2020-12-01 00:00:01 <Hello71> you didn't put --initdb
2020-12-01 00:00:07 <Hello71> but why don't you just use setup-alpine
2020-12-01 00:01:16 <asddsa> Doesn't make a difference:  https://imgur.com/a/7PkMPYQ
2020-12-01 00:01:23 <asddsa> Full disk encryption
2020-12-01 00:01:35 <asddsa> setup-alpine can't handle that
2020-12-01 00:03:34 <Hello71> setup-disk on a path then
2020-12-01 00:04:38 <asddsa> Thanks !
2020-12-01 00:04:43 <asddsa> Didn't know that !
2020-12-01 00:05:03 <asddsa> Did that command ever work ?
2020-12-01 00:06:04 <Hello71> https://git.alpinelinux.org/alpine-conf/commit/setup-disk.in?id=2d0772e8f087883d1e29f995591d57545545719a
2020-12-01 00:06:55 <asddsa> I was talking about ``` apk add --root=/mnt/root ```
2020-12-01 00:07:29 <Hello71> could be, if you add more flags
2020-12-01 03:51:49 <amom> hi guys, I tried installing sxhkd, I got an error that a library provided by xcb-util-keysyms wasnt available. Turns out, xcb-util-keysyms just wasnt available in the aarch64 repo. I built xcb-util-keysyms locally with abuild -r and sxhkd worked.
2020-12-01 03:52:44 <amom> Just thought Id let you guys know. Is there any reason xcb-util-keysyms is not in the aarch64 repos and anything I can do to help get it in there?
2020-12-01 03:54:25 <halosghost> amom: interesting
2020-12-01 03:57:18 <halosghost> amom: I wonder if the build failed for aarch64 or something, the official APKBUILD lists "all" in archs
2020-12-01 03:59:47 <hechos> on edge xorg-server-1.20.9-r1[so:libnettle.so.8] was broken?
2020-12-01 04:01:12 <amom> halosghost: I think it is the build failed
2020-12-01 04:08:20 <halosghost> hmm
2020-12-01 04:08:26 <halosghost> similar problem for otf-atkinson-hyperlegible
2020-12-01 04:08:32 <halosghost> seems to have only built for some platforms
2020-12-01 04:08:35 <halosghost> strange
2020-12-01 04:10:02 <halosghost> ikke: is there a way to request a rebuild of a package that's already in the repos?
2020-12-01 04:33:59 <amom> submitted a bug report: https://gitlab.alpinelinux.org/alpine/aports/-/issues/12152
2020-12-01 04:37:42 <halosghost> amom: oh hey, another pbp
2020-12-01 04:37:50 <halosghost> amom: how's it working for you so far?
2020-12-01 04:38:01 <amom> halosghost: I have broken the casing
2020-12-01 04:38:06 <amom> very upset about that
2020-12-01 04:38:11 <maxice8> aarch64 builder seems to be stuck
2020-12-01 04:38:15 <halosghost> amom: ):
2020-12-01 04:38:17 <maxice8> trying to build zabbix
2020-12-01 04:38:25 <halosghost> amom: I'm sorry to hear that
2020-12-01 04:38:33 <amom> It's ok. I was a bit frustrated about alot of stuff not working on aarch64 that I bought an x230
2020-12-01 04:38:48 <amom> Im gonna buy the case replacement. it's only 20$
2020-12-01 04:39:03 <amom> I love the battery life
2020-12-01 04:39:08 <halosghost> I'm slowly working on switching to the pbp as daily driver
2020-12-01 04:39:44 <halosghost> amom: I'm very excited about weight/bat-life
2020-12-01 04:42:00 <amom> halosghost: it's also very libre
2020-12-01 04:42:09 <halosghost> amom: I appreciate that as well :)
2020-12-01 04:42:57 <halosghost> amom: part of what I'm doing in the meantime is NIH-ing my own DE
2020-12-01 04:43:12 <halosghost> so that I'm running the absolute minimum OS
2020-12-01 04:44:03 <amom> halosghost: nice. I just end up using dwm
2020-12-01 04:44:17 <halosghost> amom: yeah, for now I'll be using a customized dwm
2020-12-01 04:44:22 <halosghost> amom: till I finish writing my own WM
2020-12-01 04:44:31 <halosghost> :P
2020-12-01 04:44:51 <amom> halosghost: do you have a serial cable?
2020-12-01 04:45:08 <halosghost> amom: yeah
2020-12-01 04:45:24 <halosghost> I've found it very useful; glad I got it :)
2020-12-01 04:46:22 <amom> halosghost: I tried making pinebookpro boot linux-edge kernel instead of the manjaro fork or whatever. I couldnt get it to work - I spent an entire weekend. Turns out it would help alot if I had the serial cable. Would you be interested in getting it working?
2020-12-01 04:46:29 <amom> the current kernel is outdated
2020-12-01 04:47:00 <halosghost> amom: I'd be happy to try
2020-12-01 04:47:33 <amom> halosghost: here is an example of a device that uses the linux-edge kernel on pmos - https://gitlab.com/postmarketOS/pmaports/-/tree/master/device/testing/device-cubietech-cubieboard
2020-12-01 04:47:50 <amom> It's what I was told to reference when getting pbp to boot linux-edge
2020-12-01 04:48:32 <amom> Here is the latest pbp MR where they ask for the linux-edge kernel: https://gitlab.com/postmarketOS/pmaports/-/merge_requests/1629
2020-12-01 04:48:33 <halosghost> amom: I'd probably want some more specific instruction of what to try, but I'm also not really up to do this right now
2020-12-01 04:48:49 <halosghost> amom: if you're free on the weekends, I have much more energy ($DAYJOB is kind of draining right now)
2020-12-01 04:49:12 <amom> sure. Feel free to PM me on freenode.
2020-12-01 04:49:49 <amom> but basically, you would need to edit the pinebookpro-device APKBUILD and add the linux-edge kernel, make sure the u-boot config, and extlinux config is proper
2020-12-01 04:50:25 <amom> Take care and thank you for trying =)
2020-12-01 04:50:33 <halosghost> I love the idea of being on mainline
2020-12-01 04:50:43 <halosghost> frankly, at that point, I'd probably do pure alpine
2020-12-01 04:50:52 <halosghost> amom: will do :)
2020-12-01 04:51:02 <amom> ehhhh. I think pmos has alot of useful features
2020-12-01 04:51:05 <amom> I love pmbootstrap
2020-12-01 04:51:30 <amom> fde is super easy, I can cross compile, etc.
2020-12-01 04:52:21 <halosghost> I setup FDE via LUKS manually, and I've never used pmbootstrap :)
2020-12-01 04:52:27 <amom> also, the device configs are always sane. The mouse config on manjaro was terrible (too slow) but on pmos, it's just right. Id rather not spend time tweaking that
2020-12-01 04:52:48 <halosghost> and I've made a ton of virtual packages to get rid of some stuff that pmos seems to think are required (but are fundamentally not actually deps)
2020-12-01 04:52:57 <halosghost> 🤷 :)
2020-12-01 04:53:30 <amom> halosghost: ya, I agree. for some reason networkmanager is installed when i would just prefer wpa_supplicant
2020-12-01 04:53:44 <halosghost> yeah
2020-12-01 04:53:46 <amom> I guess I have to look into that. I could not uninstall networkmanager before
2020-12-01 04:54:00 <halosghost> yeah, alpine doesn't let you break the dep chain
2020-12-01 04:54:18 <halosghost> which I actually really like, but means that packages should be really judicious about what are listed as required dependencies
2020-12-01 04:55:13 <halosghost> but, you can hack around it by making an empty virtual package
2020-12-01 04:55:19 <halosghost> I've done it lots now :P
2020-12-01 04:56:01 <amom> interesting. But ya, I got into alpine via pmos so I have a slight bias. I now use alpine on my server, and pmos on phone and pbp
2020-12-01 04:56:23 <amom> I would elsewhere but proprietary applications require glibc =(
2020-12-01 04:56:57 <halosghost> yeah ):
2020-12-01 06:54:55 <ikke> maxice8: aarch64 seems to have finished :)
2020-12-01 06:56:22 <ikke> halosghost: which package needs to be rebuilt?
2020-12-01 07:34:24 <amom> ikke: https://gitlab.alpinelinux.org/alpine/aports/-/issues/12152
2020-12-01 07:35:05 <ikke> amom: I replied
2020-12-01 07:35:23 <amom> I see thank you ikke
2020-12-01 07:36:06 <mimi89999> How long does that openjdk build take?
2020-12-01 07:36:37 <ikke> mimi89999: not this long
2020-12-01 07:36:55 <mimi89999> Still building
2020-12-01 07:37:12 <mimi89999> It's now on `-r4`, but still
2020-12-01 07:40:14 <mimi89999> I see that `-r3` is not published. That means that it wither failed or was killed.
2020-12-01 07:40:28 <ikke> mimi89999: or skipped
2020-12-01 07:40:51 <mimi89999> When I was here yesterday it was on r3
2020-12-01 08:04:03 <ikke> midasi: openjdk8 is still continuing
2020-12-01 08:04:10 <ikke> all be it slowly
2020-12-01 09:11:04 <mimi89999> Why is there only one builder for each arch and version?
2020-12-01 09:11:45 <mimi89999> All other builders are IDLE. Wouldn't it make more sense to use that HW to build other packages?
2020-12-01 09:12:26 <ikke> Each arch has a single host that shares builders for all the different versions
2020-12-01 09:12:58 <ikke> each host is also specific to that arch, it cannot build packages for other arches
2020-12-01 09:13:42 <ikke> so having 3.x builders idle does not mean that the host is idle
2020-12-01 09:25:09 <ikke> mimi89999: note that we are looking into restructuring the build infra, but this is going to take time (after first having to decide how it will actually look like).
2020-12-01 10:07:22 <ikke> mimi89999: one issue with concurrent builders is that we want to prevent packages from being built and uploaded before their dependencies
2020-12-01 10:08:21 <ikke> so there needs to be some coordination as to what builder would build what
2020-12-01 10:08:58 <ikke> Right now, each builder is the cannonical source of the repository. With more concurrent builders, this should be coordinated as well
2020-12-01 10:18:06 <ikke> I killed openjdk, it now seemed to hang (FUTEX_WAIT)
2020-12-01 10:40:32 <mimi89999> There could be one server responsible for all arch and versions that would keep build a list of package build jobs that can be done at the moment (so only deps would be listed, but not the dependent package) and builders would contact the central server and pull those jobs.
2020-12-01 10:41:02 <mimi89999> Maybe something is wrong with that host that package builds are hanging?
2020-12-01 10:41:39 <ikke> mimi89999: yes, there are ways to solve these issues, but that requries logic that we need to write
2020-12-01 10:42:08 <ikke> mimi89999: we probably want to go to some kind of container-per-build kind of system
2020-12-01 10:45:56 <ikke> mimi89999: we are aware of these deadlocks and have nothing to do with the host
2020-12-01 10:58:58 <mimi89999> `build-edge-armv7` seems also stuck on openjdk8 and `build-edge-armhf` is failed.
2020-12-01 11:03:38 <ikke> armv7 is still working
2020-12-01 11:04:04 <ikke> openjdk takes time to build, so you can not immediately say it's stuck
2020-12-01 11:48:20 <mimi89999> Why are `build-3-13` attempted? All are failing.
2020-12-01 11:53:02 <ikke> mimi89999: a new released built from scratch, that usually comes with quite a bit of build issues that need to be fixed
2020-12-01 12:22:43 <mimi89999> Where is `firefox-esr` `aarch64` build?
2020-12-01 12:23:12 <mimi89999> It's not disabled and not in the repo.
2020-12-01 12:25:54 <ikke> https://pkgs.alpinelinux.org/package/edge/community/aarch64/firefox-esr
2020-12-01 13:03:41 <mimi89999> Shouldn't `78.5.0-r0` be available?
2020-12-01 13:21:46 <Cogitri> aarch64 is currently lagging behind so it probably didn't get to building it yet
2020-12-01 14:33:49 <hechos> hi guys.
2020-12-01 14:34:52 <hechos> ikke: you right have to define --repsotory in apk but --repositories-file function seems doesnt work on alpine in live enviroment.
2020-12-01 14:36:41 <hechos> well got another question of course comming for other os, and decide leave shell as default "ash shell" right?
2020-12-01 14:38:19 <hechos> this handle any kind of .profile like zsh. because auotlogin leave just in command prompt.
2020-12-01 14:38:30 <hechos> and need to autostart xorg.
2020-12-01 14:57:29 <hechos> sorry guys i found manual, ash.
2020-12-01 15:09:50 <mimi89999> ikke: Is that `openjdk8` build still running?
2020-12-01 15:09:55 <ikke> yes
2020-12-01 15:10:01 <ikke> I'm stracing it
2020-12-01 15:10:59 <mimi89999> 😮️
2020-12-01 15:12:07 <ikke> It's single threaded writing a lot of files which take several seconds each
2020-12-01 16:20:42 <TheFatal> hey guys. i have a simple question: yesterday i builded pure php 7.3.8 package and it places locally, now i need to build a php extention for this php how can i use my php package as dependency for php extention package ?
2020-12-01 16:24:56 <programmerq> you built php 7.3.8 via abuild? or a vanilla 7.3.8 php source?
2020-12-01 16:35:26 <TheFatal> via abuild, it created a package named alpine-fpm-7.3.8-r0.apk
2020-12-01 16:39:14 <ikke> TheFatal: You can add ~/packages (expanded) to /etc/apk/repositories
2020-12-01 16:40:37 <TheFatal> i already done it, how can i update repo list in APKBUILD file ?
2020-12-01 17:26:53 <TheFatal> @ikke solved my issue the problem was in package name i used alpine-fpm-7.3.8-r0 instead of alpine-fpm, ty for help
2020-12-01 18:26:13 <mimi89999> How to get clipboard working when running Alpine in QEMU?
2020-12-01 18:27:36 <Cogitri> Depends on the DE, I suppose
2020-12-01 18:27:39 <mps> I hope this is not posible
2020-12-01 18:28:31 <mps> mimi89999: if you mean sharing clipboard between host and guest
2020-12-01 18:28:52 <mimi89999> Sharing
2020-12-01 18:28:53 <mps> except over VNC/RDP or similar things
2020-12-01 18:29:07 <mimi89999> Works with Fedora as agiest
2020-12-01 18:29:26 <mimi89999> as a guest
2020-12-01 18:30:27 <mps> if login from one of them to another then it should work
2020-12-01 18:33:29 <Hello71> did you install vdagent
2020-12-01 18:33:49 <Hello71> better just use ssh though
2020-12-01 18:33:53 <Hello71> or serial
2020-12-01 18:36:40 <mimi89999> i just ssh in now
2020-12-01 19:18:27 <cornfeedhobo> hello
2020-12-01 19:30:32 <alpine_igor> Привет
2020-12-01 19:33:07 <mps> Привет (hello)
2020-12-01 19:33:28 <ikke> Is that privjet transliterated?
2020-12-01 19:33:36 <mps> ikke: yes
2020-12-01 19:37:28 <alpine_igor> Where get link on Alpine + rtl8812bu how-to
2020-12-01 19:41:13 <mps> alpine_igor: is this driver in mainline kernel
2020-12-01 21:22:52 <fourstepper> Hi, is it possible to run the alpine docker container with OpenRC as PID 1? Is this officially documented or do I need to DIY? Thanks
2020-12-01 21:38:23 <caskd> does anyone know how to tell ifupdown to create routes for a prefix instead of the address itself? i have to add the routes in post-up as a workaround but automatic config would be better
2020-12-01 21:40:56 <kmmndr> Hi :-)
2020-12-01 21:41:58 <kmmndr> Am I missing something ? lua5.2@edge-main does not include lualdap.so file (https://pkgs.alpinelinux.org/contents?branch=edge&name=lua5.2-ldap&arch=x86_64&repo=main)
2020-12-01 21:42:34 <Hello71> caskd: uh... isn't that the default?
2020-12-01 21:43:25 <Hello71> kmmndr: i think it is a bug related to it containing /usr/lib/lua/5.1
2020-12-01 21:43:58 <Hello71> https://gitlab.alpinelinux.org/alpine/aports/-/issues/12038
2020-12-01 21:44:48 <caskd> nope, it creates routes for gateway but not for prefix
2020-12-01 21:44:56 <caskd> at least on 3.12...
2020-12-01 21:46:11 <caskd> i have 'default via 10.0.0.1 dev eth1' and '10.0.0.1 dev eth1 scope link' but no '10.0.0.0/16 via 10.0.0.1 dev eth1'
2020-12-01 21:46:17 <Hello71> what do you mean "for prefix"
2020-12-01 21:47:08 <caskd> i would want to have the route with /16 be added instead of the one with 10.0.0.1
2020-12-01 21:48:42 <kmmndr> Hello71: Thank you very much for your answer
2020-12-01 21:49:11 <Hello71> caskd: what config are you using now
2020-12-01 21:49:32 <Hello71> https://build.alpinelinux.org/buildlogs/build-edge-x86_64/main/lua-ldap/lua-ldap-1.2.5-r0.log install src/lualdap.so /home/buildozer/aports/main/lua-ldap/pkg/lua5.3-ldap/usr/lib/lua/5.1
2020-12-01 21:49:54 <Hello71> i think that's actually a file
2020-12-01 21:50:49 <Hello71> yes, I remember. I think you cannot package (empty) directories. that command installs lualdap.so to be called "5.1"
2020-12-01 21:51:41 <caskd> Hello71: https://lainsafe.delegao.moe/files/160685948842699.interfaces
2020-12-01 21:52:12 <Hello71> seems fine to me
2020-12-01 21:52:38 <Hello71> assuming your network is actually broken and you need it to go via
2020-12-01 21:53:15 <caskd> uh, it's not broken... packets go to eth0 if i don't add that for some reason
2020-12-01 21:55:38 <caskd> probably the datacenter dhcp server is returning weird stuff (hetzner)
2020-12-01 21:59:09 <MY-R> caskd: https://docs.hetzner.com/cloud/networks/connect-dedi-vswitch/
2020-12-01 22:03:10 <MY-R> looks like it isnt workaround but the way of hetzner
2020-12-01 22:10:10 <caskd> probably the datacenter dhcp server is returning weird stuff (hetzner)
2020-12-01 22:10:16 <caskd> wups
2020-12-01 22:54:55 <amk> Hey, I have an mr with a stale tag, how do i remove this? Its just waiting on review, its not broken or anything
2020-12-01 22:59:13 <ikke> amk: which one?
2020-12-01 23:02:36 <amk> https://gitlab.alpinelinux.org/alpine/aports/-/merge_requests/13983
2020-12-01 23:03:07 <ikke> removed
2020-12-01 23:04:38 <amk> cheers!
2020-12-01 23:05:05 <ikke> amk: You mentioned the tests for sublime-music fail
2020-12-01 23:05:57 <ikke> Is that something worth looking into?
2020-12-01 23:19:35 <hg> pyinstaller's test suite is being rude
2020-12-01 23:19:48 <hg> I need to buckle down and dig into it, it's just really painful since it takes a billion hours to run
2020-12-02 02:24:09 <halosghost> ikke: I don't know if you did anything, but the packages are there now that I'd expect to be there, so no worries :)
2020-12-02 03:31:17 <dave84> Hello
2020-12-02 03:31:32 <dave84> Is anybody home?
2020-12-02 03:31:39 <halosghost> dave84: yes
2020-12-02 03:31:43 <dave84> Oh cool
2020-12-02 03:31:48 <dave84> I haven't used IRC in like 20 years
2020-12-02 03:31:50 <dave84> this is cool
2020-12-02 03:32:12 <halosghost> dave84: pro-tip: IRC is a protocol of latency. there are 431 people in this channel; even if no one answers right away, they will eventually :)
2020-12-02 03:32:24 <dave84> Oh very cool
2020-12-02 03:32:32 <dave84> Hopefully one of whom can help me with boot issues
2020-12-02 03:32:44 <tehcloud> what kind of boot issues you got there?
2020-12-02 03:34:32 <dave84> Making it halfway through an encrypted boot
2020-12-02 03:34:47 <dave84> Boots to emergency shell
2020-12-02 03:35:01 <dave84> I have to manually mount my disk
2020-12-02 03:35:10 <dave84> But I get "block device required"
2020-12-02 03:36:38 <dave84> Like I unlock in Grub, then I get dumped in emergency shell. I can luksOpen and get the drive open, and then I have to mdev to get dev/mapper to show up
2020-12-02 03:37:23 <dave84> Not sure what to try next or why it's not booting
2020-12-02 03:37:30 <dave84> I wish I could explain it better
2020-12-02 03:39:01 <Hello71> irc wasn't live 20 years ago either
2020-12-02 03:39:38 <Hello71> although it was probably more live than it is today with cheap virtual servers
2020-12-02 03:41:58 <helmen[m]> I think I found a bug, should I report it to the tracker on gitlab?
2020-12-02 03:41:58 <helmen[m]> Basically when I configure alpine with setup-alpine I can configure wifi, then I backup with lbu, reboot and alpine can't reconnect, but if I manually give the command "/etc/init.d/networking restart" alpine connects.
2020-12-02 03:42:40 <tehcloud> helmen[m], you sure you've got your boot sequence correct?
2020-12-02 03:45:00 <helmen[m]> yep, how it can work otherwise when i give the command ".../newworking restart" ?
2020-12-02 03:45:14 <helmen[m]> *networking
2020-12-02 03:45:56 <tehcloud> I mean, you can put it into boot or default in openrc
2020-12-02 03:46:08 <tehcloud> the order matters, I had this issue previously on my laptop
2020-12-02 03:46:32 <helmen[m]> yes i can but alpine automatically try to reconnect at boot, but it fail
2020-12-02 03:46:37 <tehcloud> and then there's also implications if you're using networkmanager and/or wpa_supplicant
2020-12-02 03:47:32 <helmen[m]> one moment, you made me think that maybe it's wpa_supplicant that I'm relaunching and not networking
2020-12-02 03:48:39 <helmen[m]> yes i remember now, because I found the command launched in the "setup-interfaces" script
2020-12-02 03:49:10 <helmen[m]> so the problem is the same but the command is... (one moment I search it again)
2020-12-02 03:51:09 <helmen[m]> wpa_supplicant -B -c /etc/wpa_supplicant/wpa_supplicant.conf -i wlan0
2020-12-02 03:51:11 <helmen[m]> this
2020-12-02 03:51:47 <helmen[m]> Sorry but I was going to memory because I did this test yesterday.
2020-12-02 04:23:19 <helmen[m]> created: https://gitlab.alpinelinux.org/alpine/alpine-baselayout/-/issues/3
2020-12-02 04:24:48 <helmen[m]> 
ACTION sent a long message:  < https://matrix.org/_matrix/media/r0/download/matrix.org/biNZeFpGpBQfDIFWDirklVom/message.txt >

2020-12-02 04:39:02 <helmen[m]> 6. where can I found a list of boot options that i can use in syslinux.cfg ?
2020-12-02 04:39:09 <helmen[m]> with description
2020-12-02 05:34:50 <blubshep> My resolver often fails to resolve urls, is this a problem with the musl resolver or just the ISP dns
2020-12-02 05:35:13 <ikke> billiob: impossible to tell without more details
2020-12-02 05:35:44 <blubshep> true, I should do testing for a bit, I wanted to know if it's a known issue
2020-12-02 05:37:19 <ikke> there is one known issue with docker on mack
2020-12-02 05:37:24 <ikke> mac
2020-12-02 05:38:41 <blubshep> Thanks, I'll report after further testing
2020-12-02 09:38:10 <Newbyte> trying to compile something and CMake complains about not finding threads despite that musl-dev is installed. any ideas?
2020-12-02 09:40:48 <Newbyte> never mind I found an ugly hack that works for my purposes
2020-12-02 13:13:37 <mps> tehcloud: just pushed kernel 5.9.12 with ethernet driver enabled for this new and shiny machine (ZEN iirc)
2020-12-02 13:14:38 <mps> please test it when it is built and report any problem or success, to know does it works
2020-12-02 13:16:17 <tehcloud> awesome... will do it now
2020-12-02 13:16:38 <tehcloud> oh, "when it is built"
2020-12-02 13:16:41 <tehcloud> durr... just woke up
2020-12-02 13:17:50 <mps> enjoy your morning coffee first :)
2020-12-02 14:27:10 <tehcloud> mps: it's working great
2020-12-02 14:27:18 <tehcloud> gracias
2020-12-02 14:30:12 <mps> tehcloud: good to hear, thanks for report
2020-12-02 14:31:11 <tehcloud> 5.10 will be tagged as linux-lts in edge eventually I assume?
2020-12-02 14:34:53 <mps> I think so, if it was released this month, and it will probably
2020-12-02 14:35:49 <mps> latest Linus comment with 5.10-rc6 says that he is quite satisfied with it
2020-12-02 14:36:08 <mps> and also I'm :)
2020-12-02 14:36:39 <mps> I'm running 5.10-rc1 and up on my two arm64 chromebooks
2020-12-02 14:38:23 <mps> but the pity is it will not be linux-lts in next alpine 3.13 release
2020-12-02 14:38:49 <tehcloud> that's why you have to live on the edge ;)
2020-12-02 14:39:04 <mps> :)
2020-12-02 14:41:41 <miskatonic> the stable release lacks ratpoison. This will force me to live on the edge when using x11.
2020-12-02 14:42:10 <ikke> Maybe you can ask the maintainer to move it to community
2020-12-02 14:42:19 <ikke> (and give them feedback that you use it and that it works for you)
2020-12-02 14:42:57 <mps> is it better than awesome?
2020-12-02 14:43:15 <miskatonic> it is more consequently mouseless
2020-12-02 14:43:43 <tehcloud> spectrwm is where it's at
2020-12-02 14:44:18 <tehcloud> glad we have so many choices for lightweight tiling window manages
2020-12-02 14:44:29 <mps> oh, I thought 'yak shaving' is finished today :)
2020-12-02 14:45:44 <ikke> Yak shaving is nevery finished
2020-12-02 14:50:14 <mps> yes, yaks breeds a lot in last years
2020-12-02 15:52:29 <telmich> Since yesterday I get a segmentation fault after loading a game in dosbox - is anyone aware of any recent changes?
2020-12-02 15:54:47 <ikke> musl has been updated to the latest snapshot
2020-12-02 15:54:51 <ikke> on edge
2020-12-02 15:57:18 <mps> dosbox upstream is lagging, I'm thinking to replace it dosbox-staging
2020-12-02 15:57:41 <mps> with*
2020-12-02 17:47:15 <telmich> Interesting - I think the bug is/was me: I started dosbox with `dosbox .` vs. just `dosbox` - the former makes it crash after some time, the latter works fine
2020-12-02 17:52:02 <sirlami> I am trying to write up instructions to boot alpine on a RockPro64 SBC and I am having trouble with the boot repostiory. I am using packages from both main an community and have tried to use apk fetch --recursive to fetch the packages and then apk index and abuild sign to make the APKINDEX.tar.gz but this results in some errors about package mentioned in index not found. I tried pulling the actual
2020-12-02 17:52:08 <sirlami> APKINDEX.tar.gz from the main repository but then it skips installing all my packages from community. What is the correct way to set this up?
2020-12-02 17:52:54 <mps> sirlami: did you use '--repositories-file' with apk fetch
2020-12-02 17:55:15 <sirlami> mps: I did not, let me give that a try
2020-12-02 18:13:46 <sirlami> mps: I am not sure how to use this flag. I am using apk fetch --recursive --output /boot/boot-repo/aarch64 $packages to fetch the apks and place them where they need to be, how would I use this flag in this situation?
2020-12-02 18:28:02 <mps> sirlami: you create file with repositories you want to use
2020-12-02 18:28:52 <mps> or you can use -X param multiple times
2020-12-02 19:54:35 <sirlami> mps: unfortunately I am still not following. to be clear I am attempting to create my own version of the apks folder found in the generic aarch64 image. How can I make a repo that has packages from both main and community so they are verified and installed at boot? apologies if I am misunderstanding
2020-12-02 19:57:24 <mps> sirlami: I'm not sure how this can be done. You will have unpack/repack all apks to sign with your keys and create INDEX files again signed with your keys. Not simple task in my experience
2020-12-02 19:57:56 <mps> and don't see why that is needed at all
2020-12-02 20:01:53 <sirlami> mps: I can certainly confirm that it is not simple! I generate my own keys, I then fetch all the apks I need, I index them using apk index and then sign the APKINDEX.tar.gz. This works, kinda. When I boot it installs them but some of them fail with an error about the package being mentioned but not found (even though they are there). Still, it boots and installs some packages
2020-12-02 20:02:22 <sirlami> Can you elaborate why this is not needed? Is there a better way to have packages install at boot?
2020-12-02 20:05:48 <mps> we have prepared tarballs and iso for installation
2020-12-02 20:06:08 <mps> rest can be downloaded from our mirrors
2020-12-02 20:07:07 <mps> other option is to use apk --root and/or --arch options
2020-12-02 20:10:53 <sirlami> so there is no way to customize which pacakges are installled at boot? this server will not be connecting to the internet until after packages are installed so a local repostiory is my only option
2020-12-02 20:12:23 <mps> you can 'fetch' all needed packages and put them in local mirror, or you can simply mirror some of our mirrors
2020-12-02 20:12:34 <sirlami> If I play with the generic arm image I can add apks from main and community to the /apks folder as well as list them in /etc/apk/world and have them installed
2020-12-02 20:12:54 <mps> you don't need then to sign them with your keys
2020-12-02 20:13:17 <sirlami> yes, that sounds like what I am trying to achieve
2020-12-02 20:14:24 <sirlami> if I place these a list of packages in /etc/apk/world and then place their apks in a local mirror will they be automatically installed after the apkovl is applied?
2020-12-02 20:15:58 <mps> I don't use this kind of 'install' so I don't have much experience what is needed for this to work, but I read that some people do that
2020-12-02 20:17:41 <sirlami> do you know where I might find more information? I have looked extensively online and still can't get it to work reliably
2020-12-02 20:18:17 <sirlami> perhaps I am not searching for the correct terminology
2020-12-02 20:18:55 <mps> here, I think, but you have to be patient and repeat question from time to time (not to often) till someone with experience have time to help you
2020-12-02 20:20:00 <mps> though I wonder why anyone install server in 'run from mode'
2020-12-02 20:20:14 <mps> run from RAM mod
2020-12-02 20:20:25 <mps> mode*
2020-12-02 20:20:38 <mps> huh, keyboards :)
2020-12-02 20:30:17 <sirlami> fair enough, I shall wait patiently :)
2020-12-02 20:45:51 <hechos> hi guys.
2020-12-02 20:46:30 <hechos> what can expect to using zfs on basico role of desktop side?
2020-12-02 20:46:54 <hechos> maybe is good idea stick with btrfs?
2020-12-02 21:56:11 <c705> what?
2020-12-03 00:09:29 <b00b> hi. what means community repo?
2020-12-03 00:10:50 <maxice8> one of the 3 repositories used by Alpine Linux, packages in it have a promise of 6 months of bugfixes
2020-12-03 00:15:23 <b00b> yook..i mean are they made by the community? and what happend after that 6 months?
2020-12-03 00:17:42 <maxice8> They are maintained same as other repositories, after 6 months they no longer receive bugfixes.
2020-12-03 00:18:07 <maxice8> 6 months is the time (planned) between each release of Alpine Linux
2020-12-03 00:18:35 <b00b> ok the name confused me a bit ^_^ it remembered me on ubuntu repo names
2020-12-03 00:19:08 <b00b> and the main repo?
