poetry behind such a standard task as "running a build job on one of the big machines" (which is just done that way to save time, not because it is strictly necessary)...
2020-06-01 19:02:30 <maxice8> @velix sounds like an issue needs to be opened in gitlab.alpinelinux.org/alpinelinux/aports
2020-06-01 19:02:54 <velix> maxice8: I'm investigating this right now.
2020-06-01 19:03:02 <maxice8> @velix Good Luck
2020-06-01 19:03:20 <velix> maxice8: thanks ;)
2020-06-01 19:03:20 <mps> Marquel: firefox even don't build on 4GB machine
2020-06-01 19:04:08 <velix> maxice8: But I've found another issue, which needs to be reported: mapnik-config has a bash shebang, which is bad. It also works with normal shell. I'lll report it
2020-06-01 19:04:28 <maxice8> @velix, oh, thanks
2020-06-01 19:04:41 <maxice8> feel free to tag @Leo for the bash shebang issue
2020-06-01 19:04:50 <Marquel> mps: ask me about building qt-webengine and you get why i'm considering one of the less it-affected jobs in making wooden furniture. and it has absolutely nothing to do with that not-so-funny-sitcom called "home improvement". no way!
2020-06-01 19:06:25 <mps> yes, sometimes also I dream for another kind of job
2020-06-01 19:07:26 <amom> mps: https://git.sr.ht/~anjan/icecat-alpine
2020-06-01 19:07:42 <amom> It's mostly copy pasta from firefox-esr
2020-06-01 19:07:47 <amom> but I did do some work on it.
2020-06-01 19:08:23 <doggone> Marquel: what are the three major server vendors?
2020-06-01 19:08:29 <amom> Please let me know when you have it working or have given up lol.
2020-06-01 19:08:38 <mps> amom: looking
2020-06-01 19:08:40 <amom> I appreciate the help! I love icecat
2020-06-01 19:08:47 <doggone> dell, hp, ...?
2020-06-01 19:09:03 <Marquel> doggone: HPE, Dell and Fujitsu/IBM (now Lenovo) in about that order of market share.
2020-06-01 19:10:25 <doggone> and which one is the best? :)
2020-06-01 19:10:30 <doggone> asking for a friend
2020-06-01 19:11:20 <Marquel> well....
2020-06-01 19:12:42 <Marquel> as i tell my customers: they're all cooking with the same water. it's some kind of reference mainboard from the cpu vendor (which is either intel or amd), some ram from kingston or samsung, a raid controller from lsi, a networkcard from mellanox, broadcom or intel and the rest is just sugar as toppings.
2020-06-01 19:13:14 <Marquel> so. what you want is the better support for your investment as that's what's actually differentiating them.
2020-06-01 19:13:57 <Marquel> and there i can give you, of course, an advice, but that'd be biased as i don't have first-hand experience with the competition and can just relay what customers told me.
2020-06-01 19:14:48 <velix> maxice8: Ha! I've found the problem. It's not in `/usr/include/mapnik/mapbox/variant.h` as expected, but in `/usr/include/mapnik/mapnik/mapbox/variant.h`. Seems like the packager has prefixed `mapnik/` somewhere.
2020-06-01 19:15:26 <maxice8> oh
2020-06-01 19:15:39 <maxice8> see lines 58 to 60 of testing/mapnik/APKBUILD
2020-06-01 19:15:44 <doggone> Having out-of-band management is nice, but I usually just tend to build stuff myself. For some reason, regular boards boot way faster (not that important for a server, admittedly) and they are of course, way cheaper.
2020-06-01 19:15:54 <maxice8> it makes /usr/include/mapnik
2020-06-01 19:16:17 <velix> maxice8: Exactly.
2020-06-01 19:16:17 <maxice8> then move a/usr/include/mapnik to b/usr/include/mapnik, but since /usr/include/mapnik already exists it ends up on /usr/include/mapnik/mapnik
2020-06-01 19:16:27 <doggone> I used to work for a company that got good support from HPE, but I don't have experience with the others, so I was just wondering.
2020-06-01 19:16:38 <maxice8> shell has lots of <sarcasm>fun</sarcasm> pitfalls
2020-06-01 19:16:43 <velix> maxice8: And this, of course, makes the includes blow up ;)
2020-06-01 19:18:31 <Marquel> doggone: what i've heard boils down to "HPE's support depends largely on volume", but as said, that might be biased.
2020-06-01 19:20:59 <mps> amom: this will not work. in source are just 'add-ons' for icecat, not source
2020-06-01 19:22:06 <velix> maxice8: I don't know Alpine's philosophy. Since it's a header coming with mapnik, isn't it correct to throw them under /usr/include/mapnik? Because when throwing it in /usr/include, it'll end up in `/usr/include/mapnik` and `/usr/include/mapbox`, which might violate packaging rules?
2020-06-01 19:22:29 <maxice8> I think that is just a packaging error
2020-06-01 19:22:34 <maxice8> nothing worth getting philosophical about
2020-06-01 19:22:34 <amom> mps: ignore the src directory. I think thats a mistake.
2020-06-01 19:22:50 <velix> maxice8: Sorry, I'm coming from Debian and this might end of up debates of years there.
2020-06-01 19:23:22 <amom> actually some of those are patches so idk
2020-06-01 19:24:21 <mps> yes
2020-06-01 19:34:15 <mps> amom: you can look at void linux icecat package
2020-06-01 19:34:39 <amom> mps: okay I will take a look
2020-06-01 19:34:50 <amom> I should setup the alpine virtual machine
2020-06-01 19:38:15 <mps> lxc or chroot will also work
2020-06-01 20:00:23 <velix> Mapnik comes with its own set of DejaVue fonts. I think it makes sense to put them into their own package, what do you think? The bundled ones are pretty old.
2020-06-01 20:00:28 <velix> DejaVu*
2020-06-01 20:10:53 <velix> done. I like it clean.
2020-06-01 20:19:42 <velix> Uh ... eh ... how do I sed "noarch" for a subpackage?
2020-06-01 20:19:56 <maxice8> subpackages="$pkgname-foo::noarch"
2020-06-01 20:20:25 <velix> maxice8: Thanks.
2020-06-01 20:20:45 <velix> Alpine packaging is way easier than Debian packaging ;)
2020-06-01 20:23:49 <ikke> Nod
2020-06-01 20:26:19 <velix> Sorry for asking so basic things, but googling is very hard. apk is also used in Android, so all results link to android. How do I view the content of APK files? "mc" doesn't support it, 7zip doesn't support it etc.
2020-06-01 20:27:02 <maxice8> apk are just archives
2020-06-01 20:27:05 <Cogitri> tar xf
2020-06-01 20:27:15 <velix> Cogitri: uncompressed?
2020-06-01 20:27:29 <ikke> yes
2020-06-01 20:27:35 <velix> oh ok
2020-06-01 20:27:46 <velix> Just curious: why?
2020-06-01 20:28:10 <ikke> why uncompressed?
2020-06-01 20:28:13 <velix> Even gzip or zstd at lowest level give damn good results pretty often.
2020-06-01 20:28:13 <velix> Yeah
2020-06-01 20:28:16 <maxice8> you can use `tar tvf`
2020-06-01 20:28:22 <maxice8> it will list files inside of it
2020-06-01 20:28:31 <velix> maxice8: Yeah, I'm using it now :D I thought they're compressed.
2020-06-01 20:28:46 <Cogitri> Even if they are, tar xf would take care of yhat
2020-06-01 20:29:07 <ikke> Cogitri: I don't think busybox tar would
2020-06-01 20:29:30 <Cogitri> Ah, I always use GNU tar
2020-06-01 20:29:40 <velix> Wow, now mapnik-fonts has the same size as mapnik the binaries: 5 MB. I've saved all the users 50% of duplicate data.
2020-06-01 20:29:43 <ikke> maxice8: gops fails on mips
2020-06-01 20:30:01 <maxice8> oops
2020-06-01 20:30:03 <maxice8> i'll disable on mips
2020-06-01 20:30:16 <velix> "gops fails on mips"... nerds.
2020-06-01 20:30:22 <ikke> velix: I think the idea is that most packages would just contain binaries, so compression adds little, and speeds up apk
2020-06-01 20:30:26 <ikke> but not sure if that's the reasoning
2020-06-01 20:30:35 <velix> ikke: all the docs?!
2020-06-01 20:30:50 <velix> ikke: actually, all noarch stuff would win.
2020-06-01 20:31:03 <velix> all those header files are text.
2020-06-01 20:31:18 <ikke> but typically small in comparisson
2020-06-01 20:31:54 <Cogitri> ikke: are you sure it's not compressed?
2020-06-01 20:31:57 <ikke> ah, but it *is* compressed
2020-06-01 20:31:59 <ikke> yeah, just checked
2020-06-01 20:32:08 <velix> :DD
2020-06-01 20:32:08 <Cogitri> abuild spends a considerable amount of time compressing things for me
2020-06-01 20:32:10 <ikke> git-2.27.0-r0.apk: gzip compressed data, max compression, from Unix, original size modulo 2^32 16629760
2020-06-01 20:32:15 <ikke> yeah, true
2020-06-01 20:32:16 <Cogitri> I think it doesn't use multiple threads
2020-06-01 20:32:24 <ikke> it does if you have pigz installed
2020-06-01 20:32:30 <Cogitri> Oh
2020-06-01 20:32:32 <velix> ikke: nice!
2020-06-01 20:32:33 <Cogitri> I should do that
2020-06-01 20:32:37 <velix> me, too ;)
2020-06-01 20:32:48 <ikke> https://gitlab.alpinelinux.org/alpine/abuild/blob/master/abuild.in#L482
2020-06-01 20:32:51 <Cogitri> Takes like 5 minutes to compress electron debug info because it's 25gigs :D
2020-06-01 20:32:56 <ikke> that's unpack
2020-06-01 20:33:13 <ikke> https://gitlab.alpinelinux.org/alpine/abuild/blob/master/abuild.in#L1586
2020-06-01 20:33:15 <ikke> that's packing
2020-06-01 20:33:36 <Cogitri> 👍
2020-06-01 20:34:10 <ikke> So funny that I knew that fact, yet, somehow was under the impression they might be uncompressed :)
2020-06-01 20:34:39 <velix> Interesting. Why does it use sha256, I thought, it's using sha512 anywhere else?
2020-06-01 20:34:59 <Cogitri> velix: Because switching that in a package format isn't fun
2020-06-01 20:35:05 <Cogitri> We'll switch with apkv3 I think
2020-06-01 20:35:15 <Cogitri> And sha256 is still fine-ish
2020-06-01 20:35:24 <ikke> more than fine
2020-06-01 20:35:31 <velix> Sure, I just was suprised, sha512 is used on packages.
2020-06-01 20:35:33 <velix> APKBUILDs
2020-06-01 20:35:46 <Cogitri> ikke: yes, sha512 is just future proofing at this point
2020-06-01 20:35:57 <ikke> Yes, but that's just abuild that's verying during build, which is pretty easy to switch
2020-06-01 20:35:58 <Cogitri> Until quantum computers screw everything :D
2020-06-01 20:36:10 <ikke> the package format is more rigid
2020-06-01 20:37:36 <ikke> I believe it requires us at some point to rebuild all packages
2020-06-01 20:38:05 <Cogitri> Yes
2020-06-01 20:38:21 <Cogitri> But apk will probably support both the old and new packaging format for some time
2020-06-01 20:41:51 <velix> JOBS=20 abuild -r
2020-06-01 20:41:56 <velix> woops, wrong window
2020-06-01 20:42:05 <ikke> :)
2020-06-01 20:42:07 <maxice8> i think you can set in /etc/abuild.conf
2020-06-01 20:46:26 <ikke> yes, you can
2020-06-01 21:02:08 <MY-R> mps, hey! thank you for sweet linux-edge config. I dont need recompile my kernels anymore :)
2020-06-01 21:04:07 <ikke> FYI: it's in testing, so beware ;)
2020-06-01 21:05:06 <mps> ikke: why?
2020-06-01 21:05:18 <mps> MY-R: you are welcome :)
2020-06-01 21:05:29 <ikke> linux-edge is in testing, right?
2020-06-01 21:05:52 <mps> right, but that doesn't mean it is bad
2020-06-01 21:06:00 <ikke> I was not implying it was bad
2020-06-01 21:06:30 <mps> it is in testing because we don't want to maintain it stable releases
2020-06-01 21:06:43 <ikke> yes, and because we want to test features
2020-06-01 21:07:35 <mps> hmm, test features is second goal, first goal is to be useful for those who need new drivers and features
2020-06-01 21:07:39 <MY-R> full preempt and 1000hz should be default IMHO :)
2020-06-01 21:07:53 <mps> MY-R: isn't it :)
2020-06-01 21:08:03 <ikke> Was referring to: https://gitlab.alpinelinux.org/alpine/aports/-/blob/master/testing/linux-edge/APKBUILD#L6
2020-06-01 21:08:06 <MY-R> not in -lts :)
2020-06-01 21:08:23 <mps> performance is one of the goal for linux-edge
2020-06-01 21:08:49 <mps> so not much perf, debug, counters etc, just pure speed
2020-06-01 21:09:08 <MY-R> my old tp link after compile with 1000hz and tickless (openwrt) can nat 900~ mbits without problem pff
2020-06-01 21:09:29 <mps> look at configs for linux-edge
2020-06-01 21:10:03 <mps> CONFIG_HZ_1000=y
2020-06-01 21:10:26 <MY-R> ye but in linux-lts still 250hz
2020-06-01 21:10:29 <ikke> What is 1000hz about?
2020-06-01 21:10:29 <mps> and, CONFIG_PREEMPT=y
2020-06-01 21:10:55 <mps> ikke:  responsivnes
2020-06-01 21:11:09 <MY-R> web browsing and playing games are much more smooth, dont need benchmarks, done blind tests few times and always can notice difference
2020-06-01 21:11:20 <ikke> Why is it not default?
2020-06-01 21:11:29 <ikke> ie, what is the trade-off?
2020-06-01 21:11:38 <MY-R> good question, but not all devices probably can handle with it
2020-06-01 21:12:04 <Cogitri> I don't think that's the problem
2020-06-01 21:12:05 <mps> well, some specific server loads prefer less 'interrupts'
2020-06-01 21:12:23 <Cogitri> But in theory you're always trading throughout vs latency
2020-06-01 21:12:30 <mps> mostly intensive computing tasks
2020-06-01 21:12:34 <ikke> So it's about how quickly processes can be preempted?
2020-06-01 21:12:45 <Cogitri> So with 1000Hz you're getting a bit less throughput but better latency
2020-06-01 21:12:52 <Cogitri> But I don't think if it's actually a noticeable difference
2020-06-01 21:13:04 <MY-R> with no preempt (server) 1000hz working great too
2020-06-01 21:13:14 <mps> yes, CONFIG_PREEMPT_NONE and CONFIG_PREEMPT_VOLUNTARY and CONFIG_PREEMPT
2020-06-01 21:13:15 <ikke> Yeah, I can imagine throughput suffers from more context switches
2020-06-01 21:13:16 <Cogitri> Heh, changed the sentence halfway in :D
2020-06-01 21:13:32 <ikke> Cogitri: never happens to me :P
2020-06-01 21:14:24 <MY-R> the problem now is with xorg :( since I'm using linux my mouse never lag/frozen... but after update xorg to use meson it is... :\
2020-06-01 21:14:30 <mps> when you run some jobs in background and at same time play videos or browse web you can notice diffs
2020-06-01 21:15:10 <Cogitri> MY-R: time to switch to Wayland! :D
2020-06-01 21:15:20 <MY-R> if somebody is curious why: https://gitlab.freedesktop.org/xorg/xserver/-/issues/936 https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/31
2020-06-01 21:15:41 <MY-R> Cogitri, nah, tried few times, always some small issues with it :\
2020-06-01 21:15:58 <Cogitri> I use it full-time, works great
2020-06-01 21:16:06 <Cogitri> But I understand that you want to keep using Xorg :)
2020-06-01 21:16:12 <Cogitri> I hope they'll get that fixed soon
2020-06-01 21:16:29 <Cogitri> Maybe you can add an issue about it on gitlab.a.o so we don't forget about it?
2020-06-01 21:16:33 <c705> Cogitri: me too, but xclip is broken for obvious reasons. how is xclip for you?
2020-06-01 21:16:35 <Cogitri> 
ACTION hopes algitbot does its thing

2020-06-01 21:16:37 <Cogitri> Good good
2020-06-01 21:16:49 <Cogitri> c705: Dunno, I don't use it
2020-06-01 21:16:55 <bleb> i thought alpine was on sr.ht?
2020-06-01 21:17:03 <MY-R> Cogitri, for sure will do it soon :)
2020-06-01 21:17:10 <ikke> bleb: no
2020-06-01 21:17:30 <ikke> bleb: we use sr.ht software for the mailing list atm
2020-06-01 21:17:41 <bleb> ok
2020-06-01 21:17:56 <bleb> and you moved from cgit to gitlab recently?
2020-06-01 21:18:21 <Cogitri> We moved from GitHub to Gitlab a few months ago
2020-06-01 21:18:33 <ikke> And we made gitlab the cannonical source (from gitolite)
2020-06-01 21:18:44 <ikke> we still have cgit running
2020-06-01 21:26:07 <bleb> cool
2020-06-01 21:26:22 <bleb> so i'm trying to run firefox from an alpine chroot on debian
2020-06-01 21:26:37 <bleb> i've set up my chroot according to the alpine chroot wiki page
2020-06-01 21:26:46 <bleb> except i mounted /dev read/write
2020-06-01 21:27:11 <bleb> and i also did mount -o bind /tmp/.X11-unix ~/chroot/tmp/.X11-unix
2020-06-01 21:27:27 <bleb> and xhost + outside the chroot
2020-06-01 21:27:38 <bleb> now when i run firefox i get a seg fault
2020-06-01 21:27:47 <bleb> and strace outputs this: http://ix.io/2o4f
2020-06-01 21:28:26 <bleb> seems like the seg fault might have something to do with readlink("/proc/self/exe", "/usr/lib/firefox/firefox", 4095) = 24
2020-06-01 21:28:33 <bleb> but idk what to try next
2020-06-01 21:28:35 <bleb> any suggestions?
2020-06-01 21:30:36 <ikke> maybe see if you can get a stacktrace with gdb, but otherwise, no clue
2020-06-01 21:32:19 <c705> bleb: firefox needs dbus. i ran into issues because firefox couldn't acces a dbus service. ymmv
2020-06-01 21:34:37 <bleb> maybe i can run dbus in the chroot
2020-06-01 21:34:58 <c705> that's pretty much what I was doing
2020-06-01 21:35:06 <bleb> and it didnt work?
2020-06-01 21:35:15 <c705> fwiw, i was using docker instead but the concept is the same
2020-06-01 21:35:19 <joe9> does firefox even work on muslc (alpine)?
2020-06-01 21:35:25 <bleb> yeah
2020-06-01 21:35:29 <c705> kind of
2020-06-01 21:35:36 <joe9> I suspect there might be a hard dependency on glibc.
2020-06-01 21:35:43 <c705> actually, no it does, the problem I have with it are its lack of support for alsa
2020-06-01 21:35:53 <c705> joe9: not at all
2020-06-01 21:36:13 <c705> chrome i think has a hard dep on glibc
2020-06-01 21:36:16 <bleb> joe9: iirc some alpine devs made some upstream patches to get it to work on musl
2020-06-01 21:36:46 <joe9> cool..
2020-06-01 21:38:01 <joe9> bleb, I suspect you might switch to OpenBSD pretty soon...
2020-06-01 21:38:25 <bleb> heheh
2020-06-01 21:38:26 <bleb> maybe
2020-06-01 21:39:02 <artok> firefox needs pulse nowadays?
2020-06-01 21:39:04 <joe9> My hunt for a secure and reliable OS led me to it.
2020-06-01 21:39:17 <mps> artok: no
2020-06-01 21:41:09 <bleb> woah firefox is running
2020-06-01 21:41:32 <bleb> after i did setup-xorg-base
2020-06-01 21:41:43 <artok> +
2020-06-01 21:41:47 <artok> oops
2020-06-01 21:41:58 <bleb> that probably includes a bunch of stuff i dont need
2020-06-01 21:41:59 <artok> mps: that was then something I had on a dream
2020-06-01 21:42:15 <bleb> and sound will probably be more of a challenge...
2020-06-01 21:43:45 <bleb> you know what sound is working also
2020-06-01 21:43:50 <bleb> what do you know
2020-06-01 21:45:23 <mps> artok: actually I think libpulse is added to alpine firefox about month ago (not to firefox-esr) but not as hard deps
2020-06-01 21:47:33 <tehcloud> good luck on OpenBSD, the maintenance has suffered over the last few years
2020-06-01 21:49:27 <c705> mps: firefox expects pulse, yes. --enable-alsa-support is not maintained anymore by firefox as far as i know
2020-06-01 21:49:46 <velix> Wait, does alpine use /usr/share/docs or /usr/share/doc ?
2020-06-01 21:49:47 <c705> the feature is still in the code, but not maintained so who knows when that'll stop working
2020-06-01 21:50:00 <c705> tehcloud: who is talking about openbsd?
2020-06-01 21:50:11 <tehcloud> <joe9> bleb, I suspect you might switch to OpenBSD pretty soon...
2020-06-01 21:50:57 <c705> oh
2020-06-01 21:51:23 <artok> c705: so that wasn't my dream then, woah
2020-06-01 21:51:36 <artok> or I wasn't that drunk
2020-06-01 21:54:28 <velix> Is there $subpkgname ?
2020-06-01 22:06:05 <velix> Does alpine use /usr/share/docs or /usr/share/doc ?
2020-06-01 22:18:11 <velix> Seems like alpine doesn't use doc(s) at all.
2020-06-01 22:18:24 <velix> I've looked in 40 different APKBUILDs the last 15 minutes.
2020-06-01 22:18:29 <velix> None is using docs.
2020-06-01 22:18:35 <maxice8> did you look at APKBUILDs that ened with -doc ?
2020-06-01 22:19:01 <velix> maxice8: Yes, in main, there's only 1
2020-06-01 22:19:10 <velix> https://git.alpinelinux.org/aports/tree/main
2020-06-01 22:19:37 <velix> f.e. https://git.alpinelinux.org/aports/tree/main/zlib/APKBUILD
2020-06-01 22:19:40 <velix> No docs.
2020-06-01 22:19:42 <maxice8> https://pkgs.alpinelinux.org/contents?file=*&path=%2Fusr%2Fshare%2Fdoc&name=&branch=edge&repo=main&arch=x86_64
2020-06-01 22:19:43 <maxice8> yeah seems like they use neither
2020-06-01 22:19:43 <c705> uh no, i count over 2000
2020-06-01 22:20:18 <c705> apk search *-doc
2020-06-01 22:20:40 <velix> Yeah, but I want the APLBUILD not the final package ;)
2020-06-01 22:20:51 <velix> https://git.alpinelinux.org/aports/tree/main -> only 1
2020-06-01 22:21:16 <maxice8> huh ?
2020-06-01 22:21:25 <c705> so you query the aports APKBUILD file based on packages that end with -doc
2020-06-01 22:22:40 <velix> Interesting: https://git.alpinelinux.org/aports/tree/main/procps/APKBUILD
2020-06-01 22:22:49 <velix> Seems like abuild automatically moved doc files?
2020-06-01 22:22:54 <maxice8> yes
2020-06-01 22:23:01 <maxice8> into a subpackage that is suffixed with -doc
2020-06-01 22:23:10 <maxice8> just like development files are moved into a package suffixed with -dev
2020-06-01 22:23:21 <maxice8> languages go into -lang, openrc services into -openrc, and so on
2020-06-01 22:24:02 <velix> But according to GPL, I need to copy the license text to each package. Isn't this needed in Alpine?
2020-06-01 22:24:29 <velix> I've seen a lot (L)GPL pages without the license text.
2020-06-01 22:24:42 <velix> s/pages/packages
2020-06-01 22:43:56 <velix> "WARNING: mapnik-dev*: Found static archive on usr/lib/libmapnik-wkt.a but name doesn't end with -static
2020-06-01 22:43:56 <velix> " - okay, seems like it doesn't work in this case.
2020-06-02 00:28:57 <bleb> after xorg-setup-base, i only have bitmap fonts and the fonts can't show numbers
2020-06-02 00:29:04 <bleb> or directional quotes
2020-06-02 00:29:24 <bleb> is there a quick way to get numbers and directional quotes working?  i like the bitmap fonts
2020-06-02 00:32:45 <tehcloud> fonts are a sore point in this distro, lol
2020-06-02 00:33:09 <tehcloud> my advice is to just experiment with the various font packages
2020-06-02 00:46:20 <Hello71> MY-R: at that point you'd probably be better off using linux-zen
2020-06-02 00:47:13 <Hello71> maxice8: you miss *
2020-06-02 00:47:34 <maxice8> ?
2020-06-02 00:52:35 <Hello71> /usr/share/doc*
2020-06-02 00:52:54 <Hello71> not the same as *, /usr/share/doc
2020-06-02 00:54:35 <maxice8> ?
2020-06-02 01:01:45 <Hello71> 22:19 <maxice8> https://pkgs.alpinelinux.org/contents?file=*&path=%2Fusr%2Fshare%2Fdoc&name=&branch=edge&repo=main&arch=x86_64
2020-06-02 01:02:31 <maxice8> Yes?
2020-06-02 01:06:05 <Hello71> dunno what you're looking for there
2020-06-02 01:07:32 <maxice8> If alpine has stuff in either doc or docs under usr/share
2020-06-02 01:23:59 <lunaphyte> what does "trigger will generate final grub.cfg" mean here : https://gitlab.alpinelinux.org/alpine/alpine-conf/-/blob/master/setup-disk.in#L309 ?
2020-06-02 01:24:10 <lunaphyte> what trigger?
2020-06-02 01:25:49 <mcrute> lunaphyte: there's a post install trigger in the grub package (see: https://git.alpinelinux.org/aports/tree/main/grub/grub.trigger)
2020-06-02 01:26:00 <mcrute> apk will run it automatically after grub is installed
2020-06-02 01:30:45 <lunaphyte> ah, ok
2020-06-02 01:31:50 <lunaphyte> the /etc/default/ config gets put in place before the package[s] are installed, so then once the package is installed [i guess like line ~492?], the install trigger causes the config the be generated?
2020-06-02 01:51:11 <mcrute> yeah
2020-06-02 02:10:24 <lunaphyte> that makes sense, thanks
2020-06-02 02:12:03 <lunaphyte> i'm doing an install on a qcow2 image mounted via qemu-nbd --connect, and i'm still trying to figure out
2020-06-02 02:12:30 <lunaphyte> *figure out why grub-install fails
2020-06-02 06:23:53 <mps> tehcloud: yes, with current linux ecosystem bsd world looks more and more alluring
2020-06-02 07:22:57 <PhoenixMage> Has abuild $pkgdir moved? For some reason on a new package I am making it has gone from the APKBUILD dir to the src dir
2020-06-02 07:23:27 <ikke> PhoenixMage: no, it has not
2020-06-02 07:24:02 <ikke> https://gitlab.alpinelinux.org/alpine/abuild/blob/master/abuild.in#L2713
2020-06-02 07:26:27 <PhoenixMage> hmmm
2020-06-02 07:29:08 <PhoenixMage> lol, looks like its the src itself that made it
2020-06-02 07:52:02 <markand> hi
2020-06-02 07:52:10 <rodrigue7973> markand: yo
2020-06-02 07:52:11 <markand> someone is able to help me fixing this upgrade issue? https://pastebin.com/D3sR5kbM
2020-06-02 07:54:40 <ikke> You have to figure out what is keeping a dependency on any of these "libffi-dev-3.2.1-r6[libffi=3.2.1-r6] firefox-esr-68.8.0-r0[so:libffi.so.6] py3-gobject3-3.34.0-r1[so:libffi.so.6] python3-3.8.2-r0[so:libffi.so.6] python2-2.7.18-r0[so:libffi.so.6] mozjs60-60.9.0-r0[so:libffi.so.6] glib-2.62.6-r0[so:libffi.so.6] gobject-introspection-1.62.0-r1[so:libffi.so.6] llvm9-libs-9.0.0-r1[so:libffi.so.6] wayland-libs-client-1.17.0-r0[so:libffi.so.6] p11-kit-
2020-06-02 07:54:42 <ikke> wayland-libs-server-1.17.0-r0[so:libffi.so.6]"
2020-06-02 07:56:17 <ikke> There is an issue about it
2020-06-02 07:56:23 <ikke> https://gitlab.alpinelinux.org/alpine/aports/-/issues/11605
2020-06-02 07:57:29 <ikke> markand: first: can you try apk ugrade --available?
2020-06-02 07:58:00 <doggone> Is there any way to find out which program is causing infrequent kernel panics?
2020-06-02 07:58:09 <markand> same issue with --available ikke
2020-06-02 07:58:32 <doggone> (or whether it's hardware)
2020-06-02 07:59:24 <rodrigue7973> i aim problèm https://termbin.com/n4ztt
2020-06-02 08:59:18 <hbug___> is there en equivalent to parametrized services like systemd's "[..]@.service" in  openrc? can't find anything. but before i make a crappy solution with env variables i better ask?
2020-06-02 08:59:37 <ikke> hbug___: symlinks
2020-06-02 09:00:04 <ikke> ln -s foo.bar /etc/init.d/foo
2020-06-02 09:01:06 <hbug___> ikke: thank i thought about that, but how do i access the ".bar" part inside the service file?
2020-06-02 09:01:14 <ikke> with $RC_SVCNAME you get the complete service name
2020-06-02 09:01:28 <hbug___> do i have to extract it manyally from args?
2020-06-02 09:02:02 <hbug___> ikke: okey thanks :) RC_SVCNAME was the missing puzzle part.
2020-06-02 09:03:40 <hbug___> well thats even more flexible as the parametrized @.service stuff, as i can parse it as i want. nice
2020-06-02 09:04:06 <ikke> RC_INSTANCE=${RC_SVCNAME#*.}
2020-06-02 10:43:05 <rnkn> I just upgraded from 3.12_alpha (edge) to 3.12 and now my boot hangs with "Waiting for uevents to be processed ..." then errors with "ERROR: udev-settle-failed to start"; any ideas what's happened?
2020-06-02 10:44:00 <ikke> Anything in the logs?
2020-06-02 10:45:53 <rnkn> hmm which logs should I check?
2020-06-02 10:46:09 <ikke> /var/log/messages probably
2020-06-02 10:47:40 <rnkn> just the same error "/etc/init.d/udev-settle[2560]: ERROR: udev-settle failed to start"
2020-06-02 10:48:56 <rnkn> I see this is running the command `udevadm settle`
2020-06-02 10:49:17 <ikke> You could trying to run it manually
2020-06-02 10:51:22 <rnkn> ran manually; it's just hanging there
2020-06-02 10:52:42 <rnkn> it might have something to do with fuse... brb, rebooting
2020-06-02 10:57:40 <rnkn> negative on fuse... still get a 300 sec timeout with udev on boot :(
2020-06-02 11:04:58 <rnkn> also, shut down hangs trying to shut down udev
2020-06-02 11:05:08 <rnkn> udev, it would appear, is an ass
2020-06-02 11:08:11 <rnkn> I've run apk fix on all the udev apks, rebooting again.......
2020-06-02 11:17:50 <rnkn> aarrgghh same result after running apk fix on all [e]udev apks
2020-06-02 11:18:26 <rnkn> this 300 sec timeout also happens if I run `service networking restart`
2020-06-02 11:46:08 <ikke> markand: ping
2020-06-02 11:56:03 <velix> Damn... I've waited for the technican the whole morning. But he'll come TOMORROW, not today. I hate my life :(
2020-06-02 12:12:06 <maxice8> oof
2020-06-02 12:50:04 <markand> ikke, pong
2020-06-02 13:11:58 <markand> oh noes my raspberry does not boot anymore after 3.11 -> 3.12
2020-06-02 13:14:07 <markand> ikke, I've ran the commands listed on the issue, the upgrade only complains about libffi now
2020-06-02 13:17:12 <ikke> can you run the comm command?
2020-06-02 13:20:06 <markand> yes
2020-06-02 13:20:20 <markand> https://pastebin.com/mb5Ed515
2020-06-02 13:21:01 <ikke> mdocml should be mandoc
2020-06-02 13:21:59 <ikke> I'm not sure what those sdl entries are
2020-06-02 13:22:10 <ikke> and strange that vim is listed in there
2020-06-02 13:31:19 <markand> I think I have built sdl2 myself to incorporate some patches
2020-06-02 13:40:28 <thresh> hello
2020-06-02 13:40:39 <Cogitri> 👋
2020-06-02 13:40:42 <thresh> I'm trying to re-sign an apk package I've just built and I have an issue with abuild-gzplit
2020-06-02 13:40:55 <thresh> it just says "failed" when it's used to split the .apk
2020-06-02 13:41:14 <thresh> I've tried with a package I got from an upstream alpine packages repository and it's the same
2020-06-02 13:41:21 <thresh> that happens on 3.12
2020-06-02 13:43:14 <thresh> here's what happens: https://gist.github.com/thresheek/1c8b0d720d10b73ed187d92da2bab432
2020-06-02 13:43:38 <thresh> I can also reproduce that on non-docker, it's just much more convenient to show
2020-06-02 13:44:25 <thresh> did 3.11 -> 3.12 changed some apk formats?
2020-06-02 13:47:38 <Cogitri> Hm, I don't think so, the big APK format change is supposed to happen in 3.12->whetever is next
2020-06-02 13:47:58 <Cogitri> Maybe you can do some printf debugging to see when it goed to err https://gitlab.alpinelinux.org/alpine/abuild/-/blob/master/abuild-gzsplit.c#L69
2020-06-02 13:48:15 <thresh> that was my next plan :-) thanks
2020-06-02 14:02:13 <thresh> ok, someone beat me to it: https://gitlab.alpinelinux.org/alpine/abuild/-/issues/9999
2020-06-02 14:02:53 <thresh> AND got a nice issue number.
2020-06-02 14:03:24 <ikke> thresh: I was reading the code for abuild-gzsplit. You don't happen to have a data.tar.gz file in the current directory already, right?
2020-06-02 14:03:55 <thresh> ikke, nope
2020-06-02 14:11:26 <thresh> I have reverted https://git.alpinelinux.org/abuild/commit/?id=660f793d6de6291204ba044e03b37826d2e78e88 and https://git.alpinelinux.org/abuild/commit/?id=4f5b0a14bdd715a36cd276788bc79e0a67a7672e and abuild-gzsplit works now
2020-06-02 14:11:45 <thresh> reverting only https://git.alpinelinux.org/abuild/commit/?id=4f5b0a14bdd715a36cd276788bc79e0a67a7672e didnt help
2020-06-02 14:14:11 <ikke> aha, ok
2020-06-02 14:14:20 <ikke> Maybe worth mentioning that in the ticket
2020-06-02 14:14:26 <thresh> yeah
2020-06-02 14:14:28 <ikke> might be that abuild-gzsplit needs to be updated
2020-06-02 14:15:19 <thresh> indeed
2020-06-02 14:38:14 <peteco> Hi, hope you all having a good day!, anyone have any experience disabling a root ca certificate from an alpine image?
2020-06-02 14:38:58 <Cogitri> https://gitlab.alpinelinux.org/alpine/aports/-/issues/11607
2020-06-02 14:40:44 <peteco> Ok, ill remove my blindfold next time I visit the issues on alpine!
2020-06-02 14:44:41 <Cogitri> Heh, no worries, I just knew about it because someone mentioned it earlier
2020-06-02 14:45:36 <peteco> Sooo I understand that upgrading the image, if and when a new one is available is fine... but anyway to "fix" the current image I use?
2020-06-02 14:46:09 <peteco> I was hoping editing the cert files somehow will do it, but im not sure how
2020-06-02 14:55:53 <Cogitri> I guess you'd have to do the same as that debian commit that's mentioned in the issue
2020-06-02 14:56:07 <Cogitri> Namely add the cert to the blacklist and rebuild ca-certificates
2020-06-02 15:04:50 <peteco> Yeah but where is that blacklist I cannot seem to understand how that is done here
2020-06-02 16:11:09 <Celelibi> Could I propose to add some python packages to alpine? What are the conditions? Does it need to be popular or required by a popular package? Would I have to maintain it?
2020-06-02 16:14:36 <Cogitri> Celelibi: You can make package requests at https://gitlab.alpinelinux.org/alpine/aports/issues , but someone has to do the work to make the package and maintain it
2020-06-02 16:15:23 <Celelibi> I have the APKBUILD already.
2020-06-02 16:15:38 <Celelibi> Which are usually pretty straightforward for python packages.
2020-06-02 16:15:54 <Cogitri> And people usually aren't that enthusiastic about maintaining things they don't use, so your chances of getting the package into Alpine at an acceptable timeframe are somewhat low if you don't make a merge request yourself and maintain the APKBUILDs
2020-06-02 16:16:00 <Cogitri> Nice :)
2020-06-02 16:18:38 <Celelibi> The thing is. I don't know for how long I'll be maintaining that Dockerfile (building those apk for the dockerized application).
2020-06-02 16:25:44 <Hello71> run abuild yourself then?
2020-06-02 16:26:34 <maxice8> Note that by python it should be python3
2020-06-02 16:30:26 <markand> this is my current conflict issue (remaining libffi) : https://pastebin.com/VYSKsc52
2020-06-02 16:30:50 <markand> what I don't understand is why some packages wants to pull the old 3.2.1-r6 version
2020-06-02 16:32:27 <Celelibi> Hello71, that's what I do.
2020-06-02 16:32:45 <vejetaryenvampir> (Not trying to start a distro comparison war.) Some one on #gentoo says "it sacrifices a lot of things for compactness or security reasons". What do the devs say?
2020-06-02 16:33:05 <Cogitri> It being Alpine?
2020-06-02 16:33:26 <Cogitri> Well, we do sacrifice some things to be more minimal, like no PAM support by default
2020-06-02 16:34:03 <maxice8> Kinda fair i guess if the things I'm thinking when it says sacrifices what the person that said it means
2020-06-02 16:34:19 <maxice8> i am not worried, back to eating mandarin oranges
2020-06-02 16:34:33 <Cogitri> markand: The soname was changed between 3.3.x and 3.2.x in libffi. Now apk can't upgrade some package that depends on the old libffi. It can't figure out a way to solve that
2020-06-02 16:34:45 <golb> Anybody uses Alpine for KVM host? I plan to run a few FreeBSD and Windows guests with this.
2020-06-02 16:34:51 <Hello71> #gentoo sucks most of the time, but it's half-right here
2020-06-02 16:35:12 <Hello71> "security reasons" sounds like they heard it from someone who heard it from someone else
2020-06-02 16:35:26 <Cogitri> markand: It can't upgrade to new versions without breaking the dependency graph, because it can't pull in the new libffi and old libffi at the same time
2020-06-02 16:35:35 <Cogitri> So everything has to be upgraded in one go basically
2020-06-02 16:35:41 <markand> a
2020-06-02 16:35:42 <markand> h
2020-06-02 16:35:42 <vejetaryenvampir> I'm not trying to discuss something or start a distro comparison war, just trying to get info. Please don't understand it wrongly.
2020-06-02 16:36:03 <Cogitri> golb: Some people do, what's your question?
2020-06-02 16:36:22 <Hello71> for example, alpine doesn't have openssl 1.0. that arguably improves security, but that's not really why it was removed. more that nothing in the distro uses it, and nobody can be bothered to maintain it
2020-06-02 16:36:27 <vejetaryenvampir> Cogitri: Can I learn the sacrifices, please?
2020-06-02 16:36:34 <markand> btw, PAM isn't a security tool, it's more like a flexible authentication mechanism and even worse, if badly configured you can really create security holes
2020-06-02 16:36:54 <Hello71> so as with Arch, minimalism including minimal patching tends to lead to a more secure system, not the reverse
2020-06-02 16:37:09 <Cogitri> Minimal patching in Alpine?
2020-06-02 16:37:11 <Hello71> vejetaryenvampir: not much point in listing it
2020-06-02 16:37:15 <Cogitri> Sorry, but that's not really true
2020-06-02 16:37:26 <Hello71> Cogitri: well compared to debian/fedora
2020-06-02 16:37:32 <vejetaryenvampir> Hello71: I didn't understand, what did you mean?
2020-06-02 16:37:35 <Hello71> or gentoo
2020-06-02 16:37:36 <golb> Cogitri: I want to know if there's any issue at all, especially stability.
2020-06-02 16:37:37 <Cogitri> I mean we don't patch functionality, but we have *so* many downstream patches for musl, it's kind of horrifying
2020-06-02 16:37:42 <markand> fedora is total bloatware, not comparable to Alpine though
2020-06-02 16:37:59 <Cogitri> gold: I don't know of any, I guess you'll have to try
2020-06-02 16:38:11 <Cogitri> Actually, I think the libvirt initd service is a bit broke right now
2020-06-02 16:38:15 <markand> Cogitri, so any idea how I can fix it ? :D
2020-06-02 16:38:39 <Cogitri> markand: Uhh...I only know the "uninstall and reinstall at full force" way :D
2020-06-02 16:39:00 <Hello71> Cogitri: well that's still minimal though... minimal doesn't mean none, it means the minimum to have it working
2020-06-02 16:39:21 <ikke> markand: can you run so:libffi.so.6 ?
2020-06-02 16:39:25 <Cogitri> Still feels wrong to call 3.7k patches minimal patching though
2020-06-02 16:39:26 <ikke> markand: can you run apk info -r so:libffi.so.6 ?
2020-06-02 16:39:46 <ikke> Yeah, using musl accounts for a lot of patches
2020-06-02 16:40:07 <vejetaryenvampir> So, the security issues that the guy was talking about was no PAM support by default?
2020-06-02 16:40:22 <ikke> vejetaryenvampir: We don't really know what they meant
2020-06-02 16:41:00 <markand> ikke, https://pastebin.com/BEbBXpmw
2020-06-02 16:41:00 <Cogitri> I wouldn't say no PAM is a security issue, but it certainly limits functionality around login
2020-06-02 16:41:16 <Cogitri> Actually, now that I think about it, I think that's what's breaking gnome-keyring unlocking! :D
2020-06-02 16:41:46 <golb> Will definitely try it. Anyone has installed Alpine on either OVH or SYS?
2020-06-02 16:41:57 <markand> PAM is a great thing to centerize the authentication, so sensible tools don't need to reinvent the wheel
2020-06-02 16:42:56 <markand> for example, authenticating ssh, pw related tools, login, display managers, etc
2020-06-02 16:43:27 <vejetaryenvampir> ikke: I see,thank you. ^^
2020-06-02 16:44:33 <ikke> markand: trying to figure out what keeps pulling libffi.so.6
2020-06-02 16:45:18 <markand> okay
2020-06-02 16:50:02 <ikke> ncopa: any idea how to find out what holds libffi back for markand?
2020-06-02 16:51:53 <ncopa> markand: maybe `apk del so:libffi.so.6` gives a hint whats holding it back?
2020-06-02 16:52:07 <markand> a big list
2020-06-02 16:52:53 <ikke> What about `apk policy firefox-esr py3-gobject3 python3 python2 mozjs60 glib gobject-introspection llvm9-libs wayland-libs-client p11-kit wayland-libs-server`
2020-06-02 16:53:42 <ncopa> or apk version -l '?'
2020-06-02 16:55:29 <ikke> Is LIMCHARS somewhere explained?
2020-06-02 16:56:20 <ncopa> its either <>= or ?
2020-06-02 16:56:35 <ikke> Ok, and what does ? mean
2020-06-02 16:56:37 <ncopa> apk version -l '<' will list only the < packages
2020-06-02 16:56:40 <ncopa> unknown
2020-06-02 16:56:45 <ncopa> nonexistsing
2020-06-02 16:56:47 <Hello71> heh
2020-06-02 16:56:47 <ncopa> in repo
2020-06-02 16:57:07 <ncopa> it happens you apk add something and later the package disapears from repo
2020-06-02 16:57:20 <ncopa> then will it not be updated
2020-06-02 16:57:58 <ikke> heh, and here I come with comm -23 <(/etc/apk/world) <(apk search -q | sort -u) :D
2020-06-02 17:03:10 <mps> perfection is not when you don't have anything to add but when you don't need anything to remove - attributed to Leonardo da Vinchi
2020-06-02 17:03:57 <mps> so yes, minimal working system is usually more secure than bloated and easy to use ones
2020-06-02 17:07:49 <lunaphyte> that would be antoine de saint-exupery, not leonardo da vinci :)
2020-06-02 17:11:05 <mps> well, I wrote 'attributed', but that saying is known for more than 2000 years in something different words
2020-06-02 17:11:38 <mps> and is attributed to many famous masters
2020-06-02 17:11:39 <lunaphyte> yeah.  i've never seen that quote attributed to leonardo da vinci
2020-06-02 17:12:13 <lunaphyte> or anyone other than its author, i guess
2020-06-02 17:12:26 <ikke> "I would have written you a shorter letter, but I did not have the time"
2020-06-02 17:12:43 <lunaphyte> nice :)
2020-06-02 17:12:53 <mps> people rephrased it differently
2020-06-02 17:13:21 <lunaphyte> their phrasing would be attributed to them
2020-06-02 17:13:32 <lunaphyte> that's true
2020-06-02 17:13:37 <mps> ikke: this letter is too long
2020-06-02 17:13:46 <lunaphyte> what was the leonardo da vinci phrasing?
2020-06-02 17:14:34 <mps> u-boot serial input doesn't work on rpi zero w - mps da Vinchi
2020-06-02 17:15:39 <mps> trying one old from debian - mps ... :)
2020-06-02 17:17:52 <mps> hmm, maybe I know why rpi doesn't boot fully with linux-edge, probably need squashfs for initial boot
2020-06-02 17:18:07 <mps> RPizw
2020-06-02 17:45:35 <markand> ikke, ncopa, https://pastebin.com/r4CW0N61
2020-06-02 17:46:59 <markand> and the apk version -l '?' https://pastebin.com/3Y2quLWW
2020-06-02 17:50:10 <ikke> An eazy way out would probably cp /etc/apk/world /etc/apk/world.bak; echo alpine-base >/etc/apk/world; apk upgrade -a; cp /etc/apk/world.bak /etc/apk/world; apk fix
2020-06-02 17:51:05 <velix> Why do many LGPL licensed libs don't have the LGPL license text packaged?
2020-06-02 17:51:54 <Cogitri> Make sure to not reboot or open/close things while doing those commands though :)
2020-06-02 17:52:46 <Cogitri> velix: We don't package the license text for things which follow the SPDX spec
2020-06-02 17:53:05 <Cogitri> https://spdx.org/licenses/ has all the licenses and the text
2020-06-02 17:53:23 <velix> So SPDX overrides LGPLs wish for bundle the license?
2020-06-02 17:53:29 <velix> to bundle*
2020-06-02 17:56:41 <velix> Sorry for asking such questions.
2020-06-02 17:57:30 <Cogitri> Ah, no worries
2020-06-02 17:57:53 <Cogitri> Not sure to be honest, but apparently we assume that that's fine
2020-06-02 18:00:53 <velix> Actually, I'm having a greater problem. I've patched a shebang ... this triggers me to create a changelog for my edit.
2020-06-02 18:01:13 <velix> I've asked in #gnu if there is a rule for trivial edits.
2020-06-02 18:01:25 <velix> This edit is pretty trivial ;)
2020-06-02 18:01:29 <velix> bash - ba = sh
2020-06-02 18:02:19 <velix> I don't know if a single line or a patch file is enough documentation in terms of LGPL.
2020-06-02 18:03:08 <velix> That's easier in Debian. License texts and Changelogs are a must ;)
2020-06-02 18:03:31 <Cogitri> And loads of effort :D
2020-06-02 18:03:31 <velix> must have*
2020-06-02 18:03:52 <velix> Perhaps a pull-request on upstream is easier :)
2020-06-02 18:04:20 <Cogitri> Certainly, upstreaming is a good idea either way (if it's not Alpine specific)
2020-06-02 18:07:26 <Cogitri> Anyway, as long as the patch is opensource (which it is in aports),it should be fine (although it'd technically have to be licensed under a GPL compatible license AFAIU, but I doubt anyone expects that from distros)
2020-06-02 18:09:19 <opal> getting a lot of "bad signature" for packages lately
2020-06-02 18:09:35 <opal> using http://dl-cdn.alpinelinux.org/alpine/latest-stable/main and community
2020-06-02 18:10:04 <opal> can only do partial upgrade of server because of this
2020-06-02 18:11:05 <opal> currently the packages giving bad signatures are 'busybox-initscripts libltdl libmnl libnfnetlink libxdmcp', dunno if i can reproduce across systems or not
2020-06-02 18:11:59 <opal> i'll try another mirror
2020-06-02 18:12:47 <opal> eh, apk supports rsync? nice
2020-06-02 18:14:35 <opal> oh guessing not, rsync mirrors are listed on https://mirrors.alpinelinux.org/ though
2020-06-02 18:14:42 <opal> anyway yeah changing mirror fixed
2020-06-02 18:22:40 <c705> I think it should support rsync
2020-06-02 18:23:10 <opal> interesting. (and yes i did install rsync before trying)
2020-06-02 21:50:59 <hiro> " Note: We can't use the live environment from the installation isos, because we'll install the ZFS kernel module and that isn't possible with a read-only /boot (as provided by the iso)."
2020-06-02 21:51:03 <hiro> is that actually still true?
2020-06-02 21:51:41 <hiro> i remember seeing sometimes mounting a modloop, making me think above is not valid
2020-06-02 21:52:10 <hiro> the quote is from https://wiki.alpinelinux.org/wiki/Setting_up_ZFS_on_LUKS#Requirements
2020-06-02 21:55:19 <hiro> also related: this page https://wiki.alpinelinux.org/wiki/Alpine_Linux_with_root_on_ZFS_with_native_encryption#Requirements, which says "as of writing this it's easiest to use Ubuntu 19.10, which comes with ZFS pre-installed." -> can't beleive this isn't possible in another way without demanding a third OS ;)
2020-06-02 22:04:09 <gmalsack> hello all - fresh install doing setup-boot. system running diskless on /dev/sda1. it's storing my configs, however packages are gone when I reboot
2020-06-02 22:14:54 <gmalsack> thus every time it reboots I have to reinstall all of my apks
2020-06-02 22:14:57 <gmalsack> :-(
2020-06-02 22:15:55 <mps> gmalsack: do you do 'lbu commit' after install
2020-06-02 22:16:02 <gmalsack> yes
2020-06-02 22:17:35 <mps> hmm, no idea then
2020-06-03 00:38:01 <lunaphyte> i'm trying to understand what this section does: https://gitlab.alpinelinux.org/alpine/alpine-conf/-/blob/master/setup-disk.in#L412
2020-06-03 00:39:26 <lunaphyte> in the case when $APKOVL is undefined/empty, that is
2020-06-03 00:40:08 <lunaphyte> it seems to be saying that lbu package is run and then piped to tar which decompresses it and then writes the output to a file?
2020-06-03 00:40:11 <lunaphyte> but why?
2020-06-03 00:41:51 <lunaphyte> and i don't get the significance of having tar change to the $mnt dir
2020-06-03 01:56:21 <Hello71> the alpine scripts operate on the assumption that you are running from a ramdisk
2020-06-03 01:56:59 <Hello71> the way that setup-alpine works is by first configuring the current system to look like the target system, then copying the result to the target disk
2020-06-03 01:57:20 <Hello71> it's definitely a different way of thinking compared to most other distros
2020-06-03 01:57:28 <Hello71> lunaphyte:
2020-06-03 02:00:56 <bluezinc> Hello71: that's actually a cool way of doing it.
2020-06-03 02:29:42 <lunaphyte> Hello71: thanks, yeah i get all that
2020-06-03 02:30:02 <lunaphyte> it's just not clear to me what that particular portion does
2020-06-03 02:31:45 <lunaphyte> e.g. what is the point of creating a file that has a list of the file names that are in an overlay?
2020-06-03 02:34:07 <lunaphyte> i see that it's used later with apk add [--overlay-from-stdin - "Read list of overlay files from stdin"]
2020-06-03 02:34:14 <lunaphyte> but what for?  what does it do?
2020-06-03 02:41:11 <Hello71> I suspect it tells apk not to overwrite those files
2020-06-03 02:41:24 <Hello71> or create apk-new files
2020-06-03 03:09:51 <yunfan> mps: hi, i just saw the news about v3.12, is this bring the supporting for those allwinner soc ?
2020-06-03 03:37:24 <lunaphyte> Hello71: ah, i see.  thanks :)
2020-06-03 06:09:43 <mps> yunfan: it is better than previous release, but I'm not sure if everything work
2020-06-03 06:10:46 <mps> I just tested that install works and drivers for ethernet, usb, reboot/poweroff now works
2020-06-03 06:12:32 <mps> and there is now testing/linux-edge kernel which you can install, and it works really fine on my two A20 boards
2020-06-03 06:13:40 <mps> and you reminded me to remove it from my site for download :)
2020-06-03 06:16:19 <mps> yunfan: and I updated my 'flash-bananapi.sh' script to not kernel panic after boot
2020-06-03 06:18:29 <mps> missing part is the skip syslinux depends in setup-alpine to have fully 'out-of-the-box' installable alpine on these board
2020-06-03 06:18:34 <yunfan> mps: well, reboot/poweroff is what i am waiting
2020-06-03 06:19:04 <yunfan> mps: unfortunately i am live in my wife's home recently, is it possible to remote upgrade my alpine on the home via ssh?
2020-06-03 06:19:13 <mps> I think it works now with linux-lts, but sure it is better with linux-edge
2020-06-03 06:20:17 <mps> if you have previous kernel then I think reboot will not work
2020-06-03 06:21:00 <yunfan> ok, will find a chance to try that on another board
2020-06-03 06:22:18 <mps> you need someone 'on site' who can physically force board to poweroff and then poweron, or reset, by forcing button
2020-06-03 06:22:40 <mps> buttons*
2020-06-03 06:24:22 <mps> but at the end, I use latest alpine on three A20 boards and it works quite good
2020-06-03 06:30:33 <PhoenixMage> Is it possible to list a repo for pulling a single package in APKBUILD? ie, the package I need to use to build is only available in edge but I want all my other packages to come from stable
2020-06-03 06:30:51 <nacelle> you can get a box like this: https://www.apc.com/shop/us/en/products/Rack-PDU-Switched-1U-15A-100-120V-8-5-15/P-AP7900B  (they make cheaper ones, used ones, etc.) if you can get by with just power off/on and no remote human
2020-06-03 06:31:52 <nacelle> (handy for testing out reboot code, heh)
2020-06-03 06:31:59 <Cogitri> PhoenixMage: You can use repository tags for that, but keep on mind that using edge packages in a stable release is unsupported and might break at any time
2020-06-03 06:33:32 <PhoenixMage> These are just packages for me, unless anyone wants istio on alpine ;)
2020-06-03 06:34:10 <mps> nacelle: or you can build simple and cheap ones from some of arm boards
2020-06-03 06:35:57 <mps> my son made one with arduino nano and gsm module, and can do a different actions by sending SMS messages
2020-06-03 06:36:42 <PhoenixMage> Cogitri: Do you know if there is a variable or something I can edit to add the tagged repo or will I need to edit /etc/apk/repositories?
2020-06-03 06:39:23 <mps> PhoenixMage: @testing http://dl-cdn.alpinelinux.org/alpine/edge/testing
2020-06-03 06:39:35 <Cogitri> https://wiki.alpinelinux.org/wiki/Alpine_Linux_package_management#Repository_pinning
2020-06-03 06:40:38 <PhoenixMage> Yeah I know how to do it globally, I was just wondering if I can have it only for that APKBUILD
2020-06-03 06:47:32 <PhoenixMage> Actually doesnt seem to work at all "ERROR: 'bazel@edgetesting' is not a valid child dependency, format is name([<>~=]version)"
2020-06-03 06:53:52 <markand> don't know what happened on my raspberry pi, after upgrading, for some reason sudo package did not contain anything anymore Oo
2020-06-03 06:54:40 <markand> ikke, the apk fix did the trick with a new world :)
2020-06-03 06:57:59 <markand> ah did the same on my main machine
2020-06-03 07:22:29 <hiro> so nobody has an idea about my modloop question one screen up?
2020-06-03 07:30:58 <Cogitri> What question?
2020-06-03 08:37:11 <hiro> 23:50  hiro " Note: We can't use the live environment from the installation isos, because we'll install the ZFS kernel module and that isn't possible with a read-only /boot (as provided by the iso)."
2020-06-03 08:37:13 <hiro> 23:51  hiro is that actually still true?
2020-06-03 08:37:18 <hiro> 23:52  hiro the quote is from https://wiki.alpinelinux.org/wiki/Setting_up_ZFS_on_LUKS#Requirements
2020-06-03 08:37:29 <hiro> Cogitri: ^ :)
2020-06-03 08:38:05 <hiro> kinda just before sleepytime in europe, perhaps i shouldn't expect an answer hehe
2020-06-03 08:49:50 <Cogitri> Hm, I always use the Ubuntu ISO for setting things up
2020-06-03 08:50:17 <Cogitri> (And FWIW it might be a good idea to use ZFS' native encryption instead of LUKS w/ ZFS)
2020-06-03 08:50:54 <hiro> i left away quoting that part above cause it was kinda the same argument ;)
2020-06-03 08:51:17 <hiro> Cogitri: i see how if you have an ubuntu iso already anyway it shouldn't hurt to use this if you're manually bootstrapping something
2020-06-03 08:52:00 <hiro> but i was wondering if a short script for alpine wouldn't work just as well... why avoid alpine so hard?
2020-06-03 08:52:25 <Cogitri> Well, Ubuntu just works as install medium, so I didn't have a reason to use something else yet
2020-06-03 08:52:34 <Cogitri> And it's nice that it comes with ZFS by default for installing things
2020-06-03 08:53:19 <hiro> as long as you remount the modloop isn't zfs just one apk add away on alpine?
2020-06-03 08:53:25 <hiro> even on the live system?
2020-06-03 08:54:21 <hiro> the reason i'm interested in this is bec. 1) i don't want to wait for the big ubuntu to load 2) automate all of this
2020-06-03 09:01:22 <Cogitri> Probably, I only use Ubuntu to install things every other year, so it doesn't bottleneck me :)
2020-06-03 09:02:03 <Cogitri> But yes, it should be one apk add and modprobe zfs away I think
2020-06-03 09:36:49 <hiro> i like to keep very lightweight live systems around, bec. less time is spent copying things into ram :)
2020-06-03 09:37:22 <hiro> most of the time i specifically boot alpine or tinycore bec. i don't want to wait for the (already installed) ubuntu to boot
2020-06-03 09:37:44 <hiro> most of those distros just take forever to get me to a usable terminal
2020-06-03 10:12:26 <ncopa> hiro: i thikn you can use the alpine-extended iso. The zfs module should be included there
2020-06-03 10:19:18 <hiro> ncopa: thanks, will try that
2020-06-03 10:20:23 <hiro> Cogitri: hehe i forgot to mention: i don't even plan to use any encryption, it's just that there was no wiki entry for plain zfs without that.
2020-06-03 10:21:07 <hiro> Cogitri: in any case though, what is the reason you recommend the native encryption, is it more robust or something?
2020-06-03 10:21:25 <hiro> (i've been using ext on dm-crypt for a long time now)
2020-06-03 10:23:32 <ncopa> opal: the bad signatures in latest-stable repo is due to the v3.12 upgrade. the varnish cache holds the v3.11 binary for some of the urls
2020-06-03 10:24:03 <ncopa> i have ran script to purge the cache o the affected packages, but it depends on which cache you use
2020-06-03 10:24:20 <ncopa> i have purged from norway, usa and brazil at least
2020-06-03 10:24:30 <Cogitri> hiro: For ZFS native encryption is way easier to setup, faster and I think it generally plays in better with how ZFS works
2020-06-03 10:24:42 <Cogitri> Since ZFS is meant to manage the entire disk directly
2020-06-03 10:25:09 <opal> ncopa gotcha
2020-06-03 10:43:35 <hiro> Cogitri: yeah, i could imagine subtle differences in block sizes make a transparent layering more work,y eah
2020-06-03 10:44:38 <hiro> Cogitri: also now i realize you'd have to do this once per disk
2020-06-03 10:44:55 <Cogitri> Also not sure how compression interacts with LUKS vs native encryption
2020-06-03 10:45:58 <hiro> Cogitri: well you'd compress first, in zfs, so i don't see an issue there
2020-06-03 10:46:32 <hiro> *that* at least is necessarily transparent
2020-06-03 10:57:18 <Cogitri> Ah, I was just curious if the compression ratio would be the same
2020-06-03 10:59:50 <ncopa> opal: i dont know which cache you use (based on geographic location), but you can help me purge by running https://gitlab.alpinelinux.org/ncopa/purge-dlcdn-cache from your location
2020-06-03 11:03:22 <opal> ncopa you might want to modify it for wget seeing as that comes with alpine
2020-06-03 11:03:49 <opal> anyhow, running it
2020-06-03 11:05:09 <opal> k it's completed, let me try upgrade from one of my other containers which is still using the dl-cdn mirror
2020-06-03 11:07:33 <Cogitri> I think that script only lists the pkgs that have to be purged, it doesn't actually do any purging
2020-06-03 11:08:14 <opal> oh
2020-06-03 11:09:03 <opal> should do cache invalidation on server side anyhow
2020-06-03 11:10:43 <opal> so, 3.12 is out now, cool
2020-06-03 11:12:58 <ncopa> oh...
2020-06-03 11:13:00 <ncopa> lol
2020-06-03 11:13:18 <opal> hm?
2020-06-03 11:13:19 <ncopa> I thought the script does the purge but it actually doesnt :)
2020-06-03 11:13:23 <opal> aha
2020-06-03 11:13:49 <opal> should be trivial for me to modify
2020-06-03 11:13:53 <ncopa> it just list the packages that nees to be purged
2020-06-03 11:14:12 <opal> wait is a HEAD request enough to purge cache or not
2020-06-03 11:14:12 <ncopa> which means that I havent purged anythiong :)
2020-06-03 11:14:15 <ikke> curl -XPURGE dl-cdn.alpinelinux.org/alpine/latest-stable/<repo>/<package>
2020-06-03 11:14:21 <opal> ah PURGE ok
2020-06-03 11:14:25 <ikke> ncopa: I did a automatic purge as well
2020-06-03 11:14:31 <ikke> did at least solve the issue for one user
2020-06-03 11:14:43 <ikke> but that was yesterday
2020-06-03 11:15:10 <ncopa> sh dl-cdn-dupes | xargs curl -X PURGE
2020-06-03 11:16:35 <opal> running
2020-06-03 13:21:22 <clandmeter> haha
2020-06-03 13:21:39 <clandmeter> yes that was my first question about the script, why not run the purge.
2020-06-03 14:04:17 <rnkn> just done a fresh install of 3.12 and after changing LANG in /etc/profile.d/locale.sh to en_AU.UTF-8 now console (no X) is freakin out with unicode chars
2020-06-03 14:05:10 <rnkn> I've also changed the console font to terimus, but pretty sure terminus has the unicode char set...
2020-06-03 15:07:53 <rnkn> so silly of me before, I had not enabled unicode in /etc/rc.conf... sorry for the noise
2020-06-03 15:24:16 <Cogitri> No worries, glad you were able to solve it, rnkn
2020-06-03 15:26:50 <rnkn> is the requirement for abuild pub/private keys new to 3.12?
2020-06-03 15:28:08 <Cogitri> I don't think so, my old keys I created in 3.11 still work
2020-06-03 15:29:28 <rnkn> I can't remember creating keys before, or needing to add them to /etc/apk/keys
2020-06-03 15:29:51 <artok> and if the question is requirement of keys overall, they have been as long as I've had my hands on alpine
2020-06-03 15:33:50 <rnkn> would this change if I had added my user to the abuild group?
2020-06-03 15:37:39 <artok> you need yourself to be abuild group, able to sudo, and have those keys and git global settings done
2020-06-03 15:37:50 <artok> i think that was all =)
2020-06-03 15:38:28 <rnkn> ah, I think I did not set the keys but I just did sudo mv PKG /usr/local/bin
2020-06-03 15:55:27 <classsic> hi, I try to execute wget, and get this error: Error relocating wget: secure_getenv: symbol not found
2020-06-03 15:57:28 <maxice8> what version ?
2020-06-03 15:57:31 <maxice8> of Alpine*
2020-06-03 15:59:56 <afontain_> you may not have the latest version of musl
2020-06-03 16:00:47 <afontain_> apk upgrade -a may help here, if you don't have pinned packages (use -s/--simulate for a dry run)
2020-06-03 16:12:31 <classsic> now works, thanks!
2020-06-03 16:54:49 <wsinatra> Anyone here have any pointers setting up the run from ram apkvol stuff?
2020-06-03 16:55:16 <wsinatra> When I'm trying to setup a cache on /media/vda1 it throws BAD signature errors, and I'm not entirely certain why
2020-06-03 16:58:06 <wsinatra> ah wait no, I'm dumb, my partitioning looks broken
2020-06-03 18:01:35 <hiro> Cogitri: the thing is if you were to first encrypt and then compress the compression ratio would at best be 1
2020-06-03 18:02:00 <hiro> Cogitri: else your crypto would be clearly broken
2020-06-03 18:24:15 <Yatekii> hey folks
2020-06-03 18:24:29 <Yatekii> can anyone tell me how I get pip3 in newer docker images?
2020-06-03 18:24:38 <Yatekii> e.g. 3.12
2020-06-03 18:24:59 <artok> apk add py3-pip
2020-06-03 18:25:14 <Yatekii> why was it removed from python3-dev?
2020-06-03 18:27:23 <Yatekii> also wehn I install that and run pip3 install cairocffi that fails
2020-06-03 18:31:00 <artok> what does it say?
2020-06-03 18:35:37 <Yatekii> distutils.errors.DistutilsError: Command '['/usr/bin/python3', '-m', 'pip', '--disable-pip-version-check', 'wheel', '--no-deps', '-w', '/tmp/tmppuhug7l6', '--quiet', 'cffi>=1.1.0']' returned non-zero exit status 1.
2020-06-03 18:35:49 <Yatekii> the only thing I changed is the dopcker image version because :latest didn't build anymore
2020-06-03 18:36:02 <Cogitri> That's not a terribly useful output
2020-06-03 18:36:07 <Yatekii> I am aware
2020-06-03 18:36:11 <Cogitri> Can you please paste the entire output via tpaste or smth?
2020-06-03 18:36:47 <Yatekii> ohwait it warn ssth about the wheel package
2020-06-03 18:37:00 <Yatekii> https://gist.github.com/Yatekii/7a569dec94bbeefa552114cae89ca546
2020-06-03 18:37:27 <Yatekii> was that indeed changed with pip3? in newer versions? because in the old image it worked juuust fine
2020-06-03 18:38:16 <Cogitri> I don't know Python much, but some things need py3-wheel, yes
2020-06-03 18:39:36 <Yatekii> why was that changed? I don't mind, I just wonder about the reasoning because it makes everythign obscure and cumbersome to use if you don't know ...
2020-06-03 18:40:03 <Yatekii> thanks for the help, I managed to get it to work!
2020-06-03 18:46:50 <afontain_> One day I had that error, so I packaged py3-wheel. Next thing you know, it's in main
2020-06-03 18:47:28 <afontain_> So I'm not technically maintaining something in main that I know nothing about :O
2020-06-03 18:53:41 <Cogitri> Yatekii: You'll have to ask the python people about that, that's not on Alpine
2020-06-03 18:54:33 <Yatekii> Cogitri: it's the alpine package.
2020-06-03 19:02:54 <superduck> If I want a change to alpine-baselayout's profile, how do I go about submitting a pull request for that?
2020-06-03 19:06:06 <Cogitri> https://gitlab.alpinelinux.org/alpine/aports
2020-06-03 19:48:06 <superduck> @Cogitri thanks, is there a document that describes etiquette on submitting merge requests? I'm wanting to recommend a couple updates with configure flags, not sure if that's something that needs group discussion first on if the merge should be submitted.
2020-06-03 19:55:08 <Cogitri> There is no discussion without a merge request
2020-06-03 19:55:22 <Cogitri> The point of the merge request is so people can review the changes
2020-06-03 19:56:50 <mps> but opening issue with explanation is quite fine, or describe idea to alpine-devel mailing list
2020-06-03 19:58:37 <superduck> Is a merge requested preferred to opening an issue?
2020-06-03 19:58:46 <maxice8> yes
2020-06-03 20:58:14 <ikke> Not that we necesarily work that way, but a merge request can follow from an issue (gitlab even has a button to create an MR from an issue
2020-06-03 20:58:18 <helby> hello ladies, gentlemen and other
2020-06-03 20:58:35 <helby> I have few basic questions if you don't mind :)
2020-06-03 20:58:53 <helby> where I set up environment variables?
2020-06-03 20:59:11 <helby> using ash
2020-06-03 20:59:42 <ikke> .profile or /etc/profile.d/*.sh
2020-06-03 21:00:00 <ikke> ~/.profile that is
2020-06-03 21:01:43 <helby> ikke: thanks, is it key=value or set or export?
2020-06-03 21:03:09 <ikke> export, it's just a shell script that is evaluated
2020-06-03 21:03:20 <ikke> (sourced in shell terms)
2020-06-03 21:04:43 <helby> great, thanks, just need some basic setup that some things work :)
2020-06-03 21:05:08 <helby> is there alternative to read man pages *-doc other than installing man-db less ... ?
2020-06-03 21:05:34 <helby> or is it used in alpine - man-db and so on?
2020-06-03 21:06:30 <ikke> apk add docs
2020-06-03 21:06:49 <ikke> But that would still install man pages for most packages
2020-06-03 21:07:44 <helby> oh that automatic install all *doc for insalled packages I see
2020-06-03 21:08:44 <ikke> yes
2020-06-03 21:09:22 <helby> but how to read that page then?
2020-06-03 21:09:29 <helby> man packages does not work
2020-06-03 21:10:04 <mps> helby: apk add mandoc
2020-06-03 21:10:21 <mps> man-db is not as good as mandoc
2020-06-03 21:10:25 <helby> it installed autom with that
2020-06-03 21:10:59 <helby> have mandoc docs and all *doc for packages
2020-06-03 21:11:09 <helby> now how to read that *-doc ?
2020-06-03 21:11:54 <ikke> it would still be with man
2020-06-03 21:12:02 <ikke> in most cases
2020-06-03 21:12:04 <helby> oh need to create them somehow? for whatever 'No entry for ...
2020-06-03 21:12:40 <mps> if you are on older releases, then mdocml instead of mandoc
2020-06-03 21:12:41 <ikke> you can try apropos <name> to find man pages that match
2020-06-03 21:12:53 <helby> I am on edge
2020-06-03 21:12:59 <ikke> that requires mandoc-apropos though
2020-06-03 21:13:11 <helby> ah
2020-06-03 21:13:19 <ikke> you can also run apk info -L <package>-doc to find what is included
2020-06-03 21:14:03 <helby> yes, but how to read these man pages
2020-06-03 21:14:50 <helby> all installed but somehow not populated in some kind of database or whatever applies without man-db
2020-06-03 21:14:52 <mps> hmm, doesnt't 'man man' works
2020-06-03 21:15:14 <helby> oh wait, probably 'less' issue
2020-06-03 21:15:19 <helby> did not see error
2020-06-03 21:16:03 <helby> okay, it showed me less: unrecognized option: T
2020-06-03 21:16:19 <mps> hmm, do we still have man-db in repo
2020-06-03 21:16:31 <ikke> in community
2020-06-03 21:16:33 <helby> yes it's there and it install a lot of crap
2020-06-03 21:16:42 <mps> ah, I see
2020-06-03 21:17:13 <mps> but my 'apk search man-db' doesn't show it
2020-06-03 21:17:36 <mps> ahm, 3.11
2020-06-03 21:17:39 <ikke> aha
2020-06-03 21:17:50 <ikke> was about to ask if you have edge/community
2020-06-03 21:17:58 <mps> strange
2020-06-03 21:18:16 <mps> on edge it shows it
2020-06-03 21:18:24 <mps> but not on 3.11
2020-06-03 21:19:54 <mps> heh, this explains all bc987c0bdfc7110dd7baa7eee7f67188d99bd51c
2020-06-03 21:20:05 <mps> Thu Jan 9 08:55:11 2020 +0100
2020-06-03 21:20:24 <mps> but question is, why it is moved from testing
2020-06-03 21:21:13 <ikke> Is it an issue that it's moved from testing?
2020-06-03 21:22:10 <mps> meh, do we need it, question is
2020-06-03 21:22:22 <bleb> so i got my firefox chroot working
2020-06-03 21:24:23 <mps> bleb: I run devuan in lxc, can use any program from it on alpine host
2020-06-03 21:25:30 <mps> Xephir is quite fine for these things
2020-06-03 21:27:57 <mps> (and I'm somewhat disappointed how much upstreams thinks all unix/linux runs on glibc only)
2020-06-03 21:29:03 <mps> and when I started to use linux it was based on some version of bsd libc, iirc
2020-06-03 21:29:51 <bleb> i might look into lxc instead of just chroots
2020-06-03 21:30:25 <bleb> my current use is to have an alpine chroot on any number of different linux systems, so i can have a consistent firefox installation that is synced between computers
2020-06-03 21:30:50 <mps> does chroot allow sound to be played on host
2020-06-03 21:33:03 <bleb> yup
2020-06-03 21:33:44 <bleb> just worked for me, probably though the bind-mounted /dev i guess
2020-06-03 21:33:47 <mps> ah, then chroot is quite fine
2020-06-03 21:34:49 <mps> no need to complicate life with yet another 'container' system :)
2020-06-03 21:35:05 <opal> luckily the people i talked with about libc compliance bugs, they were pretty receptive
2020-06-03 21:39:58 <helby> how to get list of installed pkgs from 'testing' repo
2020-06-03 21:48:00 <Hello71> mps: my impression is that man-db is better in sense of gnu
2020-06-03 21:49:23 <helby> nvm, now digging in apk info and so on
2020-06-03 21:49:31 <mps> Hello71: I think so
2020-06-03 21:49:56 <Hello71> much more features, but poorly designed and bloated
2020-06-03 21:50:01 <helby> holy fuck, somehow at my surprise everything works without any trouble
2020-06-03 21:50:27 <mps> Hello71: agree
2020-06-03 21:50:48 <Hello71> I think man-db is not actually gnu though
2020-06-03 21:50:55 <Hello71> yeah, it is on nongnu
2020-06-03 21:51:46 <mps> my impression is that man-db is gnu, thanks for info
2020-06-03 22:45:01 <hodapp> This may or may not be Alpine-specific, but I'm running on a Raspberry Pi and have just installed the linux-rpi package. I see the 'watchdog' service in /etc/init.d, but when I try to start it, it complains that WATCHDOG_DEV is not set. I already tried setting it via a script in /etc/profile.d (which I verified is running)
2020-06-03 22:45:29 <hodapp> what is the 'right' way to set this environment variable?
2020-06-03 22:45:57 <ikke> /etc/conf.d/watchdog
2020-06-03 22:46:33 <hodapp> oh, thanks! that's easy... I hadn't even realized /etc/conf.d was a thing yet
2020-06-03 22:47:25 <ikke> /etc/profile.d is only run for login shells
2020-06-03 22:48:02 <hodapp> 
ACTION shakes fist at Stack Overflow

2020-06-03 22:48:37 <ikke> for settign environment variables for your user, that would work
2020-06-03 22:48:41 <c705> hodapp: nice to see you in here!
2020-06-03 22:48:44 <ikke> but not for services
2020-06-03 22:48:54 <hodapp> c705: uhhhhhhhhhh we've never met
2020-06-03 22:49:04 <c705> ;-)
2020-06-03 22:49:21 <ikke> lol
2020-06-03 22:50:35 <hodapp> ikke: sorry, my brain is fried after dealing with an NVidia Jetson and its NVidia JetPack AI Sandbox Installation Toolkit Manager Installation Fun-Pack of Ubuntu Shit Crammed Into 16 GB eMMC Hey Did You Want More Fucking CUDA Drivers L4T Whatever
2020-06-03 22:50:44 <hodapp> so Alpine is rather a dream to deal with in comparison
2020-06-03 23:23:19 <diazepan> :>
2020-06-04 00:01:41 <lunaphyte> if i'm doing an install manually, so to speak [e.g. instead of with setup-alpine etc], i think i want to run mkinitfs at some point during the process?
2020-06-04 00:02:49 <c705> probably yes
2020-06-04 00:03:44 <lunaphyte> ok, thanks.  i'm trying to dissect that mechanism.  what is it that generates the initramfs when the traditional installation process is used?
2020-06-04 00:05:04 <lunaphyte> i see the section that generates mkinitfs.conf in setup-disk, but so far i don't see anything that seems to actually generate the initramfs
2020-06-04 00:07:07 <c705> lunaphyte: to be honest, I was looking for the same thing but I gave up at some point
2020-06-04 00:11:13 <Hello71> probably related to installing linux-lts or mkinitfs
2020-06-04 00:11:16 <Hello71> or possibly both
2020-06-04 00:11:26 <Hello71> mkinitfs needs to be called after upgrading linux-lts already
2020-06-04 00:14:28 <c705> Hello71: but where does mkinitfs get called when running setup-disk?
2020-06-04 00:14:40 <Hello71> apk add calls the hook
2020-06-04 00:14:44 <Hello71> ... I assume
2020-06-04 00:15:32 <lunaphyte> i'm probably looking in the wrong place, but i don't see install/upgrade hooks or triggers that run it where i've looked so far
2020-06-04 00:15:36 <c705> ah. that would explain why it's not in the code anywhere
2020-06-04 00:15:37 <lunaphyte> https://git.alpinelinux.org/aports/tree/main/mkinitfs?h=master
2020-06-04 00:18:15 <c705> indeed, mkinitfs.trigger in aports/main/mkinitfs seems to be calling that
2020-06-04 00:18:34 <lunaphyte> oh
2020-06-04 00:18:37 <lunaphyte> did i miss?
2020-06-04 00:18:58 <c705> so when would *that* get called? on apk add mkinitfs?
2020-06-04 00:19:08 <lunaphyte> oh geez, right there in front of me :)
2020-06-04 00:20:24 <c705> 5     mkinitfs -o /boot/$initfs $abi_release
2020-06-04 00:20:27 <c705> yeah
2020-06-04 00:23:45 <Hello71> actually I don't think it's called when upgrading mkinitfs
2020-06-04 00:23:51 <Hello71> triggers="$pkgname.trigger=/usr/share/kernel/*"
2020-06-04 00:24:45 <lunaphyte> how does that work?
2020-06-04 00:24:57 <lunaphyte> it somehow knows if something in the directory changes?
2020-06-04 00:26:33 <lunaphyte> seems to be:
2020-06-04 00:26:34 <lunaphyte> "Listed here are directories that are watched; if a change happens to the directory, then the trigger script is run at the end of the apk add/delete"
2020-06-04 00:28:24 <lunaphyte> ah: https://wiki.alpinelinux.org/wiki/Release_Notes_for_Alpine_1.9.0_rc5
2020-06-04 00:28:31 <lunaphyte> "Initramfs images and kernel links are generated from mkinitfs trigger rather than from post-install/upgrade"
2020-06-04 00:29:51 <lunaphyte> https://wiki.alpinelinux.org/wiki/APKBUILD_Reference#triggers
2020-06-04 00:30:28 <lunaphyte> so i guess i'd need to run that in a chroot then
2020-06-04 00:30:48 <lunaphyte> since the install process is passing --root to apk
2020-06-04 00:35:49 <helby> is there something in base/kernel what takes care of cpu freq scalling or I have to install cpureqd?
2020-06-04 00:40:13 <lunaphyte> hmm, so the trigger script operates on $@ - but i can't see what shows parameters getting passed in
2020-06-04 00:50:02 <Hello71> apk runs triggers based on its own accounting of changes
2020-06-04 00:50:14 <Hello71> dunno what the api is, but presumably it'll be somewhere in apk-tools
2020-06-04 00:50:55 <lunaphyte> extrapolating, it seems to just pass in the list of directories specified in the trigger
2020-06-04 00:52:07 <helby> okay got it, only issue is I have powersave or powerfomance and 'ondemand' is not available andit should be for this proc
2020-06-04 00:55:48 <helby> from wiki if I have powersave, there is no switching, just min freq, however it is switching, while most of the time at minimum
2020-06-04 00:56:33 <helby> not sure if something like linux-firmware-intel can help or do something with it
2020-06-04 00:56:49 <helby> (I did not install any firmware)
2020-06-04 01:02:31 <Hello71> pstate only offers powersave and performance
2020-06-04 01:02:43 <Hello71> it is recommended in virtually all cases
2020-06-04 01:03:03 <helby> what is recommended?
2020-06-04 01:03:11 <Hello71> you should not determine the active frequency by polling cpuinfo, but instead use cpupower or powertop
2020-06-04 01:03:27 <helby> Yes, I don't want to fuckering myself with that
2020-06-04 01:03:28 <Hello71> also, you should configure microcode always, but it is mostly unrelated to cpufreq
2020-06-04 01:03:44 <helby> just if I need some package for it or
2020-06-04 01:05:17 <helby> but have powertop there
2020-06-04 01:06:11 <helby> of course no idea how it get there
2020-06-04 01:06:35 <helby> and what to do with it, bc. no manual
2020-06-04 01:08:36 <c705> helby: https://linux.die.net/man/8/powertop
2020-06-04 01:10:26 <helby> okay I go try to fuckering with cpupower
2020-06-04 01:21:10 <helby> useless cpupower :)
2020-06-04 01:21:30 <helby> will keep it at powersave, will see if system will respond slow
2020-06-04 01:22:16 <helby> 'ondemand' is somehow old just read
2020-06-04 01:27:20 <helby> somebody should wipe most of alpine wiki
2020-06-04 06:26:22 <MY-R> hey hey, I would like to confirm something, if anyone with last stable or edge can run web browser benchmark and try moving mouse in that time? https://web.basemark.com/ START and just wait to the 5/20 step (WARNING: it will burn your GPU!)
2020-06-04 06:29:01 <MY-R> in my case firefox frozen little bit (obvious) but worst is that mouse doing same, after check what got in "/var/log/Xorg.0.log" dunno about wayland
2020-06-04 06:33:37 <wener[m]> hi, the arch page is not very clear https://wiki.alpinelinux.org/wiki/Architecture like  https://wiki.debian.org/SupportedArchitectures
2020-06-04 06:34:12 <wener[m]> what is the minimal arm version for armhf ? is armv4t+?
2020-06-04 06:35:38 <Cogitri> I think that's armv6, but not sure
2020-06-04 06:37:31 <ikke> That was my idea as well, but only because that's what we target (rpi1 / zero)
2020-06-04 06:54:52 <wener[m]> <ikke "That was my idea as well, but on"> so the armhf is meant for rpi1 and zero ?
2020-06-04 06:55:50 <mps> alpine armhf is mostly for armv6
2020-06-04 06:56:24 <wener[m]> ok, thanks
2020-06-04 06:56:29 <mps> but in arm jungle I'm never sure
2020-06-04 06:58:57 <wener[m]> updated the wiki
2020-06-04 07:00:05 <wener[m]> If I setup the base image by using minirootfs, with chroot how can I setup the basic service ?
2020-06-04 07:00:37 <wener[m]> like bootmisc, sysctl etc. do I have to manully link them all ?
2020-06-04 07:02:12 <ikke> https://gitlab.alpinelinux.org/alpine/alpine-conf/-/blob/master/setup-alpine.in#L153
2020-06-04 07:12:37 <wener[m]>  thanks, but when I rc-update, I get `rc-update: failed to add service `networking' to runlevel `boot': No such file or directory`, checked `/etc/init.d/networking` and `/etc/runlevels/boot/` are exists
2020-06-04 07:12:59 <Cogitri> I think adding things to the boot runlevel from a chroot is bugged in openrc right now
2020-06-04 07:13:08 <Cogitri> Adding to another runlevel should work
2020-06-04 07:16:41 <wener[m]> add `rc-update --quiet add devfs sysinit` successed
2020-06-04 07:23:23 <wener[m]> also chroot apk add will not add `*-openrc` pkg
2020-06-04 07:34:39 <TBB> is that something that broke recently?
2020-06-04 07:41:44 <Cogitri> TBB; Dunno, that never worked for me
2020-06-04 07:42:03 <Cogitri> wener[m]: It should do that just fine, if openrc is installed in the root
2020-06-04 07:44:28 <TBB> I don't know... my installer didn't seem to have that specific problem, but then, I've last worked on it with 3.9 so it could be something I just haven't faced yet
2020-06-04 07:48:27 <TBB> there was that touch /run/openrc/softlevel thing tho
2020-06-04 07:48:42 <TBB> having that during setting up services under chroot, that is
2020-06-04 07:55:46 <wener[m]> manually linked, seems fine, can boot services, but how to boot into a shell ? after ` * Starting sshd ... [ ok ]` there is no shell
2020-06-04 08:29:34 <steinbach> Is Alpine compatible with DisplayLink devices? (USB Docking Stations)
2020-06-04 08:47:20 <Marquel> steinbach: the linux-kernel can do it, so alpine can do, it's just you need to find the necessary tools
2020-06-04 09:08:18 <mps> ' Wysocki said that anybody who sends patches to the kernel has an agenda — it's why they wrote the patch in the first place', https://lwn.net/Articles/821116/
2020-06-04 09:08:27 <mps> good point :)
2020-06-04 10:05:57 <T0MuX> Hi
2020-06-04 10:06:52 <T0MuX> I'm having some issues... but let start with one of those : my xterm. I'm trying to set the Terminus font trough .Xdefaults file
2020-06-04 10:07:11 <T0MuX> But I don't know the good syntax for it
2020-06-04 10:07:20 <T0MuX> Do someone has it ?
2020-06-04 10:09:20 <ikke> not for xterm
2020-06-04 10:09:22 <mps> not sure for xterm but here is for urxvt, urxvt*font:        -xos4-terminus-medium-*-*-*-12-*-*-*-*-*-*-*
2020-06-04 10:10:18 <T0MuX> What means "xos4" ?
2020-06-04 10:10:37 <mps> it is how it is seen by Xserver
2020-06-04 10:11:02 <AinNero> T0MuX: https://github.com/search?q=xterm.font&type=Code
2020-06-04 10:11:16 <mps> install xfontsel package and with it you can browse X fonts
2020-06-04 10:11:34 <AinNero> dotfiles on github usually are a good reference point to start from
2020-06-04 10:47:56 <wener[m]> alpine's rpi kernel is sooooo grate, everything is just working, 👍
2020-06-04 11:07:45 <T0MuX> thank you guys my font issue is solved :)
2020-06-04 11:52:09 <opal> wener[m] which rpi do you have
2020-06-04 11:58:48 <artok> for those who understand: https://termbin.com/cdnz
2020-06-04 11:58:55 <artok> I feel like a winner
2020-06-04 14:28:37 <rjoensen> Hello, I have got an issue booting up your installer isos for alpinelinux-standard on versions 3.10.x and 3.12.0, initramfs fails at unable to mount boot media. I have tried mounting it via supermico ipmi, spiderkvm, usbstick, usbcdrom, nothing. I edited the config to reflect usbdisk:vfat as well, but nothing works. Any ideas?
2020-06-04 14:40:15 <jdbeaud[m]> is there any equivalent to unattended-upgrade in alpine ?
2020-06-04 14:45:25 <opal> i just stuck a cron job in /etc/periodic
2020-06-04 14:45:36 <opal> i think thats the closest you will get
2020-06-04 14:50:41 <Cogitri> rjoensen: Hum, have you verified the integrity of the ISO?
2020-06-04 14:55:21 <mps> jdbeaud[m]:  apk-cron
2020-06-04 14:56:21 <opal> oh wait theres a package for that?
2020-06-04 14:56:22 <opal> nice
2020-06-04 14:57:11 <mps> yes
2020-06-04 14:57:43 <opal> ah its literally just a cron script
2020-06-04 14:57:50 <opal> fancy
2020-06-04 15:29:58 <T0MuX> little question : why my consolefont is correcly loaded immediatly after booting trough syslinux, then it is lost after "Configuring kernel parameters ..." ??
2020-06-04 15:31:23 <T0MuX> I configured consolefont with "rc-update add consolefont boot"... but, I also had to add consolefont to default as workaround :/
2020-06-04 15:32:27 <T0MuX> however, /etc/conf.d/consolefont says itself "if you need to use it you should run rc-update add consolefont boot"
2020-06-04 16:19:35 <mps> anyone know which u-boot version works with RPi zero over serial console
2020-06-04 16:50:22 <T0MuX> I also notice this, in /etc/conf.d/consolefont : "Have a look in /usr/share/consoletrans for a selection of map files you can use" ... but, this file doesnt exist :D
2020-06-04 16:50:59 <Cogitri> I think we had a similar situation with /etc/conf.d/hostname
2020-06-04 16:51:18 <Cogitri> We've copied some openRC things from Gentoo but didn't account for the difference between Alpine and Gentoo in some places I think
2020-06-04 16:53:11 <MY-R> T0MuX, it exist after install package "kbd-misc"
2020-06-04 16:58:30 <T0MuX> yes ! good call ! thank you
2020-06-04 16:59:11 <MY-R> you have got also there extra console fonts
2020-06-04 17:03:26 <T0MuX> noice :)
2020-06-04 17:04:53 <T0MuX> by any chance, MY-R, do you know how I can do to avoid than networking service stuck the boot while udhcpd is looking for ip adress ?