2020-12-03 00:20:14 <maxice8> main repo has promised 6 months of bugfixes (like community) + 18 months of security fixes (fixes for CVEs)
2020-12-03 00:20:30 <b00b> i see thx.
2020-12-03 00:21:11 <b00b> but that means for exmaple if i use vifm as filemanager after that six months i need to upgrade alpine to a new version to get fixes?
2020-12-03 00:25:18 <Jacki> hi
2020-12-03 00:34:35 <maxice8> b00b: yes
2020-12-03 00:43:47 <Jacki> any chance somebody knows libnih here?
2020-12-03 00:46:22 <ikke> No, we did not make it :P
2020-12-03 00:49:21 <Jacki> sorry im desperate
2020-12-03 00:50:09 <Jacki> i saw somewhere it was an installe package
2020-12-03 00:50:12 <ikke> I don't know it, sorry
2020-12-03 00:50:33 <ikke> There is a package, yes
2020-12-03 00:50:47 <ikke> https://pkgs.alpinelinux.org/package/edge/community/x86_64/libnih
2020-12-03 01:02:43 <halosghost> does apk have a way to determine what upgrades would happen if you were to do a full upgrade?
2020-12-03 01:10:22 <Hello71> ... apk upgrade -s?
2020-12-03 01:13:27 <halosghost> Hello71: not on the system right now, and I can't seem to find the man pages online
2020-12-03 01:13:36 <halosghost> Hello71: is that a dry-run or something?
2020-12-03 01:15:26 <Hello71> https://gitlab.alpinelinux.org/alpine/apk-tools/-/blob/master/doc/apk-upgrade.8.scd
2020-12-03 01:15:53 <halosghost> Hello71: thanks!
2020-12-03 01:16:15 <halosghost> Hello71: so…
2020-12-03 01:16:36 <halosghost> Hello71: since -s isn't documented, is that a dry-run or something? :)
2020-12-03 01:16:55 <Hello71> "*apk upgrade* supports the commit options described in *apk*(8)"
2020-12-03 01:17:35 <halosghost> ah
2020-12-03 01:17:38 <halosghost> cool
2020-12-03 01:17:40 <halosghost> so, yeah, a dry-run :)
2020-12-03 01:17:41 <halosghost> Hello71: thank you
2020-12-03 01:17:49 <Hello71> :)
2020-12-03 01:20:46 <b00b> maxice8: thanks
2020-12-03 01:21:44 <b00b> is there a discussion about adding systemd to alpine?
2020-12-03 01:21:59 <b00b> at some point in the future
2020-12-03 01:22:18 <halosghost> b00b: … wat
2020-12-03 01:22:26 <halosghost> b00b: is that serious?
2020-12-03 01:22:42 <halosghost> like, I use arch, I don't mind systemd, but boy would it seem out-of-place on alpine
2020-12-03 01:23:02 <b00b> =)
2020-12-03 01:23:09 <b00b> im glad then
2020-12-03 01:23:50 <maxice8> Only in terms of getting some of the benefits from it, not using systemd itself
2020-12-03 01:23:51 <b00b> i dont like to test a distro when its going to kill itself in the future ;)
2020-12-03 01:24:07 <halosghost> maxice8: really? that sounds fascinating; what does it involve?
2020-12-03 01:25:25 <halosghost> there are many parts of systemd I really appreciate over openrc
2020-12-03 01:25:51 <b00b> maxice8: thats really an unbeatable point for using alpine, there are not many distros that resists this crap
2020-12-03 01:25:53 <maxice8> Also systemd treating anything but glibc as verboten is an insta-killer regardless
2020-12-03 01:26:12 <halosghost> b00b: can we not have a systemd flamewar?
2020-12-03 01:26:27 <b00b> halosghost: ask santa
2020-12-03 01:27:14 <b00b> i will give alpine a try as soon as i have time for this
2020-12-03 01:28:29 <halosghost> maxice8: is there a mailing list item or a bug on gitlab tracking that discussion? I'd love to read up
2020-12-03 01:34:28 <Hello71> maxice8: systemd is already crap enough, there's no need to spread FUD
2020-12-03 01:35:21 <maxice8> Hello71 ?
2020-12-03 01:35:43 <tehcloud> lol, the systemd topic again... I think I've piled enough hate on that project in my lifetime already though
2020-12-03 01:36:05 <Hello71> systemd doesn't hate musl, it just doesn't care to support it
2020-12-03 01:36:05 <tehcloud> loving OpenRC for now and looking forward to s6 in the future
2020-12-03 01:36:56 <maxice8> Hello71: and ?
2020-12-03 01:36:57 <Hello71> it has accepted some patches for musl support, and it doesn't work against downstreams such as openembedded that work to make it compatible
2020-12-03 01:37:12 <Hello71> but it also doesn't want to be full of unnecessary ifdefs
2020-12-03 01:37:26 <Hello71> 01:25 <@maxice8> Also systemd treating anything but glibc as verboten is an insta-killer regardless
2020-12-03 01:37:32 <Hello71> this is false
2020-12-03 01:37:59 <b00b> tehcloud: you can also have a look at minit: https://www.fefe.de/minit/
2020-12-03 01:39:41 <halosghost> Thermi: wait, is an s6 transition gonna happen?
2020-12-03 01:40:20 <halosghost> tehcloud: ^
2020-12-03 01:40:23 <halosghost> Thermi: sorry ):
2020-12-03 01:40:51 <tehcloud> I've heard rumblings but I don't actually know
2020-12-03 01:40:58 <maxice8> some patches that don't conflict with their vision that glibc is the only one supported, also it would be ridiculous for them to waste time working against people that want systemd on musl, at best they get portability fixes, at worst some people will bugger them without disclosing they are using an unsupported configuration
2020-12-03 01:41:10 <tehcloud> either way it's a nice project
2020-12-03 03:02:43 <Apachez> maxice8: then devuan is a nice effort - systemdless debian :)
2020-12-03 03:05:21 <Apachez> VSZ/RSS on init for debian (systemd): 102400 11836
2020-12-03 03:05:33 <Apachez> vs 8 8 for http://www.energymech.net/users/proton/ :D
2020-12-03 04:56:19 <Shufei> Howdy everyone!  I’m a new fan of Alpine on iPad thanks to iSH.  It’s running pretty nicely so far!
2020-12-03 04:56:53 <Shufei> But I’d like to put SBCL on.  It doesn’t seem to be in the apk repo.  Any pointers please?
2020-12-03 04:57:16 <Shufei> (Or any other Common Lisp doohickeys, really.)
2020-12-03 05:13:02 <Shufei> Ah!  I shall ask on listserv.
2020-12-03 07:56:06 <helmen[m]> I want to report another bug about lbu, but i really do not understand in what section of alpine in gitlab i should post it.
2020-12-03 07:56:06 <helmen[m]> This is a very old bug, more than one year at least
2020-12-03 08:02:48 <helmen[m]> this is funny, I had already report this problem one year ago... https://gitlab.alpinelinux.org/alpine/aports/-/issues/10124
2020-12-03 08:03:07 <helmen[m]> no reply...
2020-12-03 08:21:43 <helmen[m]> I correct myself, probably this is 6 years old bug: https://gitlab.alpinelinux.org/alpine/aports/-/issues/3492
2020-12-03 08:24:12 <ikke> helmen[m]: fyi: https://gitlab.alpinelinux.org/alpine/alpine-conf/-/issues is the bucktracker for LBU
2020-12-03 08:27:57 <helmen[m]> thanks ikke  , even if my old bug report on the same bug had been transferred to aports instead of alpine-conf.
2020-12-03 08:27:57 <helmen[m]> What do you think I have to do? Do I have to rewrite the bug report again?
2020-12-03 08:29:06 <ikke> Was it moved on redmine?
2020-12-03 08:29:15 <ikke> I don't see it on that issue
2020-12-03 08:29:46 <helmen[m]> yes (from redmine: issue id 10124, created on 2019-03-17)
2020-12-03 08:30:49 <helmen[m]> i can try to write the problem another time but i don't know if it is really a good idea. The bug is the same.
2020-12-03 08:30:58 <ikke> No, I don't expect that
2020-12-03 08:31:09 <helmen[m]> uhm?
2020-12-03 08:31:20 <ikke> I don't expect you to write a new issue
2020-12-03 08:36:00 <helmen[m]> but is it understandable that it is enough to to specify only one file in the /home/user (for example /home/user/.xinitrc) to "lbu iclude" without adding also the upper directory that when rebooting the user can't access the rest of the files because they become root properties?
2020-12-03 08:36:25 <helmen[m]> ikke: also login to xfce become impossible
2020-12-03 08:37:08 <ikke> helmen[m]: reading that old bug report you meantioned, I think the expectation is that the owner should be persisted / restored
2020-12-03 08:37:49 <helmen[m]> is it not normal to have this expectation?
2020-12-03 08:40:34 <ikke> I don't see why not
2020-12-03 08:40:40 <helmen[m]> With the tests I did I realized that lbu does not keep track at all also of the change on user / group. Simply if you change these two parameters to a file that lbu already keeps track of and then do "lbu status", nothing has changed for lbu.
2020-12-03 08:43:23 <helmen[m]> my opinion is that when lbu compresses files, it keeps the owner, but this does not happen for paths if they have not been specified one by one as well. I haven't done any tests regarding the permissions, but it could have the same problem for the permissions too
2020-12-03 08:44:17 <ikke> It would be good to include these details in the issue
2020-12-03 08:44:30 <ikke> show concrete examples
2020-12-03 08:45:08 <helmen[m]> my first language is not english, this is the reason because is not simple to understand what I had written
2020-12-03 08:45:31 <ikke> right
2020-12-03 08:45:44 <helmen[m]> I have already written what happens, but perhaps I was not clear enough.
2020-12-03 08:46:08 <helmen[m]> I try to add an example
2020-12-03 08:46:20 <ikke> What usually helps is to show the output of commands and what you expect
2020-12-03 08:46:41 <ikke> The more concrete an issue is, the easier it is for people to verify / see what is happening
2020-12-03 08:50:11 <helmen[m]> I understand but sometimes is not very simple to be clear, I try another time
2020-12-03 08:59:15 <helmen[m]> 
ACTION sent a long message:  < https://matrix.org/_matrix/media/r0/download/matrix.org/zIMmHPEXMlfJOGyqiLVVZPJR/message.txt >

2020-12-03 10:56:08 <ForeverNoob[m]> hello, how do I find out which packages are present in Alpine base system? I know there's busybox, what else? I searched on the wiki as well as https://pkgs.alpinelinux.org/packages but that didn't really clear things up for me.
2020-12-03 10:56:34 <ikke> ForeverNoob[m]: https://pkgs.alpinelinux.org/package/edge/main/x86_64/alpine-base
2020-12-03 10:56:38 <ikke> check 'depends'
2020-12-03 10:58:02 <ForeverNoob[m]> thanks, that looks very minimal indeed!
2020-12-03 10:58:40 <ikke> note that alpine-setup will install additional packages (like ntp client, ssh server, etc)
2020-12-03 11:01:49 <ForeverNoob[m]> ikke: you mean setup-alpine? https://wiki.alpinelinux.org/wiki/Alpine_setup_scripts#setup-alpine
2020-12-03 11:02:03 <ikke> yes
2020-12-03 11:03:26 <ForeverNoob[m]> cool, thanks.
2020-12-03 14:37:02 <alpine_igor_ru> hi
2020-12-03 14:39:27 <alpine_igor_ru> I can't see the wifi card rtl8821cu. which module need to be connected via modpropbe?
2020-12-03 14:40:55 <mps> alpine_igor_ru: I'm not sure it is available in current -lts kernel
2020-12-03 14:41:28 <tehcloud> alpine_igor_ru, try linux-edge
2020-12-03 14:41:29 <alpine_igor_ru> another core?
2020-12-03 14:42:32 <alpine_igor_ru> I'll try
2020-12-03 14:42:45 <mps> no, linux-edge is in testing repo, you have to enable it in /etc/apk/repositories
2020-12-03 14:43:26 <alpine_igor_ru> thanks
2020-12-03 14:44:20 <mps> hmm, looks like it is not
2020-12-03 14:46:04 <mps> alpine_igor_ru: I can't find any reference in alpine kernels for this drivers
2020-12-03 14:47:23 <alpine_igor_ru> very sorry. there were hopes that I found a suitable Linux for my router.
2020-12-03 14:49:06 <mps> you can build driver locally
2020-12-03 14:50:29 <alpine_igor_ru> for debian i used dkms
2020-12-03 14:50:52 <ikke> We unfortunately do not have dkms (yet)
2020-12-03 14:51:35 <ikke> alpine_igor_ru: here is a package of a similar driver: https://gitlab.alpinelinux.org/alpine/aports/-/blob/master/community/rtl8821ce-lts/APKBUILD
2020-12-03 14:51:51 <mps> it is not complicated to install needed tools and build drivers
2020-12-03 14:52:51 <alpine_igor_ru> Thanks, I'll take a look
2020-12-03 14:54:09 <ikke> https://github.com/brektrou/rtl8821CU
2020-12-03 14:54:19 <ikke> (verify the source first, though)
2020-12-03 17:17:52 <dami0> hi. i installed alpine on a raspberry pi, but i'm having trouble finding python's pip in the packages. is it possible to install pip on alpine-aarch64?
2020-12-03 17:29:44 <Hello71> https://pkgs.alpinelinux.org/packages?name=*pip&branch=edge
2020-12-03 17:47:11 <sirlami> I am trying to replicate the apks folder found in both the raspberry pi and generic aarch64 images however I can't get this to work. I fetch all packages with apk fetch, index them with apk index and sign the APKINDEX.tar.gz with abuild-sign. While it doesn't throw an error about them being untrusted it will fail to install some packages with an error about the package being in the index but not
2020-12-03 17:47:17 <sirlami> found.
2020-12-03 17:48:50 <sirlami> Installing just alpine-base in this manner has a 4/21 failure rate, for example: ERROR: ifupdown-ng-openrc-0.10.1-r0: package mentioned in index not found (try 'apk update')
2020-12-03 17:49:46 <sirlami> What is the correct way to make an offline repository that will supply the neessary packages to satisfy /etc/apk/world at boot on a RAM installtion?
2020-12-03 17:50:53 <sirlami> Not signing the APKINDEX at all obvious completely fails due to the repository not being trusted
2020-12-03 17:58:21 <mps> sirlami: 'apk --allow-untrusted' maybe?
2020-12-03 18:03:04 <sirlami> mps: how can I do that at boot? I have an apk overlay that I can easily modify to add these settings
2020-12-03 18:04:21 <mps> ah, I thought you are trying from CLI
2020-12-03 18:04:35 <sirlami> also, why do only 4 packages fail to install at boot? I would think it was all or nothing
2020-12-03 18:06:03 <mps> I think you should have these files somewhere on overlay file, but don't know where
2020-12-03 18:07:35 <sirlami> but this is not how it is setup in the rpi/aarch64 images. There all apks are in the apks folder and /etc/apk/world dictates what will be installed
2020-12-03 18:08:09 <sirlami> I could install them all and just make an apkovl of the entire system but that isn't as clean and not how it is done on the offical images
2020-12-03 18:09:48 <mps> as I told earlier I don't much experience with 'diskless install' so can't help, sorry
2020-12-03 18:10:57 <mps> I have only one advice, install in sys mode and you will have a lot less 'headaches'
2020-12-03 18:14:31 <sirlami> that may be true but having it run in ram has some desireable benefits, it also happens to be the way Alpine is actually designed
2020-12-03 18:15:05 <sirlami> I have everything else working, only the boot repository is giving me troubles
2020-12-03 18:16:41 <mps> sirlami: alpine was designed to run from ram as small router system, but these times passed. nowadays it is mostly used on servers and workstations
2020-12-03 18:17:11 <mps> besides being dominant on docker and other containers
2020-12-03 18:20:20 <sirlami> I understand that however the fact that this is how the offical arm based images are still packaged suggests it is still the preferered way for embedded systems. it also suggests that it is quite possible, I just cannot replicate it completely
2020-12-03 18:23:53 <mps> sirlami: that is possible ofc, but by using alpine standard tools for installation, setup-alpine is enough
2020-12-03 18:33:27 <dami0> thanks Hello71
2020-12-03 18:35:32 <sirlami> I am not asking for help with some new, strange or custom way of booting Alpine, this is literally what the website claims is 'the way it is meant to be' so I do not understand your insistence that I stop going down this path. I am not asking anything that isn't officailly supported here.
2020-12-03 18:36:21 <c705> mps> sirlami: alpine was designed to run from ram as small router system, but these times passed.
2020-12-03 18:36:24 <c705> since when?
2020-12-03 18:36:59 <c705> sirlami: https://wiki.alpinelinux.org/wiki/Local_APK_cache
2020-12-03 18:37:04 <c705> hopefully that will help
2020-12-03 18:40:23 <mps> c705: since when I switched to alpine :P
2020-12-03 18:54:02 <MY-R> sirlami: you shouldnt touch anything in "/media/mmcblk0p1/apks/aarch64/"
2020-12-03 18:59:25 <MY-R> all magic happening in "/media/mmcblk0p1/cache/"
2020-12-03 19:03:19 <sirlami> c705: thank you, I have been working from this guide but perhaps I am not doing it correctly. I shall do more work on this
2020-12-03 19:08:41 <fourstepper> Hi, is it possible to use nm-applet in Alpine? I am getting a unable to get session bus error (logically)
2020-12-03 19:08:54 <sirlami> MY-R: I am trying to build this 'apks' folder from scratch as opposed to using that in the official images. I can place apks in there and use the APKINDEX.tar.gz from the official main repository, this works fine but obviously you can only use apks from main in this setup. Would the apk cache method work to install from any officail repository? I am using some packages from community that I need
2020-12-03 19:09:00 <sirlami> installed at boot too.
2020-12-03 19:09:14 <fourstepper> I have elogind installed, is there a way to emulate dbus session without systemd and get past this?
2020-12-03 19:12:18 <MY-R> sirlami: I dunno :\ I'm trying keep alpine as "vanilla" possible so can easly upgrade to next release without "fun" with rebuilding images and together with LBU backups it is smooth and more predictable
2020-12-03 19:13:02 <MY-R> sirlami: so ye you need dig in to mkimage.sh and other connected scripts to figure this out
2020-12-03 20:07:49 <Youbi> Hello, I’m trying to install a package on edge, but I get a conflict when trying to do so: http://lxp.fr/p46b05pdxp
2020-12-03 20:08:27 <Youbi> I did not upgrade this installation for a while, so I tried an apk upgrade, but it does not fix the conflict. I’m kind of noob on packages management and conflict resolution :/
2020-12-03 20:14:37 <maldridge> looks like I fell out of the channel some time ago
2020-12-03 20:15:58 <ikke> maldridge: welcome back
2020-12-03 20:16:47 <maldridge> thanks
2020-12-03 20:17:21 <maldridge> I come with a strange bug as usual.  I have an alpine machine that seems to refuse to mount cgroups.  Is there some logic somewhere that controls the mounting of special filesystems?
2020-12-03 22:13:28 <helmen[m]> ikke: I tried to find the origin of the problem on my own and I think I found it, I am not able to fix the problem on my own but I am afraid that no one will notice my report, after all it has been almost two years that my issue is open.
2020-12-03 22:13:28 <helmen[m]> This is my last message https://gitlab.alpinelinux.org/alpine/aports/-/issues/10124
2020-12-03 22:13:53 <ikke> yes, I've noticed you updated it
2020-12-03 22:14:37 <mps> #10124
2020-12-03 22:17:37 <helmen[m]> oh, what do you think, that this can be considered a problem or not? in my case the only solution is to backup everything in home and forgetting to selecting only the things i want to really backuo or work at countrary and exclude file i do not want to backup (but it is a bit ugly, because in this way I suppose that i want to backup everything)
2020-12-03 22:18:53 <helmen[m]> *countrary == opposite, excuse me for my english :P
2020-12-04 01:15:07 <computerfido> hello, I just have a question about the ninja package, I noticed that the alpine repo is using a version from 2015, is there any reason for this?
2020-12-04 01:15:30 <halosghost> is it in the community repo?
2020-12-04 01:15:45 <halosghost> if so (and true), then it's most likely because someone hasn't MR'd an update :)
2020-12-04 01:16:57 <halosghost> computerfido: actually, what package? I don't see it
2020-12-04 01:16:59 <computerfido> it is in the main repo
2020-12-04 01:17:08 <halosghost> …
2020-12-04 01:17:24 <halosghost> https://pkgs.alpinelinux.org/packages?name=*ninja*&branch=edge
2020-12-04 01:17:59 <halosghost> what version of alpine are you on?
2020-12-04 01:18:02 <tehcloud> samurai > ninja
2020-12-04 01:18:29 <halosghost> looks like 3.12 removed it from the repos
2020-12-04 01:18:53 <computerfido> actually, I was looking at the wrong version, it is 1.9 (feb 2019) however that is still an older version, I am on 3.12
2020-12-04 01:19:02 <tehcloud> ninjas have to use stealth, but samurais can attack head on and have cooler weapons, therefore the samurai software package is also superior
2020-12-04 01:19:08 <halosghost> indeed, it has been moved to unmaintained
2020-12-04 01:19:20 <computerfido> ahh ok thanks
2020-12-04 01:19:37 <halosghost> sure :)
2020-12-04 01:19:47 <halosghost> if you want to maintain it, you can MR it back to testing or community
2020-12-04 02:24:24 <cornfeedhobo> can someone clarify if one should persist the key from abuild-keygen, or is it a per-package thing without much other significance?
2020-12-04 03:23:39 <maldridge> you should persist the key
2020-12-04 05:33:33 <ikke> cornfeedhobo: the apkindex that abuild
2020-12-04 05:33:46 <ikke> that abuild generates is signed with that key
2020-12-04 05:34:06 <ikke> as well as the packages
2020-12-04 05:47:32 <helmen[m]> Off-topic , Is Better to install firefox or firefox-esr especially for addons compatibility?
2020-12-04 05:48:54 <ikke> what kind of addons?
2020-12-04 07:24:03 <rnkn> what might it mean if abuild-fetch fails with /var/cache/distfiles/TARBALL.lock: permisssion denied?
2020-12-04 07:24:19 <rnkn> I'm running as a user in the abuild group
2020-12-04 07:26:44 <rnkn> (btw, APKBUILD(5) references abuild(1) but this doesn't exist)
2020-12-04 07:52:35 <rnkn> okay so I needed to chown :abuild /var/cache/distfiles; chmod g+w /var/cache/distfiles
2020-12-04 07:53:19 <rnkn> but according to previous IRC logs, this is not supposed to be necessary as apk add abuild reportedly sets permissions
2020-12-04 09:15:57 <Youbi> Hello, I’m trying to install a package on edge, but I get a conflict when trying to do so: http://lxp.fr/p46b05pdxp
2020-12-04 09:15:59 <Youbi> I did not upgrade this installation for a while, so I tried an apk upgrade, but it does not fix the conflict. I’m kind of noob on packages management and conflict resolution :/
2020-12-04 09:19:41 <ikke> apache2-proxy is holding the musl upgrade back for some reason
2020-12-04 09:19:41 <rnkn> update on abuild issue... apk del abuild; rm -rf /var/cache/distfile; apk add abuild solved it
2020-12-04 09:20:02 <rnkn> (but apk fix abuild did not..)
2020-12-04 09:38:29 <Youbi> rnkn: tried your solution, did not fix my issue. But I tried `apk fix musl`, and got that
2020-12-04 09:38:31 <Youbi> (1/1) [APK unavailable, skipped] Reinstalling musl (1.1.24-r0)
2020-12-04 09:39:05 <Youbi> Has musl package been renamed ?
2020-12-04 09:39:26 <ikke> no, but on edge it's updated to 1.2
2020-12-04 09:43:32 <Youbi> Ok. So I should update to musl 1.2, although it would break some packages right?
2020-12-04 09:44:49 <ikke> Youbi: it _should_ not
2020-12-04 09:45:32 <ikke> maybe apache2-proxy has to be rebuilt
2020-12-04 09:46:22 <Youbi> Ok. So an apk upgrade should update musl to 1.2, right ?
2020-12-04 09:46:29 <ikke> yes
2020-12-04 09:46:42 <ikke> If you remove apache2-proxy, I think it should work
2020-12-04 09:47:13 <Youbi> apache2-proxy is not installed, I got the issue when trying to install it ^^
2020-12-04 09:47:19 <ikke> aha
2020-12-04 09:47:21 <Youbi> I think I have a bigger problem x)
2020-12-04 09:48:07 <ikke> aha, I misread it
2020-12-04 09:48:14 <ikke> it's apache2-proxy that requires the newer musl version
2020-12-04 09:48:27 <Youbi> http://lxp.fr/p46c23q6Hs
2020-12-04 09:48:31 <Youbi> Oh ok!
2020-12-04 09:48:57 <Youbi> Maybe I should stop using edge to avoid running into that kind of issues ^^
2020-12-04 09:49:35 <Youbi> I’ll search gitlab issues for apache2-proxy.
2020-12-04 09:50:00 <ikke> no, apache2-proxy seems to be alright
2020-12-04 09:50:28 <ikke> does apk version -l '?' return something
2020-12-04 09:50:53 <Youbi> Wow. Is that black magic ?
2020-12-04 09:51:15 <ikke> it lists packages that are no longer present in the repository
2020-12-04 09:51:19 <Youbi> Oh, ok
2020-12-04 09:51:35 <Youbi> So I should remove them to clean my OS ?
2020-12-04 09:51:50 <ikke> yes, that would most likely solve this issue
2020-12-04 09:51:53 <Youbi> http://lxp.fr/p46d10N4Xa
2020-12-04 09:52:19 <ikke> mdocml has become mandoc
2020-12-04 09:52:24 <ikke> and man is no longer there
2020-12-04 09:56:05 <Youbi> There is no man command anymore? Or is it provided by another package?
2020-12-04 09:56:45 <ikke> mandoc provides it
2020-12-04 09:57:11 <Youbi> Ok, nice :)
2020-12-04 09:57:30 <Youbi> Cleaned all obsolete packages, now I can upgrade to musl 1.2 \o/
2020-12-04 09:57:43 <Youbi> (and a lot of other packages)
2020-12-04 10:05:00 <ikke> nice
2020-12-04 10:11:38 <Youbi> Thank you ikke!
2020-12-04 10:54:35 <rnkn> Youbi: my issue with abuild was unrelated to yours
2020-12-04 10:56:33 <enyc> 
ACTION meows

2020-12-04 10:57:07 <ikke> 
ACTION throws a ball of yarn

2020-12-04 10:57:47 <Youbi> Oh, sorry rnkn
2020-12-04 10:58:33 <enyc> 
ACTION tries to figure out how to use yarn as part of a heath-robinson bootloader-system

2020-12-04 10:59:14 <ikke> lol
2020-12-04 10:59:31 <ikke> somehow using yarn and bootloader in one sentence gives me shivers
2020-12-04 10:59:59 <enyc> 
ACTION imagines a serial-port controlled stepper-motor using yarn as a band to send some sort of pulse code on a loop of yarn to a modified grub that accepts kernel via serial port reception of the pulses

2020-12-04 11:00:26 <enyc> 
ACTION thinks... yarn distance-controlled-booting

2020-12-04 11:01:27 <enyc> hrrrm Alpine APK
2020-12-04 11:02:02 <enyc> presumably differeent to android apk's
2020-12-04 11:02:15 <enyc> presumably more capable than openwrt-style opkg
2020-12-04 11:02:33 <enyc> 
ACTION wonders how Alpine apk compare to Slackware packages

2020-12-04 11:06:53 <rnkn> what's the closest thing to dwm on wayland?
2020-12-04 11:08:17 <fourstepper> @rnkn dwl
2020-12-04 11:08:34 <fourstepper> https://github.com/djpohly/dwl
2020-12-04 11:09:54 <rnkn> fourstepper: I had a look at dwl but it seems very early stages
2020-12-04 11:11:55 <fourstepper> I don't have any personal experience with it, so can't say
2020-12-04 11:13:38 <rnkn> what I like so much about dwm is that it seems done -- nothing to update, nothing gonna change
2020-12-04 11:17:25 <rnkn> hikari looks interesting
2020-12-04 13:54:20 <jn__> enyc: not sure if yarn (the stringy material) or yarn (the javascript package manager) would be more cursed in the context of a bootloader
2020-12-04 14:56:37 <enyc> jn__: hrrrrrrrrrrrm -- I don't know!
2020-12-04 15:51:21 <cornfeedhobo> maldridge: thanks
2020-12-04 16:53:16 <radiomark> Hi, I'm looking for an off-the-shelf tool to supervise daemons. Ie. run a foreground process in the background, transfer its stdout/stderr to syslog, restart if it crashes etc..  Already have my own but it would be ideal if there's something included in Alpine
2020-12-04 16:54:55 <radiomark> (I mean packaged in Alpine)
2020-12-04 16:55:32 <cornfeedhobo> radiomark: write an openrc script
2020-12-04 16:56:00 <cornfeedhobo> radiomark: rather, alpine seems to use openrc*
2020-12-04 16:56:18 <radiomark> Does openrc handle transferring output to syslog etc.?
2020-12-04 16:57:14 <radiomark> Reading about something called supervise-daemon
2020-12-04 16:58:32 <artok> containers could use tini
2020-12-04 16:58:48 <radiomark> It's not for containers
2020-12-04 17:05:07 <radiomark> Unfortunately openrc supervise-daemon doesn't capture output as syslog, but maybe I will follow this openrc direction
2020-12-04 17:13:27 <mps> radiomark: runit
2020-12-04 17:14:03 <radiomark> I was just taking a lookat runit, which I have used on void linux. It's a completely independent init process though?
2020-12-04 17:14:43 <mps> no on alpine but can be run on init or later 'by hand'
2020-12-04 17:15:03 <mps> I have it run default runlevel
2020-12-04 17:15:25 <radiomark> So you have openrc -> runit? And then some of your own daemons in runit?
2020-12-04 17:15:43 <radiomark> FWIW the openrc method was looking good, just supervise-daemon doesn't have a few bits of functionality
2020-12-04 17:15:54 <radiomark> runsv seems to have very different conventions and layout
2020-12-04 17:16:16 <mps> yes to all
2020-12-04 17:16:52 <mps> but we are not talking about theory but on practical usage, I think
2020-12-04 17:21:09 <radiomark> Yeah, gonna take a me a little time to understand what's going on
2020-12-04 17:22:31 <radiomark> Because the other contender here is just port my small C program over and use that in place of supervise-daemon
2020-12-04 17:22:58 <radiomark> An extra init system on top of the system one may be a bit much
2020-12-04 17:47:59 <radiomark> runit works, it's easy to make a deaemon. But looks like it reimplements its own logging and log rotation. I'm trying to see if I'm missing something and if it can just log to syslog
2020-12-04 17:50:15 <mps> use logger in log/run script
2020-12-04 17:51:10 <radiomark> I was hoping to swap out custom code for off-the-shelf. But now I'd have to use runit + custom code to bridge it to syslog
2020-12-04 17:52:11 <radiomark> For 10 years or so we just used a custom C program to supervise a daemon, then wrap that in a CentOS init script
2020-12-04 17:59:04 <radiomark> mps, artok, cornfeedhobo: Thanks for the suggestions. I think this needs more work
2020-12-04 18:15:43 <cornfeedhobo> radiomark: no problem. reading your scrollback, yes, openrc can send output where ever you want. because it's a script, there is a lot of flexibility.