2020-06-04 17:05:29 <T0MuX> I mean, my Alpine starts normally, but at the networking step, it hang until udhcpd get the ip adress. It's a little annoying
2020-06-04 17:07:03 <T0MuX> (I searched on the web, but without success 😕 )
2020-06-04 17:08:36 <MY-R> dunno, I never got problem with dhcp server in lan, it taking 2 seconds sure, you can always pass to udhcp lower timeouts or check other dhcp clients
2020-06-04 17:09:55 <T0MuX> or maybe it's the "way" that is configured ? I use the networking service, but maybe I can use another service (another way) to do the same thing, yeah like you say
2020-06-04 17:10:26 <T0MuX> at the beginning I was looking for a way yo make the service to fork in background
2020-06-04 17:10:37 <T0MuX> but don't know how to do this
2020-06-04 17:15:09 <helby> how to set aliases if my ash is non-login in sway?
2020-06-04 17:15:10 <T0MuX> so, I will keep searching
2020-06-04 17:15:38 <helby> 'export crap' is there as it's exporter from tty login, however alias ... nope
2020-06-04 17:35:01 <Marquel> morning.
2020-06-04 17:35:14 <helby> oho slowly evening here
2020-06-04 17:35:58 <Marquel> i'm unsure whether this is the right channel to ask, but since it is less software and more distribution-specific... anyway: what would i need to enable libvirtd/qemu on alpine to authenticate users using GSSAPI/kerberos?
2020-06-04 17:36:35 <Marquel> helby: according to UGT it's always morning when starting a conversation and always night when the conversation ends, regardless of the time the conversation took.
2020-06-04 17:37:28 <Marquel> (i guess that's to take near-lightspeed travellers into account who must assume that normal people's clocks run considerably faster than their own...)
2020-06-04 17:39:01 <Cogitri> https://libvirt.org/auth.html#ACL_server_kerberos I guess?
2020-06-04 17:39:48 <Cogitri> I think we're building with SASL support, so it should (just) work?
2020-06-04 17:40:26 <helby> Marquel: don't ge me wrong ... while oficially it's evening here, I did not even start my day :)
2020-06-04 17:53:19 <wsinatra> Anyone here have experience getting a salt master running on Alpine?
2020-06-04 18:15:01 <Marquel> Cogitri: then i'll have to check if mit-krb is there to support it ;)
2020-06-04 18:16:45 <Marquel> Cogitri: as it seems neither libvirtd nor qemu have pulled cyrus-sasl or any kerberos software, thus i'm really unsure if that really works out-of-the-box, yet you are correct that's exactly where i'm coming from.
2020-06-04 18:19:34 <Marquel> Cogitri: correction: krb5-{libs,conf} is installed, just no sasl implementation it seems, so i am unsure.
2020-06-04 18:19:40 <mps> Marquel: if not pulled doesn't mean you can't do that 'by hand'
2020-06-04 18:20:32 <Marquel> mps: i am unfortunately new to binary distributions, i don't really know how they operate in regards to dependencies which might or might not be installed.
2020-06-04 18:21:06 <mps> alpine tries to escape from unnecessary whenever it can
2020-06-04 18:21:27 <Marquel> mps: and i'm grateful for that.
2020-06-04 18:21:38 <mps> s/unnecessary/unnecessary deps/
2020-06-04 18:22:06 <Marquel> means in turn i need to learn how to get functionality where there is no "required" pull for dependencies... ;)
2020-06-04 18:22:11 <mps> I got this experience from mutt sasl support :)
2020-06-04 18:23:05 <mps> well, mostly, alpine is not like debian and similar distros where everything (needed and unneeded) pulled automatically
2020-06-04 18:23:52 <mps> deps which are needed (well, mostly) are pulled by installing package
2020-06-04 18:23:56 <Marquel> there's a reason i didn't opt for any of those for running my hypervisor ;)
2020-06-04 18:24:10 <helby> is there any reason why gtk+2.0 is still dependency in firefox?
2020-06-04 18:24:24 <mps> 'needed' could be replaced by 'essential', maybe
2020-06-04 18:26:19 <Marquel> so, if a feature has been compiled into a certain binary, like sasl support in qemu, but is dynamically switched off if necessary binaries for said feature are not present, they are not in the "essentials" list, not pulled in automagically when installing the requiring package, but the feature _should_ become operational once the necessary packages have been installed manually?
2020-06-04 18:26:42 <Cogitri> Marquel: i think we do build libvirtd against cyrus-sasl
2020-06-04 18:27:37 <Marquel> Cogitri: just found libsasl installed. i'm not used to alpine's apk or package names...
2020-06-04 18:30:19 <mps> Cogitri: right
2020-06-04 18:32:01 <mps> apk search cyrus-sasl, and then 'apk add cyrus-sasl-WHAT_I_NEED'
2020-06-04 18:47:18 <Hello71_> helby: gtk2 .so
2020-06-04 18:47:37 <Hello71_> probably it should be split into firefox-gtk2 subpackage, it is used only for flash suppor
2020-06-04 19:56:47 <guesto> hello ikke
2020-06-04 19:56:54 <guesto> i need some help
2020-06-04 19:57:12 <guesto> https://wiki.alpinelinux.org/wiki/LVM_on_LUKS
2020-06-04 19:57:25 <guesto> I did the GPT Grub setup
2020-06-04 19:57:35 <guesto> it boots, ask me for the password
2020-06-04 19:57:45 <guesto> but then it says it can find a device
2020-06-04 19:57:49 <guesto> cant*
2020-06-04 19:59:08 <guesto> efibootmgr is in the stable repo and i can install it (which i did)
2020-06-04 19:59:25 <guesto> but in the wiki it says its only available in the edge repos
2020-06-04 20:02:39 <ikke> It's included since 3.11
2020-06-04 20:07:00 <guesto> right
2020-06-04 20:08:42 <guesto> ive mounted the encrypted drives again
2020-06-04 20:08:50 <guesto> im in the live iso
2020-06-04 20:11:10 <guesto> okay im just gonna ditch gpt grub
2020-06-04 20:11:21 <guesto> fuck this shit
2020-06-04 20:36:17 <guesto> msdos works
2020-06-04 20:39:40 <artok> well that is great to see people flipping
2020-06-04 20:40:22 <adu> I thought GPT was a prerequisite for LUKS
2020-06-04 20:41:16 <adu> I want to rewrite setup-alpine as an ncurses installer
2020-06-04 20:56:35 <rjoensen> Cogitri: I downloaded it from alpines website.
2020-06-04 20:56:40 <rjoensen> I tested 2x isos.
2020-06-04 20:56:50 <rjoensen> The only iso I can get to work is the VIRTUAL one.
2020-06-04 21:25:44 <MY-R> LVM on LUKS working like a charm since... long time with BIOS/UEFI GPT/MBR stuff, at least that more manual way
2020-06-04 21:27:43 <MY-R> adu, when I saw OpenBSD installer then I started doubt in any gui, ui, tui, interfaces :D
2020-06-04 21:29:18 <adu> I've never used the openbsd installer
2020-06-04 21:29:24 <adu> What's it like?
2020-06-04 21:29:40 <MY-R> like a shell answer script :D
2020-06-04 21:29:55 <MY-R> simple and easy
2020-06-04 21:30:43 <helby> I like it
2020-06-04 21:31:18 <helby> openbsd has many nice and simple solutions
2020-06-04 21:31:37 <adu> Does setup-alpine have an lvm/luks question?
2020-06-04 21:31:56 <adu> I've only done the sys answer
2020-06-04 21:32:06 <MY-R> https://www.openbsd.org/faq/faq14.html#softraidFDE
2020-06-04 21:33:23 <MY-R> adu there where some changes to make that process easier but I still prefer more manual configuration
2020-06-04 21:34:08 <adu> I like to use USE_EFI=1 setup-alpine
2020-06-04 21:34:16 <adu> default everything except mbr
2020-06-04 21:34:46 <MY-R> yep
2020-06-04 23:35:15 <travankor> What tool do you guys use to suspend/hibernate?
2020-06-04 23:35:38 <travankor> pm-utils and loginctl seem like the main ones
2020-06-04 23:37:24 <maxice8> elogind
2020-06-04 23:37:26 <maxice8> which provides loginctl
2020-06-04 23:37:45 <maxice8> elogind also provides user sessions and $XDG_RUNTIME_DIR
2020-06-04 23:39:42 <travankor> thx
2020-06-04 23:40:14 <tehcloud> I use sudo to do that shit
2020-06-04 23:40:21 <tehcloud> then I don't need elogind
2020-06-04 23:40:52 <tehcloud> sudo pm-suspend && slock
2020-06-04 23:40:56 <tehcloud> that's how I roll on my laptop
2020-06-04 23:42:11 <maxice8> yeah, just use elogind :D
2020-06-04 23:49:30 <travankor> btw is Alpine-edge recommended for workstations/laptops?
2020-06-04 23:50:05 <maxice8> I wouldn't recommend personally
2020-06-04 23:50:38 <travankor> you use releases?
2020-06-04 23:51:00 <c705> i use edge, but i wouldn;t generally recommend it
2020-06-04 23:51:28 <c705> it's fairly stable, but i'm never all that surprised if a package breaks
2020-06-04 23:51:51 <travankor> stable compared to arch?
2020-06-04 23:52:09 <travankor> which breaks due to shlibs and such
2020-06-04 23:52:10 <maxice8> I use edge myself but i'm deeply involved with Alpine Linux
2020-06-04 23:52:29 <c705> oh, arch is terrible for that lol
2020-06-04 23:52:36 <c705> one reason why i left honestly
2020-06-04 23:52:38 <travankor> yeah lol
2020-06-04 23:52:42 <maxice8> apk automatically tracks shlibs like Void Linux, Debian, Fedora
2020-06-04 23:52:48 <maxice8> and basically every distro i think
2020-06-04 23:53:12 <maxice8> so the only problem would be being stopped from updating the system while there is a rebuild due to soname changes
2020-06-04 23:53:25 <maxice8> that is assuming all goes right (like someone forgetting to bump a package)
2020-06-04 23:54:18 <c705> yep, that happens to me sometimes
2020-06-04 23:54:42 <c705> not a big deal, i'll reboot and pull from v3.11 (v3.12 now) for a while before switching back to edge
2020-06-04 23:57:10 <Marquel> after about three hours of pointless searching, one question remains: how do i autoload tun module, along with kvm-amd? modules="tun kvm-amd" seems to only load kvm-amd, not tun as is expected.
2020-06-05 00:45:03 <Hello71_> echo tun > /etc/modules
2020-06-05 00:46:33 <Hello71_> arch shared libraries rebuild fails basically never. apk so dep is nice but not really a full solution
2020-06-05 01:46:01 <iso2013> hey so I've got a potato of an old laptop and I'm going to use it for a pretty low-power webserver (just local stuff). Would Alpine be a good fit? I need to be able to print documents and read from a serial port; I plan on using node.js
2020-06-05 02:35:50 <Terror> Hi team, trying to send a Merge request via gitlab (my first time on gitlab, have used gihub in the past) but can't get the linter to work. I get the following error: $ git fetch -nq $CI_MERGE_REQUEST_PROJECT_URL +refs/heads/$CI_MERGE_REQUEST_TARGET_BRANCH_NAME:refs/heads/$CI_MERGE_REQUEST_TARGET_BRANCH_NAME
2020-06-05 02:35:50 <Terror> fatal: could not read Username for 'https://gitlab.alpinelinux.org': No such device or address
2020-06-05 02:58:25 <JakeSays> hey i'm trying to use lbu. i have a alpine docker container. i ran it with /bin/sh and am able to install packages, but when i try to use lbu, it can't be found. what am i doing wrong?
2020-06-05 03:06:48 <gjabell> is there a way to keep an old kernel when upgrading a kernel in alpine?
2020-06-05 03:06:58 <gjabell> like how ubuntu/debian keeps the old version around in case it breaks?
2020-06-05 04:48:10 <ikke> JakeSays: I don
2020-06-05 04:48:26 <ikke> JakeSays: I don't think lbu makes sense in a docker container
2020-06-05 04:49:23 <ikke> JakeSays: but fyi, lbu is part of alpine-conf (which is not included in docker images because they are not used there)
2020-06-05 05:12:40 <voltagex> hey, what's the best way to create an Alpine image or ISO in a script / CI / build pipeline?
2020-06-05 05:33:46 <JakeSays> ikke: ah ok. makes sense.
2020-06-05 05:58:51 <rnkn> would anyone be interested in adding the unix calendar command to alpine?
2020-06-05 06:00:12 <rnkn> (not the cal command)
2020-06-05 06:01:44 <rnkn> https://www.commandlinux.com/man-page/man1/calendar.1.html
2020-06-05 06:02:22 <voltagex> where's the upstream source, mkn?
2020-06-05 06:02:58 <voltagex> "The -l and -w flags are Debian-specific enhancements. Also, the original calendar program did not accept 0 as an argument to the -A flag." so that manpage is from Debian.
2020-06-05 06:05:44 <rnkn> in macOS the man page says it first appeared in Version 7 AT&T Unix
2020-06-05 06:05:48 <rnkn> *UNIX
2020-06-05 06:05:59 <voltagex> found it, bsdmainutils
2020-06-05 06:06:23 <rnkn> oh sweet, so it's already in the box?
2020-06-05 06:06:35 <voltagex> sorry. that's from Debian
2020-06-05 06:07:16 <voltagex> util-linux maybe?
2020-06-05 06:08:01 <voltagex> no, again, sorry.
2020-06-05 06:08:27 <rnkn> I'm looking at the git tree of that, but you already beat me to it and no?
2020-06-05 06:09:18 <voltagex> https://aur.archlinux.org/cgit/aur.git/tree/PKGBUILD?h=bsdmainutils is probably a good start
2020-06-05 06:09:25 <rnkn> looks like it has cal but not calendar, which are deceptively different tools
2020-06-05 06:09:29 <voltagex> it seems crazy to me that they pull from Ubuntu
2020-06-05 06:09:56 <rnkn> Arch pulls from ubuntu? Crazy indeed
2020-06-05 06:11:29 <rnkn> would there be that many patches on top of the BSD origins?
2020-06-05 06:12:01 <voltagex> look at the PKGBUILD, it's not entirely straightforward
2020-06-05 06:12:23 <bleb> did they ever get plan9port working on alpine (musl)?
2020-06-05 06:13:00 <bleb> i know 9base works but if im not mistaken the graphical programs depend on glibc or at least did
2020-06-05 06:13:16 <voltagex> rnkn: https://github.com/freebsd/freebsd/tree/master/usr.bin/calendar
2020-06-05 06:13:22 <voltagex> best of luck :P
2020-06-05 06:15:55 <JakeSays> i'm trying to install gcc with apk add gcc-9.3.0-r2. i keep getting an unsatisfiable constraint gcc-9.3.0-r2 that's required by world[gcc-9.3.0-r2]
2020-06-05 06:16:09 <JakeSays> does that mean gcc has a dependency on itself?
2020-06-05 06:16:58 <rnkn> JakeSays: I think that just means it can't find it in your repos
2020-06-05 06:17:07 <JakeSays> ahh. apk add gcc worked
2020-06-05 06:32:53 <vejetaryenvampir> Is there a way to get Emacs 27.x on Alpine, other than compiling it?
2020-06-05 06:37:51 <Cogitri> You could bump the version that's available in the repos
2020-06-05 06:44:32 <Marquel> morning.
2020-06-05 06:45:32 <vejetaryenvampir> Cogitri: I'm not a native speaker, can define "bumping"?
2020-06-05 06:45:33 <rnkn> Cogitri: emacs 27 isn't released yet
2020-06-05 06:45:47 <Marquel> Cogitri: jFYI: it took me about three hours to get my Alpine libvirt/qemu HV back online - apparently with latest libvirt the monolithic daemon does not work (well enough) any longer and the modular daemons are required to become operational, maybe the team should look at that?
2020-06-05 06:46:55 <rnkn> vejetaryenvampir: an easier way might be to clone the aports repository to your $HOME, install abuild and edit the APKBUILD file to build 27.x instead of 26.x
2020-06-05 06:48:09 <vejetaryenvampir> rnkn: Will research Alpine's build system. ^^
2020-06-05 06:49:03 <mps> 'bump' is probably some kind of slang, also I'm not native speaker and I think Cogitri is not, also
2020-06-05 06:49:30 <voltagex> 'bump' - meaning increase the version number in the APKBUILD
2020-06-05 06:49:31 <mps> ime, bump is incremenet
2020-06-05 06:49:51 <rnkn> https://gitlab.alpinelinux.org/alpine/aports/-/blob/master/community/emacs/APKBUILD
2020-06-05 06:49:58 <Cogitri> Marquel: ah yes, I think the libvirt OpenRC file is broken right now
2020-06-05 06:50:07 <mps> voltagex: yes, in distro 'speaka'
2020-06-05 06:50:08 <rnkn> you'd need to change pkgver at the top, but also the shasums
2020-06-05 06:50:09 <Marquel> "bump" means, AFAIK, to just increase version number of a package without touching anything else related to that package.
2020-06-05 06:50:29 <voltagex> eh, seems to be programmer slang / forum slang mps
2020-06-05 06:50:38 <rnkn> or if someone wants to help me out with a way to ignore the shasums with abuild?
2020-06-05 06:51:01 <Cogitri> Just run abuild checksum to regenerate them
2020-06-05 06:51:24 <Marquel> Cogitri: while at it, maybe the init scripts should be checked to verify dependencies are met correctly, like virtnetworkd requires virtinterfaced and so on, to keep the order intact during sysboot ;)
2020-06-05 06:51:47 <Cogitri> Well, you have all the tools to change this upstream :)
2020-06-05 06:52:47 <Marquel> Cogitri: you are asking me to pop up in another's distro changelogs as well?
2020-06-05 06:52:58 <Marquel> do you actually know what you're bargaining for?
2020-06-05 06:54:11 <rnkn> if emacs is changed in aports upstream, perhaps it should be 27.x in edge and 26.x in community?
2020-06-05 07:22:33 <markand> hmmm, my 3.12 raspberry no longer boots, It boots and the kernel print all messages, but then it ends with [ 3.098093] hid-generic: input. blablabla
2020-06-05 07:22:37 <markand> and then cursor blinks
2020-06-05 07:22:45 <markand> I even tried init=/bin/sh just for test
2020-06-05 08:32:41 <guesto> hi ikke
2020-06-05 08:36:08 <guesto> i cant run xorg
2020-06-05 08:36:23 <guesto> it says "no screens found"
2020-06-05 08:42:04 <ikke> It probably means you are missing a video driver.
2020-06-05 08:42:08 <guesto> oh?
2020-06-05 08:42:14 <guesto> but i installed the ones
2020-06-05 08:42:23 <guesto> accordingly to the wiki
2020-06-05 08:42:54 <guesto> also lspci detects it
2020-06-05 08:43:08 <ikke> Also try to remove nomodeset from /etc/update-extlinux.conf if present and run update-extlinux
2020-06-05 08:44:07 <guesto> yeah is there
2020-06-05 08:44:25 <guesto> i removed it and ran update-extlinux
2020-06-05 08:44:38 <ikke> Now you have to reboot
2020-06-05 08:45:09 <guesto> alrighty
2020-06-05 08:46:26 <guesto> ok so
2020-06-05 08:46:38 <guesto> i installed the xf86-video-ati
2020-06-05 08:46:47 <guesto> and everything looks much smaller
2020-06-05 08:46:51 <guesto> the text
2020-06-05 08:46:56 <guesto> at tty
2020-06-05 08:47:11 <guesto> now theres a new problem
2020-06-05 08:47:26 <guesto> "libEGL.so no such file or directory"
2020-06-05 08:47:42 <guesto> (when trying to startxfce4)
2020-06-05 08:49:16 <guesto> nevermind i solved it myself
2020-06-05 08:49:27 <guesto> installed mesa-egl and voila
2020-06-05 08:52:08 <guesto> now my internet is gone
2020-06-05 08:52:30 <guesto> and its back again
2020-06-05 09:07:54 <guesto> alright so ikke
2020-06-05 09:08:07 <guesto> how did we go about installing rtl8821ce again?
2020-06-05 09:08:27 <guesto> ive installed it an wlan0 shows up, but its disconnected
2020-06-05 09:28:48 <guesto> @ikke
2020-06-05 09:52:51 <guesto> um
2020-06-05 10:15:28 <travankor> time to try wayland ;)
2020-06-05 10:15:59 <travankor> oh wait you have a different question
2020-06-05 10:19:43 <guesto> yeah
2020-06-05 11:31:46 <omni> for wireless, I've come to really apreciate iwd
2020-06-05 11:32:03 <omni> no more wpa_supplicant ftw!
2020-06-05 11:34:59 <mps> yes, I use it as client for more than a year (and even for AP in some cases)
2020-06-05 11:36:09 <travankor> iwd doesnt help if your driver is terrible
2020-06-05 11:38:24 <travankor> but completely agree the iwd's ux is 100% better than wpa_supplicant
2020-06-05 11:38:47 <travankor> and it's faster
2020-06-05 11:40:36 <mps> yes, it is first with FILS wifi implementation
2020-06-05 11:50:39 <omni> guesto: you've installed the driver and wlan0 show up, but is the link perhaps down?
2020-06-05 11:51:02 <omni> how have you configured your wlan/wifi?
2020-06-05 11:54:22 <mps> omni: are you asking about iwd?
2020-06-05 12:12:50 <omni> mps: no, trying to get more info on guestos issues
2020-06-05 12:43:56 <ncopa> anyone want help me fill in historical release data?
2020-06-05 12:44:39 <ncopa> we are moving the data from https://wiki.alpinelinux.org/wiki/Alpine_Linux:Releases to yaml so it can be used by scripts
2020-06-05 12:44:50 <ncopa> https://gitlab.alpinelinux.org/alpine/infra/infra/-/issues/10672
2020-06-05 12:47:12 <ncopa> it is moving to https://gitlab.alpinelinux.org/alpine/infra/alpine-mksite/-/blob/master/alpine-releases.conf.yaml
2020-06-05 13:15:33 <guesto> ikke are you here?
2020-06-05 13:21:15 <ikke> guesto: I'm busy, but as you notice, there are others here that can help as well
2020-06-05 13:44:54 <guesto> i see
2020-06-05 13:45:07 <guesto> can someone help me with an wifi issue?
2020-06-05 13:45:25 <guesto> im trying to use wpa_cli
2020-06-05 13:45:46 <guesto> but it "couldnt connect to wpasupplicant"
2020-06-05 13:49:27 <Hello71> did you start wpa_supplicant
2020-06-05 13:50:37 <guesto> yes
2020-06-05 13:50:41 <guesto> on boot aswell
2020-06-05 13:52:48 <mps> ctrl_interface=DIR=/run/wpa_supplicant GROUP=netdev
2020-06-05 13:53:50 <Hello71> did you set ctrl_interface
2020-06-05 14:14:34 <feuerteufel> ncopa: Are you looking for someone ist doing the typing work?
2020-06-05 14:15:16 <ikke> yes
2020-06-05 14:19:33 <ncopa> yes
2020-06-05 14:22:00 <feuerteufel> If there is no timeline, I probably able to do this if somebody gives me a hint where I'll find the data ...
2020-06-05 14:22:53 <ncopa> data is here: https://wiki.alpinelinux.org/wiki/Alpine_Linux:Releases
2020-06-05 14:23:43 <ncopa> releas notes sources are here: https://gitlab.alpinelinux.org/alpine/infra/alpine-mksite/-/tree/master/posts
2020-06-05 14:24:54 <feuerteufel> Like : eol_data: -- note: -- branch_date:
2020-06-05 14:25:09 <ncopa> if there is a missing release notes, then the release date should be added. you find that by clikcin on the link in "Directory name" on the wiki page and find the date of iso image
2020-06-05 14:25:22 <feuerteufel> Oh, let me see the second one
2020-06-05 14:25:37 <ncopa> eol data is listed on the wiki page
2020-06-05 14:26:02 <ncopa> end of support
2020-06-05 14:26:35 <feuerteufel> Is there a Timeline when it has to be done?
2020-06-05 14:27:06 <ncopa> this is an example of adding v3.8 https://gitlab.alpinelinux.org/alpine/infra/alpine-mksite/-/commit/15575a5530787e126d44b1aa6e35e8ba5ff82894
2020-06-05 14:27:12 <ncopa> not really
2020-06-05 14:27:20 <feuerteufel> Il give it a try and see how far I'll get ...
2020-06-05 14:27:35 <ncopa> if you only add one or two releases is helpful too
2020-06-05 15:14:19 <kahiru> would it be interesting for anyone to have grub for alpine compiled with support for zfs?
2020-06-05 15:23:48 <Hello71> isn't it on by default
2020-06-05 15:24:38 <Hello71> since 3.6 apparently: https://pkgs.alpinelinux.org/contents?file=zfs.mod&path=&name=&branch=v3.6
2020-06-05 15:24:43 <fujita_[m]> hi there, im interested in installing alpine linux but its my first time setting up an OS using only terminal. My question is how do I set a wireless connection before installing things like a desktop environment
2020-06-05 15:30:42 <mps> fujita_[m]: installer will ask you for parameters if it find wifi on your machine
2020-06-05 15:32:47 <fujita_[m]> will it prompt for the  name of the network connection and password?
2020-06-05 15:33:00 <mps> yes
2020-06-05 15:33:34 <mps> and will you use dhcp or manual net setup, dhcp is default
2020-06-05 15:33:52 <fujita_[m]> oh thanks
2020-06-05 15:34:20 <mps> it will ask for all network interfaces which is find during setup, you can skip some
2020-06-05 15:34:55 <guesto> hey im back
2020-06-05 15:35:11 <guesto> i got past the udhcpd lease for wlan0
2020-06-05 15:35:19 <guesto> and assigned a IP
2020-06-05 15:35:29 <guesto> but there is still no internet
2020-06-05 15:35:36 <guesto> ive added a dns server
2020-06-05 15:35:40 <guesto> obviously
2020-06-05 15:36:29 <guesto> the interface is up and running, and eth0 is down
2020-06-05 15:37:54 <afontain_> Can you ping an ip address?
2020-06-05 15:38:25 <guesto>  nope
2020-06-05 15:43:37 <guesto> it can somehow look up the ip of a server when i try to ping
2020-06-05 15:43:49 <guesto> but the actual icmp packets dont come through
2020-06-05 15:47:15 <guesto> oh wait i found something
2020-06-05 15:47:31 <guesto> "wlan0: association request to the driver failed"
2020-06-05 15:54:21 <guesto> got some new update: 4-way handshake failed
2020-06-05 16:00:40 <guesto> thats gone now
2020-06-05 16:00:53 <guesto> key negioation was a sucess
2020-06-05 16:01:08 <guesto> i ran udhcpc and got a ip, but i still cant ping
2020-06-05 16:01:15 <guesto> or do anything else
2020-06-05 16:12:42 <guesto> does anyone have a suggestion
2020-06-05 16:15:43 <c705> guesto: do you have a global route/default gateway?  `ip r`
2020-06-05 16:16:45 <guesto> yup
2020-06-05 16:28:15 <guesto> @ikke it worked last time haha
2020-06-05 16:50:01 <feuerteufel> ncopa: ikke: I have a first try here --> https://p.logout.de/?3381a6fa2e12472d#6sb9yKhnK6mooRKrnNXf7LFfkboGSXsppZAjVDqvVCd2
2020-06-05 16:50:58 <feuerteufel> Someone like to doublecheck?
2020-06-05 16:52:16 <feuerteufel> If my understanding for the System is correct, I think I found a typo:
2020-06-05 16:53:03 <feuerteufel>   - version: 3.8.5
2020-06-05 16:53:03 <feuerteufel>     date: 2020-03-23
2020-06-05 16:53:29 <ikke> that does not look right
2020-06-05 16:53:39 <ikke> oh, that's the end support date
2020-06-05 16:53:45 <feuerteufel> I think, it has to be "20-01-23"
2020-06-05 16:55:27 <feuerteufel> I take the dates form here (example) http://dl-cdn.alpinelinux.org/alpine/v3.8/releases/x86_64/
2020-06-05 16:56:39 <ncopa> I havent verified the dates but the format looks perfect
2020-06-05 16:57:28 <ncopa> feuerteufel: I'd like to credit you in the commit message. should I use your IRC nick or a real name?
2020-06-05 16:57:47 <guesto> ikke
2020-06-05 16:57:49 <guesto> please
2020-06-05 16:59:23 <ncopa> guesto: may be firmware problem for the wifi driver?
2020-06-05 17:00:50 <guesto> i dont think so
2020-06-05 17:00:50 <ncopa> feuerteufel: its weekend for me now. feel free to either create a merge request. If you are not comfortable with git and merge requests, then you can also send me the yaml file by email
2020-06-05 17:00:59 <ncopa> thanks!
2020-06-05 17:01:59 <guesto> also
2020-06-05 17:02:11 <guesto> udhcpc disconnects randomly for no reason
2020-06-05 17:02:30 <mps> guesto: could you try with iwd
2020-06-05 17:02:43 <guesto> iwd?
2020-06-05 17:03:10 <mps> apk info iwd
2020-06-05 17:04:05 <guesto> i installed it
2020-06-05 17:04:08 <guesto> what now?
2020-06-05 17:04:53 <mps> configure it and start
2020-06-05 17:05:00 <mps> read man pages
2020-06-05 17:05:09 <mps> apk add iwd-doc
2020-06-05 17:05:40 <mps> man iwd.config, man iwctl
2020-06-05 17:07:29 <guesto> right
2020-06-05 17:07:35 <guesto> i used iwdctl
2020-06-05 17:08:09 <guesto> it scanned the networks
2020-06-05 17:08:14 <guesto> i picked mine
2020-06-05 17:08:22 <guesto> and entered the password
2020-06-05 17:09:13 <mps> station wlan0 show
2020-06-05 17:09:19 <guesto> it says its connected
2020-06-05 17:09:42 <mps> ip a ?
2020-06-05 17:09:50 <mps> and 'ip r'
2020-06-05 17:10:30 <guesto> its UP
2020-06-05 17:10:46 <guesto> as for ip r
2020-06-05 17:10:53 <guesto> theres no gateway
2020-06-05 17:11:15 <mps> hmm
2020-06-05 17:11:28 <mps> default via .....
2020-06-05 17:11:28 <fujita_[m]> Does anyone here have experience secure boot mode
2020-06-05 17:11:46 <guesto> default via my gateway eth0
2020-06-05 17:11:54 <fujita_[m]> What is the settings its should be in since  my PC shows no boot device
2020-06-05 17:11:56 <guesto> not wlan0.
2020-06-05 17:12:21 <mps> guesto: ifdown eth0
2020-06-05 17:12:33 <mps> then try to restart iwd
2020-06-05 17:12:49 <guesto> done
2020-06-05 17:12:51 <mps> and, bts, apk add openresolv
2020-06-05 17:13:15 <guesto> bts?
2020-06-05 17:13:44 <mps> and set in /etc/iwd/main.conf 'NameResolvingService=resolvconf' in [Network] section
2020-06-05 17:13:56 <mps> sorry, btw
2020-06-05 17:14:12 <guesto> ip r outputs nothing now
2020-06-05 17:15:02 <mps> does your AP send 'default gateway' parameter
2020-06-05 17:15:13 <guesto> i dont have internet now lol
2020-06-05 17:15:20 <guesto> and yes
2020-06-05 17:15:53 <feuerteufel> ncopa: Use my nick, but we'll see wen Iget it done ...
2020-06-05 17:16:16 <guesto> i added eth0 back and installed openresolv
2020-06-05 17:16:28 <mps> you can stop iwd and run it in debug mode by '/usr/libexec/iwd -d' and see what happens
2020-06-05 17:17:17 <guesto> it said connectef
2020-06-05 17:17:20 <guesto> and stoppef
2020-06-05 17:17:57 <mps> also, dmesg can told you connection status
2020-06-05 17:18:19 <mps> trying to associate to ....
2020-06-05 17:18:33 <feuerteufel> ncopa: Ok, so I'll go further on like I did ...
2020-06-05 17:18:33 <mps> and, info: associated to bssid ....
2020-06-05 17:19:41 <mps> guesto: I have to go out, bbl
2020-06-05 17:19:49 <guesto> mps it says its connected
2020-06-05 17:19:56 <guesto> but it doesnt worklk
2020-06-05 17:19:58 <guesto> workk
2020-06-05 17:43:31 <wener[m]> <opal "wener which rpi do you have"> opal:  4b 3b zero
2020-06-05 17:55:58 <guesto> someone
2020-06-05 17:56:09 <guesto> please.
2020-06-05 17:57:13 <c705> guesto: ask your question and be patient please
2020-06-05 17:58:15 <artok> some power management happens on wlan chip ?
2020-06-05 17:58:27 <guesto__> already
2020-06-05 17:58:35 <guesto__> i have done that
2020-06-05 17:58:40 <guesto__> multiple times
2020-06-05 18:00:15 <c705> irc is not an instant service. if some knows they will answer. if not, be patient
2020-06-05 18:00:46 <guesto__> right
2020-06-05 18:00:48 <guesto__> sorry
2020-06-05 18:11:40 <fujita_[m]> Hi is anyone free
2020-06-05 18:13:43 <fujita_[m]> I need help booting from uefi
2020-06-05 18:15:40 <ikke> fujita_[m]: it works best to just ask your actual question
2020-06-05 18:20:44 <emr> exit
2020-06-05 18:20:46 <emr> exit
2020-06-05 18:21:24 <helby> Doas non-login shell ash read something where can I pass my aliases?
2020-06-05 18:22:31 <helby> or should I force in my terminal that ash is login shell, then it will read .profile ?
2020-06-05 18:23:25 <c705> helby: it will read .profile, yes
2020-06-05 18:24:04 <c705> you can put stuff in your `.profile` that will source aliases from a different file if you
2020-06-05 18:24:08 <c705> if you'd like
2020-06-05 18:25:36 <helby> c705: it read profile when I login in tty as it's login shell
2020-06-05 18:26:14 <helby> c705: but then I go to sway and shells in my terminal are not login shells
2020-06-05 18:26:35 <c705> yes they are
2020-06-05 18:26:41 <helby> c705: so i some variable I put in .profile with 'export key=value' that works, as it export them
2020-06-05 18:26:53 <helby> but it does not takes alias = ....
2020-06-05 18:27:13 <c705> that's not syntatically correct
2020-06-05 18:27:30 <helby> what u mean?
2020-06-05 18:27:31 <ikke> c705: I don't think non-login shells read .profile
2020-06-05 18:27:42 <helby> yes, non-login does not read it
2020-06-05 18:27:47 <afontain_> No spaces
2020-06-05 18:27:59 <ikke> alias foo='..'
2020-06-05 18:28:03 <c705> ikke: you're right, but I don't think this person is actually using a non-login shell
2020-06-05 18:28:06 <afontain_> alias ls="ls -h"
2020-06-05 18:28:11 <helby> while I can export variables from first loginshell at tty to sway
2020-06-05 18:28:31 <ikke> helby: when you start a new terminal, you get a login shell
2020-06-05 18:28:32 <helby> these exports are exported so no problem with that
2020-06-05 18:28:40 <helby> ikke: no
2020-06-05 18:28:49 <KatolaZ> @ikke mhhh no
2020-06-05 18:28:53 <KatolaZ> you get an interactive shell
2020-06-05 18:29:00 <ikke> oh, right, sorry
2020-06-05 18:29:00 <KatolaZ> you get a login shell only if you login
2020-06-05 18:29:07 <helby> yes
2020-06-05 18:29:19 <helby> so how can I set up aliases in that case?
2020-06-05 18:29:38 <helby> while I can force ash to be login shell in terminal setting ... I just need these aliases
2020-06-05 18:30:06 <c705> my point remains: put them in .profile
2020-06-05 18:30:16 <helby> c705: login shell does not read it
2020-06-05 18:30:19 <KatolaZ> helby: you cannot force a shell to be a login shell
2020-06-05 18:30:26 <helby> *non-login
2020-06-05 18:30:35 <helby> KatolaZ: yes I can
2020-06-05 18:30:41 <c705> your _interactive_ shell will read it
2020-06-05 18:30:43 <KatolaZ> how escatly?
2020-06-05 18:30:50 <helby> ash --login
2020-06-05 18:30:55 <KatolaZ> s/esc/ex/
2020-06-05 18:31:03 <KatolaZ> helby: that's a sub-shell
2020-06-05 18:31:31 <KatolaZ> i.e., it's *another* shell
2020-06-05 18:31:34 <helby> KatolaZ: how subshell?
2020-06-05 18:31:48 <helby> KatolaZ: I can force it from my terminal settings
2020-06-05 18:32:03 <helby> so if I open it it does not open ash but 'ash --login'
2020-06-05 18:32:07 <Hello71> apparently you can export ENV
2020-06-05 18:32:19 <helby> Hello71: yes can export env, but don't know how export alias
2020-06-05 18:32:22 <Hello71> echo 'export ENV=$HOME/.shrc' >> ~/.profile
2020-06-05 18:32:30 <helby> oh
2020-06-05 18:32:55 <KatolaZ> helby: I genuinely don't understand
2020-06-05 18:33:00 <KatolaZ> I bet you know better
2020-06-05 18:33:02 <KatolaZ> :)
2020-06-05 18:34:19 <helby> Hello71:so just create some file, put there aliases then export it?
2020-06-05 18:35:40 <Hello71> that is one way. if you set sh --login as terminal shell then it will override your environment settings
2020-06-05 18:36:14 <Hello71> suppose you use xterm (not sure if it is configurable, but assume), and you run PATH=whatever xterm, then it will be (mostly) ignored
2020-06-05 18:36:33 <KatolaZ> helby: using --login is a "fake" login shell
2020-06-05 18:36:40 <KatolaZ> and will probably cause more problems than not
2020-06-05 18:37:36 <helby> Hello71: thanks your export works , that was I was lookign for
2020-06-05 18:37:50 <helby> Hello71: using alacritty in sway, no xorg at all here
2020-06-05 18:38:09 <Hello71> KatolaZ: not really any problems, just that shell will reread profile, so it will override any temporary environment settings
2020-06-05 18:38:23 <KatolaZ> Hello71: that's the problem I refer to
2020-06-05 18:38:24 <KatolaZ> :)
2020-06-05 18:38:25 <helby> Hello71: now I know how to source other files in a case
2020-06-05 18:38:26 <Hello71> helby: imo on desktop better to just install zsh :p
2020-06-05 18:38:37 <KatolaZ> I don't want to have stuff re-defined in a non-login shell
2020-06-05 18:38:57 <KatolaZ> Hello71: the solution you suggested via ENV is the right one
2020-06-05 18:39:24 <helby> Hello71: oh dude, I don't have even elogind or dbus, this is just bare bone :)
2020-06-05 18:39:58 <helby> KatolaZ: yes that is okay, I was not inclined to force that --login
2020-06-05 18:40:26 <Hello71> imo elogind is not critical on single-user machine, but without dbus you lose notifications
2020-06-05 18:40:37 <helby> Hello71: notification about what?
2020-06-05 18:40:47 <helby> I don't need that soemthing pop up on my face ;)
2020-06-05 18:40:56 <Hello71> irc messages?
2020-06-05 18:41:07 <helby> that never worked for me anyway in irssi
2020-06-05 18:41:16 <Hello71> dunno, some people aggressively turn off new mail and such
2020-06-05 18:41:48 <helby> I login/logout to webmail anytime I want to read/check email
2020-06-05 18:43:11 <helby> Hello71: btw. now it's probably good idea to put everything there and have just that one line in .profile?
2020-06-05 19:24:01 <helby> I have also some other shit around if you dont mind I will share it with you :)
2020-06-05 19:25:14 <helby> As I don't use dbus and elogind, have some issue regarding firefox, if I have already opened it, when I try to open link from a terminal or any program running there of course ... not sure if I can achieve that a link will open in existing window as a tab
2020-06-05 19:25:36 <helby> bc. right now I get just error pop from firefox, that a process is already running I have ot close it or start new profile ...
2020-06-05 19:26:00 <helby> is it somethign what normally dbus or elogind take care of it?
2020-06-05 19:26:12 <Cogitri> It's definitely not elogind
2020-06-05 19:27:18 <Cogitri> Might be dbus
2020-06-05 19:32:20 <Hello71> needs dbus iirc
2020-06-05 19:36:15 <c705> firefox needs dbus, yes
2020-06-05 19:49:08 <Hello71> apparently it works without it
2020-06-05 19:49:11 <Hello71> sort of
2020-06-05 19:58:29 <c705> "sort of", yes
2020-06-05 19:58:38 <c705> certain pages crash the tab
2020-06-05 20:40:13 <helby> using it just few days, but it did not crash once, also no any issue without dbus
2020-06-05 20:40:35 <helby> just was not sure about this if it's something what I can fix somehow without dbus
2020-06-05 20:41:11 <helby> please don't send me a link, I have to copy it :)
2020-06-05 20:42:47 <helby> also interested in, what chrony do so huge, that it takes few seconds during a boot
2020-06-05 20:42:55 <helby> changed to busybox and it's done fast
2020-06-05 20:44:14 <helby> but of course not sure what busybox ntpd do there and if it does the same job as chrony
2020-06-05 20:45:59 <c705> both are ntp daemons and likely do the same thing
2020-06-05 20:46:26 <c705> chronyd activenly contacts a ntp server when started which is why the service blocks for a bit before coming online
2020-06-05 20:47:11 <Hello71> wait, it does?
2020-06-05 20:47:26 <c705> for like..maybe a second or two
2020-06-05 20:47:34 <c705> I'm assuming that's why it blocks
2020-06-05 20:49:42 <Hello71> ah, alpine sets initstepslew
2020-06-05 20:50:00 <Hello71> apparently you can override it with FAST_STARTUP, but that seems mis-designed
2020-06-05 20:51:02 <c705> my post boot init scrips already take betwen 5-10 minutes, what's another few seconds?
2020-06-05 20:59:59 <helby> I am okay with busybox if job is done, don't need additional package
2020-06-05 21:01:02 <gjabell> anyone know why there are so few erlang packages in alpine>3.10?
2020-06-05 21:02:59 <gjabell> oh I guess they were removed because of circular deps
2020-06-05 21:06:11 <gjabell> as far as actually *using* the subpackages that were removed, is the fix just to use alpine 3.11?
2020-06-05 21:22:14 <opal> wener[m]: ah, mine is older lol
2020-06-05 23:28:02 <dreadkopp> any fix for iconv //IGNORE & //TRANSLIT yet ?
2020-06-05 23:45:34 <Hello71> fix?
2020-06-06 00:58:33 <dsm42> apk search go ; is so much stuff; how does the shaman search for the golang package?
2020-06-06 01:04:19 <dsm42> this will do:   apk search go | egrep '^go-'
2020-06-06 01:47:30 <p3lim> trying to use qemu-guest-agent, the service just crashes because the device path doesn't exist
2020-06-06 01:47:55 <p3lim> it expects it to be at /dev/virtio-ports/org.qemu.guest_agent.0, but it's at /dev/vport2p1
2020-06-06 01:52:01 <p3lim> I could override the service, but I see that I can set GA_PATH somewhere to override the device, although I have no clue where I'd set that variable
2020-06-06 03:55:49 <vejetaryenvampir> Can I create virtual packages that handles dependency tree, when I install them?
2020-06-06 03:56:21 <vejetaryenvampir> Like for manually compiled programs?
2020-06-06 05:10:17 <winny> vejetaryenvampir: i'm sure that's possible, but you would find it pretty straight forward to package your software
2020-06-06 05:52:37 <ikke> vejetaryenvampir: for reference, apk add -t <virtual-name> <depenencies..>, but like winny said, packaging should be as easy
2020-06-06 07:40:45 <winny> TIL. Very cool feature :)
2020-06-06 10:58:27 <velix> Sentences, you NEVER want to read in your life - I just did :( "license was changed from GPL to a proprietary one"
2020-06-06 11:08:14 <vejetaryenvampir> ikke: If I explain what I'm talking about, it might be easier for telling me the right way. ^^ I want to have packages like `dwm`, `st`, `dmenu`, the softwares I daily use. So, I wouldn't need to have `libx11-dev`, `libxft-dev` in my world set. Any recommenations?
2020-06-06 11:24:45 <ikke> vejetaryenvampir: all these packages are already available
2020-06-06 11:24:50 <ikke> in the repositories
2020-06-06 11:25:10 <ikke> https://pkgs.alpinelinux.org/package/edge/community/x86_64/dmenu https://pkgs.alpinelinux.org/package/edge/community/x86_64/st https://pkgs.alpinelinux.org/package/edge/community/x86_64/dwm
2020-06-06 11:25:36 <ikke> So it should be as easy as `apk add dwm dmenu st` (assumed you have community enabled)
2020-06-06 11:26:18 <vejetaryenvampir> Well, I know that. The thing is, as all suckless' software, you fork the repo and patch your own that differs from the original copy.
2020-06-06 11:27:09 <ikke> You can build your own packages
2020-06-06 11:27:16 <ikke> and add patches
2020-06-06 11:27:23 <vejetaryenvampir> In there, I'm trying to manage my dependencies in a better way: one virtual package for handiling dependencies of a program that isn't in official repo or for some reason I don't want to use it.
2020-06-06 11:27:31 <vejetaryenvampir> ikke: Okay, let me see.
2020-06-06 11:27:47 <vejetaryenvampir> Don't get me wrong, in Gentoo, I just create a virual package.
2020-06-06 11:28:06 <ikke> apk has that too
2020-06-06 11:28:12 <vejetaryenvampir> Though, I totally could've create a a package, in there too.
2020-06-06 11:28:31 <ikke> But creating a custom local repo with patches is quite easy as well
2020-06-06 11:28:40 <ikke> so depends on what you prefer
2020-06-06 11:28:45 <vejetaryenvampir> I'm reasearching it. ^^
2020-06-06 11:29:19 <vejetaryenvampir> Is (the apk guide) in Developer Documentation?
2020-06-06 12:13:08 <mps> of these all I built st with patches as branch in alpine aports locally
2020-06-06 13:24:04 <MY-R> did something changed dunno, in bash lately? simple for example "echo never > /sys/kernel/mm/transparent_hugepage/enabled" return: -bash: echo: write error: Invalid argument
2020-06-06 13:24:44 <MY-R> need add "-n" to echo to avoid this "error"
2020-06-06 14:59:21 <via> i'm trying to boot alpine (generic arm) on a pinebook pro (rk3399). extlinux seems to be unhappy with the vmlinuz-lts: https://gist.github.com/via/fc8b0fab1643dac5288f2e433b614695
2020-06-06 15:00:07 <via> any idea what i'm doing wrong here? gunziping and file'ing it shows me "/tmp/bleh: MS-DOS executable", which seems odd
2020-06-06 15:20:22 <mps> via: which tarball you use armv7 or aarch64
2020-06-06 15:21:49 <via> aarch64
2020-06-06 15:26:47 <mps> u-boot shows that it retrieve vmlinuz-lts two times, strange
2020-06-06 15:30:38 <via> i've tried manually loading it and `booti`-ing it, it seems unhappy with the image format
2020-06-06 15:31:37 <mps> could you paste /extlinux/extlinux.conf to tpaste.us
2020-06-06 15:33:24 <via> https://tpaste.us/reRQ its just whats in the tarball
2020-06-06 15:33:56 <via> but yeah, i mean i still get it if i manually try to booti the image after fatload'ing it, so i don't think its extlinux. this u-boot doesn't like the kernel image format
2020-06-06 15:34:33 <mps> looks good
2020-06-06 15:35:30 <mps> I don't have this hardware to test, but in qemu it worked
2020-06-06 15:41:16 <via> ok, i'll keep playing around
2020-06-06 16:09:20 <Hello71> via: you probably need to upgrade your file
2020-06-06 16:09:51 <Hello71> and/or add -k
2020-06-06 16:10:59 <Hello71> I think even on arm, efistub is PE format
2020-06-06 16:17:24 <via> Hello71: upgrade my file?
2020-06-06 16:17:47 <via> i don't think efi is used in this boot process
2020-06-06 16:19:51 <Hello71> but the kernel is compiled as efistub format
2020-06-06 16:20:05 <Hello71> which starts with MZ magic number
2020-06-06 16:20:06 <via> i suppose that is my issue then
2020-06-06 16:20:26 <Hello71> supposed to be backwards compatible though
2020-06-06 16:20:46 <Hello71> could be your firmware is overly picky
2020-06-06 16:22:25 <Magnie> Hello! I am netbooting Alpine Linux onto a Raspberry Pi 4 and then trying to install and run k3s on it. I am running into an issue where 'nf_conntrack_max' can't be set (or found) and /lib/modules also seems to be missing. Any suggestions on how I can add 'nf_conntrack` to the modules?
2020-06-06 19:04:37 <helby> )/leve
2020-06-06 21:56:52 <nacelle> Magnie: try install nftables
2020-06-06 21:57:09 <nacelle> installing, even
2020-06-06 22:06:43 <dreadkopp> cheers comunity, any recommendations to get dompdf to run in a alpine php image ? using php-fpm image based in alpine 3.11 . tried this fix: https://ndp-studio.gitbooks.io/manual/d8/docker/dompdf.html still broken for me though
2020-06-06 23:16:31 <helby> should I use ca-certificates if want tls connection? somehow without it I got error
2020-06-06 23:16:41 <helby> just with ca-certificates-bundle
2020-06-06 23:16:54 <helby> but ca-certificates is not dependency of something I have installed
2020-06-06 23:17:47 <helby> weird to me, that I can end without a tls/ssl or whatever is behind it support
2020-06-06 23:29:09 <helby> not sure now how I installed ca-certificates at all before
2020-06-06 23:29:58 <helby> or how my tls connection works before :/
2020-06-07 06:52:57 <IvanM5> Hello! On the new release 3.12.0 bitcoind does not work out of the box: "what():  locale::facet::_S_create_c_locale name not valid". Though if I run it with "LANG=C bitcoind" it will start. What would be the proper place to put this setting to? Or maybe there is another proper way to solve this issue?
2020-06-07 07:11:19 <russkel> IvanM5, UTF-8 ?
2020-06-07 07:12:12 <ikke> IvanM5: is it an openrc service?
2020-06-07 07:13:55 <IvanM5> Yes / yes
2020-06-07 07:15:35 <ikke> You could add it to /etc/conf.d/bitcoind
2020-06-07 07:18:00 <IvanM5> ikke, thanks, it works. Shall I report this to Alpine devs?
2020-06-07 07:18:14 <IvanM5> (team)
2020-06-07 07:18:41 <ikke> Create an issue on gitlab.a.o
2020-06-07 07:19:22 <IvanM5> Thanks, will do.
2020-06-07 07:28:33 <IvanM5> Done: https://gitlab.alpinelinux.org/alpine/aports/-/issues/11620
2020-06-07 07:31:32 <IvanM5> Well, I was too fast to report it works....
2020-06-07 07:32:26 <ikke> you need to export it
2020-06-07 07:32:40 <ikke> export LANG=C
2020-06-07 07:40:44 <IvanM5> I do export it now. Though one more question - should it be /etc/conf.d/bitcoind or /etc/conf.d/bitcoin ? If the service is /etc/init.d/bitcoin
2020-06-07 07:41:55 <ikke> the latter
2020-06-07 07:42:06 <ikke> the servicename and conf.d file should match
2020-06-07 07:45:04 <IvanM5> Did - will paste to the issue (multiple lines)
2020-06-07 07:48:34 <IvanM5> Pasted. I fixed my mistakes, but it does not start being invoked from /etc/init.d/bitcoin
2020-06-07 08:13:01 <rnkn> hello, I'm messing around with abuild and the build command says "Compile and install package into $pkgdir"
2020-06-07 08:13:07 <rnkn> what should I specify as $pkgdir?
2020-06-07 08:19:43 <Cogitri> rnkn: abuild sets pkgdir itself
2020-06-07 08:19:53 <Cogitri> It's usually $pwd/pkg
2020-06-07 08:20:15 <Cogitri> It's only a temporary dir where software installs to and is later compressed into the apk package
2020-06-07 08:20:53 <rnkn> Cogitri: I see, I guess these means I'm not doing this final compress-to-apk step
2020-06-07 08:21:00 <rnkn> *this means
2020-06-07 08:21:10 <Cogitri> Just doing abuild -r should get you there
2020-06-07 08:21:30 <rnkn> I tried that but it just builds the package then deletes it
2020-06-07 08:21:44 <rnkn> it's kinda weird
2020-06-07 08:22:56 <rnkn> is this the correct way if my intent is to just build and install the program locally (not create a apk for distribution)?
2020-06-07 08:23:20 <Cogitri> abuild -r builds it and then puts the resulting apk package into ~/packages
2020-06-07 08:23:50 <rnkn> haha oh I never saw that
2020-06-07 08:24:09 <rnkn> so that's where they've all gone
2020-06-07 08:24:50 <rnkn> from that point I then install with sudo apk add ~/packages/PATH/TO/PKG.apk ?
2020-06-07 08:25:14 <Cogitri> Yup
2020-06-07 08:25:23 <rnkn> sweeet! thanks!
2020-06-07 08:25:34 <rnkn> I've been manually moving them into /usr/local/bin
2020-06-07 08:27:08 <rnkn> sorry one more little question... say I want to change something in the program's configuration (e.g. config.h file) before building, at which step should I intervene?
2020-06-07 08:27:54 <rnkn> I've been doing: fetch, verify, unpack, prepare, deps, [edit config.h], build
2020-06-07 08:29:00 <TBB> use the prepare hook for that
2020-06-07 08:29:19 <ikke> rnkn: you can also add ~/packages/<dir> to /etc/apk/repositories
2020-06-07 08:29:40 <rnkn> ikke: ah nice!
2020-06-07 08:29:44 <ikke> and your local build key to /etc/apk/keys
2020-06-07 08:29:50 <ikke> then you can just do apk add <pkg>
2020-06-07 08:30:19 <rnkn> TBB: in which case I'd be editing the config.h with sed in the APKBUILD?
2020-06-07 08:30:47 <TBB> for example, yes
2020-06-07 08:31:04 <rnkn> TBB: what if I'm not smart enough for that?
2020-06-07 08:32:00 <IvanM5_> ikke, could you advise on bitcoin+LANG? Maybe I'm doing something wrong again?
2020-06-07 08:33:29 <ikke> rnkn: you can also create a patch file and add it to source
2020-06-07 08:33:38 <ikke> rnkn: that will get applied automatically if it ends in .patch
2020-06-07 08:33:48 <rnkn> ikke: oh wow that is cool
2020-06-07 08:34:07 <ikke> make sure that patch applies from builddir
2020-06-07 08:34:29 <ikke> src/<pkgname>-<pkgver>/ usually
2020-06-07 08:36:07 <ikke> IvanM5_: if adding `export LANG=C` to /etc/conf.d/bitcoin does not work, I'm not sure either
2020-06-07 08:37:19 <IvanM5_> Thanks, let me dig further.
2020-06-07 08:43:41 <Marquel> 
ACTION thinks apk should make sure /boot is properly mounted when updating alpine's kernel or bootloader...

2020-06-07 08:44:25 <ikke> update-extlinux should do that, not apk
2020-06-07 08:44:41 <ikke> if at all
2020-06-07 08:59:53 <Marquel> ikke: well, here it was grub, and i do not know which piece puts the kernelbinary into /boot, but certainly that piece of software did not do that either.
2020-06-07 09:04:52 <mps> setup-alpine adds /boot to fstab
2020-06-07 09:06:19 <mps> alpine-baselayout
2020-06-07 09:12:41 <Marquel> mps: right, but why keep /boot mounted all the time when in fact it is only required at upgrade time?
2020-06-07 09:16:14 <mps> just in case :)
2020-06-07 09:45:15 <ikke> apk itself is not responsible for populating /boot
2020-06-07 09:45:57 <ikke> Marquel: if you don't have /boot mounted, you need to make sure yourself somehow that it's mounted at upgrade time
2020-06-07 09:47:36 <mps> and /boot can reside on / also
2020-06-07 09:47:56 <mps> no separate partition, I mean
2020-06-07 10:03:41 <wener[m]> AlpineLinux prebuild disk image https://github.com/wenerme/alpine-image/releases for thoses who don‘t want to setup manually, still trying build more arches
2020-06-07 10:04:50 <Marquel> mps: then it would be cool if the installer offered to not use a separate partition for lvm/grub setups.
2020-06-07 10:05:16 <Marquel> ikke: is there an "# apk upgrade --dry-run"?
2020-06-07 10:05:45 <mps> Marquel: apk upgrade --simulate
2020-06-07 10:09:09 <mps> and, don't know but maybe 'export BOOT_SIZE=0' before run setup-alpine disable it
2020-06-07 10:09:16 <mps> didn't tried
2020-06-07 11:44:03 <afraid> I wonder if that old big package fuckup with that dependency is fixed, yet
2020-06-07 11:44:09 <afraid> I was just going to nuke my install and go from there
2020-06-07 12:55:01 <trfl> is it a known problem that the keyboard stops working in xfce after upgrading to 3.12?
2020-06-07 12:55:19 <trfl> not even ctrl-alt-F2 does anything on my laptop at least
2020-06-07 12:59:50 <trfl> apparently I am also unable to open any terminal emulators, just get an hourglass... and that also disables the logout button in the xfce menu, making it do nothing
2020-06-07 13:08:15 <Cogitri> That's certainly not known
2020-06-07 13:13:32 <MY-R> trfl, kill X session and check "/var/log/Xorg.0.log"
2020-06-07 13:14:23 <trfl> that's the issue, I can't kill X since keyboard is dead :p currently trying to disable lxdm from single
2020-06-07 13:14:47 <trfl> ok that worked, keyboard works in TTY before X is started
2020-06-07 13:15:11 <MY-R> next time just use: ALT + SysRq + r
2020-06-07 13:15:42 <MY-R> and after can switch to TTY
2020-06-07 13:16:26 <ikke> don't you need to have magic sysrq enabled?
2020-06-07 13:16:34 <MY-R> isnt enabled by default?
2020-06-07 13:16:48 <ikke> afaik, not anymore, but not sure
2020-06-07 13:17:00 <MY-R> worked for me few days ago just fine
2020-06-07 13:17:07 <Cogitri> I think it's on by default
2020-06-07 13:18:13 <ikke> cat /proc/sys/kernel/sysrq
2020-06-07 13:18:41 <trfl> that's 1 on my box fwiw
2020-06-07 13:18:46 <ikke> then it's enabled
2020-06-07 13:18:52 <ikke> CONFIG_MAGIC_SYSRQ_DEFAULT_ENABLE=0x1
2020-06-07 13:19:38 <MY-R> it was the only way to cleanly restart PC after frozen pc caused by Intel gpu... and the reason why I had to buy some cheap radeon card
2020-06-07 13:22:31 <yunfan> mps: is it possible to use alpine to replace openwrt?
2020-06-07 13:22:54 <yunfan> i really dislike openwrt, thought i am maintain a package for it
2020-06-07 13:33:28 <trfl> my
2020-06-07 13:34:13 <trfl> my-r: thanks for the tip regarding altgr-sysrq-R, but looks like keyboard is sufficiently disables for that to do nothing as well
2020-06-07 13:34:45 <trfl> however I can switch in and out of Xorg with `service lxdm start; sleep 10; service lxdm stop` so I'll keep looking with that
2020-06-07 13:34:57 <ikke> yunfan: possibly yes, but you have to verify that Alpine runs on the specific device, and you have to do a lot more manual setup work
2020-06-07 13:35:53 <MY-R> trfl, after alt-sysrq-r cant use ctrl+alt+F1 F2 etc?
2020-06-07 13:36:28 <yunfan> ikke: though many soho router use mips, lucily, mine has a arm cpu inside, i guess that would be act like my rpi boards
2020-06-07 13:36:32 <trfl> oh! yes that works! I assumed sysrq-R would kill X somehow, my bad
2020-06-07 13:37:47 <MY-R> alt+sysrq+k should kill but need do alt+sysrq+r before sometimes too
2020-06-07 13:42:40 <mps> yunfan: yes, it is possible, I did that two years ago :)
2020-06-07 13:43:02 <mps> but as ikke says, if it works on target device
2020-06-07 13:47:05 <mps> I ran operwrt for about year and a half on lamobo R1 (allwinner A20) and about two years ago switched it to alpine
2020-06-07 13:47:50 <mps> but had to build kernel for it all the time, till a two months ago when created linux-edge for alpine
2020-06-07 13:48:01 <mps> now it is pure alpine :)
2020-06-07 13:48:32 <mps> even u-boot is in alpine, I enbabled it in repo about year ago
2020-06-07 13:50:05 <mps> another A20 board for radio amateur repeater is leemaker banana pi, also with pure alpine
2020-06-07 13:51:02 <mps> and one cubieboard A10 as a radio APRS relay is pure alpine
2020-06-07 13:58:33 <yunfan> mps: but you use that board as router?
2020-06-07 14:00:24 <mps> yes, adsl router and web, mail server
2020-06-07 14:00:41 <mps> no GUI
2020-06-07 14:00:53 <yunfan> that's cool i just searched this board
2020-06-07 14:00:59 <yunfan> it has a sata dock :D
2020-06-07 14:01:16 <mps> eh, sata is bad on it
2020-06-07 14:01:41 <mps> design bug for power on sata interface
2020-06-07 14:02:14 <mps> I tried to fix it, but it is still unstable on higher sdd disk load
2020-06-07 14:02:44 <mps> maybe adding separate power supply for sata would help
2020-06-07 14:03:18 <mps> banana pi have stable sata interface
2020-06-07 14:07:11 <yunfan> a benefit for choosing A20 cpu
2020-06-07 14:09:19 <MY-R> mps, I noticed on your linux-edge that IO scheduler is by default set to NONE, maybe set up some sheduler and try add dunno, read/write limits would help
2020-06-07 14:11:01 <mps> MY-R: intentionally
2020-06-07 14:11:17 <mps> left to user to set what it need
2020-06-07 14:11:38 <mps> I found that NONE is actually best for some mmc cards
2020-06-07 14:12:16 <MY-R> ye but not for HDD
2020-06-07 14:27:38 <mps> MY-R: yes, for hdd you can set what you find is best
2020-06-07 14:28:36 <mps> and we expect that the users of testing (and alpine) have knowledge how to set these options
2020-06-07 14:29:20 <mps> as you are :)
2020-06-07 14:29:44 <MY-R> I didnt expect NONE io scheduler at all :D
2020-06-07 14:30:18 <mps> heh, with alpine expect unexpectable :)
2020-06-07 14:30:40 <MY-R> and I think it was the problem when used HDD disk for torrent, qbitorrent just crashed and now cant reproduce it anymore
2020-06-07 14:30:44 <MY-R> :D
2020-06-07 14:33:27 <feuerteufel> mps: did you find something out about the kernel upgrade problem on rpi?
2020-06-07 14:33:29 <mps> I thought to write what is different to linux-lts
2020-06-07 14:34:57 <feuerteufel> What I yust fond is, I don't have a "/boot". I only have a "/media/mmcblk0p1/boot" ...
2020-06-07 14:34:57 <mps> feuerteufel: uhm, please remind me, I forgot sorry
2020-06-07 14:35:20 <feuerteufel> np
2020-06-07 14:35:28 <mps> feuerteufel: do you run it in sys mode or from ram
2020-06-07 14:35:28 <feuerteufel> It's WE
2020-06-07 14:36:55 <mps> 'WE' ?
2020-06-07 14:36:55 <feuerteufel> I think, it runs on ram when I forget lbu ci -dv everything new is gone :-))
2020-06-07 14:37:08 <feuerteufel> Weekend
2020-06-07 14:37:10 <mps> hehe, nice :)
2020-06-07 14:37:38 <mps> put 'lbu' in cron :)
2020-06-07 14:38:40 <feuerteufel> It's not that bad ...
2020-06-07 14:39:15 <feuerteufel> Get a running sys after did somthing wrong :-))
2020-06-07 14:42:10 <feuerteufel> It's my first rpi running AL on, I'm playing and try to learn ...
2020-06-07 14:43:38 <trfl> ok so if you have lxdm autologin enabled, xfce breaks in various ways on 3.12
2020-06-07 14:44:16 <trfl> even with autologin disabled, if you log in too quickly, you get stuck in misc broken states where either desktop doesn't load, keyboard doesn't work, autostart apps don't start, or logout breaks with "unexpected message type"
2020-06-07 14:44:28 <trfl> looks like if I let lxdm sit for 10sec before logging in, everything works
2020-06-07 15:17:26 <mps> MY-R: any wishes for linux-edge 5.7.1 (except IO scheduler change :) )
2020-06-07 15:17:29 <mps> https://lore.kernel.org/lkml/1591530257663@kroah.com/
2020-06-07 15:35:46 <ikke> GKH doesn't leave you much choice "All users of the 5.7 kernel series must upgrade." :D
2020-06-07 15:36:21 <mps> as always :)
2020-06-07 15:37:46 <winny> Good thing I never upgrade except for impactful Caves for every two weeks ;)
2020-06-07 15:37:54 <winny> CVEs
2020-06-07 15:37:58 <ikke> haha
2020-06-07 15:38:05 <ikke> I was wondering what you were talking about :)
2020-06-07 15:38:21 <winny> My phone is brilliant. Has its own ideas
2020-06-07 16:09:01 <MY-R> mps, working drm debug? :\
2020-06-07 16:10:04 <MY-R> I dont even know what frequencies my GPU is running and now discovered that it can use performance mode because wrongly detect amount of connected monitors...
2020-06-07 16:11:38 <MY-R> I dont have path like "/sys/kernel/debug/dri/" and not sure which kernel config need to enable
2020-06-07 16:12:15 <mps> MY-R: debugfs as first
2020-06-07 16:12:30 <MY-R> ye
2020-06-07 16:30:12 <mps> I lost capitaine cursors in firefox after upgrade to 3.12
2020-06-07 16:30:30 <mps> on other apps they work as before upgrade
2020-06-07 16:32:19 <MY-R> xorg or wayland?
2020-06-07 16:32:48 <mps> xorg
2020-06-07 16:33:12 <MY-R> problem with cursors I had only in wayland
2020-06-07 16:33:20 <mps> I'm not a hipster to use wayland, I think :)
2020-06-07 16:33:42 <MY-R> did you try set it up by "lxappearance" app?
2020-06-07 16:33:55 <mps> see you after succesfull reboot to 5.7.1
2020-06-07 16:34:02 <MY-R> gl! :D
2020-06-07 16:34:55 <mps> I'm back
2020-06-07 16:35:22 <mps> 5.7.1-1-gru #2 SMP PREEMPT Sun Jun 7 16:03:58 UTC 2020 aarch64 GNU/Linux
2020-06-07 16:36:27 <mps> 'gru' is gru-kevin chromebook, in uname
2020-06-07 16:36:33 <MY-R> neat!
2020-06-07 16:36:36 <artok> you build kernel as often as I did back in -97
2020-06-07 16:37:18 <MY-R> me brb, need boot some *ubuntu to check wth is going on with my gpu clocks...
2020-06-07 16:37:19 <mps> artok: at these time kernels is not released so fast as today
2020-06-07 16:37:46 <mps> nowadays*
2020-06-07 16:38:18 <mps> and I build kernels from around ~93
2020-06-07 16:53:19 <artok> I stopped that build thingie after getting mac as my main computer
2020-06-07 17:03:20 <MY-R> mps, looks like my gpu clocking up and down just correct but still would be cool to have access to that info :>
2020-06-07 17:08:28 <mps> MY-R: I think distro kernels shouldn't be built debuging but for perfomance and size, with little more security as bonus of that
2020-06-07 17:08:45 <MY-R> jesus, and now firefox, after last update from time to time tottaly frozen and only can kill it...
2020-06-07 17:08:49 <mps> actually any distro package
2020-06-07 17:09:16 <mps> users should post bug reports
2020-06-07 17:09:21 <ikke> The smallest and most performant program does nothing at all
2020-06-07 17:09:51 <mps> but developers have to build with debug and test
2020-06-07 17:10:12 <mps> ikke: are you sure? :P
2020-06-07 17:10:53 <MY-R> thought linux-edge is to test things :D
2020-06-07 17:10:54 <mps> following this reasoning javascript is best
2020-06-07 17:11:05 <mps> MY-R: no, it is not
2020-06-07 17:11:25 <mps> it is made to be usable and to be fast
2020-06-07 17:11:51 <MY-R> I wanted just check what clocks got my GPU and that was the only way
2020-06-07 17:12:03 <mps> someone wrote wrongly that linux-edge for testing purpose, but that is not true
2020-06-07 17:12:11 <artok> would be funny to see how much space would be used when whole distro would be with debug information =D
2020-06-07 17:12:27 <mps> artok: :)
2020-06-07 17:14:02 <mps> we want linux-edge as kernel with new features missing on linux-lts, for users who need it
2020-06-07 17:14:45 <mps> cu, guests on the door
2020-06-07 17:15:04 <ikke> o/
2020-06-07 17:17:37 <Hello71> https://git.alpinelinux.org/aports/tree/main/linux-lts/config-lts.x86_64: CONFIG_DEBUG_FS=y
2020-06-07 17:17:51 <Hello71> https://git.alpinelinux.org/aports/tree/main/linux-lts/config-virt.x86_64: CONFIG_DEBUG_FS=y
2020-06-07 17:19:31 <Hello71> MY-R: you only need to mount -t debugfs debugfs /sys/kernel/debug...
2020-06-07 17:21:21 <MY-R> Hello71, https://git.alpinelinux.org/aports/tree/testing/linux-edge/config-edge.x86_64#n8364
2020-06-07 17:22:02 <Hello71> heh, blame mps then
2020-06-07 17:24:09 <MY-R> no blame but I was expecting that kernel configs wont be differ so much but now I know why...
2020-06-07 18:19:29 <fc7> hi there, I have a quick question regarding apk. Is there any way to list installed packages and the repository they came from (ie: community, main, edge, etc)?
2020-06-07 18:19:55 <fc7> I searched around but couldn't find anything
2020-06-07 20:09:41 <mps> MY-R: where do you see that default IO SCHED is set to NONE
2020-06-07 20:10:52 <MY-R> mps, cat /sys/block/sd*/queue/scheduler
2020-06-07 20:11:46 <MY-R> since kernel 5 they made some changes about that
2020-06-07 20:13:43 <MY-R> maybe if mq-deadline would be compiled in kernel then it would pick it up but in linux-edge isnt and only available one is BFQ
2020-06-07 20:14:24 <mps> hmm, I'm not sure, there is not default in kernel config
2020-06-07 20:16:01 <mps> there are mq-deadline and kyber-iosched modules which can be loaded and set to be used for devices
2020-06-07 20:17:06 <mps> BFQ is already built-in
2020-06-07 20:17:54 <mps> and it is not selected by default, so user/distributor have to set it
2020-06-07 20:18:07 <MY-R> yep, so Im loading mq-deadline and then changing it in /etc/local.d/, didnt want play with udev rules for now...
2020-06-07 20:19:02 <MY-R> ye, in Arch linux I saw some rules which can add based on if it is HDD or any other
2020-06-07 20:19:22 <MY-R> not sure but I think in Arch default was mq-deadline
2020-06-07 20:21:12 <MY-R> https://git.archlinux.org/svntogit/packages.git/tree/trunk/config?h=packages/linux#n948  all compiled in
2020-06-07 20:21:39 <mps> ideally installer should test destination drive speed with different scheduler and offer user to select
2020-06-07 20:22:16 <MY-R> well, only way for HDD's is now BFQ and all rest can be NONE or others which left
2020-06-07 20:22:27 <mps> but then installation will be error prone and slow
2020-06-07 20:23:30 <MY-R> yep
2020-06-07 20:23:51 <mps> well, cat /sys/block/sd*/queue/scheduler => [none] bfq kyber mq-deadline
2020-06-07 20:24:28 <MY-R> but ye, udev rule even on Arch need to be added manualy but still pretty sure it picked up automaticly mq-deadline for my hdd and ssd
2020-06-07 20:24:32 <MY-R> https://wiki.archlinux.org/index.php/Improving_performance#Changing_I/O_scheduler
2020-06-07 20:25:13 <mps> hmm, will read later (this year) :)
2020-06-07 20:25:46 <MY-R> not much to read, just a example of udev rule
2020-06-07 20:25:53 <mps> ah, so
2020-06-07 20:26:38 <mps> but anyway, we can add somewhere in boot scripts or config for users to change it according their need and will
2020-06-07 20:26:53 <mps> let me try bfq
2020-06-07 20:27:24 <MY-R> is more about what is safer, none or anything else
2020-06-07 20:28:04 <mps> I read in last times a lot praises for BFQ
2020-06-07 20:29:02 <MY-R> since there is no single queue schedulers by default in kernel 5... BFQ is only option for HDDs
2020-06-07 20:30:14 <MY-R> and I was always happy more from CFQ...
2020-06-07 20:31:38 <mps> but it is retired
2020-06-07 20:31:48 <MY-R> yep
2020-06-07 20:42:46 <mps> hmm, what we got in kernel 5.7 'UACCE provides interface for the user process to access the hardware without interaction with the kernel space in data path'
2020-06-07 20:43:20 <mps> so, finally device driver in userspace :)
2020-06-07 21:59:40 <mps> ikke: https://cdn.kernel.org/pub/linux/kernel/v5.x/ChangeLog-5.6.17 , search cve in it. This time GKH is serious with 'all must upgrade' :)
2020-06-07 22:12:14 <winny> if i want to use a package in testing, is there a way to tell apk i only want one specific package, and don't install anything else from testing?
2020-06-07 22:13:56 <mps> winny: read about repo pinning in wiki
2020-06-07 22:15:46 <winny> thanks, keywords accelerate research B-)
2020-06-07 22:40:52 <Roenbaeck> I made a rather extensive guide when setting up Alpine 3.12 on a Raspberry Pi 4B, with a FAT32 boot partition and an additional ext4 partition for the system on the SDHC card. It also covers how to get fanshim working, macOS-friendly samba sharing, Time Machine, Cronicle and other stuff. Perhaps parts of it could be of interest for the wiki?
2020-06-07 22:42:31 <Roenbaeck> If links are allowed, you can find it here: http://www.anchormodeling.com/tinker-tailor-raspberry-pi/
2020-06-07 22:44:42 <mps> Roenbaeck: sure, it will be good to add to alpine wiki
2020-06-07 23:03:45 <pkx> https://git.alpinelinux.org/aports/tree/main/freeswitch/APKBUILD - I expect that if I run "apk add freeswitch-pgsql" I would end up with "/usr/lib/freeswitch/mod/mod_pgsql.so", but I don't. Instead, it looks like I get "/usr/lib/freeswitch/mod/mod_cdr_pg_csv.so", and freeswitch won't start because mod_pgsql is a core module
2020-06-07 23:04:59 <pkx> The actual startup error: [CRIT] switch_loadable_module.c:1785 Error Loading module /usr/lib/freeswitch/mod/mod_pgsql.so
2020-06-08 00:40:14 <lunaphyte> i just happened to notice i don't seem to have /dev/disk/.  is that normal?
2020-06-08 01:28:22 <Hello71> it is populated by udev
2020-06-08 01:29:25 <Hello71> lunaphyte: you can install eudev or enable mdev. latter may require some configuration, idk
2020-06-08 02:06:53 <lunaphyte> Hello71: ah, ok.  thanks.
2020-06-08 02:07:16 <Hello71> I've only ever used the modprobe functionality of mdev
2020-06-08 02:07:34 <Hello71> ... also tbh I'm not sure if eudev comes with /dev/disk rules, but I think it does
2020-06-08 04:51:28 <rnkn> dear oracles of alpine, I've done a sys install on an SD card (Rpi4) but given the anecdotal failure rate of SD cards, am I better to mount /home via a USB stick?
2020-06-08 04:52:11 <rnkn> (also, damn via your nick must give you some annoyances)
2020-06-08 05:41:26 <albr> McBoat[d]: You're not getting anywhere without a / volume even if your /home is intact. Better to just make proper backups imho
2020-06-08 05:42:09 <ikke> McBoat, / is usually easier to restore than the contents of your home dir
2020-06-08 05:48:45 <albr> McBoat[d]: I mean that USB sticks can fail too, so it's better to have backups than to create a false sense of security by separating your /home
2020-06-08 06:57:11 <Tuxist_> i try modify ceph-deploy to run with alpine linux but there are no init files there
2020-06-08 07:18:21 <Trieste> hi, I'm having trouble automating launching X applications; if I ssh to the machine as my regular user, set DISPLAY to ":0", and then manually run the app, it works - but if I use supervisord to run the exact same sequence of commands, as my user, I get "No protocol specified \n Unable to init server: Could not connect: Connection refused \n Error: cannot open display: :0"
2020-06-08 07:18:40 <Trieste> any pointers, what could I be looking into?
2020-06-08 07:29:54 <Cogitri> I guess because it uses a different user you don't have access to the XServer
2020-06-08 07:30:10 <Cogitri> You could use XAuthority or xhost to fix that
2020-06-08 07:33:37 <Trieste> Cogitri: it should be using the same user, unless I'm misunderstanding you?
2020-06-08 07:34:04 <Trieste> anyway, will look into XAuthority & xhost :)
2020-06-08 07:38:00 <Trieste> right, it seems that both solve the issue of running X apps as root - but I'm not running them as root (directly)
2020-06-08 10:47:29 <yunfan> mps: i had just  tested, that by install alpine 3.12 , my pcduino3 nano's reboot works
2020-06-08 10:47:42 <yunfan> thanks for the wonderful job
2020-06-08 10:47:51 <yunfan> now my k3s has a new node
2020-06-08 11:05:51 <mps> yunfan: you are welcome
2020-06-08 11:07:10 <mps> would be nice if have some time to describe or add note about install on this board so we can have it as confirmation that alpine works on pcduino3
2020-06-08 11:13:23 <yunfan> mps: there's nothing special :D
2020-06-08 11:14:29 <yunfan> mps: now i wonder if there were tools like raspAP on raspbian
2020-06-08 11:15:23 <yunfan> i want my rpi zero w to act as a portable wifi ap ,which could always have a AP open and user could connect to it then configure its upstream wifi router
2020-06-08 11:15:40 <yunfan> after that, i could add some proxy configuration
2020-06-08 11:17:24 <mps> I don't know what is raspAP
2020-06-08 11:21:10 <yunfan> mps: its a web software like those openwrt's luci suits
2020-06-08 11:21:20 <yunfan> which could configure the host's wifi
2020-06-08 11:22:02 <mps> ah, I dislike such things :)
2020-06-08 11:22:34 <yunfan> well i just find i could use some tui tools
2020-06-08 11:22:50 <yunfan> so no web suits, but is there any tui tools for doing this?
2020-06-08 11:22:51 <mps> when I used openwrt (or build it locally) first thing is to remove all traces of it
2020-06-08 11:23:31 <yunfan> mps: a luci app f**k my router's configuration yesterday:[
2020-06-08 11:23:47 <mps> CLI tools, and some shell scripts
2020-06-08 11:24:29 <mps> CLI is most user friendly interface
2020-06-08 11:31:27 <omni> openwrt has uci for the shell
2020-06-08 11:31:50 <omni> don't forget to 'uci commit' changes
2020-06-08 11:38:36 <mps> omni: yes, uci is what I used, besides of vi and cli on openwrt
2020-06-08 11:46:17 <yunfan> omni: but some luci-app cause my uci commit failed
2020-06-08 11:59:05 <markand> anyone having firefox/chromium to crash when using google meet?
2020-06-08 11:59:18 <markand> tab crash systematically after several seconds
2020-06-08 11:59:52 <Cogitri> I'm not using Google Meet, but FWIW FF hasn't crashed on me for months
2020-06-08 12:02:07 <markand> :(
2020-06-08 12:04:59 <markand> Cogitri, you're running edge isn't it?
2020-06-08 12:05:49 <mps> markand: FF is stable also for me of the 3.9 and now on edge
2020-06-08 12:06:09 <mps> though also I don't know and use Google Meet
2020-06-08 12:06:11 <markand> it only crashes when using google meet
2020-06-08 12:06:18 <markand> other than that it's perfectly stable
2020-06-08 12:06:23 <markand> [18103.687942] IncomingVideoSt[9748]: segfault at b8 ip 00007f19bcb4b0f6 sp 00007f19aba96b08 error 6 in ld-musl-x86_64.so.1[7f19bcb10000+47000]
2020-06-08 12:07:34 <Cogitri> markand: I use edge, yes
2020-06-08 12:07:53 <Cogitri> And that message isn't really useful, I'd need a complete backtrace to do anything but guess :)
2020-06-08 12:09:27 <markand> yes obviously
2020-06-08 12:09:42 <markand> I'll upgrade to edge as I'm a packager anyway
2020-06-08 12:10:51 <Cogitri> There's no shame in using stable :P
2020-06-08 12:16:08 <russkel> have no toxic rolling release attitudes in here
2020-06-08 12:16:15 <russkel> thanks
2020-06-08 12:56:05 <yunfan> err, my rpi wont output via ttyAMA0 to the host after i install 3.12.0, it used to work when 3.11 mps something changed?
2020-06-08 13:01:11 <Tuxist_> :-( now i want repackage ceph and than this https://tuxist.de/files/slowdown.png
2020-06-08 13:03:04 <Tuxist_> have they c64 based server download server ? ^^
2020-06-08 13:47:44 <mps> yunfan: for me it is still ttyAMA0
2020-06-08 13:49:17 <mps> and, what's with you? you have a lot of A20 boards and you playing with RPi :D
2020-06-08 13:49:52 <yunfan> mps: yes i had many allwinner chip inside boards
2020-06-08 13:49:56 <yunfan> and 2 rpi zero w
2020-06-08 13:50:47 <yunfan> mps: because our greate fire wall, i need a portable proxy agent, rpi zero w is very suitable for this task, before that, i had considered esp32
2020-06-08 13:52:20 <mps> aren't there A20 small boards like RPi zero w on the market?
2020-06-08 13:52:38 <yunfan> mps: i havnt found one small like rpi zero w
2020-06-08 13:52:49 <CcxWrk> robi
2020-06-08 13:52:57 <mps> (I have impression that in China is possible to buy everthing, and little more)
2020-06-08 13:52:59 <yunfan> btw, i had used to found a wifi sdcard reader which is smaller than rpi zero w
2020-06-08 13:53:05 <CcxWrk> Oops, typo.
2020-06-08 13:53:07 <yunfan> and have 128M ram
2020-06-08 13:53:42 <mps> ok
2020-06-08 13:53:48 <yunfan> there're some poland hackers used to study on this, its perfect except its too small for connect to ttl
2020-06-08 13:54:11 <yunfan> so rpi zero w is the best i had so far
2020-06-08 13:54:17 <mps> but 3.12 on RPi zero w on my desk uses ttyAMA0 without problem
2020-06-08 13:54:58 <yunfan> mps: do you have any special configuare in cmdline.txt or usercfg.txt ?
2020-06-08 13:55:21 <yunfan> and my sd card contains 2 partition, 1 is vfat another is ext4
2020-06-08 13:55:23 <mps> hmm, give me few minutes
2020-06-08 13:55:33 <yunfan> thx
2020-06-08 13:56:08 <yunfan> also i use armhf version
2020-06-08 13:57:30 <mps> yunfan: here is script I use to flash alpine release to sd card for rpi zero w, https://tpaste.us/9Nll
2020-06-08 13:59:35 <Hello71> I think sfdisk is superior to parted
2020-06-08 13:59:53 <CcxWrk> Not if you want scripted interaction.
2020-06-08 14:00:01 <Hello71> ...
2020-06-08 14:00:08 <yunfan> mps: thx, so do you use a extra power supply or ttl 5v?
2020-06-08 14:00:15 <Hello71> >sfdisk is a script-oriented tool
2020-06-08 14:00:23 <Hello71> maybe you are thinking of fdisk.
2020-06-08 14:01:33 <mps> yunfan: no, for power I use usb serial ttl adapter, attaching 5V wire to RPi
2020-06-08 14:01:43 <mps> no separate power supply
2020-06-08 14:02:37 <mps> usb serial adapter is prolific, pl2303
2020-06-08 14:02:43 <yunfan> ok will do that again,
2020-06-08 14:03:02 <yunfan> and i think your script's content is almost same like the wiki page shows
2020-06-08 14:03:24 <mps> well, yes, I picked parameters from wiki
2020-06-08 14:03:28 <yunfan> maybe the problem is about those commented lines in your script, i will check my extlinux.conf
2020-06-08 14:04:19 <mps> RPi by default doesn't use u-boot, so extlinux.conf is useless
2020-06-08 14:04:37 <yunfan> ok
2020-06-08 14:05:05 <mps> these commented lines when I installed it in qemu
2020-06-08 14:05:38 <mps> i should remove them, thanks for notice
2020-06-08 14:05:45 <yunfan> i think i might found a relevant issue, my sdcard's vfat partitions dont have boot flag
2020-06-08 14:06:06 <mps> heh, could be
2020-06-08 14:07:05 <mps> Hello71: for some time sfdisk didn't worked for me and I switched to parted, and after that didn't looked again on sfdisk
2020-06-08 14:07:11 <CcxWrk> Hello71: Sorry, confused it with cfdisk there. sfdisk has been lacking support for anything but MBR for a long time, I haven't properly investigated it since.
2020-06-08 14:07:27 <Hello71> gnu fdisk, not busybox fdisk
2020-06-08 14:07:29 <Hello71> busybox fdisk sucks
2020-06-08 14:08:09 <yunfan> is there any forth software in alpine's official repo?
2020-06-08 14:08:13 <CcxWrk> Parted is very flexible in relative positioning and different units. Not sure if sfdisk has that. OTOH for just snapshotting and restoring layout sfdisk is clearly the winner.
2020-06-08 14:09:52 <mps> yunfan: gforth :)
2020-06-08 14:10:08 <Hello71> how many options do you need? sfdisk has all fdisk options: end of device, fixed size in Ki/Mi/Gi/..., offset in Ki/Mi/Gi/...
2020-06-08 14:10:13 <yunfan> that's quite large
2020-06-08 14:10:39 <mps> yunfan: you know better one for distro?
2020-06-08 14:11:01 <mps> I'm forth fan
2020-06-08 14:11:18 <yunfan> mps: well, i dont have choices, for there were about 2 forth for every user :D
2020-06-08 14:11:37 <Hello71> according to the apocalypse os people you are supposed to write your own :p
2020-06-08 14:11:49 <mps> well, every forth programmer have to make its own :)
2020-06-08 14:12:03 <yunfan> mps: wow, so what do you think of eforth?
2020-06-08 14:12:22 <mps> didn't tried it, so don't know
2020-06-08 14:13:00 <yunfan> my goal is still not archived
2020-06-08 14:13:04 <mps> Hello71: and I wrote my forth implementation long time ago, for z80 CPU to run from ROM
2020-06-08 14:13:16 <yunfan> i used to made a pyforth, but that's too low
2020-06-08 14:13:36 <yunfan> mps: ltsy-forth is cool to me
2020-06-08 14:13:39 <mps> and extensible by SRAM, battery backup
2020-06-08 14:14:30 <yunfan> i used to read jeforth's article explaining how then reduce the pritimives needed for their own implementation
2020-06-08 14:14:48 <yunfan> but i hope there were low level impl than js
2020-06-08 14:14:50 <mps> I made trimmed version for z80, but enough for machine control (now people call it robots)
2020-06-08 14:15:41 <yunfan> i hope oneday i could use forth in backend service
2020-06-08 14:16:11 <mps> but in last 20 years I didn't made anything useful in forth, just pushed gforth to alpine to play sometimes with it
2020-06-08 14:19:28 <yunfan> mps: why you dont use it anymore?
2020-06-08 14:21:06 <mps> well, I had to work on different fields to earn money for living, and to use perl, java, go etc
2020-06-08 14:21:21 <mps> didn't had time for forth
2020-06-08 14:21:49 <mps> and didn't had where I can implement it for bussines applications/systems
2020-06-08 14:24:40 <yunfan> ok
2020-06-08 14:25:00 <yunfan> i there thinking to use forth replace lua's role on many server side services
2020-06-08 14:25:12 <yunfan> like redis and nginx
2020-06-08 14:26:08 <mps> sometimes I think to write 'bios' in forth, like it was on some old SUN machines
2020-06-08 14:26:55 <mps> but also I know that I'm nowadays lazy for that
2020-06-08 14:27:44 <yunfan> that;s openfirmware
2020-06-08 14:28:03 <mps> and thought to make linux kernel forth module to access machine on low level
2020-06-08 14:28:19 <yunfan> if you prefer bios stuff, why not turn to android cellphone's ?
2020-06-08 14:28:30 <mps> yes, openfirmware, and I think it is used on old apple machines
2020-06-08 14:28:34 <yunfan> these were very popular nowaday
2020-06-08 14:28:53 <yunfan> people need a bootable tool for flash their cellphone's main OS
2020-06-08 14:30:00 <mps> I'm lazy, and prefer to waste time on reading poetry, prose, phylosophy, history, religion etc
2020-06-08 14:30:12 <yunfan> :D
2020-06-08 14:30:20 <yunfan> i like reading history too
2020-06-08 14:30:27 <mps> and wasting time on alpine distro :)
2020-06-08 14:31:42 <mps> yunfan: you know who was 'Da Mo'?
2020-06-08 14:32:11 <yunfan> mps: you mean that finance company?