2020-12-04 18:17:25 <cornfeedhobo> radiomark: i completely agree to ignore people suggesting an init system on top of an init system. I've been at two companies that do this and i hate it. the small gain of being able to easily find company specific services is outweight by all the bullshit glue needed to get things going right. easier to lean on syslog like you are thinking. it's a true workhorse.
2020-12-04 18:17:47 <cornfeedhobo> (excuse my language)
2020-12-04 18:45:32 <radiomark> cornfeedhobo: thanks
2020-12-04 18:46:30 <radiomark> cornfeedhobo: I think openrc looks good too. I've found it one of the good thins about Alpine
2020-12-04 18:47:05 <radiomark> It's clear I can drop my exisitng C code cleanly in to do the supervision if that is needed
2020-12-04 18:47:33 <cornfeedhobo> exactly
2020-12-04 22:52:01 <cornfeedhobo> alpine-setup doesn't support ipv6?
2020-12-04 22:52:05 <cornfeedhobo> sad panda
2020-12-04 22:52:33 <cornfeedhobo> the script seems to only accept ipv4
2020-12-04 23:10:36 <Hello71> RA is the standard for ipv6 autoconfig
2020-12-04 23:11:20 <Hello71> if you want to use udhcpc6 or something you can bypass setup-interfaces
2020-12-05 00:14:53 <hechos> hi guys, need some help.
2020-12-05 00:15:14 <hechos> like always. but still cannot get working android-tools
2020-12-05 00:16:32 <caskd> hechos: did you add udev rules or started the server as root?
2020-12-05 00:17:36 <hechos> caskd: try hundread of things, like wiki from archwiki. adding vendor but cannot start server to comunicate to phone.
2020-12-05 00:17:46 <hechos> caskd: you make working please tell me how.
2020-12-05 00:18:20 <caskd> so you cannot start the server at all? do you get a error or anything back?
2020-12-05 00:23:24 <hechos> yeah.
2020-12-05 00:23:32 <hechos> let me recreate
2020-12-05 00:23:37 <hechos> what happen.
2020-12-05 00:31:39 <hechos> caskd: no show me how to add to rules
2020-12-05 00:32:26 <caskd> does it show anything?
2020-12-05 00:32:44 <caskd> i want to know if this is related to permissions or if this is something else
2020-12-05 00:32:53 <caskd> so i can guide you
2020-12-05 00:34:01 <hechos> * failed to start daemon
2020-12-05 00:34:03 <hechos> error: cannot connect to daemon
2020-12-05 00:34:09 <hechos> is from adb
2020-12-05 00:34:20 <hechos>  could not install *smartsocket* listener: Address not available
2020-12-05 00:35:30 <caskd> are you trying to use adb over tcp?
2020-12-05 00:35:44 <caskd> or only over usb?
2020-12-05 00:35:46 <hechos> cannot connect to daemon at tcp:5037: Address not available
2020-12-05 00:36:05 <hechos> i dont know. is default config.
2020-12-05 00:38:33 <caskd> reconnected because i was testing something...
2020-12-05 00:39:05 <caskd> do you have a loopback interface set up properly? adb needs one to connect to the daemon
2020-12-05 00:40:00 <hechos> just due apk add android-tools
2020-12-05 00:40:04 <hechos> what need more
2020-12-05 00:40:29 <caskd> do you see anything when running 'ip a s dev lo'?
2020-12-05 00:40:34 <hechos> in archlinux has package like android-udev
2020-12-05 00:41:29 <hechos> yes but is DOWN qlne 1000
2020-12-05 00:41:49 <caskd> well, that should be up so that adb can use it
2020-12-05 00:42:15 <caskd> try running 'ifup lo' as root
2020-12-05 00:43:13 <hechos> is not in etc/network/interface.
2020-12-05 00:44:07 <caskd> then you should add it like mentioned here https://wiki.alpinelinux.org/wiki/Configure_Networking#Loopback_Configuration_.28Required.29
2020-12-05 00:44:18 <hechos> okey.
2020-12-05 00:44:21 <hechos> let me see.
2020-12-05 00:45:35 <hechos> okey igot up.
2020-12-05 00:45:57 <caskd> try running 'adb start-server' once again
2020-12-05 00:46:35 <hechos> * daemon not running; starting now at tcp:5037
2020-12-05 00:46:37 <hechos> ADB server didn't ACK
2020-12-05 00:46:39 <hechos> Full server startup log: /tmp/adb.0.log
2020-12-05 00:46:41 <hechos> Server had pid: 13117
2020-12-05 00:46:43 <hechos> --- adb starting (pid 13117) ---
2020-12-05 00:46:45 <hechos> adb I 12-04 21:46:28 13117 13117 main.cpp:60] Android Debug Bridge version 1.0.41
2020-12-05 00:46:47 <hechos> adb I 12-04 21:46:28 13117 13117 main.cpp:60] Version 29.0.6-android-tools
2020-12-05 00:46:49 <hechos> adb I 12-04 21:46:28 13117 13117 main.cpp:60] Installed as /usr/bin/adb
2020-12-05 00:46:51 <hechos> adb I 12-04 21:46:28 13117 13117 main.cpp:60]
2020-12-05 00:46:53 <hechos> adb F 12-04 21:46:28 13117 13117 main.cpp:140] could not install *smartsocket* listener: Address not available
2020-12-05 00:46:56 <hechos> * failed to start daemon
2020-12-05 00:46:57 <hechos> error: cannot connect to daemon
2020-12-05 00:47:30 <mps> hechos: use paste service, don't spam channel
2020-12-05 00:48:01 <hechos> mps: what is services?
2020-12-05 00:48:44 <mps> huh, pastebin
2020-12-05 00:49:00 <hechos> ohh but work for terminal?
2020-12-05 00:49:05 <mps> ix.io or when you on alpine tpaste.us
2020-12-05 00:49:14 <hechos> thnaks.
2020-12-05 00:49:17 <hechos> for the tip.
2020-12-05 00:49:25 <hechos> and sorry
2020-12-05 00:49:26 <mps> apk add tpaste
2020-12-05 00:49:34 <caskd> some allow you to paste directly via pipes and curl
2020-12-05 00:49:54 <caskd> or something like tpaste, yeah
2020-12-05 00:49:59 <mps> and then you can 'run some cmd | tpaste' and post url here
2020-12-05 00:50:18 <hechos> sounds cool,.
2020-12-05 00:50:38 <hechos> still DOWN loopback.
2020-12-05 00:50:57 <hechos> maybe need to restart network service, have installed dhcpcd
2020-12-05 00:52:10 <mps> btw, adb start-server works fine here
2020-12-05 00:52:55 <hechos> cannot get work.
2020-12-05 00:53:14 <hechos> same errors add to etc/network/interfaces/
2020-12-05 00:53:41 <caskd> mps: yeah, first thing that jumped to my mind was client-daemon connection not working so i assumed it was loopback problems
2020-12-05 00:54:21 <caskd> mps: but some devices need udev rules or the daemon to run as root to load modules required
2020-12-05 00:54:42 <mps> adduser user plugdev
2020-12-05 00:55:15 <hechos> unkown group...
2020-12-05 00:55:51 <mps> hmm, so it is artifact from past on my box
2020-12-05 00:56:01 <hechos> on void
2020-12-05 00:56:14 <hechos> LOOPBACK is UP.
2020-12-05 00:56:22 <hechos> for some reason in alpine fot DOWN.
2020-12-05 00:56:27 <mps> you are trying on void linux?
2020-12-05 00:57:20 <hechos> nono.
2020-12-05 00:57:23 <hechos> ond void works.
2020-12-05 00:57:31 <hechos> android tools
2020-12-05 00:57:37 <hechos> cannot get work on alpine.
2020-12-05 00:57:54 <mps> ah
2020-12-05 00:58:08 <hechos> just looking configs. but void as runit.
2020-12-05 00:58:33 <hechos> need to update my phone and cannot due from alpine.
2020-12-05 00:58:59 <mps> ip a | tpaste
2020-12-05 01:01:31 <mps> are you in netdev group?
2020-12-05 01:02:04 <hechos> mmm/.
2020-12-05 01:03:52 <hechos> https://termbin.com/69yr
2020-12-05 01:09:07 <hechos> root account is disable all due over doas.
2020-12-05 01:09:25 <hechos> really dont know is a mystery.
2020-12-05 01:10:31 <mps> network is down
2020-12-05 01:10:38 <hechos> need to add to loop back 127.0.0.1
2020-12-05 01:10:41 <hechos> right?
2020-12-05 01:11:03 <mps> fight or right?
2020-12-05 01:11:12 <mps> sorry
2020-12-05 01:11:24 <mps> my eyes are tired
2020-12-05 01:11:32 <hechos> https://termbin.com/kbdie
2020-12-05 01:12:14 <mps> yes, this looks ok
2020-12-05 01:15:24 <hechos> mps can showme your configuratio to termbin?
2020-12-05 01:18:03 <mps> which one?
2020-12-05 01:18:42 <hechos> udev and interfaces  need to get work this.
2020-12-05 01:19:01 <mps> I don't have udev rules for it
2020-12-05 01:19:28 <mps> and interface config for loopback is same as yours
2020-12-05 01:22:02 <hechos> just got loopback up and work adb?
2020-12-05 01:22:20 <hechos> in void no need extra config. to run adb.
2020-12-05 01:22:27 <hechos> got only problems on alpine.
2020-12-05 01:22:48 <hechos> may it be im doing something wrong.
2020-12-05 01:23:31 <mps> loopback interface should be always up. why is your down
2020-12-05 01:24:38 <hechos> dont know.
2020-12-05 01:24:50 <hechos> just add dhcpcd at boot.
2020-12-05 01:24:57 <hechos> nothing fancy.
2020-12-05 01:25:10 <hechos> leet me restart machine and see.
2020-12-05 01:27:00 <hechos> mps: still donw.
2020-12-05 01:27:03 <hechos> down
2020-12-05 01:29:06 <mps> hmm, your system be seriously broken if lo is not up after boot
2020-12-05 01:30:25 <mps> what you get with 'rc-update | grep net'
2020-12-05 01:31:13 <hechos> no show nothing
2020-12-05 01:31:36 <hechos> i toll you use dhcpcd
2020-12-05 01:32:19 <mps> whatever you use for real net devices lo should be up
2020-12-05 01:32:36 <hechos> https://termbin.com/ghqp
2020-12-05 01:32:48 <mps> are you running in diskless mode, by chance?
2020-12-05 01:32:49 <hechos> eth0 to a router.
2020-12-05 01:32:54 <hechos> nooo no.
2020-12-05 01:33:01 <hechos> just my installed to chroot .
2020-12-05 01:33:49 <mps> you need 'rc-update add networking boot'
2020-12-05 01:34:32 <mps> or as root '/etc/init.d/networking start'
2020-12-05 01:35:40 <hechos> ttps://termbin.com/ca4e8
2020-12-05 01:35:49 <hechos> got ERROR
2020-12-05 01:36:37 <hg> you dropped this 'h'
2020-12-05 01:41:20 <hechos> dont know.
2020-12-05 01:41:36 <hechos> run setup-interfaces
2020-12-05 01:41:40 <hechos> and got errors
2020-12-05 01:48:00 <hechos> for what reason lo show DOWN?
2020-12-05 02:57:19 <cornfeedhobo> why is it so difficult to find where the docker daemon is storing things? sigh
2020-12-05 03:03:15 <tehcloud> it's somewhere in /var :p
2020-12-05 03:30:18 <cornfeedhobo> you would think!
2020-12-05 04:00:24 <hechos> anyone know why lo is DOWN?
2020-12-05 04:00:50 <hechos> booted live cd and ifup lo and bring UP
2020-12-05 04:01:08 <hechos> but on installed system always is DOWN
2020-12-05 05:02:37 <hechos> quit
2020-12-05 08:07:49 <rnkn> are there any adverse security implications to changing the user or group of fcgiwrap?
2020-12-05 08:08:47 <rnkn> I have a simple PR to allow the user/group to be configured, but it has gone stale
2020-12-05 13:19:36 <bahtsiz-bedevi> Hi everyone. Why can't install chromium on Alpine Linux 3.12.1? I get ""
2020-12-05 13:19:39 <bahtsiz-bedevi> ERROR: unsatisfiable constraints:   world[chromium]
2020-12-05 13:19:51 <_ikke_> bahtsiz-bedevi: you need to enable the community repo
2020-12-05 13:20:24 <bahtsiz-bedevi> I think it's enabled but let me check
2020-12-05 13:20:46 <_ikke_> https://pkgs.alpinelinux.org/packages?name=chromium&branch=v3.12
2020-12-05 13:20:49 <_ikke_> What arch are you on?
2020-12-05 13:21:17 <bahtsiz-bedevi> arm64
2020-12-05 13:21:39 <_ikke_> Should work then
2020-12-05 13:21:59 <bahtsiz-bedevi> ok making a clean install...
2020-12-05 13:24:08 <bahtsiz-bedevi> http://dl-cdn.alpinelinux.org/alpine/v3.12/mainhttp://dl-cdn.alpinelinux.org/alpine/v3.12/community
2020-12-05 13:24:20 <_ikke_> and apk --print-arch?
2020-12-05 13:24:40 <bahtsiz-bedevi> armhf
2020-12-05 13:24:48 <_ikke_> aha
2020-12-05 13:24:51 <_ikke_> that's not amd64
2020-12-05 13:24:55 <_ikke_> arm64*
2020-12-05 13:25:01 <_ikke_> that's arm32v6
2020-12-05 13:25:02 <bahtsiz-bedevi> wow
2020-12-05 13:26:09 <bahtsiz-bedevi> CoreELEC:/var/media/external # cat /proc/cpuinfoProcessor       : AArch64 Processor rev 4 (aarch64)processor       : 0processor       : 1processor       : 2processor       : 3Features        : fp asimd evtstrm crc32 wp half thumb fastmult vfp edsp neon vfpv3 tlsi vfpv4 idiva idivtCPU implementer : 0x41CPU architecture: 8CPU variant     : 0x0CPU
2020-12-05 13:26:10 <bahtsiz-bedevi> part        : 0xd03CPU revision    : 4Hardware        : AmlogicSerial          : 1f0c13003209a3a13e932786eb086f51model name      : Amlogic S905H rev c
2020-12-05 13:26:29 <_ikke_> Seems like you installed the armhf version of Alpine Linux then
2020-12-05 13:26:36 <_ikke_> which runs on aarch64
2020-12-05 13:26:53 <bahtsiz-bedevi> I just used this command
2020-12-05 13:26:54 <bahtsiz-bedevi> docker pull alpine
2020-12-05 13:27:20 <bahtsiz-bedevi> docker installed wrong acrh?
2020-12-05 13:27:23 <bahtsiz-bedevi> arch
2020-12-05 13:27:52 <_ikke_> probably, though, if the base system is aarch64, it should take aarch64 by default
2020-12-05 13:28:07 <bahtsiz-bedevi> this was my thought so
2020-12-05 13:29:07 <_ikke_> arm64v8/alpine should work
2020-12-05 13:41:36 <Apachez> its not easy when it comes to arm cores :)
2020-12-05 13:42:00 <Apachez> with or without 64bit, with or without fpu, with or without hardfloat etc
2020-12-05 14:19:15 <hechos> hello guys
2020-12-05 14:19:42 <hechos> still dealing with lo (loopback devices network is DOWN)
2020-12-05 14:19:48 <hechos> any idea how fix this?
2020-12-05 14:24:40 <artok> your /etc/network/interfaces has auto lo  and  iface lo inet loopback ?
2020-12-05 14:25:43 <hechos> artok: yes
2020-12-05 14:26:25 <hechos> im using dhcpcd on boot. but already try on openrc networking and get fails.
2020-12-05 14:27:09 <artok> it shouldn't touch your loopback, no dhcp there
2020-12-05 14:27:33 <hechos> no even without /etc/network/interfaces.
2020-12-05 14:27:54 <hechos> im install alpine over my own script to chroot.
2020-12-05 14:28:33 <hechos> what pacakge needs to networking basically?
2020-12-05 14:29:07 <_ikke_> busybox has ifupdown by default, but alpine recently switched to ifupdown-ng
2020-12-05 14:29:24 <hechos> _ikke_: hi.
2020-12-05 14:29:30 <artok> now then, if you have chroot
2020-12-05 14:29:35 <hechos> yes im all on stable repos.
2020-12-05 14:30:28 <artok> bind mounted sys and dev, proc mounted as it should?
2020-12-05 14:30:47 <hechos> system works stable but loopback as down.
2020-12-05 14:31:15 <hechos> im facing with this problem when using adb.
2020-12-05 14:31:48 <hechos> but im try live iso for call in some way, due with mkimg and lo devices wake-up correctly.
2020-12-05 14:31:55 <mps>  /etc/init.d/loopback start
2020-12-05 14:32:11 <hechos> mps: hi. thanks for help yesterday.
2020-12-05 14:32:16 <hechos> sitll looking for.
2020-12-05 14:32:18 <hechos> lol.
2020-12-05 14:33:00 <hechos> mps: WORKING THANKS!
2020-12-05 14:33:07 <mps> you probably didn't set all needed 'boot' scripts to be run on boot
2020-12-05 14:33:34 <hechos> mps so put on boot with rc-update?
2020-12-05 14:34:24 <hechos> * service loopback added to runlevel boot
2020-12-05 14:34:40 <mps> no, loopback is started automatically after clock script
2020-12-05 14:34:48 <hechos> hmm..
2020-12-05 14:34:52 <hechos> so?
2020-12-05 14:35:07 <hechos> for some reason doesnt start automatically.
2020-12-05 14:36:26 <mps> hechos: this is what I set when creating alpine on mmc cards in chroot https://tpaste.us/DMrl
2020-12-05 14:36:50 <_ikke_> normally the networking service should take care of everything
2020-12-05 14:37:12 <_ikke_> if you have networking and auto lo in /etc/network/interfaces, it should work
2020-12-05 14:37:17 <_ikke_> networking enabled*
2020-12-05 14:41:00 <hechos> mps: https://termbin.com/jy0x
2020-12-05 14:41:41 <hechos> _ikke_: no happen here, even try to disable dhcpcd and configure /etc/network/interfaces/
2020-12-05 14:41:49 <_ikke_> I don't see networking there
2020-12-05 14:42:03 <hechos> i know
2020-12-05 14:42:35 <hechos> if start networking with dhcpcd get fails.
2020-12-05 14:42:58 <_ikke_> How does your /etc/networking/interfaces look like?
2020-12-05 14:43:12 <mps> you don't need dhcpdc there, I think
2020-12-05 14:43:16 <_ikke_> correct
2020-12-05 14:43:22 <_ikke_> ifupdown starts dhcpcd
2020-12-05 14:43:36 <_ikke_> so s/udhcpcd/networking
2020-12-05 14:44:08 <hechos> _ikke_: in networking just go loopback configure.
2020-12-05 14:44:51 <_ikke_> You should add your other interface configuration there as well
2020-12-05 14:44:54 <hechos> auto lo
2020-12-05 14:44:55 <hechos> iface lo inet loopback
2020-12-05 14:45:40 <hechos> _ikke_: for some reason cannot get working networking automatic when have eth0 or wlan0 with wpa_supplicant.
2020-12-05 14:45:52 <hechos> so decide to use dhcpcd.
2020-12-05 14:46:02 <hechos> take easy path.
2020-12-05 14:46:43 <hechos> dont worry put on services if works, im done.
2020-12-05 14:46:57 <hechos> i dont iknow if tecnically bad.
2020-12-05 14:47:13 <hechos> my script tecnically sucks, but for me works.
2020-12-05 14:47:45 <mps> it doesn't, else you wouldn't ask same quetion 2 days :)
2020-12-05 14:47:46 <hechos> by the time to time get learning step with step.
2020-12-05 14:48:22 <hechos> mps: yes im asking for this more than two days.
2020-12-05 14:48:51 <hechos> dont know much about unix system, and even from gentoo.
2020-12-05 14:49:46 <hechos> sorry for troubles. well be hechos aka trouble maker or somethings like channelterror. lol.
2020-12-05 15:28:19 <hechos> mps: can confirm adb works fine right now. thanks again.
2020-12-05 15:29:23 <mps> hechos: np, enjoy :)
2020-12-05 15:31:06 <hechos> :)
2020-12-05 19:13:47 <cornfeedhobo> tehcloud: heh, it doesn't make the dirs until the first image pull. a change from ubuntu and centos that i didn't expect
2020-12-05 23:24:52 <cornfeedhobo> sshd's depend doesn't include net? that seems like an oversight, no?
2020-12-05 23:25:56 <mps> oversight?
2020-12-05 23:43:26 <MY-R> cornfeedhobo: it is just a warning because you put IP address in "ListenAddress" (/etc/ssh/sshd_config)
2020-12-05 23:45:03 <cornfeedhobo> well, it's more just something i stumbled on when looking for examples to write a custom service file
2020-12-05 23:45:27 <cornfeedhobo> sshd can't really do anything without networking, so it seems like an oversight
2020-12-05 23:46:27 <cornfeedhobo> it's most important because when one does `service networking restart` it doesn't trigger the restart of sshd as a dependent service
2020-12-05 23:47:17 <cornfeedhobo> (that might also be on purpose)
2020-12-05 23:47:24 <MY-R> cornfeedhobo: you can comment line "ListenAddress" or leave it with 0.0.0.0 and sshd will listen on all interfaces or put some IP but then is good to add in conf.d/sshd "rc_need="net.eth0" to be sure that IP is available
2020-12-05 23:47:54 <cornfeedhobo> MY-R: yeah i get that. that's not at all my point. the daemon won't bind to new networks brought up or down by networking service
2020-12-05 23:48:07 <cornfeedhobo> the daemon has to be hup'd at the very least
2020-12-05 23:48:23 <MY-R> thought sshd will deal with it by own, did you try it?
2020-12-05 23:49:05 <MY-R> cornfeedhobo: you dont get it, there is "need net" even with that warning
2020-12-05 23:49:59 <MY-R> line 36 in /etc/init.d/sshd
2020-12-05 23:50:13 <MY-R> so it will restart too
2020-12-05 23:53:02 <MY-R> oh wait, because I already mixing some stuff
2020-12-05 23:56:14 <cornfeedhobo> MY-R: i just did a fresh install. no `need net`
2020-12-05 23:56:25 <cornfeedhobo> that was the whole reason i commented :p
2020-12-06 00:07:01 <MY-R> cornfeedhobo: is probably intentional and that is why in sshd service is that tip so if you add line: rc_need="net" to /etc/conf.d/sshd then it will be restarting sshd also
2020-12-06 00:08:52 <MY-R> I just checked it to be sure :P
2020-12-06 00:09:35 <cornfeedhobo> MY-R: ooooh i see! thank you! interesting approach they have taken.
2020-12-06 00:09:54 <cornfeedhobo> this is good to know
2020-12-06 00:10:05 <MY-R> I thought you already saw warning during boot and wanted to get rid of it
2020-12-06 00:10:16 <cornfeedhobo> na, i was just digging around configs
2020-12-06 00:11:03 <cornfeedhobo> this is my first time using alpine outside of a container image, so i'm poking around
2020-12-06 00:11:27 <MY-R> I didnt even notice that sshd dont need net so I learnt something too :)
2020-12-06 00:11:37 <cornfeedhobo> haha yay!
2020-12-06 03:14:01 <andras> Hey guys my ssd doesn't show up as uefi in bios, any tips?
2020-12-06 04:34:54 <helmen[m]> I saw that on repo testing there is the package manager "nix". But shouldn't the applications not work because of musl?
2020-12-06 04:34:54 <helmen[m]> https://pkgs.alpinelinux.org/package/edge/testing/x86/nix
2020-12-06 10:39:20 <Jenkler> ikke,  .vimrc for busybox vi ?
2020-12-06 10:40:09 <Jenkler> ikke, is there a default file for that. or does busybox dont support a vim config file?
2020-12-06 10:41:04 <ikke> I rarely use bb vi, so I would not know
2020-12-06 10:41:12 <ikke> I just install vim
2020-12-06 10:42:20 <Jenkler> ahh, would be nice to only use bb vi ;)
2020-12-06 10:42:44 <Jenkler> but setting like indent would be nice
2020-12-06 10:45:19 <ikke> I don't see it reading any files on start, so I don't think it does anything like that
2020-12-06 10:47:33 <Jenkler> Few set options that work too :)
2020-12-06 10:48:19 <ikke> I see vi as disfunctional :)
2020-12-06 10:51:31 <mps> and vim is too big for base install
2020-12-06 10:51:52 <Jenkler> set expandtab hls list listchars=tab:>\ ,trail:~ shiftwidth=2 tabstop=2
2020-12-06 10:51:56 <Jenkler> that works
2020-12-06 10:52:02 <Jenkler> part of it anyway
2020-12-06 10:52:05 <Jenkler> hehe
2020-12-06 10:52:26 <Jenkler> not the list or listchars
2020-12-06 10:52:27 <mps> also i install vim as one of first packages on new 'system'
2020-12-06 10:53:09 <Jenkler> I want to keep it minimal
2020-12-06 10:53:20 <ikke> install nano :P
2020-12-06 10:53:32 <mps> :D
2020-12-06 10:53:45 <Jenkler> bb vi is nice; but wonder how the hell i read config att start ;(
2020-12-06 10:53:52 <ikke> it doesn't
2020-12-06 10:53:54 <ikke> I straced it
2020-12-06 10:54:05 <ikke> it does not look for any files
2020-12-06 10:54:44 <Jenkler> pipe in config workaround?
2020-12-06 10:57:02 <mps> "busybox vi -c 'set expandtab hls list listchars=tab:>\ ,trail:~ shiftwidth=2 tabstop=2' filename"
2020-12-06 10:57:37 <mps> and create alias with this, ofc
2020-12-06 10:59:55 <Jenkler> only tabstop=2 works
2020-12-06 11:00:11 <Jenkler> expandtab does not seam to work
2020-12-06 11:01:56 <ikke> ...............
2020-12-06 11:02:09 <ikke> sorry
2020-12-06 11:04:27 <Jenkler> mps, but -c is a nice solution. and EXINIT environment variable
2020-12-06 11:06:06 <mps> strings tell that autoindent, ignorecase, showmatch and tabstop= are supported
2020-12-06 11:08:28 <Jenkler> jepp
2020-12-06 11:08:42 <Jenkler> And the good thing is that its real space
2020-12-06 11:09:09 <Jenkler> tab makes space not tabs ;) so tabstop kinda works as it should
2020-12-06 11:09:50 <Jenkler> i Guess EXINIT is the best solution here. Maybe put it in ash startup
2020-12-06 11:13:10 <Jenkler> mps, nevermind. It seams that it uses tabs :(
2020-12-06 18:33:22 <David_9> Anyone home?
2020-12-06 19:25:20 <pinsetonl> Hi. I recently booted my laptop up and noticed that some of my applications (heavy ones) are ridiculously slow to start.
2020-12-06 19:27:44 <pinsetonl> I measured my disk speed and its ~2.7 MiB/s read and ~550 KiB/s write. Is it normal speed for SSD? I never measured this SSD before, so I have nothing to compare with.
2020-12-06 19:29:55 <pinsetonl> I measured it with: (read) ioping -A -D -s16k -c 10; (write) ioping -S64M -L -s4k -W -c 10 .
2020-12-06 20:39:07 <akiva> Hello everyone. Just a quick question: any guides on how I can attain full-disk encryption with btrfs on Alpine? looking to switch to it from Void.
2020-12-06 20:40:32 <akiva> Seems like if I choose the sys install, it has a preconceived idea as to what I want. Nor do I want LVM.
2020-12-06 20:50:53 <akiva> I cannot seem to find any mention of it (btrfs) being used for FDE. :l
2020-12-06 20:55:26 <ikke> I do see examples for zfs
2020-12-06 21:23:34 <akiva> ikke: Oh? Where? Maybe I can modify them accordingly. I am just looking at the setup installer and the wiki articles
2020-12-06 21:23:42 <akiva> Not that zfs is btrfs, but you know... It may help
2020-12-07 00:15:16 <Hello71> where do the default service setups come from? for arm and xen they're genapkovl*, but what about others?
2020-12-07 00:32:24 <Apachez> how do you mean?
2020-12-07 00:43:44 <Hello71> how does devfs get in sysinit
2020-12-07 02:00:25 <omni> akiva: the ZFS_on_LUKS or LVM_on_LUKS had what you needed?
2020-12-07 02:00:34 <omni> articles*
2020-12-07 02:01:39 <akiva> omni: Sorry, I had to step away. But, generally, no. I am looking for single DM_crypt block with btrfs
2020-12-07 02:02:16 <akiva> This is how I implement FDE on all my other machine's with Void or Arch
2020-12-07 02:07:40 <MY-R> akiva: you want just single LUKS partition with btrfs?
2020-12-07 02:08:55 <akiva> Yeah
2020-12-07 02:08:56 <akiva> Exactly
2020-12-07 02:09:29 <akiva> Sorry, my attention has been a bit spread... I should have clarified that.
2020-12-07 02:09:48 <MY-R> with help of above tutorials it is possible
2020-12-07 02:10:08 <akiva> OK, thanks. I will check those out tonight. I really appreciate it.
2020-12-07 02:12:23 <akiva> My concern with the link to the LVM was that the installer inistially seems to offern LVM, where as I didn't see any options for btrfs. Nor ZFS, so I will check that one out. I will be back later with updates. Thanks, everyone. Hoping to have success and move away from Void.
2020-12-07 02:17:20 <MY-R> well, the way to do it is more manual so you wont be using setup-alpine
2020-12-07 03:04:54 <akiva> MY-R: Excellent. I suppose that is what I was looking for. I had to find Void's manual install process, too, although that was an easy find. Seems like a lot of the links in the Alpine WIKI are dead lines. Well, maybe not a lot, but several so far.
2020-12-07 03:34:31 <MY-R> akiva: is enough to check those lvm on luks, zfs on luks etc tutorials to get idea how installation process looks, then omit points about zfs/lvm stuff and put btrfs
2020-12-07 03:36:08 <akiva> MY-R: Yes, my plan exactly. Thanks to everyone for pointing those out to me.
2020-12-07 10:11:53 <MY-R> is there a way to tell Alpine to bring up network interface (with static ip configuration) even if cable isnt pluged in to it? (or switch is not ready in my case)?
2020-12-07 10:35:20 <misterunknown> Hi. Is it possible to mount a new volume within a lxc container without restarting the container? (it's not a bind mount, but an ext4 lvm volume).
2020-12-07 11:08:36 <omni> MY-R: other than 'iface <iface name> inet static' etc in /etc/network/interfaces ?
2020-12-07 11:11:48 <omni> perhaps 'auto <iface name>' (together with the loopback interface, 'auto lo <iface name')
2020-12-07 11:12:01 <omni> interfaces(5)
2020-12-07 13:17:45 <hechos> hi guys.
2020-12-07 13:20:08 <hechos> got a rpi3 can be done installation on classic mode of OS to avoid diskless mode?
2020-12-07 13:52:42 <EF> hechos: https://wiki.alpinelinux.org/wiki/Classic_install_or_sys_mode_on_Raspberry_Pi
2020-12-07 13:53:33 <hechos> EF: yeah thanks, im reading now.