2020-06-08 14:32:24 <yunfan> aha, i got it
2020-06-08 14:32:37 <yunfan> its buddast
2020-06-08 14:32:51 <mps> Bodhidharma, Japanese calls him
2020-06-08 14:33:12 <yunfan> of course i knew, he were the chinese shaolin temple's founder :D
2020-06-08 14:33:30 <mps> yes, and man who lived 182 years
2020-06-08 14:33:46 <mps> he was also Indian prince
2020-06-08 14:33:47 <yunfan> well, i were hightly dounbt on this part
2020-06-08 14:33:55 <yunfan> i mean the age
2020-06-08 14:34:47 <mps> well, his birthdate is recorded in India, and time of his death also recorded in Chinese anals
2020-06-08 14:35:03 <yunfan> well , from a modern review, i were attracted by another india religion, jainism
2020-06-08 14:35:28 <yunfan> but i am not a believer, i just found its rules interesting and much more modern in its time
2020-06-08 14:36:03 <mps> hmm, also I'm not religious in common meaning
2020-06-08 14:36:54 <tehcloud> yunfan, how about more Western forms of thought, such as the Discourses of Epictetus?
2020-06-08 14:37:35 <mps> imo, west is behind east
2020-06-08 14:37:52 <mps> though not much
2020-06-08 14:38:09 <yunfan> tehcloud: i dont know much more about epictetus, i had read a book about philosophy, Sophie's World
2020-06-08 14:38:10 <tehcloud> depends on what angle you want to argue
2020-06-08 14:38:31 <mps> tehcloud: sure
2020-06-08 14:38:51 <tehcloud> Rome's state religion had no room for philosophy or moral teachings, so there is a greater separation between philosophy and mythology due to that
2020-06-08 14:38:57 <mps> and I don't want to argue about these
2020-06-08 14:39:06 <yunfan> and i found the story about "The Ship of Theseus" is really interesting and have much more meaning to help us thinking of AI
2020-06-08 14:40:07 <yunfan> also that story might leads us a way to trans human mind to cloud
2020-06-08 14:40:35 <tehcloud> lol, that's what the minimalist, container-oriented Alpine distro is helping accomplish ;)
2020-06-08 14:40:53 <yunfan> :D
2020-06-08 14:41:26 <tehcloud> although it has other great uses beyond that
2020-06-08 14:41:58 <tehcloud> I developed my latest project mostly on Alpine, and discovered that since it forced me to write proper standards-compliant C, my code just works on the most exotic of platforms effortlessly
2020-06-08 14:43:09 <Cogitri> Not using GNU C extensions improves portability, yes
2020-06-08 14:43:11 <yunfan> tehcloud: i would prefer to use nim
2020-06-08 14:44:10 <tehcloud> I don't do hipster languages, but I may try Rust one day :p
2020-06-08 14:44:32 <yunfan> yes, rust looks like the future
2020-06-08 14:44:57 <yunfan> even google and ms were turningto it
2020-06-08 14:45:02 <mps> tehcloud: hipster lang, nice :)
2020-06-08 14:56:49 <yunfan> mps: still no work, after follow your script.
2020-06-08 14:57:19 <yunfan> which fs type should i tag on the first vfat partition? i use w95 vfat lba currently
2020-06-08 14:59:53 <mps> here it what fdisk shows to me: /dev/sdc1  *      2048   526335   524288  256M  c W95 FAT32 (LBA)
2020-06-08 15:01:12 <mps> parted shows:  1      1049kB  269MB   268MB   primary  fat16        boot, lba
2020-06-08 15:01:19 <yunfan> that's the same with me
2020-06-08 15:01:52 <yunfan> will check it again when i got a hdmi display
2020-06-08 15:02:08 <mps> what is the part table type, dos or gpt
2020-06-08 15:02:12 <yunfan> i am doubt on gpio output problem
2020-06-08 15:02:25 <yunfan> mps: its dos
2020-06-08 15:02:41 <yunfan> and the rpi dont blink any of its led
2020-06-08 15:02:43 <mps> hmm, everything as on my card
2020-06-08 15:03:05 <mps> ah, then you have probably power problem
2020-06-08 15:04:09 <mps> or bad mmc card slot
2020-06-08 15:04:13 <yunfan> i dont think so, because it works before i do upgrading
2020-06-08 15:04:27 <yunfan> same power suply and sdcard
2020-06-08 15:07:33 <yunfan> anyway, i gave up,
2020-06-08 15:08:20 <mps> one interesting note, looks like it needs ttyAMA0 for boot loader, but when is booted it shows that the serial console is ttyS0
2020-06-08 15:09:23 <yunfan> aha, try this
2020-06-08 15:11:15 <mps> just now I reflashed my sd card with script I posted to you to check again, and it worked out-of-the-box
2020-06-08 15:15:05 <yunfan> still not work after change the ttyAMA0 to ttyS0 , but the board's hot near usb power supply, so i guess its still work
2020-06-08 15:15:20 <yunfan> ok  quit again :[
2020-06-08 15:15:43 <markand> I also have a problem with my pi after upgrade to 3.12
2020-06-08 15:15:49 <markand> kernel boots but init does not
2020-06-08 15:18:50 <mps> markand: which model
2020-06-08 15:19:05 <markand> 4
2020-06-08 15:19:33 <mps> last night someone posted url here how it installed alpine 3.12 on rpi4
2020-06-08 15:20:12 <mps> http://www.anchormodeling.com/tinker-tailor-raspberry-pi/
2020-06-08 17:11:50 <kmmndr> Hi all :-)
2020-06-08 17:14:06 <kmmndr> I've installed kolourpaint (without kde), and the tools menu are iconless
2020-06-08 17:15:23 <kmmndr> I guess it is missing stuff installed by kde groups
2020-06-08 17:15:38 <kmmndr> Any idea ?
2020-06-08 17:49:27 <monado> kmmndr set your qt icons?
2020-06-08 17:50:55 <kmmndr> hum, my question was probably a kde noobs question :-)
2020-06-08 17:51:38 <kmmndr> Isn't it a default qt icon set ?
2020-06-08 17:52:59 <monado> Not sure, don't really use qt myself
2020-06-08 17:53:25 <monado> I could try to source a wiki article for you
2020-06-08 17:53:59 <monado> https://wiki.archlinux.org/index.php/Uniform_look_for_Qt_and_GTK_applications
2020-06-08 17:54:02 <kmmndr> I'll try adding every *icon* packages till I find one matching :-)
2020-06-08 17:55:52 <kmmndr> But the real question is if we could add a "default" theme as dependency for kolourpaint and probably all kde applications
2020-06-08 17:56:56 <kmmndr> I got it "breeze-icons" solved the problem
2020-06-08 17:57:35 <kmmndr> May "breeze-icons" be a depency of "kolourpaint" ?
2020-06-08 18:02:11 <mps> kmmndr: alpine uses minimum dependencies principle
2020-06-08 18:03:24 <ikke> But some default theme would make a lot of sense
2020-06-08 18:04:07 <ikke> otherwise all applications would look very broken
2020-06-08 18:04:20 <mps> sometimes I think so, but after some more thinking I think no
2020-06-08 18:05:06 <mps> right now I'm fighting with dark theme and cursors
2020-06-08 18:05:36 <mps> after upgrade to 3.12 all is broken
2020-06-08 18:06:00 <mps> but I think we should not have any default
2020-06-08 18:06:14 <kmmndr> We could add a tip on stdout when installing any kde application: please install kde theme
2020-06-08 18:06:38 <mps> this makes sense
2020-06-08 18:06:40 <monado> Add it to the documentation :)
2020-06-08 18:07:10 <mps> but we don't 'write to console' during installing upgrading
2020-06-08 18:08:58 <mps> actually, I'm not against adding defaults and deps for some pkgs, but before that we should discuss this
2020-06-08 18:10:18 <mps> we should clearly say to developers and contributors 'no more minimal dependencies' principle is in rule
2020-06-08 18:16:49 <Hello71> imo this can be solved by virtual package "kde-icons" or something. if someone manually installed their own icons then they can create a dummy provider package
2020-06-08 18:17:03 <Hello71> my-icons-0-r0
2020-06-08 18:20:56 <mps> sure, but users coming from other distros expect that alpine is like other distros
2020-06-08 18:22:27 <mps> and actually I think it is time to decide will we keep alpine 'small, simple, secure' or 'user friendly', both can't be achieved at same time
2020-06-08 18:30:32 <Hello71> I think there are more than enough decent-quality distros for inexperienced to moderately experienced users: debian, fedora, arch... compare alpine iso size to arch. arch has all firmware, man pages, headers, etc
2020-06-08 18:32:58 <mps> heh, you absolutely right
2020-06-08 18:33:58 <tehcloud> best to keep Alpine a vanilla upstream distro and if someone wants to make a Ubuntu-style hack they are free to do so
2020-06-08 18:34:01 <mps> but we cannot say to inexperienced users 'go there'
2020-06-08 18:34:26 <tehcloud> user-friendly is a deep rabbit hole
2020-06-08 18:35:44 <mps> tehcloud: that is what i thought, i.e. keep base alpine small and simple, and someone can use it as base and add features, like pmOS
2020-06-08 18:36:14 <tehcloud> it is bound to happen to any popular distro anyway
2020-06-08 18:36:23 <Hello71> imo there's little value in a "but with" distro, e.g. "debian but with openrc". alpine should not do "arch but with musl"
2020-06-08 18:36:27 <tehcloud> Debian is usable enough in vanilla form but we have Ubuntu, and further hacks of Ubuntu
2020-06-08 18:43:36 <kmmndr> Woow, I started a longer debate
2020-06-08 18:49:30 <kmmndr> I do love alpine for its values. By using virtual packages, like "kde-icons", it could stay small and only suggest to add icons manually
2020-06-08 18:49:52 <tehcloud> I would say do something like that with fonts first ;)
2020-06-08 18:50:25 <kmmndr> That's a good idea too ! :-)
2020-06-08 20:04:59 <PHANI> @ikke
2020-06-08 20:25:42 <oliv3r[m]> Hey guys, I have a 'legal'-ish question here. Lets say I want to package a statically compiled piece of software. It simply is impossible to use dynamically. Now it uses static version of openssl or mbedTLS if you will. What do I put in the 'licenses' field of APKBUILD? 'just' the AND part enough? Can I have some freedom in saying openssl:GPL-2.0-or-later for example?
2020-06-09 00:49:41 <lunaphyte> i've done an install by hand, so to speak, and when the system boots, it gets stuck after printing "Loading initial ramdisk...'
2020-06-09 00:49:57 <lunaphyte> what can i do to troubleshoot what it's stuck on?
2020-06-09 00:51:16 <Hello71> remove quiet from command line
2020-06-09 00:52:50 <lunaphyte> i did that.  here's the current command line:
2020-06-09 00:53:45 <lunaphyte>  /boot/vmlinuz-virt root=/dev/mapper/vg1-root ro  modules=sd-mod,usb-storage,btrfs rootfstype=btrfs console=ttyS0,115200n8 console=tty0
2020-06-09 00:55:47 <lunaphyte> here's what the serial console prints: http://dpaste.com/1X2HRRQ
2020-06-09 00:56:13 <lunaphyte> that's as far as it seems to get
2020-06-09 01:04:33 <Hello71> maybe try removing console=tty0 if you want to use serial
2020-06-09 01:04:42 <Hello71> I forget the order of precedence
2020-06-09 01:11:21 <lunaphyte> the last console defined is used as tty0
2020-06-09 01:12:18 <lunaphyte> i don't want to use the serial port for the main console, it's just there as a debugging aid
2020-06-09 01:14:27 <Hello71> stuck after loading ramdisk can be caused by many things
2020-06-09 01:22:14 <Hello71> most likely your graphics driver is not working and also you didn't uncomment ttyS0 in /etc/inittab
2020-06-09 01:29:00 <lunaphyte> but without ttyS0 active in inittab, i wouldn't have seen the console output via the serial port that i put in the pastebin, right?
2020-06-09 01:53:29 <Hello71> that is kernel output. it is controlled by console kernel parameter
2020-06-09 01:54:16 <Hello71> if you reorder the consoles you will probably also get openrc output
2020-06-09 02:00:24 <Hello71> lunaphyte:
2020-06-09 02:13:43 <lunaphyte> oh, ok.  i see what you're saying.  let me give that a try
2020-06-09 02:20:13 <Hello71> it's a little bit confusing. kernel outputs given by console=, input from /dev/console, which is defined by last console= parameter. then, kernel execs init </dev/console >/dev/console. init then spawns gettys for each terminal listed in /etc/inittab
2020-06-09 02:20:40 <Hello71> so it's a mess of multiple outputs and only one input, then only one input and output, then multiple inputs and outputs
2020-06-09 02:20:48 <Hello71> all controlled by slightly different things
2020-06-09 02:22:23 <lunaphyte> thanks.
2020-06-09 02:22:25 <lunaphyte> mount: mounting /dev/mapper/vg.03bf2c6e.42d1.42c2.8a6d.3f6d8c00b362-root on /sysroot failed: No such device
2020-06-09 02:22:30 <lunaphyte> there's a good clue
2020-06-09 02:26:37 <Hello71> probably, although I don't think that will fix your graphics driver
2020-06-09 02:27:45 <Hello71> you see that if initramfs never execs real init though, and no gettys are spawned, then you will only get rescue console on tty0 (with your command line). that's why I said that you can try reordering your console arguments
2020-06-09 02:35:00 <lunaphyte> yeah, thanks.  this was really helpful.
2020-06-09 02:35:12 <lunaphyte> i wonder what it is i'm doing differently than setup-disk does.
2020-06-09 02:35:48 <Hello71> dunno, I don't use lvm
2020-06-09 05:26:50 <remexre> is there something weird about how glfw is built? https://silo.remexre.xyz/screenshots/6519b991c589a932c469e9ecd4e4e625980828ff.png
2020-06-09 05:44:29 <IvanM5> Hi ikke, I solved the problem - allowed LANG in /etc/rc.conf: rc_env_allow="LANG" . Also reported here: https://gitlab.alpinelinux.org/alpine/aports/-/issues/11620
2020-06-09 05:45:39 <IvanM5> Not sure it is the optimal way for OS though.
2020-06-09 06:59:35 <PHANI> ?
2020-06-09 07:21:34 <rpk> • /msg nickserv identify rpk Rphani!@#$%
2020-06-09 07:22:29 <ikke> rpk: I would change your password
2020-06-09 07:29:21 <ikke> rpk: What is your issue? If you need help with something, just ask here, and if someone can help you, they will respond
2020-06-09 07:32:20 <rpk> i was trying to install Smokeping from https://wiki.alpinelinux.org/wiki/Setting_up_Smokeping  there seems to be URL which is missing and also can someone give me the latest how to smokeping in alpine.
2020-06-09 09:47:32 <ThinkDao_> hi all. what dhcp client alpine-linux have in the installation image?
2020-06-09 09:48:25 <ikke> udhcpc
2020-06-09 09:48:52 <ikke> which is part of busybx
2020-06-09 09:49:14 <ThinkDao_> doesn't work
2020-06-09 09:49:57 <ThinkDao_> send back a message saying "network is down" read error but no connection
2020-06-09 09:50:37 <ThinkDao_> I ask because the alpine-interface command was not working.
2020-06-09 09:52:06 <ThinkDao> it keeps cycling trying to connect but fail
2020-06-09 09:52:23 <ThinkDao> maybe I should use static right?
2020-06-09 09:52:25 <ikke> setup-interface you mean?
2020-06-09 09:52:36 <ThinkDao> yes
2020-06-09 09:52:46 <ikke> try ifup eth0
2020-06-09 09:54:09 <ThinkDao> same ( nothing change )
2020-06-09 09:56:22 <ThinkDao> ok.. fixed.. setup a static ip and add resolv.conf file with dns information. Now I can ping the outside world. Thanks
2020-06-09 09:56:48 <ikke> check /etc/network/interfaces
2020-06-09 09:58:46 <ThinkDao> all drivers are identified correctly.
2020-06-09 09:59:04 <ThinkDao> but somehow udhcp doesn't do the job
2020-06-09 10:00:22 <ThinkDao> in any case, problem is solved
2020-06-09 10:02:29 <ThinkDao> I think this iso I download have several problems, or alpine setup is like this.. I need to setup mirrors manually?
2020-06-09 10:03:01 <ThinkDao> I wish to install cryptsetup and it doesn't find
2020-06-09 10:04:12 <ThinkDao> yes.. setup-apkrepos ok.. find
2020-06-09 10:04:15 <ThinkDao> fine*
2020-06-09 10:05:59 <ThinkDao> working... is my first time setting up a alpine-linux
2020-06-09 12:19:31 <roooooot> hello happy people
2020-06-09 12:30:17 <afontain_> Hi
2020-06-09 13:25:13 <ThinkDao> next problem.. even in documentation ( wiki ) to install grub, the only option is grub with EFI. I am not using EFI so the command should work as "grub-install /dev/sda". but it fails, saying I need to specify --target ( not sure with what parameters ) with an error about /usr/lib/grub/i386-pc/modinfo.sh 'doesn't exist'
2020-06-09 13:27:25 <ThinkDao> if I don't use EFI I can only use syslinux ( since it is not so for security ) ?
2020-06-09 13:27:48 <ThinkDao> is not so *goog*
2020-06-09 13:27:54 <ThinkDao> good*
2020-06-09 13:52:51 <ThinkDao> damn.. i386-pc libs are not installed with installing "apk add grub"
2020-06-09 13:53:01 <ThinkDao> any way to solve this problem?
2020-06-09 13:54:12 <ThinkDao> ah ok.. the package is grub-bios ^.^
2020-06-09 13:54:53 <ThinkDao> solved
2020-06-09 14:25:53 <Roenbaeck> I added my guide to the wiki. I would suggest that at least the fix for setting the date during boot makes it into the next release? It's a simple addition of a line in chrony's config. See the guide here: https://wiki.alpinelinux.org/wiki/Raspberry_Pi_4_-_Persistent_system_acting_as_a_NAS_and_Time_Machine#Keeping_the_Time
2020-06-09 14:26:34 <mps> Roenbaeck: thanks
2020-06-09 14:31:04 <mps> Roenbaeck: this chrony part is specific to sys install?
2020-06-09 14:45:39 <feuerteufel> mps: You asked me to set a reminder :-))
2020-06-09 14:45:49 <feuerteufel> [x] done
2020-06-09 14:47:11 <mps> feuerteufel: ? 'my memory is fading away'
2020-06-09 14:48:29 <mps> firedevil: reminder about what?
2020-06-09 14:48:31 <feuerteufel> ncopa: I'm done with the alpine-releases.conf.yaml file, where you want me to put the files (file and patch file)?
2020-06-09 14:48:44 <mps> ah
2020-06-09 14:48:56 <mps> that should be on ncopa, not me
2020-06-09 14:49:03 <feuerteufel> Abaut the not working kernel update on rpi
2020-06-09 14:49:11 <mps> aha
2020-06-09 14:49:26 <mps> now I understand
2020-06-09 14:49:29 <ncopa> feuerteufel: great! do you have a patch that I can `git am`?
2020-06-09 14:49:30 <mps> thanks
2020-06-09 14:50:07 <ncopa> feuerteufel: if you ahve it in git then you can: git format-patch -1 --stdout | tpaste
2020-06-09 14:50:08 <feuerteufel> I do have both
2020-06-09 14:50:57 <ncopa> or you can email me the patch if you prefer
2020-06-09 14:50:58 <feuerteufel> Oh, I'm sorry I don't know ho a git patch has to be
2020-06-09 14:51:17 <ncopa> then just: tpaste < file.patch
2020-06-09 14:51:27 <ncopa> and paste the url here
2020-06-09 14:51:30 <feuerteufel> It's only a normal "diff patch"
2020-06-09 14:51:43 <ikke> just use the command ncopa gave you
2020-06-09 14:51:43 <ncopa> that works
2020-06-09 14:51:55 <ncopa> apk add tpaste
2020-06-09 14:52:16 <ncopa> or simply paste the contents in the https://tpaste.us form
2020-06-09 14:52:33 <ncopa> or send it as attachment by email to ncopa at alpinelinux.org
2020-06-09 14:56:34 <feuerteufel> https://tpaste.us/4ElX
2020-06-09 14:56:34 <feuerteufel> https://tpaste.us/MbdP
2020-06-09 14:58:08 <ncopa> feuerteufel: thank you very much! that was awesome
2020-06-09 15:00:13 <ncopa> i see that you also fixed an error i made in the 3.8.5 release date. thanks!
2020-06-09 15:00:14 <feuerteufel> I send them also by mail so hopfully you'll have at least one of them
2020-06-09 15:00:47 <feuerteufel> I hope it's what you want?
2020-06-09 15:01:00 <ncopa> its exactly what i wanted help with
2020-06-09 15:03:44 <feuerteufel> ncopa: Pls take a look at the end of the file, I didn't know wat to do whith that so I yust command it out
2020-06-09 15:03:54 <ncopa> thats fine
2020-06-09 15:04:09 <ncopa> seems like 2.4.8 release notes is missing
2020-06-09 15:04:18 <ncopa> but  dont worry, i'll clean that up
2020-06-09 15:11:22 <feuerteufel> Oh, sorry and I double checked it this morning :-(
2020-06-09 15:12:55 <feuerteufel> mps: Sorry 2 diffrent job sites
2020-06-09 15:15:08 <mps> feuerteufel: np :)
2020-06-09 15:50:15 <pkx> Are there docs somewhere about how I can make a change to this file and test the resulting `apk add freeswitch`: https://git.alpinelinux.org/aports/tree/main/freeswitch/APKBUILD
2020-06-09 15:52:12 <ikke> https://wiki.alpinelinux.org/wiki/Creating_an_Alpine_package
2020-06-09 16:04:39 <codebam> I think there is an issue in the telegram-desktop package for alpine? for the past few releases I get this when trying to start it. https://0x0.st/iVpJ.txt
2020-06-09 16:04:54 <codebam> about incompatible qt libraries
2020-06-09 16:13:01 <Cogitri> That might be an incomplete upgrade on your side
2020-06-09 16:13:10 <Cogitri> fwiw it works just fine for me
2020-06-09 16:16:21 <codebam> oh hmm. is there a way to fix that? can I reinstall all it's dependencies recursively or something?
2020-06-09 16:17:49 <ikke> codebam: did you mix stable and edge?
2020-06-09 16:17:59 <codebam> ikke: no just edge
2020-06-09 16:18:04 <ikke> hmm, ok
2020-06-09 16:18:14 <ikke> you can try apk upgrade -U --available
2020-06-09 16:19:46 <codebam> oh yeah. https://0x0.st/iVph.txt
2020-06-09 16:20:02 <codebam> pass depends on cmd:gpg2 which is apparently not available anymore
2020-06-09 16:21:44 <codebam> although I removed pass to fix that dependency issue and I still get the same error
2020-06-09 16:25:29 <codebam> as in like the updates went through now but telegram-desktop still gives the same qt incompatibility error
2020-06-09 16:26:22 <Cogitri> Well, it's not on tg desktop, something is keeping one of your Qt libs at an older version
2020-06-09 16:26:49 <ikke> codebam: can you try apk version -l '?'
2020-06-09 16:29:18 <codebam> okay
2020-06-09 16:29:22 <codebam> https://0x0.st/iVf-.txt
2020-06-09 16:29:26 <codebam> ikke ^
2020-06-09 16:31:53 <ikke> mdocml is no longer there, replaced by man-doc
2020-06-09 16:31:55 <ikke> mandoc
2020-06-09 16:33:05 <codebam> I fixed all those except the makedepends. how do I remove those?
2020-06-09 16:33:58 <codebam> never mind
2020-06-09 16:34:02 <ikke> you can apk del them, or remove them from /etc/apk/world and run apk fix
2020-06-09 16:34:20 <codebam> fixed all of those but still getting that qt error when launching telegram-desktop
2020-06-09 16:40:23 <Cogitri> Did you do an `apk upgrade -a` or similar afterwards?
2020-06-09 16:43:17 <codebam> sorry my tab crashed
2020-06-09 16:43:40 <ikke> "Cogitri │ Did you do an `apk upgrade -a` or similar afterwards?"
2020-06-09 16:43:41 <codebam> I'd use weechat but for whatever reason it times out to the freenode server and I don't want to figure it out right now
2020-06-09 16:44:12 <codebam> Cogitri oh yeah I did `apk upgrade -U --available`
2020-06-09 16:44:34 <ikke> after you fixed those packages?
2020-06-09 16:44:47 <codebam> yeah it updated a bunch but it still doesn't work
2020-06-09 16:44:50 <ikke> k
2020-06-09 16:45:15 <codebam> the list for `apk version -l '?'` is empty now
2020-06-09 16:53:47 <codebam> thanks for all the help anyways. I might just have to go through all my packages and see which ones depend on qt
2020-06-09 16:54:10 <ikke> The error message it not very clear
2020-06-09 16:57:13 <Cogitri> Yes, unfortunately Qt doesn't say which lib is out of sync
2020-06-09 16:57:28 <ikke> perhaps an strace might clear things up?
2020-06-09 16:57:45 <Cogitri> I guess you could do apk info | grep qt and pipe that into a while loop and compare the versions of all Qt packages like that
2020-06-09 16:57:54 <Cogitri> Right, that might work too
2020-06-09 17:01:17 <codebam> this is... fun
2020-06-09 17:01:26 <codebam> I'm gonna have to reinstall so much after this is done lol
2020-06-09 17:01:46 <codebam> just going through dependencies and making sure no packages are holding back library updates
2020-06-09 17:02:50 <ikke> You can try apk version -l '>'
2020-06-09 17:03:06 <ikke> sorry
2020-06-09 17:03:13 <ikke> apk version -l '<'
2020-06-09 17:03:22 <ikke> That should list packages where an upgrade is available
2020-06-09 17:04:39 <codebam> oh okay awesome thanks
2020-06-09 17:05:14 <codebam> how do I force remove a package along with it's dependencies? (or force upgrade removing dependencies that are holding the package back)
2020-06-09 17:05:21 <codebam> ideally the latter actually
2020-06-09 17:05:39 <ikke> if you remove a package, all dependencies that are not required are automatically removed
2020-06-09 17:06:27 <codebam> oh yeah I mean manually installed dependencies
2020-06-09 17:06:46 <ikke> easiest is to check /etc/apk/world and remove anything from there that you do not need
2020-06-09 17:06:59 <codebam> okay
2020-06-09 17:07:16 <ikke> afterwards, run apk fix
2020-06-09 17:14:09 <codebam> how would I fix these unsatisfiable constraints when adding telegram-desktop
2020-06-09 17:14:12 <codebam> https://0x0.st/iVfU.txt
2020-06-09 17:16:15 <codebam> does that mean that it wants the older version, or the newer?
2020-06-09 17:17:54 <Cogitri> It wants the newer version but can't pull it in because some things still need the old version
2020-06-09 17:18:24 <Cogitri> Seems like it can't pull in the new mozjs60 version for some reason
2020-06-09 17:26:38 <codebam> I fixed that problem and got it to install (and I think I fixed the issue with telegram-desktop) but now I can't reinstall firefox due to this mess :(
2020-06-09 17:27:05 <codebam> https://0x0.st/iVfD.txt
2020-06-09 17:35:11 <codebam> I think it's due to some package holding back my python2 update, but how would I check which package that is?
2020-06-09 17:36:17 <codebam> never mind, nothing even needed python2
2020-06-09 17:36:57 <codebam> yay I fixed all the issues
2020-06-09 17:37:28 <codebam> thanks for your help Cogitri and ikke :)
2020-06-09 17:37:36 <ikke> yw
2020-06-09 17:42:12 <rapha> hi
2020-06-09 17:42:50 <rapha> how do you tell apk "please reinstall this package, like, completely, and totally ignore the possibility of overwriting any files, it's fine, really, really, REALLY, yes, I *do* mean it!"
2020-06-09 17:44:36 <ikke> try apk fix --clean-protected <package>
2020-06-09 17:48:04 <rapha> ikke: still loads of errors about not being able to rename files that are long gone
2020-06-09 17:48:50 <ikke> are these files ending in .apk-new?
2020-06-09 17:51:10 <rapha> ikke: no, but they do have to do with apk: "ERROR: py3-setuptools-47.0.0-r0: failed to rename usr/lib/python3.8/site-packages/setuptools/__pycache__/.apk.b614c05634e4e30b538bd99bf19e20e4aa8a0320a32f9d68 to usr/lib/python3.8/site-packages/setuptools/__pycache__/py34compat.cpython-38.pyc."
2020-06-09 17:52:28 <ikke> what does ls -ld return usr/lib/python3.8/site-packages/setuptools/__pycache__
2020-06-09 17:53:27 <rapha> that won't be there. this is the point where you need to get angry with me. in my attempts of resolving this, i rm rf'd usr/lib/python3.8/site-packages/setuptools*...
2020-06-09 17:58:59 <ikke> At this time, what part of that path does exist?
2020-06-09 17:59:48 <rapha> ikke: /usr/lib/python3.8/site-packages/
2020-06-09 18:00:45 <ikke> ok, and you can execute `mkdir -p usr/lib/python3.8/site-packages/setuptools/__pycache__/` ?
2020-06-09 18:01:07 <rapha> sure
2020-06-09 18:01:19 <rapha> done
2020-06-09 18:01:25 <ikke> ok
2020-06-09 18:01:43 <rapha> now the clean-protected again?
2020-06-09 18:02:01 <ikke> I don't think that flag matters at this point, but just run apk fi
2020-06-09 18:02:03 <ikke> apk fix
2020-06-09 18:02:17 <rapha> same errors
2020-06-09 18:02:49 <rapha> the heck
2020-06-09 18:02:55 <ikke> ok, lets try: apk add strace; strace apk fix py3-setuptools
2020-06-09 18:03:00 <rapha> oh
2020-06-09 18:03:02 <rapha> my fault
2020-06-09 18:03:12 <rapha> i forgot to add a slash in front of your mkdir command and i wasn't in /
2020-06-09 18:03:24 <ikke> THat command was not mean to fix anything
2020-06-09 18:03:30 <ikke> just to make sure that path could be created
2020-06-09 18:04:18 <rapha> i'm just seeing though that there's different errors before the rename ones: ERROR: py3-setuptools-47.0.0-r0: trying to overwrite usr/lib/python3.8/site-packages/__pycache__/easy_install.cpython-38.pyc owned by python3-3.8.2-r0.
2020-06-09 18:04:32 <ikke> aha
2020-06-09 18:04:47 <ikke> is this on edge?
2020-06-09 18:05:00 <rapha> no, i'm trying to upgrade 3.11 to 3.12
2020-06-09 18:05:05 <ikke> ok
2020-06-09 18:05:08 <rapha> buuuuuuut some python packages were from testing
2020-06-09 18:07:27 <ikke> Well, that is not really supported (combining stable and edge)
2020-06-09 18:08:09 <rapha> i know, and i should have undone that when i didn't need it anymore
2020-06-09 18:08:13 <rapha> instead i forgot, until now
2020-06-09 18:08:29 <rapha> the testing repo is disabled now btw
2020-06-09 18:08:33 <ikke> try to remove the edge repository, and then apk upgrade -U -a
2020-06-09 18:09:04 <rapha> that didn't do anything, and apk fix still reports those errors
2020-06-09 18:09:15 <rapha> should i delete the cache?
2020-06-09 18:11:58 <ikke> That doesn't really matter
2020-06-09 18:14:58 <rapha> aaah
2020-06-09 18:15:00 <rapha> finally
2020-06-09 18:15:11 <rapha> uninstalled all py3 stuff that could be uninstalled
2020-06-09 18:15:15 <rapha> errors went away
2020-06-09 18:16:50 <rapha> thank you ikke
2020-06-09 18:17:26 <ikke> you're welcome, even though you did most of the work :)
2020-06-09 18:18:43 <rapha> well, you know how it is, sometimes you just need a guiding light, but that, you do need :)
2020-06-09 18:34:48 <codebam_1> I removed all those packages and did updates but now I don't have the gpg command even though the gnupg package is installed?
2020-06-09 18:34:54 <rapha> ikke: interesting, /usr/bin/pon seems to forget to create /var/run/ppp now, and even after fixing that, doesn't run anymore during boot.
2020-06-09 18:36:06 <rapha> oh wait, i have /etc/local.d/uplink.start for that. _that_ is what doesn't run anymore.
2020-06-09 18:37:37 <rapha> were there any changes about /etc/local.d/ handling in 3.12?
2020-06-09 18:37:48 <ikke> Not that I'm aware of
2020-06-09 18:38:05 <rapha> it's called .start and is executable. should be called upon, no?
2020-06-09 18:39:42 <ikke> is rc-status showing local?
2020-06-09 18:40:16 <rapha> oh! no. added it. let me try another reboot...
2020-06-09 18:43:27 <rapha> ran, this time, but pppd failed to connect. manual "poff", "pon" worked though. weird. this never gave any trouble.
2020-06-09 18:45:23 <codebam> /usr/bin/gpg2 is a link to gpg but I don't have that it doesn't seem. does gnupg not provide the gpg command?
2020-06-09 18:47:01 <codebam> hmm seems like it does. maybe removing one of the other packages removed the command
2020-06-09 18:47:46 <rapha> well this will have to wait until some night where i'm awake and all others sleep deeply
2020-06-09 18:52:13 <codebam> fixed by uninstalling and reinstalling
2020-06-09 22:05:03 <ThinkDao> hi all, I am facing a weird problem here. I can access the internet with the root but I cannot access websites or git with normal user.  I need to set some special permission of normal users?
2020-06-09 22:09:32 <ThinkDao>  never mind.. I set a wrong flag to /etc/resolv.conf
2020-06-09 22:11:03 <ThinkDao> I am loving this Alpine.. First install and I am using only 50M of memory and everything is super fast. Congratulation to all developers
2020-06-10 05:54:30 <quazimodo> hola
2020-06-10 05:54:54 <quazimodo> is it normal to have a file /opt/my-file with chmod +x & when I do cd /opt; ./my-file itcomplains no such file or directory
2020-06-10 05:55:06 <quazimodo> this is with the alpine 1.13 docker image
2020-06-10 05:55:18 <quazimodo> super confusing, the file is there & it's rwx
2020-06-10 05:55:31 <Cogitri> It's probably linked against glibc
2020-06-10 05:55:42 <Cogitri> And not musl
2020-06-10 05:55:50 <ikke> it cannot find the linker, that's what it's complaining about
2020-06-10 05:55:53 <Cogitri> So it can't find the dynamic linker and as such complains about no such file or directory
2020-06-10 05:56:41 <ikke> try `ldd /opt/my-file) | head -n1`
2020-06-10 05:57:23 <quazimodo> ah
2020-06-10 05:57:29 <quazimodo> you guys are smart
2020-06-10 05:58:25 <ikke> or just bitten by similar issues ourselfs :)
2020-06-10 06:09:18 <quazimodo> it makes sense because the file is a precompiled file
2020-06-10 07:47:29 <ikke> you putting it in /opt is a dead give-away :P
2020-06-10 07:48:06 <oneinsect> hello
2020-06-10 07:48:11 <oneinsect> ikke
2020-06-10 08:58:06 <dngray> i noticed there doesn't seem to be a section about setting the locale on alpine linux in https://wiki.alpinelinux.org/wiki/Setup-alpine
2020-06-10 08:59:04 <dngray> is there a preferred way on alpinxlinux>?
2020-06-10 08:59:28 <Cogitri> I think we have setup-locale?
2020-06-10 09:00:01 <afraid> GOOD MORNING ALPINE LINUX
2020-06-10 09:00:32 <ikke> Is that a reference to "good morning vietnam"?
2020-06-10 09:53:23 <rapha> :P
2020-06-10 10:02:41 <dngray> i do not appear to have setup-locale hmm
2020-06-10 10:03:25 <dngray> maybe i need musl-locales?
2020-06-10 10:05:29 <earnestly> If you just need UTF-8, don't set LANG or LC_ at all
2020-06-10 10:05:46 <earnestly> musl's default is effectively "C.UTF-8"
2020-06-10 10:06:30 <ikke> But some programs require it to be set
2020-06-10 10:07:33 <Cogitri> Should be set by default in 3.12
2020-06-10 10:09:55 <dngray> i have 3.12 🙂
2020-06-10 10:10:25 <Cogitri> Then it should be set to C.UTF-8 via /etc/profile.d/locale.sh
2020-06-10 10:11:04 <dngray> yeah i have /etc/profile.d/locale.sh
2020-06-10 10:11:21 <dngray> and its set to export CHARSET=UTF-8
2020-06-10 10:11:21 <dngray> export LC_COLLATE=C
2020-06-10 10:11:21 <dngray> export LANG=C.UTF-8
2020-06-10 10:12:09 <dngray> but no setup-locale
2020-06-10 10:13:23 <ikke> setup-locale is part of alpine-conf, which is present by default on the live iso's, but not on installed systems
2020-06-10 10:14:12 <ikke> Cogitri: there is no setup-locale...
2020-06-10 10:14:25 <ikke> https://gitlab.alpinelinux.org/alpine/alpine-conf
2020-06-10 10:14:54 <dngray> ahh
2020-06-10 10:18:02 <dngray> ikke: are you sure https://pkgs.alpinelinux.org/contents?branch=edge&name=alpine-conf&arch=x86_64&repo=main
2020-06-10 10:19:43 <Cogitri> ikke: oh oops :D
2020-06-10 10:19:44 <earnestly> ikke: software shouldn't need it set, that may be a bug report (i.e. setlocale(LC_ALL, "");)
2020-06-10 10:20:43 <dngray> well the issue i am having is my archlinux containers are showing as C
2020-06-10 10:20:44 <dngray> i thought it might have been somehow inherited from the host
2020-06-10 10:21:07 <dngray> as it seems uncommenting the options and running locale-gen didn't fix it
2020-06-10 10:21:25 <dngray> ie https://wiki.archlinux.org/index.php/Locale#Setting_the_locale
2020-06-10 10:21:43 <earnestly> localedef --list-archive
2020-06-10 10:22:23 <Cogitri> earnestly: Things like DBus just assume that if the locale isn't UTF-8 they have to fallback to non UTF-8 which breaks things that rely on UTF-8 on DBus
2020-06-10 10:22:32 <earnestly> lol
2020-06-10 10:22:44 <earnestly> Well that's a trouble
2020-06-10 10:22:46 <dngray> <earnestly "localedef --list-archive"> that shows the correct locale
2020-06-10 10:23:23 <earnestly> dngray: Then at least it's generated and available, the problem is getting into the environment at the right time, and you'll have to figure that part out
2020-06-10 10:23:39 <dngray> oh wait maybe it is fixed now
2020-06-10 10:24:21 <dngray> was started with encoding ANSI_X3.4-1968, this should be UTF-8
2020-06-10 10:24:21 <dngray> was what i was seeing
2020-06-10 10:24:22 <dngray> but now it seems to be working, so that's good
2020-06-10 12:11:10 <yunfan> mps: i just bought 2 riscv board with 64mb ram, is it possible to run alpine?
2020-06-10 12:11:11 <yunfan> :D
2020-06-10 12:11:42 <ikke> alpine is not ported to riscv
2020-06-10 12:11:52 <yunfan> ;[
2020-06-10 12:15:07 <mps> yunfan: as ikke says, but how much they cost?
2020-06-10 12:15:37 <mps> send one to me and maybe we can have alpine on riscv in near future :)
2020-06-10 12:16:05 <yunfan> mps: it cost me 39 RMB, and to foreight consumer, its 5 euro dollar
2020-06-10 12:16:33 <yunfan> https://c-sky.github.io/docs/gx6605s.html   mps check this
2020-06-10 12:16:51 <yunfan> mps:  i am intersting of using it to impl my own forth on riscv :D
2020-06-10 12:17:11 <mps> oh, it's not too much, but 64MB of ram is not much
2020-06-10 12:17:58 <mps> for forth this ram is enough, sure
2020-06-10 12:18:51 <yunfan> mps: remind j1forth?
2020-06-10 12:19:03 <yunfan> they impl a tv game station
2020-06-10 12:19:09 <yunfan> i think this one could do that too
2020-06-10 12:19:24 <yunfan> and plus a esp8266, you got a wifi adapter
2020-06-10 12:23:24 <mps> doesn't linux in latest version got c-sky support
2020-06-10 12:27:05 <hodapp> you can use an ESP8266 as a WiFi adapter?
2020-06-10 12:31:41 <yunfan> hodapp: of course, there's some github repo for this
2020-06-10 12:32:08 <yunfan> mps: so you mean its quite possible to add a experiement alpine?
2020-06-10 12:56:32 <velix> ncopa: Do you think it's possible to build Inkscape without GTK? I'm using it for command line processing only, but it pulls 1000 MB of GUI packages :D
2020-06-10 13:07:12 <Cogitri> velix: No can do, but maybe we can split off the GUI part?
2020-06-10 13:12:44 <velix> Ooops: My 1000 MB may be wrong, since Alpine already had some packages installed.
2020-06-10 13:12:47 <velix> Cogitri: I don't know. I've just asked in #inkscape, let me check if they responded.
2020-06-10 13:12:59 <velix> Their answer was: "No" ;)
2020-06-10 13:13:21 <velix> So it needs GUI components at building stage, but I don't know about runtime.
2020-06-10 13:14:09 <Cogitri> Splitting is always about runtime and not buildtime
2020-06-10 13:17:39 <velix> Can I force the removal of single packages using apk, even if they depend on something.
2020-06-10 13:18:38 <ikke> --force-broken-world possibly
2020-06-10 13:20:16 <Cogitri> Or you could provide them via a virtual pkg
2020-06-10 13:25:45 <velix> Thanks, I'll try
2020-06-10 13:40:42 <mps> yunfan: if this CPU is c-sky, maybe alpine can work on it
2020-06-10 13:41:22 <mps> though something minimal, busybox and maybe few basic tools
2020-06-10 13:42:23 <mps> but I doubt anyone will port it in near future
2020-06-10 13:44:24 <mps> but I remember days when I run linux on 4MB ram machine
2020-06-10 13:46:13 <mps> my first hard disk was 10MB big :)
2020-06-10 13:50:36 <yunfan> mps: yes, its csky
2020-06-10 13:51:11 <yunfan> mps: in that days, you dont have so much apps
2020-06-10 13:55:51 <mps> yes, not much, mosaic browser, vi editor (elvis), gcc compiler, tcl/tk, music and video players, some games .....
2020-06-10 13:56:35 <mps> I don't use much more nowadays
2020-06-10 13:58:11 <yunfan> i think the modern web browser required too much resources
2020-06-10 13:58:16 <yunfan> but i really like it
2020-06-10 13:58:33 <yunfan> i even had bought 2 chromebook , both were arm inside
2020-06-10 13:58:42 <hodapp> it is, unfortunately, a requirement for the use of a lot of modern software
2020-06-10 13:59:36 <mps> I bought 3 chromebooks, all with 4GB ram
2020-06-10 14:01:16 <yunfan> mps: let me guess, one of google's pixel?
2020-06-10 14:01:18 <mps> one of them mostly keep dust nowadays
2020-06-10 14:01:38 <mps> no, two samsungs and one acer
2020-06-10 14:01:46 <yunfan> my chromebook is still be used, because its support for android apps, and its touch screen
2020-06-10 14:01:57 <yunfan> i use it to read book, quite convenient
2020-06-10 14:02:17 <yunfan> aha, mine is acer and samsung too
2020-06-10 14:02:26 <hodapp> I had a cheapo ARM-based Chromebook that I used as a craptop until it met its fate involving a travel bag and a poorly-sealed jar of oily substance.
2020-06-10 14:02:38 <mps> I bought them to run native linux not chromeOS, all is converted to alpine first day they arrived
2020-06-10 14:02:47 <yunfan> one is acer's chromebook flip the first edtion which had a 11inch screen
2020-06-10 14:03:06 <yunfan> another is samsung's chromebook plus
2020-06-10 14:03:41 <mps> yunfan: you mean this with rk3399 samsung?
2020-06-10 14:05:13 <yunfan> mps: yes
2020-06-10 14:05:21 <yunfan> i use it alomost everyday
2020-06-10 14:05:38 <mps> I type on it right now :) https://www.samsung.com/us/support/owners/product/chromebook-plus
2020-06-10 14:06:40 <yunfan> hahaha, we shared another same taste
2020-06-10 14:06:47 <mps> I use it daily about 3 years
2020-06-10 14:07:26 <yunfan> so do you built alpine on it too?
2020-06-10 14:07:36 <mps> ofc :)
2020-06-10 14:07:45 <yunfan> ah
2020-06-10 14:08:28 <mps> first tried debian but was somewhat slow, and for more than 2 and a half years converted it to alpine
2020-06-10 14:09:43 <mps> before this model I used samsung Peach Pi (exynos 5800) for about two years
2020-06-10 14:11:47 <mps> and in meantime I bought acer R13, mediatek SOC, and that was bad buying, it is not much supported because mediatek is not much open
2020-06-10 14:12:49 <yunfan> i had tried arch arm
2020-06-10 14:13:08 <yunfan> but the x server had some problem
2020-06-10 14:13:17 <yunfan> dont know if they had fixed this recently
2020-06-10 14:13:29 <mps> also I did, but only on console, didn't even tried to install Xorg
2020-06-10 14:13:41 <yunfan> yes, mediatek is hateful
2020-06-10 14:14:00 <yunfan> many cellphone losed value since their lack of support
2020-06-10 14:14:26 <yunfan> if they could act like allwinner, i think there would be much more devices in the list of postmarket os's wiki
2020-06-10 14:15:43 <mps> I agree
2020-06-10 14:16:16 <mps> and I know that yet another alpine developer own acer R13 mediatek chromebook
2020-06-10 14:17:13 <mps> own or have, what is proper term
2020-06-10 14:41:20 <TBB> own if it's not his, have if it is
2020-06-10 14:41:45 <TBB> ... sorry, I've been too involved with information security
2020-06-10 14:44:35 <srg> Hello! Quick question. `apk del` has option --rdepends "Recursively delete all top-level reverse dependencies too" I'm a bit confused by that description. A reverse dependency is "package(s) that depends on _this_ package", right?
2020-06-10 14:45:26 <srg> So when I run `apk del --rdepends foo`, it will delete `foo` along with the packages that _directly_ depend on `foo`?
2020-06-10 14:48:29 <Cogitri> If baz depends on bar and bar depends on foo `apk del --rdepends foo` will remove all three I think
2020-06-10 14:52:09 <srg> Hmm ok. So without --rdepends, it'll just delete `foo` and `bar`, but not `baz`? If that's the case than I think I always want --rdepends to avoid breaking anything
2020-06-10 14:55:55 <Cogitri> Without rdepends apk will refuse to remove foo because bar still needs it and baz needs it via bar
2020-06-10 14:58:31 <oneinsect> hola ladies
2020-06-10 14:58:34 <oneinsect> howdi
2020-06-10 15:14:03 <srg> Cogitri: Thanks!
2020-06-10 17:08:58 <__newguy> hello, I added coreutils in /medias/sda/apks/x86_64/, but even after an apk update, apk add coreutils it says it's missing. Any ideas ?
2020-06-10 17:09:20 <ikke> it needs to be added to the index
2020-06-10 17:09:47 <ikke> which needs to be signed by a trusted key
2020-06-10 17:16:50 <gjabell> I'm upgrading profanity from 0.8.1 to 0.9.0 and having an issue with strdup'd glib string returning an invalid pointer
2020-06-10 17:17:09 <gjabell> anyone know if there have been similar issues with glib recently that need a patch?
2020-06-10 17:20:24 <Cogitri> I haven't seen that yet and that probably is something in profanity
2020-06-10 17:22:12 <gjabell> yeah, what's odd is the code where it's breaking hasn't changed in 4 years; it works fine with 0.8.1 but breaks with the new version :P
2020-06-10 17:24:34 <Cogitri> Time to look at the diff
2020-06-10 17:30:12 <__newguy> ikkeSo I see, thx for the info, I thought I couldve just add the .apk to the iso /apks and have them x)
2020-06-10 18:24:17 <velix> Can I list the installed packaegs by size?
2020-06-10 18:24:54 <velix> https://gist.github.com/meetmatt/d5dd2ecae37fbabb483dffba2d6dbec6
2020-06-10 18:25:50 <ikke> Did you answer your own question?
2020-06-10 18:26:28 <velix> ikke: Yeah, but for the logs, you know...
2020-06-10 18:26:39 <ikke> ok
2020-06-10 18:26:43 <ikke> just wanted to be sure
2020-06-10 18:26:52 <velix> "captain's log" ;)
2020-06-10 18:27:10 <velix> But wow, this snippet is slow as hell.
2020-06-10 18:27:50 <Hello71> of course, you did -n1
2020-06-10 18:29:36 <velix> it's not my code.
2020-06-10 18:29:42 <velix> Hello71: Yeah, this works better: apk info | xargs apk info -s | xargs -n4 | awk '{print $4,$1}' | sort -rn
2020-06-10 18:32:09 <Hello71> it doesn't really work though if there are multiple versions
2020-06-10 18:32:46 <velix> Hello71: True, but it's a container. So it's fine for me.
2020-06-10 18:32:56 <Hello71> awk -F: '$1=="P"{p=$2}$1=="I"{print $2,p}' /lib/apk/db/installed | sort -rn
2020-06-10 18:33:11 <velix> ikke: --force-broken-world doesn't work, I'm still getting: "World updated, but the following packages are not removed due to ..."
2020-06-10 18:33:30 <velix> Hello71: Looks good!
2020-06-10 18:37:39 <velix> Ah, okay: "APK Tools does not support forced package removal, because that would break the graph that it uses to represent packages."
2020-06-10 18:50:53 <velix> Can I break the graph with violence?
2020-06-10 18:51:03 <velix> yelling at it didn't work
2020-06-10 18:54:23 <ikke> apk does not reward violence :)
2020-06-10 18:55:28 <ikke> velix: try apk add -t <package>
2020-06-10 18:55:37 <ikke> that would add the package as a virtual package
2020-06-10 18:56:06 <velix> ERROR: unsatisfiable constraints ...
2020-06-10 18:57:35 <ikke> So what are you actually trying to achieve?
2020-06-10 18:57:44 <ikke> and more important, why do you want to do it this way
2020-06-10 18:58:13 <velix> I want to rip out gtk+3.0 runtimes from inkscape.
2020-06-10 19:04:14 <ikke> I don't think apk itself will easily let you do that (as you already noticed)
2020-06-10 19:04:36 <velix> maybe I can just remove the files beloging to it and check if the app still starts ;)
2020-06-10 19:05:05 <ikke> most likely not
2020-06-10 19:05:19 <ikke> the linker will try to find those dependencies and will complain
2020-06-10 19:08:32 <velix> ok :(
2020-06-10 19:08:42 <velix> Thanks for your input
2020-06-10 20:19:08 <markand> hello
2020-06-10 20:19:28 <markand> to change java version one simply change the /usr/lib/jvm/default-jvm symlink?
2020-06-10 23:52:31 <lunaphyte> Hello71: thanks for the suggestions the other day.  i've figured out what was missing
2020-06-10 23:52:35 <lunaphyte> it was quite a lot :)
2020-06-10 23:55:40 <codebam> hey, would anyone know why my timezone database wouldn't contain US/Eastern? https://github.com/Alexays/Waybar/issues/630
2020-06-10 23:56:01 <codebam> I changed from Canada/Eastern just to make sure
2020-06-10 23:56:19 <codebam> I also chuckled when I realized that it was my comment a couple months ago
2020-06-10 23:57:10 <codebam> I regenerated it with `doas setup-timezone -z US/Eastern` a couple mins ago. would I need to reboot after that?
2020-06-11 00:03:03 <Hello71> well your issue is completely unrelated to maxice8's issue
2020-06-11 00:03:41 <codebam> okay. I'm going to open a new issue then
2020-06-11 00:03:50 <codebam> would this even be related to waybar though?
2020-06-11 00:04:04 <codebam> or is this a problem with how I have my timezones configured
2020-06-11 00:06:33 <codebam> if I searched harder :P https://github.com/Alexays/Waybar/issues/566
2020-06-11 00:06:37 <codebam> trying some of that now
2020-06-11 00:17:33 <codebam> nah did a complete rebuild, removed the entire git repo. same issue
2020-06-11 00:27:12 <izaki> Hello! Besides the documentation on how to make a custom iso image... Are there any docs on how to build custom alpine boot images?
2020-06-11 00:27:39 <izaki> I am trying to build a netboot image with some customizations
2020-06-11 00:28:11 <izaki> I'd like to know how to remove some of the base components, such as firmware, or virtio
2020-06-11 00:43:02 <lunaphyte> during an install, which part is responsible for populating the various runlevels?
2020-06-11 00:43:42 <lunaphyte> looking at the openrc post-install script, i'd thought it was that, but testing just now, they're all empty
2020-06-11 00:44:38 <lunaphyte> by populating, i mean enabling the basic/minimum services [like swap, networking, sysctl, etc] for the various runlevels
2020-06-11 00:47:12 <lunaphyte> setup-disk seems to be doing it [maybe indirectly], but it's not clear to me where/how
2020-06-11 00:47:51 <lunaphyte> e.g. when doing setup-disk /mnt, where it's just running install_mounted_root()
2020-06-11 00:49:55 <lunaphyte> oh, hmm.  is it something to do with lbu package - | tar -C "$mnt" -zxv > "$ovlfiles" ?
2020-06-11 00:51:02 <Hello71> probably
2020-06-11 00:51:32 <Hello71> I reckon if you do rc-update add whatever whatever; lbu commit, then reboot, it will start the thing
2020-06-11 02:26:00 <codebam> does the connman that is built for alpine come with vpn support?
2020-06-11 02:26:16 <maxice8> What is required for vpn support ? any specific packages ?
2020-06-11 02:26:17 <codebam> no matter what I do I can't get it to pick up my vpn configs
2020-06-11 02:26:29 <codebam> maxice8 I think just connman-vpnd, which is here
2020-06-11 02:27:15 <Hello71> codebam: did you check if there actually exists /etc/zoneinfo/US/Eastern
2020-06-11 02:27:27 <maxice8> openconnect, openvpn, wireguard, vpnc
2020-06-11 02:27:34 <codebam> Hello71 yes it does
2020-06-11 02:28:08 <codebam> maxice8 the end goal is wireguard, but in testing to just get it to find the config I was using the example for openconnect
2020-06-11 02:28:32 <codebam> the example here: https://git.kernel.org/pub/scm/network/connman/connman.git/tree/doc/vpn-config-format.txt
2020-06-11 02:28:58 <maxice8> well, openvpn support is disabled by default
2020-06-11 02:29:02 <maxice8> if i pass --enable-openvpn it will complain there is no openvpn binary
2020-06-11 02:29:10 <codebam> I placed it /var/lib/connman-vpn/fakevpn.config
2020-06-11 02:29:30 <codebam> or. I was using the example for l2tp
2020-06-11 02:29:46 <codebam> just to try to get it to find the config
2020-06-11 02:30:20 <codebam> maxice8 that's fine by me. I don't use openvpn anyways
2020-06-11 02:30:56 <maxice8> it apparently doesn't even have wpa_supplicant support
2020-06-11 02:31:43 <codebam> it does I think. I've used it with wpa_supplicant before. it works with iwd though I know
2020-06-11 02:32:01 <maxice8> well it doesn't detect wpa_supplicant when building
2020-06-11 02:32:25 <maxice8> yes, it has an explicit --enable-iwd i think @mps worked on it
2020-06-11 02:33:23 <maxice8> can you test !9110 ?
2020-06-11 02:33:37 <maxice8> you should be able to download the package after CI is done building it
2020-06-11 02:34:44 <codebam> sure okay, thanks!
2020-06-11 02:59:30 <codebam> okay awesome so
2020-06-11 02:59:36 <codebam> maxice8
2020-06-11 02:59:42 <maxice8> yes ?
2020-06-11 02:59:44 <codebam> I got it to detect the openvpn example config
2020-06-11 02:59:56 <maxice8> yes
2020-06-11 03:00:04 <codebam> but it doesn't detect the wireguard one
2020-06-11 03:00:06 <codebam> it's not listed
2020-06-11 03:00:17 <maxice8> this file is added after i put --enable-openvpn, usr/lib/connman/plugins-vpn/openvpn.so
2020-06-11 03:00:18 <codebam> but I can see that in the strace it's being read
2020-06-11 03:00:33 <codebam> yes I saw it read that in the strace too
2020-06-11 03:00:37 <codebam> I used strace -f
2020-06-11 03:01:50 <maxice8> then wireguard is already there
2020-06-11 03:01:58 <maxice8> and something is failing along the way
2020-06-11 03:02:48 <codebam> I can't tell. it doesn't throw any errors
2020-06-11 03:02:57 <codebam> the strace part for wireguard looks identical to openvpn
2020-06-11 03:03:27 <codebam> https://0x0.st/iVYg.txt
2020-06-11 03:03:30 <codebam> that's the strace
2020-06-11 03:03:43 <maxice8> I don't even use connman so i can't help much
2020-06-11 03:04:21 <codebam> that's okay. thanks for all your help so far. I'm going to see what I can do to get it to detect it now that I know it's in the right place
2020-06-11 07:54:53 <PhoenixMage> hmmm, so any way to use tagged repos in APKBUILD? The normal syntax doesnt work
2020-06-11 07:55:12 <ikke> I don't think it's supported
2020-06-11 07:55:53 <PhoenixMage> How unfortunate
2020-06-11 07:59:12 <PhoenixMage> I guess I could be very didgy and put an apk add in prepare()
2020-06-11 08:01:37 <ikke> I'm not sure why <package>@<pin> would not work in an APKBUILD though
2020-06-11 08:03:00 <PhoenixMage> ERROR: 'bazel@edgetesting' is not a valid child dependency, format is name([<>~=]version) is the error, maybe as simple as fixing that format check, will have a quick look at the src
2020-06-11 08:03:16 <afontain_> PhoenixMage : That's a build dependency you want to tag?
2020-06-11 08:03:40 <PhoenixMage> afontain_: yeah, I hate bazel but its the build system they are using
2020-06-11 08:04:05 <PhoenixMage> Actually after the last couple of days hate is too weak a word
2020-06-11 08:14:37 <afontain_> If that's a corporate setting, you could consider adding your repository where you cherry-pick from edge
2020-06-11 08:28:24 <PhoenixMage> Nah its my home lab
2020-06-11 08:28:58 <PhoenixMage> The limitation is you cant add a tagged package to a virtual package
2020-06-11 09:01:58 <afontain_> note that mixing stable and edge can break at any time
2020-06-11 09:02:53 <afontain_> for example, once edge upgrade to musl 1.2, it could break
2020-06-11 09:03:59 <afontain_> (it also could not, I didn't try, but there is no warranty)
2020-06-11 09:09:34 <mps> afontain_: looks like it will not break if you add musl 1.2.x to stable
2020-06-11 09:10:16 <mps> works for me 5 days without any noticeable problem
2020-06-11 09:10:59 <mps> but your advice is good, i.e. 'don't mix stable and edge'
2020-06-11 09:11:30 <Tuxist> i got it ceph 15.2.3 is now compiling :-D
2020-06-11 09:11:35 <Tuxist> https://gitlab.alpinelinux.org/Tuxist/aports
2020-06-11 09:14:23 <Tuxist> next step create init scripts and make ceph deploy runable
2020-06-11 09:38:16 <PhoenixMage> Tuxist: Are you creating a rook image or something?
2020-06-11 09:38:58 <PhoenixMage> I worked with the ceph guys to get most of the musl stuff fixed
2020-06-11 09:40:11 <Tuxist> PhoenixMage: i'am work at the moment to get deployable so i can test it
2020-06-11 09:40:25 <PhoenixMage> I have working images
2020-06-11 09:40:40 <Tuxist> PhoenixMage: cool
2020-06-11 09:40:42 <PhoenixMage> I spent a lot of time on it
2020-06-11 09:41:10 <PhoenixMage> I can share all the stuff you need
2020-06-11 09:41:20 <Tuxist> PhoenixMage: that will be cool
2020-06-11 09:41:37 <Tuxist> PhoenixMage: i have very tiny three node cluster
2020-06-11 09:42:27 <PhoenixMage> What architecture?
2020-06-11 09:43:25 <PhoenixMage> Mine is sitting at 15.2.0 I really need to upgrade it but my 2nd daughter is due Sunday so time is scarce
2020-06-11 09:43:26 <Tuxist> PhoenixMage: amd64
2020-06-11 09:43:50 <PhoenixMage> I am running a mixed cluster, 4 x arm64 and 1 x amd64
2020-06-11 09:44:26 <PhoenixMage> afontain_: I wouldnt mix edge and stable if I could at all avoid it
2020-06-11 09:45:23 <PhoenixMage> Tuxist: https://github.com/PhoenixMage/aports
2020-06-11 09:46:14 <Tuxist> PhoenixMage: that are my nodes ^^ https://tuxist.de/files/IMG_20200611_114214643_HDR.jpg
2020-06-11 09:46:22 <PhoenixMage> Not sure if all 3 patches are still necessary, they were considering back porting them
2020-06-11 09:47:20 <PhoenixMage> Damn that is slow loading
2020-06-11 09:50:40 <AinNero> Tuxist: i dont know what that is but i want it, too
2020-06-11 09:55:39 <PhoenixMage> Tuxist: My aarch64 stuff is rock pi 4's with 250Gb m.2 drives
2020-06-11 09:56:00 <PhoenixMage> The amd64 box is just a vm
2020-06-11 09:58:14 <Tuxist> PhoenixMage: n3050 dualcore atoms with 32 gb nvme for system and 2 tb harddrives
2020-06-11 09:58:52 <Tuxist> PhoenixMage: if it works good i will used it for webmyserver as backend storage
2020-06-11 10:00:15 <twb> Where does alpine keep the busybox dhclient config?
2020-06-11 10:00:52 <twb> Doesn't seem to be /etc/rc.conf or /etc/dhc*
2020-06-11 10:01:16 <ikke> There is only /usr/share/udhcpc/default.script
2020-06-11 10:02:09 <twb> Hrm
2020-06-11 10:02:20 <twb> I can see a process running called "udhcpc -b -R -p /var/run/udhcpc.eth0.pid -i eth0 -x hostname:woozy"
2020-06-11 10:02:39 <twb> So "use dhcp for eth0" must be remembered somewhere...
2020-06-11 10:02:48 <ikke> /etc/network/interfaces
2020-06-11 10:02:59 <twb> Oh.  Derp.
2020-06-11 10:03:14 <twb> I didn't even consider that it would follow ifupdown style
2020-06-11 10:03:19 <twb> thanks
2020-06-11 10:06:35 <twb> does udhcpc log anywhere?  I can't see any messages from it in /var/log/messages.
2020-06-11 10:07:11 <ikke> I don't think so, not sure
2020-06-11 10:07:12 <twb> (The background is: I suspect the dhcp server or L3 switch are being Evil at me, and I'm looking at all the DHCP clients for logs of evidence that Evil Things Happened)
2020-06-11 10:07:32 <twb> Righto
2020-06-11 10:08:29 <ikke> maybe a tcpdump might help
2020-06-11 10:10:39 <mps> or run it from console with '-f' in foreground
2020-06-11 10:11:09 <twb> Yeah, that's reasonable.  I was struggling to get into the DHCP clients to tcpdump them while they were not getting an address.  tcpdump on the DHCP server just didn't see packets at all.
2020-06-11 10:12:03 <twb> There's also STP in the mix, and I *think* weirdness is not just STP's usual delays, but it means you can't just go "I rebooted and the problem appeared", because you have to wait until after STP buggers off
2020-06-11 12:45:55 <PhoenixMage> Tuxist: I just successfully compiled ceph 15.2.3 using my github
2020-06-11 12:46:59 <PhoenixMage> Well it was successful on aarch64, armv7 failed but I will deal with that later. Should be fine for amd64, my build server is slow so will kick it off before I go to bed
2020-06-11 13:23:28 <PhoenixMage> Tuxist: I have sent you a query with more info
2020-06-11 15:27:07 <dngray> Cogitri: i applied your suggestion in https://gitlab.alpinelinux.org/alpine/aports/-/merge_requests/7846#note_95567
2020-06-11 15:29:11 <Cogitri> You probably have to regenerate the checksum for the patch now :)
2020-06-11 15:42:07 <afraid> FINALLY.
2020-06-11 15:42:37 <afraid> A week ago, some discussion on gitlab talked about the exact issue I spent ages railing against, and it was all because of fucking python2.
2020-06-11 15:42:48 <afraid> So much wondering spent on that.
2020-06-11 15:51:22 <Cogitri> Huh?
2020-06-11 15:58:28 <Xelraa> reminds me of that python talk cooper uploaded about the insecurity of python with the funny camel pictures :P
2020-06-11 16:02:24 <Han> This is a nice article https://wiki.alpinelinux.org/wiki/FreeRadius_EAP-TLS_configuration#.2Fetc.2Fraddb.2Fclients.conf but there is something missing, the way to generate a client certificate. Where can I report that problem?
2020-06-11 16:02:45 <Han> I can figure out the user 'sillysausage' maintains the page, but there is no contact data.
2020-06-11 16:03:32 <Cogitri> The best way to get it fixed is probably to make a wiki account and add it
2020-06-11 16:05:30 <Han> If I'd know what to write that would be nice.
2020-06-11 16:07:36 <Cogitri> Ah
2020-06-11 16:08:05 <Cogitri> I guess you could mention it in the wiki page's discussion then, if someone is subscribed to the page they'll get a notification I thin
2020-06-11 16:09:53 <Han> Good idea.
2020-06-11 16:14:34 <Han>     This action has been automatically identified as harmful,
2020-06-11 16:14:35 <Han> and you have been prevented from executing it. In addition, to protect Alpine Linux, your user account and all associated IP addresses have been blocked from editing. If this has occurred in error, please contact an administrator. A brief description of the abuse rule which your action matched is: Spammers not welcome
2020-06-11 16:15:25 <Han> Cogitri, that's rather unexpected.
2020-06-11 16:15:40 <Han> Creating Talk:FreeRadius EAP-TLS configuration
2020-06-11 16:18:24 <Cogitri> Huh
2020-06-11 16:18:30 <Cogitri> ikke would know about that, I suppose
2020-06-11 16:20:09 <Han> ok, thanks
2020-06-11 16:39:26 <dngray> <Cogitri "You probably have to regenerate "> oh lol
2020-06-11 16:41:09 <ikke> Han: what username?
2020-06-11 16:42:27 <gjabell> after more research it seems the issue I was having yesterday with profanity/glib strings was caused by having `-std=c99` as a compiler option
2020-06-11 16:42:46 <ikke> I think I found it and it's unblocked
2020-06-11 16:43:51 <gjabell> should that cause an incompatibility, or is that more likely an issue with an underlying library?
2020-06-11 16:43:54 <Han> ikke, hboetes
2020-06-11 16:44:00 <ikke> yeah, unblocked
2020-06-11 16:44:04 <Han> ikke, cheers!
2020-06-11 16:44:55 <Han> ikke, I' m blocked again, trying to create the talk page. Can you create the page, I suspect that I may edit it but not create it.
2020-06-11 16:45:06 <ikke> Let me try
2020-06-11 16:45:11 <ikke> Are you adding urls?
2020-06-11 16:45:34 <Han> Ah yes, as a reference to the original documentation.
2020-06-11 16:46:10 <ikke> It does not like that for new accounts <5h old
2020-06-11 16:46:31 <Han> I understand. I'll hint upon it.
2020-06-11 16:47:04 <ikke> Created
2020-06-11 16:47:06 <Han> Right, there we go: https://wiki.alpinelinux.org/wiki/Talk:FreeRadius_EAP-TLS_configuration
2020-06-11 16:47:10 <ikke> ok
2020-06-11 16:47:30 <Han> Dank je wel!
2020-06-11 16:47:36 <ikke> :)
2020-06-11 16:47:41 <ikke> Graag gedaan
2020-06-11 16:50:06 <Han> This openssl stuff requires quite a bit of concentration.
2020-06-11 16:50:54 <dngray> Han: you don't need to use openssl to create the certs
2020-06-11 16:50:55 <dngray> just run /etc/raddb/certs/bootstrap
2020-06-11 16:51:22 <dngray> see /etc/raddb/certs/README for help about that
2020-06-11 16:51:48 <dngray> https://github.com/alandekok/freeradius-server/blob/master/raddb/certs/README
2020-06-11 16:51:49 <dngray> maybe the page should have something mentioned about that
2020-06-11 16:53:05 <Han> Does that also generate client certificates?
2020-06-11 16:53:24 <dngray> https://github.com/FreeRADIUS/freeradius-server/blob/v3.0.x/raddb/certs/README rather
2020-06-11 16:53:52 <Han> I already use the bootstrap script and edited to be to use revoking.
2020-06-11 16:55:10 <Han> client certificates as in the certificates to be used by laptops/phones etc, not the radius server, which is also called a client.
2020-06-11 16:55:19 <dngray> yeah that's how i did it
2020-06-11 16:55:36 <Han> ok thanks!
2020-06-11 16:56:12 <dngray> well it mentions client certs there https://github.com/FreeRADIUS/freeradius-server/blob/v3.0.x/raddb/certs/README#L94
2020-06-11 16:56:13 <dngray> under: 		MAKING A CLIENT CERTIFICATE
2020-06-11 16:56:38 <Han> Yes, I think I misunderstood something.
2020-06-11 17:00:15 <dngray> 🙂 well make sure to update the wiki and help others
2020-06-11 17:02:50 <Han> Good idea.
2020-06-11 17:08:44 <dngray> <Cogitri "You probably have to regenerate "> done.
2020-06-11 17:10:06 <mps> dngray: you don't need to quote messages on irc
2020-06-11 17:25:17 <Hello71> ikke: I feel like the abusefilter should exclude confirmed users?
2020-06-11 17:25:37 <dngray> oh sorry bad habit 😉
2020-06-11 17:25:40 <ikke> Hello71: probably
2020-06-11 17:26:14 <Hello71> unless you trust manually confirmed users enough to make new pages but not enough to add links
2020-06-11 17:26:46 <ikke> I mean, of course it's disirable, but someone needs to implement it / find a plugin for that
2020-06-11 17:29:03 <Hello71> nangel already wrote this abusefilter, you just need to add && (user notin confirmed group) or whatever the syntax is
2020-06-11 17:30:11 <Hello71> wait, mediawiki doesn't come with confirmed group?
2020-06-11 17:30:56 <ikke> only emailconfirmed
2020-06-11 17:31:07 <Hello71> confirmed is more like autoconfirmed
2020-06-11 17:31:07 <ikke> admins, bots, bureaucrats
2020-06-11 17:31:41 <Hello71> https://wiki.alpinelinux.org/wiki/Special:SpecialPages -> https://wiki.alpinelinux.org/wiki/Special:ListGroupRights
2020-06-11 17:31:58 <Hello71> maybe we can move this to gitlab wikis :p
2020-06-11 17:32:24 <ikke> Problem with gitlab wikis (afaik) is that we explicitly need to add each user as a member to a project
2020-06-11 19:33:21 <CrawlCycle> Hi. I tried to follow the "Connecting to a wireless access point" page in the wiki.Right now I have to manually run "rfkill unblock wifi" at every boot.How to use rfkill to unblock wifi before the "automatic configuration of wifi on system boot" happens?
2020-06-11 19:34:51 <CrawlCycle> I tried to create a "/etc/modeprobe.d/rfkill.conf" file. But that didn't work. (https://forums.gentoo.org/viewtopic-t-975200-start-0.html)
2020-06-11 19:48:05 <CrawlCycle> the laptop is a hp dv2000. rfkill list shows "hp-wifi" and "phy0". if i ask rfkill to unblock wifi, the phy0 is also unblocked.
2020-06-11 20:05:43 <CrawlCycle> i can't just get rid of rfkill. "rmmod rfkill" says it is in use by hp_wmi and cfg80211
2020-06-11 20:09:44 <Marquel> CrawlCycle: of course the phy is unlocked. without it, there's no wifi.
2020-06-11 20:11:21 <CrawlCycle> initially phy is blocked. after i ask it to unblock the wifi, both the hp-wifi and the phy0 are unblocked.
2020-06-11 20:12:01 <CrawlCycle> but i still have to manually type "rfkill unblock wifi" and then manually setup wifi at every boot.
2020-06-11 20:12:14 <maldridge> ikke: wiki woes are why void is dropping the wiki
2020-06-11 20:13:20 <ikke> It's working out for Archlinux though (but a larger community)
2020-06-11 20:13:53 <Marquel> CrawlCycle: unblocking wifi is just the first step, you still have to configure it to connect anywhere.
2020-06-11 20:13:57 <CrawlCycle> how to automatically unblock the wifi before the "automatic configuration of wifi on system boot" happens?
2020-06-11 20:14:10 <Marquel> the phy and the wifi are intertwined, one can't exist without the other.
2020-06-11 20:14:32 <Hello71> CrawlCycle: did you try correctly spelling modprobe
2020-06-11 20:15:55 <CrawlCycle> @Hello71 yes. i made a "/etc/modprobe.d/rfkill.conf" with a line that says "options rfkill default_state=1". didn't work.
2020-06-11 20:16:19 <CrawlCycle> @Marquel ok. that make sense. phy and hp-wifi are linked.
2020-06-11 20:17:36 <CrawlCycle> @Marquel I am trying to follow the guide at https://wiki.alpinelinux.org/wiki/Connecting_to_a_wireless_access_point
2020-06-11 20:18:41 <Marquel> CrawlCycle: note: the wifi is the card, the phy is the chip that does the actual work for the card, it "drives the PHYsical layer", so your wired nic also has a phy (in realtek nics that's actually the only chip at all, b/c the rest is done by software in the cpu), but you get the idea.
2020-06-11 20:20:04 <CrawlCycle> I have to unblock wifi before alpine linux tries to setup wifi connection during boot. Somehow "/etc/modprobe.d/rfkill.conf" doesn't work.
2020-06-11 20:23:57 <CrawlCycle> @Marquel interesting. for now I want to install an os that is lighter than lubuntu and is reasonably secure. i don't like the snap thingy in the newer versions of lubuntu.
2020-06-11 20:25:33 <Marquel> i'm afraid i can't help much with modprobe.d/ there, i am not using alpine on a laptop, much less one that brings up the wifi as rfkill'd, that is soft-blocked.
2020-06-11 20:34:55 <CrawlCycle> I changed the "/etc/modprobe.d/rfkill.conf" file to "options rfkill default_state=-1".  "rfkill list" command now says "/dev/rfkill: No such file or directory".
2020-06-11 20:35:11 <CrawlCycle> So the boot process does consider the "/etc/modprobe.d/rfkill.conf" file that I made.
2020-06-11 20:35:35 <CrawlCycle> but i don't know the correct value for "default_state" that would unblock the wifi.
2020-06-11 20:36:20 <CrawlCycle> here is a report that setting "default_state=1" doesn't work. http://forums.debian.net/viewtopic.php?f=5&t=134610
2020-06-11 20:39:00 <mps> modinfo rfkill
2020-06-11 20:39:12 <CrawlCycle> wpa_supplicant says 'nl80211' generic netlink not found. that is not something that I see when i manually unblock wifi and setup connection.
2020-06-11 20:39:31 <CrawlCycle> modinfo rfkill only has one parm
2020-06-11 20:39:57 <CrawlCycle> parm: default_state:Default initial state for all radio types, 0 = radio off (uint)
2020-06-11 20:40:18 <mps> 2
2020-06-11 20:40:41 <izaki> Hi! I am building a customized netboot image, using a customized abuild script. I have included some additional packages using the apks variable, but it is also including packages that I don't need (such as the linux-firwares)... I am trying to find a way to remove those, but I just can't find the right way to do it. Any hints?
2020-06-11 20:40:47 <CrawlCycle> filename: /lib/modules/5.4.43-1-lts/kernel/net/rfkill/rfkill.ko
2020-06-11 20:42:11 <mps> CrawlCycle: ah, this is lts kernel
2020-06-11 20:42:17 <mps> then yes, 1
2020-06-11 20:42:48 <CrawlCycle> @mps After setting "default_state=2" and rebooting, "rfkill list" command still says hp-wifi is soft-blocked and phy0 is hard blocked.
2020-06-11 20:43:18 <mps> I think it is 0 or 1
2020-06-11 20:44:30 <mps> how do you got hp-wifi interface name
2020-06-11 20:44:54 <CrawlCycle> 1 gives same result
2020-06-11 20:45:26 <CrawlCycle> "rfkill list" command says: "0: h-wifi: wlan" and "1: phy0: wlan"
2020-06-11 20:46:18 <CrawlCycle> i can manually unblock with "rfkill unblock wifi". but i want to do that automatically during boot before setting up wifi connection.
2020-06-11 20:46:37 <mps> so, hp-wifi must be something else, bluetooth maybe
2020-06-11 20:47:31 <mps> I had to add 'blacklist asus_nb_wmi' for my asus notebook to work
2020-06-11 20:47:52 <CrawlCycle> after unblocking hp-wifi, wpa_supplicant -i wlan0 -c /etc/wpa_supplicant/wpa_supplicant.conf would work.
2020-06-11 20:48:44 <mps> you can try iwd instead of wpa_supplicant
2020-06-11 20:48:52 <CrawlCycle> some people has the same laptop 10 years ago: it is something like hp
2020-06-11 20:49:06 <CrawlCycle> https://ubuntuforums.org/showthread.php?t=1818252
2020-06-11 20:50:39 <CrawlCycle> @mps if rfkill in the kernel soft-blocks the wifi hardware, i would guess that setting up wifi connections by wpa_supplicant or other means would not work.
2020-06-11 20:51:12 <mps> sure
2020-06-11 20:51:36 <CrawlCycle> if i manually issues "rfkill unblock wifi" and then follow the steps in the wiki (https://wiki.alpinelinux.org/wiki/Connecting_to_a_wireless_access_point), it would work.
2020-06-11 20:52:03 <CrawlCycle> but i don't want to get to root account and type all of that at every boot.
2020-06-11 20:52:07 <CrawlCycle> it is not fun.
2020-06-11 20:52:09 <mps> so you can add this in boot script
2020-06-11 20:52:18 <CrawlCycle> how to do that?
2020-06-11 20:52:46 <mps> read /etc/local.d/README
2020-06-11 20:53:30 <CrawlCycle> does the boot script runs before alpine runs the wpa_supplicant service, or after?
2020-06-11 20:53:34 <mps> or, as I wrote above, you can try iwd
2020-06-11 20:59:03 <CrawlCycle> iwd - rfkill race condition: https://patchwork.kernel.org/patch/11210565/
2020-06-11 21:03:10 <CrawlCycle> Can't find a guide for setting up iwd in alpine linux. Reading /etc/local.d/README
2020-06-11 21:05:03 <CrawlCycle> The readme says: "if a file in this directory is executable and it has a .start extension, it will be run when the local service is started"...
2020-06-11 21:05:19 <CrawlCycle> when is "when"? is it before or after the local service start?
2020-06-11 21:05:54 <CrawlCycle> i guess i can try.
2020-06-11 21:08:05 <Cogitri> The local service runs them
2020-06-11 21:08:15 <Cogitri> So it runs while the service is starting
2020-06-11 21:10:34 <CrawlCycle> i am thinking about making an executable bash script called "wpa_supplicant.starts" that runs "rfkill unblock wifi".Would it run the unblock command before or after it starts the wpa_supplicant?
2020-06-11 21:11:14 <Cogitri> Depends on what runlevel your wpa_supplicant and local service are in
2020-06-11 21:15:31 <CrawlCycle> alternatively, i can setup an executable script that first unblocks wifi and then start the wpa_supplicant deamon.
2020-06-11 21:15:57 <CrawlCycle> that will guarantee that i first unblock wifi and then start the daemon. But how to run this script during boot?
2020-06-11 21:20:51 <Hello71> seems like this is a firmware configuration error
2020-06-11 21:27:18 <CrawlCycle> @Hello71 which firmware?
2020-06-11 21:28:52 <CrawlCycle> i can't find the common vulnerability and exposure (CVE) info for iwd. wpa_supplicant has lots of records in cve database. maybe iwd is more secure?
2020-06-11 21:46:39 <maxice8> It is newer and much less used
2020-06-11 22:23:06 <CrawlCycle> at the end of the boot process, my scripts in /etc/local.d finally runs. i want the scripts to run before the networking service starts.
2020-06-11 22:23:57 <CrawlCycle> have to change the dependency of the scripts in init.d in additional to "rc-update add local default"
2020-06-11 22:28:25 <CrawlCycle> why does openrc always starts the networking service before starting the wpa_supplicant? which configuration file specifies that order?
2020-06-11 22:34:04 <mps> look depend() in some /etc/init.d/ scripts
2020-06-11 22:34:07 <CrawlCycle> "/etc/init.d/wpa_supplicant" files say "depend() { need ... after ... before ...}
2020-06-11 22:34:23 <CrawlCycle> the gentoo wiki doesn't say the meaning of the after and before keywords
2020-06-11 22:34:33 <CrawlCycle> https://wiki.gentoo.org/wiki/OpenRC#Busybox_specific_init.d_files
2020-06-11 22:34:49 <CrawlCycle> and the file says need "localmount"
2020-06-11 22:35:00 <CrawlCycle> not directly "networking"
2020-06-11 22:35:58 <mps> it is not busybox init, but openrc
2020-06-11 22:36:33 <CrawlCycle> i was checking the gentoo wiki's page for openrc
2020-06-11 22:37:13 <CrawlCycle> found it in the man page: http://manpages.org/openrc-run/8
2020-06-11 22:37:49 <mps> I don't know how gentoo use openrc
2020-06-11 22:38:08 <mps> try to find openrc docs on github
2020-06-11 22:38:10 <CrawlCycle> so my script to unblock the wifi should run "before" the networking service starts.
2020-06-11 22:39:20 <mps> I use iwd for more than a year, and I'm not sure when wpa_s should be started
2020-06-11 22:40:04 <CrawlCycle> i can't find a tutorial for using iwd with alpine linux
2020-06-11 22:40:22 <mps> man pages are enough
2020-06-11 22:41:44 <CrawlCycle> erh. reading so many doc pages for openrc already made me dizzy. i guess i will try iwd later in a virtualbox.
2020-06-11 22:42:09 <mps> it have 'before net' in depends
2020-06-11 22:42:47 <CrawlCycle> but wpa supplicant is suppose to run 'after' the networking service starts.
2020-06-11 22:43:03 <CrawlCycle> net and networking probably are different
2020-06-11 22:43:39 <CrawlCycle> i will take a break now. thanks for all the help.
2020-06-11 22:43:55 <mps> networking provides net
2020-06-11 22:44:21 <CrawlCycle> then it should be "after net" not "before net" if my understanding is correct
2020-06-11 22:44:28 <mps> shorthand
2020-06-11 22:44:51 <mps> not sure for wpa_s
2020-06-11 22:46:08 <mps> and you can try to install linux-edge instead of linux-lts
2020-06-11 22:46:34 <mps> maybe it fixes your driver issue, it is 5.6.18 version
2020-06-11 22:46:41 <mps> in testing repo
2020-06-11 23:28:34 <CrawlCycle> net is a module instead of a service
2020-06-11 23:29:12 <CrawlCycle> networking is a service that "provides" the net module
2020-06-11 23:30:38 <CrawlCycle> actually not module, it is a "virtual service" according to man page of openrc
2020-06-11 23:30:51 <CrawlCycle> so i guess it is really a shorthand
2020-06-11 23:31:00 <afraid> Why do you need wpa_supplicant, CrawlCycle?
2020-06-11 23:31:16 <CrawlCycle> but needs to use "after modules net" to refer to it
2020-06-11 23:31:39 <CrawlCycle> @afraid it is a hp dv2000 laptop. and i want to connect to the internet
2020-06-11 23:31:54 <CrawlCycle> rfkill soft-blocks the wifi adapter
2020-06-11 23:32:11 <CrawlCycle> editing the configuration file of rfkill doesn't prevent the soft-block
2020-06-11 23:32:26 <CrawlCycle> i don't want to manually unblock wifi and setup connection at every boot
2020-06-11 23:33:00 <CrawlCycle> so i try to add the "rfkill unblock wifi" line to the script for the wpa_supplicant service
2020-06-11 23:33:28 <CrawlCycle> that nearly works, except somehow that line for unblocking wifi didn't get run before the networking service starts
2020-06-11 23:33:39 <CrawlCycle> don't know what happen there.
2020-06-11 23:34:33 <CrawlCycle> i add the line to the "start_pre()" section of the "/etc/init.d/wpa_supplicant" script.
2020-06-11 23:35:40 <CrawlCycle> the "wpa_supplicant" script defines the wpa_supplicant service and its "depend()" section says it should run "before" the "net" service
2020-06-11 23:36:08 <CrawlCycle> net is actually a virtual service. in script for other services, i see things like "before module net"
2020-06-11 23:36:17 <CrawlCycle> don't know why there is no "module" in this case.
2020-06-11 23:37:28 <CrawlCycle> nvm. "modules" is actually another service instead of a keyword of openrc.
2020-06-11 23:55:34 <CrawlCycle> adding "rfkill unblock wifi" line to the start section of the staticroute, wpa_supplicant, and network service doesn't work.
2020-06-11 23:56:03 <CrawlCycle> i tried to add it as the first line of the start section or start_pre section.
2020-06-12 00:02:02 <afraid> Hmmmm. Is rfkill just autoblocking, or could you modify something that doesn't softblock it?
2020-06-12 00:07:11 <CrawlCycle> rfkill is soft-blocking. if i manually issue "rfkill unblock wifi" and then redo the step in the wireless connection tutorial in the wiki, it works.
2020-06-12 00:08:15 <CrawlCycle> I made a /etc/odprobe.d/rfkill.conf file consists of a line "options rfkill default_state=1"
2020-06-12 00:08:30 <CrawlCycle> but rfkill stills blocks wifi
2020-06-12 00:08:51 <CrawlCycle> if I set the value to -1, i think the rfkill module of the kernel just don't run.
2020-06-12 00:09:08 <CrawlCycle> But set it to 1 or 2 doesn't work.
2020-06-12 00:09:38 <CrawlCycle> To work-around this, i now try to make openrc run "rfkill unblock wifi" for me...
2020-06-12 00:09:59 <CrawlCycle> but adding that line to the staticroute, wpa_supplicant, and networking service still doesn't work
2020-06-12 00:10:46 <CrawlCycle> I have to unblock the wifi before any attempt to connect to the internet through the wifi hardware.
2020-06-12 00:14:37 <bleb> how bout this
2020-06-12 00:14:47 <bleb> you have an alpine chroot directory stored on a nas or other server
2020-06-12 00:15:13 <bleb> then you always sshfs mount it and chroot into it before logging in / starting x
2020-06-12 00:15:48 <CrawlCycle> it is a dv2000 laptop. i want to use the laptop without building a cluster of laptops.
2020-06-12 00:16:19 <bleb> then all your state is stored on one remote system but the programs and shit run on your client machine and are snappy unlike with vnc/rdesktop
2020-06-12 00:16:20 <CrawlCycle> i am quite new at this. have been using lubuntu before.
2020-06-12 00:16:24 <bleb> is this an insane idea
2020-06-12 00:18:15 <CrawlCycle> @bleb does the program have to read/write to the remote hard disk all the time?
2020-06-12 00:20:02 <CrawlCycle> if the program has to do lots of calculation before reading/writing data, then it guess this is ok.
2020-06-12 00:20:10 <afraid> bleb: there's a guy in here who, on boot, pulls a clean build from a local server.
2020-06-12 00:20:27 <afraid> CrawlCycle: the mentioned user in the above message runs entirely in ram
2020-06-12 00:21:16 <bleb> CrawlCycle: well alpine is run-from-ram, so i would think you could be intentional about when to store stuff
2020-06-12 00:21:28 <bleb> afraid: do you know what nick so i can grep?
2020-06-12 00:22:01 <afraid> I don't, but definitely try "from ram" and my nick.
2020-06-12 00:22:14 <afraid> lbu was mentioned without that being the solution.
2020-06-12 00:22:53 <afraid> CrawlCycle: I swear I remember telling you a few days ago that you should read up on how rfkill is initiated and just change it there.
2020-06-12 00:23:29 <CrawlCycle> i just start trying alpiine linux today.
2020-06-12 00:23:54 <CrawlCycle> has been using lubuntu for 6 years
2020-06-12 00:24:34 <CrawlCycle> creating a /etc/modprobe.d/rfkill.conf doesn't work.
2020-06-12 00:30:29 <CrawlCycle> setting "options rfkill default_state=-1" seems to avoid starting the rfkill module in the kernel. However, setting to 1 would still make rfkill block the wifi adapter.
2020-06-12 00:40:59 <lunaphyte> i'm still trying to learn where in setup_disk install_mounted_root() the various runlevels get populated.
2020-06-12 00:41:31 <lunaphyte> when i try to emulate what's done, the runlevels are empty, but when setup-disk is run, they're there
2020-06-12 00:41:58 <CrawlCycle> what is setup_disk?
2020-06-12 00:42:14 <CrawlCycle> i was reading the doc openrc today
2020-06-12 00:42:31 <lunaphyte> the shell script which does the actual "install" of alpine
2020-06-12 00:43:12 <Hello71> lunaphyte: I thought we agreed that they were installed by lbu
2020-06-12 00:49:27 <lunaphyte> Hello71: it doesn't seem to be the case
2020-06-12 00:49:33 <lunaphyte> from my testing
2020-06-12 00:50:02 <CrawlCycle> i define a new "rfkill_unblock_all" service that would run before all other services. depend(){ before * }. it works. but it is a hack.
2020-06-12 00:50:47 <CrawlCycle> "rc-update add rfkill_unblock_all boot"
2020-06-12 00:50:56 <Hello71> lunaphyte: well I would say that assuming lbu commit saves /etc/runlevels, it follows that lbu package does too
2020-06-12 00:51:05 <Hello71> and I'm pretty sure the former is true
2020-06-12 00:51:05 <afraid> why would you not just remove rfkill from openrc? is that not how it's running?
2020-06-12 00:51:24 <CrawlCycle> i can't. the rfkill is a moudle in the kernel.
2020-06-12 00:51:30 <Hello71> yeah, isn't there supposed to be an rfkill module?
2020-06-12 00:51:50 <CrawlCycle> the wifi's driver and other things depends on the rfkill. i can't do that.
2020-06-12 00:52:03 <afraid> Where is rfkill actually being called in your boot process? Did you figure that out?
2020-06-12 00:52:12 <CrawlCycle> it is in the kernel.
2020-06-12 00:52:23 <lunaphyte> Hello71: the only thing i see lbu package used for in install_mounted_root() is to generate /tmp/ovlfiles
2020-06-12 00:52:30 <afraid> No, it isn't called In the Kernel.
2020-06-12 00:52:43 <afraid> Where, during your boot process, is rfkill being invoked to manipulate the devices?
2020-06-12 00:53:02 <afraid> When the kernel is "loaded," it doesn't execute every driver in the kernel. That would be a disaster.
2020-06-12 00:53:07 <Hello71> lunaphyte: it's z*x*v, not z*t*v
2020-06-12 00:53:37 <CrawlCycle> the wifi's driver depends on rfkill. If i don't load rfkill module, i can't load the wifi's driver.
2020-06-12 00:54:23 <Hello71> I know systemd comes with systemd-rfkill
2020-06-12 00:55:17 <Hello71> uh... https://pkgs.alpinelinux.org/contents?file=*rfkill*&path=&name=&branch=edge: "database disk image is malformed" that sounds bad
2020-06-12 00:56:08 <afraid> I did not ask you to unload rfkill's module.
2020-06-12 00:56:21 <afraid> I'm asking you WHERE IN THE BOOT PROCESS rfkill is called that leads to the softlock in the first place.
2020-06-12 00:56:35 <Hello71> afraid: in the firmware.
2020-06-12 00:56:42 <afraid> Please work through the problem in communication before you try to worry about it from there.
2020-06-12 00:56:55 <lunaphyte> Hello71: i don't think i'm following.
2020-06-12 00:56:56 <afraid> Hello71: That's not how that works.
2020-06-12 00:57:01 <Hello71> sign
2020-06-12 00:57:03 <Hello71> sigh
2020-06-12 00:57:14 <lunaphyte> lbu package - | tar -C "$mnt" -zxv > "$ovlfiles" just writes a list of the files.  right?
2020-06-12 00:57:33 <CrawlCycle> @lunaphyte when i visit "https://pkgs.alpinelinux.org/contents?file=*rfkill*&path=&name=&branch=edge" it say something malform.
2020-06-12 00:57:46 <Hello71> lunaphyte: tar -x
2020-06-12 00:57:57 <afraid> answer my goddamned question, crawlcycle.
2020-06-12 00:58:08 <Hello71> afraid: first, calm down.
2020-06-12 00:58:18 <afraid> oh no, he said the god word
2020-06-12 00:58:18 <CrawlCycle> i am a newbie.
2020-06-12 00:58:22 <afraid> better assume he's losing his mind
2020-06-12 00:58:27 <afraid> you're not a newbie to communication.
2020-06-12 00:58:42 <Hello71> second, according to my logs, nobody has mentioned rfkill in this channel since 2020-02-12 until yesterday
2020-06-12 00:58:46 <afraid> you and i were in a direct conversation, both of us using pronouns for each other, directing our attention.
2020-06-12 00:59:18 <CrawlCycle> i searched the user mail list before coming to irc. i didn't find mail about rfkill
2020-06-12 00:59:30 <afraid> i expect you to treat it as such, and not just stop responding and then reassociate randomly to come back in.
2020-06-12 00:59:36 <lunaphyte> Hello71: gah.  it does both.  of course
2020-06-12 00:59:38 <afraid> it's disingenuous.
2020-06-12 01:00:15 <CrawlCycle> i have been searching for a solution for 9 hours today. sorry a bit tired now.
2020-06-12 01:00:28 <CrawlCycle> and i really have just started to try alpine today. so i don't know.
2020-06-12 01:00:38 <afraid> just answer my question: you don't know where it is?
2020-06-12 01:00:40 <CrawlCycle> i have read lots of docs today
2020-06-12 01:01:01 <lunaphyte> Hello71: it packages up the current environment, then immediately dumps it out to the mount point, and also sets aside the list of files it extracted, for use later with apk add [...] --overlay-from-stdin
2020-06-12 01:01:06 <CrawlCycle> no. when i use lubuntu, i just copy fix in the forum...
2020-06-12 01:01:18 <Hello71> lunaphyte: yes
2020-06-12 01:01:18 <lunaphyte> Hello71: have i got that right?
2020-06-12 01:01:22 <lunaphyte> thanks
2020-06-12 01:01:33 <lunaphyte> that's silly i'd not got that
2020-06-12 01:01:53 <Hello71> presumably it would make more sense if there was an "lbu copy" command, but this works fine given the small amount of files
2020-06-12 01:02:17 <Hello71> CrawlCycle: did you check your firmware settings for any "disable wifi"
2020-06-12 01:02:27 <Hello71> or perhaps press some "enable wifi" button on your keyboard
2020-06-12 01:02:57 <CrawlCycle> @Hello71 i didn't touch the wifi switch of the computer. When i run rfkill unblock wifi manually, it works.
2020-06-12 01:03:16 <CrawlCycle> so i think it is purely a software problem not related to the hardware switch.
2020-06-12 01:03:39 <Hello71> if you have a button instead of a switch it can be both
2020-06-12 01:04:03 <Hello71> you press the button, windows turns off wifi and also tells the firmware to turn off the wifi on boot in future
2020-06-12 01:04:17 <CrawlCycle> i heard if windows disable wifi. and then i reinstall linux, it will stay that way.
2020-06-12 01:04:24 <Hello71> could be
2020-06-12 01:04:43 <Hello71> systemd comes with a service to force the setting on boot, even if your firmware is malfunctioning
2020-06-12 01:05:03 <afraid> but systemd also comes with systemd, so that's not a worthwhile solution.
2020-06-12 01:05:08 <afraid> guns come with a method of ending suffering, but you shouldn't use it.
2020-06-12 01:05:20 <Hello71> if my assumption is right that there is no rfkill init script, it's fine to write your own
2020-06-12 01:05:31 <CrawlCycle> the rfkill's default_state setting is suppose to unblock the wifi. but it doesn't work in alpine linux on my laptop.
2020-06-12 01:05:39 <Hello71> it would be better to fix the firmware but it's not a big deal
2020-06-12 01:05:53 <afraid> perhaps the way you're configuring it or placing it is flawed, crawlcycle
2020-06-12 01:06:02 <afraid> you mean the driver for his chip, hello71?
2020-06-12 01:06:08 <Hello71> no, the system firmware.
2020-06-12 01:06:15 <afraid> what system
2020-06-12 01:06:32 <afraid> there are so many firmwares on that board that what you're saying makes zero sense whatsoever.
2020-06-12 01:06:57 <CrawlCycle> @afraid forum of a dozen distros all say the config for the rfkill module is at the same spot.
2020-06-12 01:06:59 <Hello71> afraid: you may be aware that certain system settings are set at boot by the system firmware, and potentially inherited by the OS. for example, the keyboard num lock setting. this is another such setting.
2020-06-12 01:07:08 <afraid> WHICH SYSTEM FIRMWARE
2020-06-12 01:07:13 <CrawlCycle> modinfo says default_state is a "param"
2020-06-12 01:07:18 <afraid> If you mean UEFI/BIOS, say uefi/bios
2020-06-12 01:07:22 <Hello71> "system firmware" has a well-defined meaning. just because you are unaware doesn't mean it doesn't exist.
2020-06-12 01:07:31 <CrawlCycle> so i don't think i create the wrong config file for rfkill.
2020-06-12 01:07:33 <afraid> it literally does not
2020-06-12 01:07:33 <afraid> the bluetooth chip has "system firmware"
2020-06-12 01:07:43 <afraid> the ssd has "firmware"
2020-06-12 01:07:52 <Hello71> this is called "device firmware".
2020-06-12 01:07:58 <afraid> no, it really isn't
2020-06-12 01:08:03 <afraid> not by fuckboys on irc
2020-06-12 01:08:11 <afraid> well-defined doesn't fucking exist here, so when someone says
2020-06-12 01:08:12 <afraid> WHAT SYSTEM
2020-06-12 01:08:14 <afraid> you then say
2020-06-12 01:08:19 <afraid> THE SYNONYM
2020-06-12 01:08:23 <Hello71> http://google.com/search?q=system+firmware
2020-06-12 01:08:24 <afraid> and move the fuck on with a conversation
2020-06-12 01:08:25 <afraid> understood?
2020-06-12 01:08:36 <afraid> google isn't the fucking arbiter of communication here
2020-06-12 01:08:38 <afraid> you and i are
2020-06-12 01:08:39 <Hello71> afraid: kindly calm down or go away.
2020-06-12 01:08:51 <afraid> you're the one who's upset and withdrawing.
2020-06-12 01:08:56 <CrawlCycle> i create a config file for rfkill at /etc/modprobe.d/rfkill.conf. I don't know if alpine linux does this in a different way.
2020-06-12 01:09:00 <afraid> your method of dealing with conflict isn't sustainable, nor is it useful.
2020-06-12 01:09:31 <afraid> crawlcycle: real quick, do file $(which rfkill)
2020-06-12 01:09:36 <afraid> and paste output
2020-06-12 01:09:41 <afraid> since i don't have an alpine box on hand
2020-06-12 01:09:41 <Ariadne> hi
2020-06-12 01:09:55 <Ariadne> what's going on, guys?
2020-06-12 01:10:07 <afraid> hello71 being the same autistic fuckwit he's always been
2020-06-12 01:10:10 <CrawlCycle> @afraid -ash: file: not found
2020-06-12 01:10:15 <afraid> lmfao
2020-06-12 01:10:17 <Hello71> afraid: you seem like you're the type of person that always feels like you're the lone sane one. maybe consider if you're actually the insane one.
2020-06-12 01:10:19 <afraid> apk add file please
2020-06-12 01:10:29 <Ariadne> afraid: ad hominem attacks are a violation of our code of conduct, please refrain from making them.
2020-06-12 01:10:33 <afraid> you pull that stupid bullshit out every time you get called out, hello71
2020-06-12 01:10:40 <afraid> AUTISM ISN'T A FUCKING INSULT
2020-06-12 01:10:46 <afraid> IT'S A REAL FUCKING PROBLEM THAT HELLO71 HAS
2020-06-12 01:10:52 <afraid> QUIT FUCKING SIDING WITH THE PERSON WHO NEEDS HELP
2020-06-12 01:10:56 <Ariadne> "autistic fuckwit" is definitely an ad hominem attack
2020-06-12 01:11:01 <afraid> enablers are the goddamned worst
2020-06-12 01:11:06 <Ariadne> come on this is basic shit
2020-06-12 01:11:10 <afraid> yes, because you've made him into a fuckwit, which also isn't an insult, but a descriptor.
2020-06-12 01:11:15 <afraid> it's not fucking ad hominem
2020-06-12 01:11:17 <bleb> isnt it good to help people who need help
2020-06-12 01:11:21 <afraid> this isn't a fucking douglassian debate
2020-06-12 01:11:33 <afraid> bleb: actually help them, not fucking enable their autism to spiral out of fucking control
2020-06-12 01:11:34 <CrawlCycle> @afraid: the first few characters or /sbin/rfkill is ^?ELF
2020-06-12 01:11:37 <afraid> like the rest of this fucking network
2020-06-12 01:11:46 <Ariadne> anyway, last time you were here you wound up being k-lined by freenode staff.  if you'd like to repeat the cycle, that's fine by me
2020-06-12 01:11:55 <afraid> CrawlCycle: good enough for me. clever workaround. so i guess it's not a busybox thing
2020-06-12 01:12:09 <bleb> sounds like you are both being dipshits
2020-06-12 01:12:26 <Ariadne> pretty much this
2020-06-12 01:12:28 <afraid> Ariadne: I goaded them into fucking klining me, you irresponsible memoryless forticunt
2020-06-12 01:12:53 <afraid> bleb: and the morons with oper refuse to fucking ban hello71 for fistfucking every conversation with his own autistic meandering.
2020-06-12 01:13:08 <bleb> okay
2020-06-12 01:13:11 <Ariadne> anyway, if neither of you have anything productive to say to each other, i suggest not talking to each other
2020-06-12 01:13:14 <afraid> i'm in the middle of fucking helping someone and hello71 comes in to derail literally every bit of it for attention.
2020-06-12 01:13:29 <afraid> i suggest him not fucking continuing to live if his surroundings are going to rot his mind like they've continued to do
2020-06-12 01:13:40 <afraid> to perhaps teach the operators some fucking lessons about how you can't just wait for shit to boil over
2020-06-12 01:13:41 <Ariadne> right
2020-06-12 01:13:49 <afraid> you're literally the equivalent of the minneapolis police department
2020-06-12 01:14:03 <afraid> refusing to address issues until you get to wave your bancock around.
2020-06-12 01:14:15 <Ariadne> thanks for the feedback
2020-06-12 01:14:19 <afraid> maybe someone will fucking loot you and wake you the fuck up.
2020-06-12 01:14:34 <afraid> yeah, totally mature "big person" over here, having to respond with passive-aggression instead of just shutting the fuck up.
2020-06-12 01:14:41 <afraid> definitely the big boy move, ariadne; real fucking cool
2020-06-12 01:14:51 <afraid> you're such a good boy, always mommy's good boy, aren't you?
2020-06-12 01:14:53 <Ariadne> i have a question
2020-06-12 01:15:01 <Ariadne> why do you keep coming here and doing this?
2020-06-12 01:15:17 <afraid> helping people and responding when the same fucking thing happens?
2020-06-12 01:15:29 <Ariadne> calling people "autistic fuckwit" is not "helping"
2020-06-12 01:15:41 <afraid> why the fuck do i keep fucking trying when i'm just going to get fucking FREENODE SPECIAL TISM TIME'd every fucking time i try to work with and against their thorny fucking tism tussles?
2020-06-12 01:15:42 <Ariadne> suggesting that they are better off dead is also not "helping"
2020-06-12 01:15:44 <afraid> YOU FUCKING TELL ME
2020-06-12 01:15:52 <afraid> WHY I KEEP HELPING PEOPLE WHEN NO ONE ELSE WILL FUCKING HELP THEM
2020-06-12 01:16:03 <Ariadne> if this is your idea of "help" it's unwelcome here, please go away
2020-06-12 01:16:04 <afraid> this fucking guy has been here for over nine fucking hours
2020-06-12 01:16:08 <afraid> in some form
2020-06-12 01:16:18 <afraid> yeah, you'd rather this guy fucking lose his mind and ban him later
2020-06-12 01:16:23 <afraid> than to fucking deal with it
2020-06-12 01:16:28 <Hello71> CrawlCycle: in your opinion, is afraid helping?
2020-06-12 01:16:41 <afraid> how do you type so well with your head so far up your fucking ass
2020-06-12 01:16:56 <Ariadne> anyway, i am quite tired of this
2020-06-12 01:17:03 <afraid> yeah, ask now, when he's obviously left to work on it himself and read more docs
2020-06-12 01:17:04 <Ariadne> either follow the CoC or get the fuck out.
2020-06-12 01:17:07 <afraid> when we're not in the middle of me helping him
2020-06-12 01:17:17 <Ariadne> if you need help leaving, i'll be sure to secure that help for you
2020-06-12 01:17:35 <afraid> ariadne: fix your fucking antihuman capital-serving cockstuffed bullshit instead of waving it around as a savior
2020-06-12 01:17:56 <Ariadne> cool
2020-06-12 01:18:06 <Ariadne> it does not change the fact that you can either follow the CoC or get the fuck out.
2020-06-12 01:18:07 <afraid> your shitty cult of moderation and obsession with the banal is inhuman and fucking repetitive garbage
2020-06-12 01:18:25 <afraid> you're STILL hiding behind this fucking pathetic "haha cool yeah sure" habit you've never grown out of
2020-06-12 01:18:36 <afraid> your parents taught you to deal with conflict in a way that they could fucking continue manipulating you
2020-06-12 01:18:42 <afraid> you're codependent with power structures
2020-06-12 01:18:48 <afraid> and no one should have ever shown you a fucking irc client
2020-06-12 01:18:53 <Ariadne> you are sure making a lot of assumptions
2020-06-12 01:19:06 <Hello71> 4/10 troll, amusing but really quite repetitive at this point. might as well be a markov chain
2020-06-12 01:19:14 <afraid> go back to discord and twitter, where your kind can pretend to interface with other people in their own hellscapes on their own
2020-06-12 01:19:17 <Ariadne> but, these assumptions don't change the fact that you can either follow the CoC or leave
2020-06-12 01:19:26 <Ariadne> it is up to you.
2020-06-12 01:19:27 <afraid> i'll be here when you need the MAOI hookup to not fucking kill yourself
2020-06-12 01:19:30 <afraid> you alienated piece of shit
2020-06-12 01:19:41 <opal> 
ACTION pees on afraid

2020-06-12 01:19:52 <afraid> "swallow my CoCk afraid"
2020-06-12 01:19:52 <afraid> "or leave"
2020-06-12 01:20:01 <afraid> fucking operapist.
2020-06-12 01:20:12 <afraid> morning opal
2020-06-12 01:20:16 <opal> lol
2020-06-12 01:20:27 <afraid> where you been?
2020-06-12 01:20:43 <afraid> i also don't get how you don't know ncl
2020-06-12 01:20:51 <afraid> you are literally the same person
2020-06-12 01:20:53 <opal> who says i dont
2020-06-12 01:20:56 <opal> now e arent lol
2020-06-12 01:21:02 <afraid> how not
2020-06-12 01:21:16 <afraid> are you a 6'10" burly lumberjack from the bottom of the ocean
2020-06-12 01:21:47 <afraid> Hello71: grep your shithead logs for recycled content and realize you're the only one providing the same input and expecting a different output
2020-06-12 01:22:20 <opal> idgaf about codes of conduct but i agree in not seeing the productivity of spewing caps at people
2020-06-12 01:22:29 <opal> if theyre really not listening to you then dont talk with them
2020-06-12 01:22:35 <opal> better for everyone' s blood pressure
2020-06-12 01:22:40 <opal> i'd know ive been there
2020-06-12 01:22:53 <afraid> or perhaps blood pressure is an opportunity for someone to fix the problem
2020-06-12 01:22:58 <afraid> perhaps conflict is the only way progress happens
2020-06-12 01:23:08 <afraid> perhaps everyone discovered this and properly understood it as functioning adults
2020-06-12 01:23:15 <opal> in order to help people fix problems they need to be able to help themselves as well
2020-06-12 01:23:17 <afraid> and then realized that freenode was full of hopeless autists and fucking left
2020-06-12 01:23:26 <opal> if you dont think theyre doing that then spend your time on someone who will
2020-06-12 01:23:35 <Ariadne> if freenode is full of "hopeless autists," then why are you still here?
2020-06-12 01:23:36 <afraid> that's itself retarded
2020-06-12 01:23:39 <afraid> yes, it benefits me
2020-06-12 01:23:42 <afraid> but then they're left here to be fucking stupid
2020-06-12 01:23:51 <afraid> they're fully capable of improvement
2020-06-12 01:23:54 <afraid> it just takes longer
2020-06-12 01:24:03 <Ariadne> if the people on this network are beneath you, then your presence here is a waste of our time, and yours.
2020-06-12 01:24:09 <Hello71> afraid: then go ahead and leave. you've been threatening to do so for damn near half an hour, jesus christ
2020-06-12 01:24:18 <afraid> plenty of people have eventually learned on here; i've been on freenode and reconnected plentiful people with reality
2020-06-12 01:24:34 <afraid> hello71: go get plowed from both fucking ends and cum for daddy
2020-06-12 01:24:40 <Ariadne> right
2020-06-12 01:24:43 <afraid> sorry, ariaddy, in this case.
2020-06-12 01:25:02 <afraid> araidne: it's not a matter of fucking hierarchy, you massive simp for fucking power
2020-06-12 01:25:15 <afraid> you absolute dependent fuckboy oper
2020-06-12 01:25:49 <afraid> i don't have to see myself as "better" than you
2020-06-12 01:25:50 <Ariadne> i really don't care about power, but ok
2020-06-12 01:25:55 <afraid> quality doesn't imply magnitude
2020-06-12 01:26:00 <opal> look, i get into arguments and shit slinging, been doing that for years. ask around and youll see agreement
2020-06-12 01:26:05 <opal> been getting incredibly tired of doing that though
2020-06-12 01:26:07 <opal> wears down on me
2020-06-12 01:26:14 <afraid> "i weewy don't cawe about power, but if u don't wanna be nice 2 me, den i'll ban u!!!"
2020-06-12 01:26:15 <c705> Ariadne: it's probably better if you don't even respond, they'll burn themselves out
2020-06-12 01:26:16 <Ariadne> well, you clearly do, since you say freenode is full of "hopeless autists"
2020-06-12 01:26:16 <afraid> FUCKING morons
2020-06-12 01:26:29 <Ariadne> anyway, i have actual work to do
2020-06-12 01:26:32 <afraid> opal: burning yourself out isn't proving that it's bad; it's proving you're doing it wrong
2020-06-12 01:26:34 <Ariadne> freenode staff have been contacted
2020-06-12 01:26:39 <Ariadne> enjoy your k-line when you get it
2020-06-12 01:26:42 <Ariadne> and don't come back
2020-06-12 01:26:44 <Ariadne> thanks
2020-06-12 01:26:48 <afraid> ariadne: consider that "hopeless" works from both ends
2020-06-12 01:26:51 <afraid> I CALLED MOM
2020-06-12 01:26:56 <afraid> MOMMY WILL COME FIX IT ALL SOON
2020-06-12 01:26:58 <Ariadne> hell yeah i called mom
2020-06-12 01:26:59 <opal> not so much burning myself out as much as focusing my attention on more worthwhile shit
2020-06-12 01:27:06 <opal> i even have trouble with the moderation shit
2020-06-12 01:27:11 <afraid> MOM'S COMING ROUND TO PUT IT BACK THE WAY IT OUGHTTA BE
2020-06-12 01:27:15 <opal> im trying to kick people out of spaces less often
2020-06-12 01:27:16 <opal> and deal with the problems in other ways
2020-06-12 01:27:23 <afraid> opal: moving your goalposts
2020-06-12 01:27:25 <afraid> is dumb
2020-06-12 01:27:44 <afraid> "more worthwhile" is just retooling your attention instead of getting better at making people face the fucking music
2020-06-12 01:27:46 <c705> afraid: there's people you can talk to about whatever problems you're going through
2020-06-12 01:27:47 <opal> im here to enjoy chatting not burst a blood vessel
2020-06-12 01:27:51 <opal> dude
2020-06-12 01:27:56 <opal> i get physically stressed over arguments
2020-06-12 01:27:57 <c705> you don't have to be angry on irc
2020-06-12 01:28:00 <opal> i know i cant do this shit
2020-06-12 01:28:02 <afraid> c705: yeah, and those people are hello71. i don't bring other people into the fucking problem space
2020-06-12 01:28:07 <opal> it fucking ruins my health
2020-06-12 01:28:08 <afraid> i don't have to fucking keep living
2020-06-12 01:28:11 <afraid> i do it because it's worth it
2020-06-12 01:28:19 <c705> you're right, it is worth it
2020-06-12 01:28:20 <afraid> opal: you think i'm "bursting blood vessels" though
2020-06-12 01:28:28 <afraid> that's the real point
2020-06-12 01:28:39 <opal> that was figurative. what isnt figurative is the fact i have trouble breathing especially when im stressed out
2020-06-12 01:28:53 <opal> both by constant irl shit and the onslaught of online idiots i have to deal with
2020-06-12 01:29:14 <afraid> if you're getting mad at irc and not just fucking doing what's necessary to instruct the autists to go have some fucking sex and work out having an interpersonal connection for once
2020-06-12 01:29:15 <afraid> then you're doing it wrong
2020-06-12 01:29:21 <afraid> anxiety disorders are fucked, and i haven't ever had one
2020-06-12 01:29:24 <opal> im not 100% out of arguments but i do disengage quite early on compared to what ive been doing before
2020-06-12 01:29:27 <afraid> so i don't have to worry about the same thing other people do
2020-06-12 01:29:39 <opal> i mean
2020-06-12 01:29:42 <afraid> but you also don't have to push anxiety disorder onto others and claim we should all operate within the confines that would make disorder normalized
2020-06-12 01:29:46 <opal> if what youre doing works for you
2020-06-12 01:29:49 <opal> i dont really care
2020-06-12 01:29:52 <afraid> disengaging early makes it worse
2020-06-12 01:29:56 <afraid> that's like smoking weed to get rid of panic
2020-06-12 01:29:57 <opal> take or leave my advice
2020-06-12 01:30:02 <afraid> you just put off the panic until next time
2020-06-12 01:30:04 <afraid> the real advice we need
2020-06-12 01:30:17 <pandemic> I see Gen Z has discovered IRC :(
2020-06-12 01:30:22 <afraid> is how the fuck we sudo ifdown Hello71 from space
2020-06-12 01:30:27 <afraid> pandemic: they've been here long enough.
2020-06-12 01:30:40 <afraid> the discord-ants are rampant
2020-06-12 01:30:41 <pandemic> s/they/we
2020-06-12 01:30:45 <opal> just realise that youre better off putting your theory to the test on a platform that allows it because i know freenode doesnt take kindly to it
2020-06-12 01:30:48 <kplant> why do you all continue to show this person attention?
2020-06-12 01:30:51 <afraid> opal: you ever done any sdr bullshit?
2020-06-12 01:30:52 <opal> whether you agree with the policies or not doesnt really matter
2020-06-12 01:30:58 <D[_]> Ignoring them would be most effective.
2020-06-12 01:30:59 <opal> whats sdr
2020-06-12 01:31:00 <afraid> kplant: because hello71 has literal autism
2020-06-12 01:31:06 <c705> kplant: i like speaking to challenging people sometimes
2020-06-12 01:31:11 <c705> why do anything?
2020-06-12 01:31:12 <TBB> kplant: car crashes are fascinating for some people :)
2020-06-12 01:31:14 <afraid> oh i thought this was just a bunch of opers joining
2020-06-12 01:31:14 <afraid> lmfao
2020-06-12 01:31:21 <afraid> tbb: the car isn't crashing
2020-06-12 01:31:23 <afraid> it's just going really fast
2020-06-12 01:31:27 <afraid> and you know you want to drive it.
2020-06-12 01:31:40 <kplant> c705, tbb: i suppose
2020-06-12 01:31:44 <afraid> some would say the act of witnessing the car ACCELERATING has its own effects
2020-06-12 01:32:12 <c705> afraid: that sounds like a really interesting take on life
2020-06-12 01:32:18 <afraid> it makes you realize what you might be uncomfortable with in the responses of those you've trusted to protect it from crashing, for example.
2020-06-12 01:32:49 <afraid> c705: in self-contained and fragile social spaces  like IRC, it has much more merit than in socially open systems.
2020-06-12 01:32:51 <pandemic> Can we add 'don't feed the troll' in the channel topic?
2020-06-12 01:33:03 <afraid> oh no, i said system in a way that didn't mean a corporately defined product!
2020-06-12 01:33:15 <afraid> hello71's gonna come in and make sure I'm "well-defined"~
2020-06-12 01:33:45 <afraid> can we put "fucking ban both parties in a violation" anywhere in any code ever, if we're going to pretend like we're going to ever solve problems with operdom instead of humanity?
2020-06-12 01:35:00 <pandemic> evidently we need someone to separate the wheat from the chaff
2020-06-12 01:35:10 <Ariadne> we are working on expanding the mod team
2020-06-12 01:35:16 <Ariadne> i have just had more important issues to pursue
2020-06-12 01:35:30 <Ariadne> (and to be clear, my personal desire to moderate any IRC channel is near zero)
2020-06-12 01:35:34 <pandemic> afraid is making a good case for ops
2020-06-12 01:35:38 <Ariadne> indeed
2020-06-12 01:36:06 <opal> i think the commentary against afraid is equally unwelcome
2020-06-12 01:36:45 <afraid> "the wheat from the chaff"
2020-06-12 01:36:54 <afraid> dehumanize black people and put them in prison, pandemic
2020-06-12 01:37:05 <afraid> live off of their literal slave labor and call yourself a functional society.
2020-06-12 01:37:09 <pandemic> LOL
2020-06-12 01:37:21 <pandemic> I've changed my mind
2020-06-12 01:37:25 <pandemic> lets keep him
2020-06-12 01:37:27 <pandemic> he is amusing
2020-06-12 01:37:32 <afraid> yeah, the bureaucracy should just get bigger to deal with the expanding needs of the expanding bureaucracy, ariadne
2020-06-12 01:37:44 <afraid> how about you expand rapidly enough to be classified as a detonation and fucking end the moderator team
2020-06-12 01:37:59 <Hello71> opal: if you're really comparing calling people "challenging" to calling people "absolute dependent fuckboy oper", that's stupid and you know it.
2020-06-12 01:38:12 <opal> im not comparing
2020-06-12 01:38:21 <opal> fighting fire with fire is dumb
2020-06-12 01:38:21 <TBB> pandemic, recognize the movie quote? "Slightly amusing, but mostly painful."
2020-06-12 01:38:24 <opal> thats all i am saying
2020-06-12 01:38:57 <afraid> alright, time's up boys
2020-06-12 01:39:04 <opal> as much a troll as ive been and still can be, i know my way around conflict resolution
2020-06-12 01:39:09 <opal> im not retarded
2020-06-12 01:39:20 <Ariadne> anyway, responding to afraid makes you part of the problem too.  technically, that is a CoC violation as well.
2020-06-12 01:39:21 <afraid> CrawlCycle will have to come back and properly add a certain amount of "pleases" and "thank yous" to every message to deal with INTERCAL71.
2020-06-12 01:39:27 <pandemic> opal: stop feeding the troll
2020-06-12 01:39:30 <afraid> ALL HAIL THE COCK
2020-06-12 01:39:31 <opal> bringing up the coc isnt helping matters either
2020-06-12 01:39:43 <opal> pandemic at this moment im feeding you, take a moment to digest that
2020-06-12 01:39:44 <afraid> THE LIST OF RULES MADE BY ME THAT I USE TO DEPERSONIFY THE RULES
2020-06-12 01:39:52 <afraid> MAKES IT WAYYYY MORE LEGITIMATE AND UNCHALLENGABLE
2020-06-12 01:40:03 <pandemic> opal: i've tried, my system has rejected it
2020-06-12 01:40:04 <afraid> DEFINITELY DOESN'T MAKE ME LOOK CHILDISH AND MALADAPTED
2020-06-12 01:40:19 <afraid> did you try updating the system firmware
2020-06-12 01:40:21 <Hello71> impressive, all caps. I suppose next you'll graduate to unicode blackletter font face to express how much of a big man you are
2020-06-12 01:40:27 <opal> dehumanising people and expecting them to act better out of it is counterproductive. that goes for all sides here.
2020-06-12 01:40:46 <opal> i would know, ive been a fucking target of dehumanisation practicallyall my damned life
2020-06-12 01:40:54 <c705> times up anyways
2020-06-12 01:41:05 <pandemic> we are all God's children
2020-06-12 01:41:07 <pandemic> come friends
2020-06-12 01:41:07 <Ariadne> anyway
2020-06-12 01:41:08 <afraid> yes, i used all caps instead of spongebob case, because i knew it'd trigger opers when it's more "confrontational," despite spongebob case literally being more confrontational and offensive, but more familiar.
2020-06-12 01:41:12 <Ariadne> i have things to do
2020-06-12 01:41:28 <afraid> yeah, you definitely did
2020-06-12 01:41:32 <opal> if you want a better community here it needs to come from all sides
2020-06-12 01:41:33 <afraid> totally definitely needed to disengage
2020-06-12 01:41:38 <afraid> just to come back and bitch for more time, ariadne
2020-06-12 01:41:40 <afraid> you're soooo busy
2020-06-12 01:41:42 <opal> even i have shit to learn still
2020-06-12 01:41:47 <pandemic> lol
2020-06-12 01:41:50 <pandemic> opal is defending the troll...
2020-06-12 01:41:58 <Ariadne> thanks kline
2020-06-12 01:42:10 <Ariadne> i will send you whiskey
2020-06-12 01:42:10 <opal> im not defending anyone
2020-06-12 01:42:14 <opal> im telling all you to fucking can it
2020-06-12 01:42:38 <D[_]> When someone comes to look for a fight and disrupt a channel the worst you can do is buy into it. :(
2020-06-12 01:42:39 <kline> Ariadne: wasnt me
2020-06-12 01:42:42 <kline> im just eating food
2020-06-12 01:42:46 <Ariadne> o
2020-06-12 01:42:47 <Ariadne> wow
2020-06-12 01:42:52 <opal> it says in the quit message your name
2020-06-12 01:42:54 <opal> explain that
2020-06-12 01:42:56 <kline> damn
2020-06-12 01:42:57 <opal> checkmate atheists
2020-06-12 01:42:58 <D[_]> lol
2020-06-12 01:43:00 <Ariadne> anyway, opal is right, responding to that person
2020-06-12 01:43:03 <Ariadne> just makes it worse
2020-06-12 01:43:40 <pandemic> Ariadne: I believe I'm the one who said that
2020-06-12 01:43:41 <Ariadne> so Hello71 i would request you not do so in future
2020-06-12 01:43:46 <pandemic> opal was trying to emphasize his humanity
2020-06-12 01:44:03 <opal> because dehumanising people only turns them further inward
2020-06-12 01:44:13 <opal> he was doing the same shit back at you guys
2020-06-12 01:44:16 <kline> dehumanise yourself and face to klines
2020-06-12 01:44:19 <opal> i wasnt excusing him from that shit
2020-06-12 01:44:37 <pandemic> 21:41 < opal> if you want a better community here it needs to come from all sides
2020-06-12 01:44:40 <pandemic> lol
2020-06-12 01:44:49 <opal> whats false about that
2020-06-12 01:45:14 <kline> pandemic: im not really here to wade in about community standards but 1) he's right, see the famous haskell logs for example and 2) probably best to just leave it off now
2020-06-12 01:45:16 <kline> problems solved
2020-06-12 01:45:17 <pandemic> "We do not negotiate with trolls" - George Bush Jr.
2020-06-12 01:45:22 <kline> why continue to make it a problem
2020-06-12 01:45:46 <opal> >he's right
2020-06-12 01:45:51 <opal> who exactly
2020-06-12 01:45:58 <kline> you
2020-06-12 01:46:08 <opal> forgot an "s" before that then
2020-06-12 01:46:22 <kline> i was thinking of the morgan freeman meme in my head
2020-06-12 01:46:23 <kline> soz
2020-06-12 01:46:29 <opal> kek
2020-06-12 01:46:38 <c705> so hey alpine is a really cool operating system
2020-06-12 01:46:39 <opal> ok you can call me a guy if it means i get his voice
2020-06-12 01:46:47 <kline> fact
2020-06-12 01:47:16 <c705> when is edge getting bumped to v3.13_alpha*
2020-06-12 01:47:21 <c705> looking forward to seeing that
2020-06-12 01:48:04 <Hello71> well, 3.12 was just released
2020-06-12 01:48:23 <Hello71> and according to https://wiki.alpinelinux.org/wiki/Alpine_Linux:Releases, releases are about every 6 months
2020-06-12 01:48:29 <Ariadne> edge may be 3.13 or 4.0 depending on when apk-tools 3 is released
2020-06-12 01:48:42 <c705> oh boy
2020-06-12 01:48:53 <Hello71> so I would assume that preparations for the next release will start at the earliest in four or five months
2020-06-12 01:49:16 <Ariadne> also we are working on setting up CFP for the online conference discussed on alpine-devel
2020-06-12 01:49:22 <Ariadne> that should be done soon, hopefully
2020-06-12 01:49:36 <Ariadne> people have just been busy with higher priority issues
2020-06-12 01:49:43 <c705> of course
2020-06-12 01:51:32 <Ariadne> i've fired off an email to alpine-devel about the moderation issues, lets try to figure out the solution and implement it
2020-06-12 01:55:18 <opal> might follow the thread if i can remember to
2020-06-12 01:58:01 <opal> wouldnt -infra be better to discuss this on
2020-06-12 01:58:07 <opal> or is irc not considered infra
2020-06-12 01:59:22 <opal> oh man looking at the infra list, lots of multi-million offers
2020-06-12 01:59:24 <opal> cant turn those down
2020-06-12 01:59:40 <Hello71> I think it is a question of the whole project. for example, a proposal to move from gitolite to gitlab would be discussed on -devel, and then if it was agreed on, the exact technical details of how to move would be discussed on -infra
2020-06-12 01:59:56 <opal> mm
2020-06-12 02:03:23 <Ariadne> correct
2020-06-12 02:03:55 <Ariadne> the infra team reacts based on guidance from the core team or the project itself (either by rough consensus or an outright vote)
2020-06-12 02:04:13 <opal> oh its team based moreso than topic based
2020-06-12 02:04:16 <opal> i see
2020-06-12 02:04:30 <opal> not 100% familiar with how alpine's organised even still
2020-06-12 02:06:10 <Hello71> pretty sure other projects (not just linux distros, but others too) also make this split. consider infra as "IT department" in a traditional company
2020-06-12 02:06:34 <opal> yeah
2020-06-12 02:07:20 <opal> so next question is what would adding freenode-staff give for the channel, seems like the disruptions that happen here are to the scale of it not being acceptable on the network at all
2020-06-12 02:13:06 <Ariadne> it's an interim thing, an invitation for staff to join in and deal with this stuff
2020-06-12 02:13:18 <programmerq> usually these situations start smaller and tend to escalate to the point that it violates the freenode network ToS. Action taken by channel ops early on usually keeps things from devolving to that point.
2020-06-12 02:13:21 <opal> ic
2020-06-12 02:13:32 <opal> ye makes sense
2020-06-12 02:14:18 <opal> and lastly the selection criteria for new ops in here
2020-06-12 02:14:25 <opal> im sure mostly a timezone requirement
2020-06-12 02:14:33 <Ariadne> that's the point of the email
2020-06-12 02:14:40 <Ariadne> who do you want to moderate
2020-06-12 02:14:41 <opal> ah just seeing whos available when
2020-06-12 02:14:50 <Ariadne> who and do we want those people
2020-06-12 02:14:55 <D[_]> To be honest having more active ops would be more beneficial.
2020-06-12 02:15:26 <programmerq> I know I don't participate in the actual alpine development stuff, but I've been an op on another channel for several years. Adding #alpine-linux to my list of stuff to keep an eye on would be trivial.
2020-06-12 02:15:32 <Ariadne> anyway, I am up to my neck in shit to do so sorry for not taking action sooner
2020-06-12 02:15:42 <opal> what im getting at and i should just say it, is do you have anyone in mind to suggest yourself, Ariadne
2020-06-12 02:15:45 <opal> i feel that'd be a good start
2020-06-12 02:15:52 <opal> oh go do that if you need
2020-06-12 02:16:29 <Ariadne> opal I think at minimum any project member with developer privilege should be able to op themselves if needed
2020-06-12 02:16:45 <opal> sounds fair
2020-06-12 02:16:52 <Ariadne> but good moderation is obviously more than that
2020-06-12 02:16:58 <Ariadne> beyond that I don't have the answer
2020-06-12 02:17:00 <opal> yes
2020-06-12 02:17:04 <Ariadne> moderation requires consent
2020-06-12 02:17:11 <Ariadne> consent requires trust
2020-06-12 02:17:31 <Ariadne> thus it is appropriate to ask the community who they wish to be moderated by
2020-06-12 02:17:38 <opal> yeah exactly
2020-06-12 02:17:52 <Ariadne> devs don't have time to moderate anyway
2020-06-12 02:17:53 <opal> k thats all i had to ask
2020-06-12 02:17:55 <opal> i'll follow the discussion
2020-06-12 02:18:00 <Ariadne> I see that as a last resort option
2020-06-12 02:18:10 <Ariadne> like in a pinch I can op myself and ban that dude
2020-06-12 02:18:20 <Ariadne> or maxice8 or whoever
2020-06-12 02:18:53 <opal> ya i think activity would be a good requisite for ops
2020-06-12 02:19:21 <opal> not just on call but active in here enough not to need to be explicitly called on
2020-06-12 02:19:22 <opal> i suppose
2020-06-12 02:19:22 <Greyfn> anybody have any experience with aports-turbo?
2020-06-12 02:19:41 <Ariadne> I have several other resolutions I am working on right now
2020-06-12 02:19:55 <Ariadne> team maintenance, cross pollination and abuild changes
2020-06-12 02:22:23 <Greyfn> i'm getting issues right out of the gate running aports.lua: "luajit: attempt to index a number value"
2020-06-12 02:22:59 <Greyfn> I'm guessing it has to do with the fact I'm on edge; does anybody in-the-know what is the latest supported release for aports-turbo?
2020-06-12 02:23:09 <Greyfn> what does pkgs.alpinelinux.org run on?
2020-06-12 02:38:23 <Ariadne> there's a docker image on gitlab
2020-06-12 02:39:33 <Ariadne> https://gitlab.alpinelinux.org/alpine/infra/docker/aports-turbo
2020-06-12 02:40:05 <Greyfn> I'll take a look at that and see how it spins up
2020-06-12 02:40:05 <Greyfn> thanks
2020-06-12 03:49:03 <CrawlCycle> Does alpine linux's kernel have the selinux module?
2020-06-12 03:50:43 <Hello71> probably not
2020-06-12 03:50:58 <Hello71> https://git.alpinelinux.org/aports/tree/main/linux-lts/config-lts.x86_64
2020-06-12 03:51:10 <Hello71> apparently it comes with apparmor and yama
2020-06-12 03:52:14 <CrawlCycle> but it should be part of the kernel. so other non-kernel packages can't contain the selinux module.
2020-06-12 03:52:48 <CrawlCycle> oh. my bad. let me read that.
2020-06-12 03:53:38 <CrawlCycle> line 8138: CONFIG_SECURITY_SELINUX is not set
2020-06-12 03:53:48 <CrawlCycle> https://git.alpinelinux.org/aports/tree/main/linux-lts/config-lts.x86_64#n8138
2020-06-12 03:55:22 <CrawlCycle> why does the edge branch of the community repository has a libselinux package? https://pkgs.alpinelinux.org/package/edge/community/x86/libselinux
2020-06-12 03:57:00 <CrawlCycle> line 8145: CONFIG_SECURITY_APPARMOR=y. https://git.alpinelinux.org/aports/tree/main/linux-lts/config-lts.x86_64#n8141
2020-06-12 03:57:39 <Hello71> dunno, maybe you can use it if you compile your own kernel?
2020-06-12 03:58:15 <Hello71> I guess you can ask ncopa when he comes back
2020-06-12 04:00:08 <CrawlCycle> thanks. i will think about this.
2020-06-12 04:02:18 <Hello71> selinux provides questionable security on single-user machines anyways
2020-06-12 04:03:03 <Hello71> I recommend firejail for single-user machines. there is some risk of privilege escalation to root, but usually on single-user machine you already enable sudo anyways
2020-06-12 04:14:14 <lunaphyte> uh oh.  i broke pkgs.alpinelinux.org :)
2020-06-12 04:14:40 <lunaphyte> https://pkgs.alpinelinux.org/contents?file=*protected_paths.d&path=&name=&branch=edge&arch=x86_64
2020-06-12 04:24:44 <CrawlCycle> Does the apparmor module in alpine linux's kernel support network rules, dbus rules, and unix socket rules?
2020-06-12 04:36:20 <Ariadne> libselinux exists because docker
2020-06-12 04:39:38 <CrawlCycle> i haven't used docker regularly. why would libselinux be useful in that case?
2020-06-12 04:40:04 <CrawlCycle> centos?
2020-06-12 04:42:44 <dngray> Cogitri: i got that done https://gitlab.alpinelinux.org/dngray/aports/-/merge_requests/1
2020-06-12 04:42:45 <dngray> has anyone successfully got uid mapshifting lxd on alpinelinux
2020-06-12 04:43:02 <dngray> as in https://www.linuxhowto.net/how-to-add-or-mount-directory-in-lxd-linux-container/
2020-06-12 04:43:04 <dngray> https://ubuntu.com/blog/custom-user-mappings-in-lxd-containers
2020-06-12 04:43:29 <dngray> i have had some issues with this, ill post on the mailing list hopefully someone can help me
2020-06-12 04:43:31 <dngray> i think it might be some problem related to openrc or musl
2020-06-12 04:43:54 <dngray> it certainly works on ubuntu... but i am not changing my server to use ubuntu
2020-06-12 05:38:21 <Cogitri> dngray: That merge request is made against your fork, but it needs to be done against the upstream repo ( so alpine/aports instead of dngray/aports )
2020-06-12 05:51:34 <dngray> oh
2020-06-12 05:51:46 <dngray> do i need to do anything?
2020-06-12 05:52:02 <ikke> change the merge target for the merge request
2020-06-12 05:53:59 <dngray> oh
2020-06-12 05:58:45 <dngray> Cogitri: ugh i am such a gitlab neooob
2020-06-12 05:59:07 <dngray> https://gitlab.alpinelinux.org/dngray/aports/-/merge_requests/1
2020-06-12 05:59:10 <dngray> i was sure i opened that against alpine/aports
2020-06-12 06:00:03 <dngray> showing a heap of changes https://gitlab.alpinelinux.org/dngray/aports/-/merge_requests/1/diffs
2020-06-12 06:00:32 <dngray> that i did not make
2020-06-12 06:02:40 <dngray> oh i rebased it and squashed cos it seems gitlab doesn't do that
2020-06-12 06:06:23 <dngray> https://gitlab.alpinelinux.org/alpine/aports/-/merge_requests/9166 think i've done that
2020-06-12 06:06:25 <dngray> i actually thought i'd already done this, given the time i had forgotten
2020-06-12 07:32:10 <ncopa> <CrawlCycle
2020-06-12 07:32:37 <ncopa> selinux userspace is there because someone requestsed it
2020-06-12 07:32:57 <ncopa> it is not uncommon to run alpine userspace in docker images on centos or redhat kernels
2020-06-12 07:34:24 <ikke> yes, we use it in a docker pipeline
2020-06-12 07:34:43 <ikke> Not all tools require kernel support
2020-06-12 08:03:28 <MathiasHMK> Morning guys :) Where is the best place to ask advice for adding a new kernel module to Alpine?
2020-06-12 08:04:49 <ikke> #alpine-devel probably
2020-06-12 08:05:00 <ikke> If you want to add it yourself
2020-06-12 08:06:46 <MathiasHMK> Well. I think I'm not the only one who needs to use CAN in an embedded system, but I don't find it likely that it can be a part of Alpine by default. Anyways, thanks :=)
2020-06-12 08:15:28 <mps> MathiasHMK: yes, CAN should be enabled, and I think it is enabled for armv7 kernels
2020-06-12 08:28:04 <rnkn> anyone installed Alpine on a Pinebook?
2020-06-12 08:56:00 <TBB> have you asked this before?
2020-06-12 08:56:19 <TBB> because I seem to remember someone
2020-06-12 08:56:36 <rnkn> TBB: me? no I've been asking a lot of pesky Raspberry Pi q's though
2020-06-12 08:57:35 <nacelle> http://images.postmarketos.org/pinebookpro/
2020-06-12 08:57:54 <nacelle> the answer is "yes, it runs alpine", in any event
2020-06-12 08:57:57 <TBB> okay, thought I've seen you before. I did a quick grep of this year's logs and there's been discussion about pinebook and alpine
2020-06-12 08:58:54 <rnkn> TBB: I'm only eyeing a Pinebook at this stage
2020-06-12 08:59:04 <rnkn> nacelle: sweet thanks!
2020-06-12 08:59:18 <rnkn> I'd seen mention of postmarketOS but thought it was for phones
2020-06-12 09:01:21 <nacelle> the key is postmartketOS is based on alpine
2020-06-12 09:01:32 <rnkn> crafty
2020-06-12 10:23:19 <int_main> hey everyone, i'm having some problems with running alpine on a rpi
2020-06-12 10:23:36 <int_main> i look at /etc/mdev.conf and see that it says "vchiq   root:video 0660"
2020-06-12 10:23:49 <int_main> but the actual permissions for /dev/vchiq are root:root 0600
2020-06-12 10:28:04 <int_main> ah, i see, mdev isn't used when udev is
2020-06-12 10:29:07 <dngray> Cogitri: woo i just finished setting up alpinelinux and lxd
2020-06-12 10:29:32 <dngray> and its working wonderfully, all the id raw mapping and everything 😁
2020-06-12 10:49:16 <ash_worksi> `apk -U add man` yeilds: `ERROR: unsatisfiable constraints: man (missing): required by: world[man]` ... did this just get removed from pkgs.alpine-linux.org
2020-06-12 10:49:19 <ash_worksi> ?
2020-06-12 10:50:09 <ash_worksi> I guess it's not in >3.11 ?
2020-06-12 11:00:37 <ikke> ash_worksi: there are 2 packages providing it now
2020-06-12 11:00:44 <ikke> you most likely want mandoc
2020-06-12 12:30:32 <ash_worksi> ikke: 2 packages as in, one for 3.11 and one for 3.12?
2020-06-12 12:31:05 <Cogitri> ash_worksi: No, two packages in your release
2020-06-12 12:31:15 <Cogitri> apk can't decide which one to pick, so you have to explicitly install one of them
2020-06-12 12:31:37 <ash_worksi> Cogitri: but `man` doesn't show up on pkgs.alpinelinux.org for 3.12
2020-06-12 12:32:23 <ikke> it's a virtual package
2020-06-12 12:33:02 <ash_worksi> so, if I wanted to add man (over mandoc for some reason) I would have to use --virtual ?
2020-06-12 12:33:15 <ikke> no
2020-06-12 12:33:38 <ikke> There is mandoc or man-db
2020-06-12 12:33:40 <ikke> both provide man
2020-06-12 12:34:01 <ash_worksi> I saw man-db
2020-06-12 12:34:13 <ash_worksi> although I am not sure exactly what that means
2020-06-12 12:36:45 <ash_worksi> ikke: so a virtual package means it does not exist by itself, but only as a dependency to another package?
2020-06-12 13:03:01 <Hello71> ikke: can you check out the pkgs.a.o issue?
2020-06-12 13:03:17 <ikke> later
2020-06-12 13:03:49 <Hello71> ok, np. just checking you are aware
2020-06-12 13:03:57 <ikke> I wasn't :)
2020-06-12 13:15:02 <ikke> Hello71: so what's the issue with pkgs.a.o?
2020-06-12 13:15:54 <Hello71> certain types of queries return "database disk image is malformed". example: https://pkgs.alpinelinux.org/contents?file=*aaaaaaaaaa*&path=&name=&branch=edge
2020-06-12 13:16:24 <ikke> Ok, would you be able to create an issue here? https://gitlab.alpinelinux.org/alpine/infra/infra
2020-06-12 13:16:38 <Hello71> which sounds scary but I'm inclined to guess that this does not actually mean anything serious, because most normal queries seem to work fine
2020-06-12 13:16:52 <Hello71> ok. was hoping it was an easy fis :p
2020-06-12 13:17:09 <ikke> Perhaps, but I have to dive into it
2020-06-12 13:18:18 <Hello71> s/fis/fix/
2020-06-12 13:38:59 <localmans> why does trying to run an appimage give me "no such file or directory" errors?
2020-06-12 13:39:12 <ikke> "an app image">
2020-06-12 13:39:14 <ikke> ?
2020-06-12 13:39:37 <localmans> chmod +x Downloads/krita-4.2.9-x86_64.appimage
2020-06-12 13:39:53 <localmans> ./<alt-.>
2020-06-12 13:39:54 <Cogitri> localmans: https://github.com/AppImage/AppImageKit/issues/877
2020-06-12 13:39:59 <Cogitri> AppImages don't work on musl (yet)
2020-06-12 13:40:06 <Cogitri> They assume glibc is the libc
2020-06-12 13:40:24 <localmans> Cogitri: Why does it say "no such file or directory," then?
2020-06-12 13:41:07 <Cogitri> Because it looks for the glibc dynamic loader
2020-06-12 13:41:19 <Cogitri> Which doesn't exist because Alpine uses musl
2020-06-12 13:41:32 <Cogitri> Why don't you install the Alpine package for krita?
2020-06-12 13:41:35 <localmans> And then that makes BASH error that the directory doesn't exist?
2020-06-12 13:41:49 <localmans> Why don't I? I did, and it hangs on some futex
2020-06-12 13:41:58 <localmans> So then I figured I'd check if the appimage did the same, leading to this.
2020-06-12 13:42:07 <ikke> It's trying to find the linker
2020-06-12 13:42:10 <localmans> No, it's not an xy problem, because I wanted to know why bash would outright lie
2020-06-12 13:42:32 <localmans> so the real problem is bash not reporting what file it's actually looking for, which confuses the user?
2020-06-12 13:42:42 <Cogitri> It's not bash lying, literally any shell will tell you this
2020-06-12 13:42:42 <ikke> try `ldd Downloads/krita-4.2.9-x86_64.appimage | head -n1`
2020-06-12 13:44:04 <localmans> is that l or one
2020-06-12 13:44:12 <ikke> one
2020-06-12 13:44:26 <localmans> Ah, better, ld-musl-x86_64.so.1 says no such file or directory
2020-06-12 13:44:34 <localmans> still reporting it in a form that is really annoying, though
2020-06-12 13:44:44 <Cogitri> Yes, it is pretty confusing
2020-06-12 13:44:48 <ikke> yes, it is
2020-06-12 13:44:49 <localmans> "cannot load <filename>: No such file or directory"
2020-06-12 13:44:53 <Cogitri> But I'm not sure if something better can be done
2020-06-12 13:45:06 <localmans> is that bash's fault, or where does that error style come from
2020-06-12 13:45:24 <ikke> It comes from execve
2020-06-12 13:45:28 <ikke> afaik
2020-06-12 13:45:32 <Hello71> unfortunately this is the behavior of literally every linux program
2020-06-12 13:45:43 <ikke> bash tries to execute the file
2020-06-12 13:45:44 <Hello71> it calls exec*, which calls execve, which returns ENOENT
2020-06-12 13:45:49 <ikke> indeed
2020-06-12 13:46:00 <localmans> so to change that text, we have to beat up linus himself
2020-06-12 13:46:19 <Hello71> there is no additional information available to userspace unless it tries to manually investigate the file
2020-06-12 13:46:57 <localmans> i just want it to say "can't execute this file: a file required can't be found"
2020-06-12 13:47:05 <localmans> but anyway, now on to krita being broken
2020-06-12 13:47:25 <ikke> You have to report it to Linux :)
2020-06-12 13:48:07 <localmans> krita simply halts for me on edge; does anyone else get this?
2020-06-12 13:49:14 <localmans> halt is wrong; stall is the right term
2020-06-12 13:49:31 <Cogitri> Just checked it, hangs for me as well
2020-06-12 13:49:58 <Cogitri> Maybe the openexr rebuild broke that?
2020-06-12 13:50:07 <localmans> openexr?
2020-06-12 13:50:22 <Cogitri> A dependency of krita which recently got an upgrade
2020-06-12 13:51:04 <Cogitri> Sorry, was just thinking out loudly :)
2020-06-12 13:51:29 <Cogitri> I'll see if I can find why it hangs and give you a ping then, localmans
2020-06-12 13:51:37 <localmans> thx cogitri
2020-06-12 13:52:41 <artok> openexr is only file format you'd need for picture processing
2020-06-12 13:52:44 <artok> it is marvelous
2020-06-12 14:06:52 <Cogitri> localmans: https://gitlab.alpinelinux.org/alpine/aports/-/issues/11641
2020-06-12 14:07:32 <localmans> same futex call for me, too
2020-06-12 14:07:35 <localmans> at least it's deterministic
2020-06-12 14:08:04 <ikke> did you run strace with -f?
2020-06-12 14:08:38 <Cogitri> Nop, let me do that
2020-06-12 15:54:57 <opal> seeing the talk about maintainership groups/teams and their relation to bureaucracy, and i cant help but think that the idea of a central all-encompassing package repository is plenty bureaucratic in itself. goes for any distro
2020-06-12 15:55:38 <Cogitri> Contributing to aports is next to none bureaucracy compared to contributing to Debian or Fedora though
2020-06-12 15:55:40 <opal> i know its more nuanced than "offload to third party repos" but honestly i prefer that approach at least in the systems i operate
2020-06-12 15:55:45 <opal> no i know
2020-06-12 15:56:01 <opal> i have a radical mindset when it comes to centralisation of any type
2020-06-12 15:56:07 <Cogitri> Looking at the quality of some PKGBUILDs in the AUR makes me shudder
2020-06-12 15:56:07 <opal> so take what i say with a grain of salt
2020-06-12 15:56:25 <Cogitri> But it's nice how much software is available on Arch,y es
2020-06-12 15:56:42 <opal> the aur is equally a central repo, and it tends to give the false impression that theres  QC or at least consistency when there isnt
2020-06-12 15:57:16 <Cogitri> Split repos tends to be even more work (see e.g. Gentoo's overlays) in my experience though
2020-06-12 15:57:40 <Cogitri> Because now you have conflicts between the different repos, everyone does things differently and you have to search for that one nice repo that actually does what you want
2020-06-12 15:57:42 <opal> i actually really prefer overlays compared to everyone else's way of doing it
2020-06-12 15:57:51 <opal> maybe im a minority opinion
2020-06-12 15:58:09 <Cogitri> A collaborative approach results in more quality  at the end of the day instead of everyone cooking their own soup
2020-06-12 15:58:31 <opal> but gentoo's treatment of overlays as first-class even though they arent officially supported, makes things a lot smoother than arch's AUR utilities
2020-06-12 15:58:56 <opal> emerge eselect-repository, eselect repository enable whatever, emerge foobar::whatever, call it a day
2020-06-12 16:00:36 <opal> yeah i see it from both sides but i'm personally biased because i see less merit in foss if we end up trusting everything but our own eyes
2020-06-12 16:00:49 <opal> im not trying to sway alpine, i can easily add my own third-party repos just fine right now
2020-06-12 16:01:00 <opal> just openly musing about it
2020-06-12 16:01:00 <Cogitri> Does portage automatically list what repos a package is in by now?
2020-06-12 16:01:13 <Cogitri> Back when I tried it overlays were pretty messy
2020-06-12 16:01:22 <opal> how do you mean? e.g. with emerge --search?
2020-06-12 16:01:27 <Cogitri> Exherbo had a pretty neat implementation of multiple repos
2020-06-12 16:01:47 <opal> yeah exherbo treats overlays even more first-class than gentoo
2020-06-12 16:01:55 <opal> which is why i kind of want to give it a spin
2020-06-12 16:02:08 <mps> opal: I also dislike centralized bureaucracy but something have to be done for more controlled development
2020-06-12 16:03:16 <opal> Cogitri it seems like eix trumps emerge here, `emerge --search` doesnt indicate overlay membership at all
2020-06-12 16:03:25 <opal> portage is incredibly inconsistent, i dislike it
2020-06-12 16:03:27 <mps> I think we should have some kind of policy documents than teams
2020-06-12 16:03:55 <opal> atoms are =category/package-version:slot::overlay yet we barely see that syntax consistently used
2020-06-12 16:04:20 <mps> but teams are ok if they are formed for things where they needed
2020-06-12 16:07:44 <Hello71> I don't like overlays mainly because the tools for tracking upstream changes are basically non-existent
2020-06-12 16:07:59 <Hello71> also I think they interact poorly with eclasses etc, which gentoo also has way too much of
2020-06-12 16:08:09 <opal> mm
2020-06-12 16:09:06 <Hello71> I've been musing about better ways to represent ebuild structure in git, something that is more amenable to local changes (right now the default merge does a terrible job)
2020-06-12 16:09:29 <Hello71> exherbo etc have a different structure, but afaik it's not a big improvement on that front?
2020-06-12 16:10:14 <opal> local changes im assuming meaning you want to manually edit an ebuild in /var/db/repos/?
2020-06-12 16:11:42 <opal> if you do that in gentoo repo wouldnt a sync just silently overwrite your local changes?
2020-06-12 16:11:54 <opal> is this desired behaviour in your opinion? because i see it being hairy
2020-06-12 16:12:07 <opal> i'd rather have a local overlay for quick changes
2020-06-12 17:21:03 <Hello71> i'm saying for ebuild patches that you want to keep
2020-06-12 17:51:40 <opal> oh you want to utilise git more directly when syncing?
2020-06-12 17:52:14 <opal> the shallow clone logic is weird with git repo syncs
2020-06-12 17:52:40 <opal> since by default i think only clone is shallow and then subsequent pulls take in rest of history
2020-06-12 17:54:05 <albr_> Iron Oxidizer[d]: Anyone know how to do a sys install to a custom partition layout rather than nuking the entire drive?
2020-06-12 17:56:45 <mps> MathiasHMK: does this look ok? https://gitlab.alpinelinux.org/alpine/aports/-/merge_requests/9183
2020-06-12 17:59:07 <Cogitri> mps: You need a logical connection between the two licenses, so an AND or OR
2020-06-12 17:59:48 <mps> ah, thanks
2020-06-12 18:00:21 <mps> I have to look a little more on this but I think it is AND
2020-06-12 18:00:56 <mps> apkbuild-lint was ok with current
2020-06-12 18:02:29 <Cogitri> Probably something that should be added to apkbuild-lint
2020-06-12 18:02:59 <mps> AI lawyer :)
2020-06-12 22:14:04 <avbox111> What is the easiest gui for wifi setup?
2020-06-12 22:14:41 <Cogitri> I guess NetworkManager
2020-06-12 22:16:38 <avbox111> Cogitri: And how do I call the gui from command line?
2020-06-12 22:18:00 <Cogitri> ?
2020-06-12 22:18:28 <opal> nm-applet i think
2020-06-12 22:19:55 <avbox111> opal: I tried nm-applet from command line (mate desktop), but nothing happend. Any ideas?
2020-06-12 22:20:12 <opal> unsure then
2020-06-12 22:23:37 <maxice8> I don't think nm has its own GUI, it has nmcli and nmtui
2020-06-12 22:23:56 <maxice8> And GNOME has their own interface which I assume works on libnm
2020-06-12 22:29:12 <avbox111> maxice8: libnm, what package is it?
2020-06-12 22:30:33 <maxice8> networkmanager provides libnm
2020-06-12 22:31:45 <avbox111> maxice8: Ok, but how can I the gui? Do I need to configure something in /etc/NetworkManager/NetworkManager.conf?
2020-06-12 22:32:47 <maxice8> get network-manager-applet and run nm-connection-editor
2020-06-12 22:35:13 <avbox111> maxice8: nm-connection-editor works, but it does not show available hotspots (like wicd). Isn't it possible to scan the routers (scan button does not bring any routers up)?
2020-06-12 22:36:00 <maxice8> No clue, sorry
2020-06-12 22:36:08 <maxice8> i use GNOME which has their own interface that uses NetworkManager
2020-06-12 22:37:35 <avbox111> maxice8: And in gnome nm-applet does work? Just that I know it could be a mate problem.
2020-06-12 22:37:57 <TBB> it's possible allright, I ported nm-applet to Alpine for a project and got it to work just fine
2020-06-12 22:38:24 <TBB> it just requires some additional configuration, the details of which I've long since forgotten
2020-06-12 22:38:36 <maxice8> GNOME has their own tool to use instead of nm-applet
2020-06-12 22:40:34 <TBB> looks like the same thing to me
2020-06-12 22:40:48 <TBB> altho, it's gnome, who knows
2020-06-12 22:41:29 <maxice8> well it mostly does the same stuff, manage networks
2020-06-12 22:41:40 <maxice8> it is just integrated into GNOME, accessible from GNOME settings in the Networking section
2020-06-12 22:41:49 <maxice8> and obeys GNOME's HIG
2020-06-12 22:44:02 <avbox111> maxice8: Is there a gnome version of postmarketos for pinephone?
2020-06-12 22:44:29 <maxice8> I wouldn't know, you'd have to ask in the postmarketOS communication channels
2020-06-12 22:45:39 <afontain_> there is
2020-06-12 22:47:23 <afontain_> you can simply choose gnome ui on pmbootstrap init. ui are independent from devices, some may work poorly though.
2020-06-12 22:47:33 <afontain_> the postmarketOs channels if you have other questions: https://wiki.postmarketos.org/wiki/Matrix_and_IRC
2020-06-12 22:47:41 <avbox111> maxice8: Thanks, I ask in postmarketos
2020-06-12 22:56:27 <opal> "i'm a kde man myself"
2020-06-12 22:56:46 <maxice8> A Kool man
2020-06-12 22:56:56 <opal> not gonna lie kde is alright
2020-06-12 22:57:04 <opal> i just cant get myself to adapt that workflow
2020-06-12 22:58:39 <TBB> there's two reasons to use KDE
2020-06-12 22:58:44 <TBB> kate and konsole
2020-06-12 22:59:31 <TBB> and perhaps kdevelop if development is your thing
2020-06-12 23:00:25 <opal> im using neovim and st so that should give you an idea what preferences i have
2020-06-12 23:01:35 <TBB> I get it, I fluctuate between KDE, Xfce and just plain i3wm.
2020-06-12 23:05:53 <maxice8> fancy
2020-06-12 23:49:01 <Greyfn> I asked yesterday if anybody has any experience with aports-turbo; I'm still having difficulties
2020-06-12 23:49:20 <Greyfn> here is the error I'm getting: https://pastebin.com/11TcgyDQ
2020-06-12 23:49:36 <Greyfn> same error if run directly or via the provided docker container
2020-06-12 23:49:48 <Greyfn> any pointers, anybody?
2020-06-12 23:50:37 <Greyfn> I'm ignoring the warning, I'm more concerned about the `attempt to index a number value` line
2020-06-13 01:23:26 <Greyfn> solved my issues in case anybody has any future issues; aports-turbo works fine on alpine 3.8, I'm guessing it has something to do with the version of lua-turbo used
2020-06-13 01:31:31 <Greyfn> issue submitted: https://gitlab.alpinelinux.org/alpine/infra/aports-turbo/-/issues/41
2020-06-13 01:33:20 <rafaelsoaresbr> hello guys, I'm trying to set notime option in /etc/fstab for root partition, but it doens't apply after rebooting.
2020-06-13 03:52:12 <winny> i have a machine with one bootable HDD. Can also boot CDROM. it cannot boot USB. I'd like to set up encrypted root on it. Previously I did this by setting up /boot on a removable thumb drive, and configure the cryptkey feature, which creates an initramfs with the key embedded. Given I can't use a removable USB key, is there an alternate approach to separate the encrypted volume from the plaintext
2020-06-13 03:52:14 <winny> keyfile? The idea is to make disposal of this machine easy.
2020-06-13 11:45:32 <Tuxist> PhoenixMage: can i add your patch in my mere request for ceph ?
2020-06-13 11:45:54 <Tuxist> https://gitlab.alpinelinux.org/Tuxist/aports
2020-06-13 11:49:16 <Tuxist> PhoenixMage:  so we can work togehter i think only init script must be wriiten and ceph deploy support
2020-06-13 11:50:02 <Tuxist> the patch for cephdepoly is very easy to make because they sysinitv support
2020-06-13 11:51:46 <Tuxist> i will test it nextweek i cannot at they monent because my switch is broken and the new one will arrive next week
2020-06-13 11:56:31 <ikke> did you run strace with -f?
2020-06-13 11:56:34 <ikke> oh, lol
2020-06-13 11:57:03 <ikke> oh, lol
2020-06-13 11:57:30 <ikke> I must stop doing this :D
2020-06-13 12:23:22 <ikke> I must stop doing this :D
2020-06-13 12:26:01 <markand> hi
2020-06-13 12:26:13 <ikke> markand: hey
2020-06-13 12:26:19 <markand> someone has an idea why firefox isn't listed in GNOME "Default Application" settings?
2020-06-13 12:27:12 <maxice8> I think they use Midori ?
2020-06-13 12:27:34 <markand> they use GNOME Web by default but even that, I mean in the combo box I'm unable to choose firefox
2020-06-13 12:28:09 <maxice8> Hmm
2020-06-13 12:28:13 <maxice8> i have 2 firefoxes
2020-06-13 12:28:15 <maxice8> missing icons
2020-06-13 12:28:26 <maxice8> my music and video are mpv (?? no gnome music ???)
2020-06-13 12:28:32 <maxice8> my calendar is nvim
2020-06-13 12:38:09 <gray_-_wolf> Hello :) Quick question regarding building alpine images. Based on my observations, the package being builded is able to access the signing key. Is there a way to prevent such rougu software from that? For example, in archlinux I can say not to sign and sign it separately later. Is there similar flag to do that in abuild?
2020-06-13 12:39:05 <gray_-_wolf> I was using abuild-gzsplit (part of abuild package) but that one is broken in 3.12 so I am not able to upgrade my repository from 3.11 for now. So I am looking into other ways how to achieve this. Thanks :)
2020-06-13 12:39:43 <ikke> gray_-_wolf: it's the abuild index stage that is doing the signing
2020-06-13 12:40:08 <ikke> which is one of the last stages
2020-06-13 12:40:18 <ikke> so you could manual provide all the steps, except the last one
2020-06-13 12:40:44 <ikke> abuild unpack deps prepare build check rootpkg
2020-06-13 12:40:49 <ikke> oh, and undeps
2020-06-13 12:40:55 <omenius> one question: is installing nvidia non-free drivers something you can do on alpine with low to moderate effort?
2020-06-13 12:41:10 <maxice8> Aren't they glibc only ?
2020-06-13 12:41:30 <ikke> afaik, they are
2020-06-13 12:41:51 <omenius> here are some nvidia driver requirements: is installing non-free nvidia drivers something you can do with alpine? I think some deps might be missing and hard to install.
2020-06-13 12:42:04 <mps> ikke: and 'abuild index'
2020-06-13 12:42:12 <ikke> mps: that's the step they wanted to skip
2020-06-13 12:42:14 <omenius> wrong pasteclass.malware.re/
2020-06-13 12:42:20 <mps> ah, ok
2020-06-13 12:42:47 <omenius> my clipboard is not working :D
2020-06-13 12:43:06 <mps> omenius: I tried hard two years ago, but without success
2020-06-13 12:43:56 <omenius> that's long time before you gave up
2020-06-13 12:44:06 <mps> yes
2020-06-13 12:44:07 <gray_-_wolf> ikke: Thank you, will give that a try. How often does "new" step appear? (== how likely is that this will break over time?)
2020-06-13 12:44:09 <omenius> no it's just my dyslexia
2020-06-13 12:44:28 <omenius> i cant write or read today. I was thinking you tried to install them for two years
2020-06-13 12:44:37 <mps> no
2020-06-13 12:45:01 <gray_-_wolf> omenius: the user-space drivers are glibc only as already stated, I did not found a way to get that working. I've ended returning the card and buying AMD one, those work fine
2020-06-13 12:45:10 <mps> maybe I worked on that about two weeks in free time
2020-06-13 12:46:03 <omenius> i see. To be honest it's not expected to work
2020-06-13 12:47:52 <omenius> us.download.nvidia.com/XFree86/Linux-x86_64/440.82/README/minimumrequirements.html
2020-06-13 12:47:59 <omenius> here is the requirements btw.
2020-06-13 12:53:06 <mps> omenius: right
2020-06-13 12:53:49 <mps> I didn't hear that anyone managed to make it work on musl based system
2020-06-13 13:01:13 <TBB> I imagine that being such low level stuff, no translation layer will do a good enough job
2020-06-13 13:01:40 <TBB> oh well, maybe one day nvidia will change their ways...
2020-06-13 13:02:48 <mps> TBB: you are Finn?
2020-06-13 13:02:52 <mps> iirc
2020-06-13 13:03:49 <mps> you remember famous words from another famous Finn, Torvalds: F*ck you Nvidia
2020-06-13 13:04:16 <TBB> :D
2020-06-13 13:04:24 <TBB> correct, I am
2020-06-13 13:04:28 <mps> and that was publicly on one Finish TV
2020-06-13 13:04:45 <TBB> (and that's not some Yoda'esque prose)
2020-06-13 13:05:15 <mps> yes, clear, concise, and direct
2020-06-13 13:05:33 <gray_-_wolf> you mean https://www.youtube.com/watch?v=_36yNWw_07g ?
2020-06-13 13:07:24 <mps> yes
2020-06-13 13:07:43 <mps> why I thought it was on TV
2020-06-13 13:09:57 <mps> (but university is more serious institution than TV anyway)
2020-06-13 13:11:03 <omenius> i like how linus is not too polite
2020-06-13 13:12:11 <mps> I didn't ever found he is impolite
2020-06-13 13:13:38 <omenius> there is this thing https://arstechnica.com/gadgets/2018/09/linus-torvalds-apologizes-for-years-of-being-a-jerk-takes-time-off-to-learn-empathy/
2020-06-13 13:13:43 <mps> politeness is in the ears of listener
2020-06-13 13:13:44 <omenius> maybe not the best source, just first link
2020-06-13 13:14:27 <omenius> being polite saves some hurt feelings but keeps people saying also how they really feel
2020-06-13 13:15:31 <mps> yes, I read this, but that is sure because he is under pressure of this new 'political correctness movements'
2020-06-13 13:15:48 <omenius> yeah, my thoughts also
2020-06-13 13:16:42 <mps> I'd rather to hear someone sincere thought than 'political correct' lies about me
2020-06-13 13:18:13 <mps> when we start to lie in name of 'political correctness' where we stop, lie becomes our second^Wfirst nature
2020-06-13 13:19:00 <tehcloud> if it was already correct it would not need to be politically correct
2020-06-13 13:19:14 <russkel> I think that whole thing was a beat up
2020-06-13 13:19:15 <tehcloud> real solution is to simply not be an asshole, generally
2020-06-13 13:19:20 <russkel> maybe he doesn't swear any more
2020-06-13 13:19:48 <mps> tehcloud: that is what I think
2020-06-13 13:20:19 <tehcloud> you can deliver a message that is critical without being an asshole about it
2020-06-13 13:20:20 <gray_-_wolf> hm, how long does it usually take for patches to be considered/commented/rejected? ---> How often should I ping about it? I want to push one through, but don't want to be too much of an annoyance
2020-06-13 13:20:24 <mps> but I wouldn't ban anyone for words s/he use
2020-06-13 13:20:39 <tehcloud> it requires patience on the part of everyone involved
2020-06-13 13:21:13 <tehcloud> if you are not capable of handling discourse without becoming emotional over things that don't matter, then you should work on that before involving yourself in any public discourse
2020-06-13 13:21:21 <gray_-_wolf> `s/he' excludes wide variety of people, correct term is `they' (singular they) :troll:
2020-06-13 13:21:41 <mps> well, I believe that people can overcome these issues
2020-06-13 13:22:18 <ikke> gray_-_wolf: Usually it doesn't take too long, but it could have snowed under a bit
2020-06-13 13:22:20 <markand> hey
2020-06-13 13:22:28 <mps> main problem is: who draw lines
2020-06-13 13:22:29 <markand> I have this small issue when trying to call apk fix: https://pastebin.com/iDS71Pb3
2020-06-13 13:22:40 <omenius> yeah instead of swearing he could speak in caps lock to communicate his strong feelings
2020-06-13 13:22:43 <maxice8> This should be in offtopic
2020-06-13 13:23:08 <mps> maxice8: ah, see, power in action :)
2020-06-13 13:23:19 <ikke> hah :P
2020-06-13 13:24:10 <maxice8> I made mv to offtopic suggestions before someone said I could be op though
2020-06-13 13:24:37 <ikke> maxice8: Yes, but now you have an @ your 'suggestions' will never be the same :P
2020-06-13 13:24:44 <gray_-_wolf> mps: usually the lines are dwarn by the vocal minority
2020-06-13 13:24:48 <russkel> who decides what is or what isn't offtopic?
2020-06-13 13:24:49 <gray_-_wolf> drawn*
2020-06-13 13:24:51 <russkel> 
ACTION think face

2020-06-13 13:25:01 <ikke> The one with the banhammer ;-)
2020-06-13 13:25:13 <mps> :D
2020-06-13 13:25:23 <russkel> dictatorship yikes
2020-06-13 13:25:29 <russkel> I'll have no part of this.
2020-06-13 13:25:50 <tehcloud> dictatorship, historically proven most effective form of rule
2020-06-13 13:26:18 <mps> heh, and I thought I was grown up in dictator regime country
2020-06-13 13:26:33 <tehcloud> of course, I speak of Benevolent Dictator for Life positions held in free software ;)
2020-06-13 13:27:00 <omenius> tbh offtopic rules should be enforced only if there is on-topic discussion going on
2020-06-13 13:27:09 <markand> nobody nose for my py3-cairo issue?
2020-06-13 13:27:25 <ikke> omenius: which is our stance as well, we're never too strict about being off-topic
2020-06-13 13:27:26 <mps> tehcloud: true, and because that I argued for alpine to have BDFL instead of 'teams'
2020-06-13 13:27:51 <tehcloud> if a team can preserve a vision then it's fine
2020-06-13 13:28:15 <tehcloud> but we have CEOs in business structures to form and push forward a vision for a reason, and often teams complicate that
2020-06-13 13:28:20 <ikke> That does not mean the bdfl can delegate tasks to teams
2020-06-13 13:28:50 <tehcloud> if you get a team structure that changes the vision, then you end up with forks
2020-06-13 13:29:06 <russkel> markand, if you remove the package and reinstall it, does it sort the issue?
2020-06-13 13:29:15 <tehcloud> vs simply having those would-be team members create their own thing in the first place
2020-06-13 13:29:21 <russkel> looks like the folder doesn't exist
2020-06-13 13:30:06 <russkel> directory*, damn windows terminology has infected me for life
2020-06-13 13:30:10 <ikke> heh
2020-06-13 13:30:56 <markand> russkel, one minute, apk del won't let me delete it
2020-06-13 13:31:25 <ikke> you would need to remove the parent package that causes this to be installed
2020-06-13 13:32:03 <markand> there are no option to ignore dependencies temporarily?
2020-06-13 13:32:22 <russkel> oh you can't force?
2020-06-13 13:32:27 <ikke> No, you cannot
2020-06-13 13:32:34 <markand> ah
2020-06-13 13:32:35 <ikke> apk wants to keep the dependency graph intact
2020-06-13 13:32:59 <russkel> what about force reinstall?
2020-06-13 13:33:03 <russkel> -f -f -f
2020-06-13 13:33:12 <ikke> That's what apk fix is supposed to do
2020-06-13 13:33:21 <markand> :P
2020-06-13 13:33:28 <markand> nvm it was only required for one package
2020-06-13 13:33:42 <markand> fixed it
2020-06-13 13:33:43 <ikke> markand: an strace output could be helpful
2020-06-13 13:34:00 <russkel> I jump from package manager to package manager, I apologise
2020-06-13 16:15:16 <Hello71> ikke: https://gitlab.alpinelinux.org/alpine/infra/infra/-/issues/10694
2020-06-13 16:15:37 <ikke> thanks
2020-06-13 16:41:56 <lunaphyte> Hello71: the other day you mentioned "presumably it would make more sense if there was an "lbu copy" command" - regarding setup-disk
2020-06-13 16:42:46 <lunaphyte> were you saying there is an lbu copy command, and it would have made more sense to use that, or were you saying if lbu had a copy command, something like that might make more sense?
2020-06-13 16:47:30 <Hello71> I assumed there isn't
2020-06-13 16:47:51 <lunaphyte> ok, cool
2020-06-13 16:48:02 <lunaphyte> i went looking for it, to see what it did, but couldn't find it :)
2020-06-13 16:48:14 <oneinsect> perl says I can't seem to find your C library.  I've looked in the following places:
2020-06-13 16:48:27 <oneinsect> any advise friends
2020-06-13 16:48:44 <lunaphyte> is one or the other of https://git.alpinelinux.org/ or https://gitlab.alpinelinux.org/ considered more authoritative?
2020-06-13 16:52:04 <maxice8> gitlab is the authoritative
2020-06-13 16:52:45 <lunaphyte> ok, thanks
2020-06-13 17:04:40 <lunaphyte> commands like "w", "who", "last", seem to not work.  files like /run/utmp, /var/log/wtmp are present, but empty.  is this expected?  it seems like musl may deliberately not implement utmp, and instead i should use utmps if i want it?
2020-06-13 17:06:15 <ikke> lunaphyte: musl does not provide it
2020-06-13 17:06:22 <nixi> i am trying to install alpine in diskless mode in my server and i can't grasp one thing... i have an internal usb slot in which i have a flash drive with a live alpine image. where should i point `alpine-setup` to save the configuration?
2020-06-13 17:06:34 <nixi> the same flash drive?
2020-06-13 17:06:39 <ikke> usually, yes
2020-06-13 17:06:47 <ikke> but you need to make sure it is writable
2020-06-13 17:07:01 <ikke> the iso itself is not writable (due to being isofs)
2020-06-13 17:07:04 <lunaphyte> ikke: ok, thanks
2020-06-13 17:07:07 <nixi> but there is only a small partition o nthe drive
2020-06-13 17:07:12 <nixi> created with dd
2020-06-13 17:07:49 <nixi> and waht about system updates?
2020-06-13 17:08:12 <ikke> New kernels requires an update to the usb disk
2020-06-13 17:08:26 <ikke> The rest is just regular apk upgrade
2020-06-13 17:08:32 <nixi> so how does that get written to it?
2020-06-13 17:08:34 <ikke> (and lbu to make things persistent)
2020-06-13 17:10:28 <nixi> better yet, could you please explain how i would setup a system like this? i reckon i need to create a live usb stick first, which i have
2020-06-13 17:10:30 <nixi> then what?
2020-06-13 17:10:34 <nixi> expand the partition?
2020-06-13 17:11:00 <ikke> Usually it's just a single writable exfat or ext* partition
2020-06-13 17:11:22 <ikke> setup-bootable can prepare your usb media for that
2020-06-13 17:11:24 <nixi> when i dd the iso to my usb flash drive?
2020-06-13 17:11:38 <ikke> no, with dd it's just read-only
2020-06-13 17:12:33 <nixi> so i boot from one usb stick (made with dd) and do setup-bootable to a second usb stick as a target
2020-06-13 17:12:46 <ikke> no, that does not really make sense :)
2020-06-13 17:12:48 <nixi> which creates a working system on the second usb flash drive to which i will later boot from in my server?
2020-06-13 17:12:53 <ikke> oh, that can work
2020-06-13 17:13:09 <nixi> ok i will try that
2020-06-13 17:16:51 <nixi> one more thing ... when i have stuff running on the diskless system, it does not keep cached files at all i reckon. eg. logs
2020-06-13 17:17:42 <nixi> or when i use syncthing and i change settings via the web interface
2020-06-13 17:18:55 <ikke> correct
2020-06-13 17:19:11 <ikke> You can mount a disk somewhere to have persistent data
2020-06-13 17:19:40 <nixi> hmmmm
2020-06-13 17:19:51 <nixi> i guess i will rather install the system directly on the usb stick then
2020-06-13 17:20:06 <nixi> is that a good idea?
2020-06-13 17:20:37 <nixi> i will have a raid logical drive mounted to the system
2020-06-13 17:21:34 <ikke> depends on your usecase
2020-06-13 17:23:26 <nixi> i had two disks in raid 1 for system before, but both died during recovery...
2020-06-13 17:23:31 <nixi> that is not alpine related though
2020-06-13 17:28:50 <lunaphyte> i've installed utmps, and started it, but commands like who still don't work
2020-06-13 17:29:22 <lunaphyte> does musl not supporting this mean that it doesn't matter if i install something like utmps?  commands like who just aren't going to work, period?
2020-06-13 17:49:25 <tomreid> Hi all, does anyone know anything about hosting the apkrepos on s3? I've given it a try, but unfortunately apk fails to fetch packages that contain special characters (libstdc++ as an example) since s3 expects libstdc%2B%2B
2020-06-13 17:49:44 <tomreid> is there any place I can dig in for url encoding in apk requests?
2020-06-13 17:50:03 <ikke> apk uses libfetch, might be related to that
2020-06-13 17:50:17 <ikke> https://gitlab.alpinelinux.org/alpine/apk-tools/-/tree/master/libfetch
2020-06-13 17:50:18 <tomreid> I'm just digging around the edges of that
2020-06-13 17:50:32 <ikke> Much more I don't know about it
2020-06-13 17:51:36 <tomreid> thanks @ikke
2020-06-13 18:38:57 <Hello71> lunaphyte: I would assume you need to recompile each program you want to write to utmp
2020-06-13 19:07:20 <Hello71> the program needs to know to contact the daemon in order to do the IO
2020-06-13 19:09:20 <AinNero> lunaphyte: alpine is not really supporting utmps
2020-06-13 19:10:15 <AinNero> like, its packaged, but many programs lack the code to drive it
2020-06-13 21:33:38 <lunaphyte> Hello71, AinNero: that makes sense, thanks
2020-06-14 05:38:56 <rnkn> hello! I'm having trouble allowing user-level mounting of an external exfat HDD. I'm sure I've followed all the instructions to the letter, with my /etc/fstab with:
2020-06-14 05:39:01 <rnkn> LABEL=BABEL	/media/BABEL		exfat	rw,user,noauto	0 0
2020-06-14 05:39:08 <rnkn> and my /etc/fuse.conf with:
2020-06-14 05:39:15 <rnkn> user_allow_other
2020-06-14 05:40:23 <rnkn> the HDD drive mounts fine with sudo/root, and with user it's correctly finding the labelled drive as /dev/sdb2, but still telling me I must be superuser to mount; so I'm wondering if this is a busybox limitation?
2020-06-14 05:41:42 <rnkn> p.s. my user account is also a member of group "disk"
2020-06-14 05:57:06 <ikke> you can test by adding regular mount
2020-06-14 05:57:31 <rnkn> ikke: good idea, brb
2020-06-14 05:57:35 <ikke> it's part of util-linux
2020-06-14 06:05:04 <rnkn> huh, okay real mount (not busybox) was already installed
2020-06-14 06:05:49 <rnkn> I'm not sure what changed since last time but now I get "fusermount: unknown option 'user=willaim' " and no successful mount
2020-06-14 06:07:16 <rnkn> should I be using fuse or fuse3 ?
2020-06-14 06:09:08 <rnkn> (answering my own q, fuse-exfat requires fuse..)
2020-06-14 06:16:26 <rnkn> ah how frustrating... it seems that fuse-exfat cannot mount devices as unprivileged users :C
2020-06-14 06:18:01 <ikke> Didn't linux get an exfat driver now?
2020-06-14 06:19:12 <ikke> 5.7 does
2020-06-14 06:20:38 <mps> ikke: 5.6 also
2020-06-14 06:21:10 <ikke> aha ok
2020-06-14 06:21:20 <rnkn> wow, okay I just purged a bunch of stuff for .makedepends-webkit2gtk (including util-linux) and now I can mount the exfat drive as user
2020-06-14 06:21:25 <mps> 5.6 got driver from microsoft, but it is replaced by driver from samsung in 5.7
2020-06-14 06:21:32 <ikke> ah, yeah, ok
2020-06-14 06:22:17 <mps> uhm, I need a coffee
2020-06-14 06:22:30 <mps> 5.4 got driver from microsoft
2020-06-14 06:23:07 <mps> but second part is right, in 5.7 it is replaced by samsung exfat drivers
2020-06-14 06:23:12 <rnkn> haha now it works perfectly
2020-06-14 06:23:26 <rnkn> using busybox mount
2020-06-14 06:23:31 <mps> f4831272936cfff8f9b1287c201a6e57d612b757
2020-06-14 06:23:47 <mps> algitbot: ^
2020-06-14 06:24:14 <mps> looks like 'he' also need coffee :)
2020-06-14 06:24:28 <ikke> rnkn: hah, that's a first :P
2020-06-14 06:30:53 <rnkn> only thing is that exfat doesn't do permissions, so everything is still owned by root, and restic just can't deal
2020-06-14 06:37:49 <rnkn> can confirm, reinstalling util-linux (because I needed column) makes mount fail with exfat-fuse user option
2020-06-14 06:52:28 <ikke> usually you can specify uid/gid
2020-06-14 06:53:21 <rnkn> tried uid and gid but no luck. I'll install thunar-volman and see
2020-06-14 09:09:16 <LACampbell> is there any reason to use awall over ufw? other than the python dependency
2020-06-14 09:09:33 <LACampbell> not sure if i should invest the time in learning awall. ufw is in testing, which makes me uneasy
2020-06-14 09:10:06 <LACampbell> this is for a server, not a router
2020-06-14 09:21:15 <LACampbell> or just learn iptables, I guess. I think I'll do that
2020-06-14 10:12:14 <clandmeter> LACampbell: awall is not that bad
2020-06-14 10:12:29 <clandmeter> you can  use the predefined profiles
2020-06-14 10:12:32 <clandmeter> and modify a bit
2020-06-14 10:12:50 <clandmeter> policies i mean
2020-06-14 10:13:43 <clandmeter> but awall still needs basic understanding of iptables and firewalls in general.
2020-06-14 15:25:42 <fujita_[m]> Hi there, I just installed alpine and when I do an apk update I got errors. Is there something wrong with my set up
2020-06-14 15:26:08 <afontain_> that's vague
2020-06-14 15:26:10 <afontain_> which error?
2020-06-14 15:26:41 <fujita_[m]> 
ACTION uploaded an image: VectorImage_2020-06-14_112633.jpg (368KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/CHfXjusFZsgvhPapexNVGCSG >

2020-06-14 15:27:09 <afontain_> are you connected to the internet?
2020-06-14 15:27:28 <afontain_> ping dl-cdn.alpinelinux.org
2020-06-14 15:27:40 <fujita_[m]> I did when booting from usb
2020-06-14 15:27:45 <fujita_[m]> Must I connect again?
2020-06-14 15:28:46 <fujita_[m]> If so what's the command to get internet set up again?