2020-12-07 17:05:28 <hs> So I am running Alpline x86_64 in Docker. I recently see a message related to  pyenv shell 3.8.2/ # pyenv shell 3.8.2/bin/sh: pyenv: not found
2020-12-07 17:05:41 <hs> Where does it come from ?
2020-12-07 17:13:38 <Apachez> where did you see that message?
2020-12-07 17:13:40 <artok> the shell ?
2020-12-07 17:14:15 <artok> what is your dockerfile and what is run command and what are you trying to do ?
2020-12-07 17:15:03 <artok> termbin hastebin etc to use
2020-12-07 17:25:06 <hs> > Executing task: docker run --rm -it  osc4ar/nodejs-chrome:latest <pyenv shell 3.8.2pyenv shell 3.8.2/ # pyenv shell 3.8.2/bin/sh: pyenv: not found/ #
2020-12-07 17:25:32 <hs> My Dockerfile: is really just FROM alpine
2020-12-07 17:29:59 <hs> Apachez the message happes with I start an interactive shell, but before the prompt shows.
2020-12-07 17:30:49 <hs> artok this is the /bin/sh -> /bin/busybox
2020-12-07 17:31:03 <midasi> We try to execute 'git pull' from https://git.alpinelinux.org/aports but always get the error: RPC failed; HTTP 429 curl 22 The requested URL returned error: 429 - fatal: the remote end hung up unexpectedly
2020-12-07 17:34:34 <Cogitri> midasi: You could pull from gitlab for the time being (https://gitlab.alpinelinux.org/alpine/aports)
2020-12-07 17:36:06 <midasi> cogitri: that works, thanks!
2020-12-07 17:36:27 <Cogitri> fwiw pulling from the link you sent works for me though
2020-12-07 17:38:30 <midasi> perhaps the issue is related to using a proxy? your link works without issues though
2020-12-07 18:19:17 <ikke> midasi: who is we?
2020-12-07 18:19:54 <ikke> And, with what are you pulling?
2020-12-07 19:44:45 <mps> anyone know for guide how to set users in dendrite (matrix server) for local usage
2020-12-07 19:45:37 <ikke> I could just register them
2020-12-07 19:45:47 <ikke> through a client\
2020-12-07 19:45:59 <mps> ah, which client
2020-12-07 19:46:19 <mps> i tried with gomuks but it doesn't work
2020-12-07 19:47:35 <ikke> element
2020-12-07 19:48:34 <mps> no such package
2020-12-07 19:50:16 <mps> ikke: do you run http proxy in front of dendrite
2020-12-07 19:50:58 <ikke> not in my test setup
2020-12-07 19:53:22 <mps> ok, thanks. I will try to search more
2020-12-07 19:55:13 <midasi> ikke: 'we' refers to the team in the enterprise I'm working for. We are pulling from command line on alpine
2020-12-07 19:56:21 <ikke> midasi: Reason I'm asking is because we put some rate limiting in place for some specific situations because some client / system is generating a lot of traffic
2020-12-07 21:36:10 <halosghost> anyone know of a way to contact Jens Gustedt?
2020-12-07 21:36:33 <halosghost> I want to send them a C&D about the C proposals…
2020-12-07 21:37:06 <ikke> THat name does not ring a bell
2020-12-07 21:37:14 <ikke> C&D meaning cease and desist?
2020-12-07 21:37:18 <halosghost> yeah
2020-12-07 21:37:22 <halosghost> I don't have the legal right to send one
2020-12-07 21:37:26 <halosghost> but something's gotta give
2020-12-07 21:37:57 <halosghost> namely, all the proposals to make C closer to C++
2020-12-07 21:37:59 <halosghost> just why
2020-12-07 21:38:04 <c705> you have every legal right to send it. doesn't say anything about whether or not it's enforcable
2020-12-07 21:38:09 <ikke> What has that to do with alpine :)
2020-12-07 21:39:22 <halosghost> ikke: there are people here who know C and it isn't ##c
2020-12-07 21:39:28 <halosghost> oh ##c, the bane of all C
2020-12-07 21:39:32 <halosghost> c705: fair
2020-12-07 21:39:39 <halosghost> c705: s/send/enforce/
2020-12-07 21:44:32 <mps> halosghost: hope he will not touch forth :)
2020-12-07 21:50:08 <Apachez> so a cease or desist for using c coding?
2020-12-07 21:50:15 <Apachez> what kind of drugs are you on? :)
2020-12-07 21:53:55 <halosghost> Apachez: a C&D for adding C++ things to C
2020-12-07 22:23:18 <hs> Does anyone have an idea about the
2020-12-07 22:24:28 <hs> Does anyone have an idea about where the 'pyenv' command would get interjected from?  Is it something Docker would do as part of starting running an image interactively?
2020-12-07 22:44:26 <alxsdf_> So I am having trouble with installing python's cryptography binary on latest edge
2020-12-07 22:44:39 <alxsdf_> specifically the 1.5.2 version that mitmproxy requires
2020-12-07 22:44:55 <alxsdf_> waaait
2020-12-07 22:45:27 <alxsdf_> nope
2020-12-07 22:45:39 <alxsdf_> I have tried installing libressl-dev but it doesn't replace openssl-dev
2020-12-07 22:45:53 <alxsdf_> it gives 'OK' but the header files specifically still have 'const' in them that causes the mismatch
2020-12-07 22:46:05 <alxsdf_> (what kind of person uses const in a fucking function prototype anyway ahhh)
2020-12-07 22:58:56 <Hello71> ... what?
2020-12-07 22:59:09 <Hello71> you mean like "const char *f(void)"?
2020-12-07 23:02:42 <halosghost> alxsdf_: to show intentionality? :)
2020-12-07 23:02:45 <halosghost> I do it frequently
2020-12-07 23:02:59 <alxsdf_> no I mean in function arguments
2020-12-07 23:03:28 <alxsdf_> it does communicate but practically speaking it does nothing except generate type mismatches between people who bother to write const and people who don't
2020-12-07 23:45:18 <Hello71> it's called "code correctness"
2020-12-07 23:45:50 <Hello71> you might as well go back to the bad old days where everything was int
2020-12-07 23:45:53 <Hello71> pointers are int
2020-12-07 23:46:31 <Hello71> hell, throw out function declarations entirely and just use implicit int
2020-12-07 23:55:55 <maxice8> Reject modernity go back to int
2020-12-08 00:03:02 <halosghost> alxsdf_: yeah, I definitely put them in function arguments (again, just to essentially guarantee callers that I'm not gonna touch the thing they pass in)
2020-12-08 00:03:18 <halosghost> alxsdf_: plus, it should always be safe to cast a mutable thing to a const-thing
2020-12-08 00:03:29 <halosghost> so, having it in a function parameter shouldn't cause anyone else headaches
2020-12-08 00:20:32 <Hello71> c... automatically does that?
2020-12-08 01:10:10 <akiva> So, I suppose if I only have the one PC available, I have to create my own customised Alpine bootable install in order to have extra kernel modules available to install from, like btrfs
2020-12-08 01:10:55 <akiva> Found it.
2020-12-08 02:27:10 <akiva> Does anyone know of a way I could build my own customised Alpine to put on USB? I want to add support for a few kernel modules and add a few packages to those pre-installed
2020-12-08 09:13:57 <PureTryOut[m]2> libucontext-dev doesn't cleanly upgrade from 0.12 to 0.13: `failed to rename usr/lib/.apk.56ddbe4a4e20d454c6af781a2b260c68864952ab06a50aa1 to usr/lib/pkgconfig`
2020-12-08 11:59:21 <radiomark> Is there a helper for generating APKFILES for python3 packages already in "pip" ?
2020-12-08 12:00:40 <ikke> I'm not aware of any
2020-12-08 12:00:51 <ikke> newapkbuild -y should be very close
2020-12-08 12:02:16 <radiomark> I see, maybe that is the starting point
2020-12-08 12:02:30 <radiomark> I wonder where people get the https://files.pythonhosted.org/ URLs from
2020-12-08 12:02:42 <maxice8> grepping a previous apkbuild
2020-12-08 12:02:43 <radiomark> I rather wondered if these were being automatically imported from pip information
2020-12-08 12:03:15 <maxice8> there is apkbuild-pypi but it is old compared to apkbuild-cpan which is amazing
2020-12-08 12:03:46 <radiomark> Yeah, I was hoping for an NPM one as well
2020-12-08 12:03:55 <radiomark> Porting over all the dependencies for this project
2020-12-08 12:04:01 <ikke> Tim Legge put a lot of work in the cpan helper
2020-12-08 12:05:28 <radiomark> Is there an irony that apkbuild-pypi is written in Perl
2020-12-08 12:05:39 <maxice8> no
2020-12-08 12:23:26 <radiomark> This gets messy very quickly, as the python setup.py automatically starts pulling dependencies
2020-12-08 12:25:02 <radiomark> There must be an art to this
2020-12-08 12:25:07 <radiomark> Or a lot of manual work is being done
2020-12-08 12:26:38 <ikke> I have not seen setup.py pulling dependencies
2020-12-08 12:26:49 <ikke> Something new?
2020-12-08 12:27:00 <radiomark> Maybe it's specific to this library: python-stripe
2020-12-08 12:27:39 <ikke> Problem is that python is going a way that will make it harder and harder for distros to package modules
2020-12-08 12:27:54 <radiomark> Yeah, that an npm as well
2020-12-08 12:28:47 <radiomark> I do not object to using their package managers, but I a couple of simple requirements: to freeze the package without an internet connection, and ensure I can use exact same package on development systems, test and deployment
2020-12-08 12:29:18 <ikke> And not require a compiler on prod :)
2020-12-08 12:29:25 <radiomark> That would be nice, too
2020-12-08 12:30:16 <radiomark> We operate very lean here, so it is preferable to freeze these components and change them only when warranted, not always be picking up the latest, which means keep finding bugs, changes etc.
2020-12-08 12:31:47 <radiomark> also if you "pip install" you then get a mix of packages already installed via APK, versus those coming via Pip
2020-12-08 12:33:23 <radiomark> I think these dependencies are probably coming in via strip-python's own build process
2020-12-08 14:35:47 <PureTryOut[m]2> The PyPi urls are quite easy really. https://pypi.python.org/packages/source/<first letter of the name of the package>/<name of the package>/<name of the package>-<version of the package>.tar.gz
2020-12-08 20:55:08 <iggy> you can't really pin alpine pkg versions anyway (since the mirrors only store the latest version), may as well use the language native pkg management
2020-12-08 21:18:00 <fourstepper> On a system with systemd, pipewire is run as a --user .socket - should this in alpine be handled just starting pipewire in shell profile or also somehow via openrc?
2020-12-08 22:31:47 <Hello71> running it from profile is the closest
2020-12-08 22:31:56 <Hello71> unless you run a user inetd or something
2020-12-09 03:45:40 <ipcloud> .
2020-12-09 04:19:39 <Magnie> Hello! I am netbooting Alpine Linux and trying to mount an SSD as a data volume (/var). However, despite being in /etc/fstab, it never mounts. I am also unable to manually mount the drive to /var because it is always "Resource Busy". However, it does mount /media/sda2. In general, it feels like Alpine isn't listening to /etc/fstab at all even if I
2020-12-09 04:19:39 <Magnie> comment/remove all mounts (it will still mount to /media/sda2 even if it isn't in /etc/fstab). Does anyone have any advice?
2020-12-09 06:31:46 <rnkn> where is the appropriate place for me to set TERM env var system-wide? I currently have it in /etc/profile
2020-12-09 06:34:48 <ikke> rnkn: if it's for users, /etc/profile is an appropriate place. For services, /etc/rc.conf
2020-12-09 06:35:11 <ikke> I'm not aware of any single place for the whole system
2020-12-09 06:35:28 <MY-R> rnkn: /etc/profile.d/*.sh
2020-12-09 06:35:50 <ikke> that's the drop-in variant of /etc/profile
2020-12-09 06:36:03 <MY-R> ye, but no reason to touch /etc/profile
2020-12-09 06:36:27 <rnkn> cool, thanks
2020-12-09 07:25:12 <kimmyk> can anyone recommend a guide to upgrading alpine?
2020-12-09 07:26:17 <kimmyk> mainly i'm trying to understand how people go from one verison to the next
2020-12-09 07:27:07 <kimmyk> (: or does everyone just install it readonly and replace the whole os at one time?
2020-12-09 07:42:58 <fourstepper> Hi kimyk, the guide is on the wiki
2020-12-09 07:43:14 <fourstepper> https://wiki.alpinelinux.org/wiki/Upgrading_Alpine
2020-12-09 08:10:10 <maldridge> perhaps not useful to your particular solution, but I prefer to just rebuild the universe to contain the version of alpine in it I want now
2020-12-09 08:11:53 <ikke> reference: https://xkcd.com/378/
2020-12-09 08:12:05 <maldridge> damn straight
2020-12-09 08:12:30 <maldridge> more realistically though its because I'm operating in a world of immutable golden images
2020-12-09 08:21:32 <kimmyk> thanks fourstepper i saw that and was trying to figure out how it fit into the recommended install where you run everything out of ram
2020-12-09 08:22:43 <kimmyk> maldridge: so do you install alpine to a small root partition as a read-only image and keep the rest somewhere else?
2020-12-09 08:23:04 <kimmyk> is that how alpine runs from ram?
2020-12-09 08:23:06 <maldridge> no I just rotate machines out of service to do upgrades where I can wipe the entire local disk
2020-12-09 08:23:15 <kimmyk> oh i see
2020-12-09 08:23:20 <maldridge> data lives on remote systems which ofc require special handling
2020-12-09 08:23:52 <kimmyk> (: i guess that makes more sense when you're running a whole shop not 1 or 2 systems at home
2020-12-09 08:24:24 <maldridge> yeah this is how I do it for a fairly significant number of hosts, one or two gets down into pet territory
2020-12-09 08:24:29 <kimmyk> so do you also have your apps live on remote systems next to the data?
2020-12-09 08:24:57 <maldridge> no apps fetch data on demand from other servers that store it via NFS or object storage protocols
2020-12-09 08:25:07 <maldridge> the alpine boxen that manage the data have many disks
2020-12-09 08:37:22 <kimmyk> i think i should download an old version and see what happens when i upgrade it a few times
2020-12-09 08:38:00 <maldridge> fwiw I've never had an alpine upgrade go bad, and I've done stupid stuff like upgrading live production hosts that were serving traffic from one version to another
2020-12-09 10:29:12 <sam_son> hey, I'm looking to migrate from centos (now that redhat killed it...) to alpine. My requirements are based on email servers (postfix/opendkim/spamassassin/etc) and I've found everything I need, except OpenDMARC. Anyone knows if there is a separate repo that has opendmarc? thanks
2020-12-09 10:30:55 <ikke> sam_son: apparently it's in testing
2020-12-09 10:31:12 <ikke> which means it's only available in edge (unstable branch)
2020-12-09 10:31:28 <ikke> https://pkgs.alpinelinux.org/package/edge/testing/x86_64/opendmarc
2020-12-09 10:31:46 <sam_son> ah I see, I was looking at v3.12
2020-12-09 10:31:55 <sam_son> ikke, thank you
2020-12-09 10:41:42 <sam_son> one more question, from the FAQ pages, it is not clear to me if its possible to upgrade from one major version (for example 3.11) to the next one (3.12) and so on. Could someone clarify this please?
2020-12-09 10:44:07 <ikke> sam_son: yes, that is certainly possible
2020-12-09 10:44:26 <sam_son> so there is no need to do a complete reinstall?
2020-12-09 10:44:31 <ikke> sam_son: correct
2020-12-09 10:45:15 <sam_son> what kind of procedure does alpine use? does it just upgrade individual packages like Fedora does?
2020-12-09 10:45:21 <ikke> yes
2020-12-09 10:45:43 <sam_son> ah I see, thanks for the help
2020-12-09 10:45:45 <ikke> you point to the repos to the new version and run apk upgrade --available
2020-12-09 10:47:05 <sam_son> which is something that needs to be done within the 2 years of support, is that correct?
2020-12-09 10:49:21 <fourstepper> Is there somewhere on the net a ready-made alpine container with a running openrc-init as PID1? Just for testing purposes in CI to test for physical machines.
2020-12-09 10:49:40 <fourstepper> I tried looking around for a bit but couldn't find much that would work well
2020-12-09 10:50:03 <ikke> fourstepper: not aware of any
2020-12-09 10:50:32 <ikke> sam_son: after the 2 years, that repo rarely gets any updates anymore
2020-12-09 10:50:58 <ikke> but you can certainly still upgrade afterwards (though it can become more difficult as the amount of changes grows larger)
2020-12-09 10:52:16 <sam_son> hmm
2020-12-09 10:53:39 <mps> 'apk update' && 'apk upgrade -a -s' weekly
2020-12-09 10:53:55 <sam_son> I need to adjust my expectations and practices, I've been spoiled by centos's 8 years of support.
2020-12-09 10:54:18 <ikke> right. We're a smallish community distro, so we cannot afford to give that long time support
2020-12-09 10:54:59 <sam_son> sure, that is understood, as I mentioned I need to adjust my practices
2020-12-09 10:56:48 <sam_son> ok one last question and I'll stop pestering you. Does alpine have a similar security mechanism to SELinux? for granular security contexts etc
2020-12-09 10:59:29 <mps> heh, no
2020-12-09 11:00:09 <sam_son> ok
2020-12-09 11:00:10 <mps> I call this 'false sense of security'
2020-12-09 11:00:38 <mps> security is process and not 'product'
2020-12-09 11:01:41 <sam_son> its a security layer and has been very helpful to minimize security issues for me, so its a tried and tested system, indeed security is a process and SELinux and been a valuable part of that process.
2020-12-09 11:02:44 <mps> heh ;)
2020-12-09 11:03:51 <mps> 
ACTION bacported selinux on debian when it first introduced to the previous stable release

2020-12-09 11:04:20 <mps> unofficial backport though
2020-12-09 11:04:28 <sam_son> it is just another security layer on top of other layers...
2020-12-09 11:05:18 <mps> having many/some number of layers can also be security problem (and nightmare for sure)
2020-12-09 11:05:48 <mps> but yes, layers are good to get money
2020-12-09 11:06:40 <sam_son> obviously you have some issues with SELinux that you need to resolve...
2020-12-09 11:07:09 <detha> mps: how much effort was it to get selinux on debian? person-weeks/months/years?
2020-12-09 11:07:17 <ikke> it also requires a lot of effort to get right
2020-12-09 11:07:32 <sam_son> ikke, that is very true
2020-12-09 11:08:57 <mps> detha: I can't remember, that was more than a decade ago, but about one month
2020-12-09 11:09:53 <mps> I mean, month to backport it to previous release which didn't had it at all
2020-12-09 11:10:20 <detha> Less than I expected. With the current centos mess, I see an opportunity for something stable, not-centos, and having selinux
2020-12-09 11:11:03 <mps> sam_son: I'm not trying to tell that you shouldn't 'invest' in security, it is always good
2020-12-09 11:11:25 <mps> detha: iirc debian have it
2020-12-09 11:11:39 <detha> very half-heartedly
2020-12-09 11:12:50 <mps> but I think apparmor is maybe better option, less risky, less work needed, and is more resource friendly
2020-12-09 11:13:40 <sam_son> many business are looking to jump ship now that centos is dead, providing something that is attractive may bring a lot more attention to you than you'd expect.
2020-12-09 11:14:00 <detha> I have never been a fan of apparmor, for various reasons. selinux made a lot more sense
2020-12-09 11:15:00 <detha> Initially selinux was a total pain, until the tooling around it grew up
2020-12-09 11:15:42 <mps> I dislike all them, even grsec
2020-12-09 11:16:39 <sam_son> SELinux won't stop a breach... but it will minimize the damage. I've seen it work to great effect on compromised web servers, the intruder could breach the web application but couldn't even run a simple 'ls -la'. I don't see many other systems providing that kind of restriction.
2020-12-09 11:17:26 <sam_son> in most systems, once there is a breach... its over... everything is compromised.
2020-12-09 11:18:44 <sam_son> meeting...bbl
2020-12-09 11:18:45 <detha> Exactly. In the days I was running centos servers I have seen a few 'Bad CVE against $foo, everybody patch immediately' where I could sit back with a 'I'll patch next week or so, selinux catches it already'
2020-12-09 11:20:38 <detha> On the other hand, doing anything even slightly non-standard meant more work so make a profile for the thing
2020-12-09 11:24:31 <fourstepper> At work, we are currently also considering to which distro we will move as CentOS is on it's way out. We have settled on using Alpine and Debian as an alternative for systems that would be hard to get up and running with Alpine (say a Zimbra mail server that we use for our clients). Looking forward to what the switch will bring to the table
2020-12-09 11:52:18 <DivanSantana> For those considering an enterprise alternative to centos, alpine might be a good one. Could one run an enterprise cassandra/percona/java8/kafka etc all on alpine? Some of these projects may not directly support or mention alpine. But perhaps it would work fine, via the rpm package on alpine.  Lack of systemd might require to adapt start up of some of these apps.
2020-12-09 11:54:12 <ikke> DivanSantana: alpine is based on musl, so packages from glibc based systems would not work directly on alpine (though there are some shims that could help)
2020-12-09 12:09:10 <fourstepper> As far as apparmor goes, is the support for it included in the kernel that comes with alpine linux or does a custom kernel have to be built?
2020-12-09 12:09:50 <fourstepper> I see it's included as a package in the repositories, but not sure if it will work out of the box just installing and enabling it at boot if we were to use it
2020-12-09 12:10:30 <DivanSantana> ikke:interesting. OK. Will have to research into that. Thanks
2020-12-09 12:11:01 <ikke> DivanSantana: most of these are java applications, so they should work
2020-12-09 12:11:25 <ikke> A dockerized kafka based on Alpine for example: https://github.com/blacktop/docker-kafka-alpine/blob/master/2.6/Dockerfile
2020-12-09 12:12:44 <DivanSantana> ikke: Great.
2020-12-09 12:32:57 <gjabell> anyone know if javafx isn't in the alpine repos because it doesn't work, or because no one has packaged it yet?
2020-12-09 12:34:55 <Cogitri> fourstepper: Apparmor should work just by installing it and setting `security=apparmor` in the boot parameters
2020-12-09 12:35:11 <Cogitri> You could check `aa_status` once you're booted into it
2020-12-09 12:45:42 <fourstepper> Thanks Cogitri will check it out
2020-12-09 12:53:25 <markand> anyone running edge and has firefox systematically crashing?
2020-12-09 12:56:56 <mps> markand: I have that on 3.12, every morning it is killed by OOM
2020-12-09 12:57:38 <markand> me it opens, but immediately reports "Your tab has just crashed"
2020-12-09 12:58:09 <markand> I get a lot of: (/usr/lib/firefox/firefox:5825): Pango-WARNING **: 12:57:51.590: font_face status is: out of memory
2020-12-09 12:58:26 <markand> which is strange since I'm definitely *not* out of memory
2020-12-09 12:58:32 <mps> on edge I rebuilt it without dbus, wifi, pipewire and libpulse, and it stable
2020-12-09 13:00:12 <Cogitri> markand: What font do you have for your system&FF?
2020-12-09 13:00:40 <markand> defaults in GNOME, Cantarell, but I've tried using something different just for test, same errr
2020-12-09 13:01:24 <Cogitri> Hm yes, definitely shouldn't happen with Cantarell
2020-12-09 13:01:38 <Cogitri> Seems like some folks have that error w/ pango with Windows fonts, that's why I was asking :)
2020-12-09 13:01:48 <ikke> markand: is this esr?
2020-12-09 13:01:48 <markand> I also have errors like: Sandbox: seccomp sandbox violation: pid 6121, tid 6121, syscall 16, args 2 21523 140728634258600 13 1 140278967713020.
2020-12-09 13:02:08 <markand> ikke, no plain firefox
2020-12-09 13:02:23 <Cogitri> That error certainly is more fatal :)
2020-12-09 13:02:56 <Cogitri> That'll certainly kill your tab process
2020-12-09 13:03:01 <markand> firefox-esr works
2020-12-09 13:03:12 <markand> I'll stick to it for the time being
2020-12-09 13:03:17 <Cogitri> Hum, what's syscall 16
2020-12-09 13:03:18 <ikke> huh. strange
2020-12-09 13:03:31 <ikke> markand: my experience (when testing it) was the opposite
2020-12-09 13:03:36 <markand> ikke, :P
2020-12-09 13:03:42 <ikke> esr crashing while the latest version did not have issues
2020-12-09 13:04:49 <Cogitri> I guess we might have to punch another hole through the seccomp filter with __NR_ioctl, but weird that that only happens on your machine
2020-12-09 13:04:56 <Cogitri> And I had thought the sandbox already allowed that
2020-12-09 13:30:15 <fourstepper> How does apk -U upgrade interact with packages that have been pinned to a different apk source? Say I add the edge repo as @edge in /etc/apk/sources and run apk add app@edge and then after a while would run apk -U upgrade, how would the interaction look there?
2020-12-09 13:32:55 <TedRoastBeef> God Day! does someone know how i can Tell Apline to run a script after the initramfs is done loading? im trying to boot from pxe without a overlayfs and run a script which sets things up
2020-12-09 13:33:51 <TedRoastBeef> I should also mention that this script should only run when no overlayfs is defined with apkovl
2020-12-09 14:08:29 <yZ5vlALg86lP> if nothing else you can always adapt your init script, sucks on kernel updates, but is a sure way to do it
2020-12-09 14:09:48 <TedRoastBeef> that what im trying right now ^^ maybe i will find a neat way :)
2020-12-09 14:12:11 <oliv3r[m]> Hey All, I'm trying to figure out how the public keys are named. They contains a crc32 it seems, but when running crc32 on the file, the tool actually figures and finds the crc32, but complains about it not matching. Where is the repo/code that generates these alpine-keys? I know the APKBUILD has the files pre-generated.
2020-12-09 14:14:16 <ikke> abuild-keygen creates them
2020-12-09 14:14:35 <ikke> oliv3r[m]: https://gitlab.alpinelinux.org/alpine/abuild/-/blob/master/abuild-keygen.in
2020-12-09 14:32:08 <oliv3r[m]> perfect, thanks :)
2020-12-09 14:32:30 <ikke> seems like it's a hex format of the date
2020-12-09 14:33:04 <oliv3r[m]> yeah but it also reminded me, I could have just used abuild-keygen instead, maybe i used openssl to have stronger keys though
2020-12-09 14:33:57 <ikke> You can specify the amount of bits with abuild-keygen
2020-12-09 14:34:03 <oliv3r[m]> funny that it's the date, not the crc32 of the file itself
2020-12-09 14:34:08 <ikke> I don't think the name of the key matter, only that it remains the same
2020-12-09 14:34:22 <oliv3r[m]> i saw that some where the name is important
2020-12-09 14:34:36 <ikke> the name is stored and used to match the correct public key
2020-12-09 14:34:48 <oliv3r[m]> but the content itself, doesn't matter indeed, as a colleague just randomly named it :p maybe even copy/pasted the hex from a diff key
2020-12-09 14:43:04 <Hello71> gjabell: probably nobody has packaged it yet, there are not many users
2020-12-09 14:46:28 <gjabell> Hello71: alright, thanks :)
2020-12-09 14:47:13 <Hello71> i think there are a bunch of dependencies for a clean packaging
2020-12-09 14:47:36 <Hello71> fourstepper: https://git.alpinelinux.org/aports/tree/main/linux-lts/config-lts.x86_64 CONFIG_SECURITY_APPARMOR=y
2020-12-09 14:47:48 <Hello71> also see linux-edge and linux-virt
2020-12-09 14:49:03 <fourstepper> Thank you, should've been obvious to go check the APKBUILD repo
2020-12-09 14:51:27 <Hello71> apk -U shouldn't do anything if your system is working properly
2020-12-09 14:52:18 <Hello71> upgrade -a is the useful one
2020-12-09 14:52:30 <mps> Hello71: I'm tempted long time to disable CONFIG_SECURITY_APPARMOR :)
2020-12-09 14:53:16 <Hello71> firejail uses it by default, but firejail is a security nightmare anyways
2020-12-09 14:53:28 <Hello71> at least from host security standpoint
2020-12-09 14:53:48 <mps> though after long resistance I enabled BPF yesterday for linux-edge
2020-12-09 16:11:54 <omni> where did the resistance lie?
2020-12-09 16:35:59 <Hello71> at mps
2020-12-09 16:40:47 <omni> thought so, just checkin'
2020-12-09 16:54:05 <fourstepper> BPF? Also, what's the problem with the CONFIG_SECURITY_APPARMOR=y mps?
2020-12-09 17:42:13 <mps> omni: Hello71: the idea to have interpreter in kernel doesn't look good, imo
2020-12-09 17:43:28 <mps> and this is not only my opinion, alpine developers (most of them) also think so
2020-12-09 17:44:10 <mps> and this is debated many times on lkml (linux kernel ML)
2020-12-09 17:46:25 <mps> fourstepper: well, it was a small joke about APPARMOR, personaly on kernels I run on workstation I have minimum options and drivers enabled, though on servers I use default alpine kernels
2020-12-09 17:48:05 <mps> so no, I never wanted to disable APPARMOR actually, but every time I work on kernel upgrades and see it I wonder why it is enabled and not some other also
2020-12-09 17:58:56 <l1x> hi guys, i am working on a project to use Alpine as the host and guest system for microVMs. the tool that we use is called Firecracker. I am able to use alpine as host but when I try to use it as guest it does not see the network interface at all
2020-12-09 17:59:30 <l1x> i am trying to figure out how an alpine rootfs can cause this, because the kernel I am using is the same for all linux distros
2020-12-09 17:59:53 <l1x> could somebody point me the right direction to read on how /etc/network/interfaces work in alpine?
2020-12-09 18:00:39 <mps> l1x: mostly like it works on other distros
2020-12-09 18:01:20 <mps> config is simple, like debian to some degree
2020-12-09 18:01:26 <l1x> ip: can't find device 'eth0'
2020-12-09 18:01:26 <l1x> ip: ioctl 0x8913 failed: No such device
2020-12-09 18:01:35 <l1x> yeah mean the next step after that
2020-12-09 18:01:48 <l1x>  /etc/init.d/networking restart
2020-12-09 18:01:59 <mps> cat /proc/net/dev
2020-12-09 18:02:01 <l1x> i am trying to understand what could be missing
2020-12-09 18:02:16 <l1x> thanks mps !
2020-12-09 18:02:17 <l1x> cat /proc/net/dev
2020-12-09 18:02:17 <l1x> Inter-|   Receive                                                |  Transmit
2020-12-09 18:02:17 <l1x>  face |bytes    packets errs drop fifo frame compressed multicast|bytes    packets errs drop fifo colls carrier compressed
2020-12-09 18:02:17 <l1x>     lo:       0       0    0    0    0     0          0         0        0       0    0    0    0     0       0          0
2020-12-09 18:02:25 <l1x> it does not see eth0 at all
2020-12-09 18:02:38 <mps> yes, driver is not loaded
2020-12-09 18:02:57 <l1x> if i have a totally static kernel and the driver is the virtio net driver
2020-12-09 18:03:06 <l1x> isnt it supposed to be working?