2020-06-14 15:29:24 <afontain_> I use `sudo nmtui`
2020-06-14 15:29:34 <afontain_> but that's for wifi network
2020-06-14 15:30:47 <afontain_> wired is (if done manually) `ip link set eth0 up; udhcpc -i eth0`
2020-06-14 15:30:47 <fujita_[m]> Sudo not found
2020-06-14 15:31:07 <afontain_> you are aready root
2020-06-14 15:31:15 <fujita_[m]> I did it via wifi network
2020-06-14 15:31:19 <fujita_[m]> Yup
2020-06-14 15:32:28 <fujita_[m]> I have reset my PC to avoid any confusion
2020-06-14 15:33:23 <fujita_[m]> 
ACTION uploaded an image: VectorImage_2020-06-14_113317.jpg (274KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/qFOhdgrKFYETXxHyJyxqlNCc >

2020-06-14 15:40:08 <fujita_[m]> Do I type root and just follow the connecting to wireless access point guide?
2020-06-14 15:40:08 <ikke> temporary error usually indicates dns errors
2020-06-14 15:41:10 <afontain_> you can, yes
2020-06-14 15:41:20 <pickfire> Hi, I recently just updated alpine about a week ago on my raspberry pi but now I cannot even resolve any dns.
2020-06-14 15:41:43 <pickfire> /etc/resolv.conf have been set to nameserver 1.1.1.1 1.0.0.1 but still it failed to resolve.
2020-06-14 15:41:56 <pickfire> Does anyone know how to solve the issue? I never seen this issue before.
2020-06-14 15:42:21 <ikke> can you ping those servers?
2020-06-14 15:42:41 <ikke> note that each nameserver option takes only one nameserver
2020-06-14 15:42:55 <pickfire> No
2020-06-14 15:43:04 <ikke> Then this is not a DNS issue
2020-06-14 15:43:37 <pickfire> Oh
2020-06-14 15:44:22 <ikke> You first need to make sure your network is working
2020-06-14 15:44:29 <pickfire> I can connect in.
2020-06-14 15:44:32 <pickfire> By SSH.
2020-06-14 15:44:48 <ikke> Can you ping anything else in on the host>
2020-06-14 15:44:50 <ikke> ?
2020-06-14 15:45:08 <pickfire> I can ping my own computer.
2020-06-14 15:46:06 <fujita_[m]> Might the problem be due to me setting ssh as none?
2020-06-14 15:46:31 <pickfire> Wait, I cannot even ping my own computer.
2020-06-14 15:46:49 <pickfire> No, I can ping my own computer.
2020-06-14 15:47:02 <pickfire> Setting ssh as none?
2020-06-14 15:47:20 <pickfire> I also disabled firewall but still cannot go anywhere.
2020-06-14 15:47:25 <afontain_> fujita_: ping dl-cdn.alpinelinux.org
2020-06-14 15:47:36 <ikke> pickfire: what does `ip route` return?
2020-06-14 15:47:46 <pickfire> default via 192.168.1.1 dev eth0
2020-06-14 15:47:47 <pickfire> 192.168.1.0/24 dev eth0 proto kernel scope link src 192.168.1.122
2020-06-14 15:48:02 <ikke> ok, looks normal
2020-06-14 15:48:11 <pickfire> Different from my own computer though.
2020-06-14 15:48:16 <pickfire> Let me check with my dad.
2020-06-14 15:48:23 <ikke> oh
2020-06-14 15:48:27 <ikke> yeah, I see one issue
2020-06-14 15:48:33 <pickfire> Most likely routing fault, I didn't get to though those so I don't know much about it.
2020-06-14 15:48:41 <pickfire> Thanks for helping out. :D
2020-06-14 15:48:53 <ikke> your default route is outside of your local subnet
2020-06-14 15:49:07 <ikke> your default gw is 192.168.*1*.1
2020-06-14 15:49:22 <ikke> ok, ignore me, it's correct
2020-06-14 15:53:13 <pickfire> Yeah, my default gateway on laptop is 192.168.1.2
2020-06-14 15:53:20 <ikke> hmm
2020-06-14 15:53:27 <ikke> ok, then you need to fix that
2020-06-14 15:54:23 <pickfire> Looks like the default gateway was changed, I didn't know that.
2020-06-14 15:54:58 <pickfire> By the way, I just realized weechat improved a lot.
2020-06-14 15:56:47 <pickfire> Also weird that I never once got `sudo reboot` working on pi, I need to manually plug off the power on pi and put it back on for alpine linux.
2020-06-14 16:07:02 <lunaphyte> with a typical installation, what mounts /proc/?  i see /etc/init.d/procfs, but that doesn't appear to be in any run levels?
2020-06-14 16:07:27 <lunaphyte> i'd like to adjust the options used for mounting it
2020-06-14 16:07:45 <maxice8> most likely mounted with the initramfs
2020-06-14 16:16:26 <lunaphyte> i did see that, and rebuilt a new initramfs with the options i wanted, but it doesn't seem to have worked
2020-06-14 16:17:08 <lunaphyte> i've unpacked the new initramfs, and i do see that it has my modifications, so i'm confident that part worked
2020-06-14 16:17:32 <lunaphyte> but after booting, the /proc/ mount is no different than before
2020-06-14 16:29:54 <fujita_[m]> Hi there, is there anyway to set up alpine wirelessly?
2020-06-14 16:31:45 <maxice8> Wirelessly as in, fetch packages during installation via WiFi ?
2020-06-14 16:33:36 <fujita_[m]> Yes
2020-06-14 16:34:29 <fujita_[m]> I managed to install connect to my network when booting for my usb
2020-06-14 16:34:37 <fujita_[m]> Connect*
2020-06-14 16:35:22 <fujita_[m]> Now that alpine is installed, how do I scan for my network again since it didn't prompt for anything
2020-06-14 17:04:20 <lunaphyte> when it put the proc mount in fstab, it works
2020-06-14 17:04:47 <lunaphyte> that makes me curious though.  proc is mounted regardless of if it's in fstab, so why do i have to put it there to get what i want?
2020-06-14 17:05:41 <lunaphyte> br
2020-06-14 17:05:44 <lunaphyte> oops
2020-06-14 17:07:02 <maxice8> most likely it remounts with your options
2020-06-14 17:07:58 <lunaphyte> what is remounting it?
2020-06-14 17:08:52 <lunaphyte> init in initramfs?
2020-06-14 17:14:20 <maxice8> localmount service in OpenRC i guess
2020-06-14 17:17:14 <lunaphyte> if fstab doesn't contain proc though, it wouldn't be remounting it though, would it?
2020-06-14 17:29:59 <omenius> trying to wget the installer and getting "certificate's owner does not match hostname"
2020-06-14 17:31:14 <omenius> from http://dl-cdn.alpinelinux.org
2020-06-14 17:31:27 <omenius> as I paste it I notice it's http
2020-06-14 17:35:42 <lunaphyte> you're using an http:// url and wget is complaining about a certificate?
2020-06-14 18:12:56 <fujita_[m]> Hi there is anyone able to help me
2020-06-14 18:13:17 <fujita_[m]> On how to get a wireless connection
2020-06-14 18:40:57 <peter44> Hello can I ask you is it possible to run zoom-client on alpine? I have downloaded some non specific package from their website and I dont know how to run it. thank you
2020-06-14 18:43:57 <u0jQx9gPyrYg> if it is a binary then you probably need a glibc chroot to do so
2020-06-14 18:44:11 <u0jQx9gPyrYg> if it is a java thing, then it should work.
2020-06-14 18:44:47 <u0jQx9gPyrYg> i doubt that gcompat will be enough to have it running
2020-06-14 18:45:12 <peter44> it is a binary but if i try to run it nothing happens
2020-06-14 18:45:28 <afontain_> fujita_: just use nmtui
2020-06-14 18:46:35 <u0jQx9gPyrYg> yeah, then it is probably some glibc thingy. you want a glibc chroot to run this. or i heard soem people using flatpacks, if there's such a version available use that...
2020-06-14 19:04:37 <Hello71> also zoom is a malware
2020-06-14 19:04:40 <Hello71> use the web client if you have to
2020-06-14 19:05:09 <nacelle> run it in its own world
2020-06-14 19:14:09 <Cogitri> Unfortunately the webclient is rather limited
2020-06-14 19:14:15 <Cogitri> I use the Flatpak
2020-06-14 21:28:57 <LACampbell> clandmeter: turns out what I want to do is not that complicated in iptables, so I'm just going to figure that out for now
2020-06-14 22:12:13 <omeniux> this alpine install was really nice experience
2020-06-14 22:12:50 <omeniux> only thing i missed was guided lvm encryption
2020-06-14 22:30:26 <LACampbell> alpine is the only linux system small enought that I feel like I can understand a decent chunk. which is weird as people consider it to be for unix experts
2020-06-14 22:56:02 <tehcloud> you will become one by using it
2020-06-14 22:56:19 <tehcloud> for some people it is easier to understand something basic that is "for experts"
2020-06-14 22:56:35 <tehcloud> vs the supposedly easy, full-featured, user-friendly, bug-ridden type stuff
2020-06-15 00:07:46 <LACampbell> yeah you see this dichotomy in a lot of areas of computing. CSS vs SASS... I'd also argue CSS is easier
2020-06-15 06:39:23 <yunfan> mps: i got the c-sky board now.
2020-06-15 06:39:38 <yunfan> its the original version which is not a riscv core
2020-06-15 06:50:17 <yunfan> also how to build alpine from source?
2020-06-15 06:50:21 <yunfan> any wiki pages?
2020-06-15 06:59:26 <TBB> clone aports and look at the scripts/ directory
2020-06-15 07:01:24 <TBB> you'll find a bootstrap script there
2020-06-15 07:37:29 <mps> yunfan: maybe you can create forth OS on c-sky instead of linux/alpine
2020-06-15 07:38:18 <oneinsect> alpine can be created for c-sky
2020-06-15 07:38:22 <oneinsect> not hard
2020-06-15 07:38:47 <oneinsect> I ported alpine to many boards including ones with allwinner etc
2020-06-15 07:39:29 <mps> depends. how much RAM board have
2020-06-15 07:39:39 <oneinsect> 64MB?
2020-06-15 07:43:30 <yunfan> mps: that would lack of too many software
2020-06-15 07:43:46 <yunfan> oneinsect: any wiki pages on how to do that?
2020-06-15 07:43:59 <yunfan> oneinsect: yes, buitin 64 MB
2020-06-15 07:45:04 <mps> yunfan: how much software (in the wild) can be used in 64MB (rhetorical question)
2020-06-15 07:45:37 <yunfan> mps: many?
2020-06-15 07:45:53 <ikke> Certainly not any browser :P
2020-06-15 07:45:56 <yunfan> i think it could be used as a personla agent
2020-06-15 07:46:08 <yunfan> like web camera agent :D
2020-06-15 07:46:15 <yunfan> or hosting your irc bot
2020-06-15 07:47:00 <Cogitri> Hm, even my raspberry pi that only runs dnsmasq and uhttpd needs 74MB RAM 🤔
2020-06-15 07:47:39 <yunfan> Cogitri: you mean using raspbian?
2020-06-15 07:47:55 <Cogitri> I think yes, I don't run Alpine on that thing
2020-06-15 07:48:38 <Cogitri> But that's just because setting up PiHole is so convenient on Raspbian :)
2020-06-15 07:48:54 <yunfan> have you tried dietpi? i am quite sure it wont need that much ram at the beginning
2020-06-15 07:50:51 <mps> there were times when I run linux (slackware) in 4MB RAM but now ...
2020-06-15 07:52:08 <yunfan> well i remember back to 2009 when i bought my first PC at college, it has huge ram of 512M, i had tried much tiny linux  :D
2020-06-15 07:52:43 <yunfan> many of them were install in my usb disk in usb floppy mode :D
2020-06-15 07:52:53 <yunfan> also there were web browser w3m
2020-06-15 07:54:40 <mps> my first PC had 512KB, though in these times linux didn't exists
2020-06-15 07:55:33 <mps> KB, no MB :)
2020-06-15 07:56:05 <mps> though my first computer was with 32 or 64 KB
2020-06-15 07:56:08 <yunfan> cool
2020-06-15 07:56:26 <yunfan> i guess you use tape at the time
2020-06-15 07:57:01 <mps> actually, yes
2020-06-15 07:57:10 <mps> even tapes
2020-06-15 07:58:27 <mps> though also had some kind of cartridges and floppy disk drives
2020-06-15 08:00:26 <mps> even once I made program for control simple machine only using hex editor, writing instructions as hexes
2020-06-15 08:00:39 <mps> that must be long time ago :)
2020-06-15 08:02:08 <mps> experience nearly like programming with physical switches, as computer pioneers programmed first computers
2020-06-15 08:03:04 <ikke> pkgs.a.o will be unavailable for a minute to fix a database issue. Should be back soon.
2020-06-15 08:05:00 <ikke> Ok, it's up again
2020-06-15 08:06:28 <winny> ran into a funny gotcha with respect to alpine. Turns out using UUIDs to reference logical volumes can cause to gotchas when lvm snapshots are used, because the snapshot will share the same UUID. just had to fix my root= to use the lvm /dev/VG/LV because using the UUID would mount my pre upgrade snapshot on root
2020-06-15 08:07:33 <ikke> Is that alpine specific?
2020-06-15 08:07:46 <winny> only in that setup-disk assumes you always want UUID in the root parameter
2020-06-15 08:09:03 <winny> so not particularly, but it did give me pause as i wondered why none of my upgrades appeared to be installed, and no devices detected because old kernel modules were available, but was using the new kernel =)
2020-06-15 08:09:14 <yunfan> mps: is that the time you fall in love with forth?
2020-06-15 08:10:09 <ikke> using UUIDs in fstab is considered best-practice
2020-06-15 08:10:38 <winny> yeah this feels like a weird edge case
2020-06-15 08:10:46 <ikke> indeed
2020-06-15 08:10:58 <winny> blkid reporting the same UUID and PARTUUID for both LVs
2020-06-15 08:11:46 <mps> yunfan: can't remember exactly, but yes, first forth I started to learn on those 8-bit computers
2020-06-15 08:11:49 <winny> for what it's worth LVM refuses to acknowledge multiple VGs with the same name
2020-06-15 08:13:41 <yunfan> so i had cloned aports, looks like i should run this on alpine?
2020-06-15 08:15:34 <ikke> bootstrapping outside of alpine is more challenging and depends on your situation how to it
2020-06-15 08:16:08 <yunfan> ok, will find a compiling machine for this
2020-06-15 08:16:12 <ikke> You'd first need to create a chroot with gcc/musl (possibly cross-compiled)
2020-06-15 08:16:29 <yunfan> my old pc had failed for months
2020-06-15 08:19:59 <crdb2> Hello
2020-06-15 08:21:07 <crdb2> My Intel wifi card firmware is missing from the extended image. As /.modloop is read only how can I provide the correct firmware to iwlwifi driver?
2020-06-15 09:09:52 <TBB> interesting, one would think iwlwifi is so standard it's in there
2020-06-15 09:21:01 <crdb2> TBB my specific firmware is in linux-firmware-other but not on the installation ISO
2020-06-15 09:21:10 <crdb2> and can't install it from the ISO
2020-06-15 09:21:24 <crdb2> I'm stuck having to make a custom ISO ...
2020-06-15 09:25:07 <Marquel> crdb2: no way to slap a wired interface on the system?
2020-06-15 09:27:14 <crdb2> Marquel well... not if I want to validate my wifi interface priori to installation
2020-06-15 09:28:15 <crdb2> prior*
2020-06-15 09:28:47 <Marquel> crdb2: why not? slap a wired in and see if you can get your wifi working in live system?
2020-06-15 09:28:47 <Marquel> which card is that anyway?
2020-06-15 09:30:10 <ikke> crdb2: I think you should be able to remount the modloop writable
2020-06-15 09:30:30 <crdb2> well if you know how I am all ears
2020-06-15 09:30:40 <crdb2> @marquel
2020-06-15 09:31:02 <crdb2> as I said the problem is that I can't install the firmware package in the live system
2020-06-15 09:31:15 <ikke> try: mount -o remount,jw /.modloop
2020-06-15 09:31:19 <ikke> try: mount -o remount,rw /.modloop
2020-06-15 09:32:22 <crdb2> the squashfs can be used RW O_o
2020-06-15 09:33:12 <ikke> perhaps not :)
2020-06-15 09:34:38 <Marquel> might not be permanent on your live-image but maybe able to trick the system into being able to install the fw package.
2020-06-15 09:34:51 <Marquel> however - which exact card are we talking about?
2020-06-15 09:35:28 <crdb2> an old 6300 somthing
2020-06-15 09:35:42 <crdb2> needs iwlwifi-6000-4 > 6 .ucode
2020-06-15 09:36:21 <Marquel> if it helps you: it will work. definitely.
2020-06-15 09:36:53 <Marquel> i've tried 3xxx, 6xxx, 8xxx and 9xxx - all operational under linux w/o any issues.
2020-06-15 09:39:16 <crdb2> OK can't remount rw on modloop
2020-06-15 09:39:30 <crdb2> @marquel oh I know it works been using it for ages
2020-06-15 09:39:39 <crdb2> just not from the install CD
2020-06-15 09:39:42 <crdb2> :(
2020-06-15 10:21:34 <crdb2> is there a doc somewhere that specify how to had something to modloop ?
2020-06-15 10:22:05 <crdb2> seems adding packages in the mkimage profile isn't sufficient it only adds the package in ISO repo
2020-06-15 10:34:46 <ikke> crdb2: there seems to be a variable called $modloop_addons
2020-06-15 10:35:03 <ikke> https://git.alpinelinux.org/aports/tree/scripts/mkimg.base.sh#n20
2020-06-15 10:41:48 <crdb2> @ikke yeh can't to find how to make out something with that
2020-06-15 10:44:27 <ikke> crdb2: I think you should add it to https://git.alpinelinux.org/aports/tree/scripts/mkimg.standard.sh#n25
2020-06-15 10:45:58 <crdb2> I mean what's the format of the value inside the variable
2020-06-15 10:51:00 <ikke> The variable it appears in is called _pkgs, and the existing values all seem to point to packages :)
2020-06-15 10:51:24 <ikke> I don't know any more then you do, I just try to read the code
2020-06-15 10:58:47 <crdb2> well it's pointless anyway cause it seems my card is incompatible with my new AP
2020-06-15 10:58:49 <crdb2> >_<
2020-06-15 10:59:08 <ikke> d'oh
2020-06-15 10:59:49 <artok> give a man a fish and you feed him for a day, teach a man to fish and he wants you to do the fishing for him -internet proverb
2020-06-15 10:59:53 <artok> oh gosh
2020-06-15 11:00:18 <artok> AP doesn't have some settings for supported standards?
2020-06-15 11:01:47 <crdb2> it seems to support the basic standards ... and other devices seem to be happy with it ... Let's try latest intel's firmware
2020-06-15 11:11:28 <crdb2> ok it's the last version, previous version crash with microcode SW error
2020-06-15 11:28:15 <crdb2> at least this one cleanly deauth :D
2020-06-15 13:22:47 <wsinatra> Anyone here done a deep div into Selinux on Alpine?
2020-06-15 13:49:04 <Hello71> how deep do you need to div
2020-06-15 13:49:11 <Hello71> kernel part, user part, done
2020-06-15 13:49:29 <TBB> deep dives into unknown waters tends to be life ending experiences
2020-06-15 13:50:57 <wsinatra> No joke, selinux in and of itself is daunting, easy to get lost.
2020-06-15 13:51:42 <wsinatra> I'm curious about kernel & user implementation. If anyone has any experience implementing it.
2020-06-15 13:52:43 <wsinatra> The modus being that CentOS is being suggested for having a solid selinux baseline, and I really am not convinced that the same thing can't just be done on Alpine, and that it shouldn't be a justification to use that specific distro
2020-06-15 13:56:07 <TBB> there was at some time discussion about what route Alpine should take; at that point I think apparmor seemed to gather a lot of support, but that was a good while ago
2020-06-15 13:59:37 <wsinatra> Hmm I've never looked that far into apparmor, I've always been under the impression that selinux provided better coverage
2020-06-15 13:59:47 <wsinatra> maybe I should look into it
2020-06-15 14:00:16 <ikke> It would add a huge maintenance burden though
2020-06-15 14:00:23 <ikke> making sure all the right profiles are included
2020-06-15 14:00:50 <maxice8> Void Linux went with apparmor, it seemed painless at the time
2020-06-15 14:00:54 <detha> selinux is a pretty huge beast. And for it to be effective almost everything needs a policy
2020-06-15 14:01:33 <detha> I would love to see selinux in alpine, but it's a lot of maintenance
2020-06-15 14:01:49 <wsinatra> ikke: absolutely, just trying to see if I could implement it on the servers I'm in charge of seems like a huge undertaking. It would require some dedicated people to ensure coverage across the entire distro
2020-06-15 14:02:57 <wsinatra> I'm being stubborn on the matter. I have control of the stack I'm building, and I want to use Alpine for that. But I also want to implement some level of security, to make a solid argument of "Alpine can do what CentOS does" to management
2020-06-15 14:03:03 <detha> enable selinux in kernel, start with a completely open policy, tighten as required
2020-06-15 14:03:31 <wsinatra> detha: that's a solid suggestion
2020-06-15 14:03:33 <ikke> yeah, you usually start in auditing mode to find what is necessary
2020-06-15 14:04:03 <detha> and lots of audit2allow
2020-06-15 14:08:47 <wsinatra> is audit2allow even packaged for Alpine? I'm not seeing it
2020-06-15 14:10:19 <ikke> Most likely not. There are a few tools packages, but not everything
2020-06-15 14:11:56 <wsinatra> I didn't think so, I can probably fix that though. Kind of feels like a prerequisite before I dive too far into this
2020-06-15 14:46:40 <mps> wsinatra: centos already can't do what alpine can
2020-06-15 14:48:42 <wsinatra> mps: care to elaborate on that? I'm personally biased towards Alpine, I've moved from Arch to it, and I'm trying to get away from RHEL/CentOS, but I would love any ammunition I could use when trying to defend that bias to my employer
2020-06-15 14:49:07 <mps> when the first debian released with selinux support I backported it to the previous debian release. forgot versions, that was 15-20 years ago
2020-06-15 14:50:04 <mps> probably details can be found on some debian ML archive
2020-06-15 14:50:40 <mps> after that I saw how it was waste of time and effort
2020-06-15 14:52:00 <mps> nowadays it would be probably easier to make it for distro, but again I think it will be waste of time and effort
2020-06-15 14:52:31 <wsinatra> Yeah I'm not certain that it would be something that I could implement properly by myself, without investing an absolutely massive amount of time towards doing so
2020-06-15 14:52:54 <mps> but anyone who want can do that, ofc
2020-06-15 14:53:09 <wsinatra> Then even if I could, or did, manage that the complexity of Selinux requires a lot of forethought, and outside eyes to make sure things are done right, otherwise it's just a tangle mess that doesn't actually help
2020-06-15 14:53:30 <mps> right
2020-06-15 14:53:43 <wsinatra> Agreed, if there are people interested in Selinux for Alpine, that's the beauty of FOSS
2020-06-15 14:54:09 <mps> some other of these LSM could be easier and less error prone
2020-06-15 14:54:14 <wsinatra> I'm more concerned with making an argument "this is secure" than adhering Selinux over Alpine really
2020-06-15 14:55:21 <mps> I would print one paper with words "Alpine is Secure" and glue it to some box running alpine :)
2020-06-15 14:56:02 <mps> nothing is secure, there is just more or less *wrong* perceptions that something is secure
2020-06-15 14:56:34 <wsinatra> hahaha I'll have to include it in the motd :)
2020-06-15 14:56:36 <mps> and false perception of security is worse than no security
2020-06-15 14:57:29 <mps> looking in real life you can notice that complicated 'things' are usually less secure and stable than simple 'things'
2020-06-15 14:57:41 <mps> same is in computer security
2020-06-15 14:57:57 <mps> keep in mind who invented selinux
2020-06-15 14:58:28 <wsinatra> That's a fair argument
2020-06-15 14:58:49 <wsinatra> very much runs against the Unix mentality of "do one thing, do it well"
2020-06-15 14:59:11 <SpaceDandy> I'm setting up Alpine on a Raspberry Pi 4, following the instructions on the Alpine wiki.  Everything is working fine, except lbu commit seems to be ignoring changes in /etc/init.d/.  Is this a known issue?
2020-06-15 14:59:32 <wsinatra> did you lbu include /etc/init.d?
2020-06-15 15:00:22 <SpaceDandy> No, there were no instructions to that effect in the wiki.  I can try that.
2020-06-15 15:00:52 <wsinatra> I'm not positive that it's part of the base include, but if you lbu include it it should pick it up for you
2020-06-15 15:00:59 <mps> I implemented yet another 'security framework' on linux (similar to grsec and with RBAC) and used grsec, but nowadays I'm more inclined to monitoring as security measure
2020-06-15 15:02:15 <wsinatra> I think the benefit of selinux/apparmor is the proactive prevention model. I'll have Nagios watching every server under my purview, but if there's some small way to help ensure that something doesn't spread quickly it gives me more time to react
2020-06-15 15:05:15 <mps> suricate/snort is better for this than nagios
2020-06-15 15:06:36 <wsinatra> I've never heard of suricata, though we've got snort through our Merakis and that's excellent stuff
2020-06-15 15:07:15 <mps> apk info 'one if them' :
2020-06-15 15:07:27 <mps> :)
2020-06-15 15:07:57 <wsinatra> I'm trying to figure out how best to provision AWS infra for multiple small businesses. Obviously taking the isolated VPC + bastion approach, but trying to find a balance between managing it all at scale between two people, and implementing something secure and sane
2020-06-15 15:09:18 <SpaceDandy> lbu include /etc/init.d didn't work.  I can see it added a line to /etc/apk/protected_paths.d/lbu.list, but lbu commit -n still doesn't show any changes to /etc/init.d
2020-06-15 15:09:24 <wsinatra> +10 on the suricata suggestion mps, this looks excellent
2020-06-15 15:11:50 <wsinatra> SpaceDandy: Are you trying to preserve the entirety of the init.d directory? You might need to add context per configuration
2020-06-15 15:12:31 <wsinatra> I think the basis is explicit deny unless specified? I'm not entirely positive as I've only touched on lbu to build a couple of SBCL arm build systems
2020-06-15 15:13:30 <mps> btw, I think we should remove snort from alpine or move to unmaintained
2020-06-15 15:13:54 <wsinatra> Is it not getting updated?
2020-06-15 15:14:18 <mps> it is not well maintained in alpine
2020-06-15 15:15:17 <mps> and it is 'proprietary', went from free to non free license, because that suricata was born
2020-06-15 15:15:46 <wsinatra> I think the non-free argument is pretty convincing
2020-06-15 15:15:48 <mps> suricata is born as free snort version, when snort went non free
2020-06-15 15:15:54 <SpaceDandy> wsinatra: Yes, I thought the point was that lbu should preserve any changes.  Ideally I'd like the overlay to contain any changes made.  Where is the master list of preserved paths?
2020-06-15 15:16:28 <ikke> lbu persists committed changes
2020-06-15 15:16:39 <wsinatra> lbu creates an overly tarball, so if the changes aren't part of that they'll be lost
2020-06-15 15:16:56 <ikke> somewhere in /etc/lbu
2020-06-15 15:16:59 <ikke> forgot the exact path
2020-06-15 15:17:02 <wsinatra> you need to lbu commit after any changes you made, otherwise it will return the system to the last saved state
2020-06-15 15:17:43 <mps> actually, snort is still free, but free version is somewhat limited
2020-06-15 15:18:14 <wsinatra> So lbu is a great system for doing something like installing a base user, saving something that doesn't change often (abuild keys for example) and then allowing you to toss out any changes to the system made between reboots
2020-06-15 15:18:30 <mps> company behind it have commercial offering, and you know results of such crippled things
2020-06-15 15:18:57 <wsinatra> mps: the enterprise edition gets rolled into Cisco offerings a lot, they make a ton of money that way.
2020-06-15 15:19:03 <wsinatra> The writing on the wall is VERY clear there
2020-06-15 15:19:12 <mps> yes
2020-06-15 15:19:30 <mps> and I dislike to call such 'things' free
2020-06-15 15:19:39 <wsinatra> suricata has lua configuration too
2020-06-15 15:20:02 <wsinatra> which means I can use Fennel to extend it, and anywhere I can sneak more lisp things is a boon
2020-06-15 15:20:17 <mps> heh, nice for you :)
2020-06-15 15:20:42 <wsinatra> It's always come back around to lisp somehow :D
2020-06-15 15:20:56 <mps> maybe you will be 'Talos of suricata' one day :)
2020-06-15 15:22:29 <wsinatra> Hahah now that would be something. I bet Grecians would program their automatons in lisp too ;)
2020-06-15 15:23:20 <mps> what/who is/are 'Grecians'
2020-06-15 15:24:12 <wsinatra> Grecians as in the Greeks, Talos was a greek automaton right? I could be misremembering my Greek mythology
2020-06-15 15:24:29 <mps> ah
2020-06-15 15:24:54 <mps> yes. but they are Greeks not Grecians
2020-06-15 15:25:35 <wsinatra> Huh my dad always referred to them as Grecians, didn't realize it wasn't correct
2020-06-15 15:25:39 <SpaceDandy> wsinatra,ikke: Yes, I understand how lbu is supposed to work, and that is how I'm trying to use it.  The issue is that lbu commit isn't picking up all changes.  Specifically it is ignoring changes to /etc/init.d, despite my explicit inclusion of that path as suggested above.
2020-06-15 15:27:27 <mps> wsinatra: now you can give back something to father in 'teaching fields'
2020-06-15 15:27:47 <wsinatra> SpaceDandy: Hmm, it would be annoying to have to include everything in /etc/init.d but lbu commit /etc/init.d/whatever-init would probably force the persistence to happen
2020-06-15 15:28:05 <wsinatra> but you're right, it should just pick it up with /etc/init.d if you lbu commit after you make your change
2020-06-15 15:28:39 <wsinatra> mps: I'll have to! I'm curious where he stumbled across the term now!
2020-06-15 15:29:19 <mps> uh, this is story about 2500 years old :)
2020-06-15 15:29:20 <SpaceDandy> wsinatra: No, it doesn't even work if I include the specific file.  In this case it is /etc/init.d/dropbear.  I've included that directly and it still ignores it.
2020-06-15 15:30:10 <mps> in some languages something like 'Grecians' can be used, but in most short form is used
2020-06-15 15:32:21 <mps> grecian term 'walked' from some near east languages, iirc
2020-06-15 15:33:33 <mps> it is more something like adjective form (though I'm not sure if same in english apply)
2020-06-15 15:34:23 <wsinatra> If I had to guess honestly, it's some small influence from my Scillian heritage. My Father is a second generation immigrant, and his parents didn't speak much English.
2020-06-15 15:35:22 <mps> Scillian or Sicillian
2020-06-15 15:35:47 <wsinatra> SpaceDandy: You could manually remove the tarball and recreate it, but running lbu diff might give you more info as to whether it's seeing the change and writing it, or just not seeing anythinga t all
2020-06-15 15:35:58 <oneinsect> https://pi3g.com/2019/01/10/alpine-boot-process-on-the-raspberry-pi/
2020-06-15 15:36:25 <oneinsect> helps understand init process as well including overlays
2020-06-15 15:36:33 <oneinsect> for lbu commits
2020-06-15 15:36:37 <wsinatra> Sicilian* dyslexia always kills my spelling
2020-06-15 15:37:02 <oneinsect> lbu commits will work if you use setup-alpine on rpi
2020-06-15 15:37:31 <mps> heh, then that explain why your father use this term
2020-06-15 15:38:11 <mps> for him this is right term
2020-06-15 15:40:44 <mps> Sicily was long time under occupation from some Arab speaking near east people
2020-06-15 15:41:26 <wsinatra> I didn't think my selinux frustration would end up in a language/history lesson :) part of why I love this community
2020-06-15 15:41:46 <mps> :)
2020-06-15 15:42:37 <wsinatra> It absolutely was, it was pretty well contested during the crusades, due to the ability to command large parts of the Mediterranean and be reasily defended. It's pretty cool stuff
2020-06-15 15:42:37 <mps> point taken
2020-06-15 15:43:11 <SpaceDandy> wsinatra, oneinsect: lbu diff shows no changes.  I have already run setup-alpine and I know the lbu system is working in principle.  I may just blow away the overlay and start again to see what happens.
2020-06-15 15:43:17 <SpaceDandy> Thanks for the boot sequence link.
2020-06-15 15:43:56 <oneinsect> i had similar issues long long back but i realized you need setup-alpine and a few things and it will work
2020-06-15 15:44:10 <oneinsect> to* setup-alpine
2020-06-15 15:44:35 <oneinsect> try it, i dont have rpi with me at the moment but could have tried
2020-06-15 15:45:27 <oneinsect> okie back to my drakonian glibc port work
2020-06-15 15:45:37 <oneinsect> almost completing
2020-06-15 15:46:04 <oneinsect> glibc is a drakula, seriously it sucks blood bad time
2020-06-15 15:46:47 <mps> and you are trying to feed it generously :)
2020-06-15 15:46:54 <oneinsect> LOL
2020-06-15 15:50:07 <SpaceDandy> Well, I started again from scratch and have the same issue.  lbu commit is saving all changes except for those in /etc/init.d. :(
2020-06-15 15:50:27 <mps> oneinsect: I follow your effort and I think you will learn a lot even if you do not end with your plan
2020-06-15 15:51:02 <SpaceDandy> I'm out of time now, so will have to pick it up tomorrow.  I'll log an issue and see if any has a bright idea overnight.
2020-06-15 15:51:48 <wsinatra> oneinsect: do you disseminate any kind of info on your port anywhere?
2020-06-15 15:52:03 <mps> lbu include -h
2020-06-15 15:53:15 <oneinsect> yeaa mps: its a terribly time consuming work but lots of learning but the product will come, i will not leave it behind
2020-06-15 15:54:13 <oneinsect> not yet wsinatra: probably some ramblings here
2020-06-15 15:54:14 <oneinsect> https://github.com/Engincs/releases/tree/master/backup
2020-06-15 15:54:44 <oneinsect> some are on an offline server (infact lots of them) i will upload once ports completes this week
2020-06-15 15:55:26 <wsinatra> I look forward to it :)
2020-06-15 15:56:00 <mps> oneinsect: I like stubborn people, they change world :)
2020-06-15 15:56:19 <mps> not conformists
2020-06-15 15:57:06 <oneinsect> lol
2020-06-15 15:57:42 <oneinsect> it never takes so much time, just that glibc and gcc combo are making mess of things
2020-06-15 15:58:02 <Lord>  /b #wee
2020-06-15 15:59:16 <oneinsect> what did Lord say?
2020-06-15 15:59:27 <oneinsect> :D could not understand
2020-06-15 15:59:41 <Lord> *you saw nothing interesting in here*
2020-06-15 15:59:50 <oneinsect> lol
2020-06-15 15:59:53 <Lord> *you'll forget it*
2020-06-15 16:01:39 <oneinsect> wsinatra: whats your github handle?
2020-06-15 16:02:11 <wsinatra> durrendal -> https://gitlab.com/Durrendal
2020-06-15 16:02:28 <wsinatra> I use github very sparingly, everything I do actively is on gitlab
2020-06-15 16:04:23 <oneinsect> sent an invite
2020-06-15 16:04:27 <hodapp> I need to start using Gitlab more
2020-06-15 16:05:17 <hodapp> also, surprised to see RetroArch, emulationstation, and libretro-* are in the repos. I have spent so many hours futzing around with RetroPie configuration and wonder if it may just be vastly easier to do all this in Alpine and have the configuration changes centralized in one place
2020-06-15 16:07:24 <wsinatra> oneinsect: awesome! Thanks!
2020-06-15 16:08:37 <wsinatra> I'll poke around when I get a chance today, really interested to see your progress, it looks like there's a ton of work here
2020-06-15 16:09:51 <oneinsect> there is lots of more work not uploaded, will do a public dump once done with this
2020-06-15 16:17:22 <wsinatra> Absolutely, any project this large is going to take time, I'm impressed to say the least though, you've done a ton in the past few months
2020-06-15 16:19:49 <datastry> Hi all!   I'm completely new to Alpine but I've been working in Linux for years (usually Ubuntu).
2020-06-15 16:20:26 <wsinatra> welcome to the part datastry
2020-06-15 16:20:30 <wsinatra> party*
2020-06-15 16:20:37 <oneinsect> i think ikke ncopa mps Cognitri and lots of others here are the behind not just the work but the help and support that goes in to anything
2020-06-15 16:21:18 <wsinatra> oneinsect: I feel you there, I wouldn't have gotten anywhere on the packages I maintain without their suggestions and just generally amiable dispositions
2020-06-15 16:21:18 <datastry> Thanks wsinatra!
2020-06-15 16:22:01 <Lord> ls
2020-06-15 16:22:27 <wsinatra> file1 file2 file3 file4 ...
2020-06-15 16:22:36 <Lord> :-þ
2020-06-15 16:22:36 <datastry> I'm hoping to discover some resources (docs, tutorials, etc.) that are relevant to a particular endeavor that I have planned.
2020-06-15 16:23:08 <wsinatra> datastry, what're you trying to build? I'm sure we can point you in some sort of a direction
2020-06-15 16:23:51 <datastry> (I'm using an IRC app that I'm not familiar with...   I haven't figured out how to send multi-line messages.   Apologies for my messages trickling out... It's not my goal to stretch it out for suspense :P  )
2020-06-15 16:24:09 <ikke> irc doesn't have multiline messages :)
2020-06-15 16:24:34 <datastry> ikke, aha!  Well that explains why I can't figure it out, haha
2020-06-15 16:27:36 <datastry> wsinatra, I'm interested in booting a minimal system into RAM from a LiveDVD.   The LBU feature of Alpine looks super promising for saving configuration changes, this is a feature that is very important to me.   However, I'm interested in saving configuration changes back to the DVD -- assuming I can boot a DVD that isn't finalized.
2020-06-15 16:28:21 <wsinatra> everyone has lbu questions today haha
2020-06-15 16:28:37 <wsinatra> That sounds like a neat idea, why a DVD and not a usb drive, or a hdd?
2020-06-15 16:30:07 <datastry> I'm sort of playing musical chairs with data and media.   Obviously I should just buy another HDD, but I'm trying to save money.
2020-06-15 16:30:53 <wsinatra> Monetary constraints are a good reason to be creative :)
2020-06-15 16:31:34 <wsinatra> Assuming that the DVD isn't finalized and you can continue to write to it, you should be okay to do that. But I've never tried using LBU outside of qemu personally
2020-06-15 16:32:21 <datastry> Cool!  Do you recommend any really useful -- and relevant -- docs?
2020-06-15 16:33:38 <wsinatra> I found the wiki helpful when I was looking at lbu https://wiki.alpinelinux.org/wiki/Alpine_local_backup
2020-06-15 16:33:45 <wsinatra> but that's really just a start
2020-06-15 16:34:09 <wsinatra> https://dev.to/nesterow/setup-minimal-kiosk-environment-with-alpine-linux-27b this article was of particular interest to me
2020-06-15 16:34:17 <datastry> Also I should point out that I'm pretty agnostic about whether I actually make use of LBU for this project.   So if there's a better tool for the job, I'm all ears.    I just stumbled upon it somehow and it sounded promising.
2020-06-15 16:34:42 <datastry> Awesome, I will take a look at these.   I appreciate these leads!
2020-06-15 16:35:00 <wsinatra> What exactly are you trying to do with it?
2020-06-15 16:35:13 <wsinatra> http://lambdacreate.com/posts/11 <- something I did with it if you're curious
2020-06-15 16:35:54 <wsinatra> for me, I needed a reproducible aarch64 instance to do SBCL compilation on, if you're trying to set something up that has a basic configuration, and you don't really care what happens outside of your baseline, then lbu is a great tool
2020-06-15 16:36:24 <wsinatra> Like, if you just want to keep things like etc configs, ssh keys, some userland stuff, and throw the rest of it out on a reboot, then you're looking in the right direction with lbu
2020-06-15 16:39:35 <datastry> Yeah I think we're on the same page here.   I'm really ok with keeping the system very minimal and simple.   For me, I think it's going to be network config, sshd config, and exFAT support (if it's not included already by default).
2020-06-15 16:40:21 <ikke> linux 5.6/5.7 include exfat support
2020-06-15 16:41:28 <datastry> ikke awesome.. one less worry
2020-06-15 16:42:22 <ikke> alpine itself is still on 5.4 though (because it's an lts release)
2020-06-15 16:42:26 <wsinatra> Yeah that sounds a lot like what I was using mine for. Happy to help out if you hit any stumbling blocks
2020-06-15 16:42:41 <wsinatra> I'm on IRC most days
2020-06-15 16:42:50 <ikke> there is linux-edge in testing that is 5.6, though
2020-06-15 16:43:28 <datastry> Near future I'd like a dead simple file server;  longer future I'd like to use VirtualBox with a remote X to run graphical apps.  Even longer future, I'd like to explore other virtualization and/or container options in place of VirtualBox
2020-06-15 16:43:53 <wsinatra> I suggest Qemu/KVM over virtualbox
2020-06-15 16:44:08 <wsinatra> It has nicer remote GUI capabilities, and a better OSS license
2020-06-15 16:44:23 <wsinatra> double plus on better features, stability, and just everything else honestly.
2020-06-15 16:45:19 <wsinatra> That said, a run from ram hypervisor sounds pretty interesting. I imagine persisting the VMs wouldn't be an issue, but it will make your lbu tarball rather large, and init times might suffer because of that
2020-06-15 16:45:38 <ikke> you would use a disk for persisting vms
2020-06-15 16:45:41 <wsinatra> Unless they're on a completely separate disc and use just persist fstab rules, which would be sensible here
2020-06-15 16:46:03 <datastry> I have years of experience with VirtualBox, so there's zero learning curve there -- in fact I already have a bunch of system images taking up space on my storage media.      But I think we are in agreement that I need to graduate to a more permanent solution.    I've been hearing about KVM for awhile... I'm interested in actually getting some
2020-06-15 16:46:04 <datastry> experience with it.
2020-06-15 16:46:10 <wsinatra> ikke: absolutely, but I don't think lbu would prevent you from trying
2020-06-15 16:46:46 <ikke> it wouldn't
2020-06-15 16:47:18 <ikke> It would just silently judge you
2020-06-15 16:47:36 <datastry> hahaha
2020-06-15 16:47:51 <wsinatra> hahah, just because of that I'm going try
2020-06-15 16:48:31 <mps> I installed alpine in sys mode on one usb flash disk and using it as rescue system
2020-06-15 16:48:48 <mps> imo, simple and stable solution
2020-06-15 16:49:04 <wsinatra> datastry: I've actually got in my mind to build a run-from-ram docker host at some point, to run an ssh-bastion and suricata. So I think we'll likely talk about this quite a bit :)
2020-06-15 16:49:14 <ikke> wsinatra: datastry left
2020-06-15 16:49:18 <mps> I can add/remove/upgrade it easy
2020-06-15 16:49:22 <wsinatra> huh I guess he did
2020-06-15 16:49:51 <wsinatra> Oh well, hopefully we set him on the right path :)
2020-06-15 16:50:08 <wsinatra> mps: good solution for a rescue disk. I typically just use the installation ISOs
2020-06-15 16:50:29 <ikke> and back
2020-06-15 16:50:51 <ikke> yeah, alpinelinux and arch live iso's are good rescue systems on their own
2020-06-15 16:51:02 <datastry> ikke , it's good to be back!
2020-06-15 16:51:08 <ikke> :)
2020-06-15 16:51:08 <wsinatra> datastry: you missed this, but ->  I've actually got in my mind to build a run-from-ram docker host at some point, to run an ssh-bastion and suricata. So I think we'll likely talk about this quite a bit :)
2020-06-15 16:51:52 <datastry> wsinatra marvelous :D   Someone to commiserate with!!
2020-06-15 16:52:49 <ikke> wsinatra: run-from-ram docker host should be fairly easy to setup
2020-06-15 16:53:15 <ikke> I would use a disk for /var/lib/docker though
2020-06-15 16:55:29 <datastry> I'm in agreement about using the disk to store config as much as possible.... well I mean I guess installing the full system 100% to disk would be "as much as possible", right?   But I guess my point is that I'm trying to be reasonable and keep the boot disc minimal, including the stuff I want to persist to the disc.
2020-06-15 16:56:11 <wsinatra> I think so too. I'm actually trying to set this up in AWS, so I'm not sure where my mileage is here. But I think it's doable
2020-06-15 16:56:28 <datastry> Cool :D
2020-06-15 16:57:00 <wsinatra> datastry: right, your DVD would host the image and mount points, but your overlay tarball and lbu stuff would be on a disk that gets mounted during init.
2020-06-15 16:57:08 <wsinatra> So you're thinking about it correctly I think :)
2020-06-15 16:57:19 <wsinatra> but bonus points if you sneak a tiny VM into the lbu volume, just because
2020-06-15 16:57:58 <datastry> Well hang on because I may or may not have missed something along the way...
2020-06-15 16:58:30 <datastry> So the current linux kernel in Alpine DOES or DOES not have exFAT support?
2020-06-15 16:58:42 <ikke> not
2020-06-15 16:58:52 <ikke> it's still 5.4, which does not include it
2020-06-15 16:59:18 <ikke> linux-edge, which is in the testing repository (so only available on edge), is 5.6, which has microsofts exfat driver
2020-06-15 16:59:28 <ikke> 5.7 swapped that out for samsungs exfat driver
2020-06-15 17:00:28 <mps> ikke: I think we have exfat in lts, from staging kernel area
2020-06-15 17:00:33 <datastry> Right now I have two disks (HDDs) available.   One is NTFS formatted, it is the internal HDD of this old PC I'm using.   The other is exFAT formatted, it is external via USB-3.0.    Both have data on them, but the internal HDD is mostly full and the external HDD is mostly empty.
2020-06-15 17:00:38 <ikke> mps: aha ok
2020-06-15 17:01:25 <datastry> My latest attempts were using Ubuntu on DVD.   I wasn't able to use a casper-rw image on either of those HDDs because those file systems weren't recognized at boot time.
2020-06-15 17:01:26 <mps> config-lts.x86_64:CONFIG_EXFAT_FS=m
2020-06-15 17:01:44 <ikke> mps: ok, good
2020-06-15 17:01:55 <ikke> datastry: so it _is_ available then
2020-06-15 17:02:06 <mps> but you are right about 5.7 swap to samsung
2020-06-15 17:02:10 <datastry> oh really.   fantastic
2020-06-15 17:02:40 <mps> and we have in testing exfatprogs for linux-edge exfat
2020-06-15 17:03:21 <mps> for older exfat driver I think exfat-utils from community should work
2020-06-15 17:03:28 <wsinatra> I use exfat for my persistent disk on my sbcl lbu systems
2020-06-15 17:03:46 <wsinatra> so it should work on edge, just as some additional validation
2020-06-15 17:05:01 <datastry> Ok, cool.   If my exFAT disk is recognized at boot time, then game on.    I thought I was going to be unable to access it until later, and that's what made me interested in persisting a minimal set of changes back to the DVD
2020-06-15 17:05:32 <ikke> datastry: You probably need to make sure the exfat driver is included in initramfs
2020-06-15 17:06:24 <datastry> True.  How's that saying go?  "No plan survives first contact with the enemy"?
2020-06-15 17:06:56 <oneinsect> if you guys have a spare alpine installation somewhere say in chroot
2020-06-15 17:07:01 <oneinsect> can you please install these
2020-06-15 17:07:02 <oneinsect> pax-utils grep pcre pcre2 git python3 bison texinfo
2020-06-15 17:07:13 <oneinsect> and share me the /lib/apk/db/installed file
2020-06-15 17:07:35 <oneinsect> would help me speed up the work
2020-06-15 17:08:10 <wsinatra> does arm64 work? or does it have to be x86_^4?
2020-06-15 17:08:21 <oneinsect> anything is fine
2020-06-15 17:08:42 <oneinsect> just need to check owner packages
2020-06-15 17:09:25 <mps> root
2020-06-15 17:10:05 <mps> anything else is suspicious
2020-06-15 17:10:10 <oneinsect> lol
2020-06-15 17:16:01 <wsinatra> https://drive.google.com/file/d/1_ikf0cEdKI4pbjwafG16o_e5EODCALNc/view?usp=sharing
2020-06-15 17:16:12 <wsinatra> oneinsect: hopefully that helps
2020-06-15 17:17:41 <datastry> Folks, it's getting late on my end here so I should get some sleep.   I appreciate your wisdom, insights, and suggestions.
2020-06-15 17:18:02 <ikke> datastry: o/
2020-06-15 17:18:18 <wsinatra> night mate, good luck!
2020-06-15 17:18:21 <datastry> Do you keep this chat open most of the time?
2020-06-15 17:18:48 <oneinsect> good nite and thank you
2020-06-15 17:18:59 <wsinatra> Yeah I typically have it open when I'm at work
2020-06-15 17:19:02 <ikke> I do
2020-06-15 17:19:12 <wsinatra> M-F US EST here
2020-06-15 17:19:14 <ikke> I have weechat in a tmux session on a vps
2020-06-15 17:19:22 <wsinatra> smart design ikke
2020-06-15 17:19:33 <datastry> Fantastic..  then hopefully we'll chat again in the future!
2020-06-15 17:19:41 <wsinatra> absolutely :D
2020-06-15 17:19:41 <ikke> And weechat has relay options for mobile clients, really nice
2020-06-15 17:19:57 <oneinsect> this is perfect thank you, going off now
2020-06-15 17:19:59 <datastry> Have a good one everybody!
2020-06-15 17:20:00 <oneinsect> good night all
2020-06-15 17:20:03 <ikke> good night
2020-06-15 17:20:05 <wsinatra> I want my nokia n900 to run Alpine AND be a daily driver, so I don't have to do any of that haha
2020-06-15 17:21:01 <ikke> The idea is that I just have a single entry to irc, and no matter whether I'm on a pc, phone or tablet, I have access to everything
2020-06-15 17:21:30 <wsinatra> It's a good design honestly. I juggle configurations between systems.
2020-06-15 17:21:44 <wsinatra> Atleast with PMOS I can just move my weechat conf onto the nokia, makes it a little easier
2020-06-15 17:21:51 <wsinatra> but not quite as good as what you've got
2020-06-15 17:23:41 <mps> I have irssi on always online server in tmux, and mostly use mosh from different machines to access it
2020-06-15 17:23:52 <ikke> yeah, I use mosh as well
2020-06-15 17:24:02 <ikke> Works well for that usecase
2020-06-15 17:55:34 <eydaimon> does the installer come with options to install ZFS as root ?
2020-06-15 17:55:40 <eydaimon> what's the default ?
2020-06-15 17:55:53 <ikke> no, it does not
2020-06-15 17:55:59 <wsinatra> ext4 is the default I think
2020-06-15 17:56:00 <ikke> use uses ext4
2020-06-15 17:56:02 <ikke> yeah
2020-06-15 18:00:04 <eydaimon> Hm. So https://wiki.alpinelinux.org/wiki/Alpine_Linux_with_root_on_ZFS_with_native_encryption  looked like it did
2020-06-15 18:01:13 <eydaimon> at least, it looks like it's possible to install ZFS on root
2020-06-15 18:01:29 <eydaimon> so I assume you answered mhy question whther it was provided in the installer or not
2020-06-15 18:02:21 <ikke> They create the zfs root first, then install alpine
2020-06-15 18:03:08 <ikke> "Please follow Installing Alpine Linux in a chroot to setup a base install of Alpine Linux. "
2020-06-15 18:03:11 <ikke> so it's not using the installer
2020-06-15 18:03:42 <Cogitri> I created that wiki entry, I installed via Ubuntu
2020-06-15 18:04:03 <eydaimon> Cogitri: I was just seeing that's what you recommended
2020-06-15 18:04:22 <eydaimon> Cogitri: I may give it a go to see if I can get it done :)
2020-06-15 18:04:57 <eydaimon> I was considering ubuntu, but I feel like alpine would be less bloated
2020-06-15 18:07:31 <Cogitri> Sure, but I figured it doesn't really matter if the install medium is bloated
2020-06-15 18:07:41 <Cogitri> So I just went with something that's easy to use and has ZFS preinstalled
2020-06-15 18:08:02 <Cogitri> I think the extended Alpine ISO comes with ZFS too so you can use that if you prefer that, but I personally haven't used it
2020-06-15 18:22:20 <eydaimon> did you do any speed comparison between ZFS encryption and LUKS + ZFS ?
2020-06-15 18:26:13 <eydaimon> I'd be moving over from FreeBSD and I suspect that I'd have to have some intermediate step using LUKS before moving over to ZFS encryption since at no time do I want to have a harddrive that's not encrypted
2020-06-15 18:29:18 <Cogitri> I haven't done any speed comparisons, but it's so nice and easy how it works with native encryption that I wouldn't want to switch
2020-06-15 19:03:31 <eydaimon> yeah, that's the idea. it's for my NAS so as long as I can stream from it and whatever, speed isn't super critical
2020-06-15 20:46:00 <TBB> according to one comparison zfs does outperform luks by a bit
2020-06-15 22:38:43 <LACampbell> first day using alpine linux at work. no disasters so far
2020-06-15 22:39:36 <LACampbell> (web dev, so pretty much just git and some horribly bloated node.js build script monstrosity I inherited)
2020-06-15 23:29:27 <eroa> I know a network connection is required for the standard ISO. But does the extended ISO required an internet connection?
2020-06-15 23:29:55 <codebam> okay so I finally found the problem with connman and wireguard
2020-06-15 23:30:04 <codebam> it's not loading the plugin because of this
2020-06-15 23:30:06 <codebam> Jun 15 19:23:50 komodo daemon.err connman-vpnd[32560]: Can't load /usr/lib/connman/plugins-vpn/wireguard.so: Error relocating /usr/lib/connman/plugins-vpn/wireguard.so: __vpn_ipconfig_foreach: symbol not found
2020-06-16 00:04:04 <codebam> was looking through the connman code and it comes from https://git.kernel.org/pub/scm/network/connman/connman.git/tree/vpn/vpn-ipconfig.c
2020-06-16 02:15:43 <herag> Hello all, just installed Alpine on my main desktop to see how it fares as a regular desktop environment. One thing I'm having issues with is getting my speakers and headphone jack to work.
2020-06-16 02:15:57 <herag> Is this the right place to ask about that?
2020-06-16 02:23:07 <maxice8> yes, this is the right place to ask about issues you're having with Alpine Linux
2020-06-16 02:23:24 <maxice8> You're using alsa ?
2020-06-16 02:45:21 <LACampbell> if you're dual booting with win10, make sure you shut it down first, don't just hit restart from the menu then boot into linux
2020-06-16 06:30:12 <ky0ko> can someone please tell me how to fix this: https://paste.sr.ht/~ky0ko/fa7713f3cab873401c744f8fc37be3f332f12141
2020-06-16 06:41:02 <danci1973> Hello...
2020-06-16 06:41:38 <danci1973> I upgraded to 3.12.0 and found out that Libreswan doesn't work.
2020-06-16 06:42:12 <danci1973> There are a couple of 'authpriv.warn pluto[2406]: NSS: AEAD decryption using AES_GCM_16_128 and PK11_Decrypt() failed (SECERR: 2 (0x2): security library: received bad data.)' lines and finally the '
2020-06-16 06:42:12 <danci1973> ABORT: ASSERTION FAILED: test_gcm_vectors(&ike_alg_encrypt_aes_gcm_16, aes_gcm_tests) (in test_ike_alg() at ike_alg_test.c:41)' ...
2020-06-16 06:42:35 <maxice8> Can you open an issue on gitlab.alpinelinux.org/aports/alpine ?
2020-06-16 06:46:55 <danci1973> maxice8: I'll try. :)
2020-06-16 06:47:10 <maxice8> Thanks, feel free to tag @Leo in the issue
2020-06-16 06:49:21 <ky0ko> managed to fix my thing. i thought apk del -r vlc might work, but it segfaulted. instead ran `apk info | grep vlc | xargs apk del` and it seemed to resolve whatever apk was hanging up on.
2020-06-16 06:50:55 <clandmeter> danci1973: you can also cc fabled
2020-06-16 06:55:01 <mps> danci1973: long time ago I switched from *swan to racoon/ipsec-tools for ipsec, and found it more stable and simpler
2020-06-16 06:56:03 <mps> though nowadays (two years ago) I deleted latest ipsec config for systems I manage
2020-06-16 06:56:21 <mps> wireguard is more simple and stable
2020-06-16 06:57:07 <mps> so, if you can, switch to WG, imo
2020-06-16 06:59:15 <danci1973> I need to communicate with other network gear, like Cisco, Checkpoint, ...
2020-06-16 07:01:03 <mps> understand
2020-06-16 07:01:29 <danci1973> Which kinda leaves me with (some implementation) of IPSEC.
2020-06-16 07:02:56 <mps> iirc, Timo (fabled) (our core developer) worked on racoon. I remember his name when i was acitve on racoon ML
2020-06-16 07:52:30 <danci1973> @mps - when you say racoon/ipsec-tools, do you mean http://ipsec-tools.sourceforge.net/ ?
2020-06-16 07:54:55 <mps> danci1973: yes
2020-06-16 07:55:07 <mps> apk info ipsec-tools
2020-06-16 07:55:31 <danci1973> @mps - not very confidence inspiring web site :D
2020-06-16 07:56:00 <mps> ok.
2020-06-16 07:56:23 <mps> I just wrote about my experience, nothing more
2020-06-16 07:58:05 <adria_> Hello everybody
2020-06-16 07:58:26 <adria_> Sorry for my English!
2020-06-16 07:58:58 <adria_> I'm a beginner with linux.
2020-06-16 07:59:22 <adria_> I want to use Alpine Linux server with docker,
2020-06-16 07:59:55 <adria_> with the following alpications run in the docker
2020-06-16 07:59:55 <adria_> plex + hw transcoding, tautulli, heimdall, jellyfin, pihole, transmission,
2020-06-16 07:59:55 <adria_> tvheadend, xteve, nextcloud, domoticz or openhab ...
2020-06-16 08:00:25 <adria_> Is this distribution good for running all these applications in the docker?
2020-06-16 08:01:29 <Cogitri> Well, any distro will do as a docker host, I suppose
2020-06-16 08:01:44 <adria_> I want to use for media server and backup (urbackup)
2020-06-16 08:02:03 <Cogitri> I run a Plex container and Nextcloud (although not via docker) on my Alpine server and that works greatr
2020-06-16 08:03:34 <adria_> hw transcodig works on plex?
2020-06-16 08:04:30 <adria_> I want to use an asrock j5005-itx motherboard
2020-06-16 08:05:50 <Cogitri> Not sure, the thing runs on a Ryzen 2600 and I usually watch at original quality
2020-06-16 08:07:15 <adria_> ok,thanks
2020-06-16 08:34:55 <omni> I boot with cgroup_no_v1=all on the cmdline and in my rc.conf I have rc_cgroup_mode="unified", rc_cgroup_controllers="cpu memory io pids" and rc_controller_cgroups="NO", still init.d/cgroups:cgroup1_controllers() seem to run and I get related errors at boot
2020-06-16 08:49:44 <omni> the errors read "mkdir: can't create directory '/sys/fs/cgroup/<enabled service>': No such file or directory" for each enabled service
2020-06-16 08:54:55 <omni> hmm... not sure that's it... cgroup1_controllers() that is
2020-06-16 10:43:49 <herag> Yeah, I'm using Alsa and Pulse. Perhaps I should just Jack Instead?
2020-06-16 10:44:23 <maxice8> There is no sound at all ?
2020-06-16 10:44:28 <herag> Oh I should have mentioned, the HDMI speakers on my monitor work, but they suck bad.
2020-06-16 10:44:58 <herag> My 'soundcard' is the Intel builtin 'card'
2020-06-16 10:45:05 <maxice8> I see, could you install pavucontrol and see if everything is alright ? (nothing is muted, the correct audio output is selected)
2020-06-16 10:45:35 <herag> Oh yeah, I did that. That's the only way I even got HDMI sound working.
2020-06-16 10:49:13 <herag> I know there's probably something simple that I haven't done, but I'm not sure what. If I knew how to edit the conf file I would, but there is so much about sound I don't know that I'm a little bit afraid to touch it.
2020-06-16 10:52:33 <Cogitri> Do you have udev installed & started?
2020-06-16 10:53:00 <Cogitri> Without that the appropriate module for your soundcard might not be loaded so pa has no chance of seeing it
2020-06-16 11:01:00 <herag> I have eudev Installed. But do you think I need to find the kernel mod and enable it?
2020-06-16 11:08:44 <Cogitri> Well, do you have the udev services started via openrc?
2020-06-16 11:08:59 <Cogitri> You could install pciutils and then run `lspci -nnk` and see if the module for your hda is loaded
2020-06-16 12:02:24 <LACampbell> how do you select a different mirror?
2020-06-16 12:03:02 <ikke> the mirrors used are located in /etc/apk/repositories
2020-06-16 12:03:43 <ikke> mirrors.alpinelinux.org has a list of mirrors
2020-06-16 12:03:47 <ikke> mirrors.a.o
2020-06-16 12:03:52 <ikke> https://mirrors.alpinelinux.org/
2020-06-16 12:03:58 <TBB> or, if you're lazy ...
2020-06-16 12:04:04 <TBB> ... setup-apkrepos, isn't it?
2020-06-16 12:04:21 <ikke> ah yes, I was looking for that
2020-06-16 12:04:24 <ikke> from alpine-conf
2020-06-16 12:05:49 <mps> iirc, setup-apkrepos add mirrors, not change
2020-06-16 12:06:35 <ikke> it does
2020-06-16 12:07:13 <LACampbell> of course I'm lazy! sweet all the setup stuff have their own binary, smart design
2020-06-16 12:08:25 <herag> Oh I will do that. Then check back with you.
2020-06-16 12:09:54 <LACampbell> fastest mirror for my australian server is in... the western US :D of course it is
2020-06-16 12:10:14 <ikke> even faster then the aarnet one?
2020-06-16 12:10:42 <LACampbell> the aarnet one is down
2020-06-16 12:11:28 <ikke> heh, that *just* happened
2020-06-16 12:11:35 <ikke> 20 minutes ago
2020-06-16 12:11:59 <herag> Looks like it's loaded.
2020-06-16 12:12:34 <ikke> apparently these things happen when they are pushing out big updates or something like that
2020-06-16 12:13:49 <herag> Intel Corp 200 Series PCH HD Audio
2020-06-16 12:13:51 <LACampbell> I belive data in australia is transferred through a series of dusty mirrors
2020-06-16 12:15:55 <herag> snd_hda_intel driver in use
2020-06-16 12:16:45 <LACampbell> herag:just as a sanity check, you did shut down windows 10 right? not sure if applicable
2020-06-16 12:18:21 <herag> Yep... well actually, when I installed Alpine, windows disappeared. It is on a separate ssd and it no longer appears in the boot menu. I could disconnect that drive because I actually only keep it around for testing. I havent used windows in over a decade.
2020-06-16 12:21:06 <herag> I did 'egrep -r -i hdmi /proc/asound/*' and it is showing a bunch of configs, but after 'pin default' is says [Jack]. I don't have Jack installed... could this be my problem?
2020-06-16 12:36:29 <grant> Okay, I'm on my PC now. This is who herag was. hehehe... I was on an android app before.
2020-06-16 20:50:24 <ahills> in-place upgrades sequentially from 3.7 through 3.12 mostly succeeded with minimal intervention (switching from linux-hardened to linux-vanilla and running apk fix a few times)
2020-06-16 20:50:38 <ahills> is there a way to find packages that have been orphaned between releases?
2020-06-16 20:55:01 <ahills> oh weird, I thought "apk search" vs. "apk info" in comm would help, but it looks like vim isn't in the repos?
2020-06-16 20:55:59 <Cogitri> https://pkgs.alpinelinux.org/packages?name=vim&branch=v3.12
2020-06-16 20:56:59 <ahills> yeah, there's definitely something wrong with the local package cache, guess I claimed success too soon
2020-06-16 21:00:37 <mps> in 3.12 (and 3.11) linux-vanilla is out-of-date, use linux-lts
2020-06-16 21:00:51 <ahills> yep, got that
2020-06-16 21:02:26 <LACampbell> does anyone else mosh into an alpine box? I cannot figure out what ports and protocol it wants. the docs say UDP and 60000:61000, but it seems to need both TCP and UDP
2020-06-16 21:02:59 <ebb> I think it's 22/tcp and that range/udp
2020-06-16 21:03:08 <ebb> for mosh in general; I haven't used on alpine
2020-06-16 21:03:44 <ahills> py3-setuptools and python3 both own site-packages/setuptools?
2020-06-16 21:03:47 <LACampbell> omfg, of course it uses 22/tcp to establish the thing
2020-06-16 21:04:04 <LACampbell> ebb: I was going crazy but that rings a bell, thanks
2020-06-16 21:48:14 <srg> Hi! I'm trying to set the timezone in 3.12. I see conflicting advice, but it seems that `setup-timezone` is the way to do this. However, it's not installed on my system. Is there a package I'm missing?
2020-06-16 21:48:58 <srg> I don't see any info on the Wiki page about needing to install a certain setup package first.
2020-06-16 21:50:21 <srg> Note I'm running this as the alpine 3.12 Docker image
2020-06-16 21:55:02 <LACampbell> at a guess the stripped down docker image doesn't have the setup packages installed, but I'm an alpine noob myself
2020-06-16 21:56:07 <Cogitri> apk search cmd:setup-timezone
2020-06-16 22:02:00 <srg> Cogitri: Got it! So it's the alpine-conf package that I need to install. So should I install it, `setup-timezone`, then uninstall that package?
2020-06-16 22:03:26 <srg> I suppose I could keep it, but I'm trying to make the image minimal
2020-06-16 22:04:25 <Cogitri> Yes, that'd work
2020-06-16 22:04:52 <srg> Thank you!
2020-06-16 22:06:59 <Cogitri> Alternatively just install tzdata and cp the appropriate file to /etc/localtime
2020-06-16 22:13:17 <mps> for example: ln -s  /etc/zoneinfo/Europe/CITY_NAME /etc/localtime
2020-06-16 22:25:43 <Hello71> that's not copying though
2020-06-16 22:26:31 <mps> sure, but why copy it
2020-06-16 22:27:40 <mps> though I don't see why not install alpine-conf if tzdata are already installed
2020-06-16 22:28:37 <LACampbell> is that all setting up the time zone does? that's kind of cool.
2020-06-16 22:30:23 <mps> I think that is enough, though someone can extract needed file from tzdata and copy it to /etc/localtime, and remove tzdata apk
2020-06-16 22:30:47 <mps> for those who want to preserve few KB
2020-06-16 22:31:35 <Cogitri> Yup, in docker containers it usually makes sense to just cp the one file you actually need
2020-06-16 22:31:49 <mps> unix/linux is simple but some distro makers like to complicate it :)
2020-06-16 22:48:50 <Hello71> mps: setup-timezone already does this :p
2020-06-16 22:55:10 <mps> Hello71: you missed my previous msg :)
2020-06-16 23:32:46 <LACampbell> using alpine to serve my web app now. ram usage is way down, though I should probably profile requests. it wouldn't surprise me if node.js is slower with musl
2020-06-17 01:10:45 <guest2391> I followed this guide but can't use my keyboard to enter my passphrase. https://wiki.alpinelinux.org/wiki/LVM_on_LUKS#Introduction
2020-06-17 01:11:14 <guest2391> It is like my keyboard doesn't respond, no caps-lock light, return key doesn't work.
2020-06-17 01:12:09 <guest2391> It is a desktop computer and I can provide more details on request.
2020-06-17 01:20:28 <Hello71> maybe you need to add usb module or something
2020-06-17 01:24:50 <guest2391> Hello71 It is added by default.
2020-06-17 01:25:00 <guest2391> I mean, it was in my mkinitfs.conf.
2020-06-17 01:27:21 <guest2391> I pretty much have an exact setup of my laptop on my desktop but, as you know, it has somewhat internal keyboard. I mean, I think it can make a difference.
2020-06-17 01:28:58 <guest2391> My mkinitfs.conf: features="ata base ide scsi usb virtio ext4 lvm keymap kms cryptsetup"
2020-06-17 01:54:12 <guest2391> I'll come back later and ask again. It's too late now and I should sleep. :D
2020-06-17 01:54:25 <guest2391> Anyways, take care. ^^
2020-06-17 01:55:58 <FesterJester> Hello all. Anyone able to help me setup mariadb? Installation goes as planned, but trying to start the service fails.
2020-06-17 01:56:26 <FesterJester> [ERROR] InnoDB: preallocating 12582912 bytes for file ./ibdata1 failed with error 95
2020-06-17 02:01:10 <Jiggy> FesterJester: That usually means it is unable to write to the file. Often due to lack of space or appropriate permissions
2020-06-17 02:02:05 <Jiggy> is this a baremetal install or within a container? if baremetal, do you have available space?
2020-06-17 02:03:28 <FesterJester> container
2020-06-17 02:03:41 <FesterJester> LXC container on Proxmox to be exact
2020-06-17 02:04:54 <Jiggy> Does the host machine have space available?
2020-06-17 02:05:14 <FesterJester> 100+GB
2020-06-17 02:05:34 <FesterJester> container has 7.8G free
2020-06-17 02:06:01 <Jiggy> is mariadb running in that container as if it's baremetal or is that a containerized service ie docker?
2020-06-17 02:07:07 <Jiggy> I see it's refering to ibdata1 relatively instead of with an absolute path, suggesting that the directory you're currently in (or that is configured as the working directory in whatever you're using for service management); are permissions of that directory suitable for writing from the user/group mariadb is running as?
2020-06-17 02:08:24 <FesterJester> bare metal and exact command I am running is "mysql_install_db --user=mysql --datadir=/var/lib/mysql"
2020-06-17 02:10:28 <Jiggy> Does the /var/lib/mysql/ directory already exist? If so, what is the owner and group assigned to /var/lib/mysql/? What are the permissions assigned to /var/lib/mysql/?
2020-06-17 02:10:30 <FesterJester> and /var/lib/mysql permissions "drwxr-sr-x mysql mysql"
2020-06-17 02:13:05 <Jiggy> Mine are the same on a default Alpine Linux install installed via the docs here - https://wiki.alpinelinux.org/wiki/MariaDB
2020-06-17 02:14:00 <FesterJester> Steps taken so far 1.apk add mysql mysql-client  2./etc/init.d/mariadb setup
2020-06-17 02:14:18 <FesterJester> Yup, that is what I am using for a reference
2020-06-17 02:14:42 <Jiggy> And you're running into the issue when running /etc/init.d/mariadb start for the first time?
2020-06-17 02:15:15 <FesterJester> Yes
2020-06-17 02:15:29 <FesterJester> That's what is so confusing to me
2020-06-17 02:15:46 <Jiggy> sounds to me like something in proxmox could be preventing anything from being written there, then considering you have space. Can you write to a file in that directory as root? what about as mysql?
2020-06-17 02:16:29 <Jiggy> touch /var/lib/mysql/testfile ; echo "test" > /var/lib/mysql/testfile
2020-06-17 02:17:05 <FesterJester> The setup appears to have started writing ibdata1 inside /var/lib/mysql/ it is 0 in size though.
2020-06-17 02:17:25 <FesterJester> and yes I was able to touch a file inside there
2020-06-17 02:17:45 <Jiggy> Were you able to write anything to that file? What about as mysql user?
2020-06-17 02:17:45 <FesterJester> and edit it
2020-06-17 02:18:50 <FesterJester> How do I change users? "su mysql" tells me "This account is not available"
2020-06-17 02:18:55 <Jiggy> You said you are running "mysql_install_db --user=mysql --datadir=/var/lib/mysql", i thought the issue was occurring when running "/etc/init.d/mariadb start" after a successful "/etc/init.d/mariadb setup"?
2020-06-17 02:19:13 <Jiggy> I would use doas to run as another user
2020-06-17 02:20:45 <FesterJester> doas: operation not permitted
2020-06-17 02:22:11 <FesterJester> looking at the /etc/init.d/mariadb script shows "mysql_install_db --user=mysql --datadir=/var/lib/mysql" as the command run for the setup switch
2020-06-17 02:23:48 <FesterJester> I may just go to using a Debian container instead.
2020-06-17 02:24:05 <Jiggy> kk so you aren't running "/etc/init.d/mariadb start", and are running into an issue at "/etc/init.d/mariadb setup", just wanted to make sure I understoof
2020-06-17 02:24:31 <FesterJester> Oh, yes.
2020-06-17 02:24:46 <Jiggy> considering this is a fresh install, have you tried reinstalling mariadb and doing the setup script again?
2020-06-17 02:24:51 <FesterJester> Did I type it wrong somewhere?
2020-06-17 02:25:09 <Jiggy> No, I just wasn't clear enough with a question I asked so I misunderstood
2020-06-17 02:26:00 <FesterJester> So, do apk del and then apk add?
2020-06-17 02:26:39 <Jiggy> check and make sure nothing related is running with ps/kill, then apk del, then remove the /var/lib/mysql/ directory
2020-06-17 02:27:15 <Jiggy> ps aux|grep -i "sql\|maria"
2020-06-17 02:33:18 <FesterJester> same result
2020-06-17 02:34:09 <FesterJester> removed packages, rebooted container, installed packages, ran setup
2020-06-17 02:39:28 <FesterJester> Know what, I am going to either setup a proper virtual machine or physical machine with alpine and see what the results are. Just to rule out the proxmox container setup.
2020-06-17 02:42:13 <FesterJester> Jiggy: Thanks for your help so far!
2020-06-17 02:42:45 <FesterJester> I will hang out here for a while in case someone has any more ideas to try
2020-06-17 02:43:06 <Jiggy> FesterJester: That's what I would do; it does sound like an issue with the container/hypervisor to me. Best of luck
2020-06-17 02:45:11 <Jiggy> https://bugzilla.proxmox.com/show_bug.cgi?id=2281
2020-06-17 02:47:12 <Jiggy> https://forum.proxmox.com/threads/mariadb-not-working-on-debian-9-lxc.53234/
2020-06-17 02:48:18 <Jiggy> I'd check the permissions of the container
2020-06-17 02:48:27 <Jiggy> privileges*
2020-06-17 03:02:39 <FesterJester> Jiggy: Already knew about that issue from a different container. Made sure to make a privileged container with nesting and nfs options enabled. :D
2020-06-17 03:03:00 <FesterJester> Also tried without options and unprivileged.
2020-06-17 03:37:03 <yunfan> mps: finally tried alpine 3.12 on my cubieboard2 , it works perfectly
2020-06-17 03:58:04 <FesterJester> Jiggy: so everything works correctly inside a virtual machine. Now I know the steps I am performing work. Guess it's something with Proxmox or LXC. Not sure which.
2020-06-17 04:02:00 <maldridge> how likely are things to break if I jump multiple alpine versions in one go?
2020-06-17 04:02:07 <maldridge> looking to go 3.9 to 3.12
2020-06-17 04:49:04 <ikke> The kernel was renamed from vanilla to lts
2020-06-17 04:49:22 <maldridge> even in virt?
2020-06-17 04:49:32 <ikke> I think so, yes
2020-06-17 04:49:48 <ikke> oh, no
2020-06-17 04:49:52 <ikke> virt is a separate kernel
2020-06-17 04:49:54 <ikke> that's still there
2020-06-17 04:49:56 <maldridge> yeah, I thought so
2020-06-17 04:50:14 <maldridge> aside, I'm still more than willing to upstream my work on building AWS AMIs
2020-06-17 04:51:10 <maldridge> went through earlier today and upgraded my base machine image to 3.12; this was when I discovered that I still had 3.9 and "3.8" machines in the fleet
2020-06-17 04:51:24 <maldridge> (I don't think they're actually running 3.8, I think they date from when my VMs were still long lived)
2020-06-17 04:52:15 <ikke> There is no literal `python` package anymore. You have to explicitly install python2 and/or python3
2020-06-17 04:52:35 <maldridge> hmm, I'll keep an eye out for that.
2020-06-17 04:53:47 <maldridge> just checked, I was already explicitly installing python3
2020-06-17 04:56:18 <Jiggy> Thanks for the update FesterJester, good luck getting it going! I haven't used either so I got nothing
2020-06-17 04:58:36 <yunfan> ikke: is it possible to add kernel parameters to enable hdmi overscan for alpine ?
2020-06-17 05:00:59 <ikke> yunfan: I don't know exactly
2020-06-17 05:02:16 <yunfan> ikke: :[ will try googling
2020-06-17 05:19:41 <ikke> I know you can set this for raspberry pi's, but I don't think this is a general kernel config option
2020-06-17 05:19:47 <ikke> any reference points to xrandr
2020-06-17 05:27:21 <yunfan> ikke: well i only need tty mode
2020-06-17 05:27:35 <yunfan> but i had tried add overscan_left parameters
2020-06-17 05:27:45 <yunfan> failed, so will try to install a X
2020-06-17 05:27:55 <yunfan> and then gave xrandr solution a try
2020-06-17 05:43:01 <yunfan> is xf86 for x86?
2020-06-17 05:43:33 <ikke> no
2020-06-17 05:43:52 <ikke> https://www.xfree86.org/
2020-06-17 05:44:06 <yunfan> ok
2020-06-17 05:44:31 <yunfan> installing xorg-server take too much software, any tiny impl?
2020-06-17 05:45:34 <ikke> The only alternative to xorg is wayland
2020-06-17 05:45:36 <ikke> afaik
2020-06-17 05:45:43 <ikke> but no clue if it's smaller or not
2020-06-17 05:45:50 <yunfan> :[
2020-06-17 05:48:24 <ikke> But I doubt that xrandr helps in the case of framebuffers
2020-06-17 05:49:01 <yunfan> well framebuffers reminds me, there were many desktop on fb :D
2020-06-17 05:49:09 <yunfan> i should turn to them for help
2020-06-17 06:24:56 <mps> yunfan: nice to hear. so you successfully run alpine on pcduino and cubieboard2
2020-06-17 06:27:15 <yunfan> mps: on cb2 , yes, for pcduino not verified, since the board were lay down at the apartment in the city, i am at my parent's home
2020-06-17 06:28:51 <mps> hmm, iirc you told previous week that you tested it on pcduino?
2020-06-17 06:29:28 <yunfan> mps: i think that's not the 3.12 version
2020-06-17 06:30:16 <mps> ah. but even if it is 3.11 that means alpine works
2020-06-17 06:31:09 <mps> alpine userspace works fine on any board I tried
2020-06-17 06:31:55 <mps> only problem could boot loader and kernel with drivers or options in kernel config
2020-06-17 06:32:05 <mps> could be*
2020-06-17 06:33:33 <yunfan> mps: yes, the current news is reboot work too :D
2020-06-17 06:33:48 <yunfan> mps: so that's perfect for using it as the board's os
2020-06-17 06:34:54 <mps> it is time to prepare document with confirmed boards where it work out-of-the-box
2020-06-17 06:36:00 <mps> I don't have much free time, first is my $day_job
2020-06-17 06:36:19 <yunfan> mps: where is the location? i could confirmed that
2020-06-17 06:36:37 <mps> and for alpine there are a lot of pkgs which I'm trying to fix or improve
2020-06-17 06:37:26 <mps> yunfan: we can create wiki page about alpine on arm but I dislike wiki
2020-06-17 06:37:41 <yunfan> mps: which format do you prefer?
2020-06-17 06:37:44 <yunfan> markdown?
2020-06-17 06:37:47 <yunfan> or latex?
2020-06-17 06:37:54 <mps> markdown
2020-06-17 06:38:06 <yunfan> well, not a problem to me
2020-06-17 06:38:24 <yunfan> i use mardown as my blog'sformat
2020-06-17 06:38:30 <mps> also for me, but we should publish it somewhere
2020-06-17 06:39:03 <yunfan> i guess wiki support markdown with plugins
2020-06-17 06:39:13 <mps> you have blog? mind you to post url
2020-06-17 06:39:41 <yunfan> mps: i dont have a english blog, mine is written in chinese
2020-06-17 06:39:46 <mps> yunfan: yes, but our wiki doesn't have this plugin
2020-06-17 06:40:26 <mps> (uh, why people speak different languages)
2020-06-17 06:40:57 <mps> and program in different languages :D
2020-06-17 06:41:00 <yunfan> mps: as a developer, i had wished many times that why my mother tone is not english
2020-06-17 06:41:47 <yunfan> mps: perhaps i could setup a english blog for practise it :D
2020-06-17 06:41:47 <mps> I'm joking, it is good that people speak different languages
2020-06-17 06:42:31 <yunfan> mps: but i am serious, if i could use english as my native lang, i can read so many tech documents and science fiction :D
2020-06-17 06:43:45 <mps> no one can think clearly in language which is different of their mothers language
2020-06-17 06:44:24 <mps> that is our destiny
2020-06-17 06:44:57 <yunfan> do you know alpinelinux?
2020-06-17 06:45:56 <mps> I don't understand question
2020-06-17 06:46:30 <yunfan> sorry, its tinycore linux
2020-06-17 06:46:59 <mps> ah, I read about it but didn't tried it
2020-06-17 06:47:46 <yunfan> its quit small with a X server
2020-06-17 06:48:30 <mps> yes, I remember
2020-06-17 06:48:54 <mps> but I'm not sure if it is ok for my use case
2020-06-17 06:48:56 <yunfan> shold i install glibc for compiling when there is ctypes error?
2020-06-17 06:49:21 <mps> I want functional and small, where alpine fits quite fine
2020-06-17 06:50:57 <yunfan> mps: then how to do hdmi overscan without X ?
2020-06-17 06:50:58 <mps> I think you don't need glibc, in most cases
2020-06-17 06:51:59 <yunfan> mps: i am compiling  fbpad, which told me ctypes error
2020-06-17 06:52:37 <yunfan> no such file or directory for ctype.h
2020-06-17 06:52:50 <mps> huh, could you check what needs to be patched in 'fbpad' (whatever it is)
2020-06-17 06:53:39 <mps> apk info -W /usr/include/ctype.h => /usr/include/ctype.h is owned by musl-dev-1.2.0-r0
2020-06-17 06:54:16 <yunfan> ah, musl-dev
2020-06-17 06:55:50 <yunfan> mps: back to the wiki issue, is it possible to have a github repo to host such tech documents?
2020-06-17 06:55:58 <yunfan> so people could just send pr to it
2020-06-17 06:56:27 <mps> no github, please
2020-06-17 06:56:43 <mps> it is non-free
2020-06-17 06:57:07 <Cogitri> We have our own Gitlab instance, so might as well use that
2020-06-17 06:58:02 <mps> Cogitri: yes, and I'm waiting infra team, when they will 'amend' use its wiki options
2020-06-17 06:58:10 <yunfan> okay, how about host that in your own git
2020-06-17 06:58:44 <yunfan> and i think the markdown format articles could be render into a static site
2020-06-17 06:58:57 <mps> I'm for it
2020-06-17 06:59:46 <yunfan> by instaling musl-dev helped me solve the ctypes problem, then i met no linux/vt.h
2020-06-17 06:59:47 <mps> waiting for my daughter to finish their hugo learnig, and I can host this on my web page, till we can have it gitlab
2020-06-17 07:00:09 <mps> yunfan: kernel-headers?
2020-06-17 07:00:35 <yunfan> mps: i had a daughter too :D  in 2 month age
2020-06-17 07:01:31 <mps> sorry, linux-headers contains vt.h
2020-06-17 07:02:43 <mps> yunfan: yes, I remember when you as happy father told me about your newborn daughter :)
2020-06-17 07:02:49 <yunfan> cool compiling works
2020-06-17 07:02:57 <yunfan> i love ansi c project
2020-06-17 07:04:58 <mps> standard C (ansi) is only portable language, besides perl :)
2020-06-17 07:07:01 <LACampbell> what standard? :p
2020-06-17 07:07:51 <LACampbell> oh I see, Standard C is another term for C89
2020-06-17 07:08:00 <yunfan> i only use perl before i havnt learnt python
2020-06-17 07:08:07 <yunfan> that's a decade ago
2020-06-17 07:09:16 <mps> python is invented for people who can't remember few characters, i.e. '{};'
2020-06-17 07:09:46 <yunfan> no worry all of this will be solved in python 6
2020-06-17 07:10:12 <mps> :D
2020-06-17 07:11:12 <LACampbell> perl had that culture of "one liners" which ruby and python never really had
2020-06-17 07:12:39 <mps> C and perl are most hated lang but they keep computers and internet working
2020-06-17 07:12:53 <mps> langs*
2020-06-17 07:13:05 <yunfan> i prefer one liners style
2020-06-17 07:13:10 <yunfan> even in python
2020-06-17 07:13:18 <yunfan> by abusing list comprehension
2020-06-17 07:13:27 <LACampbell> I can't be bothered hating languages anymore. I just hate over-engineering
2020-06-17 07:14:57 <LACampbell> yunfan: oh yeah I remember those, they were great. a shame they never caught on to other langs
2020-06-17 07:16:04 <yunfan> LACampbell: you could try those on lisp or haskell or esnext
2020-06-17 07:16:30 <LACampbell> wait javascript is getting list comprehensions?!
2020-06-17 07:16:59 <yunfan> not completely, but js is being more and more pythonic
2020-06-17 07:17:22 <yunfan> i remember when i learning es6 features, the turtorial just said, its like blah blah in python :D
2020-06-17 07:18:03 <LACampbell> yeah JS really isn't that bad these days. are you a developer?
2020-06-17 07:18:38 <LACampbell> https://developer.mozilla.org/en-US/docs/Archive/Web/JavaScript/Array_comprehensions removed from the standard :(
2020-06-17 07:19:01 <mps> a man who understand forth could understand anything ;)
2020-06-17 07:19:20 <maxice8> Try understanding steamed hams in the style of nier automata
2020-06-17 07:20:18 <mps> :)
2020-06-17 07:24:09 <yunfan> LACampbell: i am a backend developer , sometimes were forced to fix some frontend issues
2020-06-17 07:24:30 <yunfan> LACampbell: and i use typescipt in the last year for backend services
2020-06-17 07:24:58 <yunfan> mps: yes, why not just boot to forth and let it handle everything :D
2020-06-17 07:25:22 <yunfan> maxice8: chunk moore the founder of forth, even make chips :D
2020-06-17 07:25:40 <LACampbell> yunfan: oh cool. I am a front end developer these days. cool to see another webdev using alpine
2020-06-17 07:25:48 <maxice8> Hope they are as tasty as my bbq ruffles
2020-06-17 07:26:34 <yunfan> LACampbell: dont get me wrong, the online service dont use alpine, but in the previous company, there were only one coder, which is me, so i had use alpine based nodejs image for deploy services
2020-06-17 07:26:48 <yunfan> for python service, its quite easy to met problem
2020-06-17 07:27:10 <yunfan> but for nodejs , alpine based image is usable in most case
2020-06-17 07:27:54 <yunfan> LACampbell: since you are front end developer, which stack were you using?
2020-06-17 07:28:29 <mps> yunfan: chuck moore, not chunk :)
2020-06-17 07:28:45 <LACampbell> yunfan: for my professional work I use react.js. it's so-so. I do all my dev on alpine WSL
2020-06-17 07:28:47 <yunfan> mps: ok
2020-06-17 07:29:06 <mps> though real name is Charles H. Moore
2020-06-17 07:29:09 <yunfan> LACampbell: i had fall in love with hyperapp some days
2020-06-17 07:29:23 <LACampbell> oh, I read about that one! have you used it?
2020-06-17 07:30:04 <LACampbell> it was like 2kb or something. I'm big on keeping frontend code small
2020-06-17 07:30:33 <yunfan> LACampbell: yes, i use that to make a chatroom for learning python's websocket supporting
2020-06-17 07:31:00 <yunfan> LACampbell: but when i want to make a real small site, i found it lacks of features
2020-06-17 07:31:31 <LACampbell> oh yeah, which features?
2020-06-17 07:31:52 <yunfan> LACampbell: network flow fee is a huge bill in china than the outside
2020-06-17 07:32:06 <yunfan> so keep these assets small is a money problem
2020-06-17 07:32:20 <mps> yunfan: if I don't understand how industry works and what drives it, I would wonder why we don't actually 'boot to forth'
2020-06-17 07:32:40 <yunfan> LACampbell: i dont remember, since i am not a professional front end developer :D
2020-06-17 07:32:54 <mps> but industry wouldn't earn so much money by using simple and elegant solutions
2020-06-17 07:33:04 <yunfan> mps: consider forth as lua
2020-06-17 07:33:30 <yunfan> i think the problem is forth community is full of elder developer who dont had interesting of modern PC
2020-06-17 07:33:47 <mps> no
2020-06-17 07:33:55 <yunfan> 8th's founder is focus on PC/cellphone, i think its interesting
2020-06-17 07:34:04 <yunfan> and i had talked with him serveral times
2020-06-17 07:34:17 <mps> Israels military and aviation use forth
2020-06-17 07:34:30 <LACampbell> mps: I don't care what the tech industry things anymore. I just help businesses make money, and you can make them happy with simple and elegant
2020-06-17 07:34:33 <LACampbell> *thinks
2020-06-17 07:34:54 <TBB> back to basics: do everything in COBOL
2020-06-17 07:34:57 <TBB> 
ACTION hides

2020-06-17 07:35:00 <yunfan> mps: wow, i dont konw he has this background
2020-06-17 07:35:32 <yunfan> mps: since i love algriculture too, hope someday i could saw forth in the controller of iiregation system :D
2020-06-17 07:35:41 <LACampbell> yunfan: I am serving a simple website from alpine, works fine
2020-06-17 07:36:05 <yunfan> LACampbell: which other domain do you like?