2020-12-09 18:03:27 <mps> well, I don't know what is firecracker and how it works
2020-12-09 18:03:55 <l1x> imagine qemu without the XML frontends and all that crap
2020-12-09 18:04:10 <l1x> it emulates a super simple CPU and few devices with extremely low overhead
2020-12-09 18:04:27 <mps> ah, if it is like qemu then it should load drivers on boot
2020-12-09 18:04:38 <l1x> yes, this is what I thought
2020-12-09 18:04:50 <l1x> it is so weird it doesnt
2020-12-09 18:05:07 <Hello71> are you using virtio mmio or virtio pci
2020-12-09 18:05:28 <mps> you have to look are all init services are started on boot, especially hwdrivers
2020-12-09 18:07:28 <mps> l1x: alpine runs fine for a lot of users in qemu
2020-12-09 18:07:48 <l1x> yeah
2020-12-09 18:08:10 <l1x> ok, i am going to go another deep dive into this, maybe i screw up something with the kernel config
2020-12-09 18:08:15 <l1x> it must work
2020-12-09 18:08:24 <l1x> Hello71: i think so
2020-12-09 18:08:34 <Hello71> uh.
2020-12-09 18:10:18 <mps> btw, I always use qemu from cli or script, not XML as you say 'all that crap' :)
2020-12-09 18:11:15 <Hello71> yes, qemu doesn't use any xml afaik
2020-12-09 18:11:39 <mps> right
2020-12-09 18:11:51 <Hello71> qom is some dsl
2020-12-09 18:13:17 <mps> though it have libxml2-dev in makedepends, and I don't know why, and never looked
2020-12-09 18:14:31 <Hello71> apparently need for gdbstub
2020-12-09 18:16:07 <mps> looks like qemu build system is somewhat intelligent to skip things it didn't found during configure and use those it finds
2020-12-09 18:17:39 <l1x> yeah it is optional
2020-12-09 18:18:19 <mps> imagine, we have libpulse in qemu makedepends, :D
2020-12-09 18:18:20 <l1x> i think the problem is the kernel i am using
2020-12-09 18:18:26 <l1x> hehe
2020-12-09 18:18:29 <l1x> fun time
2020-12-09 18:18:39 <mps> yes
2020-12-09 18:19:12 <l1x> firecracker is just a super lean qemu that emulate 2 kind of cpu, x86 / arm and only one type in each arch
2020-12-09 18:19:20 <l1x> so the performance loss is minimal
2020-12-09 18:19:35 <l1x> it is designed to run production workloads in amazon
2020-12-09 18:19:37 <mps> is it free and open source?
2020-12-09 18:19:40 <l1x> yes
2020-12-09 18:19:58 <l1x> https://github.com/firecracker-microvm/firecracker
2020-12-09 18:20:07 <l1x> uses kvm under the hood
2020-12-09 18:20:19 <mps> thanks, will look
2020-12-09 18:20:31 <l1x> alpine is a perfect match to this and they use alpine in their docs to demo the tool
2020-12-09 18:27:26 <mps> oh, rust :P
2020-12-09 18:27:26 <mps> no thanks
2020-12-09 18:27:26 <l1x> we could replace some of the docker workloads with this, because fc also supports cgroups out of the box so you get a super lean VM + container at the same time
2020-12-09 18:28:14 <l1x> haha what is wrong with it?!
2020-12-09 18:28:21 <mps> everything is wrong with rust
2020-12-09 18:28:22 <eipi> hello all, i am having some trouble with 3.12.1 on vmware. i cannot get the shared folders to automount on boot. I have tried everything in the book. if i mount them manually everything is hunky dory, but it when i edit /etc/fstab and reboot, no luck. has anyone had some experience with this? I used the exact same command as the open-vm-tools
2020-12-09 18:28:22 <eipi> release notes here: https://github.com/vmware/open-vm-tools/blob/stable-11.1.x/ReleaseNotes.md I also tried the other version from vmware here: https://kb.vmware.com/s/article/60262
2020-12-09 18:28:29 <ipcloud> podman in alpine?
2020-12-09 18:28:31 <l1x> mps: yeah there are some things that are wrong but the fact that you can build with your choice of libc (or no libc at all) out of the box without hacking makes it pretty convenient
2020-12-09 18:28:32 <eipi> can someone show me how to enable messages from fstab mount on boot?
2020-12-09 18:28:32 <l1x> ipcloud: sort of
2020-12-09 18:28:32 <l1x> eipi: cat /etc/fstab?
2020-12-09 18:28:32 <eipi> i can see the file
2020-12-09 18:28:32 <ipcloud> l1x: have alpine 'pods' running on fedora/centos ( :( ) with networking ~working in the pods
2020-12-09 18:28:32 <eipi> ahh thanks
2020-12-09 18:28:32 <mps> eipi: 'rc_verbose=yes' in /etc/rc.conf
2020-12-09 18:28:32 <l1x> ipcloud: yeah i think i screwed up the kernel config
2020-12-09 18:28:32 <ipcloud> starting some eval of alpine as workstation and looking to see if podman on alpine would be a resonable environment to work in
2020-12-09 18:28:32 <l1x> i am trying to create a super minimal kernel and i might have accidentally removed something that i should not have
2020-12-09 18:28:32 <mps> l1x: rust is made to better c++ and as such it is bad because c++ is bad. and rust added their 'cruft' a top
2020-12-09 18:28:32 <l1x> ipcloud: what is your usecase?
2020-12-09 18:29:01 <ipcloud> l1x: dev workstation running containers for intial development (my company custom applicaiton(s) and dB servers and httpd servers ...)
2020-12-09 18:29:25 <l1x> mps: sure, i agree with you. however look at the tooling. show me one more language that can do rustc --target powerpc-unknown-linux-musl
2020-12-09 18:29:41 <l1x> ipcloud: this is exactly what we are working on too
2020-12-09 18:29:42 <omni> "XML and all that crap" - libvirt?
2020-12-09 18:29:59 <mps> l1x: did you looked zig
2020-12-09 18:30:00 <l1x> omni: yeah i think i was running into xml there
2020-12-09 18:30:07 <l1x> mps: absolutely, yes
2020-12-09 18:30:18 <eipi> im getting no messages from the fstab during boot
2020-12-09 18:30:27 <l1x> it is years away from being ready for prime time
2020-12-09 18:30:35 <eipi> if i type mount -a on the command line then my share is mounted, but it just wound mount from fstab
2020-12-09 18:30:40 <ipcloud> l1x: have some work on this using nspawn but seems heavy
2020-12-09 18:30:48 <mps> l1x: also crystal to some degree but it is not intended as system lang
2020-12-09 18:31:25 <ipcloud> mps: l1x: is golang fine
2020-12-09 18:31:51 <mps> ipcloud: yes, but also not intended to be system lang
2020-12-09 18:32:15 <l1x> mps: totally
2020-12-09 18:32:16 <mps> go is more 'service' oriented
2020-12-09 18:33:03 <ipcloud> mps: point is building the system v running a service (daemon) using golang, if I understand correctly.
2020-12-09 18:33:15 <l1x> ipcloud: sure
2020-12-09 18:33:22 <mps> eipi: yes
2020-12-09 18:33:28 <mps> sorry
2020-12-09 18:33:36 <mps> ipcloud: yes
2020-12-09 18:33:48 <eipi> is there a way to see the order during boot, perhaps the fstab mounts prior to open-vm-tools loading?
2020-12-09 18:34:09 <mps> eipi: does all services run on boot, rc-update will show them
2020-12-09 18:34:48 <mps> rc-update will show services with their runlevel
2020-12-09 18:36:08 <eipi> rc-update seems to be alphabetical order
2020-12-09 18:36:23 <eipi> i see open-vm-tools is on boot
2020-12-09 18:36:24 <eipi> as is fuse
2020-12-09 18:36:34 <eipi> but when is the fstab executed?
2020-12-09 18:37:14 <eipi> why else is my mount not visible after boot, but it is visible on mount -a ? that means everythign is correct, but just not wokring at boot
2020-12-09 18:41:53 <mps> eipi: did you installed alpine using setup-alpine or some 'by hand' method
2020-12-09 18:42:10 <eipi> setup-alpine
2020-12-09 18:42:13 <eipi> on sys mode
2020-12-09 18:42:26 <eipi> root
2020-12-09 18:42:36 <mps> hmm, then it is very strange
2020-12-09 18:43:09 <mps> and you don't see any 'mounting ....' message during boot?
2020-12-09 18:43:35 <eipi> no, i agree very strange
2020-12-09 18:43:52 <eipi> and if i boot and login right away and just do mount -a it works fine
2020-12-09 18:44:04 <eipi> i already tried deleting the VM and install again from scratch
2020-12-09 18:44:08 <eipi> same problme
2020-12-09 18:45:09 <mps> 'Mounting /run ...' and similar
2020-12-09 18:45:49 <mps> '* Remounting root filesystem read/write ... [ ok ]' etc
2020-12-09 18:47:38 <eipi> i have mounting /run and /dev/mqueue. mountig debug filesystem, moutning local filesystems
2020-12-09 18:47:40 <eipi> etc
2020-12-09 18:47:59 <eipi> starting open-vm-tools and fuse is much later
2020-12-09 18:48:06 <eipi> maybe this is the problem?
2020-12-09 18:48:25 <mps> I have no experience with wmvare
2020-12-09 18:49:38 <eipi> im curious whre can i edit the order of these items on boot
2020-12-09 18:50:34 <mps> rc-update del svcname runlevel
2020-12-09 18:50:41 <mps> rc-update add svcname runlevel
2020-12-09 18:56:26 <eipi> i tried rc-update del open-vm-tools boot and then rc-update add open-vm-tools sysinit same problem
2020-12-09 18:56:48 <eipi> boot log says, "mounting local filesystems" and then the next line is "starting open vm tools"
2020-12-09 18:57:00 <eipi> so still cant get it to be first
2020-12-09 18:57:28 <fourstepper> Thanks for explanation regarding apparmor mps :D
2020-12-09 18:57:39 <eipi> after login mount -a still no problem
2020-12-09 19:00:16 <mps> eipi: you should set somewhere in open-wm-tools option 'before localmount' (or something like that). but again I don't know how open-wm-tools work
2020-12-09 19:02:30 <mps> fourstepper: yaw :)
2020-12-09 19:11:42 <eipi> @mps i got it.
2020-12-09 19:11:51 <eipi> I moved open-vm-tools to sysinit
2020-12-09 19:12:07 <eipi> and then rc-update add local default
2020-12-09 19:12:23 <eipi> then just added mount command to /etc/local.d
2020-12-09 19:12:29 <eipi> the problem is the services not started
2020-12-09 19:13:02 <mps> ah, that explains
2020-12-09 19:13:02 <eipi> i will try to post somewhere online maybe in a gist for future google search people to save them my headaches of 12 hours
2020-12-09 19:13:14 <eipi> thanks
2020-12-09 19:13:33 <eipi> this is my first time working with alpine and i am very happy how smart it is
2020-12-09 19:13:51 <eipi> anyway now centos is dead im very happy already to find my new home
2020-12-09 19:14:01 <mps> yes, and it is documented many time that when add something in local service should be enabled
2020-12-09 19:14:31 <eipi> i didnt mean that local service not started, i know that from the docs
2020-12-09 19:14:47 <mps> I just told on one other alpine channel that I expect a lot refuges here :D
2020-12-09 19:14:51 <eipi> i meant that when fstab mount the vm share, the open-vm-tools service not yet mounted
2020-12-09 19:15:02 <eipi> not yet running*
2020-12-09 19:15:15 <eipi> so this solution not the best but will work
2020-12-09 19:15:36 <eipi> its a good time for alpine enthusiats to write some blog and social media posts
2020-12-09 19:15:42 <eipi> welcome centos refugees
2020-12-09 19:15:51 <eipi> and will get a lot of people looking for sure
2020-12-09 19:16:03 <eipi> thanks for your help very much
2020-12-09 19:16:31 <mps> eipi: np :)
2020-12-09 19:17:19 <fourstepper> I would really love to see some statistics of iso downloads ever since yesterday :D
2020-12-09 19:18:50 <mps> we don't spy on our users
2020-12-09 19:19:29 <ikke> And due to all the mirrors, even impossible
2020-12-09 19:19:39 <mps> yes
2020-12-09 19:19:52 <mps> and, who cares
2020-12-09 19:20:29 <maxice8> mps: the person who just asked
2020-12-09 19:20:59 <mps> though we can ask mirrors to send us statistic ;p
2020-12-09 19:21:53 <mps> maxice8: person added ':D' so I feel person doesn't care much for this
2020-12-09 19:21:56 <tehcloud> good, more refugees to Alpine means more testing against musl and then I have less broken C codebases to fix on my own
2020-12-09 19:22:51 <Hello71> assuming optimistically that they won't add broken fixes or spread fud about musl being broken
2020-12-09 19:23:19 <mps> tehcloud: I hope for more serious and knowledgeable developers
2020-12-09 19:23:36 <ikke> One can dream
2020-12-09 19:23:45 <tehcloud> there is a certain codebase I have my eyes on right now that has some good old fashioned UB
2020-12-09 19:24:12 <mps> Hello71: yes, that is what we will see on the beginning, I fear
2020-12-09 19:25:00 <mps> only solution for this is to appoint clandmeter as BDFL ;)
2020-12-09 19:26:04 <Hello71> also centos users are annoying, but i don't want more ubuntu users either
2020-12-09 19:26:45 <tehcloud> this is not a distro that attracts Ubuntu users
2020-12-09 19:27:02 <tehcloud> at least, not en masse
2020-12-09 19:27:09 <mps> heh, most refuges from other distros are annoying at beginning
2020-12-09 19:28:41 <mps> but I hope serious ones will accept alpine as it is, and rest will wander away
2020-12-09 19:29:09 <mps> and write blog how alpine/musl/openrc are bad :P
2020-12-09 19:29:10 <tehcloud> yes, I like this distro how it is now, very minimal
2020-12-09 19:29:33 <ipcloud> mps: I think I am here and will work to be serious
2020-12-09 19:29:37 <Hello71> i think there will be a fair sum of centos users who will go to ubuntu because it is "popular"
2020-12-09 19:29:59 <Hello71> which is bad for the whole ecosystem
2020-12-09 19:30:12 <mps> ipcloud: you are welcome
2020-12-09 19:31:30 <eipi> Alpine really is a breath of fresh air
2020-12-09 19:31:46 <eipi> i havent been this excited to work with linux for 10+ years
2020-12-09 19:32:00 <eipi> i have been on centos for 10+ years
2020-12-09 19:32:35 <eipi> Hello71 I agree that Ubuntu has "name recognition" and will grab
2020-12-09 19:34:04 <tehcloud> it is the closest thing to the feel of FreeBSD when it was in its prime for me
2020-12-09 19:34:05 <mps> I have to make shortcut/macro for irssi which write 'https://docs.alpinelinux.org' message
2020-12-09 19:41:01 <clandmeter> mps: are you in need of my support?
2020-12-09 19:45:44 <mps> clandmeter: without your support and encouragement I never will be involved in alpine, and most important your friendship
2020-12-09 19:46:30 <clandmeter> You have lts my friend :)
2020-12-09 19:48:07 <mps> thanks, I have no doubts
2020-12-09 19:54:24 <eipi> is there an easy way to do custom scripting of alpine post install?
2020-12-09 19:57:17 <fourstepper> Me and another collegue from our company feel like jumping to Ubuntu makes us potentially quite vulnerable to a similar thing that just happened to centos, which is why we are now between Alpine and Debian and have decided to try both, with Debian as a fallback - I would consider ourselves to be quite commited in that way
2020-12-09 19:58:43 <tehcloud> my company is likely doing CentOS stream, which is somewhat worrying
2020-12-09 20:05:34 <Hello71> fourstepper: it's pretty unlikely. ubuntu is in a very different position (paid downstream) than centos (paid upstream)
2020-12-09 20:06:57 <fourstepper> I am not talking about completely the same thing happening, but similar + we already dislike some of the modifications that ubuntu is shipping right now, for example netplan
2020-12-09 20:07:22 <fourstepper> I am not saying it's a bad distro, it's just a mutual dislike between both of us :)
2020-12-09 20:08:02 <fourstepper> + for me personally alpine is quite exciting even if it wasn't for the problems with CentOS
2020-12-09 20:08:32 <fourstepper> tehcloud I am looking forward to what is going to happen as far as centos stream goes
2020-12-09 20:09:04 <tehcloud> our company does old, heavy, crufty legacy stuff that is extremely fragile
2020-12-09 20:09:30 <Hello71> imo there's virtually no reason to use ubuntu server
2020-12-09 20:09:50 <Hello71> it's just milking the name
2020-12-09 20:10:19 <fourstepper> Hello71 that's the other thing, debian just seems like a better alternative in all ways
2020-12-09 20:11:04 <fourstepper> tehcloud I really am wondering about the actual stability hit that stream is going to take on and how will some projects approach releasing their software repos
2020-12-09 20:14:39 <fourstepper> Another team at our company has decided to move to Ubuntu (even before the shit hit the fan with centos), mostly because they liked a few features it brings out of the box (like live kernel patching, 2FA on SSH and some other things), though
2020-12-09 20:15:19 <Hello71> knowing ubuntu it's probably insecure
2020-12-09 20:15:30 <fourstepper> :D
2020-12-09 20:15:33 <fourstepper> well
2020-12-09 20:15:39 <tehcloud> my previous company was on Ubuntu, I worked hard to rid them of it and did so successfully
2020-12-09 20:15:59 <Hello71> https://news.ycombinator.com/item?id=25053238
2020-12-09 20:16:22 <fourstepper> yeah i found that funny
2020-12-09 20:16:48 <fourstepper> the other day, the automatic upgrades also upgraded docker for a lot of people which just killed the daemon and brought their systems down
2020-12-09 20:17:01 <fourstepper> as I heard, didn't experience it myself
2020-12-09 20:17:33 <fourstepper> = anyone who had docker from ubuntu repos and auto upgrades + containers running would be down that day, which just seems like a ridiculous thing
2020-12-09 20:17:58 <fourstepper> https://www.reddit.com/r/docker/comments/k4jiyp/woke_up_to_my_docker_daemon_being_stopped_no_logs/?utm_medium=android_app&utm_source=share
2020-12-09 20:41:08 <mps> just reading debian-devel ML (where I'm still subscribed) and read this https://lists.debian.org/debian-devel/2020/12/msg00118.html
2020-12-09 20:42:47 <fourstepper> thanks for sharing, interesting
2020-12-09 20:46:30 <c705> Never expect any kind of reasonable decision from enterprise IT
2020-12-09 20:46:30 <c705> management. You can only earn disappointment.
2020-12-09 20:46:35 <c705> truth ^
2020-12-09 20:49:37 <mps> that sentence is why I posted this here
2020-12-09 20:52:23 <Hello71> this email is completely false!
2020-12-09 20:52:31 <Hello71> as we see today, centos is actually worse than debian -.-
2020-12-10 07:15:23 <batterydude> Will 3.6 update?
2020-12-10 07:19:19 <maxice8> 3.6?
2020-12-10 07:21:34 <maldridge> pretty sure that 3.6 is a fixed decimal constant, I would recommend contacting your vendor for a supply of decimal numbers if you require updates
2020-12-10 09:29:55 <batterydude> maldridge: i am stucked on 4.9 and notebook doesn't support the version greater than it
2020-12-10 09:31:05 <ikke> At least 3.7 still has linux 4.9
2020-12-10 09:31:11 <ikke> but these releases are all out of support
2020-12-10 09:31:33 <ikke> so unless there are specific things you need fixes for, they won't be updated anymore
2020-12-10 12:16:08 <LACampbell> maldridge: I enjoyed that dad joke :D
2020-12-10 13:43:32 <l1x> does anybody use here rpi4 with alpine? I can install it to the mmcblk using vfat + ext4 but when I am trying to swap out the / to the ext4 dhcp starts to fail
2020-12-10 14:11:33 <mps> l1x: did you read 'catb' 'smart questions'
2020-12-10 14:12:58 <mps> http://www.catb.org/~esr/faqs/smart-questions.html
2020-12-10 17:02:35 <l1x> mps: yes thanks.
2020-12-10 17:12:38 <Odin-> So I was trying to download an image and noticed that all the links on https://www.alpinelinux.org/downloads/ gave me 404s.
2020-12-10 17:13:03 <Odin-> The page also says the most recent release was released today, and that _isn't_ on the front page, so ... should I wait a bit? :)
2020-12-10 17:15:32 <govynnus> Hello! I'm planning to convert my desktop to alpine with disk encryption and home on a separate drive. Is LUKS + LVM the best way to go with this, or could I try ZFS? I've read through some of the wiki pages and feel like there are a lot of options. Thanks for any tip :-)
2020-12-10 17:16:12 <govynnus> I downloaded 3.12.1 earlier today so probably is to do with the new release
2020-12-10 17:17:46 <Hello71> Odin-: you can download the previous release manually and upgrade it
2020-12-10 17:17:53 <Hello71> unless you need new kernel
2020-12-10 17:19:40 <Odin-> Yeah, not an issue. I'm just perfectly able to wait for an hour or two ... I'm just messing about.
2020-12-10 17:59:30 <steinbach> evening :)
2020-12-10 18:06:38 <mps> l1x: I didn't understood you question. and yes some people reported that they run fine alpine on RPi4
2020-12-10 18:22:25 <greyxor> Hi everyone :)  i wonder if they are, somewhere, an roadmap, with date or speculative date of next alpine version (23.12.2, 3.13 etc.) ? Thanks
2020-12-10 18:23:05 <mps> greyxor: 3.13 is 'on the horizon'
2020-12-10 18:23:44 <greyxor> mps i'm no specially looking for 3.13, just an roadmap with planned date
2020-12-10 18:24:16 <greyxor> but ok thanks ;)
2020-12-10 18:24:56 <mps> usually at 6 months cycles, december and may
2020-12-10 18:25:15 <minimal> greyxor: next release? 10 minutes after it finishes building ;-)
2020-12-10 18:25:39 <greyxor> minimal 10 minutes to upload payload ? seems long ^^
2020-12-10 18:25:59 <minimal> not exactly joking - 3.13.0 is currently in that situation - it might be released today
2020-12-10 18:26:27 <greyxor> i'm looking for alpine CI then :)
2020-12-10 18:26:32 <greyxor> i will check on gitlab
2020-12-10 18:27:47 <minimal> greyxor: related info but for historical releases: https://wiki.alpinelinux.org/wiki/Releases
2020-12-10 18:27:54 <mps> (obviously different people have different meaning for word 'release')
2020-12-10 18:28:13 <greyxor> mps for me a release is a new tag in git
2020-12-10 18:29:26 <minimal> mps: you mean like Microsoft where their software isn't exactly released, rather it escapes leaving a trail of destruction in its wake? ;-)
2020-12-10 18:29:33 <mps> today was 3.12.2 taged
2020-12-10 18:29:34 <steinbach> release the kracken!
2020-12-10 18:29:56 <steinbach> kracken the release!
2020-12-10 18:30:04 <mps> hehe
2020-12-10 18:51:18 <l1x> mps: thanks, i figured it out myself. the alpine wiki entry is mostly working only a minor step is not clear but it is fine
2020-12-10 18:52:08 <l1x> mps: i think somebody trying to install alpine on rpi4 is still need to do a bit of hacking compare to other distros, i guess because of the non-standard boot procedure
2020-12-10 18:53:08 <halfbit> are there any tools to setup like a cross compile chroot of sorts
2020-12-10 18:53:32 <halfbit> I know of buildroot... and yocto, was sort of hoping to avoid that route
2020-12-10 18:56:14 <minimal> l1x: I use Alpine on RPI3 using a "sys" install. If you need any help let me know
2020-12-10 18:57:17 <l1x> minimal: thanks a lot. Do you know if there is a simpler approach than this : https://wiki.alpinelinux.org/wiki/Raspberry_Pi_4_-_Persistent_system_acting_as_a_NAS_and_Time_Machine#Tinkering_for_Persistence
2020-12-10 19:04:03 <minimal> l1x: have sped-read that page and it doesn't seem too clear - it talks about creating a small boot partition and of downloading the RPI tarfile - I'm guessing they mean you unpack it inside there
2020-12-10 19:04:21 <mps> l1x: I run alpine on different arm32 and and arm64 machines, and yes always have to do manual install, but this is not big problem
2020-12-10 19:05:04 <mps> and we don't yet plan to make ready made images for arm
2020-12-10 19:05:29 <minimal> l1x: the big issue for RPI on Alpine in general is that (some) upgrades don't work right due to packaging of the DTB files
2020-12-10 19:05:43 <mps> minimal: :)
2020-12-10 19:07:34 <minimal> mps: you called? ;-)
2020-12-10 19:08:23 <mps> minimal: I think you know what I wanted to say with ':)'
2020-12-10 19:09:02 <minimal> you wanted me not to go there? lol
2020-12-10 19:09:28 <mps> regarding our discussion about subject
2020-12-10 19:09:35 <mps> few days ago
2020-12-10 19:09:55 <minimal> l1x: I do build my own ready-to-go disk images for RPIs. At some point I should make them available for anyone interested, haven't got around to that yet
2020-12-10 19:10:45 <minimal> mps: linux-edge kernel for aarch64 you mean? haven't tried it yet
2020-12-10 19:12:02 <mps> minimal: I had this idea about year (and some months ago) and we discussed this on devel channel, but no one is interested in that so we didn't do anything about that
2020-12-10 19:12:45 <mps> minimal: no, I meant our discussion about dtb files for RPis
2020-12-10 19:13:00 <minimal> mps: yeah I intended to put in the MR for raspberrypi-bootloader as a first step but keep getting side-tracked with other issues
2020-12-10 19:13:45 <mps> we will not switch to 5.10 as -lts for 3.13 so no need to hurry
2020-12-10 19:16:56 <minimal> mps: to fix the whole RPI situation also involves addressing the /boot config files - they're currently not packaged as all, they're added to the release image by make-chroot-image (or whatever its called) - to deal with config.txt, usercfg.txt, etc would make adding them to some package as templated files and modifying setup-alpine to create/configure them accordingly - so there's a fair bit of work
2020-12-10 19:18:18 <mps> yes, I know it is not simple fix
2020-12-10 19:18:59 <minimal> so its a case of chipping away at it bit by bit. So the MR I've in mind for the bootloader is my intended step 1...
2020-12-10 19:19:00 <mps> and I'm not much interested now when we can use u-boot and make rpi kernel from mainline
2020-12-10 19:20:22 <minimal> well u-boot is still loaded by bootcode.bin and it still needs config.txt as well configured to refer to the u-boot.bin file, right?
2020-12-10 19:20:48 <mps> right
2020-12-10 19:21:04 <minimal> so that still needs to be taken care of even with u-boot - and need any upgrades to the bootcode.bin file to be handled
2020-12-10 19:21:08 <mps> but dtb files are not needed for it
2020-12-10 19:21:50 <mps> and dtbs can be (and they are) loaded by u-boot as on other arms
2020-12-10 19:22:13 <minimal> so then setup-alpine on a RPI in future needs to let you choose u-boot or normal bootloader and it to set up the config.txt file appropriately
2020-12-10 19:22:50 <mps> reminding you on my short notes about this http://arvanta.net/alpine/mailine-rpizw/
2020-12-10 19:23:34 <mps> minimal: yes, setup-alpine should/must be adapted
2020-12-10 19:24:07 <minimal> yupe, I have that stored already for reference. So stage 2 I guess would be to modify setup-alpine and package around the config files in /boot for both u-boot and stock bootloader
2020-12-10 19:24:21 <minimal> then step 3 would be to deal with DTB files
2020-12-10 19:24:30 <minimal> then step 4 would be a large beer or 3 :-)
2020-12-10 19:24:56 <minimal> think I've theoretically mapped out how to resolve this issue lol
2020-12-10 19:25:22 <mps> complicated, yes. and because that I'm for bare bootloader and u-boot setup only. remember alpine moto 'small, _simple_ ...)
2020-12-10 19:26:15 <minimal> yupe, that's why my step 1 includes moving bootcode.bin to a seperate raspberrypi-bootloader-common package so that the normal, cutdown, and experimental subpackage depends on that
2020-12-10 19:26:37 <mps> actually I'm for option that all owners of RPis should put their devices in 'trashcan'
2020-12-10 19:26:48 <minimal> at minute I have to install raspberrypi-bootloader when I'm using raspberrypi-bootloader-cutdown
2020-12-10 19:27:13 <mps> minimal: yes, your proposal sounds good
2020-12-10 19:27:39 <mps> but that is to consider after 3.13 release
2020-12-10 19:28:23 <minimal> mps: most SBCs have limitations - what the RPI has going for it is (a) realtively low cost, (b) easy availability, (c) lots of 3rd HATs and other addons, (d) PoE support
2020-12-10 19:29:04 <minimal> RPI is certainly easier to deal with than SheevaPlugs and PogoPlugs I used in past
2020-12-10 19:29:37 <mps> RPis are bad because of their heart => broadcom
2020-12-10 19:30:36 <minimal> yeah, can you suggest a low-cost alternative with onboard PoE and USB3 (or ideally SATA)?
2020-12-10 19:32:52 <mps> I'm not so much deep in this so can't, but for most my devices I choose allwinner and rockchip SOCs
2020-12-10 19:33:17 <mps> they are more open and linux 'friendly'
2020-12-10 19:37:55 <minimal> allwinner? have things changed? I remember when they first "supported" linux it wasn't really true
2020-12-10 19:39:01 <minimal> they was a guy who convinced them to opensource stuff and I remember him getting frustrated as they didn't really get the whole thing, think they "threw some source over the wall once"  and thought they were done
2020-12-10 19:40:18 <mps> allwinner published all docs they can, that is enough for open source
2020-12-10 19:41:42 <mps> some docs they couldn't publish but that was small part and not because they wanted to hide them but because they obliged by some other manufacturers of components
2020-12-10 19:41:58 <minimal> maybe its changed, its was a few years back I'm thinking of. Thought the sunxi people's mainlining of support was *in spite* of Allwinner, not with their help. Perhaps my recollection is wrong
2020-12-10 19:42:21 <mps> you are right
2020-12-10 19:42:59 <mps> allwinner didn't helped much about linux, but didn't hide things
2020-12-10 19:43:11 <minimal> so isn't that actually worse in a way than RPI/Broadcom as at least theres mainlining is being done with (some) co-operation
2020-12-10 19:43:50 <mps> all we need is docs and specifications
2020-12-10 19:44:20 <mps> think nvidia
2020-12-10 19:44:30 <minimal> I'd rather not, lol
2020-12-10 19:44:46 <minimal> a lot of the nvidia support is via reverse engineering
2020-12-10 19:46:08 <mps> no matter how much some entity help linux if 'it/they' hide something I don't believe them and dislike to use anything from them
2020-12-10 19:47:21 <mps> you read Iliad I hope
2020-12-10 19:52:26 <minimal> I prefer the other Homer ;-)
2020-12-10 19:55:25 <minimal> even AMD hides stuff, e.g. related to the Digital Rights Protection/HDCP
2020-12-10 20:07:48 <l1x> minimal: right
2020-12-10 20:08:17 <l1x> minimal: i would love to work on that, if you have anything to share i would pretty much interested
2020-12-10 20:08:41 <minimal> l1x: you mean the pre-build images?
2020-12-10 20:11:50 <l1x> yes and how to boot that up easily on many rpis
2020-12-10 20:14:16 <minimal> What I have is tailored towards my own uses but I can tweak to make more generic. Basically I have written a *large* Ansible playbook that creates/partitions/formats disk images and installs Alpine into them (via chroot) and then configures them in varying ways and then end result are disk images that can be "dd"ed onto a SD card ready to boot on a RPI.
2020-12-10 20:16:03 <minimal> The disk images are "somewhat" hardened (i.e. SSH config) and they tend to include cloud-init with a separate small partition for its YAML config files. So the first time a machine is booted cloud-init sets up its network config, adds users, and various other things and the machine is ready to go. I do the same for x86_64 physical and virtual machines
2020-12-10 20:16:24 <l1x> wooo
2020-12-10 20:16:31 <l1x> needless to say i am a big ansible user :)
2020-12-10 20:16:41 <l1x> now you got me excited!
2020-12-10 20:16:57 <minimal> I'm currently sorting out LVM, LUKS, and LVM-on-LUKS for the disk images
2020-12-10 20:18:53 <minimal> any I have disk images tailored for servers vs desktops, base machine vs docker machine, v3.11 vs v3.12 vs Edge etc. So a set of distinct YAML files are used with the playbook to build the various images
2020-12-10 20:20:41 <minimal> I do intend to release the playbook at some stage - just need to tidy it up/overhaul it as its been developed bit-by-bit for about 1 year now.
2020-12-10 20:23:15 <l1x> amazing, let me know if you need any help. i am building images for this new microVM project, the use case is much simpler  but i need to automate somehow the image creation and i was thinking about ansibel
2020-12-10 20:27:13 <minimal> l1x: well for quite a bit of the stuff my playbook does I ended up using "command" and "shell" especially for the chroot portions as, at the time, I didn't figure out an alternative method. For partitioning and formatting filesystems etc generally Ansible has native modules for that
2020-12-10 20:29:34 <minimal> l1x: when you say microVM are you referring to Firecracker or something similar?
2020-12-10 20:30:04 <kpcyrd> I'm trying to setup ipv6, I added `iface eth0 inet6 auto` as described in the wiki but this breaks the networking service with `ifup: unknown method "auto"`. the ipv6 module is loaded.
2020-12-10 20:31:06 <ikke> kpcyrd: "iface eth0 inet6 auto" is incorrect
2020-12-10 20:31:27 <ikke> afaik, at least
2020-12-10 20:31:37 <kpcyrd> what should I write instead? :)
2020-12-10 20:32:11 <kpcyrd> I can't find anything else in the networking wiki page except static ipv6 (which I don't want)
2020-12-10 20:33:53 <ikke> kpcyrd: normally auto eth0 should already be enough
2020-12-10 20:34:24 <ikke> linux should automatically do slaac, unless you have forwarding enabled
2020-12-10 20:34:41 <minimal> according to Debian's interfaces manpage that's a valid entry for DHCPv6 SLAAC. Alpine theoretically is Debian-compatible
2020-12-10 20:35:07 <minimal> kpcyrd: what's the full contents of your interfaces file?
2020-12-10 20:35:10 <ikke> kpcyrd: what version of alpine?
2020-12-10 20:36:34 <l1x> minimal: exactly firecracker
2020-12-10 20:37:02 <l1x> minimal: i want to replace docker with fc (and build alpine images, i need python, java and dotnet)
2020-12-10 20:44:07 <minimal> l1x: never got round to looking at Firecracker beyond its initial announcement. As for the minimal emulated devices, yes for my Alpine VMs I'm basically only using the virtio devices.
2020-12-10 20:45:53 <minimal> l1x: got a pointer to a simple "how to use Firecracker" guide? Would like to compare it to QEMU
2020-12-10 20:46:04 <l1x> yes
2020-12-10 20:46:16 <l1x> minimal: you are very lucky because i wrote an article about it
2020-12-10 20:46:24 <l1x> and i also wrote a python tool
2020-12-10 20:46:29 <l1x> i can give you early access
2020-12-10 20:46:36 <l1x> it will be opensourced 1 week from now
2020-12-10 20:46:54 <l1x> if you msg me a github account i am going to add you to the repo and you can check out everything
2020-12-10 20:47:12 <l1x> https://dev.l1x.be/posts/2020/11/22/getting-started-with-firecracker-on-raspberry-pi/
2020-12-10 20:47:41 <minimal> I see their FAQ indicated they how can handle initrds............so wondering how to quickly try it with an existing kernel, initrd, and disk image
2020-12-10 20:47:52 <l1x> it is super easy
2020-12-10 20:48:25 <l1x> i have 3 kernels that i have compiled for the firecracker use case, i can give you the config if you need. it produces super small kernel
2020-12-10 20:48:39 <l1x> we have removed most of the unnecessary modules
2020-12-10 20:48:55 <kpcyrd> minimal, ikke: got it to work, not sure why I thought I had to configure it excplicitly. thanks!
2020-12-10 20:49:29 <l1x> minimal: or you can try existing kernel
2020-12-10 20:50:39 <minimal> l1x: yeah I started creating a local minimal Alpine kernel package for my disk images with a lot of unneeded stuff stripped out
2020-12-10 20:55:32 <l1x> great, we can join forces, it is useful for firecracker too
2020-12-10 21:04:22 <minimal> Nomad supports Firecracker too - cool
2020-12-10 21:07:21 <maldridge> support might be a bit broad for that, its not at feature parity with other drivers
2020-12-10 21:08:19 <minimal> yeah its a community driver
2020-12-10 22:01:52 <fourstepper> Is there a way to allow a user run pm-suspend without editing the sudoers file(for sudo)?
2020-12-10 22:02:32 <fourstepper> I havent really found a suitable group for it and couldnt really find the answer on the net either
2020-12-10 23:01:02 <mps> fourstepper: suspend to disk or to ram?
2020-12-10 23:01:29 <mps> anyway I use acpid for that, LID and PWR button
2020-12-10 23:15:51 <BugHunter1000> Hey I'm checking out vuls.io and it says it's compatible with Alpine 3.3+ but the current download is 3.1 - what do?
2020-12-10 23:16:02 <BugHunter1000> nm I can't read
2020-12-10 23:56:56 <LACampbell> do you need a display manager in linux these days? I used to just run 'startx'
2020-12-10 23:57:36 <BugHunter1000> a lot of distros don't give you startx by default
2020-12-10 23:57:51 <BugHunter1000> https://wiki.alpinelinux.org/wiki/Xfce_Setup
2020-12-10 23:58:47 <LACampbell> yeah that's what I've been going off
2020-12-10 23:59:00 <LACampbell> I am reading my xorg logs, eliminating the problems one by one
2020-12-10 23:59:15 <LACampbell> just making sure this isn't The Wrong THing To Do now
2020-12-11 00:01:44 <LACampbell> "Warning: couldn't open module fbdev". I am not sure why it's trying to do that. I have the xf86-video-intel thing installed. I did setup-xorg-base. hmm
2020-12-11 00:01:47 <BugHunter1000> you did remove the install cd from vbox right
2020-12-11 00:01:54 <BugHunter1000> it doesnt autoeject like debian does
2020-12-11 00:02:08 <LACampbell> yeah that thing is gone
2020-12-11 00:02:28 <LACampbell> it's a regular sys install with a root password when I reboot
2020-12-11 00:03:48 <LACampbell> ok right I am looking at the wrong thing, it's just a warning for a thing I don't need
2020-12-11 00:05:14 <BugHunter1000> What are our controls on what "community" is allowed to pull in?  Are we talking PPA/AUR or does Alpine vet the packages?  This distro is completely worthless for desktops if its main sell is security and it's not vetting the base desktop packages.
2020-12-11 00:06:51 <maxice8> community is not like PPA/AUR
2020-12-11 00:08:10 <BugHunter1000> maxice8, Can you help me understand how it differs?
2020-12-11 00:08:26 <maxice8> community is a repo just like main/ and testing/
2020-12-11 00:09:42 <BugHunter1000> What quality controls exist for these repositories?
2020-12-11 05:50:11 <soft_concrete> GPG keys here https://alpinelinux.org/downloads/ give error 404
2020-12-11 05:50:50 <soft_concrete> When I said key, I believe I meant signature
2020-12-11 05:51:03 <ikke> RIght, checking
2020-12-11 05:52:02 <ikke> apparently the signatures were not uploaded for the last release
2020-12-11 05:55:29 <soft_concrete> Alright... I'll check back tomorrow
2020-12-11 08:03:45 <clandmeter> ikke: yes there is also an ml entry about that issue
2020-12-11 08:40:45 <Marquel> morning. out of curiosity: is there a timeframe when openssl-1.1.1i will appear for Alpine-3.12?
2020-12-11 08:41:36 <fourstepper> Hi, found myself in a bit of a pickle. How should I handle allowing myself to use networkmanager as a non-root user? Is the best way to do this through polkit, or somehow else? Thanks!
2020-12-11 08:58:14 <Cogitri> fourstepper: You could either use polkit or add a sudo rule for it to allow yourself to access nmtui
2020-12-11 08:58:30 <Cogitri> Maybe adding yourself to the plugdev group also does the trick?
2020-12-11 08:59:27 <fourstepper> @Cogitri I am in plugdev I believe, will have to check when I am back to the PC - I will probably go the polkit way, seems cleaner
2020-12-11 09:05:22 <mps> netdev group
2020-12-11 09:22:22 <fourstepper> @mps netdev I have added to my user but hasn't helped as far as I have seen
2020-12-11 09:32:03 <mps> not sure but iirc adding user to netdev was enough when I tested NM, I don't use policykit or any other kit
2020-12-11 09:32:54 <mps> and sure be in netdev is enough for iwctl, iwgtk and connman
2020-12-11 09:34:40 <fourstepper> interesting
2020-12-11 09:34:45 <fourstepper> let me check that again then
2020-12-11 09:34:47 <fourstepper> thanks
2020-12-11 09:46:20 <fourstepper> mps indeed seems to not be enough - when trying to add a new connection, I get "insufficient permissions "
2020-12-11 10:58:23 <ikke> Marquel: It still needs to be backported
2020-12-11 10:59:13 <Marquel> oh? what is the latest release of Alpine then?
2020-12-11 10:59:37 <ikke> Marquel: it's fixed in edge
2020-12-11 10:59:42 <ikke> but not in a stable release yet
2020-12-11 10:59:50 <ikke> fixed / updated
2020-12-11 14:39:15 <kpcyrd> I'm on edge and trying to remove the deprecated ifupdown-ng-openrc package since the files are now provided by openrc, I have the latest version of openrc installed but I'm getting this error when trying to uninstall ifupdown-ng-openrc: https://paste.debian.net/plainh/72c91016
2020-12-11 14:39:49 <kpcyrd> do I need to remove this with one of the --force-* flags?
2020-12-11 14:40:10 <maxice8> The apkbuild has `install_if="$pkgname openrc"`
2020-12-11 14:40:25 <maxice8> so it will get automatically installed whenever you have openrc and ifupdown-ng
2020-12-11 14:45:26 <kpcyrd> maxice8: the package is orphaned though
2020-12-11 14:45:34 <ikke> But how to get rid of it now it has been removed from the repos
2020-12-11 15:02:13 <kpcyrd> huh, I've tried --force-broken-world and -f but it still won't remove it
2020-12-11 15:36:26 <Hello71> apk upgrade
2020-12-11 17:06:53 <kpcyrd> Hello71: does nothing
2020-12-11 17:09:39 <MY-R> kpcyrd: what about: apk -i -a upgrade
2020-12-11 17:10:37 <MY-R> probably nothing too but on edge better always use -a option
2020-12-11 17:14:47 <kpcyrd> MY-R: that worked, thanks!
2020-12-11 17:15:43 <MY-R> kpcyrd: another reason to always use -a when upgrade :)
2020-12-11 17:17:56 <kpcyrd> MY-R: it also "upgraded" a bunch of packages, but I didn't understand why: `(5/25) Replacing libc-utils (0.7.2-r3 -> 0.7.2-r3)`
2020-12-11 17:21:35 <MY-R> kpcyrd: ye that is when using edge, sometimes something changing in dependencies of some package and that block others from upgrade
2020-12-11 17:24:13 <MY-R> kpcyrd: you could notice that by doing "apk upgrade" and then run "apk version" and you would still see some packages arent upgraded so then using apk -a upgrade and should do it (but not always if something block those packages, then need to start delete/reinstall their dependencies)
2020-12-11 17:27:28 <MY-R> kpcyrd: but that libc-utils upgrade looks like you didnt use -a when was switching from stable to edge
2020-12-11 17:28:37 <MY-R> kpcyrd: https://wiki.alpinelinux.org/wiki/Upgrading_Alpine#Upgrading_to_Edge  -a = --available
2020-12-11 17:29:28 <kpcyrd> cool, thanks!
2020-12-11 17:29:35 <MY-R> enjoy!
2020-12-11 17:33:09 <EF> Fishie02
2020-12-11 17:33:43 <EF> Sorry, computer gone mad
2020-12-11 17:34:44 <ikke> Time to change a password?
2020-12-11 17:35:55 <EF> It's changed already
2020-12-11 17:35:58 <ikke> heh
2020-12-11 18:53:47 <mtnt> hi there! Not sure it's a right place to ask, but i'll try. "modeset(0): disabling kernel dirty updates ... not required" in Xorg.log Mice won't move inside dwm. What may be the cause? It is latest Alpine x86
2020-12-11 20:00:02 <mtnt> sorry, gtg. thanks!
2020-12-11 20:36:44 <omni> ooh, firehol in testing, interesting
2020-12-11 20:37:39 <omni> a 12314 line long bash script, but anything is more tolerable than raw iptables
2020-12-11 20:38:03 <omni> and only 22 (potential) bashisms...
2020-12-11 20:40:04 <Hello71> nft > iptables > firehol
2020-12-11 20:41:14 <mps> does busybox ash support .inputrc, or something similar
2020-12-11 20:44:24 <omni> Hello71: yeah, perhaps I should learn nftables
2020-12-11 20:45:09 <omni> I've mostly handled iptables through ferm the past few years
2020-12-12 00:16:21 <yxprogramming> Hello everyone! I'm trying out alpine on a pi 4 (4GB if that's relevant), and was having some issues getting x to work. The error is "No screens found" which I assume has something to do with a graphics driver for the rpi. However, I don't really know how to fix it, I'm not familiar with alpine. Can anyone give me some advice?
2020-12-12 00:23:35 <yxprogramming> I seem to have made progress now, it turns out that I was indeed missing a log line specifying the package I needed to install for the driver. Now it's just a matter of hunting down xterm in the config files ;) Sorry I asked unnecissarily
2020-12-12 01:09:16 <omni> I was happy to notice the addition of man pages to apk-tools (I'm at edge until v3.13.x)
2020-12-12 01:46:42 <omni> yeah,gotta dig into nft https://bugzilla.netfilter.org/show_bug.cgi?id=1434
2020-12-12 06:12:04 <lazywalker> hi, is there python3.9 for alpine?
2020-12-12 06:22:52 <maxice8> No
2020-12-12 06:23:05 <maxice8> Most likely there won't be until 3.13 is released
2020-12-12 06:28:20 <maxice8> Same with Lua 5.4
2020-12-12 09:36:52 <maxice8> https://lists.alpinelinux.org/~alpine/devel/%3C86c7a81e-1640-7f82-9e13-dfdbe1aad07b%40gmail.com%3E
2020-12-12 11:30:23 <LACampbell> https://pkgs.alpinelinux.org/package/edge/testing/x86_64/xf86-video-vboxvideo is there any way to get native resolution in virtualbox without using this edge package?
2020-12-12 11:30:43 <LACampbell> xrandr doesn't have my resolution
2020-12-12 11:46:57 <ikke> LACampbell: Isn't vbox using vmware drivers nowadays? xf86-video-vmware
2020-12-12 11:47:02 <ikke> https://pkgs.alpinelinux.org/package/edge/community/x86_64/xf86-video-vmware
2020-12-12 11:47:38 <ikke> If you select vmsvga as video device
2020-12-12 11:50:22 <LACampbell> thanks for the tip ikke I think you maybe right
2020-12-12 13:59:21 <Hello71> should be simply remove nomodeset
2020-12-12 15:28:58 <ipcloud> .
2020-12-12 16:20:06 <Apachez> it got both vmsvga and its own
2020-12-12 16:20:28 <Apachez> vboxvga, vmsga, vboxsvga, none
2020-12-12 16:20:45 <Apachez> I usa vboxsvga with linux as host and windows as guest
2020-12-12 16:21:16 <Apachez> and vmsvga where the guest is *nix
2020-12-12 17:30:23 <Hello71> linux has drm driver for vbox since 5.2
2020-12-12 19:35:06 <nicolascccc> Guys, is xedit available as a package on Alpine?
2020-12-12 19:35:15 <nicolascccc> I can't seem to find any package that contains it
2020-12-12 20:00:01 <maxice8> is xedit a binary (executable installed to /bin|/usr/bin) ?
2020-12-12 20:00:42 <nicolascccc> Yes
2020-12-12 20:00:51 <nicolascccc> It's a text editor
2020-12-12 20:01:30 <nicolascccc> On Debian it's inside xorg-extras
2020-12-12 20:03:20 <maxice8> As far as I can see, no.
2020-12-12 20:03:57 <nicolascccc> Yeah, I thought so. A shame, because it's the lightest text editor with a GUI
2020-12-12 20:04:06 <nicolascccc> Leafpad is okay, but not that lightweight
2020-12-12 20:04:24 <nicolascccc> I might build it myself, but Leafpad is good enough
2020-12-12 21:40:34 <Hello71> so apparently the location where default rc services are set is mkinitfs of all places
2020-12-12 22:24:19 <systemdlete> I made the apparently blundering choice of having /boot on the same fs with /.  I used the following instructions to install alpine:  1.  mount /dev/sda2 /target; 2. mkdir /target/boot; 3.  setup-alpine -f answers (where answers can be seen at http://paste.debian.net/1176640/)
2020-12-12 22:25:04 <systemdlete> the install process goes exceedingly well.  No errors, except for my (bad) lbu config, which the wiki claims should not be called for "sys" installs.
2020-12-12 22:25:34 <systemdlete> But when I boot, I get a crash, saying no filesystem could mount root filesystem.
2020-12-12 22:26:05 <systemdlete> I'm using a grub, multiboot gpt hard disk.  So I'm certain I'm doing something incorrectly/incompletely.
2020-12-12 22:26:48 <systemdlete> I re-ran grub-install from a devuan partition after the alpine install, just to make sure it would be up-to-date with the new installed partition.
2020-12-12 22:28:06 <fourstepper> how does your partitioning layout look like now? systemdlete
2020-12-12 22:28:10 <systemdlete> (there is at least one page on the wiki talking about adjusting the paths in extlinux.conf, but I'm still not sure I have that right)
2020-12-12 22:28:28 <systemdlete> you mean my gpt?
2020-12-12 22:28:42 <fourstepper> ah it's all on one partition
2020-12-12 22:28:46 <systemdlete> yes.
2020-12-12 22:28:52 <systemdlete> that was the bug-a-boo.
2020-12-12 22:29:01 <fourstepper> damn, haven't seen that ever. I am not even sure that's something valid to do
2020-12-12 22:29:08 <systemdlete> I'm pretty sure that if I had opted for separate /boot this would likely not be an issue.
2020-12-12 22:30:16 <fourstepper> i am like 99% sure that the /boot is needed on a seperate partition, but I am not 100% sure as I have never questioned this myself
2020-12-12 22:30:18 <systemdlete> see https://wiki.alpinelinux.org/wiki/Installing_Alpine_on_HDD_dualbooting
2020-12-12 22:30:46 <systemdlete> (Yeah, I seem to recall this issue from an attempt to install alpine a year or two ago)
2020-12-12 22:31:17 <systemdlete> fourstepper:  About 1/2-way down that page, there is an admonition about /boot on the root partition.
2020-12-12 22:31:56 <systemdlete> I'm wondering if this is a grub-syslinux interaction problem?
2020-12-12 22:32:13 <systemdlete> grub <-> syslinux I mean
2020-12-12 22:32:32 <fourstepper> I stand corrected, I am unsure then. To understand better - one disk, two OS, devuangrub takes care of getting the system booted
2020-12-12 22:32:38 <fourstepper> devuan's*
2020-12-12 22:33:02 <systemdlete> more than 2 OS's, actually.  There's several.  This is a test box.
2020-12-12 22:33:07 <systemdlete> but no matter.
2020-12-12 22:33:30 <systemdlete> yeah, devuan grub-install/grub-mkconfig does all the grub stuff.
2020-12-12 22:33:47 <systemdlete> I think this is what they call chaining.
2020-12-12 22:33:51 <systemdlete> but I am not really sure.
2020-12-12 22:34:34 <systemdlete> I'd have to put on my grub-thinking hat for this... I've gotten into the guts of grub2 a few times now.  It comes back to me, slowly, but it is still a steep curve each time.
2020-12-12 22:35:31 <systemdlete> fourstepper:  I am not even sure who is issuing the error message.  Grub or syslinux.  (Do you happen to recognize that error?)
2020-12-12 22:35:41 <systemdlete> I think it is coming from syslinux, but I am not entirely certain.
2020-12-12 22:36:20 <systemdlete> chainloading.  That's it.
2020-12-12 22:37:54 <fourstepper> hmm....
2020-12-12 22:37:58 <systemdlete> OK, Maybe the problem is that, for whatever reason,
2020-12-12 22:37:59 <fourstepper> I honestly am unsure
2020-12-12 22:38:30 <tehcloud> systemdlete, nice name
2020-12-12 22:38:31 <systemdlete> grub-mkconfig did not generate an initrd for the partition.  It generates the kernel image line (linux) though.
2020-12-12 22:38:38 <fourstepper> I don't recorgnize the error. is it possible that the GUID if maybe wrong?
2020-12-12 22:38:51 <fourstepper> tehcloud was about to say the same when I saw it :D
2020-12-12 22:39:20 <systemdlete> No, I checked that.  grub-mkconfig found the correct guid for the partition.
2020-12-12 22:39:30 <systemdlete> Let me try adding the initrd and see what that does.
2020-12-12 22:39:32 <systemdlete> bbs
2020-12-12 22:39:52 <fourstepper> good luck
2020-12-12 22:44:30 <systemdlete> yep.  THat was it.
2020-12-12 22:44:37 <systemdlete> thanks for your feedback, though.
2020-12-12 22:45:01 <systemdlete> It looks like I was NOT chainloading; I was actually driving the boot from grub alone.
2020-12-12 22:45:25 <systemdlete> However, it seems to be hanging after it discovers my usb kvm switch... grrrr.
2020-12-12 22:48:54 <fourstepper> do you have the udev service running on alpine?
2020-12-12 22:49:05 <fourstepper> actually
2020-12-12 22:49:17 <fourstepper> disregard that, I am not sure that could be related
2020-12-12 23:13:49 <systemdlete> whoa... I finally got a boot prompt!
2020-12-12 23:13:55 <systemdlete> I didn't even have to log in?
2020-12-12 23:14:52 <fourstepper> idk, empty root pw?
2020-12-12 23:15:03 <systemdlete> nope.  I'm in a rescue shell
2020-12-12 23:15:13 <systemdlete> maintenance shell, whatever
2020-12-12 23:15:20 <Hello71> are you definitely booted into alpine
2020-12-12 23:15:34 <systemdlete> df shows nothing besides the device fs and tmpfs
2020-12-12 23:15:34 <Hello71> well, unbooted but does it say alpine
2020-12-12 23:15:41 <systemdlete> hmmm...
2020-12-12 23:15:59 <Hello71> what is the output of uname -a
2020-12-12 23:16:08 <systemdlete> well, uname -a shows it is alpine
2020-12-12 23:16:14 <systemdlete> but hostname is not set
2020-12-12 23:16:22 <Hello71> then your kernel command line is wrong
2020-12-12 23:16:27 <systemdlete> aha
2020-12-12 23:16:44 <systemdlete> I should be able to get that from the extlinux.conf?
2020-12-12 23:17:00 <systemdlete> I can just copy-paste that to the grub cfg, right?
2020-12-12 23:17:01 <Hello71> or your initramfs doesn't have the right modules
2020-12-12 23:17:18 <Hello71> i doubt default grub.d will chainload syslinux
2020-12-12 23:17:21 <systemdlete> Hello71:  Keep in mind, I followed mainly the defaults for install.
2020-12-12 23:18:23 <systemdlete> You can see my answer file to alpine-setup at http://paste.debian.net/1176640/
2020-12-12 23:18:55 <systemdlete> Other than running alpine-setup -f answers, I did not change the installation time env
2020-12-12 23:19:02 <Hello71> also grub-mkconfig from devuan won't pick up alpine boot params
2020-12-12 23:19:05 <systemdlete> (just mounted the target file system
2020-12-12 23:19:07 <systemdlete> )
2020-12-12 23:19:40 <systemdlete> so... can I just copy-paste them from the extlinux.conf?
2020-12-12 23:20:10 <Hello71> with the modules and GRUB_CMDLINE_LINUX_DEFAULT
2020-12-12 23:20:14 <Hello71> https://git.alpinelinux.org/aports/tree/main/grub/alpine-mkconfig.patch
2020-12-12 23:20:24 <systemdlete> ah, thank you!
2020-12-12 23:20:32 <Hello71> but if you edit grub.cfg it'll be overwritten on grub update
2020-12-12 23:20:59 <systemdlete> you know what?  I think I may just resort to reinstalling with separate /boot drive.
2020-12-12 23:21:43 <systemdlete> It looks like I have a rather unusual configuration.  setup-alpine expects separate /boot and does not tolerate other options well I guess.
2020-12-12 23:22:25 <systemdlete> If one chooses to do it differently, as I have, then it might not be well-supported by default.
2020-12-12 23:22:56 <systemdlete> As it is, I am using grub as my boot manager, so that is probably enough right there to spark a revolt.
2020-12-12 23:24:53 <systemdlete> Hello71:  The only thing I am concerned about is whether the patch is against a compatible version of grub.d
2020-12-12 23:25:18 <systemdlete> Will I need the patch even if I set up separate /boot?
2020-12-12 23:25:33 <systemdlete> Or should it work OK even with grub so long as /boot is separate?
2020-12-12 23:26:07 <Hello71> your issue is not related to separate /boot
2020-12-12 23:26:36 <systemdlete> so, at this point, it is just the boot params (and maybe initramfs modules)
2020-12-12 23:26:39 <Hello71> your problem is you are not properly combining grub across multiple distros
2020-12-12 23:28:28 <systemdlete> so the short answer is, yeah, I will need the patch either way
2020-12-12 23:29:14 <systemdlete> is there a grub option for alpine-setup?   (I should have asked this a while back)
2020-12-12 23:29:26 <systemdlete> or is the patch that solution?
2020-12-12 23:30:23 <Hello71> https://git.alpinelinux.org/alpine-conf/tree/setup-disk.in
2020-12-12 23:37:12 <mps> docs.alpinelinux.org
2020-12-12 23:37:15 <systemdlete> Well, looks like I coudl try setting USE_EFI, but this is not a true EFI gpt.  It is only a grub-bios configuration
2020-12-12 23:38:05 <Hello71> it would seem that the intention is to export BOOTLOADER=grub
2020-12-12 23:38:34 <systemdlete> hmmm.  I saw that also, but I wasn't sure yet if that might be an internal var
2020-12-12 23:38:43 <systemdlete> but it looks like that could do it
2020-12-12 23:39:44 <systemdlete> see, the help message for setup-disk says "if BOOTLOADER is specified..." which made me think that was supposed to be an option specified to the program.
2020-12-12 23:39:49 <systemdlete> But,yeah.
2020-12-12 23:40:04 <systemdlete> I guess I could export that right into the env of setup-alpine
2020-12-12 23:40:47 <systemdlete> Usually, the terminology would be something like "if BOOTLOADER is set..."
2020-12-12 23:40:55 <systemdlete> so that threw me.
2020-12-12 23:41:13 <systemdlete> OK, now that I have looked at the source (thanks for pointing this out), I see it should work.
2020-12-12 23:41:41 <systemdlete> I would think I could even set BOOTLOADER=grub right there in the answers file?
2020-12-12 23:42:09 <mps> https://docs.alpinelinux.org/user-handbook/0.1a/Installing/manual.html
2020-12-12 23:42:49 <systemdlete> mps:  Perfect.  Thanks!
2020-12-12 23:43:38 <systemdlete> thanks to you Hello71, fourstepper, mps.
2020-12-12 23:43:56 <mps> systemdlete: np :)
2020-12-13 00:56:09 <LACampbell> is there any way to get the setup scripts from apk?
2020-12-13 00:57:36 <LACampbell> hmm maybe it's 'setup-box'
2020-12-13 00:58:57 <LACampbell> no that wasn't it
2020-12-13 01:02:28 <LACampbell> ah, alpine-conf. which I could get by doing `apk search alpine-setup`
2020-12-13 02:00:45 <systemdlete> Well, that worked.  Thanks again.  The only issue left is why grub-mkconfig is not generating entries for the other OS's on the disk.
2020-12-13 02:00:53 <systemdlete> I see grub-probe is in the path.
2020-12-13 02:14:03 <systemdlete> I could cross-mount devuan and steal its grub.cfg (well, most of it anyway).  But if there is a "clean" way to get the other OS's, I'd prefer it.   I'm looking through all of the grub docs and I'm not seeing what I need.
2020-12-13 02:22:41 <systemdlete> gpt
2020-12-13 02:23:04 <systemdlete> I've read that some people have problems with os-prober on gpt disks.  So that's a possibility.
2020-12-13 02:26:26 <systemdlete> Then again... devuan can find the other partitions and their OS's.  And I'd think alpine had at least the same source version or greater for grub.  So...
2020-12-13 02:26:28 <systemdlete> ?
2020-12-13 02:37:00 <systemdlete> I see.  os-prober is not in the repos (but it *is* in edge)
2020-12-13 02:37:14 <systemdlete> so for now, I'll just steal it from devuan
2020-12-13 02:49:26 <ipcloud> d
2020-12-13 02:55:49 <Hello71> really the solution is to throw out grub-mkconfig
2020-12-13 02:56:09 <Hello71> it's good for ubuntu and nobody else
2020-12-13 03:00:37 <Hello71> this alpine patch is really broken anyways, because it assumes all kernels should take modules parameter. otoh GRUB_CMDLINE_LINUX_DEFAULT is broken in the same way
2020-12-13 03:01:15 <Hello71> the whole grub-mkconfig idea is broken. XBOOTLDR is better, or hell, even standard ESP
2020-12-13 03:40:34 <LACampbell> Hello71: are you dual booting? I used to but I am too dumb for that in this windows 10 UEFI error
2020-12-13 03:40:43 <LACampbell> *era. maybe a freudian slip
2020-12-13 03:44:31 <Hello71> i used to have windows and linux share mbr booting. in fact i once had ntldr chainloading grub
2020-12-13 04:08:56 <omni> I'm not sure the first comment here is valid, it sound as if alpine was made with docker in mind rather than happened to be suitable for container images, nor relevant https://wiki.alpinelinux.org/wiki/Alpine_Linux:FAQ#General
2020-12-13 04:18:29 <trwww> I'm installing alpine in kvm via https://blog.ruanbekker.com/blog/2018/02/20/setup-a-kvm-hypervisor-on-ubuntu-to-host-virtual-machines/
2020-12-13 04:18:29 <trwww> I've started setup-alpine and am at:
2020-12-13 04:18:36 <trwww> Which one do you want to initialize? (or '?' or 'done') [eth0] ?
2020-12-13 04:18:55 <trwww> but the command from that blog has:
2020-12-13 04:18:57 <trwww> --network bridge:virbr0,model=virtio \
2020-12-13 04:19:35 <trwww> could someone suggest how I get bridged networking? I'm assuming if I choose 'eth0' I won't get bridged networking?