2020-06-17 07:36:07 <LACampbell> lol I've been doing programming for agricultural companies for the past few years
2020-06-17 07:36:10 <mps> heh, forth cpu RTX2000 is only one CPU chip which can survive nuclear radiation, you understand rest :)
2020-06-17 07:37:03 <LACampbell> yunfan: domain, like front-end/back-end?
2020-06-17 07:38:07 <yunfan> LACampbell: no, like algriculture to me
2020-06-17 07:38:41 <LACampbell> ahh. whatever makes me money I guess :D I'd love to do something with language learning but I don't thibk there's any money in it
2020-06-17 07:39:08 <yunfan> LACampbell: but when you eared money, which domain would you like to use on ?
2020-06-17 07:39:33 <yunfan> eating? buying? traveling?
2020-06-17 07:39:40 <yunfan> or growing like me :D
2020-06-17 07:39:50 <LACampbell> oh, not related to work/money? like hobbies?
2020-06-17 07:40:10 <yunfan> mps: i think spacex's recently success work would make that useless
2020-06-17 07:40:32 <yunfan> the future belongs to redundancy
2020-06-17 07:40:39 <yunfan> LACampbell: yep
2020-06-17 07:41:13 <mps> future belongs to future, not nowadays thinking
2020-06-17 07:41:31 <LACampbell> yunfan: don't have much time for hobbies! even alpine is for work
2020-06-17 07:41:45 <yunfan> LACampbell: ;[
2020-06-17 07:41:56 <LACampbell> trying to get started as an independent web developer
2020-06-17 07:42:09 <LACampbell> consultant/freelancer whatever you want to call it. have one client
2020-06-17 07:43:24 <yunfan> LACampbell: i think this cornos issue created many jobs for you front end developer
2020-06-17 07:43:36 <yunfan> people need video/audio tools
2020-06-17 07:44:03 <yunfan> but your western people were afraid of privacy issue, so a local host solution might be sellable
2020-06-17 07:44:30 <LACampbell> haha yes. you're from china?
2020-06-17 07:44:31 <yunfan> LACampbell: why not using those webrtc tech to build such tools for enterprises which has self host needs
2020-06-17 07:44:36 <yunfan> LACampbell: yes, china mainland
2020-06-17 07:45:23 <LACampbell> well it's not a tech issue, it's a sales/marketing issue
2020-06-17 07:46:02 <yunfan> i dont have chance to verify it :[
2020-06-17 07:46:36 <LACampbell> you need to find people with money and convicne them you can solve their problem. that's my bottleneck
2020-06-17 07:47:47 <yunfan> LACampbell: my bottleneck is skills :D
2020-06-17 07:47:56 <yunfan> that;s why i recently want to learn front end
2020-06-17 07:47:59 <LACampbell> haha, you are good at sales then?
2020-06-17 07:48:12 <yunfan> i should first make a prototype and then got money
2020-06-17 07:48:16 <LACampbell> if you like alpine, and like minimal things, the front end culture will annoy you like it annoys me maybe :D
2020-06-17 07:48:27 <LACampbell> they pull in many many libraries that do almost nothing
2020-06-17 07:48:31 <yunfan> LACampbell: i dont know, but i think i could talk than ordinary developer
2020-06-17 07:49:11 <yunfan> LACampbell: my master degree is public management
2020-06-17 07:49:19 <yunfan> i am self taught coder
2020-06-17 07:49:31 <LACampbell> I can talk with business people once we are working together, but sales is another skill
2020-06-17 07:51:01 <yunfan> LACampbell: i havnt been a saleman, so i cant promise that i could do that easily. but anyway, we could have a try when there were chance
2020-06-17 08:05:55 <winny> I'd like to get debug symbols for the parted package. I don't see a -dbg package. How can I install debug symbols for parted?
2020-06-17 08:06:42 <Cogitri> You'll probably have to add it to the APKBUILD
2020-06-17 08:06:52 <ikke> and build it yourself
2020-06-17 08:07:10 <ikke> providing debug symbols for all pacakges is requiring a lot of storage
2020-06-17 08:07:13 <Cogitri> Unfortunately we don't have -dbg subpackages for all packages since that'd take up too much space on the mirrors apparently
2020-06-17 08:07:25 <winny> is that the preferred method? (The other approach I was considering is simply modifying my build env to not strip)
2020-06-17 08:07:51 <ikke> winny: it depends on what you prefer yourelf
2020-06-17 08:07:57 <ikke> yourself*
2020-06-17 08:09:38 <winny> Right, I have two good options now =) Another question, might be asking too much, because it sounds like a pain to set up: is it possible to install source code for line-by-line debugging information? For example on gentoo I can configure portage to install sources, and it'll also embed the source path in the installed ELF files, so when I load up gdb it'll have the sources all set up across all
2020-06-17 08:09:39 <winny> packages
2020-06-17 08:11:17 <ikke> When you build from an APKBUILD with debugging, the source is already present
2020-06-17 08:14:48 <mps> heh, just read this https://tonsky.me/blog/syncthing/ when we talked about free and software for money
2020-06-17 08:15:22 <winny> right, though i imagine manually configuring source locations get a bit much because most applications will cross several package boundaries
2020-06-17 08:16:20 <winny> actually... maybe i can just tel abuild to never clean up
2020-06-17 08:17:11 <mps> winny: abuild unpack
2020-06-17 08:22:58 <mps> or, abuild -K, -K  Keep buildtime temp dirs and files (srcdir/pkgdir/deps)
2020-06-17 08:35:56 <winny> oh I see, even better
2020-06-17 08:36:17 <winny> looks like most of the vars in the stackframes are <optimized out>
2020-06-17 08:36:49 <winny> time to see what compile flags parted is being built with
2020-06-17 08:38:00 <Cogitri> You'd have to built with -O1 or even -O0 to not have that, I guess
2020-06-17 08:38:13 <winny> ah insult to injury, only now did I realize that abuild has a usage function
2020-06-17 08:38:15 <Cogitri> And we build with -Os by default
2020-06-17 08:39:12 <winny> ahh right, -Os!
2020-06-17 08:52:54 <winny> Rebuild musl and parted both with CFLAGS="-O2 -ggdb" and the segfault went away
2020-06-17 10:37:39 <guest1532> Is there any IRC logs stored in somewhere? I asked a question in fairly late around the world and want to see if it is answered or not. :)
2020-06-17 10:37:56 <guest1532> late time i mena
2020-06-17 10:37:58 <guest1532> mean*
2020-06-17 10:38:04 <ikke> https://irclogs.alpinelinux.org
2020-06-17 10:38:07 <artok> https://irclogs.alpinelinux.org/
2020-06-17 10:38:17 <artok> too slow again
2020-06-17 10:38:24 <guest1532> thank all J3
2020-06-17 10:38:34 <guest1532> <3*
2020-06-17 10:41:52 <guest1532> Nope, nobody has answered yet.
2020-06-17 10:43:08 <guest1532> Basically I do what https://wiki.alpinelinux.org/wiki/LVM_on_LUKS tells me to do but my keyboard doesn't work when I want to type my passphrase in the password prompt.
2020-06-17 10:44:19 <guest1532> Like caps-locks' light doesn't work, return key doesn't work (like if you press return with a wrong password, including empty password, it will tell you that it is not matched).
2020-06-17 10:44:45 <ikke> It means the correct driver / module is missing
2020-06-17 10:44:51 <ikke> key is to find out what module
2020-06-17 10:44:58 <guest1532> Someone told me to add `usb` to the mkinitfs.conf but it was already there when I was addding `cryptsetup` to the end of the list.
2020-06-17 10:45:00 <ikke> (no pun intended :P)
2020-06-17 10:45:14 <guest1532> How can I do that?
2020-06-17 10:45:34 <ikke> maybe check lsmod for anything that looks related
2020-06-17 10:45:56 <guest1532> Unfortunately, I have only one keyboard to test, BTW.
2020-06-17 10:46:21 <ikke> I had the same issue on archlinux. Laptop keyboard worked, but usb didn't (did not solve it yet)
2020-06-17 10:46:24 <guest1532> Let me boot up the machine with the USB stick again.
2020-06-17 10:46:57 <guest1532> Did you try different keyboards?
2020-06-17 10:47:11 <ikke> no
2020-06-17 10:47:21 <guest1532> I was just able to use different ports of the machine, hence I don't have any plus keyboard.
2020-06-17 10:47:23 <guest1532> I see.
2020-06-17 10:47:35 <guest1532> s/use/try/
2020-06-17 10:48:07 <mps> hmm, my PS2 keyboard form 90-ies works over usb adapters always
2020-06-17 10:48:20 <mps> even AT keyboard
2020-06-17 10:48:44 <guest1532> I can show how my keyboard looks like, or whatever information I can provide.
2020-06-17 10:48:51 <mps> AT to PS2 then to USB adapter
2020-06-17 10:49:16 <ikke> guest1532: Wouldn't help me honestly
2020-06-17 10:49:35 <ikke> lsmod output might, but not sure
2020-06-17 10:50:03 <guest1532> Will do it, just a sec.
2020-06-17 10:54:38 <guest1532> ikke http://i
2020-06-17 10:54:49 <guest1532> Oh, sorry, the ISO/ANSI thing.
2020-06-17 10:54:59 <guest1532> http://ix.io/2por
2020-06-17 10:55:03 <guest1532> There you go ^
2020-06-17 10:56:21 <guest1532> ikke It's what you want, right?
2020-06-17 10:56:26 <ikke> yes
2020-06-17 10:56:39 <guest1532> Good then. :)
2020-06-17 10:57:27 <guest1532> I assumed I didn't need to unlock and possibly mount the encrypted parts, is it okay?
2020-06-17 10:58:06 <ikke> yes, this is enough
2020-06-17 11:02:37 <ikke> so features refers to a features file in /etc/mkinitfs/features.d
2020-06-17 11:03:57 <guest1532> What do you mean?
2020-06-17 11:04:24 <adria_> Hello everybody
2020-06-17 11:04:39 <adria_> How to install vainfo to alpine linux?
2020-06-17 11:04:57 <ikke> guest1532: I was trying to figure out what those features in /etc/mkinitfs/mkinitfs.conf meant
2020-06-17 11:05:11 <guest1532> I can provide that file, if you want?
2020-06-17 11:05:19 <ikke> no, not necessary
2020-06-17 11:05:28 <ikke> adria_: libva-utils is in testing
2020-06-17 11:05:48 <ikke> adria_: so you would need to run on edge to use them
2020-06-17 11:06:00 <guest1532> I'm probably not as good as you are understanding English. What do you want from me, e
2020-06-17 11:06:04 <guest1532> exactly?*
2020-06-17 11:06:07 <adria_> ok,thanks
2020-06-17 11:06:36 <ikke> guest1532: I'm more talking out load
2020-06-17 11:06:38 <ikke> loud*
2020-06-17 11:06:53 <ikke> trying to figure out what is missing
2020-06-17 11:07:46 <guest1532> Oh, I see. Sorry for bothering then. :D
2020-06-17 11:11:02 <mps> ikke: I think libva-utils can be run safely on 3.12
2020-06-17 11:11:05 <mps> stable
2020-06-17 11:16:34 <guest1532> I think I lost connection for a while.
2020-06-17 11:20:33 <guest1532> ikke You sure you don't need any other information? Maybe, I can find something useful from dmesg?
2020-06-17 11:23:32 <ikke> Could you try something like rmmod hid_lenovo?
2020-06-17 11:23:47 <ikke> basically removing modules until your keyboard stops working
2020-06-17 11:25:03 <guest1532> Which modules should I remove?
2020-06-17 11:25:13 <ikke> did you try hid_lenovo?
2020-06-17 11:25:40 <guest1532> hid_lenovo stopped my keyboard from working !
2020-06-17 11:25:47 <guest1532> I think it is good news, right?
2020-06-17 11:25:47 <ikke> ok, bingo
2020-06-17 11:25:50 <ikke> yes
2020-06-17 11:26:16 <ikke> Now you need to reboot and unlock your drive
2020-06-17 11:26:31 <guest1532> Can I use ssh to add that module back?
2020-06-17 11:26:38 <ikke> sure
2020-06-17 11:26:41 <guest1532> I started ssh. ^^
2020-06-17 11:26:54 <ikke> ok, but not sure if you can login as root though
2020-06-17 11:27:26 <guest1532> Oh, but I can't login as root because of the confings, derp. :D
2020-06-17 11:29:13 <guest1532> Waiting for the mirrors to sort themselves.
2020-06-17 11:33:49 <guest1532> Okay, I booted and unlocked the encrypted partition.
2020-06-17 11:34:06 <guest1532> Now I guess I'll mount the necessary parts under `/mnt`, correct?
2020-06-17 11:34:06 <ikke> Ok, now mount it somehwere (probably /mnt
2020-06-17 11:34:09 <ikke> yes
2020-06-17 11:34:13 <guest1532> yes, sir.
2020-06-17 11:34:17 <ikke> / proc sys dev
2020-06-17 11:34:19 <ikke> and boot
2020-06-17 11:34:27 <guest1532> Okay.
2020-06-17 11:34:32 <guest1532> Just gimme a sec.
2020-06-17 11:35:18 <ikke> for F in proc sys dev boot; do mount -orbind /$F /mnt/$F; done
2020-06-17 11:35:30 <ikke> assuming you mounted / on /mnt already
2020-06-17 11:35:39 <ikke> and boot was mounted on /boot
2020-06-17 11:37:09 <guest1532> I lost connection again.
2020-06-17 11:37:18 <guest1532> I did what wiki showed me:
2020-06-17 11:37:20 <guest1532> # mount -t proc /proc /mnt/proc# mount --rbind /dev /mnt/dev# mount --make-rslave /mnt/dev# mount --rbind /sys /mnt/sys
2020-06-17 11:37:27 <ikke> should work
2020-06-17 11:37:28 <guest1532> ikke okay?
2020-06-17 11:37:29 <ikke> ok
2020-06-17 11:37:31 <guest1532> okay. ^^
2020-06-17 11:37:39 <guest1532> What should I do now?
2020-06-17 11:37:41 <ikke> You miss /boot
2020-06-17 11:37:43 <guest1532> Everything is mounted.
2020-06-17 11:37:50 <guest1532> No, I did mount boot and home.
2020-06-17 11:37:52 <ikke> ok
2020-06-17 11:37:57 <ikke> chroot /mnt /bin/sh
2020-06-17 11:38:25 <guest1532> done and read /etc/profile
2020-06-17 11:38:28 <ikke> ok
2020-06-17 11:38:46 <ikke> now create a file /etc/mkinitfs/features.d/lenovo.modules
2020-06-17 11:39:04 <guest1532> IT is a file file, right?
2020-06-17 11:39:13 <ikke> sorry?
2020-06-17 11:39:33 <guest1532> like you do `touch lenovo.modules` instead of `mkdir lenovo.modules`, right?
2020-06-17 11:39:37 <ikke> yes
2020-06-17 11:39:43 <ikke> or directly open it with an editor
2020-06-17 11:39:58 <guest1532> Done,
2020-06-17 11:41:09 <ikke> add 'kernel/drivers/hid/hid-lenovo.ko' to that file
2020-06-17 11:41:25 <guest1532> Like literal text?
2020-06-17 11:41:41 <ikke> yes
2020-06-17 11:41:48 <ikke> without the quotes
2020-06-17 11:41:55 <guest1532> done
2020-06-17 11:42:13 <ikke> now update /etc/mkinitfs/mkinitfs.conf and add lenovo to features
2020-06-17 11:42:32 <guest1532> Is there an order to add that?
2020-06-17 11:43:02 <ikke> probably after usb
2020-06-17 11:43:13 <guest1532> Done.
2020-06-17 11:43:41 <ikke> now run mkinitfs
2020-06-17 11:43:47 <guest1532> without any args?
2020-06-17 11:43:56 <ikke> You need some args
2020-06-17 11:44:08 <guest1532> Listening.
2020-06-17 11:44:56 <ikke> hold on
2020-06-17 11:44:56 <guest1532> There's a command on the wiki page I linked, but it does that not inside the chroot.
2020-06-17 11:45:12 <guest1532> `# mkinitfs -c /mnt/etc/mkinitfs/mkinitfs.conf -b /mnt/ $(ls /mnt/lib/modules/)`
2020-06-17 11:45:19 <ikke> right
2020-06-17 11:45:27 <guest1532> But outside the chroot.
2020-06-17 11:46:09 <ikke> just mkinitfs -c /etc/mkinitfs/mkinitfs.conf should work
2020-06-17 11:46:13 <ikke> inside the chroot
2020-06-17 11:46:19 <guest1532> OKay, sir.
2020-06-17 11:46:40 <ikke> ah, you also need to specify the kernel version I ugess
2020-06-17 11:47:02 <ikke> so just use the same snippet
2020-06-17 11:47:09 <ikke> without /mnt
2020-06-17 11:47:20 <guest1532> It said `==> initramfs: creating /boot/initramfs-lts`.
2020-06-17 11:47:27 <ikke> ok, should be fine then
2020-06-17 11:47:42 <guest1532> Well, I'll do what you told me if it doesn't work. :D
2020-06-17 11:47:46 <guest1532> Are we done now?
2020-06-17 11:47:54 <guest1532> I mean, can I unmount everything and reboot?
2020-06-17 11:48:03 <ikke> yes, I suppose so
2020-06-17 11:48:08 <guest1532> Okay.
2020-06-17 11:48:55 <guest1532> Still doesn't work.
2020-06-17 11:49:15 <guest1532> Now, let me try the `$(ls /lib/modules/)` part as well.
2020-06-17 11:58:08 <guest1532> Doing it now.
2020-06-17 11:59:15 <guest1532> Done.
2020-06-17 11:59:25 <guest1532> I've got the same output, though.
2020-06-17 12:00:56 <guest1532> ikke Nope, it doesn't work.
2020-06-17 12:01:28 <ikke> ok, then I'm out of ideas myself :(
2020-06-17 12:01:39 <guest1532> :(
2020-06-17 12:01:59 <ikke> It's kind of hard to inspect the current situation
2020-06-17 12:02:27 <guest1532> I would like to provide as much information I can.
2020-06-17 12:02:49 <ikke> Understood, but it's not that you can run lsmod right now
2020-06-17 12:03:26 <guest1532> IS there something like ssh to unlock?
2020-06-17 12:03:49 <ikke> nothing provided by AL
2020-06-17 12:03:58 <guest1532> I see.
2020-06-17 12:04:34 <guest1532> Having the exact setup work on my laptop but not my desktop as well kinda feels bad. :D
2020-06-17 12:05:00 <KatolaZ_> guest1532: computer science is neither fair nor deterministic
2020-06-17 12:05:05 <KatolaZ_> ;)
2020-06-17 12:05:09 <guest1532> lol
2020-06-17 12:06:10 <mps> KatolaZ_: something like astrology combined with true random generator, as I told my son this morning
2020-06-17 12:08:36 <KatolaZ_> reality is that most of the stuff we use relies on layers upon layers upon layers of unnecessary complications, but I guess that's a rant for another day
2020-06-17 12:08:39 <KatolaZ_> sorry :)
2020-06-17 12:08:49 <KatolaZ_> sorry for not being able to help guest1532
2020-06-17 12:08:59 <guest1532> No worries. :)
2020-06-17 12:09:18 <guest1532> But agreed about the layers of layers of abstractions.
2020-06-17 12:09:39 <ikke> I don't know mkinitfs very well, so it might be that I forgot something
2020-06-17 12:10:10 <guest1532> I think I never had any issue with luks+lvm before in any distro is because of I always used my internal keyboard of laptop.
2020-06-17 12:10:19 <mps> yes, emmc kernel driver worked fine more than 8 days, yesterday I rebooted it to check something and now it crashed 3 times
2020-06-17 12:10:23 <guest1532> I recently got some money to build my pc.
2020-06-17 12:10:43 <guest1532> I mean, desktop computer, not pc.
2020-06-17 12:12:59 <KatolaZ_> guest1532: maybe you just need to have a kernel with USB support build it
2020-06-17 12:13:19 <KatolaZ_> betterm with support for your keyboard
2020-06-17 12:13:26 <KatolaZ_> in theory the module should be in the intramfs
2020-06-17 12:13:42 <guest1532> I have `usb` in the mkinitfs.conf.
2020-06-17 12:13:48 <KatolaZ_> well, that might not be enough
2020-06-17 12:13:50 <guest1532> I think it is for usb suppor.
2020-06-17 12:14:02 <guest1532> Hmm, what should I add then?
2020-06-17 12:14:05 <KatolaZ_> that's just the generic support I guess
2020-06-17 12:14:31 <KatolaZ_> guest1532: which modules you have loaded with that keyboard on a running distro
2020-06-17 12:14:34 <KatolaZ_> ?
2020-06-17 12:14:47 <guest1532> I have no other keyboard, but I can confirm that my mouse's light is working.
2020-06-17 12:14:49 <ikke> KatolaZ_:  $(ls /mnt/lib/modules/)
2020-06-17 12:14:52 <ikke> oops
2020-06-17 12:15:08 <guest1532> I'm on my laptop with the same setup with Alpine.
2020-06-17 12:15:15 <ikke> http://ix.io/2por
2020-06-17 12:15:27 <guest1532> you want lsmod of my laptop?
2020-06-17 12:16:12 <KatolaZ_> guest1532: you mentioned you are using an external keyboard right?
2020-06-17 12:16:26 <guest1532> On my desktop, correct.
2020-06-17 12:16:39 <KatolaZ_> you should check which module gets loaded when you plug the keyboard in
2020-06-17 12:17:05 <guest1532> I don't know how can I do that.
2020-06-17 12:17:13 <KatolaZ_> dmesg
2020-06-17 12:17:16 <KatolaZ_> will tell you
2020-06-17 12:17:34 <guest1532> On my laptop, correct?
2020-06-17 12:17:44 <ikke> KatolaZ_: one thing we found out is that the keyboard stopped working when they removed hid_lenovo
2020-06-17 12:17:56 <guest1532> I'm he, BTW.
2020-06-17 12:18:09 <KatolaZ_> ok
2020-06-17 12:19:17 <KatolaZ_> guest1532: do you have that hid_lenovo module in the initramfs?
2020-06-17 12:19:30 <KatolaZ_> no waut
2020-06-17 12:19:32 <KatolaZ_> wait
2020-06-17 12:19:43 <KatolaZ_> hid_lenovo is the lenovo-specific hid driver
2020-06-17 12:19:44 <guest1532> I think I have. Because ikke guided me to do that, I assumed.
2020-06-17 12:19:53 <KatolaZ_> it's not the same you need in your desktop, most probably
2020-06-17 12:20:19 <KatolaZ_> just boot a plain alpine on your desktop
2020-06-17 12:20:25 <KatolaZ_> and see which hid-related modules are loaded there
2020-06-17 12:20:33 <guest1532> This is what I have: http://17c4dcd7f91259d8cc66-f5932f6db0039e8c02f89a70c334ff0e.r2.cf1.rackcdn.com/wp-content/uploads/sites/2/48086.jpg
2020-06-17 12:21:10 <KatolaZ_> yeah guest1532 but I am not employed at lenovo :D
2020-06-17 12:21:10 <guest1532> Okay, how can I list loaded modules (and possibly grep them)?
2020-06-17 12:21:19 <KatolaZ_> guest1532: man lsmod
2020-06-17 12:22:12 <KatolaZ_> sorry have to go
2020-06-17 12:22:29 <guest1532> Okay, just plain `lsmod` is okay.
2020-06-17 12:28:24 <guest1532> `lsmod|grep hid` output: http://ix.io/2poS
2020-06-17 13:17:23 <guest1532> I gave up. :/
2020-06-17 16:37:21 <Piraty> how to list all files an installed package provides?
2020-06-17 16:37:32 <maxice8> apk info -L
2020-06-17 16:37:56 <Piraty> ah subcommand has more subcommand
2020-06-17 16:37:57 <Piraty> thanks
2020-06-17 16:56:31 <n0a110w> im looking to install wireguard on my rpi4
2020-06-17 16:56:59 <n0a110w> but im getting this conflict:  breaks: wireguard-rpi4-5.4.46-r0[linux-rpi4=5.4.46-r0]
2020-06-17 16:58:00 <guest1532> I'll install alpine without encryption for now. :/
2020-06-17 17:10:41 <p3lim> I don't get this patch file for podman: https://git.alpinelinux.org/aports/tree/testing/podman/libpod-config.patch
2020-06-17 17:11:19 <p3lim> Because the "cni-plugins" package installs the cni plugins to "/usr/libexec/cni-plugins", not "/usr/share/cni-plugins/bin"
2020-06-17 17:17:44 <Piraty> how to resolv reverse dependencies?
2020-06-17 17:18:07 <maxice8> resolv ?
2020-06-17 17:21:22 <TBB> +e
2020-06-17 17:26:40 <maldridge> who might I talk to about getting docker abuild updated to 3.12?
2020-06-17 17:28:54 <maldridge> I would have thought the branch would be cut by algitbot
2020-06-17 17:28:56 <maldridge> but it seems not
2020-06-17 17:30:04 <ikke> Clandmeter
2020-06-17 17:32:08 <maldridge> ikke: do you know if in the meantime that dockerfile is somewhere public?  I could just build my own version locally until a new release is made
2020-06-17 17:32:31 <maxice8> This repo i think ? https://gitlab.alpinelinux.org/alpine/docker-abuild
2020-06-17 17:33:07 <maldridge> I see
2020-06-17 17:33:43 <ikke> Making an MR can also help
2020-06-17 17:33:55 <maldridge> that would require me to first identify what needs to be changed
2020-06-17 17:33:58 <maldridge> I'll work on that now
2020-06-17 17:35:23 <ikke> https://gitlab.alpinelinux.org/alpine/docker-abuild/-/blob/master/.gitlab-ci.yml#L34
2020-06-17 17:36:15 <maxice8> .gitlab-ci.yml needs to be updated, dockerfiles.sh also need to update RELEASES
2020-06-17 17:37:24 <maldridge> I se
2020-06-17 17:37:35 <maldridge> so the list in the Makefile isn't used to generate the rest
2020-06-17 17:53:41 <p3lim> ncopa: mind taking a look at this MR? https://gitlab.alpinelinux.org/alpine/aports/-/merge_requests/9350
2020-06-17 17:54:13 <guest1532> I have a rx 5500 xt gpu and I've done setup-xorg-base with `xf86-video-amdgpu` but can't start xorg. I couldn't find something useful on the logs but I can provide them on request.
2020-06-17 17:56:16 <maldridge> ikke: alright, #62 is now open
2020-06-17 17:57:55 <Cogitri> guest1532: fwiw that works fine for me on a 5700XT I think
2020-06-17 17:58:03 <Cogitri> Haven't used Xorg for quite some time tho
2020-06-17 17:58:07 <guest1532> Here is the xorg log: http://ix.io/2pq4
2020-06-17 17:58:35 <guest1532> Cool!
2020-06-17 17:58:46 <guest1532> Are you using wayland or tty?
2020-06-17 18:00:41 <Cogitri> GNOME Wayland
2020-06-17 18:01:50 <guest1532> I fllowed this guide btw: https://wiki.alpinelinux.org/wiki/Alpine_setup_scripts#setup-xorg-base
2020-06-17 18:01:54 <guest1532> followed*
2020-06-17 18:02:27 <Cogitri> > open /dev/dri/card0: No such file or directory
2020-06-17 18:02:34 <Cogitri> Maybe you forgot to start all udev services?
2020-06-17 18:03:24 <guest1532> doesn't setup-xorg-base already do that for you?
2020-06-17 18:05:33 <Cogitri> Dunno, you can do `rc-update` and check for udev services
2020-06-17 18:05:36 <ikke> maldridge: the jobs are not picked up for some reason
2020-06-17 18:06:13 <maldridge> this sounds like a problem for -infra
2020-06-17 18:06:19 <ikke> Yup
2020-06-17 18:07:17 <guest1532> Cogitri I couldn't see any value for `udev` but `udev-postmount` is `default`, and `udev-trigeer` is `sysinit`.
2020-06-17 18:12:34 <Cogitri> You probably want to add udev-settle and udev itself tlo
2020-06-17 18:12:40 <Cogitri> s/tlo/too/
2020-06-17 18:14:02 <guest1532> udev was already started but just in case I restarted it, and I started udev-settle but still the same error.
2020-06-17 18:21:42 <guest1532> Sorry for the connection lost. Did I miss anything?
2020-06-17 18:24:58 <Hello71> I think modesetting is still recommended over amdgpu unless you really need xrender for some reason
2020-06-17 18:25:26 <Hello71> you need /dev/dri/card0 to exist for any reasonable performance
2020-06-17 18:25:39 <guest1532> What I need to do?
2020-06-17 18:25:40 <Hello71> did you remember to remove nomodeset from kernel command line?
2020-06-17 18:26:09 <guest1532> I didn't do anything. I just let setup-alpine to handle everything.
2020-06-17 18:26:22 <Hello71> you need to remove nomodeset from kernel command line to enable graphics
2020-06-17 18:26:35 <guest1532> And as the wiki guided me, installed necessary xorg components with `setup-xorg-base`.
2020-06-17 18:27:17 <guest1532> Okay, let me remove the nomodeset from the params.
2020-06-17 18:27:23 <guest1532> Then reconfigure the grub.
2020-06-17 18:27:29 <guest1532> I'll be back in a sec. ^^
2020-06-17 18:29:20 <guest1532> I guess that was it. Thank you Hello71!
2020-06-17 18:29:39 <Hello71> probably better to remove xf86-video-amdgpu unless you need it for something
2020-06-17 18:29:43 <Hello71> but also no major harm keeping it installed
2020-06-17 18:29:45 <guest1532> But, why do I have `nomodeset`?
2020-06-17 18:30:05 <Hello71> because it's set by setup-disk, because it's set on the iso
2020-06-17 18:30:05 <guest1532> What is the cons or pros of keeping or not keeping it?
2020-06-17 18:30:15 <Hello71> if you set it then you will have no graphics support :)
2020-06-17 18:30:16 <guest1532> I chose sys.
2020-06-17 18:30:27 <guest1532> lvm then sys
2020-06-17 18:30:46 <guest1532> I mean, that's what the `?` said me to do.
2020-06-17 18:31:28 <Hello71> sure but that's irrelevant
2020-06-17 18:32:27 <guest1532> Oh, I'm sorry. I thought you said `data`.
2020-06-17 18:32:31 <guest1532> meant*
2020-06-17 18:32:40 <guest1532> `disk` -> `data`
2020-06-17 18:33:10 <Hello71> uh... what?
2020-06-17 18:34:07 <guest1532> I implicitly read that as "because it is set by setup-disk data".
2020-06-17 18:34:26 <Hello71> still no idea what that means, but whatever
2020-06-17 21:33:08 <codebam> would anyone here have an idea of what Daniel means when he says it was a problem with the link loader looking up the symbols? https://lists.01.org/hyperkitty/list/connman@lists.01.org/thread/WQS76SZ2L3ZBW5SFGOAEJOXI665Y5CC5/
2020-06-17 21:33:31 <codebam> in particular how I'd fix the error if it is in fact a problem with the link loader
2020-06-18 03:36:38 <LACampbell> I thought using alpine for docker went out fashion?
2020-06-18 03:37:11 <bougyman> It did for us.
2020-06-18 03:37:25 <bougyman> We hit the dns resolver not upgrading to TCP thing and bailed.
2020-06-18 03:38:15 <LACampbell> I never heard of that issue. I'm serving a web app from alpine directly, with no problems
2020-06-18 03:41:37 <bougyman> LACampbell: you would only hit it if you had a DNS request that returned so much data that it had to upgrade from UDP to TCP
2020-06-18 03:41:40 <bougyman> it's an edge case.
2020-06-18 03:41:47 <bougyman> But for us: caused a 4 hour outage.
2020-06-18 03:41:53 <bougyman> It was _not_ easy to root cause.
2020-06-18 03:42:11 <bougyman> mostly due to the fact these containers were running in a SaaS managed service (AWS ECS)
2020-06-18 03:42:17 <bougyman> still, it was a big eye-opener.
2020-06-18 03:42:30 <bougyman> You can't really control how large a DNS response you ask for is going to be.
2020-06-18 03:46:10 <LACampbell> interesting. I just did a quick search and couldn't find an issue or people talking about it. any links?
2020-06-18 04:05:59 <maldridge> LACampbell: its an exceedingly rare edge case
2020-06-18 04:06:13 <maldridge> one that typically will only ever happen if you abuse DNS to do something it was never designed to do
2020-06-18 04:27:54 <LACampbell> maldridge:would still be keen to see a link if you have one handy
2020-06-18 04:29:07 <maldridge> I don't unfortunately
2020-06-18 04:29:37 <maldridge> the gist though is that past a cutoff size, DNS results don't fit in a single UDP datagram, and so have to be sent as a TCP stream
2020-06-18 06:19:48 <egrain> pi4 running alpine and kodi. good idea?
2020-06-18 06:46:10 <LACampbell> no, you definitely need systemd and glibc for that
2020-06-18 06:53:36 <Cogitri> Huh, we have kodi in our repos so I don't see why you'd need that
2020-06-18 06:55:23 <ikke> sounds almost like a sarcastic remark
2020-06-18 06:55:42 <afontain_> I've run it on my computer, and even on pinephone using the Alpine package
2020-06-18 06:55:50 <afontain_> I'd say it works
2020-06-18 07:03:27 <Cogitri> ikke: Ah, didn't have my morning coffee yet, so I'm not able to detect that yet :D
2020-06-18 07:03:38 <ikke> Well, I said almost
2020-06-18 07:11:27 <LACampbell> Cogitri:I am being sarcastic, but maybe I would've thought you needed that a few months ago as well
2020-06-18 07:12:23 <oneinsect> what did Cogitri do?
2020-06-18 07:14:28 <LACampbell> nothing I just made a smart ass comment and he hadn't had his coffee yet :D
2020-06-18 07:14:39 <oneinsect> lol
2020-06-18 07:16:00 <ikke> LACampbell: my sarcasm detector was inconclusive
2020-06-18 07:18:30 <egrain> shiny.
2020-06-18 07:18:39 <egrain> thanks a bunch.
2020-06-18 07:22:49 <egrain> why no blender? not that i need it. i was just curious. krita and ardour are there though, so what has blender ever done to you?
2020-06-18 07:27:37 <ikke> egrain: most like being difficult to build on musl for some reason :)
2020-06-18 07:27:47 <ikke> likely*
2020-06-18 07:28:21 <egrain> then i should be double grateful that void has it.
2020-06-18 07:28:44 <egrain> anyway, thanks again. gotta run. i'll be back later with more dumb questions.
2020-06-18 07:28:47 <ikke> If void has it, it should be doable to get it on Alpine as well
2020-06-18 07:29:03 <egrain> oh, i don't need it. i was just curious. so don't build it on my account.
2020-06-18 07:29:24 <egrain> i'm just here, becuase void doesn't support pi4 yet it seems.
2020-06-18 07:29:36 <egrain> and you guys were recommended to me over there.
2020-06-18 07:30:05 <egrain> anyway, laters.
2020-06-18 07:33:17 <codebam> oh he left
2020-06-18 07:33:22 <codebam> you can get blender using flatpak
2020-06-18 07:34:10 <mps> huh, probably also we can have 'ls' in flatpack :)
2020-06-18 07:34:27 <codebam> I use the firefox flatpak to watch or listen to things with drm (spotify, netflix)
2020-06-18 07:36:18 <codebam> mps oh yeah I just was saying because he was saying that alpine doesn't have blender in our repos, because it's difficult to build on musl
2020-06-18 07:40:14 <LACampbell> is some stuff dififcult to build because it uses non standard glibc stuff? I saw a few of them listed on the musl page
2020-06-18 07:41:07 <mps> testing/blender
2020-06-18 07:41:56 <mps> though it is disabled for now
2020-06-18 08:27:22 <EF> Hello.
2020-06-18 08:28:16 <EF> Does anyone know where the mga x driver is or has it been specifically omitted?
2020-06-18 09:01:53 <egrain> alpine isn't a rolling release if i understand that correctly. so how does that work? do i have to reinstall every time a new release comes out or i can just update stuff like i'm used to?
2020-06-18 09:03:02 <mps> egrain: apk upgrade (so simple)
2020-06-18 09:03:54 <ikke> egrain: edge is rolling release
2020-06-18 09:04:02 <ikke> but we cut off stable branches as well
2020-06-18 09:04:24 <ikke> upgrading is a matter of updating the repositories in /etc/apk/repositories to the new version and running apk upgrade --update --available
2020-06-18 09:04:39 <ikke> There might be some backwards incompattible changes that you need to account for though
2020-06-18 09:12:15 <egrain> i want this for a pi4. so updating shouldn't be a problem, right?
2020-06-18 09:12:24 <egrain> maybe for a bananapi too. don't know yet.
2020-06-18 09:12:47 <egrain> anyway, i appreciate you guys being so kind and explain everything.
2020-06-18 09:12:51 <LACampbell> ikke: why doesn't that update the version number? I guess that's just stuck on whatever one you install. /etc/os-release
2020-06-18 09:13:15 <ikke> LACampbell: It should be updated
2020-06-18 09:13:26 <ikke> unless you changed it
2020-06-18 09:13:36 <ikke> it's part of alpine-base
2020-06-18 09:14:52 <yunfan> mps: i had use fbpad to avoid the overscan problem, :D
2020-06-18 09:15:08 <yunfan> and i had ask author gave me a patch to add the feature
2020-06-18 09:15:30 <yunfan> it looks like he live in iran which caused the github repo unwritable
2020-06-18 09:16:06 <LACampbell> ikke: I'll put it down to a bit of WSL weirdness. I should try and get this thing working on a usb stick at some point
2020-06-18 09:16:14 <yunfan> how can i request a new package to alpine? i think fbpad tool is really useful for people who dont need X
2020-06-18 09:17:21 <ikke> yunfan: gitlab.a.o
2020-06-18 09:17:33 <ikke> create an issue there
2020-06-18 09:17:41 <ikke> or provide a merge request with a new APKBUILD
2020-06-18 09:21:46 <egrain> when i only run kodi, can i use its power off feature? over at void it doesn't work, because i need sudo to reboot/shut down.
2020-06-18 09:23:40 <afontain_> ^ I'm interested into the answer
2020-06-18 09:23:59 <yunfan> ikke: will try build it myself, if failed, then create the issue :D
2020-06-18 10:39:51 <LACampbell> god I hate git so much. maybe I should set up an alpine mercurial server
2020-06-18 10:41:37 <mps> yunfan: if I know what fbpad is
2020-06-18 10:41:59 <yunfan> mps: is a terminal emulator using framebuffer
2020-06-18 10:42:04 <mps> but I believe you, and will look at it
2020-06-18 10:42:17 <mps> ah, thanks
2020-06-18 10:42:43 <yunfan> mps: after the author gave me the patch, i could easily custimized the overscan in configuration file
2020-06-18 10:43:26 <yunfan> mps: i had put all of the work to my github repo  github.com/yunfan/fbpad   in branch overscan
2020-06-18 10:44:07 <mps> good, will look later, now I'm on meeting
2020-06-18 10:45:16 <yunfan> ok
2020-06-18 11:30:29 <EF> Anyone any esperience of Alpine and Matrox G200 setup on a HP Gen8 Microserver?
2020-06-18 11:44:53 <adac> Hi there! I try to start my ruby on rails application in an alpine docker but I get this error: https://pastebin.com/pk4WsQrv
2020-06-18 11:44:58 <adac> any ideas?
2020-06-18 11:45:45 <ikke> wait, what?
2020-06-18 11:46:01 <Cogitri> adac: you're probably using a glibc binary
2020-06-18 11:46:28 <Cogitri> Use a binary/library compiled against musl or compile it yourself, then it'll work
2020-06-18 11:46:31 <adac> Cogitri, ok that could well be I see
2020-06-18 12:04:12 <yunfan> EF: isnt that just x86 server?
2020-06-18 12:15:26 <Hello71> EF: doesn't the matrox driver support kms now? did you forget to remove nomodeset?
2020-06-18 12:18:47 <EF> kms = in kernel? Yes, just a x86_64 server. Will read how to do that in Alpine. Vesa and modesetting just spew errors
2020-06-18 12:37:36 <yunfan> ikke: is it possbile to compiling alpine on my arm boards?
2020-06-18 12:37:45 <yunfan> and for none arm arch
2020-06-18 12:43:47 <ikke> You would need a cross-compiler
2020-06-18 12:53:56 <yunfan> ikke: is that provided by alpine's official repo?
2020-06-18 12:54:16 <ikke> No
2020-06-18 12:54:33 <yunfan> aha
2020-06-18 12:55:06 <yunfan> so how you guys generate those images for arm? by using real board?
2020-06-18 12:55:19 <ikke> yes
2020-06-18 12:55:50 <ikke> We have ARM build servers
2020-06-18 12:59:42 <yunfan> :[
2020-06-18 13:26:47 <EF> Hello71: removing nomodeset was the key. Thanks.
2020-06-18 13:35:15 <Hello71> where is alpine-iso now?
2020-06-18 13:36:24 <Hello71> never mind, I found it
2020-06-18 13:38:50 <arinov> Hello guys, i am trying to upgrade and my busybox app download failed several times. But i can download cdn files with my Firefox. How to diag this issue?
2020-06-18 13:39:08 <arinov> ERROR: busybox-1.31.1-r19: network error (check Internet connection and firewall)
2020-06-18 15:27:04 <twosecslater> hey all, I've just got a RPi 4 and I'm trying to install Alpine onto it. I've made a 256MB FAT32 partition on the SD card with the MSDOS/MBR partition table, and extracted the aarch64 tarball to it. however, when I insert the SD card into the RPi and try booting it, it's continuously blinking the green light 7 times with the rainbow on-screen. anyone know about this?
2020-06-18 15:32:20 <twosecslater> maybe also worth noting, the microSD card is 64GB SDXC, although I kinda don't feel like that would affect anything here
2020-06-18 17:56:09 <GekkePrutser> Has anyone run into issues with migrating from 3.11 to 3.12 by changing the repos?
2020-06-18 17:56:18 <GekkePrutser> I know it's not officially supported to do that :)
2020-06-18 17:56:27 <GekkePrutser> But I don't think there is an official upgrade path right?
2020-06-18 17:57:08 <Cogitri> I don't think that it's not supported, we just currently don't have great mechanisms to detect orphaned packages and that might cause issues down the road
2020-06-18 17:57:22 <ikke> But that kind of is the way to upgrade
2020-06-18 17:57:46 <ikke> Except from orphaned packages, it should mostly work
2020-06-18 17:59:16 <GekkePrutser> "Should mostly" -> that's what I mean by not supported ;)
2020-06-18 17:59:30 <GekkePrutser> But just wanted to get some real life feedback to see if anything broke
2020-06-18 17:59:49 <ikke> GekkePrutser: To be honest, we do that all the time in Alpine itself
2020-06-18 17:59:56 <ikke> on our infra
2020-06-18 18:00:01 <GekkePrutser> Oh cool!
2020-06-18 18:00:03 <GekkePrutser> Good to know!
2020-06-18 18:01:45 <GekkePrutser> How would an orphaned package manifest itself? Failing to run with a segfault or linking error perhaps?
2020-06-18 18:01:53 <GekkePrutser> Just wondering what I should be looking for if something weird happens ;)
2020-06-18 18:01:58 <ikke> no, more like apk complaining
2020-06-18 18:02:26 <GekkePrutser> Ah ok that's not so bad
2020-06-18 18:02:44 <GekkePrutser> I was thinking more of impact to things running
2020-06-18 18:04:02 <GekkePrutser> At least with Alpine things are easy to troubleshoot because everything is set up so simple <3
2020-06-18 18:10:40 <agrisis> hello, does apk support cross compiling natively?
2020-06-18 18:10:49 <ikke> apk does not compile at all ;-)
2020-06-18 18:11:00 <agrisis> ha, true enough!
2020-06-18 18:11:21 <agrisis> well, I'm considering moving from arch to either alpine or void, void xbps supports cross compiling quite well
2020-06-18 18:11:28 <agrisis> I'm curious what alpine has to offer here?
2020-06-18 18:11:39 <ikke> To be honest, not a lot
2020-06-18 18:11:44 <agrisis> I on x86_64 host and need to build for arm and aarch64
2020-06-18 18:11:56 <agrisis> so really just chroot/qemu then?
2020-06-18 18:12:47 <ikke> Alpines own approach is to compile on native hardware, but that also means we don't put a lot of effort in supporting cross-compilation
2020-06-18 18:13:24 <agrisis> ah, that makes a lot of sense
2020-06-18 18:13:47 <agrisis> might be hard for me considering my target doesn't have a lot processing power nor RAM and I can't necessarily dedicate it to compiling
2020-06-18 18:14:05 <agrisis> thanks anyway and I love alpine on my pinephone =)
2020-06-18 18:24:35 <artok> agrisis: in case of alpine target, there can be for example llvm toolchain that builds arm and aarch64
2020-06-18 18:25:31 <artok> just have sysroot
2020-06-18 18:25:42 <artok> aaand sysroot is simple to deploy with apk
2020-06-18 18:26:40 <wsinatra> agrisis: I do a substantial amount of compilation on armv7/aarch64 for complex projects (SBCL), qemu + alpine run from ram with an LBU apkvol is the way I went.
2020-06-18 18:27:17 <wsinatra> Not exactly cross compiling, but it's functional, and if you take the extra effort the make apkbuilds for whatever you compile the installation can be made trivial
2020-06-18 18:27:58 <artok> for smaller projects, it is ok, but really hell for even mid sized stuff
2020-06-18 18:28:26 <artok> 20min vs 2hrs build time cross compile vs qemu
2020-06-18 18:29:32 <wsinatra> Last time I compiled SBCL in qemu it took 6 hours :) it's definitely not perfect
2020-06-18 18:30:12 <wsinatra> But it does work, and I'm happy to work around it because the distro brings so much to the table
2020-06-18 18:30:43 <artok> https://www.embeddedreality.com/projects/rpi/crossbuilding-with-llvm-toolchain/
2020-06-18 18:30:49 <artok> written slightly drunk
2020-06-18 18:33:00 <wsinatra> very interesting, and you've got a dockerfile in there too :)
2020-06-18 18:33:27 <wsinatra> I might have to play around with this, it seems interesting
2020-06-18 18:33:31 <agrisis> thanks! yes I'm using qemu now and it's horrible for compiling a lot of packages
2020-06-18 18:33:45 <artok> that now builds llvm from git repo
2020-06-18 18:34:04 <artok> so you'll get AVR target too! or did I enable it =D
2020-06-18 18:34:31 <artok> anyway, that is 2-stage llvm toolchain with gnu assembler added into the picture
2020-06-18 19:33:45 <gjabell> just did a fresh install of alpine and my smartcard isn't being recognized by gpg as a normal user (works with root)
2020-06-18 19:34:22 <gjabell> there are some similar issues online but there doesn't seem to be an issue with permissions, since getfacl shows my user as having rw permissions to the card
2020-06-18 19:34:30 <gjabell> anyone know if there's an extra step I'm missing?
2020-06-18 19:53:38 <c705> gjabell: do you have a udev rule in place so regular users can acces the device?
2020-06-18 19:54:52 <gjabell> atm I'm just running with the ruleset provided by the gnupg-scdaemon package, which provides a rule for the specific card I'm using (yubikey 4)
2020-06-18 19:55:36 <gjabell> if I run getfacl [path to usb device] it shows my user as having rw access to the card, but gpg --card-status still fails saying it can't find the device
2020-06-18 19:55:58 <gjabell> running it with root works, so the drivers/program setup should be ok as far as communicating with the card
2020-06-18 19:56:50 <c705> https://termbin.com/n1pz
2020-06-18 19:56:56 <c705> something like this (udev rules)
2020-06-18 19:58:06 <gjabell> in /etc/udev/rules.d, or somewhere else?
2020-06-18 19:59:02 <c705> correct
2020-06-18 19:59:10 <c705> I also had to drop this in my profile:
2020-06-18 19:59:10 <c705> export GPG_TTY="$(tty)"
2020-06-18 19:59:32 <c705> as well as:
2020-06-18 19:59:52 <c705> # send ssh connection calls thru gpg-agent (for gpg card)
2020-06-18 19:59:52 <c705> export SSH_AUTH_SOCK="$HOME/.gnupg/S.gpg-agent.ssh"
2020-06-18 19:59:53 <c705> gpg-connect-agent updatestartuptty /bye &> /dev/null
2020-06-18 20:00:10 <c705> not sure if the gpg-connect-agent is required or not though
2020-06-18 20:02:57 <gjabell> yeah unf that doesn't seem to fix it
2020-06-18 20:03:20 <gjabell> I'm *pretty sure* the rules provided with scdaemon should do it, but for some reason something is going wrong in between
2020-06-18 20:03:38 <gjabell> I also use the SSH_AUTH_SOCK, but it doesn't really matter unless I can get it to read the card afaik
2020-06-18 20:05:24 <TBB> what's your smartcard, btw?
2020-06-18 20:11:19 <gjabell> ykman says it's a yubikey 5 but lsusb says it's a yubikey 4, not sure why that's different
2020-06-18 20:11:37 <gjabell> fwiw it works fine on another of my computers and afaict there aren't any differences in the setups
2020-06-18 20:15:04 <gjabell> I guess might be worth mentioning that ykman also doesn't have permissions to the card
2020-06-18 20:15:12 <gjabell> so maybe it's not a gpg thing but something else entirely
2020-06-18 20:16:45 <gjabell> aaaand now it works...wtf
2020-06-18 20:16:58 <gjabell> I think it may have been related to a BIOS setting, although I'm not sure which
2020-06-19 05:32:20 <albr> TuxThePenguin[d]: Reposting from the other discord only channel because I'm an idiot lol:
2020-06-19 05:32:21 <albr> TuxThePenguin[d]: Is there a reason alpine doesn't package chromium for armv7? I'm trying to use alpine to make a lightweight 32 bit chromium chroot for my aarch64 system so I can use flash player and widevine
2020-06-19 05:33:28 <maxice8> Most likely fails to build on such arch
2020-06-19 05:36:13 <albr> TuxThePenguin[d]: That's weird especially since it's packaged for aarch64 and works fine on other distros on armv7
2020-06-19 06:16:17 <iggy> other musl distros?
2020-06-19 06:20:05 <maldridge> maxice8: alpine is entirely native right?
2020-06-19 06:20:08 <maldridge> y'all don't cross
2020-06-19 06:20:20 <maxice8> Only on initial bootstrap i think
2020-06-19 06:20:34 <maldridge> kk, looking at trying to compile some godforsaken qt stuff
2020-06-19 06:20:38 <maldridge> qmake does not cross well
2020-06-19 06:20:44 <maxice8> I'm not versed in the toolchain
2020-06-19 06:21:34 <maxice8> qmake still exists unfortunately
2020-06-19 06:22:31 <maldridge> indeed
2020-06-19 06:22:55 <maldridge> qlcplus combines the inherent clarity of qmake with the simplicity of various gnu extensions
2020-06-19 06:23:13 <Cogitri> At least they didn't switch to QBS :D
2020-06-19 06:23:48 <maldridge> point
2020-06-19 06:31:45 <albr> TuxThePenguin[d]: Hmm I had a better look and it seems like armv7 used to have chromium but the last version packaged was 77
2020-06-19 06:32:53 <Cogitri> We had to disable it because it failed to build and there were some really bad security vulnerabilities that were fixed by the upgrade
2020-06-19 07:22:28 <yunfan> how can i boot the iso file from usb ?
2020-06-19 07:23:13 <albr> breadtard[d]: Flash it to said usb
2020-06-19 07:23:45 <albr> breadtard[d]: On windows use rufus, on mac etcher, on another linux distro dd
2020-06-19 07:25:10 <yunfan> ok i just found that
2020-06-19 07:25:14 <ikke> Note that the iso uses isofs, so the usb would be read-only
2020-06-19 07:27:39 <yunfan> want to use my z8700 box to compile things
2020-06-19 07:33:17 <yunfan> ikke: it works, but since it use the whole disk and read-only, it cant satified me, is it possible to use only 1 partitions on the usb disk?
2020-06-19 07:33:31 <yunfan> in that case i could use the left partition to store data and files
2020-06-19 07:34:51 <ikke> yunfan: not with if you `dd` the iso
2020-06-19 07:35:19 <ikke> setup-bootable in alpine can make a writable usb disk
2020-06-19 07:35:32 <ikke> otherwise, it's fairly simple to do it manually as well
2020-06-19 07:35:45 <yunfan> ikke: can i reuse the same usb disk when using setup-bootable?
2020-06-19 07:36:39 <ikke> I think you need to stop the modloop service, but after that, should be possible
2020-06-19 07:36:54 <yunfan> ikke: thanks wil try it
2020-06-19 08:21:02 <yunfan> the target box's wifi cannot be used by iwconfig ...
2020-06-19 08:21:44 <yunfan> when executing ifconfig wlan0 up, i got ioctl 0x8914 failed: Operation not permitted
2020-06-19 08:22:29 <albr> breadtard[d]: sudo
2020-06-19 08:23:46 <yunfan> its root user which i use to executing the command
2020-06-19 08:53:21 <nacelle> yunfan: might need firmware
2020-06-19 09:35:30 <yunfan> nacelle: i just gave up :D
2020-06-19 09:44:32 <Skinny> Good day everyone! I'm creating a sort of automated install script for alpine+my app based on the default setup-alpine script. After completing this script I'd like to know the used root partition/device (e.g /dev/sda2) but cannot seem to find a good solution. Does anyone has an idea or tip ?
2020-06-19 09:45:02 <Skinny> Because the setup script unmounts everything when it's done it's hard to find the right device
2020-06-19 12:37:29 <Jenkler> anyone uses alpine in azure?
2020-06-19 12:40:06 <artok> I think many people
2020-06-19 12:41:03 <Jenkler> artok, but its a lot of tweeks. It does not work with .iso files
2020-06-19 12:41:21 <Jenkler> https://github.com/tomconte/packer-alpine-azure
2020-06-19 12:41:39 <Jenkler> must be a better way doing this
2020-06-19 12:44:25 <artok> not working with virtual iso?
2020-06-19 12:48:14 <Jenkler> artok, no, azure dont support iso installation
2020-06-19 12:48:27 <Jenkler> Windows is kinda broken as usual ;)
2020-06-19 12:52:28 <artok> oh yeah they didn't have alpine on there
2020-06-19 12:53:43 <artok> my bad, memory serves wrongly
2020-06-19 12:53:50 <Jenkler> NP ;)
2020-06-19 14:45:20 <oneinsect> how do i simulate a pkg without installing it?
2020-06-19 14:45:50 <oneinsect> i still would want it to populate the /etc/apk/world and /lib/apk/db/installed files
2020-06-19 14:47:01 <oneinsect> ikke: can you throw some light on that
2020-06-19 14:56:16 <ikke> maybe apk add -t foo alpine-base?
2020-06-19 14:56:36 <oneinsect> dont want to create a virtual package
2020-06-19 14:56:42 <oneinsect> assuming i already have an apk
2020-06-19 14:56:45 <oneinsect> physical apk
2020-06-19 14:57:29 <ikke> I don't think you can
2020-06-19 14:58:02 <oneinsect> aaah i wish that feature was there
2020-06-19 14:58:04 <oneinsect> never mind
2020-06-19 16:49:51 <vejetaryenvampir> Anyone able to run MultiMC?
2020-06-19 16:50:09 <vejetaryenvampir> On the amd64 version.
2020-06-19 16:50:23 <c705> just ask your question please
2020-06-19 16:50:35 <c705> oh i misread, disregard
2020-06-19 16:51:14 <vejetaryenvampir> c705: no worries. ^^
2020-06-19 16:52:08 <vejetaryenvampir> I've been trying to install the dependencies that show in the build of the musl version void linux but just cound't get it working.
2020-06-19 16:52:10 <vejetaryenvampir> couldn't*
2020-06-19 16:52:38 <c705> I've bee  half looking for a way to run mc again, mAybe I'll lok into this later
2020-06-19 16:53:08 <vejetaryenvampir> My problem is specific to MultiMC; to be honest.
2020-06-19 16:53:16 <vejetaryenvampir> I didn't even tried the vanilla launcher.
2020-06-19 16:53:30 <c705> the vanilla won't work without glibc
2020-06-19 16:53:34 <c705> https://aur.archlinux.org/packages/multimc-git/
2020-06-19 16:53:46 <c705> It's arch linux, but you might get sme build/debug info here
2020-06-19 16:53:47 <vejetaryenvampir> c705: you sure?
2020-06-19 16:54:16 <c705> I'm not sure, but it's an educated guess. The bin that minecraft ships won't run and I suspect it's because it's built with glibc
2020-06-19 16:54:41 <vejetaryenvampir> I think the launcher also written in java.
2020-06-19 16:55:10 <c705> nah, that launcher is fine
2020-06-19 16:55:30 <c705> a year or 2 ago, mc stopped shipping the java launcher and started shipping some c# or c++ bullshit launcher
2020-06-19 16:55:51 <vejetaryenvampir> microsoft...
2020-06-19 16:55:58 <c705> the backend is still java I believe, but I never really gave enough of a shit to really look into it
2020-06-19 16:56:32 <vejetaryenvampir> I think this one only depends on java: https://aur.archlinux.org/packages/minecraft/
2020-06-19 16:58:22 <c705> vejetaryenvampir: I still think it's expecting a glibc runtime. I've tried extracting the bin out of the deb and that's what failed to run
2020-06-19 16:58:27 <c705> I could easily be wrong however
2020-06-19 16:58:44 <vejetaryenvampir> trying ^^
2020-06-19 16:59:46 <vejetaryenvampir> It just says that the file I'm trying to run doesn't exist.
2020-06-19 16:59:59 <c705> yeah, that's what i get
2020-06-19 17:00:04 <vejetaryenvampir> As far as I understand, it means some library is missing.
2020-06-19 17:00:07 <c705> My educated guess is you're missing glibc
2020-06-19 17:00:28 <c705> Minecraft would have to ship a musl build for us, and obviously that's not gonna happen
2020-06-19 17:00:52 <c705> You can pack in glibc + whatever other crap in a chroot and run it like that, but that's pretty much when I ran out of fucks
2020-06-19 17:01:32 <vejetaryenvampir> Makes sense.
2020-06-19 17:01:56 <vejetaryenvampir> What about multimc?
2020-06-19 17:02:04 <vejetaryenvampir> It is c++.
2020-06-19 17:03:10 <vejetaryenvampir> brb
2020-06-19 17:03:12 <c705> I have no idea. The code is open, so at least you can fork it and patch what you need to to get a musl build going
2020-06-19 17:20:38 <wej> Hi! I've been trying to use domoticz on alpine and starting it as a system daemon fails unfortunately. As it turned out, the /etc/init.d/domoticz init script does some weird stuff. Although it tells domoticz via arguments to daemonize itself and write a pid file, it is also configured with command_background=YES, which causes start-stop-daemon to write its own pid file, which in turn causes
2020-06-19 17:20:39 <wej> domoticz to crash because it is unable to overwrite that file due to permissions problems. Removing both the -daemon and -pidfile flags from command_args makes it run just fine. I would prefer not to fiddle with the init script though. Any chance for a fix for that?
2020-06-19 17:22:24 <c705> wej: did you install from a package?
2020-06-19 17:22:32 <wej> c705: yes
2020-06-19 17:22:56 <c705> wej: you can file a bug on the package:
2020-06-19 17:22:56 <c705> https://gitlab.alpinelinux.org/groups/alpine/-/issues
2020-06-19 17:22:59 <wej> domoticz-4.10717-r1
2020-06-19 17:23:07 <wej> ah
2020-06-19 17:24:11 <wej> what's also strange is, that also the package says version 4.10717 when running it, it says 4.9700
2020-06-19 17:24:18 <wej> not sure what's the deal there
2020-06-19 17:25:03 <c705> wej: maintainer might have forgot to bump the version in the APKBUILD. not sure
2020-06-19 17:25:28 <Cogitri> c705: No, if the pkgver wasn't bumped then it wouldn't say that version number during install
2020-06-19 17:25:44 <Cogitri> Probably upstream that forgot to increase the version number so the program itself still prints the old version
2020-06-19 17:26:12 <Cogitri> That initd file thing definitely is a bug though, a report (or even better a merge request :) would be appreciated
2020-06-19 17:26:17 <c705> Cogitri: from what i'm reading though, prog version > package version
2020-06-19 17:26:32 <c705> anyways yes, a bug report would be lovely
2020-06-19 17:57:33 <jhass> Is anyone archiving old packages?
2020-06-19 17:57:58 <jhass> like the ALA https://wiki.archlinux.org/index.php/Arch_Linux_Archive
2020-06-19 17:58:29 <ikke> jhass: no, we want to do something like that, but requires a lot of storage space
2020-06-19 17:58:40 <jhass> gotcha
2020-06-19 18:03:01 <c705> hi, i have problems with adding grub on the post rigger
2020-06-19 18:03:02 <c705> https://termbin.com/c20q
2020-06-19 18:03:25 <c705> I'll file a bug in a sec if it's not something silly I'm doing
2020-06-19 18:21:16 <wej> i've created a bug report for that problem
2020-06-19 18:21:58 <c705> Cool
2020-06-19 18:22:03 <c705> tracking no?
2020-06-19 18:22:14 <wej> https://gitlab.alpinelinux.org/alpine/aports/-/issues/11671
2020-06-19 18:22:17 <c705> cheers
2020-06-19 18:22:41 <c705> Oh what, no?
2020-06-19 18:22:49 <c705> This is for domoticz
2020-06-19 18:22:58 <c705> Oh, you meant your problem, not mine..lol
2020-06-19 18:23:09 <wej> oh, sorry
2020-06-19 18:23:12 <c705> np
2020-06-19 18:23:39 <c705> you can also submit a mr if you're comfortable with that
2020-06-19 18:26:52 <oneinsect> one quick advise, say i have to build libxcrypt and libxcrypt-compat, can i use the same APKBUILD and generate them
2020-06-19 18:27:30 <oneinsect> compatible version has slightly different configure options
2020-06-19 18:27:49 <oneinsect> do they need to be different packages??
2020-06-19 19:12:24 <wej> c705: i've submitted a mr for that domoticz bug :)
2020-06-19 19:13:00 <c705> cool
2020-06-19 20:05:36 <twb> How do I check what version of alpine I'm running?
2020-06-19 20:05:46 <twb> I don't have lsb_release installed and it's not mentioned in /etc/motd
2020-06-19 20:05:54 <twb> Derp.  /etc/alpine-release
2020-06-19 21:45:41 <codebam> just curious, but why the r1 release on linux-lts (edge)?
2020-06-19 21:46:06 <codebam> that means a rebuild right?
2020-06-19 21:48:07 <gtkspert> Weird question, for a package on aports to be rebuilt and published, do you need to bump the version / release?
2020-06-19 21:48:48 <mps> pkgrel field
2020-06-19 21:49:06 <mps> man APKBUILD
2020-06-19 21:57:30 <gtkspert> Yep. Already read the manual. In fact, someone specifically didn't bump the pkgrel. I wanted to confirm if the bug I see is valid.
2020-06-19 22:01:11 <mps> if you upgrade pkgver pkg will be rebuilt, and if you do not upgrade pkgver and want/need package rebuild then you have 'bump' pkgrel
2020-06-19 22:04:32 <gtkspert> Ok, so this change wouldn't cause a rebuild then?
2020-06-19 22:04:33 <gtkspert> https://gitlab.alpinelinux.org/alpine/aports/-/commit/20666afae7f741f12b7fc32c745fd03e40f815ba
2020-06-19 22:07:20 <mps> it depends, what is MR number?
2020-06-19 22:08:20 <gtkspert> Not sure. I could only find the commit id. Sidenote, is aports built by gitlab?
2020-06-19 22:08:25 <mps> !9426
2020-06-19 22:08:43 <mps> algitbot: ^
2020-06-19 22:08:47 <mps> :)
2020-06-19 22:10:07 <mps> no, aports are built by builders (separate machines)
2020-06-19 22:10:38 <gtkspert> Sorry, I meant using gitlab's ci functionality.. I see there's travis and drone files in the root.
2020-06-19 22:12:31 <mps> CIs are also separate machines (actually qemu VM) but they are triggered by merge requests create/change
2020-06-19 22:13:16 <gtkspert> Ok, cool. I couldn't find the run for the commit before when I searched. I thought perhaps I was looking in the wrong tool (travis, etc.)
2020-06-19 22:16:01 <gtkspert> Raised a merge request to bump pkgrel anyway. Why not right? Worst case, it's still filtering through CDNs. Best case it gets resolved quicker.
2020-06-19 22:16:13 <mps> heh, although I use from day one and every day I have no clear understanding how it works :)
2020-06-19 22:16:45 <mps> I user it* (gitlab.a.o)
2020-06-19 22:17:22 <gtkspert> Sounds like the definition of a good system! Stays out of your way, and Just Works (tm)
2020-06-19 22:18:32 <mps> praise for our infra team :)
2020-06-19 22:53:24 <wrycode> Which image should I use for a KVM VPS on Ramnode?
2020-06-19 22:54:35 <wrycode> I'm guessing the virtual x86_64?
2020-06-19 22:55:34 <wrycode> and can I set up security (ssh) by modifying the image locally before uploading it? Or what do people usually do?
2020-06-19 22:58:34 <Cogitri> I usually boot a sysrescue CD image on my hosting provider and set up the system from there
2020-06-19 22:59:04 <Cogitri> If you don't want to do that, maybe you can just VNC into the vps?
2020-06-19 23:02:20 <wrycode> hmm well the vps doesn't exist yet. I have to select an image as a boot source. They don't have sysrescue as an option, but could I choose some other generic linux and then install Alpine from that?
2020-06-19 23:04:09 <Cogitri> Yup, any linux will do
2020-06-19 23:04:51 <Cogitri> I usually follow the wiki entry here: https://wiki.alpinelinux.org/wiki/Alpine_Linux_in_a_chroot in some live medium
2020-06-19 23:05:57 <wrycode> Cogitri: ah that's what I was looking for. Thanks.
2020-06-19 23:17:56 <wrycode> god this is complicated
2020-06-19 23:55:55 <HyperTables> hello. I'm trying to set up `postgresql` ... so I installed it along with `openrc` but when I do `/etc/init.d/postgresql start` I am told it is already starting. `ps -A` does not show it. where does debug output go? it must be failing on some missing config
2020-06-19 23:59:55 <HyperTables> ok so I'm learning that one does not normally start openrc in a docker container
2020-06-20 00:03:28 <artok> yeah, docker is it's own thing
2020-06-20 00:16:15 <HyperTables> `performing post-bootstrap initialization ... sh: locale: not found` how can I fix this? (running in docker btw)
2020-06-20 00:16:24 <HyperTables> postgres seems to want "locales"
2020-06-20 00:21:15 <HyperTables> are locales no longer a thing I should worry about on alpine?
2020-06-20 01:51:27 <LACampbell> I thought the idiomatic docker thing was to spin up another container for your DB?
2020-06-20 01:51:57 <LACampbell> isn't it kind of missing the point if you're ssh'ing in and installing stuff?
2020-06-20 02:11:05 <HyperTables> well partitioning a complex system can be part of the point of docker but it also has portability benefits of containers even if you keep them monolithic
2020-06-20 03:17:34 <c705> hi. what's the default network tool on edge builds. I know /etc/init.d/networking, but where's the source included?
2020-06-20 04:36:10 <LACampbell> I have messed up my repos badly. can't seem to change them either
2020-06-20 04:37:22 <LACampbell> wget: bad address 'mirrors.alpinelinux.org' --- that's no good
2020-06-20 04:40:15 <LACampbell> editing by hand fixes it. apparently setup-apkrepos just appends stuff to the end of the ifle
2020-06-20 04:43:05 <LACampbell> https://gist.github.com/LewisAndrewCampbell/58e2b189cd551d55dd8f63aa3d7f7bcc ahh I've noobed this up big time.
2020-06-20 04:50:38 <LACampbell> tried deleting the cache, no luck
2020-06-20 04:53:28 <c705> still waiting for a question
2020-06-20 04:54:25 <LACampbell> ok. how download curl? me know no
2020-06-20 04:58:57 <c705> apk add curl
2020-06-20 05:00:39 <LACampbell> no work!
2020-06-20 06:13:52 <afontain_> apk add cmd:curl ?
2020-06-20 06:28:55 <LACampbell> turns out it was a firewall issue. which I probably should have guessed as I was noobing my way through my providers firewall before
2020-06-20 07:39:14 <mps> anyone/anywhere use mISDN linux drivers nowadays? thinking to disable all them in kernel
2020-06-20 08:36:46 <Jenkler> issue with upgrade?
2020-06-20 08:36:49 <Jenkler> conflicts: libffi-3.2.1-r6
2020-06-20 08:36:56 <Jenkler> from 3.11 to 3.12?
2020-06-20 08:40:12 <maxice8> You probably has a package that was dropped in 3.12 and thus depends on libffi-3.2.1 instead of libffi 3.3
2020-06-20 08:40:23 <maxice8> Please send the full output in a paste
2020-06-20 08:40:28 <Jenkler> seams to be python
2020-06-20 08:40:36 <ikke> yes
2020-06-20 08:40:56 <ikke> you need to change python to either python2 or python3
2020-06-20 08:41:23 <Jenkler> ikke, are we going to remove python2 soon?
2020-06-20 08:41:37 <ikke> Jenkler: yes
2020-06-20 08:41:43 <maxice8> We have been trying it for a while now
2020-06-20 08:41:51 <Jenkler> wonder if python2 is smaller and faster then python 3
2020-06-20 08:41:57 <ikke> The target was 3.12
2020-06-20 08:42:03 <maxice8> Can't remove it without causing collateral damage
2020-06-20 08:42:05 <Jenkler> ikke, ok
2020-06-20 08:42:12 <ikke> so now we aim for 3.13
2020-06-20 08:42:18 <ikke> (/4.0)
2020-06-20 08:42:24 <Jenkler> ahh
2020-06-20 08:43:53 <iggy> yeah, I just `apk del python` and then went on my way
2020-06-20 08:44:26 <Jenkler> i see "py-setuptools" depend on python3 now
2020-06-20 08:44:50 <maxice8> py3-setuptools provides= and replaces= py-setuptools like all packages that were made py3-only
2020-06-20 08:44:51 <ikke> best to check /etc/apk/world and change py-* to py3-*
2020-06-20 08:46:56 <Jenkler> found out that we had lego in repo ;) I have been using docker certbot.
2020-06-20 08:47:10 <Jenkler> Do we have WALinuxAgent in the repo?
2020-06-20 08:48:53 <ikke> I don't think so
2020-06-20 08:53:36 <Jenkler> cat find it anyway so, yes the answer is no ;)
2020-06-20 08:55:20 <ikke> feel free to provide an MR with an APKBUILD :), (or a package request on gitlab.a.o, but then someone needs to care enough to provide it)
2020-06-20 08:55:21 <Jenkler> is --available needed to clean all old pkg
2020-06-20 08:55:47 <ikke> It does not remove packages that are not present anymore
2020-06-20 08:56:17 <Jenkler> so, for a maximum cleanup apk -U upgrate is enought ?
2020-06-20 08:56:36 <Jenkler> or is the some more cleaning that i can do?
2020-06-20 08:57:09 <ikke> if you do an upgrade to a new major version, passing --available is recommended
2020-06-20 08:57:21 <ikke> but you should manually verify /etc/apk/world as well
2020-06-20 08:58:26 <Jenkler> if i issed the --available option, is it possible to run the command again to get a better resault or does this only work once
2020-06-20 08:58:57 <afontain_> shouldn't do anything more on second run
2020-06-20 08:58:58 <ikke> No, you can run it again
2020-06-20 08:59:14 <afontain_> ah
2020-06-20 08:59:16 <ikke> BUt yes, it doesn't do more if you run it twice
2020-06-20 08:59:36 <Jenkler> nice, yes indeed. I got a few more updates ;)
2020-06-20 08:59:40 <Jenkler> thaks
2020-06-20 09:01:29 <Jenkler> why is not --available default? change to break stuff?
2020-06-20 09:01:48 <ikke> Yes, it ignores any dependency constrains you have set
2020-06-20 09:05:28 <Jenkler> when you say "you have set", do you mean specialconfing. because I use plain alpine nothing speciall att all.
2020-06-20 09:06:03 <ikke> Jenkler: you can do something like apk add 'pkg<1.5'
2020-06-20 09:06:22 <ikke> or apk add pkg=1.5
2020-06-20 09:06:25 <ikke> to pin versions
2020-06-20 09:06:54 <Jenkler> ikke, ok but if i dont do that --available is safe
2020-06-20 09:07:42 <ikke> yes
2020-06-20 09:09:34 <Jenkler> nice, at the moment I am replacing coreos with alpine in azure.
2020-06-20 10:39:28 <PhoenixMage> Anyone ever had to deal with bazel when trying to compile a package, it is truly the worst
2020-06-20 10:46:48 <Cogitri> Fortunately no, but the "How to make package maintainers cry" talk made me avoid Bazel where I could :)
2020-06-20 11:32:03 <winny> i should watch that talk
2020-06-20 11:34:04 <winny> added to watchlist, thanks
2020-06-20 13:11:54 <Jenkler> after update to 3.12 I cant connect to KVM from virsh eller Virtual Manager
2020-06-20 13:14:30 <Jenkler> Also stoppning hangs
2020-06-20 13:14:35 <Jenkler> service virtlogd stop
2020-06-20 13:14:35 <Jenkler>  * Stopping libvirtd ...
2020-06-20 13:15:42 <Cogitri> #11602 ?
2020-06-20 13:15:56 <Cogitri> Err, apparently algitbot doesn't work here
2020-06-20 13:15:58 <Cogitri> https://gitlab.alpinelinux.org/alpine/aports/-/issues/11602
2020-06-20 13:24:21 <Jenkler> Cannot read CA certificate '/etc/pki/CA/cacert.pem': No such file or directory
2020-06-20 13:24:45 <Jenkler> if i start it in the fg. (/usr/sbin/libvirtd  --listen)
2020-06-20 13:25:25 <Jenkler> i have ca-certificates installed, do I need to generate this?
2020-06-20 13:29:43 <Jenkler> Yes, if killing " {daemon-init} /usr/sbin/libvirtd -d" it seams to work
2020-06-20 13:36:01 <Jenkler> but it still broken, so i guess we need 6.4
2020-06-20 13:37:42 <Cogitri> Maybe you can comment on thay issue then
2020-06-20 13:39:01 <Jenkler> Yes, I think backport would be best
2020-06-20 13:53:19 <Jenkler> Cogitri, comment done ;)
2020-06-20 13:54:30 <Jenkler> Cogitri, how has the auth to backport pkg. It it the maintainers or only ncopa?
2020-06-20 14:01:29 <maxice8> Anyone that can push to that branch i think
2020-06-20 14:01:42 <Cogitri> Yup, anyone can backport
2020-06-20 14:01:56 <Cogitri> We'd have quite the high bus factor if only one member was able to to do
2020-06-20 14:01:59 <Cogitri> to do that*
2020-06-20 14:10:34 <Jenkler> hehe thats true
2020-06-20 14:13:30 <arash> hey guys
2020-06-20 14:14:19 <arash> How can i install a package from aports non-free?
2020-06-20 14:14:36 <maxice8> build them with abuild and then add apk the .apk file
2020-06-20 14:16:24 <arash> tnx mate
2020-06-20 14:34:33 <Jenkler> there is a lot of virt* service files. What services need to be started?
2020-06-20 14:36:32 <egrain> is there a wiki showing how to put the aarm64 thing on the card for the pi4?
2020-06-20 14:37:09 <afontain_> Alpine wiki > raspberry pi?
2020-06-20 14:39:22 <egrain> oh, on the right.
2020-06-20 14:39:31 <egrain> my bad.
2020-06-20 14:39:33 <egrain> thanks.
2020-06-20 14:43:03 <Cogitri> Jenkler : The ones you need presumably, probably libvirt and the one with qemu in its name
2020-06-20 14:58:17 <HyperTables> what's a good minimal httpd for alpine? it doesn't ship with busybox httpd for some reason
2020-06-20 14:58:32 <Jenkler> nodejs ?
2020-06-20 14:58:45 <Jenkler> or php
2020-06-20 14:59:01 <Jenkler> niginx is kinda small
2020-06-20 14:59:14 <HyperTables> no I just need to serve a 1KB file called a PAC file for configuring chrome with proxy settings
2020-06-20 14:59:21 <afontain_> what?
2020-06-20 14:59:33 <afontain_> darkhttpd is minimalist
2020-06-20 14:59:53 <Jenkler> or use golang to fix a server
2020-06-20 15:00:08 <Jenkler> not much code for that!
2020-06-20 15:00:26 <Cogitri> Or the py3 http server if you have py3 installed already
2020-06-20 15:00:27 <HyperTables> ok will give darkhttpd a shot
2020-06-20 15:00:33 <afontain_> you likely have python3, if your definition of minimal is "something I have already"
2020-06-20 15:00:50 <HyperTables> is this python3 server something I can start from the cli without writing a .py?
2020-06-20 15:00:51 <afontain_> `python3 -m htttp.server` IIRC
2020-06-20 15:01:05 <fabled> HyperTables, busybox httpd comes in busybox-extras package
2020-06-20 15:22:41 <rnkn> hello, recently some man pages have not been rendering properly for me, giving me something like this for udisksctl: https://termbin.com/6l6b
2020-06-20 15:23:12 <rnkn> I have mandoc apk installed; is that the right apk for rending man pages?
2020-06-20 15:24:58 <Hello71> seems suspiciously like you have locale or TERM set incorrectly
2020-06-20 15:25:20 <afontain_> I think I can relate
2020-06-20 15:25:46 <rnkn> TERM=dumb
2020-06-20 15:26:23 <rnkn> scratch that, TERM=linux in console, but dumb within emacs
2020-06-20 15:26:58 <Hello71> oh, actually hold on
2020-06-20 15:27:07 <afontain_> TERM=alacritty in my case, but /usr/share/terminfo/ doesn't contain it
2020-06-20 15:27:16 <Hello71> so if you actually have ^H or 0008 displaying then that's a problem with TERM
2020-06-20 15:27:33 <afontain_> hmm
2020-06-20 15:27:37 <rnkn> LANG=en_AU.UTF-8
2020-06-20 15:28:07 <mps> afontain_: /etc/terminfo/a/alacritty
2020-06-20 15:28:26 <afontain_> I never see ^H or such, just man showing .SH and stuff
2020-06-20 15:28:34 <afontain_> mps, ah, it's there indeed
2020-06-20 15:29:14 <afontain_> for some reason, I've only got urxvt in share (/usr/share/terminfo/r/rxvt-unicode*)
2020-06-20 15:29:26 <mps> maxice8 and I moved most used terminfo in /etc/terminfo, ncurses-base-terminfo pkg
2020-06-20 15:29:54 <rnkn> I'm not using a terminal emulator, I'm on the console (i.e. no X)
2020-06-20 15:29:54 <mps> urxvt have its own terminfo
2020-06-20 15:30:53 <mps> console mostly use linux terminfo defs
2020-06-20 15:30:56 <Hello71> I think this man page is generated wrong. maybe caused by ash instead of bash
2020-06-20 15:31:43 <rnkn> this is in ash and zsh
2020-06-20 15:32:20 <Hello71> supposed to look like this: https://bpa.st/TUIA (from Debian)
2020-06-20 15:32:51 <mps> rnkn: do you have ncurses-base-terminfo installed
2020-06-20 15:33:04 <Hello71> aw man it has so many makedeps
2020-06-20 15:33:16 <Hello71> mps: did you try it yourself? the man page is broken
2020-06-20 15:33:29 <mps> which one
2020-06-20 15:34:06 <Hello71> well rnkn said udisksctl
2020-06-20 15:34:15 <Hello71> it has all these spaces at the beginning of lines in the man page
2020-06-20 15:34:34 <Hello71> and I guess man macros have to start at the beginning of the line
2020-06-20 15:34:41 <mps> aha, in which pkg is it
2020-06-20 15:34:49 <mps> udisks2?
2020-06-20 15:34:56 <rnkn> mps: yes ncurses-terminfo-base is installed
2020-06-20 15:35:22 <rnkn> udisks2-doc
2020-06-20 15:35:27 <mps> rnkn: why is is your TERM is dumb then
2020-06-20 15:35:32 <rnkn> but that's just an example, there have been plenty
2020-06-20 15:35:59 <rnkn> mps: TERM is only dumb within emacs, in console it says TERM=linux
2020-06-20 15:36:18 <Hello71> I think it is caused by interaction between docbook-xsl and ash
2020-06-20 15:36:28 <Hello71> hm, maybe i also need to rebuild docbook itself with bash
2020-06-20 15:37:10 <rnkn> but then the terminal emulator emulator says TERM=eterm-color and it has the same man rendering problems... :(
2020-06-20 15:37:30 <mps> yes, man page is broken
2020-06-20 15:37:37 <Hello71> no time now though, maybe later
2020-06-20 15:37:57 <Hello71> if nobody finds it then file an issue and ping me
2020-06-20 15:38:28 <mps> all man page in this pkg, or most of them are broken
2020-06-20 15:39:02 <mps> something is wrong with pkg
2020-06-20 15:39:45 <rnkn> mps: you have the same issue?
2020-06-20 15:40:24 <mps> yes
2020-06-20 15:41:07 <rnkn> this must have happened only recently, because about 2/3rds of my man pages are okay
2020-06-20 15:41:41 <mps> man pages from other pkgs are ok
2020-06-20 15:42:00 <rnkn> I've got the same issue with dbus man pages
2020-06-20 15:44:49 <rnkn> okay looks like it's just udisks2-doc and dbus-doc
2020-06-20 15:45:11 <mps> I don't install much -doc apks, so can't check
2020-06-20 15:45:36 <rnkn> should I file a bug?
2020-06-20 15:45:47 <afontain_> I know there's others
2020-06-20 15:45:52 <afontain_> I've found others
2020-06-20 15:46:33 <afontain_> man gnome-keyring
2020-06-20 15:46:47 <afontain_> here's at least one more
2020-06-20 15:48:39 <rnkn> so it's a known thing? no need to notify anyone?
2020-06-20 15:49:29 <mps> rnkn: now it is known :)
2020-06-20 15:49:48 <afontain_> gnome-books, gnome-disks, gnome-software, gnome-shell, but not gnome-latex or gnome-calculator
2020-06-20 15:49:51 <mps> and creating bug report is good idea
2020-06-20 15:58:50 <rnkn> issue created
2020-06-20 15:58:56 <rnkn> thanks!
2020-06-20 17:11:35 <Hello71> created... where?
2020-06-20 17:12:23 <ikke> rnkn already left
2020-06-20 18:16:27 <oneinsect> sometimes abuild -r waits along time saying Analyzing dependencies...
2020-06-20 18:16:38 <oneinsect> even -v doesnt throw much useful info
2020-06-20 18:16:50 <oneinsect> any idea why?
2020-06-20 18:17:03 <oneinsect> a long time*
2020-06-20 18:20:46 <oneinsect> naa found the reason
2020-06-20 18:20:59 <oneinsect> requires i mount sys dev proc etc in alpine chroot
2020-06-21 02:26:37 <p3lim> running a couple custom services with openrc, how do I see the logs from them?
2020-06-21 02:26:53 <L1Cafe> Did you write the unit files yourself?
2020-06-21 02:26:58 <p3lim> yes
2020-06-21 02:27:14 <L1Cafe> Then you either look in the typical place for the specific binary or...
2020-06-21 02:27:17 <p3lim> copied and customized from other existing services from aports
2020-06-21 02:27:34 <L1Cafe> I don't think there's an "or" actually. OpenRC doesn't manage logs
2020-06-21 02:27:55 <p3lim> so anything stdout/stderr is voided?
2020-06-21 02:28:02 <L1Cafe> They should be usually in /var/log/<name>.log, or perhaps in the folder youre running them from
2020-06-21 02:28:15 <p3lim> yeah nothing there, first place I looked
2020-06-21 02:28:28 <p3lim> nothing in messages neither, other than rc starting/stopping the services
2020-06-21 02:28:45 <L1Cafe> Hmmm... I don't know what to tell you, I'll shut up so someone more knowledgeable than me can handle this :D
2020-06-21 02:30:08 <p3lim> also wondering if openrc has some sort of drop-in like systemd
2020-06-21 02:30:16 <p3lim> to extend/override init.d files
2020-06-21 02:30:35 <p3lim> without modifying the original files so that they can get updated by apk
2020-06-21 05:50:34 <mps> p3lim: no, but you can install runitd and create run files, which are shell scripts. also same can be done with s6 tools but I never used the,
2020-06-21 06:07:38 <rnkn> hey I'm pretty sure this is a Rpi problem unrelated to Alpine, but when I plug in both a USB 3 stick and USB 3 drive into the 2x USB 3.0 ports on my Rpi4 everything unmounts (and /home is on the USB 3 stick!)
2020-06-21 06:08:08 <rnkn> I'm thinking this is hardware, but hoping maybe there's some config limit to the USB 3.0 power somewhere..?
2020-06-21 06:10:18 <wrycode> does the alpine docker container run in diskless, data, or sys mode?
2020-06-21 06:12:16 <wrycode> diskless seems like it'd be redundant but idk
2020-06-21 06:15:15 <mps> rnkn: did you looked in dmesg output
2020-06-21 06:17:45 <rnkn> mps: hang on, I will plug in, which means I will likely lose /home and thus be disconnected...
2020-06-21 06:19:30 <rnkn> I get: [sdb] Spinning up disk... (click.. click... click....)
2020-06-21 06:20:58 <mps> looks like not enough power from power supply
2020-06-21 06:21:41 <rnkn> yeah, sad :(
2020-06-21 06:22:09 <rnkn> how do I watch dmesg, like tail?
2020-06-21 06:22:27 <rnkn> (rather than calling dmesg and printing the whole thing each time)
2020-06-21 06:23:38 <mps> '-w'
2020-06-21 06:24:30 <rnkn> mps: thanks!
2020-06-21 06:26:39 <mps> this is for dmesg from util-linux, busybox dmesg don't have follow option
2020-06-21 06:27:35 <rnkn> I have util-linux
2020-06-21 06:27:46 <rnkn> I wish util-linux was split up more
2020-06-21 06:28:10 <rnkn> the culprit seems to be this Anne Pro 2 keyboard
2020-06-21 06:29:06 <oneinsect> why do you want util-linux to be split more?
2020-06-21 06:29:17 <oneinsect> and what did anne pro 2 do?
2020-06-21 06:30:37 <rnkn> Anne Pro seems to be drawing too much power on Rpi4 causing unmounting of drives, including the USB stick with /home
2020-06-21 06:31:41 <rnkn> util-linux seems antithetical to the whole alpine ethos of splitting packages into their most singular forms
2020-06-21 06:32:06 <oneinsect> hmmmm
2020-06-21 06:32:33 <oneinsect> do you have a multimeter with you?
2020-06-21 06:33:09 <rnkn> lol
2020-06-21 06:33:27 <rnkn> is that the thing the Ghostbusters use?
2020-06-21 06:35:59 <oneinsect> :D
2020-06-21 06:36:01 <oneinsect> lol
2020-06-21 06:36:23 <oneinsect> its used to measure your current draw so that we can make sure that is the problem
2020-06-21 06:36:57 <mps> yes, if you think of smoke from 'smoke tests' :)
2020-06-21 06:38:08 <rnkn> sorry I do not
2020-06-21 06:40:04 <rnkn> tangential q: when mounting as user, udisks mounts at /run/media/USER/LABEL; given that /run is tmpfs, does this mean mounting drives under /run eats up memory?
2020-06-21 06:40:47 <afontain_> I don't think so
2020-06-21 06:41:02 <TBB> of course it does
2020-06-21 06:41:24 <afontain_> if you mount it over a regular filesystem, do you copy what's in the mounted disk?
2020-06-21 06:41:29 <TBB> but the same amount as any mount really
2020-06-21 06:41:56 <TBB> nope, no copying happens
2020-06-21 06:43:14 <rnkn> okay, I can plug the USB 3.0 HDD into a 2.0 port and mount with udisks fine, but when unmounting then plugging into a 3.0 port dmesg says:
2020-06-21 06:43:16 <rnkn> usb usb2-port2: over-current change #5
2020-06-21 06:43:23 <rnkn> usb usb2-port3: over-current change #5
2020-06-21 06:43:28 <rnkn> usb usb2-port4: over-current change #5
2020-06-21 06:43:33 <rnkn> etc.
2020-06-21 06:44:18 <rnkn> with that always reassuring *click.. click.. click..*
2020-06-21 06:45:27 <rnkn> TBB: so mounting elsewhere (other than tmpfs /run) will make no difference to memory impact?
2020-06-21 06:46:34 <ikke> no
2020-06-21 06:46:43 <ikke> it's just a mount point
2020-06-21 06:47:38 <rnkn> cool
2020-06-21 06:48:03 <ikke> just like mounting a big disk on your root file system does not fill up your root filesystem
2020-06-21 06:48:29 <rnkn> I'm guessing "over-current" in dmesg is a clear indication that I don't need to buy a multimeter to know this is a power issue?
2020-06-21 06:48:41 <rnkn> ikke: cool, that makes sense
2020-06-21 07:00:34 <rnkn> g2g  thanks for the help!
2020-06-21 07:06:08 <mps> hmm, a hacker without multimeter ? :)
2020-06-21 07:07:24 <mps> (not to mention oscilloscope, microscope, logic analyzers)
2020-06-21 07:08:11 <afontain_> I've still to buy these too
2020-06-21 07:09:03 <mps> hmm, maybe we will need also telescopes soon :D
2020-06-21 07:10:12 <afontain_> wait, why?