2020-12-13 04:20:09 <trwww> `ip addr` shows the virbr0 interface
2020-12-13 04:22:24 <tomleb> Hi, I am having issues with installing wireguard on alpine diskless mode due to /lib/modules being read-only. Is there a fix to this? Or should I not run diskless (on rpi4)
2020-12-13 04:43:35 <trwww> nevermind, I guess eth0 is the guest interface. defaults work as expected.
2020-12-13 05:17:11 <dngray> ah fcolista looks cool https://www.youtube.com/watch?v=C6VU8qRzp40
2020-12-13 05:41:11 <LACampbell> I have alpine running on a usb, but when I try and startx it loads 'glamoregl' then crashes. which is for a radeon card, I have an nvidia, so no idea why it's loading (lspci output is correct). am I best off writing a custom Xorg.conf to disable it?
2020-12-13 05:46:05 <MY-R> LACampbell: be sure you have installed "mesa-dri-nouveau" and "xf86-video-nouveau" packages
2020-12-13 05:48:59 <LACampbell> MY-R: ah I've misread the situation. I've got an on board intel so just downloaded that, will install those two and see how I go thanks
2020-12-13 05:49:14 <LACampbell> I guess if it shows up in lspci I need a driver for it
2020-12-13 05:49:25 <MY-R> :)
2020-12-13 05:59:18 <LACampbell> MY-R: that worked, thanks so much. got wireless keyboard, mouse and window manager working on actual hardware. it's a start
2020-12-13 05:59:55 <MY-R> yeah, gj! :)
2020-12-13 12:13:22 <omni> tomleb: you want the wireguard-rpi4 kernel, right? https://pkgs.alpinelinux.org/packages?name=wireguard-rpi4&branch=v3.12
2020-12-13 12:17:24 <jbgoldberg1> Hey, newbie here! I just started to play with Alpine. Very sweet stuff.
2020-12-13 12:18:59 <jbgoldberg1> But there is some surprisingly difficult things to figure out... I having some hard time with basic stuff.
2020-12-13 12:20:01 <mps> climbing Alp(s)ine require training before and effort when start :)
2020-12-13 12:21:01 <ius> tomleb: interestingly the modloop init script mounts /lib/modules rw only if the overlay module is loaded, but doesn't attempt to do so by itself
2020-12-13 12:21:17 <jbgoldberg1> For example: until now I was unable to mount my pendrive (256G) on the system. I tried some different aproaches. NTFS, ExtFAT, formatted in another machine, in the alpine.... Never mounts: invalid parameter... hard stuff.
2020-12-13 12:21:35 <ius> tomleb: if you specify modules=overlay on the kernel command line it'll get probed early on boot, and thus modloop will use overlayfs
2020-12-13 12:22:15 <ius> not sure if i fail to understand the rationale, or that modloop should just try to probe overlay..
2020-12-13 12:22:19 <jbgoldberg1> mps: Yes. But I am in love with the SO. I spend A LOT of time searching for something really minimalist in Linux domains...
2020-12-13 12:22:51 <mps> jbgoldberg1: you probably need to install mount, apk add mount
2020-12-13 12:23:09 <ius> jbgoldberg1: did you specify the filesystem type when mounting?
2020-12-13 12:23:21 <mps> busybox mount is minimalistic
2020-12-13 12:23:34 <ius> ^ because of that, yes
2020-12-13 12:24:02 <jbgoldberg1> mps: I assumed that (at least) the mount was available! kkkkkk
2020-12-13 12:24:21 <mps> it is, but bare one
2020-12-13 12:24:24 <jbgoldberg1> Actuall it tries to mount....
2020-12-13 12:24:34 <cads> jbgoldberg1, what options did you pass to mount?
2020-12-13 12:24:35 <jbgoldberg1> I see...
2020-12-13 12:25:38 <mps> busybox applets are enough in most cases, for some other things and features proper packages have to be installed
2020-12-13 12:26:08 <cads> busybox's built in mount takes the -t FSTYPE parameter like I'm used to - I think it should work
2020-12-13 12:26:28 <jbgoldberg1> cads: mount -t ntfs /dev/sda1 /mnt
2020-12-13 12:26:33 <mps> cads: if driver is loaded, yes
2020-12-13 12:27:03 <cads> jbgoldberg1, blow a -v across that line and lets see what it really thinks?
2020-12-13 12:27:38 <jbgoldberg> ERROR: unsatisfiable constraints:
2020-12-13 12:27:38 <jbgoldberg> mps: Doesn't worked: caronte:/mnt# apk add mount
2020-12-13 12:27:38 <jbgoldberg>     required by: world[mount]
2020-12-13 12:27:38 <jbgoldberg>   mount (missing):
2020-12-13 12:28:17 <maxice8> cmd:mount
2020-12-13 12:28:46 <cads> jbgoldberg, hope that was not for me - I'm asking for verbose output from the mount command that fails: mount -v -t ntfs /dev/sda1 /mnt
2020-12-13 12:29:04 <mps> apk add util-linux
2020-12-13 12:29:14 <mps> maxice8: thanks
2020-12-13 12:29:38 <artok> best in apk are cmd: and lib:
2020-12-13 12:30:34 <jbgoldberg> Now we are talking:  :P
2020-12-13 12:30:35 <jbgoldberg> caronte:/mnt# apk add cmd:mount
2020-12-13 12:30:35 <jbgoldberg> (2/9) Installing libcap-ng (0.7.10-r1)
2020-12-13 12:30:35 <jbgoldberg> (1/9) Installing blkid (2.35.2-r0)
2020-12-13 12:30:35 <jbgoldberg> (3/9) Installing setpriv (2.35.2-r0)
2020-12-13 12:30:35 <jbgoldberg> (4/9) Installing findmnt (2.35.2-r0)
2020-12-13 12:30:36 <jbgoldberg> (5/9) Installing mcookie (2.35.2-r0)
2020-12-13 12:30:36 <jbgoldberg> (6/9) Installing hexdump (2.35.2-r0)
2020-12-13 12:30:37 <jbgoldberg> (7/9) Installing lsblk (2.35.2-r0)
2020-12-13 12:30:37 <jbgoldberg> (8/9) Installing sfdisk (2.35.2-r0)
2020-12-13 12:30:38 <jbgoldberg> (9/9) Installing util-linux (2.35.2-r0)
2020-12-13 12:30:38 <jbgoldberg> Executing busybox-1.31.1-r19.trigger
2020-12-13 12:30:39 <jbgoldberg> OK: 356 MiB in 116 packages
2020-12-13 12:30:46 <artok> let's not talk that way
2020-12-13 12:30:50 <mps> stop paste
2020-12-13 12:31:41 <mps> jbgoldberg: use some paste service when posting output of cmds
2020-12-13 12:32:58 <cads> where does cmd:mount live at https://pkgs.alpinelinux.org/package/ ?
2020-12-13 12:33:57 <jbgoldberg> Thanks, guys! Now it worked perfectly!
2020-12-13 12:34:10 <mps> cads: apk search cmd:mount
2020-12-13 12:34:11 <jbgoldberg> I really appreciate your help.
2020-12-13 12:34:18 <maxice8> cads: it is a provides
2020-12-13 12:35:56 <jbgoldberg> My suggestion is to create a "trail" in documentation site to show this particularities from the SO. They are very different from other Linuxes and make newbies like me spend A LOT of time chasing the own tail.
2020-12-13 12:36:09 <maxice8> artok: I think you mean so: instead of lib:, there is also pc:
2020-12-13 12:36:18 <maxice8> Ans hopefully in the future we get py:
2020-12-13 12:37:20 <mps> jbgoldberg: high bar for newbies is sometimes good to have :)
2020-12-13 12:37:26 <artok> oh, true, so:
2020-12-13 12:37:29 <jbgoldberg> Thinks like lbu, this so minimalistic aproach that have no full blown mount are AWESOME. At same time, this is hard to know where to find "the right name for the mount package".... :P
2020-12-13 12:38:02 <artok> usually have it on dockerfile, my template has it already so haven't really written it many times =)
2020-12-13 12:38:13 <mps> alpine is 'small' and also in documentation
2020-12-13 12:38:29 <ikke> jbgoldberg: https://pkgs.alpinelinux.org/contents?file=mount&path=&name=&branch=edge&arch=x86_64
2020-12-13 12:38:37 <jbgoldberg> mps: Hahahahaha Make sense. But let take in consideration that "does not know the right package name for the package" isn't exactly "high bar".... hahaha
2020-12-13 12:39:23 <jbgoldberg> ikke: Thanks.
2020-12-13 12:40:29 <mps> jbgoldberg: alpine is different in basic concepts from most other distros so new users have to learn new approach
2020-12-13 12:40:43 <john17> hello, how are you guys?
2020-12-13 12:41:02 <mps> if it is like debian or centos why anyone would switch to alpine
2020-12-13 12:41:07 <jbgoldberg> mps: Yes. That is clear to me, now.
2020-12-13 12:41:56 <jbgoldberg> mps: No, PLEASE. NEVER change the aproach of the SO. This is the reason I loved it in the first place.
2020-12-13 12:42:21 <mps> jbgoldberg: 'SO' ?
2020-12-13 12:42:31 <john17> hey i tried to put alpine on my usb drive (dd and balenaetcher) but this does not work, can anyone recommend me any good arch linux tools for putting iso images on usbs
2020-12-13 12:42:50 <john17> well for linux
2020-12-13 12:42:51 <jbgoldberg> 'OS' (Sorry, translation issues)
2020-12-13 12:43:15 <mps> john17: dd always works for me
2020-12-13 12:43:23 <artok> john17: target platform is ?
2020-12-13 12:43:40 <john17> im running arch linux, trying to install alpine linux
2020-12-13 12:44:37 <ikke> yeah, dd should just work
2020-12-13 12:44:59 <john17> maybe i did the wrong way of using the command
2020-12-13 12:45:18 <jbgoldberg> Once more: thanks a lot, people!
2020-12-13 12:50:32 <cads> mps, so cmd:mount is provided by many packages, include ceph, fuse, cifs-utils, sshfs, and more. I'm missing something about the nature of a provides. would installing cmd:mount install the mount drivers from all 16 packages that come up when I do apk search cmd:mount?
2020-12-13 12:51:30 <cads> 
ACTION has cornered themselves into reading the manual

2020-12-13 12:52:57 <ikke> if there are multiple providers without any priority, then you need to explicitly install one
2020-12-13 12:54:14 <cads> can we tell which one got installed in jbgoldberg's terminal logs from five minutes ago?
2020-12-13 12:55:20 <ikke> not by the output, but I see it installed util-linux
2020-12-13 12:55:38 <cads> it looks like it defaulted to util-linux in some way that apk search cmd:mount does not hint at
2020-12-13 12:55:56 <ikke> right, it does not show that kind of info
2020-12-13 12:56:07 <ikke> it just lists packages which provide cmd:mount
2020-12-13 12:56:36 <cads> but now who wins the popularity contest by default, as it were
2020-12-13 12:57:17 <ikke> Good question, to which I don't have an answer atm
2020-12-13 12:58:19 <cads> doing apk info cmd:mount gives us util-linux's description
2020-12-13 12:58:37 <cads> so maybe that's a reliable indicator of what resolves by default
2020-12-13 12:59:10 <cads> lets see if the wiki page discusses such subtleties https://wiki.alpinelinux.org/wiki/Alpine_Linux_package_management
2020-12-13 12:59:58 <ikke> I think search is not exact
2020-12-13 13:00:08 <cads> I would think apk policy would have something to say about cmd:mount, but surprisingly not
2020-12-13 13:00:30 <ikke> fuse3-3.10.1-r0 provides:
2020-12-13 13:00:33 <ikke> cmd:fusermount3
2020-12-13 13:01:00 <ikke> so apk search cmd:mount returns fuse3, but it procides cmd:fustermount3
2020-12-13 13:01:18 <ikke> minus all the typos :)
2020-12-13 13:04:08 <ikke> ubase does provide 'cmd:mount', so not sure why that one is not picked
2020-12-13 13:05:14 <cads> fuse3's apkbuild says nothing about the cmd:mount https://git.alpinelinux.org/aports/tree/main/fuse3/APKBUILD
2020-12-13 13:05:33 <mps> cads: cmd: so: are just hints, not 'real things'
2020-12-13 13:05:44 <cads> but he package contents contain a /usr/sbin/mount.fuse3 https://pkgs.alpinelinux.org/contents?branch=edge&name=fuse3&arch=x86_64&repo=main
2020-12-13 13:06:04 <ikke> these provides are automatically determined
2020-12-13 13:08:39 <cads> ikke, I see, so you're wondering why some other alternative isn't chosen, while I'm still trying to figure out how the package infrastructure decides that a package provides cmd:mount
2020-12-13 13:10:35 <mps> by crystal ball, called 'scanelf'. in util-linux mount is linked with libmount
2020-12-13 13:10:43 <ikke> cads: https://gitlab.alpinelinux.org/alpine/abuild/-/blob/master/abuild.in#L1189
2020-12-13 13:11:39 <ikke> cads: it just checks for executable files in the standard *bin directories
2020-12-13 13:11:57 <mps> ikke: ah, thanks
2020-12-13 13:12:03 <cads> very cool, thanks for the code spelunking!
2020-12-13 13:12:10 <mps> I thought it scans libs
2020-12-13 13:12:25 <ikke> that's for so: provides
2020-12-13 13:12:33 <mps> yes
2020-12-13 13:12:38 <ikke> or more, dependencies
2020-12-13 13:13:15 <mps> never looked at prepare_command_provides()
2020-12-13 13:13:47 <ikke> me neither :p
2020-12-13 13:14:09 <ikke> these provides can be easily found by checking what files are in the package
2020-12-13 13:16:02 <cads> ikke, are these kinds of meta packages mentioned in the wiki currently? looking at  https://wiki.alpinelinux.org/wiki/Alpine_Linux_package_management I don't see it, but I don't think that's the entire apk manual
2020-12-13 13:16:47 <ikke> I don't think so
2020-12-13 13:17:06 <systemdlete> Hello71:  Actually, the problem is that os-prober is missing from the packages.  Or, at least, I could not install it.
2020-12-13 13:17:44 <cads> 
ACTION is going to continue going on as if :provides packages don't exist, lol

2020-12-13 13:17:59 <ikke> cads: yes, they are not that important
2020-12-13 13:18:38 <cads> unless you're a new user trying to get the mount command working :/
2020-12-13 13:19:25 <ikke> pkgs.alpinelinux.org is probably a bit easier for new users
2020-12-13 13:21:21 <cads> could use a way to search these abstract packages. I searching for the mount file in the package and found just a 5 or so of the 16 results given by apk search
2020-12-13 13:21:38 <mps> systemdlete: os-prober is in testing repo
2020-12-13 13:21:59 <ikke> cads: apk search does a wildcard search
2020-12-13 13:22:07 <ikke> so it returns more then just mount
2020-12-13 13:22:40 <cads> eg, mount.fuse3
2020-12-13 13:23:03 <cads> okay, I buy it
2020-12-13 13:23:26 <cads> maybe I'll pitch in and sketch an update for the docs
2020-12-13 13:24:29 <cads> Is it possible to track when the docs and apk went out of sync, or determine if that's what really happeend here?
2020-12-13 13:25:22 <cads> ie, if there's some other place where the wiki discusses these provides, all the main manual needs is a link to that document
2020-12-13 13:26:52 <systemdlete> mps:  I don't usually like to mix different levels of software.  If the testing repo is compat with the current release, I would try it.  But that's OK.  For now, I kinda stole the devuan-generated grub.cfg and modified that.  It works now.  Thanks again for your help.
2020-12-13 13:27:49 <cads> also is it possible to write a pull request against the online package search? does code for this thing live in the public:  https://pkgs.alpinelinux.org/packages ?
2020-12-13 13:30:16 <ikke> cads: aports-turbo
2020-12-13 13:30:25 <ikke> https://gitlab.alpinelinux.org/alpine/infra/aports-turbo
2020-12-13 13:32:28 <cads> apart from cmd: and lib: , do any other provides exist?
2020-12-13 13:32:34 <maxice8> pc:
2020-12-13 13:32:42 <ikke> so:
2020-12-13 13:32:50 <maxice8> dbus:
2020-12-13 13:32:59 <maxice8> there is no lib: as far as I'm aware
2020-12-13 13:33:14 <ikke> is dbus: already being used?
2020-12-13 13:33:41 <maxice8> I think yes
2020-12-13 13:33:47 <maxice8> @Cogitri, @PureTryOut
2020-12-13 13:34:47 <Cogitri> We only use it for libsecret
2020-12-13 13:35:13 <Cogitri> Since keepassxc, GNOME Keyring and KWallet can expose that DBus interface
2020-12-13 13:35:27 <Cogitri> But right now we just manually slapped it onto those APKBUILDs
2020-12-13 13:35:38 <cads> do appear :-tags appear relatively recently, in software archeology terms?
2020-12-13 13:35:58 <cads> err, that first "appear" needs to go
2020-12-13 13:36:12 <ikke> cads: I think the so: provides are older
2020-12-13 13:36:16 <ikke> the rest is fairly recent
2020-12-13 13:36:34 <cads> could explain the documentation lag
2020-12-13 13:40:18 <cads> so we have this control dir and at this abuild.in phase of the process, we shove stuff into files in the control dir: .provides-command, .provides-pc, and so on. Then at some later phase we will look at that control dir and the .provides- files in it
2020-12-13 13:41:08 <ikke> yes, that's my understanding
2020-12-13 13:41:08 <cads> https://gitlab.alpinelinux.org/alpine/abuild/-/blob/master/abuild.in#L1421   for example here, where we do something with sed that is so ugly that it makes me weep, and then we add that to the package's pkginfo
2020-12-13 13:42:14 <ikke> I don't know why it's done like that
2020-12-13 13:42:50 <cads> oh, it's not doing anything complicated, that's just sed being a bad api for touching config files
2020-12-13 13:42:51 <ikke> maybe some kind of backwards compatibility?
2020-12-13 13:43:18 <ikke> it's just prefixing every line with 'provides = cmd:'
2020-12-13 13:49:45 <john17> hey i try to boot alpine linux using dd on my usb and my laptop only detects the mbr version and when i boot the mbr version it reboots back to my main os
2020-12-13 13:50:42 <john17> well its a usb hdd but it should not be a issue (idc if i lose data as there is non)
2020-12-13 13:50:51 <john17> (only alpine linux)
2020-12-13 13:51:03 <john17> (installer)
2020-12-13 13:55:14 <tomleb> ius: I will try this!
2020-12-13 13:55:28 <cads> ikke,  I know, I just think that if we do "provides = cmd:$f" at  line 1198 (compat is not an issue)
2020-12-13 13:56:18 <cads> then, we can expect .provides-command to be properly formatted at line 1420. And it better documents what's happening at 1198 (before we were compiling a list of command names, now we're really listing key = value pairs)
2020-12-13 13:56:24 <cads>      https://gitlab.alpinelinux.org/alpine/abuild/-/blob/master/abuild.in#L1198
2020-12-13 13:56:57 <ikke> well, the backwards compattibility is if .provides-command is used by older versions
2020-12-13 13:57:36 <cads> ouch
2020-12-13 13:57:57 <ikke>  I have no idea if that's the case, but the way this is being done seems to indicate that
2020-12-13 13:58:21 <cads> I would say it's entirely plausible and needs to be checked
2020-12-13 13:58:30 <cads> how fun
2020-12-13 13:59:02 <cads> but hey, I now understand a lot more than I did about abuild
2020-12-13 13:59:19 <john17> hey i try to boot alpine linux using dd on my usb and my laptop only detects mbr (for the hdd i dont know why) and when i boot the mbr version it reboots back to my main oswell its a usb hdd but it should not be a issue (i dont really care if i lose data as there is non)
2020-12-13 14:00:43 <cads> ikke, and thanks for the aports turbo link. This code is so clean!
2020-12-13 14:01:56 <cads> makking additions to search for these meta packages wouldn't be very hard, it would more be about how not to make them more intrusive than their importance really deserves
2020-12-13 14:03:26 <ikke> and performance also needs to be taken into account
2020-12-13 14:04:44 <cads> currently to simulate search apk cmd:mount, we need to know to search for mount files - with or without wildcard, as our preference might be (the apk command uses wild card) - it would be possible to, in the background, simply turn cmd:mount into the appropriate file search
2020-12-13 14:05:35 <cads> ikke, do file owner searches tax the database?
2020-12-13 14:08:35 <mps> I doubt wildcards (or worse regex) will be accepted in apk
2020-12-13 14:13:13 <cads> might need to extend the aports data model keep explicity keep track of these :-packages - that gives a zero regex query, and lets you implement:   1)  searching cmd: provides users a paged listing of all command provides    2) searching cmd:mount gives users the same results as apk search cmd:mount
2020-12-13 14:14:21 <cads> this requires parsing of the provides = lines during aport's PKGINFO ingest phase (which I have not inspected yet)
2020-12-13 14:15:05 <cads> https://gitlab.alpinelinux.org/alpine/infra/aports-turbo/-/blob/master/model.lua   might be a method of the model, not sure
2020-12-13 14:17:49 <cads> ikke, do you know if the shape of the database is determined by the model declarations plus an ORM? If so, what lines should I look at to get a quick idea of what kinds of properties and tables aports stores?
2020-12-13 14:18:30 <cads> here we go https://gitlab.alpinelinux.org/alpine/infra/aports-turbo/-/blob/master/tools/import.lua
2020-12-13 14:24:26 <cads> ikke, thanks for the help, it's been enormous! I feel like I've seen enough source code to be dangerous, and I've solved some mysteries about how the apk/abuild/aports ecosystem ticks along
2020-12-13 14:25:52 <cads> ikke, if you wish, I'll ping you for your opinion if I come up with any improvements to aports or abuild!
2020-12-13 14:26:31 <ikke> sure
2020-12-13 14:26:44 <cads> before I send a pr, always helps to get some input
2020-12-13 14:26:45 <ikke> note that I don't maintain either systems
2020-12-13 14:26:47 <ikke> right
2020-12-13 14:31:17 <cads> cool, well, glad we could chat today
2020-12-13 14:44:15 <cads> ikke, is it a requirement for abuilds.in to be a shell script?
2020-12-13 14:44:49 <ikke> well, it integrates tightly with APKBUILDs
2020-12-13 14:45:32 <cads> does it need to be run in an stripped down, sandboxed environment where we don't have, eg, python?
2020-12-13 14:46:00 <ikke> yes, its part of the bootstrap chain
2020-12-13 14:46:09 <ikke> you want as little dependencies as possible
2020-12-13 14:46:20 <cads> yeah, that makes sense
2020-12-13 14:48:11 <ikke> You don't just rewrite abuild in another language
2020-12-13 14:48:17 <ikke> it would be a big endeavour
2020-12-13 14:48:32 <cads> is the idea that any system that runs alpine should also be able to run abuild with minimal additional dependencies?
2020-12-13 14:49:03 <ikke> when bootstrapping a new arch, you want abuild to be available very early
2020-12-13 14:50:02 <ikke> the more dependencies, the more work you need to do manually
2020-12-13 14:51:56 <cads> oh, I wasn't aware of that benefit
2020-12-13 14:53:06 <cads> so the only reasonable language to rewrite it in would be c
2020-12-13 14:53:21 <ikke> and the question is whether that's worth it
2020-12-13 14:53:28 <cads> and I don't think that would be at be reasonable in itself
2020-12-13 14:53:34 <ikke> right
2020-12-13 14:54:34 <cads> lol, I hate that we have a sed but not a python
2020-12-13 14:54:55 <cads> but oh well, I guess that's cross platform life
2020-12-13 14:55:34 <mps> sed is busybox applet by default
2020-12-13 14:56:05 <cads> I'm gonna slip micropython into there
2020-12-13 14:56:15 <cads> shhh don't tell
2020-12-13 14:56:37 <mps> will not pass, don't worry
2020-12-13 14:57:51 <cads> assuredly not, lol, but the patch would be hilarious
2020-12-13 14:59:19 <mps> just if you want bad karma :)
2020-12-13 15:00:21 <cads> I'm going to laugh demonically when I find out that micropython has over 20x the line count of busybox. And then I will write that patch and burn in hell later
2020-12-13 15:00:48 <cads> just for having the patch on my harddrive at one point in time, soul lost
2020-12-13 15:01:20 <Apachez> we can burn you free of charge
2020-12-13 15:01:48 <cads> that something so evil and affronting to the holy purity should be ever exist, basically voids one's soul into unfathomable suffering
2020-12-13 15:02:12 <cads> to the holy purity of busybox*
2020-12-13 15:02:33 <cads> lol, lets hope my patch is just as well written as my english, too ;)
2020-12-13 15:03:03 <Apachez> or just make python part of busybox =)
2020-12-13 15:03:26 <Apachez> ln -s /usr/bin/python /usr/bin/busybox%20python
2020-12-13 15:03:29 <cads> well, that's what the patch will by - adds micropython to busybox
2020-12-13 15:03:39 <tehcloud> or purge python from reality
2020-12-13 15:03:57 <tehcloud> like it never existed at all
2020-12-13 15:04:02 <Apachez> while you are at it also add microsystemd to busybox =)
2020-12-13 15:04:13 <cads> tehcloud, what if doing that meant everything that is now python becomes... typescript X(
2020-12-13 15:04:27 <tehcloud> lol, well... you could say that about perl too
2020-12-13 15:04:33 <tehcloud> there are worse things out there for sure
2020-12-13 15:05:07 <cads> tehcloud, the thing about perl is that only resentful neckbeards push it anymore these days, all the smart kids are doing python
2020-12-13 15:06:21 <cads> "I'm holding the weight of the world on my shoulders by maintaining this awful perl script and nobody cares as I suffer in silence. But i do it. Because it is my duty. Because I hold up part of the internet with this perl script. And I must. Go. On""
2020-12-13 15:06:25 <cads> ^ these types
2020-12-13 15:07:08 <Apachez> "there are many scripts like this but this one is mine" etc? ;)
2020-12-13 15:07:44 <cads> lool, I have a few of those these days, now that I'm spending so much time in alpine/ash
2020-12-13 15:08:19 <cads> I have read the ash manual end to end so many times I think I could recite it - luckily it's so damn short that I now officially know ash much much better than I ever knew bash
2020-12-13 15:08:46 <tehcloud> "smart kids" using python... lol
2020-12-13 15:08:57 <cads> tehcloud, you know, the phd kids
2020-12-13 15:09:11 <tehcloud> in the field of data science, which has all the buzz, sure
2020-12-13 15:09:32 <cads> making filters that can amplify the dog content of any image, irrespective of initial dog content
2020-12-13 15:09:56 <tehcloud> I am personally a resentful facebeard, so that means I push C instead of perl
2020-12-13 15:10:02 <cads> and all sorts of wacky stuff we most definitely asked them NOT to create
2020-12-13 15:10:42 <cads> tehcloud, honesty the more responsible and yay, commendable of the two choices, lol
2020-12-13 15:11:17 <tehcloud> I ended up on this distro largely because of C
2020-12-13 15:11:27 <tehcloud> using it to test my codebases with musl more or less
2020-12-13 15:11:50 <tehcloud> but then I realized this distro is awesome, so when I built my next "main desktop" I installed Alpine there as well
2020-12-13 15:14:31 <cads> I think I am here because docker didn't and still doesn't have a real great story for unikernels, and so docker alpine images are the next best thing - then I noticed that apk package installs were blazing fast, and that I could convert apt dependencies into apk dependencies with no trouble in most cases
2020-12-13 15:15:31 <cads> then I started using a windows workstation for work, but wsl came to the rescue and has given me a fully integrated hybrid windows/alpine workstation with near native linux performance
2020-12-13 15:17:30 <cads> and using sh for a daily driver is not that bad
2020-12-13 15:20:33 <tehcloud> I use bash on here
2020-12-13 15:20:44 <tehcloud> and some other GNU tools
2020-12-13 15:21:19 <tehcloud> but mostly I like the minimalism and the ability to find shit codebases that don't work with musl
2020-12-13 15:27:00 <Apachez> cads: and full support of billions of malware =)
2020-12-13 15:27:03 <Apachez> I do the other way around
2020-12-13 15:27:09 <Apachez> using ubuntu as host and wintendo in a vm
2020-12-13 15:27:44 <cads> Apachez, the are part of my body now
2020-12-13 15:29:51 <cads> it's not like I'm in windows because i'm a gamer or anything. But the reason I'm here is almost as bad - technical debt in /my/ chosen cad design platform (I chose a windows only platform because it was super cheap but also had decent APIs)
2020-12-13 15:32:02 <cads> I could be anywhere else I wanted to be, apart from this one app, which pukes itself into unusable pixelated oblivion in a vm, with NO vendor support for that use case
2020-12-13 15:33:59 <cads> but at least thanks to alpine, I can still drive a local shell in a fully sane linux environment, where I can /forget/ that underneath me is a writhing mass of closed source snakes, coiling around me like undiscovered CVEs
2020-12-13 15:34:48 <tehcloud> I have to keep a Windows VM around that can never be updated in order to maintain some old Windows software :|
2020-12-13 15:35:04 <cads> tehcloud, I hate that soooo much!
2020-12-13 15:35:49 <cads> arguably I'm experiencing that with my /entire base OS/, but at least I get to update
2020-12-13 15:37:16 <cads> but I'm honestly waiting for windows to fold itself away into a gui server for WSL, so I've started saying that Windows 10 + Alpine WSL is my favorite linux distro
2020-12-13 15:38:11 <cads> year of linux on the desktop is 2020 thanks to wsl 2, lol
2020-12-13 15:38:25 <tehcloud> it has been the year of the Linux desktop since the 90s
2020-12-13 15:38:50 <cads> I know, didn't lunduke do a hilarious video on that?
2020-12-13 15:39:09 <mps> tehcloud: yes, my desktop is only linux from '93
2020-12-13 15:39:31 <cads> secret of comedy: repetition. for 30. years.
2020-12-13 15:39:58 <mps> I don't care what other people use
2020-12-13 15:40:27 <tehcloud> me neither, I just care that I can use what I want without too much hassle
2020-12-13 15:44:44 <cads> aren't there ppl still actively developing amiga os? that's where it's at. That's love.
2020-12-13 15:45:45 <cads> like, it will be the post singularity and there will still be unmodified humans who insist on remaining unmodified so they can continue maintaining mainline amiga
2020-12-13 15:47:41 <cads> "it's not authentic mainline amiga unless it's mainline devs doing it for fun, don't run the fork made by the saturn hypermind. Yes it fixes mutexes but they used hyperturing machines to do that shit."
2020-12-13 16:01:36 <fourstepper> Is it possible to do something similar to the answer as this for GRUB for syslinux? https://stackoverflow.com/questions/62470055/add-an-option-to-grub-command-line-in-etc-default-grub#62470463
2020-12-13 16:02:56 <fourstepper> Basically extra options (https://wiki.alpinelinux.org/wiki/Docker#Extlinux) , just included instead of directly added to /etc/update-extlinux.conf
2020-12-13 16:10:08 <Hello71> i don't think so
2020-12-13 16:14:05 <fourstepper> Alright, thanks :)
2020-12-13 18:17:14 <David_9> Aloha
2020-12-13 18:34:08 <ipcloud> David_9: Greetings
2020-12-13 19:17:21 <xtrWrithe> hello!, somebody had luck running i2c on rock64 boards using alpine?