2020-06-21 07:11:13 <mps> to hack spacex sitellites (and others maybe), they use musl libc, and u-boot
2020-06-21 07:11:32 <afontain_> lol
2020-06-21 07:20:22 <yunfan> mps: help
2020-06-21 07:21:15 <yunfan> mps: when i try to use the bootstrap.sh in aports:/scripts/ , it gave me many errors
2020-06-21 07:21:26 <yunfan> like unable to read database stte no such file or directory
2020-06-21 07:22:37 <mps> yunfan: what are you trying to do with bootstrap.sh, create new arch
2020-06-21 07:23:04 <yunfan> mps: yes, support for the csky 610
2020-06-21 07:23:04 <mps> and did you installed alpine-sdk
2020-06-21 07:23:43 <yunfan> mps of course, i had followed the wiki said
2020-06-21 07:24:08 <mps> hmm, then it should work
2020-06-21 07:24:50 <mps> except if it doesn't :)
2020-06-21 07:25:07 <yunfan> the example aarch64 works right now
2020-06-21 07:25:22 <yunfan> but for csky, it just failed
2020-06-21 07:26:13 <yunfan> is that because of binutils lack of supporting for csky?
2020-06-21 07:26:20 <mps> have no idea why, and I don't have much experience with bootstrap.sh
2020-06-21 07:26:46 <mps> could be, I just thought of binutils
2020-06-21 07:26:53 <yunfan> ok will try to googling
2020-06-21 11:58:10 <p3lim> mps: you want me to replace the entire init system just to get stdout/err into files?
2020-06-21 11:58:30 <p3lim> if that's the only way then openrc is seriously flawed
2020-06-21 11:59:51 <p3lim> for the record, I'm not using alpine for containers, I'm using the virt image
2020-06-21 12:00:20 <mps> p3lim: if I read your question correctly, you asked for systemd which can't work (thanks God) on musl based system
2020-06-21 12:00:50 <p3lim> didn't ask for systemd, just the same convenience with outputs from the processes run as services
2020-06-21 12:01:07 <p3lim> also not asking for journald, (r)syslog is sufficient
2020-06-21 12:01:08 <mps> and I told you for other pkgs which can be used to get what you asked for
2020-06-21 12:01:43 <p3lim> which is why I asked if the _default_ init system used by alpine can't handle outputs from processes
2020-06-21 12:01:45 <mps> and by installing runitd or s6 on alpine you don't need to replace init system
2020-06-21 12:02:15 <ikke> p3https://unix.stackexchange.com/questions/445427/how-to-view-daemon-stdout-in-openrc
2020-06-21 12:02:18 <mps> just add/change services you want to run under them
2020-06-21 12:02:24 <russkel> anyone have any experience with things like babeltrace/lttng?
2020-06-21 12:02:32 <ikke> russkel: no
2020-06-21 12:03:06 <russkel> :'(
2020-06-21 12:03:21 <p3lim> ikke: thanks, I'll give that a look
2020-06-21 12:03:35 <p3lim> I kind of assumed that `use logger` would do something, but I suppose not
2020-06-21 12:03:42 <ikke> that's just a dependency
2020-06-21 12:03:47 <ikke> nothing more
2020-06-21 12:04:50 <p3lim> kind of troubling that I couldn't find many services in the repositories handling process output
2020-06-21 12:04:58 <p3lim> although I didn't look at a lot of them
2020-06-21 12:08:46 <Cogitri> Yes, that's a bit annoying, something like `systemctl status $service` would be very nice to have with OpenRC
2020-06-21 12:09:15 <Cogitri> But I suppose we won't get that since OpenRC is pretty dead
2020-06-21 12:09:24 <p3lim> the output from systemctl status sucks anyways, it's always too short to be relevant and truncated by default
2020-06-21 12:09:48 <Cogitri> Hm, it usually is long enough for me to diagnose why a service doesn't start
2020-06-21 12:10:33 <ikke> and journalctl if you need more
2020-06-21 12:11:19 <winny> In case you're wondering if any distro has a tool like systemctl for openrc, gentoo has 'eselect rc' which unified all the enable,disable,start,stop,etc tasks to have a simpler language. Nobody uses it though. Openrc's ergonomics are pretty good.
2020-06-21 12:12:06 <Cogitri> I don't think `service $service $action` is too bad either
2020-06-21 12:12:16 <Cogitri> I just wished the log situation was different with openRC :)
2020-06-21 12:13:37 <mps> that is why I proposed to switch to runit
2020-06-21 12:13:39 <winny> It might be worthwhile to replace BusyBox syslogd with rsyslog, and use a tool such as lnav to view logs
2020-06-21 12:16:28 <p3lim> mps: switching to runit for custom services is fine, but I'd have to make custom ones for every service I install that is provided in aports
2020-06-21 12:17:04 <Cogitri> mps: I think switching to runit isn't worthwhile anymore since it's also dead :/
2020-06-21 12:17:17 <p3lim> isn't void using it?
2020-06-21 12:17:21 <Cogitri> Maybe s6-rc will become nicer to use down the road or wrappers like tt or 66 will take over
2020-06-21 12:17:36 <Cogitri> p3lim: Yes, but upstream abandoned it in favour of s6-rc
2020-06-21 12:17:55 <Cogitri> And runit has loads of missing functionality, e.g. service dependency management
2020-06-21 12:18:31 <Cogitri> So if you forget to start a service A that service B needs than runit won't start A for you, it'll just let B spin, checking if A has been started since
2020-06-21 12:18:58 <Cogitri> Which was a common source of problems for Void, e.g. when people forget enabling dbus and then wonder why the NetworkManager service does nothing
2020-06-21 12:54:30 <mps> Cogitri: service dependency is easy with runit
2020-06-21 12:54:52 <mps> s6 is not good for init 1
2020-06-21 12:55:48 <mps> runit is old and have some quirks but it is only viable option if we want to change from openrc
2020-06-21 12:55:54 <mps> IMO, ofc
2020-06-21 12:56:30 <mps> and, I'm not proposing change, for me everything is fine now
2020-06-21 13:02:08 <p3lim> hrm, `output_logger="logger"` doesn't seem to do anything
2020-06-21 13:02:21 <p3lim> must be something I'm missing, since that works fine just running logger by itself
2020-06-21 13:02:41 <p3lim> tried adding it both to init.d and conf.d
2020-06-21 13:18:59 <p3lim> `supervisor="supervise-daemon"` should send everything to syslog according to the docs, weird that it doesn't do that
2020-06-21 13:19:20 <artok> damn
2020-06-21 13:20:56 <artok> someone care to send color emoji to channel and see that my terminal dies? =)
2020-06-21 13:21:41 <ikke> 😉
2020-06-21 13:21:59 <artok> hmm. that has stripped colors
2020-06-21 13:22:04 <ikke> 👍
2020-06-21 13:22:21 <artok> those work
2020-06-21 13:22:34 <ikke> 😔
2020-06-21 13:23:24 <artok> next release of libxft might have it already fixed but they aren't that keen to do new releases, it seems
2020-06-21 13:24:12 <mps> xft? I'm running st with libxft
2020-06-21 13:24:49 <artok> yeah, st dies with https://termbin.com/lzqw depending on fonts
2020-06-21 13:25:40 <mps> hm, didn't had this for long time
2020-06-21 13:26:11 <mps> 'st -f Cousine:bold:size=8:antialias=true'
2020-06-21 13:30:19 <artok> cousine
2020-06-21 13:31:11 <mps> I keep most of fonts on $HOME/.fonts/
2020-06-21 13:31:25 <mps> just two from distro
2020-06-21 13:32:14 <mps> same for .icons and .themes
2020-06-21 13:33:58 <artok> let's see if dropping noto fonts do the trick
2020-06-21 13:35:54 <mps> I use Simbola fonts for emoticons, but they can't be distributed by license, only can be used if user downloads it
2020-06-21 13:36:48 <tehcloud> font licensing is far more insane than open source software licensing, from my experience
2020-06-21 13:37:17 <mps> tehcloud: licensing as such is insane
2020-06-21 13:57:23 <p3lim> this is so inconsistent, in one service's conf.d I'm able to override the supervisor set in init.d, but in another I can't
2020-06-21 14:22:15 <Mee> Hello
2020-06-21 14:28:30 <Mee> I have a small problem with my Alpine install, I tried it now in VirtualBox and on actual hardware but everytime I get to the setup-apkrepos script wget fails to download the mirror list an prints 'wget: bad address `mirrors.alpinelinux.org`' i've tried to circumvent that by manually adding a mirror to the system but apk fails to update the repos
2020-06-21 14:28:31 <Mee> no matter what I do. I've tried the links on other Computers in my network and no problems there. I've also checked the network - dhcp does work and pings to google.de do work as well. Does anyone have any advice for me, thx.
2020-06-21 14:41:54 <p3lim> Mee: does `ping mirrors.alpinelinux.org` work?
2020-06-21 14:43:43 <Mee> this results in 'ping: bad address `mirrors.alpinelinux.org`'
2020-06-21 14:44:22 <ikke> what about getent hosts mirrors.alpinelinux.org
2020-06-21 14:44:48 <jhass> any difference for ping -4 vs ping -6 ?
2020-06-21 14:46:55 <Mee> ikke: runs for a second then terminates without output and return code 2
2020-06-21 14:47:31 <Mee> jhass: no at least nothing obvious
2020-06-21 14:47:46 <bougyman> Mee: can you dig mirrors.alpinelinux.org on one of those other computers in your network?
2020-06-21 14:51:46 <Mee> I've used drill hope thats ok (dig is not installed) `mirrors.alpinelinux.org.        3161    IN      A       172.105.78.12`
2020-06-21 15:02:07 <bougyman> That's it, just one ip returns?
2020-06-21 15:02:25 <bougyman> huh. same here.
2020-06-21 15:02:38 <bougyman> so yeah, you have a weird problem.
2020-06-21 15:06:54 <fwm> how best to perform package maintaince? I want to try and find orphaned packages, or get a list of all top level packages with no dependacies
2020-06-21 15:07:18 <fwm> sorry, packages that aren't dependacies themselves
2020-06-21 15:07:51 <Mee> Thx :]  thought I missed something simple.
2020-06-21 15:18:21 <Hello71> ... nobody suggested to check /etc/resolv.conf?
2020-06-21 15:18:24 <mps> fwm: there is no orphaned packages in alpine, officially
2020-06-21 15:18:45 <Hello71> oh, I didn't read enough
2020-06-21 15:18:48 <mps> but there is unmaintained repo
2020-06-21 15:19:30 <Hello71> Mee: what if you just run wget http://mirrors.alpinelinux.org/
2020-06-21 15:19:49 <mps> fwm: but there is a lot of outdated and badly maintained packages
2020-06-21 15:20:07 <Hello71> actually, hm
2020-06-21 15:20:55 <Hello71> I guess still check what is in resolv.conf
2020-06-21 15:21:40 <Mee> Hello71: /etc/resolv.conf list my local dns server with its ipv4 address only (no ipv6 although available)
2020-06-21 15:22:22 <Mee> Hello71: there is no difference in running wget from the script or by itself
2020-06-21 15:22:55 <Hello71> can you wget http://172.105.78.12/
2020-06-21 15:23:13 <Hello71> I think it should give ssl error
2020-06-21 15:23:45 <afontain_> fwm: to get the list of packages explicitely installed, you simply look at /etc/apk/world
2020-06-21 15:24:03 <Mee> Hello71: yes it does after that connection reset by peer
2020-06-21 15:24:13 <Hello71> what error exactly?
2020-06-21 15:25:48 <Mee> ssl_client: <ip>: certificate verification failed: unable to get local issuer certificate
2020-06-21 15:25:51 <Mee> wget: error getting response: Connection reset by peer
2020-06-21 15:26:27 <Hello71> hm. so everything is fine except gai
2020-06-21 15:26:50 <Hello71> next I would probably try downloading extended iso for more troubleshooting tools
2020-06-21 15:27:18 <Hello71> or if you can fetch ldns-tools/strace externally then install manually with apk add
2020-06-21 15:27:35 <Mee> I currently run it on a older notebook sitting next to me with the x86-extended version of the newest alpine release
2020-06-21 15:28:57 <Hello71> oh, you already have extended? what does drill mirrors.alpinelinux.org return on the affected machine then
2020-06-21 15:29:06 <Mee> `alpine-extended-3.12.0-x86.iso` this to be exact
2020-06-21 15:29:52 <Mee> drill not found ...
2020-06-21 15:30:36 <Hello71> uh, maybe you need to apk add ldns-tools first
2020-06-21 15:32:07 <Mee> apk installs 3 packages resulting in `OK: 11MiB in 26 packages` but drill still is not found
2020-06-21 15:32:51 <Mee> ahhh `apk add drill` fixed it
2020-06-21 15:33:24 <Mee> drill has zero answers for me
2020-06-21 15:33:45 <Mee> and rcode: SERVFAIL
2020-06-21 15:36:57 <Hello71> probably you need to troubleshoot the network then
2020-06-21 15:37:08 <Hello71> extended comes with tcpdump too
2020-06-21 15:37:43 <Mee> I changed the nameserver in /etc/resolve.conf  to 8.8.8.8 and now it works
2020-06-21 15:38:31 <Mee> I run a pi-hole as my dns server and the addresses work with my normal machine. total guess here but does alpine not work with dnssec???
2020-06-21 15:40:08 <ScrumpyJack> has anyone managed to replace the stock linux distro of a hosted EFI booting VM with alpine linux?
2020-06-21 15:41:04 <ScrumpyJack> i've done it numerous time with legacy boot VMs, but can't figure out what's needed for EFI boot VMs
2020-06-21 15:45:16 <ScrumpyJack> aha! https://www.alextsang.net/articles/20191006-063049/index.html
2020-06-21 15:46:43 <Hello71> Mee: well it won't work if you force dnssec
2020-06-21 15:47:03 <Hello71> also musl doesn't support TCP
2020-06-21 15:48:17 <Hello71> ScrumpyJack: this is a pretty bad guide
2020-06-21 15:49:15 <Hello71> writing directly to disk will probably get you a corrupted iso
2020-06-21 15:49:55 <ScrumpyJack> what do you mean?
2020-06-21 15:50:19 <Hello71> although in this case you might be ok because the ESP is bigger than alpine-virt
2020-06-21 15:50:52 <Hello71> additionally the password command is not "password"
2020-06-21 15:52:31 <ScrumpyJack> dd to disk works fine for me for legacy boot, but not sure if it will work for a virtual bios that looks for an EFI partition
2020-06-21 15:53:04 <ScrumpyJack> i mean dd the iso to disk works fine for legacy boot, etc
2020-06-21 15:53:18 <Hello71> "virtual bios" ugh.
2020-06-21 15:53:47 <Hello71> how do you think the firmware boots from external media then?
2020-06-21 15:53:55 <ScrumpyJack> well whatever the hypervisor does. I'm being generic
2020-06-21 15:55:02 <Mee> I have no force option on the pi-hole only a use DNSSEC switch with the following warning `[...] If a domain fails validation or the upstream does not support DNSSEC, this setting can cause issues resolving domains.  [...]`
2020-06-21 15:58:52 <fwm> afontain_, Mee, many thanks
2020-06-21 15:58:58 <Hello71> I don't think pi hole has a force dnssec option
2020-06-21 15:59:25 <Hello71> iirc dnsmasq falls back to tcp more aggressively than is necessary
2020-06-21 15:59:31 <Hello71> or it doesn't support edns0? can't remember
2020-06-21 16:00:45 <Hello71> oh, and besides that wouldn't affect drill
2020-06-21 20:11:23 <c705> is anyone else having issues with the post trigger after apk add grub?
2020-06-21 20:12:03 <c705> I can't for the life of me understand what is happening or how to fix it:
2020-06-21 20:12:07 <c705> #11673
2020-06-21 20:12:13 <c705> https://gitlab.alpinelinux.org/alpine/aports/-/issues/11673
2020-06-22 00:07:31 <quazimodo> hi all
2020-06-22 00:08:02 <quazimodo> this could just be me, but I believe that when I run dockerised alpine with bash installed, the terminal is still a bit ... unfriendly
2020-06-22 00:08:37 <quazimodo> is this just because alpine is quite sparse, that i'd need some convenience packages that do things like autocomplete, better formatting etc?
2020-06-22 01:12:31 <LACampbell> quazimodo: I don't have bash installed, just the standard ash shell with busyboy
2020-06-22 01:12:33 <LACampbell> I have autocomplete
2020-06-22 01:14:07 <LACampbell> I am somewhat amused by the amount of people who need to ssh into their containers though :D
2020-06-22 01:15:26 <quazimodo> LACampbell: i don't ssh into them, but sometimes run /bin/bash for diagnostic reasons
2020-06-22 01:15:31 <quazimodo> rare but does happen
2020-06-22 01:15:56 <quazimodo> production containers are never ssh'd into, we actually cant
2020-06-22 01:16:04 <quazimodo> but in local hostdevelopment it's different
2020-06-22 01:18:58 <dustfinger> Hi, I am wondering if there is an ofono plugin for network manager for alpine linux? I tried adding it to my config and network manager attempted to load libnm-settings-plugin-ofono.so which does not exist on my system.
2020-06-22 01:23:47 <LACampbell> quazimodo: fair enough. I just rsync from an alpine dev environment to an alpine server. seems to work
2020-06-22 01:23:52 <LACampbell> regardless, I have autocomplete without bash
2020-06-22 01:24:00 <LACampbell> what formatting looks weird?
2020-06-22 01:27:31 <quazimodo> LACampbell: perhaps weird was the wrong word. Spartan, there's not much information in the shell, which is probably just the $PS1. But also autocomplete didn't exist  & navigation seemed fussy, certain ctrl/alt commands weren't moving the cursor but instead inserted control characters
2020-06-22 01:27:59 <quazimodo> i'll have to double check for an exact scenario, it was last week that I noticed but we don't have much work today so I thought I'd ask
2020-06-22 01:28:52 <LACampbell> I'd check it out on an alpine VM install
2020-06-22 01:28:57 <LACampbell> just to make sure it's not docker
2020-06-22 01:29:13 <LACampbell> like I run alpine in WSL, and I notice a few weird things I don't notice when I have alpine running properly on my server
2020-06-22 01:29:49 <LACampbell> in general though alpine is pretty spartan. I kind of like it that way, but YMMV
2020-06-22 14:06:08 <hugoc15> I am trying to create a local apk index of some local packages. I'm having some issues with the "builddate" configuration, it doesn't get set upon creating the index, even though it is set in every package. Does someone have a clue how to solve this?
2020-06-22 14:13:49 <hugoc15> Nvm, I solved it.
2020-06-22 15:21:13 <skyroveRR> Heya guys, I'm attempting to compile strace 4.18 for ARM aarch64 platform. I'm not using alpine linux, but I am using a custom distribution based on musl libc and I use alpine patches for my distro. I'm getting an error while compile strace: "In file included from syscall.c:1070: ./linux/aarch64/arch_regs.c:24:22: error: field 'aarch64_r' has incomplete type struct user_pt_regs aarch64_r ERROR: compile
2020-06-22 15:21:16 <skyroveRR> syscall.c failed." I googled the error and only came across one discussion from this page: https://github.com/skycocker/chromebrew/issues/1378 , and I tried using those options, but the --host option isn't application, and --enable-mpers=check is also not recognised by the strace 4.18's configure script. I'm using skarnet's gcc 8.2.0 toolchain, and here's the build log:
2020-06-22 15:21:18 <skyroveRR> http://pktsurf.in/p/strace.log.txt . Appreciate any assistance on this one :)
2020-06-22 15:25:32 <Hello71> isn't that three years old
2020-06-22 15:26:17 <skyroveRR> Yeah, but it is in alpine repos, so I decided to give it a try. It compiled fine on 32bit ARM, but not 64bit.
2020-06-22 15:27:04 <Hello71> not in any *current* repos
2020-06-22 15:28:32 <skyroveRR> Ah shit.
2020-06-22 15:28:35 <skyroveRR> One sec.
2020-06-22 15:29:43 <skyroveRR> Ah, my bad. Earlier during the day I did try 5.7, but it gave me a bunch of other errors, but I wasn't using alpine patches.
2020-06-22 15:35:13 <skyroveRR> Hello71: same exact error: http://pktsurf.in/p/strace.1.log.txt
2020-06-22 19:05:42 <abenz> hi, I'm looking into using alpine (diskless) full time, upon experementing with the system to learn more, I set lbu backup to 3, and enrypted my apkovls
2020-06-22 19:06:22 <abenz> all was going well until I did something that resulted in me having a localhost.apkovl (as opposed to myhostname.apkvol)
2020-06-22 19:08:22 <abenz> I figured, I will just delete the localhost.apkvol and reboot, which will leave only myhostname.apkovl and restore my system.. I tried deleting localhost.apkovl by running alpine-setup, and choosing localhost for my hostname (to match that of the offending apkovl), once that was done, I issued lbu ci -d. but this did not delete localhost.apkovl, instead it deleted myhostname.apkovl !
2020-06-22 19:10:07 <abenz> no worries, I figured, since I have 3 backups, I just need to delete localhost.apkovl, rename one of the backups and I will be back to a good state. after realizing that lbu ci -d will not delete localhost.apkovl, I rw,remount the /media/usb and deleted the file, renamed one of the old backups, and rebooted
2020-06-22 19:10:31 <abenz> apologies for walls of text
2020-06-22 19:11:15 <abenz> upon reboot, my hostname, passwords..etc came back correctly, but hardware is not detected.. my second monitor doesn;t come up
2020-06-22 19:11:56 <ikke> is the modloop service started?
2020-06-22 19:11:58 <abenz> consulting the wiki, I dont know where I messed up, where did I go wrong? and is my system currently in broken state?
2020-06-22 19:12:30 <ikke> sounds like you are missing either kernel modules or firmware or something like that
2020-06-22 19:13:51 <abenz> I was under the impression by messing with apkovls, I can simply restore that to have my system exactly how it was. Do you think running alpine-setup again was a bad idea (I Ctrl+C out of it as soon as I reached the stage to set my hostname)
2020-06-22 19:14:19 <ikke> perhaps
2020-06-22 19:15:34 <abenz> is it considered normal practice to rw,remount to manipulate apkovls files? I ran alpine-setup simply to make use of lbu ci -d (ie adhere to alpine way of deleting the backup).. but that backfired
2020-06-22 19:16:40 <ikke> Depends on whether you know what you are doing :)
2020-06-22 19:16:56 <ikke> but I don't think most people manually manipulate apkovls
2020-06-22 19:17:07 <ikke> (some might generate them automatically)
2020-06-22 19:17:27 <ikke> What were you adjusting?
2020-06-22 19:17:54 <abenz> if in /media/usb I have hostname1.apkovl and hostname2.apkovl (created by mistake), whats the proper way of deleting the second apkovl? thats waht I was trying to do
2020-06-22 19:18:20 <ikke> just deleting it, I would say
2020-06-22 19:18:53 <abenz> ah, so rw,remount is the right way all along
2020-06-22 19:19:02 <abenz> got it
2020-06-22 19:19:29 <ikke> I don't think lbu ci -d allows you to remove arbitrary apkovls
2020-06-22 19:20:05 <ikke> it says 'remove old apk overlay files', but not sure what it considers old
2020-06-22 19:20:19 <abenz> I thought it deleted the apkovl that matches the running system's hostname?
2020-06-22 19:21:26 <abenz> ikke: thanks for the tips
2020-06-22 19:23:08 <ikke> "local rmfiles=$(ls "$mnt/"*.apkovl.tar.gz* 2>/dev/null)"
2020-06-22 19:23:13 <ikke> so it seems to remove all of them
2020-06-22 19:39:20 <abenz> cheers
2020-06-22 23:25:26 <dan45> Has anybody got audio working on a raspberry pi 3b+?
2020-06-22 23:26:27 <dan45> Been banging my head against the wall for a while now... I've got alsa but no /proc/asound
2020-06-22 23:27:30 <dan45> Also no 'snd' modules in lsmod
2020-06-22 23:44:34 <dan45> looks like I just needed to add snd_bcm2835 to /etc/modules
2020-06-23 05:00:45 <twb> OK so I have an EFI-based VM and I need to switch it to BIOS-based
2020-06-23 05:01:36 <twb> I did "apk add e2fsprogs syslinux" and I'm trying to add a label to the rootfs with tune2fs, but it's not in my path...
2020-06-23 05:02:35 <twb> Oh I guess never mind, the comment says "# root device - if not specified, will be guessed using blkid -o export /dev/root"
2020-06-23 05:03:36 <twb> And /dev/vda has an MBR on it already also, so as long as extlinux's gpt driver is there, I should be good...
2020-06-23 05:33:58 <twb> After fucking up a few times, I think I should have just turned off EFI and then tried it -- I think grub probably would have Just Worked
2020-06-23 06:46:43 <twb> How do I do "chattr -i" on alpine?  e2fsprogs doesn't seem to have chattr
2020-06-23 06:48:37 <afontain_> > apk search cmd:chattr
2020-06-23 06:48:39 <afontain_> e2fsprogs-extra
2020-06-23 06:48:49 <twb> ah thanks
2020-06-23 06:49:01 <twb> I didn't know how to do search
2020-06-23 06:49:18 <afontain_> you can even use `apk add cmd:chattr` directly
2020-06-23 06:49:33 <twb> I wonder if that works via ansible :P
2020-06-23 06:56:04 <twb> Hrmmm, I'm doing SOMETHING wrong here...
2020-06-23 06:56:22 <twb> I have an alpine VM that's trying to provision the disks of other (old, garbage) VMs
2020-06-23 06:57:34 <twb> I've done "mkfs.ext4 -L print03-rootfs /dev/vdi", then "mount /dev/vdi /srv/backup/print03-rootfs", then rsync, then this crap: http://ix.io/2pVm
2020-06-23 06:58:02 <twb> Which AFAICT is pretty much the same process I've used before for partition-less extlinux
2020-06-23 06:58:43 <twb> Oh FFS
2020-06-23 06:58:53 <twb> I was using the wrong disk.
2020-06-23 07:00:22 <twb> Now at least it's booting the extlinux menu and just failing to load the kernel
2020-06-23 07:34:23 <twb> I put a partition table and mbr.bin on there, and now mbr.bin can't find ldlinux.sys on the ext4 drive.  I hate BIOS.
2020-06-23 08:06:45 <twb> does alpine have memtest86 or memtest86+ ?  I can't see it in "apk search memtest"
2020-06-23 08:14:04 <ikke> I don't see it either
2020-06-23 08:14:05 <ikke> https://pkgs.alpinelinux.org/contents?file=memtest&path=&name=&branch=edge
2020-06-23 08:34:58 <twb> OK I still don't know what the hell is going on
2020-06-23 08:35:09 <twb> but from extlinux (booting off an ext4 rootfs), I can run rosh.c32
2020-06-23 08:35:22 <twb> And in there I can run "ls /boot" and it gives me a bunch of files that definitely *aren't* in /boot
2020-06-23 08:36:04 <twb> e.g. sane-umas-config-doc.html which isn't even on that drive
2020-06-23 09:03:49 <twb> OK guess what
2020-06-23 09:04:01 <twb> I downgraded the rootfs from ext4 to ext3, and extlinux started working properly
2020-06-23 09:04:21 <twb> So something has changed in ext4.ko or mke2fs.conf that extlinux doesn't like.
2020-06-23 09:06:03 <twb> It might be this option which is in alpine's mke2fs but not Debian's:
2020-06-23 09:06:07 <twb> [options] fname_encoding = utf8
2020-06-23 09:06:29 <twb> ...although I did both "mkfs.ext4" and "mkfs.ext3" from alpine, so it shouldn't be...
2020-06-23 12:25:29 <EF> Has anyone got any pointers please. Got Alpine installed on a LVM2 RAID10 install but struggling with grub. /boot and / are on lvs
2020-06-23 12:26:11 <EF> Pointers to useful readmes would be good
2020-06-23 13:26:09 <egrain> how big does that dos partition have to be on the aarch64 ssd?
2020-06-23 14:23:48 <steinbach> ...
2020-06-23 14:24:54 <egrain> apk install sshd? is that right?
2020-06-23 14:25:01 <egrain> pi is running, but no ssh on it yet it seems.
2020-06-23 14:26:16 <egrain> oh, apk add.
2020-06-23 14:26:17 <egrain> okay.
2020-06-23 14:29:59 <egrain> hm, eth0 is down.
2020-06-23 14:44:49 <egrain> false alarm. pulled the plug.
2020-06-23 14:44:52 <egrain> happens to the best of us.
2020-06-23 14:45:10 <egrain> this setup-alpine is nice for newbies.
2020-06-23 14:45:13 <egrain> much appreciated.
2020-06-23 14:55:49 <Hello71> ... LVM RAID0?
2020-06-23 14:56:31 <egrain> openrc, right?
2020-06-23 14:56:35 <egrain> how do i restart sshd?
2020-06-23 14:57:36 <ikke> rc-service sshd restart
2020-06-23 15:08:17 <egrain> thanks.
2020-06-23 15:08:19 <egrain> your adduser is weird.
2020-06-23 15:08:35 <egrain> -G makes the main group, and you do'nt have -m but -h /home/dir
2020-06-23 15:08:49 <egrain> anyway, how do i add me to users and wheel.
2020-06-23 15:08:53 <egrain> -G doesn't like ,
2020-06-23 15:09:14 <egrain> addgroup afterwards?
2020-06-23 15:13:53 <egrain> if i run kodi, and change stuff in there, i ahve to always do lbu commit -d?
2020-06-23 15:20:09 <egrain> apk search kodi gives me nothing. but in packages on alpine.com it says it exists for aarch64.
2020-06-23 15:24:07 <egrain> oh, why you have community commented out?
2020-06-23 15:24:22 <egrain> now the question is, which one do i install?
2020-06-23 15:26:54 <egrain> i'm guessing kodi and the rest are addons? plugins? one of those.
2020-06-23 15:29:10 <mps> wiki.a.o
2020-06-23 15:33:28 <egrain> does alpine have a kodi wiki?
2020-06-23 15:33:32 <egrain> i haven't found anything so far.
2020-06-23 15:54:24 <egrain> what's with the percentages counting up everytime i boot the pi4?
2020-06-23 16:01:57 <egrain> can i get out of this diskless mode somehow?
2020-06-23 16:02:00 <egrain> like for good?
2020-06-23 16:08:37 <hodapp> what percentages are you talking about?
2020-06-23 16:10:39 <egrain> like this:
2020-06-23 16:10:48 <egrain> 30% ################################################
2020-06-23 16:10:56 <egrain> 100% is one line of #####
2020-06-23 16:11:13 <egrain> then it seems to actually boot, with the messages and all.
2020-06-23 16:11:29 <egrain> i wouldn't mind, but it takes a while.
2020-06-23 16:12:30 <egrain> also now it says starting kodi but nothing happens. except right afterwards starting sshd.
2020-06-23 16:12:48 <egrain> i could really use a wiki here or something.
2020-06-23 16:14:24 <afontain_> I guess the percentages at boot is apk fetching and installing every package you want on the fly
2020-06-23 16:15:35 <egrain> like the lbu commit thing?
2020-06-23 16:15:46 <egrain> so if i unplug the network nothing works anymore?
2020-06-23 16:16:45 <egrain> no. unplugged it. it's still counting up.
2020-06-23 16:23:11 <egrain> if i change ro to rw in /etc/fstab i should have ended the diskless scenario, right?
2020-06-23 16:24:07 <c705> egrain: read the wiki
2020-06-23 16:24:58 <egrain> i'm trying. i can't find anything in there. how do i navigate it? the search thing on the side brings up random things.
2020-06-23 16:25:35 <ikke> c705: it would help if you link to concrete pages
2020-06-23 16:25:37 <c705> https://wiki.alpinelinux.org/wiki/Installation
2020-06-23 16:26:30 <egrain> c705, that doesn't help me.
2020-06-23 16:26:35 <egrain> i'm on a pi4.
2020-06-23 16:26:40 <egrain> i was never asked about diskless.
2020-06-23 16:26:51 <egrain> i even did the setup-alpine thing again, thinking i must have overlooked it.
2020-06-23 16:27:01 <egrain> i just want it to run kodi.
2020-06-23 16:27:18 <c705> egrain> can i get out of this diskless mode somehow?
2020-06-23 16:27:28 <ikke> alpine only supports diskless setups for raspberry pi
2020-06-23 16:27:54 <ikke> But users have gotten a disk setup to work
2020-06-23 16:28:17 <ikke> https://wiki.alpinelinux.org/wiki/Classic_install_or_sys_mode_on_Raspberry_Pi
2020-06-23 16:28:23 <ikke> not sure about how good that works
2020-06-23 16:29:22 <egrain> yeah, then this is my cue.
2020-06-23 16:29:34 <egrain> i thank you for your support.
2020-06-23 16:29:37 <afontain_> postmarketOS raspberry pi are also using some kind of sys mode, so it must work well enough
2020-06-23 16:31:21 <mps> setup-disk need mmcblk addition, to recognize them as possible targets for sys mode
2020-06-23 16:32:05 <mps> I know this issue long time but never found time to fix it and test
2020-06-23 17:04:36 <KatolaZ> mps: it was not needed in my case (tested on both rpi2 and rpi3)
2020-06-23 17:04:49 <KatolaZ> the tutorials linked in the wiki worked seamlessly
2020-06-23 17:05:48 <mps> KatolaZ: yes, if create partitions and run setup-disk with mounted target FS
2020-06-23 17:06:24 <mps> some kind of 'manual' installation
2020-06-23 17:09:54 <KatolaZ> it was enough for me
2020-06-23 17:10:09 <KatolaZ> I mean, I am still able to create an fs and mount a partition :P
2020-06-23 17:10:26 <KatolaZ> it's much better to keep things simple, IMHO
2020-06-23 17:10:30 <KatolaZ> and this is quite simple
2020-06-23 17:25:07 <mps> agree :)
2020-06-23 17:25:39 <mps> that is probably reason I didn't added mmcblk to setup-disk
2020-06-23 21:01:59 <guest170> hi
2020-06-23 21:03:59 <guest170> I'm using alpine from RAM. When I create initramfs and boot that, I get the following error: "mount: mounting /dev/loop0 on /.modloop failed: Invalid argument"
2020-06-23 21:04:38 <guest170> What am I missing? I just run "mkinitfs" and adjusted grub to boot the new initramfs
2020-06-23 21:05:16 <ikke> I don't think that's going to work (running mkinitfs with run-from-ram)
2020-06-23 21:05:42 <guest170> do I need to create a new ISO?
2020-06-23 21:06:40 <ikke> Something like this: https://wiki.alpinelinux.org/wiki/Upgrading_Alpine#Upgrading_Alpine_Linux_on_other_removable_media_.28such_as_CF.2FUSB.29
2020-06-23 21:08:04 <guest170> yeah, I did that all
2020-06-23 21:08:31 <guest170> maybe not "all", but almost
2020-06-23 21:08:45 <guest170> I think, my initramfs create process is broken
2020-06-23 21:09:23 <ikke> did you run setup-bootable?
2020-06-23 21:09:38 <guest170> https://s12.directupload.net/images/200623/pwrpbcat.png
2020-06-23 21:10:49 <guest170> I don't know, if our understanding is the same. I need a custom initramfs, because I want to passthrough PCI to a KVM guest.
2020-06-23 21:11:13 <guest170> so I want to swap out the current initramfs on my USB stick with my custom initramfs
2020-06-23 21:11:36 <guest170> an initramfs with vfio support
2020-06-23 21:12:35 <ikke> hmm, ok, and did you do that before as well, or just started doing it now
2020-06-23 21:12:55 <guest170> how do you mean?
2020-06-23 21:13:00 <ikke> because the run-from-ram setup is not the same as a normal sys install, which is you see involving a modloop
2020-06-23 21:13:40 <guest170> so, if I want to swap out initramfs, I also need to create a new modloop?
2020-06-23 21:13:51 <ikke> but sadly enough I don't know enough about it to be of concrete help
2020-06-23 21:14:23 <guest170> I had a similar problem many many years ago :D
2020-06-23 21:14:36 <guest170> I think, I just created a new alpine ISO
2020-06-23 21:15:09 <guest170> I wanted to add flashcache support
2020-06-23 21:16:18 <ikke> Not sure if usable, but apparently there is an update-kernel script: https://gitlab.alpinelinux.org/alpine/alpine-conf/-/blob/master/update-kernel.in
2020-06-23 21:16:57 <guest170> I'm not sure, but maybe I need to set the initfs_features?
2020-06-23 21:16:59 <guest170> https://github.com/alpinelinux/aports/search?q=initfs_features&unscoped_q=initfs_features
2020-06-23 21:42:45 <codebam_> why is gnutls not trusting my letsencrypt certificates?
2020-06-23 21:42:55 <codebam_> firefox and chromium are trusting them
2020-06-23 21:43:15 <codebam_> it's causing issues with weechat because weechat isn't trusting any of the letsencrypt certs on my IRC servers
2020-06-24 00:16:56 <TBB> codebam, it's probably connected to what cert chains those use; could be there's a difference and gnutls is using one that doesn't include letsencrypt
2020-06-24 00:17:05 <TBB> this is purely a guess tho
2020-06-24 01:08:49 <Hello71> you didn't include the intermediate
2020-06-24 12:36:06 <yunfan> mps: i just found an old notebook, which has a via's nano cpu
2020-06-24 13:22:37 <mps> yunfan: happy hacking :)
2020-06-24 13:23:27 <mps> does alpine boot on it?
2020-06-24 13:25:50 <yunfan> mps: it used to boot on it, so i think its not a problem
2020-06-24 13:26:00 <yunfan> remember i told you that the X is a issue
2020-06-24 13:30:04 <mps> aha, understand. driver problem?
2020-06-24 13:30:38 <yunfan> yep
2020-06-24 15:17:41 <ncopa> yunfan: is it a hp notbook?
2020-06-24 15:18:00 <ncopa> i had one of those
2020-06-25 06:58:56 <yunfan> ncopa: nope, its a chinese brand, and unfortunately its screen driver since broke up
2020-06-25 06:59:03 <yunfan> so i havnt consitune yesterday
2020-06-25 06:59:55 <yunfan> btw, the setup-disk dont work on arm boards since it cant install syslinux mps
2020-06-25 07:11:04 <mps> yunfan: this "can't found syslinux" is harmless though annoying
2020-06-25 07:11:13 <mps> in setup-disk
2020-06-25 07:13:41 <mps> and, yes, it doesn't work for arm boards
2020-06-25 07:14:29 <mps> we have to create setup-flash for every particular arm board :)
2020-06-25 07:14:56 <yunfan> mps: i dont know why its difference for arm board
2020-06-25 07:15:27 <mps> every board need its specific u-boot
2020-06-25 07:16:18 <yunfan> mps: but since the script is running in a alpine linux, isnt that mean the uboot has already been installed in the usb media?
2020-06-25 07:16:18 <mps> yes, we can detect on boot board type and then use specific u-boot
2020-06-25 07:16:40 <yunfan> how about extract the firmware and offset from the usb medis itself
2020-06-25 07:17:07 <mps> it is installed on boot device/media but not on target
2020-06-25 07:17:38 <mps> extracting would be error prone
2020-06-25 07:17:59 <yunfan> :[
2020-06-25 07:18:28 <yunfan> i hope those arm board could boot like rpi
2020-06-25 07:18:52 <mps> uhm
2020-06-25 07:19:36 <mps> u-boot is quite fine
2020-06-25 07:22:47 <mps> I don't know for any distro or OS which can be installed automatically on arm boards
2020-06-25 07:23:32 <mps> only specifically tailored for particular boards can be
2020-06-25 08:45:45 <yunfan> mps: if all arm boards's bootloader acts like rpi's i think that wont be a problem
2020-06-25 08:45:49 <markand> hello
2020-06-25 08:46:08 <yunfan> but linus dont care of this, so no one saying f work to them, so no improvement
2020-06-25 08:46:30 <markand> is there something else to install to make oauth working with firefox? on some websites trying to identify just throw me a minimalist error that it was unable to authenticate me (it does not even tries to open oauth external windows)
2020-06-25 08:48:28 <mps> yunfan: Linus didn't created boot loaders and arm boards, why he have to care about this insanity
2020-06-25 08:49:35 <mps> this insanity is created by SOC designers which want to monetize their design
2020-06-25 08:50:11 <KatolaZ> yunfan: the way the bootloader works on rpi is insane
2020-06-25 09:58:07 <mps> KatolaZ: yes, I agree. RPis are worst things in arm jungle
2020-06-25 09:59:31 <yunfan> mps: because his pressure helped the creation of dtb
2020-06-25 09:59:45 <yunfan> KatolaZ: its just convenient
2020-06-25 10:02:15 <KatolaZ> no yunfan, it's just sick
2020-06-25 10:02:42 <KatolaZ> and useless, since other SOC are able to avoid that much complexity without problems
2020-06-25 10:02:51 <mps> yunfan: no, it is not for convenience but for 'locking' users to their closed systems
2020-06-25 10:03:30 <yunfan> mps: KatolaZ you could build a opensourced one , and acts like it
2020-06-25 10:03:43 <yunfan> i mean read from the partition's configuration
2020-06-25 10:03:56 <yunfan> no need to dd sdcard
2020-06-25 10:04:07 <mps> yes, we can, but that doesn't mean we don't see bad things
2020-06-25 10:04:25 <mps> and we have rights to say our observations
2020-06-25 10:05:15 <KatolaZ> yunfan: an opensourced what>
2020-06-25 10:05:17 <KatolaZ> ?
2020-06-25 10:05:41 <yunfan> KatolaZ: opensourced firmwarm to be prepared for those boards
2020-06-25 10:05:51 <mps> and, btw, I made boot loader for 6502 (arm predecessor) long time ago :)
2020-06-25 10:06:55 <mps> and, it is not complicated thing to make
2020-06-25 10:07:44 <KatolaZ> yunfan: how exactly?
2020-06-25 10:07:49 <yunfan> i think convenient is important, look at arduino, its successful based on this
2020-06-25 10:07:57 <yunfan> KatolaZ: i dont know
2020-06-25 10:07:59 <KatolaZ> yunfan: uboot is convenient
2020-06-25 10:08:08 <KatolaZ> having to use rpi-custom-kernel
2020-06-25 10:08:14 <KatolaZ> that is not convenient at all
2020-06-25 10:08:32 <yunfan> KatolaZ: that's not the feature i point to
2020-06-25 10:08:34 <KatolaZ> yunfan: then if you don't know, how can you say it's easy? :)
2020-06-25 10:08:44 <yunfan> KatolaZ: because i am a user
2020-06-25 10:08:48 <KatolaZ> oh sure
2020-06-25 10:08:52 <yunfan> KatolaZ: its steps is easy to me
2020-06-25 10:08:55 <KatolaZ> so everything should be easy
2020-06-25 10:09:00 <KatolaZ> yep
2020-06-25 10:09:04 <yunfan> i dont need to dd every time
2020-06-25 10:09:10 <KatolaZ> then complain to rpi foundation :)
2020-06-25 10:09:31 <KatolaZ> 'cause they decided to keep stuff secret, and to lock-in users
2020-06-25 10:09:37 <KatolaZ> in the name of opensource and sharing
2020-06-25 10:09:41 <mps> yunfan: you don't need to dd every time with u-boot, also
2020-06-25 10:09:53 <KatolaZ> indeed
2020-06-25 10:10:10 <mps> do it once and forget it
2020-06-25 10:10:17 <yunfan> mps: different board need to dd to different offset
2020-06-25 10:10:22 <yunfan> its quite sick
2020-06-25 10:10:51 <KatolaZ> yeah, they also decided to call each borad with a different name
2020-06-25 10:10:55 <KatolaZ> :P
2020-06-25 10:11:12 <mps> that's true, but how often someone move sdcards (boot media) to different devices
2020-06-25 10:12:35 <mps> I don't think other SOCs are perfect (they are not) but RPis are worst, imo
2020-06-25 10:13:04 <KatolaZ> I think the whole ARM market is a bit sick, since it allows a lot of room for lock-in
2020-06-25 10:13:26 <KatolaZ> but I don't see a dd offset as a massive problem
2020-06-25 10:13:30 <mps> because that I call that 'arm jungle'
2020-06-25 10:13:44 <KatolaZ> being unable to use a vanilla kernel is much more disturbing to me
2020-06-25 10:14:34 <mps> heh, I managed to boot mainline 5.6 on RPi zero w
2020-06-25 10:14:51 <mps> and also u-boot on it
2020-06-25 10:15:40 <mps> but a lot of work is in front of me to get alpine working on it out-of-the-box
2020-06-25 10:16:18 <KatolaZ> good mps :)
2020-06-25 10:16:48 <mps> :)
2020-06-25 10:17:51 <mps> and I do this in 'really' free time, I don't have much interest in using RPis
2020-06-25 10:18:41 <mps> allwinners and rockchip are my preferred choices
2020-06-25 10:21:04 <KatolaZ> mps: free time has driven most of the great innovations in IT in the last 50 years
2020-06-25 10:21:11 <KatolaZ> :)
2020-06-25 10:21:20 <albr> TuxThePenguin[d]: > allwinners and rockchip are my preferred choices
2020-06-25 10:21:21 <albr> TuxThePenguin[d]: You might like Pine64 :P
2020-06-25 10:22:47 <mps> albr: well, I use rk3399 in notebook (chromebook) for nearly 3 years
2020-06-25 10:22:57 <albr> unowe[d]: Does Alpine work well on the Pine64?
2020-06-25 10:23:15 <albr> TuxThePenguin[d]: I tried PostmarketOS on my PBP for a bit and it worked nicely
2020-06-25 10:24:15 <mps> userspace works on every arm I tried, kernel and boot loader could be problem
2020-06-25 10:27:25 <mps> KatolaZ: OT question, but, are you KatolaZ of Devuan project
2020-06-25 10:28:13 <KatolaZ> I was KatolaZ of Devuan project :)
2020-06-25 10:29:01 <mps> :) good (great) work
2020-06-25 10:29:48 <mps> I installed it month ago on one old notebook to have it just in case
2020-06-25 10:30:29 <mps> I didn't manage to repair f2fs on alpine with alpine but worked flawlessly from Devuan
2020-06-25 11:31:58 <hiro> KatolaZ: what happened, devuan dead?
2020-06-25 11:37:34 <KatolaZ> hiro: uh? Devuan is alive and kicking, as am I
2020-06-25 11:40:15 <hiro> why the past tense then? "I was"
2020-06-25 11:42:42 <ikke> Most likely reason is that they are not involved anymore
2020-06-25 11:43:47 <KatolaZ> it just means that I am not actively contributing to Devuan
2020-06-25 12:00:17 <hiro> ah. any reasons?
2020-06-25 12:00:27 <hiro> i always fear the next level drama...
2020-06-25 12:05:27 <KatolaZ> hiro: do you get around asking all people why they don't contribute any more to the last project they used to contribute to? :)
2020-06-25 12:05:30 <mps> hiro: found Alpine :)
2020-06-25 12:06:12 <hiro> KatolaZ: yes! :)
2020-06-25 12:07:06 <KatolaZ> :)
2020-06-25 12:13:57 <hiro> i can imagine so many reasons...
2020-06-25 12:25:13 <yann-kaelig> Hi
2020-06-25 12:26:00 <yann-kaelig> Is there an issue with busybox with zsh ? I can not list my directory when I execute ls. Or am I doing something wrong ?
2020-06-25 12:27:12 <yann-kaelig> ls /home > ls: unrecognized option: human-readable
2020-06-25 12:27:53 <ikke> yann-kaelig: `type ls`
2020-06-25 12:28:07 <ikke> ls is an alias that provides an option that busybox ls does not support
2020-06-25 12:30:52 <yann-kaelig> ikke: Hi. It's a first time for me on busybox. How can I list the current directory content or a directory ? On busybox website  the command is ls ( ls 92 List directory contents )
2020-06-25 12:31:26 <ikke> try; unalias ls;
2020-06-25 12:32:06 <ikke> the problem is aliases defined in your shell
2020-06-25 12:32:24 <ikke> ls /home works with bb ls
2020-06-25 12:33:00 <ikke> but `ls --human-readable /home` does not
2020-06-25 12:36:41 <yann-kaelig> ikke: thx you, Indeed I have an alias on my zshrc which I have to adapt for these case
2020-06-25 12:43:19 <mps> bb ls have this option as '-h'
2020-06-25 12:44:03 <mps> busybox ls -lh
2020-06-25 13:10:20 <yann-kaelig> thx
2020-06-25 13:11:59 <yann-kaelig> well, I'm going to build custom packages. It is not possible to read the content of a apk binairy package ? There is a dedicated tools for that ?
2020-06-25 13:13:06 <yann-kaelig> I tried to open an apk package in a same way I'm doing for gentoo or Arch but the content is not huamn readable
2020-06-25 13:18:11 <mps> yann-kaelig: tar tvf apk-name
2020-06-25 13:18:53 <mps> or, apk info -L pkgname
2020-06-25 13:23:57 <yann-kaelig> thx, well  I found that downloading the apk and readint it with ARK works, but not directly from firefox. Perhaps I need an plugin for that, I will tried to find how to do this
2020-06-25 14:12:15 <yann-kaelig> Am I doing something in a wrong way. I have defined SRCDEST=$HOME/pkgsrc/ in my /etc/abuild.conf, next I'm executing abuild checksum in the folder where I have created the APKGBUILD, but still I get the src directory created in this folder, which I don't want.
2020-06-25 14:13:23 <ikke> Those are 2 separate concepts
2020-06-25 14:13:58 <ikke> https://gitlab.alpinelinux.org/alpine/abuild/blob/master/abuild.in#L2658
2020-06-25 14:14:05 <ikke> It's using srcdir, which is hardcoded
2020-06-25 14:14:24 <ikke> You can overwrite it I see
2020-06-25 14:14:45 <Hello71> SRCDEST is for distfiles
2020-06-25 14:15:01 <ikke> which get symlinked in the srcdir
2020-06-25 14:35:08 <yann-kaelig> the maintainer line in the APKBUILD  has a link with the variable Maintainer I can see in the abuild.conf ? Do I need to manually write this in the APKBUILD ?
2020-06-25 14:40:25 <ikke> The variable is only used by newapkbuild
2020-06-25 14:40:37 <ikke> the one in abuild.conf
2020-06-25 14:42:14 <yann-kaelig> ok, thx
2020-06-25 14:42:42 <ikke> so yes, you need to make sure it's set in the APKBUILD
2020-06-25 15:10:36 <yann-kaelig> this is another thing that strange for me. I have created a $HOME/apkgbuild and inside this dir a have a subdir for my softwares which contain the APKBUILD. Also in my abuild.conf I have REPODEST=$HOME/pkgbin/v3.12/main. I run abuild -r and the build finisehd he's creating a apkgbuild folder in my $HOME/pkgbin/v3.12/main/apkbuild . Is it possible to not create this direcotry inside my REPODEST ?
2020-06-25 15:12:00 <yann-kaelig> I would like to have $HOME/pkgbin/v3.12/main/x86_64 and not $HOME/pkgbin/v3.12/main/apkgbuild/x86_64
2020-06-25 15:18:18 <ikke> the notion of a repo is hardcoded in abuild
2020-06-25 15:19:20 <ikke> https://gitlab.alpinelinux.org/alpine/abuild/blob/master/abuild.in#L1658
2020-06-25 15:20:19 <ikke> So don't think you can prevent that from happening
2020-06-25 15:20:32 <goofie> is there a repo for alpine with wakeonlan available? I am having such a hard time finding it
2020-06-25 15:21:42 <ikke> I don't think so
2020-06-25 15:22:59 <goofie> ah, looks like 'awake' will work
2020-06-25 15:52:48 <yann-kaelig> Do I need to manually run abuild-sign or is it something automatically done after abuild -r ?
2020-06-25 15:54:05 <ikke> automatically done
2020-06-25 15:54:09 <yann-kaelig> thx
2020-06-25 16:10:39 <Eskuero> Hi
2020-06-25 16:11:05 <Eskuero> Has someone run into issues trying to install openjdk?
2020-06-25 16:11:11 <Eskuero> remnant:~# apk add openjdk11-jreERROR: unsatisfiable constraints:  libffi-3.2.1-r6:    conflicts: libffi-3.3-r2    satisfies: python3-3.8.2-r0[so:libffi.so.6] python2-2.7.18-r0[so:libffi.so.6] py3-cffi-1.13.2-r0[so:libffi.so.6]  libffi-3.3-r2:    conflicts: libffi-3.2.1-r6    satisfies: p11-kit-0.23.20-r5[so:libffi.so.7]
2020-06-25 16:15:08 <ikke> what does apk version -l '?' return?
2020-06-25 16:16:42 <Eskuero> Installed:                                Available:py3-cffi-1.13.2-r0                      < 1.14.0-r2 libffi-3.2.1-r6                         < 3.3-r2 python3-3.8.2-r0                        < 3.8.3-r0
2020-06-25 16:16:59 <Eskuero> y3-cffi-1.13.2-r0                      < 1.14.0-r2
2020-06-25 16:17:06 <Eskuero> libffi-3.2.1-r6                         < 3.3-r2
2020-06-25 16:17:10 <Eskuero> python3-3.8.2-r0                        < 3.8.3-r0
2020-06-25 16:17:36 <Eskuero> apk upgrade never said anything
2020-06-25 16:21:31 <c705> i'm on edge. I cannot replecate
2020-06-25 16:23:44 <Eskuero> I'm too
2020-06-25 16:36:07 <Eskuero> apk fix libffi
2020-06-25 16:36:24 <Eskuero> Did the trick and installed it, then the remaining upgraded fine
2020-06-25 16:36:50 <Eskuero> What happened, I don't know, thanks everyone
2020-06-25 18:30:08 <yann-kaelig> there is something like conflicts= variable in apkgbuild ?
2020-06-25 18:30:41 <yann-kaelig> I do not see a conflicts variable in the apkgbuild from the aport
2020-06-25 18:31:34 <ikke> depends="!package"
2020-06-25 18:33:05 <yann-kaelig> ok, and something like backup= ? to preserve some configuration files after an update
2020-06-25 18:33:46 <ikke> that's the default
2020-06-25 18:33:55 <ikke> everything in /etc at least
2020-06-25 18:34:13 <yann-kaelig> ok thx
2020-06-25 18:48:46 <yann-kaelig> is there a list of group ID that are not used in alpine linux ?
2020-06-25 18:49:16 <ikke> >1000? :)
2020-06-25 18:49:29 <c705> yann-kaelig: just lok in /etc/group for a list of what is used
2020-06-25 18:49:31 <yann-kaelig> oups, sry, my bas system group I mean
2020-06-25 18:49:34 <yann-kaelig> bad
2020-06-25 18:50:16 <ikke> https://gitlab.alpinelinux.org/alpine/aports/-/blob/master/main/alpine-baselayout/group
2020-06-25 18:51:52 <yann-kaelig> ikke, thx, but for exampel in this lsit there is no such apache group for example 82, and others group ID that could be used by another program
2020-06-25 18:52:07 <yann-kaelig> later
2020-06-25 18:53:57 <yann-kaelig> well 24 seems free
2020-06-25 18:53:59 <yann-kaelig> thx
2020-06-25 18:54:45 <ikke> best to just let the system chooise one honestly
2020-06-25 19:43:26 <jammers> trying to install alpine std to disk, process completes, but only syslinux files in boot partition, so epic fail. no kernel etc
2020-06-25 19:45:59 <jammers> how do I NOT download/install all 700mb of firmware/drivers during this process.?  Dont need them :(
2020-06-25 19:46:16 <ikke> make sure linux-firmware-none is installed
2020-06-25 19:47:09 <ikke> but no idea if you can prevent it during installation
2020-06-25 19:50:38 <jammers> is there VM images already made for 3.12 x86 somewhere? its during install that it loads all firmware/drivers
2020-06-25 19:52:38 <mps> does linux-virt kernels depends on firmware-*
2020-06-25 19:53:18 <ikke> Doesn't seem so
2020-06-25 19:53:34 <ikke> nop
2020-06-25 19:53:35 <ikke> e
2020-06-25 19:53:58 <mps> thought so, thanks for confirming
2020-06-25 19:54:05 <feuerteufel> That's the one I use for VM installations:
2020-06-25 19:54:15 <feuerteufel> http://dl-cdn.alpinelinux.org/alpine/v3.12/releases/x86/alpine-virt-3.12.0-x86.iso
2020-06-25 19:54:49 <feuerteufel> As far as I know thre is no Firmware in
2020-06-25 19:55:21 <mps> feuerteufel: right, just checked my VM
2020-06-25 19:56:11 <artok> hm.. people using alpine on rpi: how do you format your sd card ?
2020-06-25 19:58:38 <feuerteufel> I do
2020-06-25 19:59:32 <feuerteufel> I format the card like the howto say's :-))
2020-06-25 19:59:51 <Alina-malina> artok, use balena or etcher
2020-06-25 20:00:35 <artok> Alina-malina: for alpine?
2020-06-25 20:00:52 <artok> that is tar.gz of the boot
2020-06-25 20:00:54 <artok> not image
2020-06-25 20:01:16 <mps> artok: mkfs.$preferdFS :)
2020-06-25 20:01:41 <jammers> @feuerteufel thanks, that worked. dont know why the std iso borked up the boot portion of the install, firmware aside.
2020-06-25 20:02:29 <artok> well what is that prefered.. as if I use my SD's, I get that problem I've told many times, need to copy kernel and initramfs into root and edit config.txt =)
2020-06-25 20:03:26 <mps> artok: on SD/emmc/mmc cards I use mostly f2fs
2020-06-25 20:03:56 <mps> though for RPis you must have boot as FAT
2020-06-25 20:04:10 <mps> boot partition
2020-06-25 20:04:48 <feuerteufel> Yes, like it say's here ..:
2020-06-25 20:04:55 <feuerteufel> https://wiki.alpinelinux.org/wiki/Raspberry_Pi
2020-06-25 20:05:18 <artok> still getting different outcomes than people seem to have =)
2020-06-25 20:05:23 <artok> oh well, try again
2020-06-25 20:06:29 <feuerteufel> mps: Did you find a way for updating the kernel on RPI?
2020-06-25 20:07:16 <mps> feuerteufel: iirc, it works but I don't have time to check now
2020-06-25 20:07:40 <mps> you mean is sys mode?
2020-06-25 20:08:11 <feuerteufel> artok: I did it exaktly the way the wiki say's, exept I left out the wifi stuff becourse I don't need it
2020-06-25 20:08:47 <feuerteufel> mps: No, the ram mode
2020-06-25 20:09:10 <mps> ah, didn't even tried that
2020-06-25 20:09:11 <feuerteufel> "the normal mod"?
2020-06-25 20:10:25 <jammers> any better screen resolution in virt? something akin to vga=791
2020-06-25 20:11:12 <ikke> jammers: check if nomodeset is as kernel parameter (it shouldn't)
2020-06-25 20:11:23 <ikke> though, I think that was never the case for virt
2020-06-25 20:11:57 <mps> feuerteufel: sorry, these days I'm short with time, but next time when I install RPi in 'ram' mode I will try upgrade kernel
2020-06-25 20:12:48 <feuerteufel> mps: No Prob ... I'll set a reminder again :-))
2020-06-25 20:13:02 <mps> :)
2020-06-25 20:13:40 <mps> jammers: you can set different resolution with qxl driver, but I forgot exact parameters
2020-06-25 20:15:34 <mps> '-display sdl -vga qxl'
2020-06-25 20:47:25 <c705> what's going on with grub/ is it conformed upstream issues?
2020-06-25 20:47:30 <c705> confirmed, even
2020-06-26 00:45:27 <quesker> is the cron from busybox?  does it support RANDOM_DELAY?
2020-06-26 00:49:00 <quesker> oh I have perl in this docker anyway I can use perl to sleep for a random time
2020-06-26 02:33:16 <shubha> Hello
2020-06-26 02:33:46 <shubha> Is there a way to kill tcp connections in alpine linux?
2020-06-26 02:36:44 <maldridge> firewall?
2020-06-26 02:36:55 <maldridge> standard mechanisms that are available to Linux
2020-06-26 02:39:13 <Shubh> actually I want to test my application which connects to a database and its behaviour after killing those connections (I want to replicate a network trouble kind of test where application can not connect to DB )
2020-06-26 02:39:41 <maldridge> ah, you probably want tc
2020-06-26 02:39:46 <maldridge> you can inject an unreliable network there
2020-06-26 02:42:20 <Shubh> I was looking for something like tcpkill, but looks like alpine does not support it ,  sorry I'm new to alpine linux wanted to know other methods
2020-06-26 02:43:04 <Shubh> it is docker app running on k8s and I dont want to change anything in docker image
2020-06-26 07:03:52 <markand> even though openrc isn't that fast, it still boots faster than Fedora 32
2020-06-26 08:42:04 <mps> feuerteufel: installing/upgrading kernel on RPi zero W running in 'RAM' mode doesn't work in my test
2020-06-26 08:46:33 <korner> has there been some known issue with xorg on alpine and dgraphics overall? i use uhd 620 as rendering GPU and no matter what i could not ever get into the DE
2020-06-26 08:46:38 <korner> maybe display manager if i got lucky
2020-06-26 09:14:28 <artok> mps oh you tested that already
2020-06-26 09:14:59 <mps> artok: yes
2020-06-26 09:15:55 <mps> maybe it can be in offline, i.e. mounting mmc on other computer and put new kernel there
2020-06-26 10:59:11 <nub> hi
2020-06-26 11:11:48 <yann-kaelig> Hi
2020-06-26 11:14:59 <yann-kaelig> So I have created my apk and everythings works as expected, then I decided to test and install my packages on another alpine installation like I was another user who want to intall them, but I get an unstrusted signature. How shoudl I proceed to truste them ?
2020-06-26 11:16:00 <yann-kaelig> How the users who want to install my packages should proceed
2020-06-26 11:17:33 <ikke> You need to copy your public key in /etc/apk/keys (with the same name as they were created with)
2020-06-26 11:17:48 <ikke> which is most likely located in ~/.abuild/
2020-06-26 11:21:05 <yann-kaelig> ikke: Hi. Ok, yes I see my pub key in this folder . Is there another method than copying my pub key manually ? I mean I don't have to copy the key manually when I'm using alpine repository, so how are they available on my system ?
2020-06-26 11:28:24 <ikke> they are part of the iso
2020-06-26 11:28:39 <ikke> https://pkgs.alpinelinux.org/packages?name=alpine-keys&branch=edge&arch=x86_64
2020-06-26 11:29:09 <yann-kaelig> ok, interesting, thx
2020-06-26 13:36:10 <yann-kaelig> What is the alternative agetty used in alpine linux for opening a tty and get a prompt login ?
2020-06-26 13:37:30 <ikke> alternative in what sense?
2020-06-26 13:37:58 <ikke> either getty from bb, or agetty from util-linux
2020-06-26 13:38:17 <yann-kaelig> I need agetty for my work, but there is no such software. I foudn that a part of ubase which is in testing alpine repo
2020-06-26 13:38:36 <ikke> util-linux provides agetty
2020-06-26 13:38:37 <yann-kaelig> ok
2020-06-26 13:39:09 <yann-kaelig> well, so perhaps I made a mistake somewhere in my work
2020-06-26 13:39:50 <ikke> https://pkgs.alpinelinux.org/contents?file=*getty*&path=*bin&name=&branch=edge&arch=x86_64
2020-06-26 13:40:20 <mps> hmm, simple as 'apk search agetty'
2020-06-26 13:40:43 <mps> or 'apk search cmd:agetty'
2020-06-26 13:43:48 <yann-kaelig> which agetty return nothing, but apk info -i util-linux is well installed. I also tried find / -name "*agetty*" , nothing
2020-06-26 13:47:09 <mps> apk info -i util-linux, is not for that
2020-06-26 13:47:29 <mps> apk info -e util-linux
2020-06-26 13:47:47 <yann-kaelig> I tried to be sure that util-linux is well installed, but ls: /sbin/agetty: No such file or directory
2020-06-26 13:48:02 <mps> if answer is empty it means package is not installed
2020-06-26 13:48:46 <yann-kaelig> ha, so what does that mean in this case when I run apk info -i util-linux and  I see > util-linux-2.35.2-r0 installed size:
2020-06-26 13:48:53 <mps> so, apk add util-linux
2020-06-26 13:49:11 <yann-kaelig> installed size doesn't mean it's installed ?
2020-06-26 13:49:26 <mps> no, '-i' list install_if packages
2020-06-26 13:50:13 <mps> right,  installed size show how much packages takes when installed
2020-06-26 13:50:31 <mps> not that the package is actually installed
2020-06-26 13:50:58 <yann-kaelig> ok, sry my fault then
2020-06-26 13:56:05 <mps> btw, *getty is always in /sbin
2020-06-26 13:56:26 <mps> except for modern /usr only distros :D
2020-06-26 13:56:45 <maxice8> it is in /sbin as well in usr-merge distros
2020-06-26 13:57:00 <maxice8> -/sbin is a symlink to /usr/sbin or /usr/bin
2020-06-26 13:57:03 <mps> symlink?
2020-06-26 13:57:13 <maxice8> So you can access *getty via /sbin/getty /usr/sbin/getty or /usr/bin/getty
2020-06-26 13:57:19 <mps> heh, my intuition works :)
2020-06-26 14:34:03 <yann-kaelig> So, there is nothing like optdepends in alpine linux.
2020-06-26 14:34:43 <yann-kaelig> optional dependencies
2020-06-26 14:35:28 <ikke> No
2020-06-26 14:40:02 <yann-kaelig> well, $pkgname.post-install is an idea
2020-06-26 14:40:46 <yann-kaelig> with a message about optionnal packages depending the need of the useer
2020-06-26 14:41:18 <wsinatra_> You could make a meta package
2020-06-26 14:41:22 <ikke> A thread on the alpine mailing list about it: https://lists.alpinelinux.org/~alpine/devel/%3C883dca1a-b7f3-6137-059d-f561ef22c126%40cosmoborsky.com%3E
2020-06-26 14:43:49 <Hello71> imo the arch system is fine
2020-06-26 14:55:26 <mps> hm, there is 'policy' in alpine, *minimal dependencies*
2020-06-26 15:00:09 <yann-kaelig> I think in my opinion 'optional' mean the system run without the optional packages even if the user experience is not fully satisfied ( for example XFCE case without the icons)  UNTIL the user decide to install the optional package(s).
2020-06-26 15:00:12 <yann-kaelig> What was optional for the system before installation is no longer for the system of the user who decided to install the package and whatever happens later it must be treated as if it were part of the system.
2020-06-26 15:08:02 <yann-kaelig> let's sya i have a package where btrfs, lvm2, zfs are optional. When the user decide to install one or more of these package, what was optional becomes 'mandatory' and when the user deicde to remove the package with the optionals packages,  it should be impossible because it could simply prevent the proper functioning of the user system.
2020-06-26 15:10:15 <yann-kaelig> in fact, what was optional from the point of view of the packager, could be mandatory for the user after he decide to install on of these packages
2020-06-26 15:19:09 <mps> imo, that doesn't make sense
2020-06-26 15:19:46 <mps> I want freely to add/remove packages, no distro should decide for me
2020-06-26 15:20:32 <tehcloud> I like this policy
2020-06-26 15:22:30 <mps> and, creating metapackages with apk is really easy, if someone need them
2020-06-26 15:23:30 <yann-kaelig> mps: this is exactly what I'm trying to explain :)  But from the link posted above "Natanael" wonders about " What should happen when you do `apk del pkg`? should it keep the optional deps installed or should they be deleted?
2020-06-26 15:23:39 <mps> example, apk info xfce4
2020-06-26 15:23:55 <mps> and, apk info -R xfce4
2020-06-26 15:24:52 <mps> if you delete metapackage all dependencies will be deleted
2020-06-26 15:25:30 <mps> except if one/some of them are 'added' by 'apk add'
2020-06-26 15:26:31 <mps> or I don't understand question
2020-06-26 15:40:42 <yann-kaelig> Case1) User install A with optional B, user want to remove A but removing B can break the system, so A can not be removed. User has to manually remove first B knowing what he's doing before A can be removed. Case 2) User install A and optionnal B, user want to remove A, A can be removed but B is not.A note inform the user that B is an orphan of A, the user has the choice to delete B manually in the same way as he chose to
2020-06-26 15:40:43 <yann-kaelig> install it.
2020-06-26 15:45:29 <mps> then B must depend on A
2020-06-26 15:45:47 <mps> iiuc your question
2020-06-26 15:52:58 <yann-kaelig> It is not obvious indeed, or perhaps report to the user if A is goign to be removed that an optional package B has been installed and removing A while B has been installed can 'break' the system.  Then it is up to the user to fix his system if something is broken :)
2020-06-26 15:56:06 <yann-kaelig> You are removing A while B is installed as optionnal package. Take a coffee and things about it again" :D
2020-06-26 16:01:10 <mps> well, apk requires from user to think a little
2020-06-26 16:01:31 <wsinatra_> Right now if you attempt to do something like apk del lua, which weechat-lua is installed apk updates /etc/apk/world but doesn't remove the package to prevent system breakage
2020-06-26 16:01:46 <wsinatra_> and in so doing reports that fact
2020-06-26 16:02:18 <wsinatra_> so if a user has an dependency (even if that's "optional" through a metapackage) you should have the same result of getting a message saying "do due diligence"
2020-06-26 16:36:45 <yann-kaelig> Well. Alpine linux with S6 and 66 tools. http://dpaste.com/0HCZEJR
2020-06-26 16:37:12 <maxice8> classic is a longterm ?
2020-06-26 16:40:42 <yann-kaelig> Classic service is pure s6 service, meaning no database, no compilation at all. You can compare it to runit service. This kind of service do not support inter dependencies like tty
2020-06-26 16:41:02 <maxice8> Fancy
2020-06-26 16:41:10 <yann-kaelig> https://web.obarun.org/software/66/
2020-06-26 16:41:36 <yann-kaelig> If you want more info about 66, ask eric the devs of 66 on #obarun
2020-06-26 16:41:50 <maxice8> Sorry I have no interest besides OpenRC
2020-06-26 16:42:26 <yann-kaelig> I know that lot of people like openrc :) No problem at all
2020-06-26 18:37:18 <feuerteufel> mps: Thank's! So you are able to verify my problem. You see any way out of it?
2020-06-26 18:45:07 <mps> feuerteufel: I don't have much experience with 'run from ram' mode, but some packages I tested installation worked fine
2020-06-26 18:45:27 <mps> kernel must be somewhat specific
2020-06-26 18:45:41 <feuerteufel> Same I did
2020-06-26 18:45:56 <mps> for now I have no idea how to fix it
2020-06-26 18:46:12 <feuerteufel> Install a Package is no problem
2020-06-26 18:46:32 <mps> yes
2020-06-26 18:47:09 <feuerteufel> Do you know if this Prob is only in V. 3.12?
2020-06-26 18:47:38 <mps> not sure, didn't tried kernel install on previous releases
2020-06-26 18:48:22 <feuerteufel> It's my first rpi, I don't have any experience
2020-06-26 18:49:33 <mps> also I have only one RPi, and I borrowed it actually, it is not mine
2020-06-26 18:49:46 <feuerteufel> Or exactly my first rpi whith AL on it ...
2020-06-26 18:50:57 <mps> maybe someone with other arches  who run from ram can shed some light on this
2020-06-26 18:52:21 <feuerteufel> mps: It's not a big deal ... it's a playground, it's a rpi 1b, I don't hav eany jab for that one :-))
2020-06-26 18:53:12 <mps> ok, but anyway would be nice to find solution for problem
2020-06-26 18:53:32 <feuerteufel> 
ACTION Ups, to hot ... better dont Write

2020-06-26 18:53:44 <mps> also, I don't use RPi for anything serious, just for testing installations
2020-06-26 18:54:45 <mps> I use other arm SBCs for serious things :)
2020-06-26 18:59:33 <feuerteufel> SBC = "Sparky Single Board"?
2020-06-26 19:00:27 <ikke> Single board computer
2020-06-26 19:01:26 <feuerteufel> Ah, thank's ikke!
2020-06-26 19:01:59 <feuerteufel> My tranlation was the first hit @ go....
2020-06-26 19:08:41 <qman__> well, it's not a small block chevy
2020-06-26 19:31:06 <feuerteufel> qman__: You most likely right :-))
2020-06-26 19:41:08 <mps> feuerteufel: 'ls -l /lib/modules' => rwxrwxrwx    1 root     root            17 Jan  1  1970 /lib/modules -> /.modloop/modules
2020-06-26 19:41:46 <mps> 'mount | grep modloop' => /dev/loop0 on /.modloop type squashfs (ro,relatime)
2020-06-26 19:42:13 <mps> read-only FS
2020-06-26 19:44:17 <feuerteufel> So I have to change something?
2020-06-26 19:44:32 <feuerteufel> It's the same here ...
2020-06-26 19:44:49 <mps> I don't know
2020-06-26 19:44:53 <ikke> modloop is read-only, yes
2020-06-26 19:44:58 <mps> there is where I stop
2020-06-26 19:45:04 <ikke> it's a mount backed by a file
2020-06-26 19:45:06 <mps> this*
2020-06-26 19:45:46 <mps> yes, I understand quite fine loop mounts
2020-06-26 19:46:35 <mps> but have no idea will how it can be mounted 'rw' on run-from-ram system
2020-06-26 19:47:04 <mps> 'mount -o remount,rw /.modloop' doesn't work
2020-06-26 19:51:40 <mps> huh, but I don't have time now for this, have to forward port some patches for my local st term
2020-06-26 19:52:06 <ikke> it's a service
2020-06-26 19:52:09 <ikke> modloop
2020-06-26 19:53:17 <mps> what? stop it?
2020-06-26 19:55:39 <ikke> that would unmount the modloop
2020-06-26 19:59:08 <mps> aha, I see, ebegin "OverlayFS detected, mounting modloop rw"
2020-06-26 20:29:58 <feuerteufel> I'm sorry, but I'm not able to follow up what that means for me?
2020-06-26 20:30:30 <ikke> Not sure what you want to achieve
2020-06-26 20:34:04 <feuerteufel> I'm trying to update the kernel on a rps in ram mode, it dosn't work and mps try to help ...
2020-06-26 20:34:34 <ikke> I don't think a normal system upgrade works for that
2020-06-26 20:35:33 <feuerteufel> It worked exept the kernel
2020-06-26 20:35:46 <ikke> yes, that's the part that does not work
2020-06-26 20:35:51 <mps> feuerteufel: listen the man with experience, i.e. ikke :)
2020-06-26 20:36:00 <feuerteufel> I went from 3.11 to 3.12
2020-06-26 20:36:11 <ikke> that's quite normal for run-from-ram systems
2020-06-26 20:36:18 <ikke> you need to update the boot medium
2020-06-26 20:36:27 <ikke> but in case of rpis, the boot medium is the main storage
2020-06-26 20:36:33 <ikke> well, still run-from-ram
2020-06-26 20:36:43 <ikke> I'm not sure if setup-bootable works for rpi images
2020-06-26 20:36:50 <ikke> but that's normally what you need to do
2020-06-26 20:37:20 <ikke> the new modloop comes from the iso
2020-06-26 20:37:35 <ikke> as well as the kernel
2020-06-26 20:38:43 <feuerteufel> Ah ok, so I have to burn a new card?
2020-06-26 20:39:01 <ikke> yes, which setup-bootable does
2020-06-26 20:39:25 <feuerteufel> Ok like the wiki say's
2020-06-26 20:40:30 <feuerteufel> Is there a way to save the installed packages, so I'm able to reinstall them?
2020-06-26 20:40:48 <ikke> the apkovl is kept
2020-06-26 20:41:14 <ikke> so you don't lose anything
2020-06-26 20:41:29 <ikke> (as long as it's been committed)
2020-06-26 20:45:36 <feuerteufel> You mean this one "lbu ci -dv", right?
2020-06-26 20:49:46 <feuerteufel> ikke: So when I start "setup-bootable" I'll get the latest kernel?
2020-06-26 20:51:30 <ikke> you need to pass it an iso
2020-06-26 20:51:56 <ikke> (note that I have no experience with alpine on RPI's at all(
2020-06-26 20:52:44 <feuerteufel> That's what I'm trying to get :-))
2020-06-26 20:53:01 <ikke> Understood
2020-06-26 20:53:07 <ikke> The wiki mentions something about it
2020-06-26 20:53:14 <ikke> "Beware, though, that the contents of /boot will be ignored when the Pi boots: it will use the kernel, initramfs, and modloop images from the FAT32 boot partition. To update the kernel, initfs or modules, you will need to manually (generate and) copy these to the boot partition or you could use bind mount so that manually copy the files to boot partition is not needed. "
2020-06-26 20:56:41 <feuerteufel> Yes, I have seen that, but I'm not sure how to do this ...
2020-06-26 21:48:09 <artok> ehheh
2020-06-26 21:48:33 <artok> support.russianbridesnetwork.com ukk0.internationalconspiracy.org
2020-06-26 21:48:49 <artok> what a nice list of domains that chrony connects for ntp =D
2020-06-26 21:53:19 <wrycode> lol those are default servers?
2020-06-26 21:53:52 <TBB> well the owners of those domains truly know what time it is :P
2020-06-26 21:58:36 <artok> they are from pool.ntp.org
2020-06-26 23:34:09 <yann-kaelig> apk it's a nice tool
2020-06-26 23:34:15 <yann-kaelig> well
2020-06-26 23:34:23 <yann-kaelig> thank you for your help
2020-06-26 23:34:25 <yann-kaelig> cu later
2020-06-27 00:43:33 <p3lim> Anyone mind flagging coredns? 1.7.0 came out on the 18th, page just reloads when I submit
2020-06-27 00:46:22 <p3lim> And alertmanager, 0.21 came out the 17th
2020-06-27 01:39:50 <maldridge> reloads for me as well, I think the pkg site may be borken
2020-06-27 03:52:24 <abenz> I'm using diskless mode, my amdgpu won't initalize, upon searching online I think I need to set some kernel boot parameters
2020-06-27 03:52:49 <abenz> reading the wiki, I think I need to use update-extlinux?
2020-06-27 03:53:25 <abenz> any links on how to add a simple kernel bootarg (its amdgpu.dc=0 btw)
2020-06-27 03:54:17 <abenz> btw I see a similar problem here: https://gitlab.alpinelinux.org/alpine/aports/-/issues/11081
2020-06-27 03:54:24 <abenz> dont know if that guy ever solved it
2020-06-27 10:36:44 <lorailo> anybody there?
2020-06-27 10:36:56 <lorailo> I mean here
2020-06-27 10:37:06 <ikke> no, nothing to see here ;-)
2020-06-27 10:37:53 <lorailo> oh alright, it might be this ircII that doesn't show logged nicks
2020-06-27 10:38:05 <lorailo> I have a pressing issue
2020-06-27 10:38:25 <lorailo> is there another alpine channel we more ppl?
2020-06-27 10:38:33 <ikke> No
2020-06-27 10:38:37 <ikke> there are 411 idlers here
2020-06-27 10:38:48 <ikke> lorailo: it helps to actually ask a question
2020-06-27 10:39:01 <ikke> But it's weekend, so the amount of active people is limited
2020-06-27 10:39:07 <lorailo> I have a pressing issue I don't know how to solve it
2020-06-27 10:39:15 <ikke> You already mentioned :)
2020-06-27 10:39:35 <lorailo> i'll show you what I am talking about
2020-06-27 10:40:22 <lorailo> (EE) AIGLX error: dlopen of /usr/lib/xorg/modules/dri/swrast_dri.so failed (Error loading shared library /usr/lib/xorg/modules/dri/swrast_dri.so: No such file or directory)
2020-06-27 10:41:22 <lorailo> (II) AIGLX: Screen 0 is not DRI2 capable
2020-06-27 10:41:29 <ikke> https://pkgs.alpinelinux.org/contents?file=swrast_dri.so&path=&name=&branch=edge&arch=x86_64
2020-06-27 10:43:26 <lorailo> everything else is kinda working well, quite impressed with the 32MB iso size.
2020-06-27 10:44:04 <lorailo> at first I thought there must some mistake, no way a 32mb iso can boot anything.
2020-06-27 10:45:14 <lorailo> ikee I see
2020-06-27 10:45:21 <lorailo> but what about x86?
2020-06-27 10:46:13 <lorailo> found it
2020-06-27 10:46:20 <lorailo> logging on/off, brb
2020-06-27 10:59:58 <lorailo> heyr
2020-06-27 11:00:08 <lorailo> I see, now irssi displays logged nicks
2020-06-27 11:00:38 <lorailo> what, I am lagging )-:
2020-06-27 11:01:10 <lorailo> no dice with swrast )-:
2020-06-27 11:01:20 <lorailo> devuan bloated glbic works fine
2020-06-27 11:06:43 <lorailo> is there a different login manager? I can't start Xorg with startx
2020-06-27 11:07:42 <afontain_> there are lots of display managers
2020-06-27 11:07:45 <afontain_> the arch wiki is probably a good guide for them
2020-06-27 11:08:05 <lorailo> is there something lighter than xfce?
2020-06-27 11:08:14 <lorailo> lxde or lxqt or rox manager?
2020-06-27 11:08:43 <lorailo> afontain_: I followed the install guide and the xfce instructions.
2020-06-27 11:09:18 <afontain_> there's no lxqt
2020-06-27 11:09:25 <afontain_> IIRC, no lxde either
2020-06-27 11:09:33 <lorailo> ow crop
2020-06-27 11:09:40 <lorailo> now what
2020-06-27 11:09:44 <afontain_> you can simply use a window manager though
2020-06-27 11:10:05 <lorailo> argh, your right with that
2020-06-27 11:10:18 <afontain_> I just use sway, but some in the minimalist crowd don't like wayland
2020-06-27 11:10:54 <afontain_> at least, launching it is simple: just type "sway" in a console
2020-06-27 11:10:56 <Arahael> afontain_: What? Why don't the minimalist crowd like wayland?
2020-06-27 11:10:57 <lorailo> the problem with *box is that st/rxvt/xterm don't work
2020-06-27 11:11:08 <lorailo> afontain_: how to install it?
2020-06-27 11:11:24 <afontain_> (though dbus-run-session sway may be better)
2020-06-27 11:11:31 <afontain_> apk add sway
2020-06-27 11:11:33 <afontain_> simple
2020-06-27 11:11:40 <lorailo> afontain_: and how to start it?
2020-06-27 11:12:00 <afontain_> > at least, launching it is simple: just type "sway" in a console
2020-06-27 11:12:12 <lorailo> afontain_: as user?
2020-06-27 11:12:18 <afontain_> > (though `dbus-run-session sway` may be better, so you get dbus support)
2020-06-27 11:12:19 <afontain_> yes
2020-06-27 11:13:08 <afontain_> Arahael: the suckless kind of minimalist people probably aren't fond of wayland compositors
2020-06-27 11:13:27 <Arahael> afontain_: They do realise that the composition isn't being done on the CPU?
2020-06-27 11:13:39 <afontain_> or the KISS Linux guys
2020-06-27 11:13:56 <Arahael> *sigh*, I guess they don't realise that X11 is insanely complex.
2020-06-27 11:14:35 <lorailo> Arahael: x11 has been modularized
2020-06-27 11:15:07 <Arahael> Yes, but it's still big and the protocol is still a bit crazy - especially when you add DRM and opengl and all that to the mix.
2020-06-27 11:15:52 <afontain_> well, in X.org's defense, for postmarketOS, it tend to work fairly well on the old manufacturers kernel
2020-06-27 11:16:13 <afontain_> while most wayland UI require DRI
2020-06-27 11:16:18 <afontain_> (/DRM)
2020-06-27 11:16:21 <Arahael> X.org was a huge improvement.  (Actually, to be honest, I've yet to use wayland seriously - I mainly use my linux as a headless system)
2020-06-27 11:17:45 <afontain_> (but, hey, if that gives people a huge incentive to mainline their device, it's potentially a good thing :P)
2020-06-27 11:18:12 <lorailo> we gonna need minimum as hell with the onslaught of IOTs
2020-06-27 11:18:43 <lorailo> or else ppl wont' be able to configure their toasters.
2020-06-27 11:19:02 <lorailo> can you imagine having to toast a couple of bread loafs with the terminal?
2020-06-27 11:19:16 <ikke> The interface would at least be functional :P
2020-06-27 11:19:38 <ikke> imagine the automation possible :P
2020-06-27 11:19:41 <Arahael> lorailo: I'm not so sure - RaspberriPi is getting big. Plus, even a freaking CRO scope has windows integrated.
2020-06-27 11:19:59 <lorailo> i mean, ppl familiar with cli will do straigh away this: toast --help
2020-06-27 11:20:09 <Arahael> lorailo: I imagine the next 'toaster', if it's an "internet of things", will at least be as powerful as a Rpi.
2020-06-27 11:20:15 <afontain_> also smart watches are ridiculously underpowered
2020-06-27 11:20:37 <lorailo> then go on with: toast -n 2m
2020-06-27 11:20:57 <Arahael> afontain_: smart watches are pretty good, considering.
2020-06-27 11:21:05 <afontain_> Arahael: for my next desktop, I'll consider buying a toaster then
2020-06-27 11:21:07 <Arahael> It's ultimately a matter of power.
2020-06-27 11:21:19 <Arahael> afontain_: Good idea.  Or a fridge.
2020-06-27 11:21:25 <lorailo> so smartwatches, what X do they use?
2020-06-27 11:21:57 <Arahael> lorailo: Apple's uses WatchOS, and is probably pretty darn close to the framebuffer.
2020-06-27 11:22:09 <Arahael> I assume Android would be similar.
2020-06-27 11:22:24 <lorailo> interesting, brb with sway though
2020-06-27 11:22:25 <Arahael> Pebble, while they were still around, had a raw interface to teh display, I think?
2020-06-27 11:22:34 <afontain_> three years ago, someone did run doom on X.org on a smartwatch https://postmarketos.org/blog/2017/09/03/100-days-of-postmarketos/
2020-06-27 11:23:05 <afontain_> with Alpine at the core
2020-06-27 11:23:12 <ikke> echo 'toast -n 2m' | at 0800
2020-06-27 11:23:48 <Arahael> afontain_: Nice.  All this stuff makes me sad at how limited and restrictive Apple systems are. :(
2020-06-27 11:23:57 <Arahael> (Yet, that's precisely why I like them as a consumer)
2020-06-27 11:24:30 <afontain_> ikke oh
2020-06-27 11:24:32 <afontain_> that's great
2020-06-27 11:25:04 <afontain_> I'd have gone with cron
2020-06-27 11:25:11 <afontain_> 
ACTION adds a tool to his toolbelt

2020-06-27 11:26:01 <Arahael> Trouble with cron is that it doesn't stop you from toasting too many toasts at once.
2020-06-27 11:26:26 <Arahael> For that, add 'flock' to your toolbelt, too. :)
2020-06-27 11:26:51 <afontain_> "too many toasts"? is that a problem?
2020-06-27 11:27:08 <Arahael> It is when your toaster can only do two at a time.
2020-06-27 11:27:26 <afontain_> ah, yes
2020-06-27 11:27:45 <afontain_> I've seen it already, but have yet to use it though
2020-06-27 11:31:26 <lorailo> XDG_RUNTIME_DIR not set
2020-06-27 11:31:31 <lorailo> can't run sway
2020-06-27 11:31:41 <lorailo> does it need mesa gallium driver?
2020-06-27 11:32:17 <afontain_> install elogind and mesa and log in again
2020-06-27 11:32:30 <lorailo> got it
2020-06-27 11:51:09 <abenz> remounting diskless as rw, and modifying /boot/syslinux/syslinux.cfg (to append to kernel bootargs), I re-ran syslinux but upon reboot the changes don't take effect
2020-06-27 11:51:29 <abenz> any idea on what I should be doing to make my kernel boot parameters permanent?
2020-06-27 11:51:44 <abenz> this is on diskless
2020-06-27 11:52:25 <mps> afontain_: and that is minimalists system ;p
2020-06-27 11:53:30 <afontain_> well, technically elogind isn't needed
2020-06-27 11:53:46 <afontain_> `export XDG_RUNTIM_DIR=$(mktemp -d)` is good enough
2020-06-27 11:53:53 <mps> look what kiss linux people do
2020-06-27 11:54:00 <afontain_> *RUNTIME
2020-06-27 13:09:14 <Hello71> ugh, reminds me of bad freebsd articles that say to do XDG_RUNTIME_DIR=/tmp
2020-06-27 13:12:40 <afontain_> Hello71: why do you think I suggested elogind instead :P
2020-06-27 13:18:56 <earnestly> Ideally that directory would be 0700 as well
2020-06-27 13:19:50 <Hello71> apparently qt gets upset when XDG_RUNTIME_DIR isn't owned by the user
2020-06-27 13:20:01 <Hello71> so some guy in #freebsd was going on for hours about how to mount /tmp as their user
2020-06-27 13:20:29 <Hello71> apparently that's not easy on freebsd? otoh they seemed very confused in general, so idk if that's a good reference
2020-06-27 13:27:15 <earnestly> XDG_RUNTIME_DIR was added by Lennart and its list of requirements are absurd
2020-06-27 13:27:38 <earnestly> I wrote up the summary here: https://wiki.archlinux.org/index.php/XDG_Base_Directory#User_directories
2020-06-27 13:46:58 <sandro> Hello frens. Has anybody here tried to install generic ARM on pine64's RockPro?
2020-06-27 14:07:37 <antranigv> hey all, how are you? I see that the ejabberd package is very much outdated, I'd like to upgrade it. is there a documentation to get me started?
2020-06-27 14:08:49 <Cogitri> https://wiki.alpinelinux.org/wiki/APKBUILD_Reference
2020-06-27 14:08:57 <Cogitri> Although it's a bit outdated, hopefully I'll get to rewriting it soon
2020-06-27 14:09:38 <antranigv> Okay I see that it's very similar to Gentoo Ebuilds, hopefully I'll be able to pick it up fast.
2020-06-27 14:10:07 <antranigv> Cogitri thank you very much friend.
2020-06-27 14:10:09 <Cogitri> Hm, it's a bit similar to ebuilds, but simpler (as in less magic)
2020-06-27 14:11:15 <antranigv> personally prefer FreeBSD Ports, but the magic in Gentoo's Ebuilds was a lot, specially the Funtoo flavour.
2020-06-27 14:16:52 <mps> more like Arch PKGBUILD
2020-06-27 14:25:23 <Hello71> really the only similarity to ebuilds is using shell and variables and functions
2020-06-27 14:25:40 <maxice8> so basically 99% of ports systems
2020-06-27 14:26:22 <Hello71> fbsd ports use make though
2020-06-27 14:26:33 <Hello71> (poorly)
2020-06-27 14:33:01 <antranigv> Make is one of the reasons why I like it, comes with the system, everyone knows it, and really easy to debug.
2020-06-27 14:35:21 <Cogitri> When looking at fbsd makefiles it looks anything but easy
2020-06-27 14:35:34 <earnestly> (Because few know how to use it)
2020-06-27 14:36:15 <Cogitri> Make and generally autotools aren't quite my cup of tea
2020-06-27 14:37:48 <earnestly> Hello71: To be fair just about everyone uses make poorly, and from my experience it's split fairly evenly between simply not understanding it's purpose and trying to make it do more than it was designed to address
2020-06-27 14:38:41 <earnestly> its*
2020-06-27 14:46:14 <Hello71> ironically I think bsd ports have both
2020-06-27 14:48:42 <lerelel> hei
2020-06-27 14:49:15 <lerelel> weird thing with sway just a tree and an island withou any icons, menus, right click, left click.
2020-06-27 14:49:28 <lerelel> just like a museum painting.
2020-06-27 14:50:03 <lerelel> it's able to appear with virgl
2020-06-27 14:50:09 <lerelel> the xfce looks like a bug.
2020-06-27 14:50:34 <lerelel> xfce logs but stay with a black screen.
2020-06-27 15:19:32 <yann-kaelig> Hi
2020-06-27 15:38:41 <yann-kaelig> Someone use xen on alpine linux ?
2020-06-27 16:51:19 <Hello71> no, absolutely nobody, that's why there is xen, xen-dev, xen-doc, xen-hypervisor, and xen-libs packages since v3.3
2020-06-27 16:57:14 <artok> i've kept myself over an hour not to tell that joke =)
2020-06-27 16:58:28 <mps> yann-kaelig: guest or host? I use standard alpine virt as guest on some xen VM provider
2020-06-27 16:59:12 <mps> and some people reported here that they use xen as dom0, but sorry I'm not sure who and how much
2020-06-27 17:08:28 <yann-kaelig> mps: Hi. On host. I'm totally noob in this domain, Xen is something new for me.  I start reading the wiki, not too hard to get a first amd global idea of the software and others part like Xapi. About that I have not seen any distro, not to mention xcp-ng which offers packages for this toolstack.
2020-06-27 17:11:58 <yann-kaelig> I was on opennebula community before, but they have decided to closed the database upgrade part, so I have decided to move away from this project
2020-06-27 17:13:42 <dron2065> Hello to all, I have a question about installing Alpine Linux on UEFI machine with existing partitions. When I boot from USB, do setup-alpine (without installing on disk), do setup-disk on clean ext4 partition, mounted at /mnt, and try to boot installed system with previously installed rEFInd (selecting file vmlinuz-lts) - it fails loading with message like "can't mount some file on some mount
2020-06-27 17:13:48 <dron2065> point - file not found". What I do wrong?
2020-06-27 17:20:19 <mps> yann-kaelig: as I wrote I don't have much experience with xen so can't tell anything from my experience. but some people reported that they use alpine xen as dom0
2020-06-27 17:20:30 <mps> I prefer qemu
2020-06-27 17:24:07 <yann-kaelig> mps: I understand. That was the purpose of my question to get alpine users feedback in dom0 use case
2020-06-27 17:24:12 <yann-kaelig> thx mps
2020-06-27 17:26:01 <mps> yann-kaelig: np. better time to ask such questions would be on workdays and probably around CET time because most of devs and users of alpine is in this TZ
2020-06-27 17:29:05 <yann-kaelig> Right. I have studied all the other possibilities, including proxmoxd and I have chosen Xen which meets our needs
2020-06-27 18:52:09 <artok> should I now start learning how to use alpine from ram?
2020-06-27 18:52:12 <artok> eh
2020-06-27 18:54:56 <artok> mount
2020-06-27 18:54:58 <artok> ooh
2020-06-27 18:55:37 <artok> too many terminals
2020-06-27 19:37:36 <uplime> hi! I just installed fcron from apk, and tried to edit a crontab, but it can't set the real uid: https://clbin.com/ybUnG is there something else that needs to be done post install? I've been messing around with different permissions and capabilities for it, but I still get that same error every time
2020-06-27 19:57:42 <nacelle> artok: s/too many/not enough/ :-)
2020-06-27 19:58:43 <Hello71> uplime: check that fcrontab has +s and that you have not mounted the volume with nosuid
2020-06-27 20:00:08 <uplime> Hello71: volume is just rw,relatime. /usr/bin/fcrontab is ---s--s--x
2020-06-27 20:00:46 <Hello71> check it is owned by root:fcron and fcron is 65533
2020-06-27 20:02:46 <uplime> it looks to be: https://clbin.com/BnrFS
2020-06-27 20:07:30 <Hello71> well in this paste it is -s, and also ptrace blocks suid
2020-06-27 20:07:47 <Hello71> oh, the error changed too
2020-06-27 20:08:11 <Hello71> getent group fcron
2020-06-27 20:09:48 <uplime> oh, im dumb
2020-06-27 20:09:50 <uplime> thanks Hello71
2020-06-27 20:10:07 <uplime> I set CAP_UID but not CAP_GID
2020-06-27 20:39:25 <artok> nacelle: actually lack of keyboards seems to be teh problem
2020-06-27 20:39:55 <artok> TV connected to rpi3b+, monitor to rpi4, this laptop with external hdmi monitor
2020-06-27 20:40:05 <artok> only one keyboard
2020-06-27 21:01:21 <uplime> Hello71: last thing - https://clbin.com/UBWcZ do you know where it tries to create that uplime.orig file? It doesn't seem to be in /var/spool/fcron
2020-06-27 21:17:36 <artok> there.. freshly installed alpine in rpi4, rpi3, hp620 laptop
2020-06-27 21:35:23 <abenz> found all my answers with vi /sbin/setup-bootable! Didn't occur to me to check it first, all steps are documented with comments.. unlike the wiki :)
2020-06-27 21:35:38 <abenz> cheers
2020-06-27 21:37:49 <nacelle> 
ACTION hands artok more keyboards

2020-06-27 21:37:52 <nacelle> i know that feeling
2020-06-28 02:14:57 <DaRock> hey guys
2020-06-28 02:15:18 <DaRock> I'm trying to work my head around my first permanent alpine install on rpi - I've been playing in vms so far to get some experience with it
2020-06-28 02:16:20 <DaRock> so I have to boot the rpi from the sd card, and run alpine diskless, but I'm thinking of the lbu and package cache as well as home on a usb hdd
2020-06-28 02:16:57 <DaRock> and possibly a redundancy for lbu and cache using nfs
2020-06-28 02:19:09 <DaRock> so if I partition hdd with 2 partitions: one for cache and one for home. Then mount /home (obviously), then mount the cache dir where?
2020-06-28 02:20:08 <DaRock> and as admin I'd have to take care of lbu commit - as an ordinary user is this necessary?
2020-06-28 02:21:02 <DaRock> and where is the configuration that determines these locations permanently?
2020-06-28 03:08:27 <abenz> DaRock: https://wiki.alpinelinux.org/wiki/Create_a_Bootable_USB
2020-06-28 03:09:33 <abenz> in my setup, I have 2 partitions on my usb: 1 is vfat for alpine and configs and cache, and 2 is ext4 for stuff that I need (and not related to alpine)
2020-06-28 03:25:21 <DaRock> thx abenz. I'm trying figure out the best solution for a desktop os for a user. Learning a new distro in the process :-)
2020-06-28 03:27:48 <abenz> that makes two of us buddy.. I have a system where I cant use HDDs, so alpine fits with its diskless mode
2020-06-28 07:25:06 <abenz> I see with df -h, that alpine allocates a lot of RAM for /dev/shm
2020-06-28 07:25:58 <abenz> am I right in assuming to change those defaults I need to write entries for shm and / size in /etc/fstab ?
2020-06-28 07:26:05 <ikke> no
2020-06-28 07:28:08 <ikke> You need to provide rootflags to the kernel parameters
2020-06-28 07:29:30 <ikke> rootflags=size=...
2020-06-28 07:29:55 <abenz> kernel parameters as in syslinux.cfg and grub.cfg ?
2020-06-28 07:30:39 <ikke> yes, the kernel cmline
2020-06-28 07:30:46 <ikke> cmdline
2020-06-28 07:31:27 <ikke> for run-from-ram it would just be a matter of changing it on the boot media
2020-06-28 07:31:57 <ikke> https://gitlab.alpinelinux.org/alpine/mkinitfs/-/blob/master/initramfs-init.in#L561
2020-06-28 07:32:00 <abenz> yes this is for diskless mode
2020-06-28 07:34:05 <abenz> ikke: what would be a reasonable size to leave for alpine? this will be for a buildbot machine. compiling stuff, sending binaries, then erasing. Repeatedly
2020-06-28 07:34:43 <abenz> 64GB system ram, I need around 53GB for tmpfs (/)
2020-06-28 07:35:05 <abenz> does that seem reasonable or too ambitious?
2020-06-28 07:38:34 <arahael2> That's not a normal amount of RAM.
2020-06-28 07:39:05 <arahael2> I use an SSD and I let my RAM cache it.
2020-06-28 07:39:18 <afontain_> lol
2020-06-28 07:39:30 <arahael2> You can always wipe the storage
2020-06-28 07:40:45 <abenz> I have a 5400rpm drive..
2020-06-28 07:47:07 <abenz> Iit worked!
2020-06-28 07:47:12 <abenz> s,lit,it
2020-06-28 07:47:41 <arahael2> Curious, I'd have invested in an SSD. :)
2020-06-28 07:50:04 <abenz> arahael2: the system will cross compile for embedded systems, thousands of small files. I dont know how much caching will help prolong an ssds' life, but I figured I do everything in a ramdisk, copy all thats needed from ro mount and dont touch disk again until next reboot
2020-06-28 07:52:07 <arahael2> To be honest, I haven't worried about write amplification in years.
2020-06-28 08:56:03 <ryan-c> The GNU screen package is slightly broken - the config is at /etc/screenrc, but the binary looks at /usr/etc/screenrc
2020-06-28 08:56:09 <ryan-c> Where can I file a bug report?
2020-06-28 08:57:34 <mps> ryan-c: are you sure? ./configure have '--sysconfdir=/etc'
2020-06-28 08:57:39 <ryan-c> yes
2020-06-28 08:57:44 <ryan-c> i'm on 3.12.0
2020-06-28 08:57:58 <ryan-c> settings in /etc/screenrc have no effect
2020-06-28 08:58:13 <ryan-c> i ran strace on it, and saw it's looking at /usr/etc/screenrc
2020-06-28 08:58:55 <mps> yes, you are right
2020-06-28 08:59:16 <mps> strings with grep shows this
2020-06-28 08:59:25 <ryan-c> copying the config to /usr/etc/screenrc does work
2020-06-28 08:59:40 <mps> ofc, needs to be fixed
2020-06-28 09:00:27 <mps> I can look after breakfast to fix it but you can fill bug report on gitlab.a.o aports section
2020-06-28 09:00:52 <mps> hmm, gitlab. alpinelinux.org
2020-06-28 09:01:14 <mps> algitbot: you still not wake up :)
2020-06-28 09:02:21 <ryan-c> i'm looking at the APKBUILD on aports and it does indeed have that configure flag
2020-06-28 09:02:26 <ryan-c> idkwtf
2020-06-28 09:02:45 <mps> will look soon
2020-06-28 09:03:29 <mps> well, maybe this ' --with-sys-screenrc=path to the global screenrc file'
2020-06-28 09:04:27 <ryan-c> wat
2020-06-28 09:04:53 <ryan-c> that does sound like it might be the issue
2020-06-28 09:06:32 <mps> yes, 'strings screen-4.8.0 | grep screenrc' => /etc/screenrc
2020-06-28 09:06:46 <mps> tested on local build
2020-06-28 09:06:49 <mps> right now
2020-06-28 09:07:19 <mps> need some other fixes for creating package in package() function
2020-06-28 09:07:46 <mps> but have to go for breakfast first :)
2020-06-28 09:52:40 <ryan-c> hmm, did you mean --with-system_screenrc
2020-06-28 09:53:18 <ryan-c> curiously, the screen source specifies the default as /etc/screenrc
2020-06-28 10:25:01 <mps> ryan-c: no, exactly '--with-sys-screenrc=/etc/screenrc'
2020-06-28 10:31:36 <ryan-c> okay
2020-06-28 10:32:46 <mps> but I'm looking to enable some other options, --enable-telnet, --enable-rxvt_osc maybe
2020-06-28 10:33:27 <mps> and maybe '--with-socket-dir=/run/screens'
2020-06-28 10:33:53 <ryan-c> my workaround for now is to drop a file at /usr/etc/screenrc that just has 'source /etc/screenrc' in it
2020-06-28 10:34:53 <ryan-c> i was making an aws ami that has my prefered configs installed at the systemwide locations
2020-06-28 10:34:54 <mps> did you opened but report
2020-06-28 10:35:08 <ryan-c> where do i open the bug report?
2020-06-28 10:35:35 <mps>  gitlab. alpinelinux.org section aports
2020-06-28 10:35:36 <ryan-c> i see the gitlab, which project?
2020-06-28 10:35:43 <mps> aports
2020-06-28 10:36:37 <mps> you can login with gitlab or github account if you have any of them, else you have to register
2020-06-28 10:37:15 <mps> hmm, have to register self (I need to fix keyboard)
2020-06-28 10:40:44 <ryan-c> https://gitlab.alpinelinux.org/alpine/aports/-/issues/11697
2020-06-28 10:55:27 <mps> !9698
2020-06-28 10:55:55 <mps> oh, algitbot really doesn't work here
2020-06-28 10:56:20 <mps> ryan-c: here is merge request https://gitlab.alpinelinux.org/alpine/aports/-/merge_requests/9698
2020-06-28 10:56:33 <artok> what
2020-06-28 10:57:25 <artok> didn't it before?
2020-06-28 10:57:56 <mps> ryan-c: here you can download screen for your arch https://gitlab.alpinelinux.org/mps/aports/pipelines/24447, if you want to test it before it will be in repos
2020-06-28 10:58:03 <mps> artok: iirc, it worked
2020-06-28 11:37:15 <josh[m]3> I delete a package, it shows no errors, but the package hasn't been deleted. I'm not sure why. Does anyone know?
2020-06-28 11:37:15 <josh[m]3> https://paste.ubuntu.com/p/zDcB9jrdmT/
2020-06-28 11:57:14 <ryan-c> mps: I don't have a host handy with a build toolchain on it
2020-06-28 12:00:42 <ryan-c> mps: okay, i downloaded the build artifact apk, that does fix it
2020-06-28 12:04:39 <PureTryOut[m]> josh: pro-tip: since this is a majority IRC channel, try to not use Matrix-specific features like multiline messages. It doesn't propagate through nicely
2020-06-28 12:56:59 <afontain_> josh: can you try adding the package and removing it again?
2020-06-28 12:58:49 <ikke> josh[m]3: it most likely means that something still depends on that package
2020-06-28 13:02:33 <mps> ryan-c: yes, I forgot to tell that you can download artifacts
2020-06-28 13:03:18 <mps> so, it is ok now? at least regarding screenrc
2020-06-28 13:38:03 <adu> I can't sudo
2020-06-28 13:40:43 <afontain_> man sudoers
2020-06-28 13:41:12 <adu> visudo: /etc/sudoers: Permission denied
2020-06-28 13:42:20 <adu> So I mounted -v ./sudoers:/etc/sudoers and it's owned by root, but every time I try and sudo, I get: sudo: /etc/sudoers is owned by uid 1, should be 0
2020-06-28 14:18:39 <artok> adu on wnere?
2020-06-28 14:18:53 <artok> where, that is
2020-06-28 14:19:21 <adu> artok: in docker?
2020-06-28 14:19:26 <artok> write better docker image that you don't need to sudo
2020-06-28 14:19:31 <artok> but what is your use case?
2020-06-28 14:19:41 <adu> I'm trying to build an apk
2020-06-28 14:20:12 <adu> and the tutorial told me to use a non-root user
2020-06-28 14:20:13 <artok> and never mess your host etc files with bind mounts
2020-06-28 14:20:34 <adu> ok, so how to I made a non-root user without editing /etc files?
2020-06-28 14:21:04 <artok> don't bind mount, that is the thingie
2020-06-28 14:21:35 <artok> you create that builder, non-root user inside docker image, not to your host
2020-06-28 14:22:46 <artok> check https://gitlab.alpinelinux.org/alpine/docker-abuild
2020-06-28 14:25:44 <adu> I'm having trouble understanding this line
2020-06-28 14:25:48 <adu> printf -- >> /etc/apk/repositories '/home/builder/packages/%s\n' main community testing
2020-06-28 14:26:28 <adu> what does it mean when command line arguments appear after redirections
2020-06-28 14:31:05 <artok> there is %s replacement
2020-06-28 14:33:40 <adu> I've never seen redirections before arguments
2020-06-28 14:33:55 <adu> I didn't know that was even allowed
2020-06-28 14:34:19 <adu> where do you learn that this is allowed?
2020-06-28 14:38:13 <artok> shell manual
2020-06-28 14:52:27 <adu> artok: do you mean this? https://pubs.opengroup.org/onlinepubs/9699919799/utilities/V3_chap02.html#tag_18_09_01
2020-06-28 14:55:39 <artok> alpine scripts should be busybox ash runnable
2020-06-28 14:56:14 <adu> https://hastebin.com/raw/esiriyofiq
2020-06-28 14:58:34 <adu> ok back to task, here is what I'm trying to convert to an apk
2020-06-28 14:58:38 <adu> https://github.com/andydude/dockerfiles/blob/master/xar/Dockerfile
2020-06-28 15:02:12 <adu> what should ALPINE_TAG, ALPINE_REL, and ALPINE_IMG be in the Dockerfile/
2020-06-28 15:04:54 <adu> how is it that printf seems to run multiple times?
2020-06-28 15:07:32 <adu> is this UB?
2020-06-28 15:09:08 <mokulus> can you repost? just joined
2020-06-28 15:11:17 <adu> I'm trying to understand the first line of https://gitlab.alpinelinux.org/alpine/docker-abuild/-/blob/master/Dockerfile.in
2020-06-28 15:11:46 <adu> printf -- >> /etc/apk/repositories '/home/builder/packages/%s\n' main community testing
2020-06-28 15:12:07 <adu> So I now understand that this is equivalent to
2020-06-28 15:12:19 <adu> printf -- '/home/builder/packages/%s\n' main community testing >> /etc/apk/repositories
2020-06-28 15:13:00 <adu> but now I'm trying to understand what it means when the number of arguments doesn't match the number of format specifiers
2020-06-28 15:13:57 <ikke> clandmeter: ping
2020-06-28 15:13:59 <artok> iterations
2020-06-28 15:19:47 <adu> Is this documented?
2020-06-28 15:21:25 <artok> indeed, but I think you're now lost in small details over what you're trying to accomplish
2020-06-28 15:22:12 <artok> you should check what teh dabuild script does
2020-06-28 15:22:47 <adu> ok it is standard, the spec says $(printf "%5d%4d\n" 1 21 321 4321 54321 0) would print 3 lines
2020-06-28 15:24:20 <adu> Sorry, I get it now
2020-06-28 16:17:27 <adu> So I'm still not sure how to build a package
2020-06-28 16:18:36 <adu> What does this mean: "cat <<- EOF > "$HOME"/.abuild/abuild.conf"
2020-06-28 16:19:05 <adu> is '> "$HOME"/.abuild/abuild.conf' part of the here-document, or part of the command?
2020-06-28 16:19:43 <mokulus> adu: https://www.tldp.org/LDP/abs/html/here-docs.html
2020-06-28 16:20:40 <artok> that is
2020-06-28 16:20:47 <artok> redirection
2020-06-28 16:21:16 <artok> but are you now trying to create docker image for building or trying to create that apple xar ?
2020-06-28 16:23:38 <adu> I'm familiar with here-documents, but not the part of the standard which says that commands and reidirections can follow the limit string
2020-06-28 16:28:44 <lunaphyte> that's not anything that has to do with a heredoc
2020-06-28 16:29:04 <lunaphyte> the heredoc command is "cat <<- EOF"
2020-06-28 16:29:06 <lunaphyte> that's it
2020-06-28 16:56:38 <lerelel> hei
2020-06-28 16:59:07 <lerelel> BLACK TERMINALS MATTER
2020-06-28 16:59:15 <lerelel> 
ACTION proceeds to loot mp3s, mkv releases, and m$$ paying software.

2020-06-28 16:59:58 <lerelel> just kiddin
2020-06-28 17:00:02 <lerelel> anybody out there?
2020-06-28 17:01:23 <lerelel> I followed the sway wiky, added the rest of packages, alacrity, dmenu, etc, and used the command to start sway from there.
2020-06-28 17:02:29 <lerelel> still the same sway desktop painting, no right/left click, no menu, nothing.
2020-06-28 17:02:57 <Cogitri> I guess you have to add it to some config file too?
2020-06-28 17:03:16 <lerelel> Cogitri: well, no mention in the wiki.
2020-06-28 17:05:11 <lerelel> I was missing xorg-server-xwayland, but I doubt that got anything to do with getting a menu in sway
2020-06-28 17:05:55 <mokulus> dmenu is xorg only im fairly sure
2020-06-28 17:08:56 <lerelel> do you mean this?? --> "An example config is provided at /etc/sway/config. Copy it to ~/.config/sway/config and read through it to learn the default keybindings. "
2020-06-28 17:09:11 <mokulus> no
2020-06-28 17:09:17 <mokulus> sway is wayland i3
2020-06-28 17:09:33 <mokulus> dmenu is written for x only, so it wont work
2020-06-28 17:13:15 <Hello71> ...
2020-06-28 17:15:04 <mokulus> please use unicode elipsis …
2020-06-28 17:16:11 <lerelel> devuan --> [DRI] DRI driver: virtio_gpu
2020-06-28 17:16:26 <mps> hmm ...
2020-06-28 17:18:18 <lerelel> alpine missing  "virtio_gpu"
2020-06-28 17:18:44 <lerelel> I am using the cloud image, could it be that the kernel isn't compiled with virtio_gpu?
2020-06-28 17:20:11 <lerelel> it is loaded.
2020-06-28 17:20:25 <lerelel> this is annoying
2020-06-28 17:22:33 <mps> linux-lts, config-virt.x86_64 => CONFIG_DRM_VIRTIO_GPU=m
2020-06-28 17:22:59 <ikke> https://git.alpinelinux.org/aports/tree/main/linux-lts/config-virt.x86_64#n3059
2020-06-28 17:23:20 <lerelel> mps: yes it is loaded I just checked with 'lsmod'
2020-06-28 17:23:39 <lerelel> presumably the driver is inside mesa-dri-gallium.
2020-06-28 17:24:50 <Hello71> sway doesn't come with a menu....
2020-06-28 17:25:03 <Hello71> nor does i3
2020-06-28 17:25:05 <Hello71> it's not openbox
2020-06-28 17:25:09 <lerelel> Hello71: so how's supposed to be used?
2020-06-28 17:25:44 <lerelel> I typed this:  XDG_RUNTIME_DIR=/tmp sway
2020-06-28 17:30:06 <lerelel> I want to reduced mem footprint to stop climate change once and for all.
2020-06-28 17:30:18 <mokulus> lerelel: drunk?
2020-06-28 17:31:19 <lerelel> wait, changing dm, brb
2020-06-28 17:52:08 <lerelel> w00t, problem solutionized
2020-06-28 17:54:46 <lerelel>  modeset(0): [DRI2]   DRI driver: virtio_gpu
2020-06-28 17:54:55 <lerelel> Initializing extension DRI3
2020-06-28 17:55:03 <lerelel>  GLX: Initialized DRI2 GL provider for screen 0
2020-06-28 17:55:07 <lerelel> cute as hell!!
2020-06-28 17:55:17 <lerelel> so, xfce broken then?
2020-06-28 18:13:41 <lerelel> lol, youtube HD vids play smooth with better fps that xterm :(
2020-06-28 18:17:05 <lerelel> one thing is solved then another issue shows up
2020-06-28 18:19:47 <lerelel> lol@freenode, bait and switch policy.
2020-06-28 18:22:08 <lerelel> there are no man pages?
2020-06-28 18:25:23 <artok> install them
2020-06-28 18:29:36 <lerelel> :( i've got an outdate image, musl 1.1, gcc 9
2020-06-28 18:29:55 <lerelel> are there images ready with musl 1.2?
2020-06-28 18:30:02 <lerelel> and gcc 10 at least?
2020-06-28 18:30:32 <ikke> alpine has not upgraded to gcc 10 yet
2020-06-28 18:31:05 <ikke> an neither to musl 1.2 (which is not trivial for 32-bits systems)
2020-06-28 18:31:27 <lerelel> overall I am satisfied with its performance, waay better than any *butntu systemDown, susebloat, etc.
2020-06-28 18:32:09 <lerelel> ikke: yeah I read about it in the musl page.
2020-06-28 18:32:39 <lerelel> all I need is firefox, a terminal.
2020-06-28 18:32:51 <lerelel> as long as those are supported in 32bit.
2020-06-28 18:33:22 <Hello71> what's going on with gcc 10? too hard, or just nobody has worked on it so far?
2020-06-28 18:33:33 <mokulus> ikke: are you alpine developer? I've noticed that apk is really fast, what makes is so performant compared to other package managers?
2020-06-28 18:33:41 <Hello71> afaik between gentoo and fedora most of the fno-common issues have been worked out
2020-06-28 18:34:12 <lerelel> mokulus: I noticed this too.
2020-06-28 18:34:13 <mokulus> Hello71: I think it's a matter of testing, it's a change that affect almost everything
2020-06-28 18:34:21 <Hello71> mokulus: apk is not that fast compared to pacman. apt and yum/dnf are very slow due to slow/"safe" design
2020-06-28 18:34:56 <lerelel> apk feels like a lil toy.
2020-06-28 18:35:01 <ikke> mokulus: I help with Alpine yes, but I'm not involved with development of APK
2020-06-28 18:35:24 <ikke> apk tries to handle data only once
2020-06-28 18:35:26 <Hello71> apk and pacman have somewhat optimized package formats, but that's really only a result of general simplified design
2020-06-28 18:35:26 <mokulus> I've always used pacman with -yy, so maybe that's why I find apk faster
2020-06-28 18:36:00 <Hello71> well sure, if you hold the knife from the blade end then a hammer will do a better job chopping
2020-06-28 18:36:07 <lerelel> and simpler, I never understood apt-get/apt-cache to install and search
2020-06-28 18:37:12 <lerelel> I am wondering if distros are over engineering things after trying this.
2020-06-28 18:37:33 <lerelel> specially opensuse was insane.
2020-06-28 18:37:47 <Hello71> arch and alpine share some philosophy of aggressive simplicity
2020-06-28 18:38:32 <lerelel> isn't arch still with systemd?
2020-06-28 18:38:56 <Hello71> to be fair many things are lost in that. for example, pacman doesn't sync, and I think apk doesn't either
2020-06-28 18:39:26 <lerelel> I mean after trying this, why aren't major distros attempt, at leas, a leaner approach?
2020-06-28 18:40:16 <lerelel> plymouth <-- lololol!!
2020-06-28 18:40:33 <Hello71> systemd *is* the simple approach at this point. so is glibc. despite their flaws, it is undeniably more work to use openrc and musl
2020-06-28 18:40:56 <Hello71> it is arguably a "leaner" system though, which is one of the ways in which arch and alpine differ
2020-06-28 18:41:18 <lerelel> Hello71: heellll nah!! I just tried here for minmum systeam load.
2020-06-28 18:41:40 <lerelel> musl wins hands down. I can see in htop yellow bars.
2020-06-28 18:41:46 <Hello71> your premise is dumb in the first place, memory usage has no contribution to "climate change"
2020-06-28 18:42:16 <lerelel> any systemDownsyndrom htop with solid 100% yellow bars.
2020-06-28 18:42:25 <Hello71> 
ACTION bails

2020-06-28 18:42:47 <lerelel> Hello71: no, mem usage, something with the cpu hitting 100% accross the cores.
2020-06-28 18:42:52 <mokulus> yellow bars are cache
2020-06-28 18:43:22 <lerelel> and those without even running firefox.
2020-06-28 18:43:51 <lerelel> Hello71: are you defending systemDowngrade?
2020-06-28 18:45:31 <Hello71> sigh, this is the problem with any "alternative" distro. you inevitably attract this sort of "anti-mainstream" troll
2020-06-28 18:46:24 <ikke> Hello71: we are well aware of that fact
2020-06-28 18:46:47 <ikke> lerelel: if systemd worked with musl, Alpine might have chosen to use it
2020-06-28 18:47:36 <lerelel> don't do it!
2020-06-28 18:47:46 <lerelel> there are tons of other inits out there
2020-06-28 18:48:28 <mokulus> so why was musl chosen exactly?
2020-06-28 18:48:40 <mokulus> busybox + systemd LOL
2020-06-28 18:49:28 <lerelel> systemDown will never use musl for the simple fact that they won't be able to hook/hide their backdoors.
2020-06-28 18:50:17 <lerelel> whereas in glibc they've got plenty of room for the mischevious practices.
2020-06-28 18:52:40 <lerelel> lol --> https://latesthackingnews.com/2019/01/13/multiple-critical-security-vulnerabilities-discovered-in-linux-systemd/
2020-06-28 18:53:32 <lerelel> anyways I am not trying to win an argument, just glad that alpines isn't saturating the yellow color bar in htop
2020-06-28 18:53:54 <ikke> maybe read this: https://www.linuxatemyram.com/
2020-06-28 18:54:33 <lerelel> ikke: not the MEM bar, the CPU bars.
2020-06-28 18:55:13 <mokulus> there is no yellow in cpubars in htop
2020-06-28 18:55:48 <lerelel> mokulus: yes it got three, blue, red and yellow
2020-06-28 18:55:58 <mokulus> on cpu?
2020-06-28 18:56:02 <Hello71> https://www.google.com/search?hl=en&q=syslog%20vulnerability
2020-06-28 18:56:04 <lerelel> yep
2020-06-28 18:56:09 <mokulus> not on cpu
2020-06-28 18:56:12 <mokulus> press ? in htop
2020-06-28 18:56:40 <mokulus> maybe you have non standard color scheme
2020-06-28 18:57:04 <mokulus> ahhh
2020-06-28 18:57:09 <lerelel> in the latest version
2020-06-28 18:57:10 <mokulus> htop has themes
2020-06-28 18:57:16 <mokulus> i probably have different one
2020-06-28 18:57:23 <mokulus> totally forgot
2020-06-28 18:57:25 <mokulus> sorry
2020-06-28 18:57:37 <mps> ikke: I hope you are wrong sage about musl/systemd combo :)
2020-06-28 18:58:51 <lerelel> alright, my cpu bars got three colors, green/red/yellow
2020-06-28 18:59:04 <lerelel> the mem bar, green/blue/yellow
2020-06-28 19:02:07 <mokulus> can you press ? and describre cpu colors
2020-06-28 19:04:30 <lerelel> blue(low priority)/green(normal)/red(kernel)/turquois(virtualiz)
2020-06-28 19:04:48 <mokulus> so where is yellow?
2020-06-28 19:05:06 <mokulus> yellow shouldn't be on cpu then
2020-06-28 19:05:09 <lerelel> that's the description, but htop shows yellow, so something it's broken.
2020-06-28 19:06:58 <lerelel> using fg white and bg black.
2020-06-28 19:10:49 <lerelel> which is the best cpu to run musl? instruction-wise
2020-06-28 19:11:49 <mokulus> what do you mean?
2020-06-28 19:12:00 <lerelel> bust arch
2020-06-28 19:15:24 <mps> risc-v :)
2020-06-28 19:16:30 <lerelel> damn! this arch is affordable! 60$
2020-06-28 19:49:39 <lerelel> lol, no flash
2020-06-28 19:50:03 <lerelel> 
ACTION goes back to windoze

2020-06-28 20:11:52 <artok> eh
2020-06-28 21:25:53 <Zeppel> Hi! I tried to setup wireguard on alpine 3.12 (raspberry pi 3b) but i only receive mkinit-error. did someony made it?
2020-06-28 21:29:49 <remexre> is the cert on dl-cdn.alpinelinux.org borked?
2020-06-28 21:41:09 <Zeppel> No, not a signature problem. i cant build new kernel modules with a LBU :-/
2020-06-28 21:41:29 <Zeppel> Cant find any solution in mailinglist-history or on alpinelinux-wiki
2020-06-28 21:49:51 <MY-R> remexre, dl-cdn. is from fastly.net and working cert is under this address https://alpine.global.ssl.fastly.net/alpine/ (you can check status/addresses of mirrors here: https://mirrors.alpinelinux.org/)
2020-06-28 21:50:13 <ikke> recently we put our own cert on fastly
2020-06-28 21:50:27 <ikke> so https://dl-cdn.alpinelinux.org should work (and is working for me)
2020-06-28 21:50:44 <ikke> huh
2020-06-28 21:50:56 <maldridge> ikke: do y'all pay for fastly?
2020-06-28 21:50:56 <MY-R> ikke, oh nice, ye working for me too
2020-06-28 21:51:12 <ikke> maldridge: no, they sponsor us
2020-06-28 21:51:46 <maldridge> ah
2020-06-28 21:51:48 <artok> I've got unsecure connection... hmmm
2020-06-28 21:52:13 <ikke> it's a letsencrypt cert
2020-06-28 21:53:08 <escala> hello?
2020-06-28 21:53:19 <escala> I am using this weird ircII that I found on alpine.
2020-06-28 21:53:26 <Hello71> Zeppel: you speak german?
2020-06-28 21:53:44 <escala> what's the difference between irssi and ircII
2020-06-28 21:53:49 <Hello71> ircii is older
2020-06-28 21:54:02 <artok> mmm that classic
2020-06-28 21:54:11 <escala> so not reccomended or something?
2020-06-28 21:54:39 <Hello71> usually weechat is a better option for new users
2020-06-28 21:54:56 <Hello71> I think ircii had some security problems
2020-06-28 21:55:00 <Zeppel> Ja
2020-06-28 21:55:13 <Hello71> Zeppel: I can tell by unnecessary use of hyphens :p
2020-06-28 21:55:30 <escala> alse ircII connects directly to freenode
2020-06-28 21:55:38 <Hello71> also you are using vodafone, but honestly the hyphens are a better clue
2020-06-28 21:55:57 <Zeppel> :D  OK, gut erkannt ;)
2020-06-28 21:56:17 <Hello71> germans and their hugecompoundwords :p
2020-06-28 21:56:50 <Zeppel> germans always love hyphens :D
2020-06-28 21:57:19 <ikke> Very similar to dutch
2020-06-28 21:58:09 <escala> weird question
2020-06-28 21:58:15 <remexre> MY-R: hm, maybe my issue is the comcast dns mitm, then
2020-06-28 21:58:30 <remexre> apk update works from here, but gets "SSL certificate subject doesn't match host dl-cdn.alpinelinux.org" on the work CI system
2020-06-28 21:58:32 <escala> is there some flash enabler package?
2020-06-28 21:58:39 <ikke> ah yes, dns mitm
2020-06-28 21:59:16 <MY-R> :\
2020-06-28 21:59:46 <escala> i would hate having to go back to winblowz just for flash
2020-06-28 22:00:21 <escala> since m$$ is capable of playing flash content.
2020-06-28 22:00:40 <remexre> escala: does flash projector work for you?
2020-06-28 22:01:56 <escala> remexre no such a package
2020-06-28 22:02:10 <remexre> escala: er you'd need to get it from adobe's site
2020-06-28 22:04:39 <remexre> hm, wait, I just got the "SSL certificate subject doesn't match host dl-cdn.alpinelinux.org" locally now too
2020-06-28 22:05:02 <remexre> is it plausible that fastly isusing the wrong cert on some servers but not others?
2020-06-28 22:05:06 <escala> upgrade ca-certificates
2020-06-28 22:06:23 <ikke> It's been more than a week ago that we changed it
2020-06-28 22:06:25 <remexre> ah, okay; would be nice to get the updated one into the alpine docker containers
2020-06-28 22:09:39 <remexre> wait, no, after installing ca-certificates it just came back
2020-06-28 22:09:44 <remexre> it's really intermittent, looks like
2020-06-28 22:09:52 <remexre> reproduces a lot more reliably on the work machines
2020-06-28 22:11:19 <escala> delete apk cache?
2020-06-28 22:11:35 <ikke> that does not make sense
2020-06-28 22:11:50 <escala> or reboot
2020-06-28 22:11:58 <ikke> it's not windows :)
2020-06-28 22:12:11 <remexre> these are all docker ctrs, so I think those're both the same, heh
2020-06-28 22:12:32 <remexre> would I just run `apk cache -U`?
2020-06-28 22:13:53 <ikke> no
2020-06-28 22:14:04 <ikke> apk cache just downloads a package to the local cache
2020-06-28 22:14:31 <lerelel> how sad, any other way to put firefox to sleep than ctrl-z on the terminal?
2020-06-28 22:14:43 <ikke> put to sleep?
2020-06-28 22:15:04 <mps> ikke: doesn't linux want to be next windows? :)
2020-06-28 22:15:11 <lerelel> oh wai, I've looged with twice
2020-06-28 22:15:26 <mps> so reboot is mandatory for any change
2020-06-28 22:16:40 <lerelel> ikke: firefox iddling is like sprinting through the cores.
2020-06-28 22:18:08 <lerelel> amazing! putting firefox to sleep I am able to get a single core at 0.7!!
2020-06-28 22:19:33 <lerelel> all others at 0
2020-06-28 22:20:14 <MY-R> and when firefox not "sleeping"?
2020-06-28 22:25:00 <lerelel> MY-R: ctrl-z in the terminal
2020-06-28 22:25:50 <MY-R> ye but what usage you have got when firefox is awake
2020-06-29 02:38:52 <adu> I don't understand initd and confd
2020-06-29 02:45:25 <adu> I'm still trying to make a package
2020-06-29 02:45:27 <adu> https://hastebin.com/ezulununit.sh
2020-06-29 03:14:03 <maxice8> initd and  confd are OpenRC service files
2020-06-29 03:17:55 <adu> ok I think I got it working
2020-06-29 03:18:53 <adu> and I ran docker-abuild, and it says it compiled my package, then it seems to have deleted it...
2020-06-29 03:26:37 <adu> how do I make it not delete it?
2020-06-29 03:29:16 <maxice8> It shouldn't, it should be under the directory mentioned in REPODEST= in /etc/abuild.conf
2020-06-29 03:32:37 <adu> oh I found it
2020-06-29 03:33:23 <adu> it was in ../../../packages/edge/andydude/x86_64/
2020-06-29 03:33:43 <adu> thanks!
2020-06-29 03:34:06 <adu> https://hastebin.com/mavohukilo.sh
2020-06-29 03:35:24 <adu> So does this have to be in a certain location, or can I apk add http://...
2020-06-29 03:35:35 <maxice8> you can add it to /etc/apk/respositories
2020-06-29 03:36:10 <maxice8> you can also add an apk file
2020-06-29 03:36:12 <adu> so add the web server to there, then $(apk add xar)
2020-06-29 03:36:22 <maxice8> or  apk add /path/to/xar.apk
2020-06-29 03:36:47 <maxice8> or you can add the path in REPODEST= variable of /etc/abuild.conf to your /etc/apk/repositories
2020-06-29 03:37:01 <adu> I would prefer to setup a web server, since I don't have any persistent alpine boxes
2020-06-29 03:37:48 <maxice8> Then just have it serve the APKINDEX.tar.gz together with the .apk archives and add it to /etc/apk/repositories
2020-06-29 03:45:53 <adu> ok I did that
2020-06-29 03:45:57 <adu> and then I got "UNTRUSTED signature"
2020-06-29 03:46:08 <maxice8> You did that which ?
2020-06-29 03:46:35 <adu> I made a web server, and put x/aports and x/packages under it
2020-06-29 03:47:06 <adu> I added http://.../packages/edge/andydude to /etc/apk/repositories
2020-06-29 03:48:25 <maxice8> Also need to add public key used to sign in /etc/apk/keys
2020-06-29 03:48:28 <maxice8> see alpine-keys
2020-06-29 03:48:40 <adu> is that a parameter to dabuild?
2020-06-29 03:48:49 <maxice8> no
2020-06-29 03:49:41 <adu> is it an output of dabuild?
2020-06-29 03:50:21 <maxice8> no, just look in $HOME/.ssh
2020-06-29 03:54:21 <adu> oh .abuild/ has some stuff n/m
2020-06-29 03:54:51 <adu> thank you
2020-06-29 03:55:20 <adu> where would I put these on the host I'm trying to install xar?
2020-06-29 03:55:35 <adu> in /etc/apk/keys?
2020-06-29 03:58:54 <adu> I put my public key in /etc/apk/keys, and I still get "UNTRUSTED signature"
2020-06-29 05:46:14 <abenz> something is causing inconsistency in unpacking my apkovl when rebooting.. sometimes the extraction bar goes slowly (extracting all my stuff) and loading normally. Other times it zips by very quickly, hardware detection doesn;t work (eg amdgpu, kms doesn;t load) and after a few reboots, it loads again as normal
2020-06-29 05:46:21 <abenz> any idea what could be causing this?
2020-06-29 05:49:56 <abenz> I noticed that if I interrupt grub (by pressing up or down) and letting things "settle" for a few seconds, I have a much better chance of correctly booting and unzipping my apkovl-ays..
2020-06-29 05:55:18 <ikke> issues with the boot medium?
2020-06-29 05:56:23 <abenz> I suspected that, but my hostname shows up (ie user/pass that I set, and other settings).. just not everything. I disabled "fast boot" in bios as well.
2020-06-29 05:57:29 <abenz> I will test with another USB stick and see how it goes. Although a partial load of settings makes me less suspecious of a faulty usb
2020-06-29 07:21:56 <ecp18>  /quit
2020-06-29 07:30:35 <nacelle> abenz: if its an sbc, maybe the power supply is inadequate?
2020-06-29 08:28:05 <moosl> hei
2020-06-29 08:28:28 <moosl> are there many day to day software available on musl?
2020-06-29 08:28:49 <moosl> darktable, gimp, inkscape, mplayer, blender etc?
2020-06-29 08:29:11 <moosl> I wonder how do they perform
2020-06-29 08:29:30 <moosl> is it realistic to switch main OS for musl?
2020-06-29 08:30:10 <ikke> darktable, gimp, inkscape are available
2020-06-29 08:30:16 <ikke> mplayer and blender are not (yet)
2020-06-29 08:30:22 <moosl> w00t, rawtherapee availble
2020-06-29 08:30:59 <mps> mpv instead of outdated mplayer
2020-06-29 08:31:06 <moosl> ikke: so slowly it's getting to become a full blown desktop OS?
2020-06-29 08:31:28 <mps> moosl: no, it is already
2020-06-29 08:32:04 <mps> for long time
2020-06-29 08:33:26 <afontain_> you'll still find less things packaged
2020-06-29 08:34:00 <mps> yes, microsoft office doesn't work on musl :)
2020-06-29 08:34:06 <afontain_> also, forget proprietary software
2020-06-29 08:34:06 <moosl> latex/luatex in it?
2020-06-29 08:34:17 <afontain_> (though flatpak works)
2020-06-29 08:34:22 <afontain_> moosl: yes
2020-06-29 08:34:40 <afontain_> that works
2020-06-29 08:35:09 <moosl> wow even wine is in it.
2020-06-29 08:35:23 <moosl> how to search by categories? I want to search for games and emulators
2020-06-29 08:35:34 <ikke> We don't have categories
2020-06-29 08:35:36 <moosl> like dolphin, ppsspp etc.
2020-06-29 08:35:55 <afontain_> for games, there's minetest
2020-06-29 08:36:49 <afontain_> gnome-games is packaged, I haven't tested that, but I've heard it's a good emulator thing?
2020-06-29 08:36:50 <moosl> zsnes, interesting.
2020-06-29 08:36:53 <afontain_> (not sure if any or every backend is here)
2020-06-29 08:37:53 <moosl> if wine is in it, theoretically I could install and play Modern Warfare?
2020-06-29 08:38:58 <ikke> note that wine is mostly only usefull on 32-bits systems
2020-06-29 08:39:08 <ikke> we don't have multilib
2020-06-29 08:39:22 <moosl> ikke: a lot of dx9 and dx11 are still solely 32bit
2020-06-29 08:39:34 <moosl> for the sake of testing musl.
2020-06-29 08:40:02 <moosl> that's it I will try and report back.
2020-06-29 08:44:37 <abenz> nacelle: sbc?
2020-06-29 08:44:49 <ikke> Single board computer
2020-06-29 08:46:23 <abenz> ah, no its a desktop. I can (almost) consistently avoid the problem by simply interrupting grub and waiting for a few seconds, then booting. So its fine for now.. no point in updating grub.cfg since my use case is always me in front of PC (diskless)
2020-06-29 08:47:45 <abenz> next will be setting up virt-manager on alpine diskless, and controlling it from a laptop. Then going headless by passing through the main GPU to a VM!
2020-06-29 08:50:21 <abenz> libvirtd wont run if auto-started, it gets stuck when added to runlevel default. I have to (you guessed it) wait for a while after system runs (dbus?) and then start it..
2020-06-29 08:52:28 <ikke> maybe an issue with lack of entropy?
2020-06-29 08:54:25 <abenz> never occured to me, will look into it
2020-06-29 08:56:08 <moosl> abenz: if it's going to be headless, why not use plain qemu, less insane deps. Besides qemu headless absolutely no need for dbus.
2020-06-29 08:56:48 <moosl> after trying musl 32bit I realize that 64bit is really a bloat arch.
2020-06-29 09:04:55 <abenz> moosl: indeed a good point
2020-06-29 09:05:44 <abenz> just fancied using virt-manager with qemu://headless-alpine.IP and controlling knobs and whatnot remotely
2020-06-29 13:46:03 <eiro> hello people
2020-06-29 13:47:25 <Cogitri> 👋
2020-06-29 14:07:46 <mokulus> hello
2020-06-29 17:35:22 <mps> I have interesting behavior with X cursor on one machine. when X starts I can move it for 2-3 seconds and after that it is frozen on last position
2020-06-29 17:35:54 <mps> xev and evtest shows thad touchpad device works
2020-06-29 17:37:06 <mps> even, cursor changes shape when I move it (blindly) around, but it stays on same position on screen
2020-06-29 17:37:28 <mps> and, I think it is not alpine related, but who knows
2020-06-29 17:38:40 <mps> searching net doesn't give sensible answers
2020-06-29 17:43:31 <mps> and it is unrelated to pointer device, tried with usb mouse and the behavior is same
2020-06-29 17:56:35 <MY-R> mps what is your xorg gpu driver?
2020-06-29 17:58:59 <MY-R> try with < Option "SWcursor" "on" > in your < Section "Device" >
2020-06-29 17:59:52 <mps> modesettinfg with kms
2020-06-29 18:00:31 <MY-R> so ye, check with SWcursor turned on
2020-06-29 18:01:00 <mps> forgot to mention that all is fine with old 3.18 kernel
2020-06-29 18:01:25 <mps> but with 5.4 it freeze
2020-06-29 18:02:10 <MY-R> mps, anything at the end of "/var/log/Xorg.0.log" ?
2020-06-29 18:02:45 <mps> switching from X to console and back, resets it and again can move it few seconds, but it again freeze after few seconds
2020-06-29 18:03:14 <MY-R> with swcursor on?
2020-06-29 18:03:19 <mokulus> ok, but you need to give some information
2020-06-29 18:03:29 <mokulus> post log
2020-06-29 18:03:32 <mps> yes, on every switch to console and back it shows in log that the driver is removed
2020-06-29 18:04:28 <mps> when it is frozen I can move windows with it, just cursor stays on same position
2020-06-29 18:05:14 <mps> btw, consolation works fine
2020-06-29 18:06:20 <mokulus> cool
2020-06-29 18:07:13 <mps> from Xorg.0.log => [  4445.280] (II) event2  - Elan Touchpad: device removed
2020-06-29 18:07:47 <MY-R> so SWcursor doesnt change anything at all? :\ because not sure if you tried it or not
2020-06-29 18:08:19 <mps> MY-R: didn't tried yet
2020-06-29 18:08:32 <mps> I want to find 'real' problem
2020-06-29 18:08:57 <mps> because it works fine with different kernel
2020-06-29 18:09:20 <mps> so, must be some relation kernel <=> X
2020-06-29 18:09:31 <MY-R> so check it first, so will know if it have something commong with gpu or not... in newer kernel gpu stuff is much more updated
2020-06-29 18:10:51 <mps> yes, and I'm using non standard kernel, but with a lot of patches for mediatek, acer arm64 chromebook
2020-06-29 18:11:30 <mps> and I will try with SWcursor later
2020-06-29 18:11:48 <mps> but doubt it will fix the problem
2020-06-29 18:11:58 <MY-R> even here just found: https://bbs.archlinux.org/viewtopic.php?id=250128
2020-06-29 18:13:28 <mps> MY-R: thanks, I will try
2020-06-29 18:16:27 <mps> MY-R: I'm asking myself why I'm stubborn and didn't followed your advice straightly :)
2020-06-29 18:16:41 <mps> yes, it works with SWcursor
2020-06-29 18:17:42 <mps> MY-R: thanks for insisting to try it
2020-06-29 18:18:21 <MY-R> ye is weird that working on older kernel just fine but MAYBE it have something to do with missing "silken" mouse stuff after many distros started building xorg with meson
2020-06-29 18:19:05 <mps> but hey, yet another chromebook running alpine with mainline kernel (and not chromeOS kernels) though with some patches
2020-06-29 18:19:52 <mps> only serious missing part is suspend-to-ram
2020-06-29 18:20:03 <MY-R> heh, gj, maybe try with any newer kernel... found that mouse issue suposed to be fixed in 5.3 but looks like nope
2020-06-29 18:20:14 <mps> well, suspend-to-ram work but resume doesn't :)
2020-06-29 18:20:29 <MY-R> "work" :D
2020-06-29 18:20:48 <mps> works?
2020-06-29 18:21:11 <mps> I told many times that I will never learn English ;)
2020-06-29 18:21:12 <MY-R> how you can be sure that suspended fine if cant resume? :D
2020-06-29 18:21:31 <mps> I can ssh to it after resume
2020-06-29 18:21:58 <MY-R> ah, so again gpu problem probably, screen is just off, right?
2020-06-29 18:22:00 <mps> so I see it works but the screen is blank
2020-06-29 18:22:07 <mps> yes
2020-06-29 18:22:09 <MY-R> ye
2020-06-29 18:22:49 <mps> but, that's a life with arm machines
2020-06-29 18:23:24 <MY-R> ye, every got own quirks
2020-06-29 18:25:33 <MY-R> mps, did you try use anything else than modesetting, what gpu device got chromebook?
2020-06-29 18:26:02 <adu> I did it!
2020-06-29 18:26:23 <mps> but I have dmesg crash message, [ 5975.324574] [CRTC:39:crtc-0] vblank wait timed out
2020-06-29 18:26:57 <mps> good starting point
2020-06-29 18:27:25 <adu> https://hastebin.com/raw/ruxodekixa
2020-06-29 18:27:31 <adu> https://hastebin.com/raw/jokojemunu
2020-06-29 18:27:42 <adu> I finally learned how to make packages :)
2020-06-29 18:28:33 <mokulus> cool
2020-06-29 18:28:59 <mps> adu: you don't need to have 'alpine-sdk' in makedepends
2020-06-29 18:29:08 <adu> you don't? ok
2020-06-29 18:29:11 <mps> it is default
2020-06-29 18:30:09 <mps> install is empty, so not needed
2020-06-29 18:30:22 <mps> and builddir is wrong
2020-06-29 18:31:38 <adu> what is builddir
2020-06-29 18:31:59 <ikke> a variable that tells abuild where to find the files to build
2020-06-29 18:32:18 <adu> so that should be $srcdir/$pkgname-$pkgver?
2020-06-29 18:32:36 <ikke> yes, but if it's the default, then you should leave it out
2020-06-29 18:32:50 <adu> does that mean I don't need to $(cd $pkgname-$pkgver)?
2020-06-29 18:32:56 <ikke> https://gitlab.alpinelinux.org/alpine/abuild/blob/master/abuild.in#L2677
2020-06-29 18:33:02 <ikke> adu: correct
2020-06-29 18:33:21 <adu> I didn't use abuild I used dabuild
2020-06-29 18:33:33 <ikke> dabuild uses abuild
2020-06-29 18:34:10 <adu> and I think that was generated by newapkbuild and I never editted it
2020-06-29 18:34:52 <adu> so what should I do for zchunk, it requires a separate builddir, should builddir=$srcdir/build?
2020-06-29 18:35:20 <mps> MY-R: forgot to tell that authors of patches for this kernel mentioned that gpu driver is unstable (and bugy)
2020-06-29 18:36:00 <MY-R> :\
2020-06-29 18:37:15 <ikke> adu: https://gitlab.alpinelinux.org/alpine/aports/-/blob/master/community/midori/APKBUILD#L15
2020-06-29 18:37:37 <adu> ikke: thanks
2020-06-29 18:37:48 <mps> but I'm ok with it now, battery on this machine died so I don't have any reason to put it in suspend-to-ram
2020-06-29 18:50:34 <adu> ikke: that didn't work, I still need to mkdir -p $pkgname-$pkgver/build; cd $pkgname-$pkgver/build
2020-06-29 18:51:07 <ikke> You need the prepare step as well
2020-06-29 18:51:13 <adu> oops
2020-06-29 18:51:58 <adu> that works
2020-06-29 18:52:48 <ikke> btw, I did `ag -G APKBUILD 'builddir=.*build'` to find that example
2020-06-29 18:53:14 <ikke> ofcourse, it returned many results, so I looked for one that was most cannonical
2020-06-29 18:53:54 <adu> what's ag?
2020-06-29 18:54:02 <ikke> the_silver_searcher
2020-06-29 18:54:28 <ikke> grep -r on steroids
2020-06-29 18:54:47 <ikke> (there is also ripgrep, but I never bothered to look into it)
2020-06-29 18:54:48 <adu> like mg is emacs on steroids?
2020-06-29 18:54:55 <ikke> I don't know mg :)
2020-06-29 18:56:22 <adu> it's in alpine/main
2020-06-29 18:58:54 <adu> I think emacs is like 60 MiB, mg is 170 KiB
2020-06-29 18:58:56 <ikke> Ofcourse the readme does not tell anything about what it is
2020-06-29 19:01:35 <adu> supposedly mg (or a fork of it) is Linus' favorite text editor
2020-06-29 19:04:59 <maldridge> is there a formal process to abandon packages?
2020-06-29 19:05:25 <ikke> not very formal
2020-06-29 19:05:39 <adu> is there a metric about how many downloads packages are getting?
2020-06-29 19:05:51 <maldridge> ikke: is there a process at all?
2020-06-29 19:06:14 <ikke> maldridge: similar to becomming a maintainer of a package ;-)
2020-06-29 19:06:19 <ikke> adu: no
2020-06-29 19:06:26 <ikke> adu: alpine does and cannot track that
2020-06-29 19:06:42 <ikke> there are lots of mirrors where people can download packages from
2020-06-29 19:06:52 <maldridge> do I just move it to unmaintained?  or is there an orphan@ "maintainer" I can list
2020-06-29 19:07:08 <maxice8> Leave the field empty
2020-06-29 19:07:10 <maxice8> # Maintainer:
2020-06-29 19:07:29 <maldridge> need a revbump to regenerate the database?
2020-06-29 19:07:35 <mps> maldridge: if you are maintainer just create MR where you remove self from APKBUILD
2020-06-29 19:07:37 <ikke> mps: yes
2020-06-29 19:07:41 <ikke> maldridge: *
2020-06-29 19:08:28 <ikke> maldridge: We do not have a list or some anouncement of packages becomming unmaintained
2020-06-29 19:08:55 <maldridge> got it
2020-06-29 19:09:31 <ikke> depending how much a package is used, you could opt to leave it in the same repo for the time being
2020-06-29 19:09:53 <maldridge> I don't know if anyone besides the group I was with used it, probably best to leave it for now
2020-06-29 19:10:04 <maldridge> it was in testing anyway
2020-06-29 19:10:11 <ikke> ah ok
2020-06-29 19:10:25 <ikke> then it would make more sense to move it to unmaintained
2020-06-29 19:10:42 <maldridge> they already seem to believe that I'm some evil open source villain, I'd rather not give them anything they could use to substantiate that claim
2020-06-29 19:11:45 <ikke> alpine is luckily still small enough to not require a lot of red tape to get things done
2020-06-29 19:11:58 <maldridge> yeah, its something I highly respect within the project
2020-06-29 19:16:04 <maldridge> I'd like to think that void is still that low friction, but I also realize the reality that void is getting too big to keep the tiny feel
2020-06-29 19:23:45 <adu> where are CHOST, CTARGET documented?
2020-06-29 20:07:18 <bovis> Running 'hdajacksensetest' gives me this output, showing it's not recognizing a plugged in microphone (this is a 2-in-1 3.5mm jack for headphone/mic).
2020-06-29 20:07:39 <bovis> http://sprunge.us/qX6skz
2020-06-29 20:07:56 <bovis> Suggested in various forums: Placing 'options snd_hda_intel index=0 model=[model]' in /etc/modprobe.d/alsa-base.conf doesn't help because modprobe doesn't recognize the model. http://sprunge.us/H9VKpl
2020-06-29 20:08:22 <bovis> Everything else with the sound card seems to work fine. How can I get the hotplugging to work?
2020-06-29 20:08:35 <programmerq> I wonder if it could be due to AHJ vs OMTP (not all TRRS jacks are equal)
2020-06-29 20:08:50 <programmerq> the mic and ground pin are flipped between those two
2020-06-29 20:09:11 <programmerq> I usually keep an adapter around in case I run into an issue like that. some devices seem to care, others don't.
2020-06-29 20:10:39 <programmerq> I had one device that wouldn't recognize my mic on my main headset no matter what. ended up getting a separate usb sound device and used that instead (which did correctly detect the mic).
2020-06-29 20:21:39 <bovis> programmerq: You're talking about the pins in the headphones?
2020-06-29 20:21:49 <programmerq> yeah
2020-06-29 20:25:27 <bovis> I have two pairs of headphones (w/ mic), one adapter for those headphones that splits them into separate green/red plugs, and another set of headphones that's got a bespoke split into green/red plugs. None of them are recognized as plugged in.
2020-06-29 20:48:11 <bovis> programmerq: When you say you got a USB sound device, you're referring to USB-based headphones? Was it trivial to get Alpine to recognize that connection?
2020-06-29 20:49:13 <programmerq> I got a USB sound device with a TRRS jack built in. I didn't face that specific microphone-not-recognized issue on a linux device, but your issue sounded a lot like teh issue I had
2020-06-29 20:49:44 <programmerq> but in general that particular USB sound device has worked quite well on a large variety of systems
2020-06-29 20:50:01 <programmerq> I've successfully used it on mac, windows, linux, and even android with a usb c->usb A adapter.
2020-06-29 20:51:23 <Hello71> what's the url for ancient alpine releases?
2020-06-29 20:51:36 <Hello71> ... wait, it is "ancient"
2020-06-29 20:51:47 <Hello71> thanks
2020-06-29 20:53:24 <ikke> :)
2020-06-29 20:53:43 <ikke> quack
2020-06-30 03:27:51 <abenz> do any of the alpine releases have a kernel with vfio built in (not as an external module) ?
2020-06-30 03:28:55 <abenz> because if there is, I can simply specify vfio vfio-pci.ids=... in grub at boot, to grab a gpu for passthru without needing to touch any /etc/modules
2020-06-30 03:55:26 <amom> abenz: when I used alpine for pci passthrough, I didnt need to worry about that
2020-06-30 03:55:33 <amom> I just used the standard release of alpine
2020-06-30 03:55:56 <amom> I just specified vfio vfio-pci.ids=... in grub
2020-06-30 03:57:02 <abenz> amom: I tried that but vfio-pci won't grab the gpu, afaik grub kernel parameters only work with built-in modules, don't they?
2020-06-30 03:58:21 <amom> abenz: Not sure but I dont think that's the reason for vfio-pci not grabbing the gpu
2020-06-30 03:58:39 <amom> did you remember to pass in the id of the audio device on your gpu?
2020-06-30 03:58:46 <abenz> yes
2020-06-30 03:59:04 <amom> lemme check the write up I did for pci passthrough. I might have something there for you
2020-06-30 03:59:30 <abenz> the way I;m doing it now, in /etc/modprobe.d/vfio.conf ... softdep pre amdgpu, vfio-pci grabs the gpu but not the audio
2020-06-30 03:59:40 <abenz> so yea if you know how to do it from grub it'd be much cleaner
2020-06-30 03:59:51 <amom> abenz:Ya, I do pci passthrough using grub too
2020-06-30 04:01:22 <amom> abenz: editting modprobe should not be needed if you enterred vfio-pci.ids correctly in the kernel boot parameters
2020-06-30 04:01:53 <amom> iirc, those get loaded and used before modprobe or amdgpu gets a chance
2020-06-30 04:02:46 <abenz> I entered them correctly. in fact, just specifying vfio in commandline should give some feedback in dmesg about VFIO being loaded, in my case that never happens unfortunately
2020-06-30 04:02:55 <amom> abenz: In my grub, I wrote `vfio-pci.ids=1002:67df,1002:aaf0` as my kernel parameter
2020-06-30 04:03:06 <abenz> what else?
2020-06-30 04:03:16 <abenz> can you pastebin or send the whole thing in a pm please?
2020-06-30 04:03:26 <amom> abenz: sure
2020-06-30 04:03:31 <abenz> also intel or amd ?
2020-06-30 04:03:37 <amom> amd for gpu and cpu
2020-06-30 04:03:41 <abenz> nice
2020-06-30 04:05:04 <amom> abenz: I sent a personal message with my grub config
2020-06-30 04:05:31 <amom> also, here is a guide I wrote for pci passthrough. It's for void linux but should apply to alpine as well - https://github.com/void-linux/void-docs/pull/217/files
2020-06-30 04:05:40 <abenz> cheers
2020-06-30 04:05:44 <amom> I wrote it after I got tired of alot of the outdated guides on pci passthrough
2020-06-30 04:06:02 <amom> Good luck and remember to backup the configs you hasseled around with! abenz
2020-06-30 04:06:18 <abenz> Always do :)
2020-06-30 06:57:20 <abenz> amom: still can't figure out why vfio won't load
2020-06-30 06:57:46 <abenz> btw I'm using diskless mode, EFI system (so only updating /boot/grub/grub.cfg)
2020-06-30 06:58:10 <amom> abenz: ah, thats probably why
2020-06-30 06:58:24 <amom> you should never ever ever edit /boot/grub/grub.cfg
2020-06-30 06:58:28 <abenz> other parameters (like iommu=pt) are recognized and I see them reflected in dmesg. Just vfio wont show anything..
2020-06-30 06:58:41 <amom> if you can, please edit the kernel boot parameters when you reach the grub screen on boot
2020-06-30 06:58:56 <abenz> I did, still nothing
2020-06-30 06:58:59 <amom> it's very easy to make a mistake and put your kernel parameters in the wrong if statement
2020-06-30 06:59:06 <amom> thats odd....
2020-06-30 06:59:41 <amom> what audio driver is the kernel loading for your gpu audio?
2020-06-30 07:00:01 <abenz> when I specify nomodeset, then amdgpu doesn't load (so gpu is free to be grabbed), yet vfio still doesn't. After I login to the system, if I modprobe vfio and specify the gpu ids, it loads sucessfully!
2020-06-30 07:00:13 <abenz> snd_hda_intel
2020-06-30 07:00:39 <amom> abenz: hmmm. I may have been wrong
2020-06-30 07:00:51 <amom> maybe take a look at the modprobe /etc config after all
2020-06-30 07:01:49 <abenz> oh noes, I thought I could e into grub and change my machine from personal to work on the fly
2020-06-30 07:26:31 <abenz> amom: it finally worked, by setting softdep in modprobe for: drm, amdgpu, snd_hda_intel
2020-06-30 07:28:20 <abenz> I will save it as a apkovl now and swap later as needed
2020-06-30 07:28:33 <abenz> thanks for your help
2020-06-30 07:30:25 <amom> abenz: no problem
2020-06-30 07:30:43 <amom> If you can add your knowledge to the wiki, that would be great =)
2020-06-30 07:32:42 <abenz> will do
2020-06-30 07:38:13 <abenz> amom: did you run qemu with passthru as root?
2020-06-30 07:38:52 <amom> abenz: you have to change a setting in libvirtd and add yourself to that group to get it working as a normal user
2020-06-30 07:38:52 <abenz> my user is in qemu and kvm groups, but I get permission errors (failed to open /dev/vfio/17: Permission denied)
2020-06-30 07:39:02 <amom> did you log out and log back in?
2020-06-30 07:39:32 <abenz> i dont have libvirtd installed, just qemu (minimal)
2020-06-30 07:39:59 <abenz> yes, rebooted actually (to grab the gpu) :)
2020-06-30 07:40:21 <amom> what are the permissions of /dev/vfio/17?
2020-06-30 07:41:51 <abenz> crw-------    1 root     root
2020-06-30 07:42:10 <amom> hmmm. I would look into how that device's permissions are set
2020-06-30 07:42:21 <amom> usually udev sets them for /dev
2020-06-30 07:42:30 <amom> but since this is a vfio device, I dont know
2020-06-30 07:42:41 <amom> Ive never gone that minimal with my vfio setup.
2020-06-30 07:43:01 <amom> Well, Im lying, I did once but I got sick of it after 4 months cause I needed features offered by virt-manager
2020-06-30 07:44:08 <abenz> heh, I was actually setting up virt-manager the other day (with remote control from a laptop) but then one mentioned minimalism.. and since I;m on alpine.. :)
2020-06-30 07:44:24 <amom> if you ever get tired of doing it the minimal way, my guide for voidlinux shows how to set the permissions for libvirtd
2020-06-30 07:44:48 <amom> ya, alpine is great for pci passthrough
2020-06-30 07:45:14 <amom> I loved it but I unfortunately need matlab for school so Im stuck on glibc
2020-06-30 07:47:17 <amom> the void guide should work on alpine too btw
2020-06-30 07:47:24 <amom> not plugging void.
2020-06-30 07:47:53 <abenz> no worries
2020-06-30 07:54:38 <steevec> Hi there
2020-06-30 07:55:09 <mokulus> hi
2020-06-30 07:55:38 <amom> steevec: hello
2020-06-30 07:56:06 <steevec> I've increased ulimit rules in /etc/conf.d/docker, but now docker service failed with the following error: prot_exec|prot_write failed
2020-06-30 07:56:19 <steevec> do you know how can I fix that?
2020-06-30 07:56:42 <amom> steevec: This might be wrong but I think the kernel also has a ulimit.
2020-06-30 07:57:11 <amom> so docker might be hitting the limit and the kernel might be saying you've reached the limit.
2020-06-30 07:57:18 <amom> You may need to increase kernel's ulimit
2020-06-30 07:58:30 <steevec> do you know what's the equivalent of /etc/security/limits.conf on alpine ?
2020-06-30 07:59:02 <steevec> looks like this file doesn't exists
2020-06-30 07:59:20 <steevec> or where to set kernel ulimit?
2020-06-30 07:59:35 <amom> steevec: this is what I could find http://polarhome.com/service/man/?qf=limits.conf&af=0&tf=2&of=Alpinelinux
2020-06-30 08:00:23 <abenz> steevec: I'm facing this exact issue (although with gpu passthru run as a user, not docker)
2020-06-30 08:00:45 <abenz> there's no /etc/security/..., so I will create that folder and see if the kernel honours it
2020-06-30 08:01:11 <steevec> same here, let's create it and see what happend
2020-06-30 08:01:33 <steevec> abenz: I'll share my knowledge here
2020-06-30 08:01:46 <abenz> yup
2020-06-30 08:02:02 <amom> steevec: you can set ulimit temporarily too
2020-06-30 08:02:09 <amom> just to see if it fixes your issue
2020-06-30 08:02:28 <amom> using the ulimit command
2020-06-30 08:03:10 <steevec> that's great, I'm going to start with that then, thanks amom to point it out
2020-06-30 08:17:22 <abenz> limits.conf makes no difference in my case
2020-06-30 08:28:08 <TBB> it's a PAM thing, and Alpine by default doesn't use PAM
2020-06-30 08:28:14 <steevec> abenz: same here
2020-06-30 08:28:40 <abenz> TBB: where can we specify memlock ulimits?
2020-06-30 08:29:06 <steevec> I guess we should install https://pkgs.alpinelinux.org/package/edge/main/x86/linux-pam
2020-06-30 08:29:27 <abenz> there has to be a simpler way
2020-06-30 08:31:04 <steevec> abenz: it's not a full linux kernel it's just a module lib
2020-06-30 08:31:59 <steevec> I guess ulimits are by default defined during kernel config
2020-06-30 08:32:05 <steevec> not sure by the way
2020-06-30 08:32:25 <steevec> it would be nice if someone can confirme
2020-06-30 08:32:25 <amom> steevec: I think you can also write to a file the ulimits
2020-06-30 08:32:28 <steevec> confirm*
2020-06-30 08:32:58 <steevec> amom: I'd like to… the point is: in which file? ^^
2020-06-30 08:33:13 <amom> uhhhhh. I dont know off the top of my head
2020-06-30 08:33:14 <amom> let me see
2020-06-30 08:34:10 <amom> https://www.linuxtechi.com/set-ulimit-file-descriptors-limit-linux-servers/
2020-06-30 08:34:15 <amom> /proc/sys/fs/file-max
2020-06-30 08:42:28 <abenz> thats not related to ulimit
2020-06-30 08:42:36 <abenz> I was able to set ulimit for my user
2020-06-30 08:42:41 <abenz> hacky way though
2020-06-30 08:42:46 <abenz> su root
2020-06-30 08:42:57 <abenz> ulimit -l <set amount in kb>
2020-06-30 08:43:04 <abenz> su <your user>
2020-06-30 08:43:43 <steevec> abenz: what about the next reboot?
2020-06-30 08:44:06 <abenz> gonee
2020-06-30 08:44:20 <abenz> not even reboot, if you switch to another terminal its gone
2020-06-30 08:44:49 <steevec> so it's kinda complicate to set it for a given service then
2020-06-30 08:45:25 <steevec> amon setting /proc/sys/fs/file-max doesn't work after reboot too
2020-06-30 08:45:55 <amom> steevec: Oh good to know
2020-06-30 08:46:06 <amom> sorry if I let you onto  the wrong patch
2020-06-30 08:46:09 <amom> it was a guess
2020-06-30 08:46:31 <amom> I ssh'd into my alpine phone
2020-06-30 08:46:41 <amom> alpine should have sysctl enabled by default
2020-06-30 08:47:08 <amom> https://www.linuxtechi.com/set-ulimit-file-descriptors-limit-linux-servers/
2020-06-30 08:47:18 <amom> this website explains how to set sysctl settings
2020-06-30 09:09:39 <steevec> well, I've updated `fs.file-max` which persist the result of /proc/sys/fs/file-max even after reboot. But my services still fail with: PROT_EXEC|PROT_WRITE failed.
2020-06-30 09:09:46 <steevec> I don't know why
2020-06-30 09:38:50 <mps> anyone know is there mediatek IRC channel somewhere. DDG search can't find anything
2020-06-30 10:03:13 <jhass> alis knows #linux-mediatek
2020-06-30 10:19:30 <mps> jhass: thanks
2020-06-30 10:34:23 <EF> hello. is booting from a non-uefi gpt drive doable in Alpine please?
2020-06-30 10:35:13 <TBB> it is. gpt is by no means uefi-specific.
2020-06-30 10:35:13 <mps> EF: yes
2020-06-30 10:37:23 <EF> I am struggling. Got an HP Gen8 Microserver and straight install to sda works, installing grub into sda1 bios_grub with /boot on a raid1 and / on raid10, get a non-system disk error and no grub. Any suggestions please?
2020-06-30 10:38:44 <mps> try with syslinux instead of grub
2020-06-30 10:40:00 <EF> Can syslinux boot an md based /boot? Take it still need a 1M bios_boot partition still?
2020-06-30 10:43:32 <mps> huh, much time passed when I worked with linux raid, so can't tell anything useful because I forgot nearly all about booting
2020-06-30 10:45:55 <EF> It isn't a raid thing, it may be a grub thing.
2020-06-30 10:46:38 <EF> so creating a /boot partition as raid1 and an everything else on raid10 should be okay for syslinux?
2020-06-30 10:47:42 <mps> not sure, but afaik syslinux read boot record from disk like grub, so it should work but again not sure
2020-06-30 10:48:25 <mps> meh, syslinux is in boot record :)
2020-06-30 10:50:00 <EF> okay, going to try. Thanks for the advice
2020-06-30 10:52:37 <mps> hehe, this is my method, try and try again ... :)
2020-06-30 11:54:32 <EF> mps: well it has installed but the raid arrays have renamed themselves but Alpine is aline!
2020-06-30 12:03:46 <mps> EF: well, if it works such cosmetic 'things' are not important :)
2020-06-30 12:20:24 <EF> indeed. It just causes a few error messages when booting and could cause issues if root was on an array too.
2020-06-30 12:20:50 <EF> Just need to sort crypto out and EF will be a very happy bunny. Thanks again.
2020-06-30 12:23:42 <EF> Oh and for the record, syslinux does not require a 1MB partition on a gpt drive so that must be a grub thing.
2020-06-30 12:30:11 <Hello71> syslinux always uses embedding
2020-06-30 12:30:20 <Hello71> grub doesn't use embedding unless you specify grub-install --force
2020-06-30 12:37:43 <EF> Hopefully I can find the time to do an install writeup and share it. Even Arch's wiki isn't clear about that.
2020-06-30 12:51:32 <id4rk> Hello, how can I see all the partitions on Alpine!?
2020-06-30 12:51:46 <abenz> fdisk -l
2020-06-30 12:51:51 <abenz> or apk add lsblk
2020-06-30 12:51:54 <abenz> lsblk
2020-06-30 12:52:41 <id4rk> Fisk
2020-06-30 12:52:54 <id4rk> fdisk -l helped
2020-06-30 12:53:02 <id4rk> thx
2020-06-30 12:53:34 <id4rk> The Swap partition is there.
2020-06-30 12:53:58 <id4rk> Before starting the install I used SWAP_SIZE=0 in command line
2020-06-30 12:56:05 <id4rk> Ah... I forgot to write “export” before
2020-06-30 12:57:31 <id4rk> export SWAP_SIZE=0
2020-06-30 12:57:49 <id4rk> Is that correct?
2020-06-30 13:12:51 <Hello71> better to do SWAP_SIZE=0 setup-disk or SWAP_SIZE=0 setup-alpine
2020-06-30 13:17:53 <id4rk> I just reinstalled it with export, it worked but I take note of your tip.
2020-06-30 13:41:31 <id4rk> Many of the tutorials of the wiki are outdated.
2020-06-30 13:41:46 <id4rk> Or half made
2020-06-30 14:03:38 <EF> Can syslinux copy with opening a Luks encrypted / if it has a seperate /boot?
2020-06-30 14:03:49 <EF> -copy +cope
2020-06-30 14:11:21 <EF> Digging into the wiki, the answer appears to be yes but there are a lack of non-uefi gpt disk examples. C'est la vie
2020-06-30 14:44:32 <FesterJester> Hello all! Anyone know how to setup a TFTP server on alpine 3.11? I installed tftp-hpa, but have found no docs to give me an idea how to proceed.
2020-06-30 14:45:43 <FesterJester> I am also open to using a different server package, doesn't have to be tftp-hpa
2020-06-30 14:45:49 <ikke> I used dnsmasq of tftp
2020-06-30 14:45:53 <ikke> for*
2020-06-30 14:46:22 <mps> yes, dnsmasq is better and complete
2020-06-30 14:46:30 <mps> also I used it
2020-06-30 14:47:01 <FesterJester> Ok, dnsmasq it is. I will see if I can get that running. :D
2020-06-30 14:47:47 <ikke> should be fairly straight-forward
2020-06-30 14:48:39 <FesterJester> Well, off to a decent start. Found the dnsmasq.conf file
2020-06-30 14:50:10 <ikke> enable-tftp
2020-06-30 14:50:43 <ikke> tftp-root
2020-06-30 14:53:34 <EF> kill off dhcp and dns parts though.
2020-06-30 14:54:48 <FesterJester> Oh, trust me. That is why I am reading the ENTIRE config file. Did that once in a small office setup and it took me a while to solve that problem.
2020-06-30 14:55:08 <FesterJester> Just not with the dnsmasq service
2020-06-30 16:55:49 <moosl> hei
2020-06-30 16:55:50 <moosl> question
2020-06-30 16:56:03 <moosl> which apk do I need in order to see flash content?
2020-06-30 16:56:19 <FesterJester> flash content?
2020-06-30 16:56:30 <FesterJester> like adobe flash?
2020-06-30 16:56:43 <moosl> glibc/windoze got the option of 'activate flash' on web content
2020-06-30 16:57:54 <moosl> and systemDrown based distros
2020-06-30 16:58:02 <mps> is flash open sourced?
2020-06-30 16:58:25 <mps> no, it is not, afaik
2020-06-30 16:58:52 <FesterJester> no, not yet anyway
2020-06-30 16:58:56 <moosl> by any means,
2020-06-30 16:58:57 <mps> only option is to ask 'upstream' (huh ;) ) to build it on musl
2020-06-30 16:59:11 <FesterJester> Ubuntu has/had icedtea
2020-06-30 16:59:27 <moosl> or some open sauce.
2020-06-30 16:59:45 <FesterJester> some kind of open source version of adobe flash
2020-06-30 16:59:46 <moosl> it's an outdated flash an simple content, but I need to see nonetheless
2020-06-30 17:00:37 <mps> FesterJester: maybe you are looking for this 'apk info icedtea-web'
2020-06-30 17:00:44 <FesterJester> I do see icedtea in apk
2020-06-30 17:01:25 <moosl> that's java
2020-06-30 17:01:28 <FesterJester> Not sure if you will have any luck with that, but good luck
2020-06-30 17:02:04 <moosl> FesterJester: I doubt 'good luck' got anything to do with having a working apk
2020-06-30 17:02:31 <moosl> linuxmint, ubuntu, etc, works everywhere
2020-06-30 17:02:59 <mps> moosl: alpine is musl based
2020-06-30 17:03:19 <mps> all other you mention are glibc based
2020-06-30 17:04:30 <mps> not all upstream know/care to make their software to work musl, some are even opposed to that
2020-06-30 17:04:44 <mps> especially commercial ones
2020-06-30 17:04:45 <moosl> oohh I see. it's an upstream issue
2020-06-30 17:05:03 <moosl> mps: what about with some sort of wrapper?
2020-06-30 17:05:19 <moosl> musl <-> glibc wrapper?
2020-06-30 17:05:37 <mps> there is glibc-compat but it is not much useful, ime
2020-06-30 17:05:38 <FesterJester> Yes, I know I have trouble with MariaDB and it is not an Alpine problem
2020-06-30 17:05:44 <moosl> I believe flash is just a .so file
2020-06-30 17:31:51 <nshp> no stable release yet, so no apk, but you could build https://lightspark.github.io/
2020-06-30 17:32:41 <nshp> GNU's gnash is now unmaintained I think, but it also supported enough of flash to make simple things work
2020-06-30 17:50:59 <moosl> nshp: lightspark, that was the one
2020-06-30 17:51:38 <moosl> lolol!  I can try compile it.
2020-06-30 18:13:54 <moosl> firefox 78.0 out
2020-06-30 18:14:50 <mps> yes
2020-06-30 18:15:06 <moosl> does apk add new version  or I have to wait for a new release?
2020-06-30 18:16:04 <mps> when the maintainer find time and fix potential issue it will be updated for alpine
2020-06-30 18:16:22 <mps> Cogitri: ^
2020-06-30 18:16:49 <moosl> how to refresh apk tree and is there a way to show available updates?
2020-06-30 18:17:03 <Cogitri> I'll probably take a look tomorrow
2020-06-30 18:17:11 <ikke> moosl: what do you mean with the apk tree?
2020-06-30 18:18:05 <mps> Cogitri: no need to hurry
2020-06-30 18:18:48 <mps> I just mentioned to you that, saying 'when maintainer find time'
2020-06-30 18:19:20 <moosl> ikke: apk repo,
2020-06-30 18:19:42 <moosl> ikke: that option that shows up available version upgrades of programs
2020-06-30 18:19:49 <hodapp> oh dear, flash
2020-06-30 18:19:52 <mps> aports repo
2020-06-30 18:20:16 <moosl> mps: the command is 'aports repo'?
2020-06-30 18:20:22 <ikke> sno
2020-06-30 18:20:54 <mps> aports repo is git repository
2020-06-30 18:21:19 <mps> apk is/are built packages
2020-06-30 18:21:25 <ikke> moosl: do you want to update your local system, or are you working with the APKBUILDS and wnat the latest versions of those?
2020-06-30 18:21:41 <moosl> got it --> update    Update repository indexes from all remote repositories
2020-06-30 18:21:50 <moosl> from 'apk --help'
2020-06-30 18:21:52 <ikke> right
2020-06-30 18:21:54 <ikke> so the former
2020-06-30 18:22:16 <mps> :)
2020-06-30 18:22:40 <ikke> apk version -l '<' should show you the packages that can be updated
2020-06-30 18:22:50 <moosl> what about, no 'pretend' option?
2020-06-30 18:22:56 <ikke> yes
2020-06-30 18:22:57 <moosl> ikke: ah ok
2020-06-30 18:23:04 <ikke> but you need to run apk udpate first
2020-06-30 18:23:10 <moosl> I did
2020-06-30 18:23:13 <ikke> (apk udpate does not update any packages yet)
2020-06-30 18:23:18 <moosl> ikke: with '<' included?
2020-06-30 18:23:22 <ikke> there is apk upgrade -s as well
2020-06-30 18:23:23 <ikke> yes
2020-06-30 18:24:33 <moosl> just one, so left package is OLD and right is NEW?
2020-06-30 18:24:49 <mps> hmm, 'apk version' is enough in most cases
2020-06-30 18:24:52 <moosl> somehow the arrew < pointing the old version is not that intuitive
2020-06-30 18:24:53 <ikke> mps: correct
2020-06-30 18:25:13 <ikke> mps: heh, didn't even know that :D
2020-06-30 18:25:20 <ikke> and the output is exactly the same
2020-06-30 18:25:27 <moosl> wait,  < as 'bigger than'
2020-06-30 18:25:30 <mps> yes :)
2020-06-30 18:25:33 <ikke> < is smaller hten
2020-06-30 18:25:35 <ikke> then
2020-06-30 18:25:44 <ikke> meaning, your local version is older than in the repo
2020-06-30 18:25:53 <moosl> so it's a math operand rather than pointing the direction of the upgrade.
2020-06-30 18:25:59 <ikke> correct
2020-06-30 18:26:31 <moosl> not quite n00b friendly but ok.
2020-06-30 18:26:52 <ikke> moosl: mps pointed out that `apk version` exists that gives the same output
2020-06-30 18:29:05 <moosl> so for instance now it shows gtk+ upgrade to 3.24.21-r0
2020-06-30 18:29:26 <moosl> but 'apk add gtk+' installed 2.24.31-r1 instead
2020-06-30 18:29:42 <maxice8> gtk+3.0 ?
2020-06-30 18:29:48 <ikke> apk add does not upgrade
2020-06-30 18:30:21 <moosl> im trying to remove it --> ERROR: No such package: gtk+2.24.31-r1
2020-06-30 18:30:42 <ikke> apk del gtk+2
2020-06-30 18:30:50 <ikke> apk del gtk+
2020-06-30 18:31:51 <mps> 'apk add -u pkgname' upgrade pkg
2020-06-30 18:32:23 <moosl> mps: damn! your genious, it worked.
2020-06-30 18:32:43 <ikke> mps: wow, two things I've learned today :D
2020-06-30 18:33:38 <mps> moosl: no, I'm not genius, just learned to use '-h' switch for most 'things'. 'apk add -h'
2020-06-30 18:34:02 <moosl> ah alright, that's a subhelp option
2020-06-30 18:35:20 <moosl> is busybox hard coded or alpine can use util-linux?
2020-06-30 18:35:31 <ikke> it can use util-linux
2020-06-30 18:35:53 <moosl> not that I needed, but out of curiosity
2020-06-30 18:35:59 <mps> ikke: when you switch your workstation to alpine your knowledge will shine :)
2020-06-30 18:49:30 <adu> Is there a formal process of submitting new packages to edge/community?
2020-06-30 18:49:47 <ikke> Yes: the process is: 1) make a merge request
2020-06-30 18:49:50 <ikke> done
2020-06-30 18:50:23 <mps> adu: no in community but first in testing
2020-06-30 18:50:45 <ikke> right packages should generally start in testing
2020-06-30 18:51:01 <ikke> There can be exceptions if it's suddently required as a dependency for example
2020-06-30 18:51:14 <adu> Do you think zck and car would be tools other people might want?
2020-06-30 18:51:25 <ikke> I don't know them, so no clue
2020-06-30 18:51:26 <adu> Xar*
2020-06-30 18:52:08 <adu> They're useful in unzipping fedora and macos packages respectively
2020-06-30 18:52:42 <ikke> They might be usefull in pipelines
2020-06-30 18:52:55 <mps> I don't think of these as much useful for alpine
2020-06-30 18:53:16 <mps> and I don't understand why we have dpkg
2020-06-30 18:53:32 <ikke> mps: CI/CD
2020-06-30 18:53:38 <mps> also, rpm and xbps
2020-06-30 18:54:11 <adu> That's incredibly powerful
2020-06-30 18:54:17 <mps> ikke: yes, I understand but still ...
2020-06-30 18:54:24 <ikke> It's nice to be able to build packages in alpine docker pipelines
2020-06-30 18:54:36 <ikke> as long as they do not require compilation
2020-06-30 18:54:37 <adu> To be the distro people use to build other distros
2020-06-30 18:55:09 <ikke> mps: I'm quite happy that those tools are available
2020-06-30 18:55:21 <mps> understand
2020-06-30 18:55:39 <mps> but I'm to conservative, as you probably know :)
2020-06-30 18:55:45 <mps> too*
2020-06-30 18:56:59 <mps> and actually I'm not against those pkgs, just didn't used them so my feeling comes from zero experience
2020-06-30 18:58:33 <FesterJester> So, I have a MariaDB server up and running, I can login locally, but I cannot login from a remote client. What have I missed?
2020-06-30 18:58:39 <mps> adu: yes, I think some users will find those pkgs quite useful to have them on alpine
2020-06-30 18:59:24 <ikke> FesterJester: most likely is that it listens ons a unix socket instead of on a TCP port
2020-06-30 19:09:41 <FesterJester> ikke: How do I change it to tcp port on Alpine?
2020-06-30 19:09:50 <ikke> Same as on any other system
2020-06-30 19:09:56 <ikke> in the mysql config
2020-06-30 19:11:51 <FesterJester> I already have bind-address=0.0.0.0 and port=3306
2020-06-30 19:15:35 <FesterJester> Figured it out. Missed the "skip-networking" entry in my config
2020-06-30 19:15:53 <ikke> :)
2020-06-30 21:05:03 <moosl> can anybody reccomend a good ftp client?
2020-06-30 21:07:52 <c705> lftp
2020-06-30 21:10:25 <FesterJester> +1
2020-06-30 21:11:06 <moosl> c705: tahnks, i'll try it
2020-06-30 21:47:08 <amom> I can see acme.sh is available for x86_64 but when I do `apk add acme.sh` I get unsatisfiable constraints
2020-06-30 21:48:43 <afontain_> amom: sanity check: `wc -l /etc/apk/world`
2020-06-30 21:49:56 <amom> afontain_: 14
2020-06-30 21:50:36 <afontain_> ok, grat
2020-06-30 21:50:39 <afontain_> *great
2020-06-30 21:50:57 <afontain_> can you be more specific on the error message?
2020-06-30 21:51:23 <afontain_> what's the error exactly? (hint: paste service)
2020-06-30 21:52:14 <amom> acme.sh (missing):
2020-06-30 21:52:17 <amom>     required by: world[acme.sh]
2020-06-30 21:53:41 <Cogitri> I guess you haven't added the community repo to /etc/apk/repositories
2020-06-30 21:53:50 <amom> Cogitri: ah, thank you
2020-06-30 21:53:51 <afontain_> weird
2020-06-30 21:53:54 <amom> That is the issue
2020-06-30 21:53:55 <afontain_> ah
2020-06-30 22:39:54 <MY-R> is it me or google services are overloaded? long connection times to their apis cdns etc and many dns records return after few seconds from 8.8.8.8 server?