2020-12-14 04:04:22 <cornfeedhobo> this would be nice to have packaged https://github.com/skeeto/endlessh
2020-12-14 04:05:09 <TBK[m]> drop at request at https://gitlab.alpinelinux.org/alpine/aports/-/issues
2020-12-14 04:05:23 <TBK[m]>  * drop a request at https://gitlab.alpinelinux.org/alpine/aports/-/issues
2020-12-14 04:07:04 <cornfeedhobo> okay
2020-12-14 04:15:59 <cornfeedhobo> thanks
2020-12-14 05:25:49 <sam_son> hello everyone
2020-12-14 05:26:04 <TBK[m]> hi
2020-12-14 05:26:57 <sam_son> I need to run rabbitmq, which appears to be available in edge. I switched to edge, run apk add rabbitmq-server but it fails because erlang packages are not found. I'd appreciate some help.
2020-12-14 05:27:10 <sam_son> hi TBK[m]
2020-12-14 05:35:05 <TBK[m]> sam_son: rabbitmq-server was apparently disabled because of the erlang pkg got upgraded to 23 - https://git.alpinelinux.org/aports/log/testing/rabbitmq-server/APKBUILD
2020-12-14 05:35:05 <TBK[m]> Many of the erlang subpackages got removed. Comment on why can be found here - https://git.alpinelinux.org/aports/commit/community/erlang?id=53aebeddaa4ae6901b1c40652c1cdd21b57106cb
2020-12-14 05:35:05 <TBK[m]> MR/Patches to fix rabbit are welcome :)
2020-12-14 05:35:48 <sam_son> oh damn
2020-12-14 05:36:39 <sam_son> hmm
2020-12-14 05:37:02 <ikke> they are part of erlang itself now
2020-12-14 05:37:17 <ikke> no longer separate subpackages
2020-12-14 05:37:41 <sam_son> to be honest, I don't have time to look into it, so I'll need to boot another distro
2020-12-14 05:38:09 <sam_son> at least mariadb installed and worked with no issues
2020-12-14 05:39:19 <sam_son> oh well, thank you all for the quick help
2020-12-14 05:39:31 <ikke> sam_son: hmm, if I look at rabbitmq-server on edge, it does not seem to depend on the erlang packages
2020-12-14 05:40:24 <sam_son> but 'apk add rabbitmq-server' fails with errors about erlang packages
2020-12-14 05:41:09 <ikke> Yes, I see
2020-12-14 05:41:10 <TBK[m]> https://tpaste.us/R4do
2020-12-14 05:43:51 <ikke> Right, if I look at the APKBUILD, I see them
2020-12-14 08:37:22 <rnkn> how might I go about troubleshooting no sound? lsmod lists the sound modules, aplay -l finds the correct soundcard, and aplay WAVFILE acts as if it's playing the file, but no sound emits from speakers or headphones
2020-12-14 08:37:50 <ikke> is the sound muted?
2020-12-14 08:37:58 <ikke> amixer / alsamixer
2020-12-14 08:38:04 <rnkn> not according to alsamixer
2020-12-14 08:43:12 <rnkn> oh hang on, all the way to the right is Left/Right Headphone Mixer, which were muted, now I get some weird crackles/pops, but not sound
2020-12-14 08:46:07 <MY-R> rnkn: you sure that unmuted everything and volumed up PCM or even "front" channel?
2020-12-14 08:46:39 <MY-R> could be that some aggressive power saving mode so disable auto-mute stuff
2020-12-14 08:47:27 <rnkn> there are a lot of things like ALC Capture Decay Time, etc that I don't think are about sound output
2020-12-14 08:47:40 <rnkn> should DAC be 100%?
2020-12-14 08:48:11 <MY-R> dunno what sound card you have got
2020-12-14 08:48:28 <MY-R> and dunno which one should be your default one
2020-12-14 08:48:58 <MY-R> maybe the one on different slots is your default, tried check what got under alsamixer -c1
2020-12-14 08:49:03 <MY-R> -c2 etc
2020-12-14 08:50:08 <ikke> also check amixer
2020-12-14 08:50:17 <rnkn> I just have the one soundcard, only alsamixer -c0 works
2020-12-14 08:52:24 <rnkn> amixer outputs about 20 mixer controls
2020-12-14 08:55:55 <MY-R> rnkn: paste your aplay -l and amixer output somewhere
2020-12-14 08:56:26 <rnkn> aplay -l http://ix.io/2HYK
2020-12-14 08:56:27 <rnkn>  
2020-12-14 08:56:54 <rnkn> amixer pc
2020-12-14 08:57:06 <rnkn> * amixer http://ix.io/2HYL
2020-12-14 08:57:36 <MY-R> not sure if you have to be in group "audio" but just to be sure add your user to it
2020-12-14 08:58:13 <rnkn> yep, both me and root are in audio
2020-12-14 08:59:40 <rnkn> amixer output shows the Headphone Mixer is "turned up to 11"
2020-12-14 08:59:46 <rnkn> (;
2020-12-14 09:01:54 <rnkn> upon attempting any playback I'm getting a short "pop" sound, but nothing else
2020-12-14 09:03:31 <MY-R> it is some sound card hat so there can be something deeper
2020-12-14 09:05:27 <rnkn> MY-R: a hat?
2020-12-14 09:06:08 <MY-R> dunno, you tell me how it looks or on what it working :)
2020-12-14 09:06:35 <MY-R> some arm board? or usb?
2020-12-14 09:06:58 <MY-R> did it work on different distribution?
2020-12-14 09:09:59 <rnkn> it's on a Pinebook Pro, I can't actually remember if it worked on Manjaro... I think so
2020-12-14 09:10:06 <MY-R> rnkn: run those both:
2020-12-14 09:10:09 <MY-R> amixer set 'Right Headphone Mixer Right DAC' on
2020-12-14 09:10:15 <MY-R> amixer set 'Left Headphone Mixer Left DAC' on
2020-12-14 09:11:19 <MY-R> ah it is already on, nevermind
2020-12-14 09:11:49 <rnkn> ah I thought we'd found a breakthrough...!
2020-12-14 09:11:51 <rnkn> alas
2020-12-14 09:14:48 <MY-R> rnkn: so headphone not working, speakers too?
2020-12-14 09:15:10 <rnkn> both emit the pop but not audio
2020-12-14 09:17:48 <rnkn> I've muted the internal speaker for the time being to isolate to just the headphones
2020-12-14 09:23:06 <MY-R> rnkn: anything weird in dmesg? dmesg | grep sound
2020-12-14 09:26:31 <rnkn> oh yeh, upon booting ALSA says no soundcard found
2020-12-14 09:26:37 <rnkn> also, lspci shows nothing
2020-12-14 09:26:51 <rnkn> (what's the flag for grep to give N lines of context?)
2020-12-14 09:27:17 <rnkn> (found it)
2020-12-14 09:27:36 <MY-R> hmm so what the hell aplay -l showed :\
2020-12-14 09:28:15 <rnkn> dmesg http://ix.io/2HZ2
2020-12-14 09:30:13 <MY-R> there should be probably more sound devices like hdmi so maybe some module is missing or cant load
2020-12-14 09:30:49 <MY-R> rnkn: lsmod |grep snd
2020-12-14 09:34:47 <rnkn> lsmod | grep snd http://ix.io/2HZ4
2020-12-14 09:35:27 <rnkn> technically pipped thru column, because aesthetics are important
2020-12-14 09:38:19 <MY-R> :)
2020-12-14 09:39:32 <cads> does anyone recommend the speed and lightness of alpine with the determinism and completeness of nixos?
2020-12-14 09:39:48 <cads> thinking of running some tests
2020-12-14 09:41:50 <MY-R> rnkn: which kernel do you using? in Alpine I dont even see that sound module enabled at all or Im just missing something :\
2020-12-14 09:43:50 <rnkn> MY-R: it's actually postmarketOS, the kernel is 5.9.something... where do I find the kernel version again?
2020-12-14 09:44:13 <rnkn> of course, I realise it's a bit unfair to ask #alpine-linux about pmOS... sorry!
2020-12-14 09:44:24 <cads> also, does anyone feel there could be a musl-related reason why certain rust packages build fine in WSL Alpine, but throw segfault as soon as I run them?
2020-12-14 09:44:24 <ikke> rnkn: uname -r
2020-12-14 09:44:40 <rnkn> 5.9.0-rc8
2020-12-14 09:44:40 <cads> examples include onefetch and, tealdeer
2020-12-14 09:46:39 <rnkn> according to pmOS, it's running "the 5.9-rc kernel directly from kernel.org + one patch from linux-next"
2020-12-14 09:46:46 <rnkn> although not sure what that one patch is
2020-12-14 09:47:24 <MY-R> rnkn: so all that should work out of the box, you for sure got installed package "device-pine64-pinebookpro" and that should do the job
2020-12-14 09:47:34 <mps> rnkn: did you tried alpine kernel linux-edge, 5.9.14
2020-12-14 09:48:10 <rnkn> MY-R: apk info -e check out
2020-12-14 09:48:31 <rnkn> mps: I have not, is it a case of install from apk?
2020-12-14 09:48:49 <mps> rnkn: yes, testing repo
2020-12-14 09:49:18 <MY-R> can install linux-edge in pmOS?
2020-12-14 09:49:55 <mps> ah, rnkn runs pmOS and not alpine
2020-12-14 09:50:16 <MY-R> ye
2020-12-14 09:50:24 <MY-R> in Alpine # CONFIG_SND_SOC_ES8316 is not set
2020-12-14 09:50:24 <mps> :(
2020-12-14 09:50:51 <rnkn> well, glad I didn't go ahead and replace the kernel then :)
2020-12-14 09:51:26 <rnkn> there's always the possibility that this PBP is just a lemon.. it has had quite a few issues
2020-12-14 09:51:50 <MY-R> https://gitlab.com/postmarketOS/pmaports/-/tree/master/device/testing/device-pine64-pinebookpro/ucm
2020-12-14 09:52:03 <MY-R> looks like sound card should be configured already
2020-12-14 09:52:20 <mps> afaik halosghost runs alpine on pbp
2020-12-14 09:52:47 <MY-R> but it is in "testing" dunno like it looks in pmOS
2020-12-14 09:53:25 <rnkn> is there a way to see which repo an apk is in via `apk info`?
2020-12-14 09:53:46 <mps> apk policy pkg
2020-12-14 09:54:12 <rnkn> mps: thanks
2020-12-14 09:55:10 <mps> MY-R: well, if no one test and report back to us, or pine don't send me pbp how can we know what should be enabled in kernels and other 'things'
2020-12-14 09:55:49 <MY-R> rnkn: so you dont have that paclkage installed? "device-pine64-pinebookpro"
2020-12-14 09:56:07 <rnkn> MY-R: sorry, yes it is installed
2020-12-14 09:56:28 <MY-R> ye so all should work but doesnt
2020-12-14 09:56:29 <rnkn> as is -nonfree-firmware
2020-12-14 09:56:59 <rnkn> I have also pulled the PBP apart and checked that the UART switch is set to audio
2020-12-14 09:58:19 <rnkn> it's possible that I broke a wire when disassembling/reassembling, but I'm sure this would not affect both speakers and headphone
2020-12-14 09:58:36 <rnkn> (although I've checked all wires)
2020-12-14 10:03:14 <MY-R> but in pmOS they got own way to adjust some stuff
2020-12-14 10:04:57 <MY-R> "alsaucm" command
2020-12-14 10:06:41 <MY-R> rnkn: paste "aplay -L"
2020-12-14 10:11:13 <rnkn> stop the presses I have some audio!
2020-12-14 10:12:05 <MY-R> ye I wanted give you that line: alsaucm -c HERE-MODULE set _verb HiFi set _enadev Headphone
2020-12-14 10:12:26 <MY-R> that is probably better way to switch between headphone and speakers
2020-12-14 10:13:01 <rnkn> I think alsaucm reset may have done it
2020-12-14 10:13:42 <MY-R> ye, in Alpine wont find that :)
2020-12-14 10:14:19 <MY-R> so ye, enjoy!
2020-12-14 10:15:02 <rnkn> thanks!
2020-12-14 10:15:21 <rnkn> oh darn, I did a `while true; do aplay WAVFILE; done` and now can't ctrl-C/killall it...
2020-12-14 10:15:33 <ikke> try to hold ctrl+c
2020-12-14 10:15:37 <ikke> that sometimes works :)
2020-12-14 10:15:45 <MY-R> heheh
2020-12-14 10:16:07 <ikke> (from experience)
2020-12-14 10:16:40 <rnkn> nah, it's fine, I'll just reboot when I'm done testing
2020-12-14 10:16:55 <MY-R> oh I have got a question, how quit from that mode after when you type "' and enter" in ash....
2020-12-14 10:17:01 <rnkn> okay so weird thing, opening alsamixer and if I turn the headphone mixer up into the red, it cuts off sound
2020-12-14 10:17:09 <ikke> MY-R: ctrl+d
2020-12-14 10:17:12 <rnkn> this may have been why I was not getting sound
2020-12-14 10:17:29 <rnkn> ikke: ctrl-D doesn't wanna kill it either
2020-12-14 10:17:33 <MY-R> ikke: wth.... I was doing that so many times and didnt work................
2020-12-14 10:17:42 <ikke> rnkn: that was a response to MY-R
2020-12-14 10:18:37 <MY-R> ikke: rly I had to done something more weird because ctrl + d didnt work before... now working, thanks!
2020-12-14 10:18:53 <MY-R> rnkn: that is why you need use alsaucm thing
2020-12-14 10:19:05 <ikke> MY-R: some quantum wave that collapsed :P
2020-12-14 10:19:44 <MY-R> rnkn: you can check how it work here, so can adjust your modifications if using alsamixer maybe: https://gitlab.com/postmarketOS/pmaports/-/blob/master/device/testing/device-pine64-pinebookpro/ucm/HiFi
2020-12-14 10:20:39 <rnkn> haha gitlab really does not like non-javascript browsers
2020-12-14 10:20:52 <rnkn> (I'm on console, using Emacs EWW for web)
2020-12-14 10:21:08 <MY-R> rnkn: ok this is raw file: https://gitlab.com/postmarketOS/pmaports/-/raw/master/device/testing/device-pine64-pinebookpro/ucm/HiFi
2020-12-14 10:22:38 <MY-R> ikke: many times I forgot close some one liner script with single quote and I was ending up with closing terminal... was hiting half of keyboard letters together with CTRL and never worked :D
2020-12-14 10:23:29 <ikke> MY-R: I know the feeloing
2020-12-14 10:23:36 <rnkn> MY-R: thanks, I've saved that for future reference
2020-12-14 10:23:46 <ikke> closing the quote should also have helped
2020-12-14 10:23:57 <MY-R> but ye, thanks, heh :)
2020-12-14 10:24:14 <MY-R> oh ye...
2020-12-14 10:25:13 <rnkn> okay this is enough pink noise for me, I'm gonna disconnect/reboot, thank you for all the help!
2020-12-14 10:25:56 <MY-R> and now I feel stupid but also smarter... next time wont smash all keyboards keys :) closing working too of course...
2020-12-14 14:04:58 <hechos> hi guys.
2020-12-14 14:05:25 <hechos> what is the best choice to bootloader to zfs, grub is harder and buggy to configure.
2020-12-14 14:05:48 <Cogitri> I usually just use rEFInd
2020-12-14 14:05:57 <Cogitri> But I have a vfat /boot
2020-12-14 14:20:35 <hechos> Cogitri: is ease to configure use refind on void uefi btrfs just work fine and easy but never configure with zfs
2020-12-14 14:22:58 <Hello71> just put kernel on esp
2020-12-14 14:24:40 <klesk> fset
2020-12-14 15:03:21 <greyxor> hi :)  when 3.13 will be tagged ?
2020-12-14 15:04:33 <ikke> greyxor: community is still being buitl
2020-12-14 15:04:38 <ikke> built*
2020-12-14 15:05:49 <greyxor> what does that mean ? i ask because 2 days ago, mps tell me that 3.13 is tagged soon. but i can't see it in git.alpine
2020-12-14 15:06:12 <ikke> greyxor: We can't see exactly when, but we are working hard to get it done
2020-12-14 15:06:31 <ikke> making usre all packages are building is quite bit of work
2020-12-14 15:07:09 <greyxor> yeah sure, nice jobs. alpine's very useful for my usage, thanks to all staff :)
2020-12-14 15:07:53 <ikke> It's a community distribution, everything is being done by volunteers
2020-12-14 15:07:56 <ikke> no staff :)
2020-12-14 15:08:15 <greyxor> thanks to every contributors then
2020-12-14 15:08:25 <ikke> you're welcoem
2020-12-14 15:10:12 <mps> greyxor: I told that the 3.12.2 is tagged
2020-12-14 15:12:37 <greyxor> ah, i see 3.13
2020-12-14 15:17:00 <perrypal> hello
2020-12-14 15:17:19 <hechos> kernel 5.10 is lts right?
2020-12-14 15:17:20 <omni> hechos: I boot with syslinux, but have /boot on a ext4 partition (on mdraid)
2020-12-14 15:17:34 <hechos> omni: aha
2020-12-14 15:17:35 <ikke> hechos: not yet, but will be
2020-12-14 15:17:50 <hechos> so kernel gone bump to?
2020-12-14 15:18:00 <ikke> hechos: ncopa is testing it
2020-12-14 15:18:20 <hechos> ikke: alight.
2020-12-14 15:18:25 <hechos> alright.
2020-12-14 15:18:39 <perrypal> quick question, how would I go about setting up a cron job
2020-12-14 15:18:44 <perrypal> in alpine
2020-12-14 15:19:33 <ikke> either through /etc/periodic/*
2020-12-14 15:19:37 <ikke> or crontab
2020-12-14 15:23:23 <perrypal> I did add a script in /etc/periodic but I dont think it works
2020-12-14 15:23:27 <perrypal> I will try crontab then
2020-12-14 15:23:45 <ncopa> im configuring the 5.10 kernel now. let me know if there are any specific kernel options you are interested in
2020-12-14 15:23:53 <omni> perrypal: https://wiki.alpinelinux.org/wiki/Alpine_Linux:FAQ#My_cron_jobs_don.27t_run.3F
2020-12-14 15:23:56 <ikke> perrypal: is crond running?
2020-12-14 15:25:35 <perrypal> lemme check
2020-12-14 15:26:09 <perrypal> it says that crond is started
2020-12-14 15:29:25 <caskd> ncopa: maybe CONFIG_CHECKPOINT_RESTORE would be useful, mesa drm can have some problems with os_same_file_description and throws warnings about it
2020-12-14 15:30:51 <caskd> related issue: https://gitlab.freedesktop.org/mesa/mesa/-/issues/2923
2020-12-14 15:35:57 <ncopa> what architectures?
2020-12-14 15:41:50 <caskd> x86_64
2020-12-14 15:42:24 <caskd> probs x86 as well, not sure
2020-12-14 15:49:23 <mps> yes, this ok to enable, and on aarch64 too
2020-12-14 15:49:58 <mps> I intended to enable it on linux-edge but forgot
2020-12-14 15:51:01 <mps> and CONFIG_PREEMPT=y
2020-12-14 15:51:07 <mps> on all
2020-12-14 15:56:52 <artok> I need to build and configure uboot for rpi
2020-12-14 15:57:43 <artok> removing boot media, mounting that into laptop for configure, putting it back to rpi.. not fun =)
2020-12-14 16:15:36 <minimal> artok: I've been chatting with mps about rpi and u-boot recently. I sent in a couple of MRs during the weekend to start to tidy things up (adding missing deps). I was intending to next look at how to get setup-alpine to support both normal bootloader and u-boot
2020-12-14 16:30:40 <hechos> anyone using qutebrowser feels pretty buggy on musl right?
2020-12-14 16:31:05 <hechos> is a chromium related or python3?
2020-12-14 18:06:21 <hechos> where have to put on aprots/scripts to build iso zfs already add to world but have to modprobe zfs
2020-12-14 18:06:45 <hechos> also add to cmdline and features.
2020-12-14 19:25:53 <caskd> hechos: tried to use quotebrowser but it has more than 1 bug and the debug messages are so useless i didn't know what to work on to fix it
2020-12-14 19:26:34 <caskd> would gladly add patches to it if you could tell me where the problems lie
2020-12-14 19:27:11 <caskd> that's the main reason why i ported ungoogled-chromium to alpine...
2020-12-14 19:27:52 <caskd> one thing i know for sure is that seccomp causes most crashes
2020-12-14 19:28:45 <caskd> but what part of seccomp exactly is still a mistery to me, maybe adding seccomp code from the official chromium project as patches would help but i'm unsure
2020-12-14 23:30:22 <emlen> :)
2020-12-15 09:05:32 <telmich> I get segmentation fault while starting the network on some systems
2020-12-15 09:05:40 <telmich> `ip` seems to work, buf `ifup` gets a segmentation fault
2020-12-15 09:09:09 <telmich> [10:08] router2.place5:~# ifup bond0.3 -> Segmentation fault
2020-12-15 09:49:00 <telmich> Just created https://github.com/ifupdown-ng/ifupdown-ng/issues/138
2020-12-15 16:40:21 <mpy> Hi, how do I use ipsets with awall (Alpine Wall). I'm able to create empty sets, but how am I supposed to get data into them? Can't find anything about it in the documentation, on google or the source code.
2020-12-15 16:47:47 <ikke> I don't think awall manages ipsets
2020-12-15 16:48:07 <ikke> Oh, it does, sorry
2020-12-15 16:48:22 <ikke> But I think the ipset needs to already exist
2020-12-15 16:48:53 <mpy> documentation says: "Any IP set referenced by rule objects should be created by awall."
2020-12-15 16:49:04 <ikke> https://gitlab.alpinelinux.org/alpine/awall#ip-sets
2020-12-15 16:49:06 <ikke> right
2020-12-15 19:32:41 <hechos> hello guys
2020-12-15 19:33:06 <hechos> how can get /dev/disk for persistens names tp zds
2020-12-15 19:33:09 <hechos> zfs
2020-12-15 19:33:15 <hechos> udevadm tigger?
2020-12-15 19:35:54 <hechos> yeah need udev. nevermind.
2020-12-15 22:29:09 <tehcloud> anyone with an RDNA AMD GPU experiencing regressions on the 5.10 kernel?
2020-12-16 00:19:33 <maldridge> is there a way to specify persistent mount options for things like /run?
2020-12-16 00:21:59 <Hello71> fstab?
2020-12-16 00:22:46 <maldridge> yeah, that's my standby if there's no more specific way that alpine is expecting
2020-12-16 00:28:23 <mz41> hi, when i just updated edge it purged ifupdown-ng-openrc, i'm just wondering what the replacement is to get networking on startup
2020-12-16 00:39:52 <TBK[m]> mz41: look for "2020-12-15 11:38:44 <nmeum> why was the networking services dropped from ifupdown-ng?" in https://irclogs.alpinelinux.org/%23alpine-devel-2020-12.log
2020-12-16 00:44:28 <mz41> thanks. running apk fix openrc fixed it. strange it doesn't turn up when i search /etc/init.d/networking on the package finder https://pkgs.alpinelinux.org/contents?file=&path=%2Fetc%2Finit.d%2Fnetworking&name=&branch=edge
2020-12-16 00:48:53 <TBK[m]> mz41: "networking" is the initd file, not a path - https://pkgs.alpinelinux.org/contents?file=networking&path=&name=openrc&branch=edge
2020-12-16 11:21:28 <[diablo]> Good afternoon #alpine-linux , could anyone pass comment on https://paste.centos.org/view/40a6ff4f please?
2020-12-16 11:23:51 <MY-R> [diablo]: is just warning, "done" at the end means that everything is ok
2020-12-16 11:23:59 <[diablo]> hi MY-R
2020-12-16 11:24:07 <[diablo]> ah ok, nothing to worry about then...
2020-12-16 11:24:11 <MY-R> yep
2020-12-16 11:24:22 <[diablo]> no way of surpressing it?
2020-12-16 11:25:47 <[diablo]> will brb, just rebooting... my irc client runs within Alpine...
2020-12-16 11:25:53 <[diablo]> and thank you MY-R
2020-12-16 11:30:34 <MY-R> [diablo]: many tried to silence those "warnings" but in general on bug reports answers is "just ignore it, is harmless"
2020-12-16 11:30:45 <[diablo]> cool
2020-12-16 14:13:58 <abenz_> hey, anyone using libvirtd (virt-manager) ? I cant seem to restart the service (because mine got stuck), it just hangs at "Stopping libvirtd ..."
2020-12-16 14:14:02 <abenz_> is there any trick?
2020-12-16 14:14:22 <ddeimeke> Hi! I need a helping hand switching the kernel from 4.9.65 to 5.4.83. Maybe I am just to dump to find any documentation on the procedure. Any support is highly appreciated. Thank you.
2020-12-16 14:15:31 <ikke> ddeimeke: What kind of setup do you have?
2020-12-16 14:16:07 <ikke> disk setup or run from ram?
2020-12-16 14:16:19 <ddeimeke> ikke: I installed Alpine from an iso some years ago, most probably it was the standard iso.
2020-12-16 14:16:48 <ikke> But you don't need this to boot, right? You have everyting installed on disk?
2020-12-16 14:17:40 <ddeimeke> ikke: No, the system (a VirtualBox VM) is running fine ever since. No attached cdrom, everything is on disk.
2020-12-16 14:18:19 <ikke> what does `cat /etc/alpine-release` and `uname -a` return?
2020-12-16 14:19:00 <ddeimeke> ikke: 3.12.3 and "Linux alpine 4.9.65-1-hardened #2-Alpine SMP Mon Nov 27 15:36:10 GMT 2017 x86_64 Linux"
2020-12-16 14:19:14 <ikke> ah
2020-12-16 14:19:37 <ikke> apk upgrade --available
2020-12-16 14:19:42 <ddeimeke> ikke: Last update: "(1/1) Upgrading linux-lts (5.4.83-r0 -> 5.4.84-r0)"
2020-12-16 14:19:58 <ddeimeke> So the lts-kernel is installed but not the default one
2020-12-16 14:20:11 <ikke> the lts kernel is the default one
2020-12-16 14:20:28 <ikke> but you are running the hardened kernel, which is an older one
2020-12-16 14:20:36 <ddeimeke> :-)
2020-12-16 14:20:40 <ikke> It got renamed
2020-12-16 14:20:48 <ikke> first to vanilla, and now to lts
2020-12-16 14:20:50 <wener[m]> dose alpinelinux v3.12 has rbd module ?
2020-12-16 14:20:53 <ddeimeke> Ok, I rephrase, I like to switch to the lts kernel
2020-12-16 14:21:08 <ikke> apk del linux-hardened
2020-12-16 14:21:18 <wener[m]> modprobe rbd return Module rbd not found in directory
2020-12-16 14:22:26 <ddeimeke> ikke: This is too easy :-) - "Linux alpine 5.4.84-0-lts #1-Alpine SMP Wed, 16 Dec 2020 11:01:26 UTC x86_64 Linux"
2020-12-16 14:22:42 <Hello71> https://git.alpinelinux.org/aports/tree/main/linux-lts/config-lts.x86_64
2020-12-16 14:22:50 <ddeimeke> ikke: Thanks a lot for your support. I have no idea where the hardened kernel came from.
2020-12-16 14:22:56 <Hello71> # CONFIG_BLK_DEV_RBD is not set
2020-12-16 14:23:01 <ikke> ddeimeke: it was still there from when you installed alpine
2020-12-16 14:23:32 <ddeimeke> ikke: But now it is the "correct" one. Thank you.
2020-12-16 14:23:57 <Hello71> wener[m]: i think you can use qemu-nbd if not concerned about performance
2020-12-16 14:24:08 <wener[m]> should alpine include the rbd module ? make CONFIG_BLK_DEV_RBD=m ?
2020-12-16 14:24:51 <wener[m]> yes, but this caused by someone else who use rook ceph in alpinelinux
2020-12-16 14:25:18 <wener[m]> They don't know how to handle this in k8s
2020-12-16 14:36:21 <ggg> I would like to know when will Realtek 8125B be supported in alpinelinux?
2020-12-16 14:36:33 <ggg> Hihi!
2020-12-16 14:36:36 <ggg> I would like to know when will Realtek 8125B be supported in alpinelinux?
2020-12-16 14:41:16 <wener[m]> emmmmm, all virt's rbd is enabled, but all lts rbd is not set ~
2020-12-16 14:43:40 <ikke> ggg: depends on whether it's upstreamed in linux or not
2020-12-16 14:44:21 <ikke> ggg: I don
2020-12-16 14:44:30 <ikke> I don't see it in the linux tree
2020-12-16 14:46:21 <ggg> @ikke
2020-12-16 14:46:25 <ggg> ikke Please help me with exactly what this means. I'm a pretty basic user. I remember it was already supported from the 5.8 or 5.9 kernel.
2020-12-16 14:47:03 <wener[m]> v3.12 is 5.4
2020-12-16 14:47:06 <wener[m]> edge is 5.10
2020-12-16 14:47:08 <ikke> ggg: I'm looking here https://github.com/torvalds/linux/tree/master/drivers/
2020-12-16 14:49:37 <ggg> ikke Can't I find it there, any other option?
2020-12-16 14:50:27 <ggg> wener[m] Can I somehow use the edge under extended?
2020-12-16 14:50:59 <ggg> wener[m] Can i somehow use the edge under extended?
2020-12-16 14:51:02 <ikke> I don't think using the kernel from edge will help
2020-12-16 14:52:13 <ggg> ikke I found such a description, but honestly it’s very high for me. https://www.fedux.org/articles/2020/10/05/setup-alpine-linux-diskless.html
2020-12-16 14:53:30 <ikke> 404
2020-12-16 14:53:49 <mps> ikke: drivers/net/ethernet/realtek/r8169_main.c 'static const struct ephy_info e_info_8125b[]'
2020-12-16 14:55:27 <ikke> ah, 8169 is enabled
2020-12-16 14:55:33 <mps> yes
2020-12-16 14:55:57 <ikke> about one month ago
2020-12-16 14:57:18 <mps> I don't know is it enabled in new -lts
2020-12-16 14:57:45 <Hello71> isn't r8169 very old?
2020-12-16 14:57:53 <ikke> mps: yes
2020-12-16 14:58:18 <ikke> oh, wait, novermber 2019
2020-12-16 14:58:24 <ikke> so more then one year ago
2020-12-16 14:58:29 <ggg> I pay close attention and cheer ikke and mps :)
2020-12-16 14:58:48 <tehcloud> I asked this yesterday, but I'll ask again
2020-12-16 14:59:01 <tehcloud> anyone in here having regressions with kernel 5.10 using an AMD GPU?
2020-12-16 14:59:17 <Hello71> i saw your question, and didn't answer because your question is excessively vague
2020-12-16 14:59:18 <tehcloud> just taking a quick poll before I dig deeper for the root cause
2020-12-16 14:59:28 <Hello71> what problems are you having, if any
2020-12-16 14:59:41 <ikke> ggg: it should be available on 3.12
2020-12-16 14:59:42 <tehcloud> can't keep X running for longer than a minute until graphical corruption and hard reset
2020-12-16 14:59:55 <Hello71> did you try netconsole
2020-12-16 14:59:56 <tehcloud> modesetting driver lasts a little longer than amdgpu
2020-12-16 15:00:45 <ggg> ikke I tried it first but couldn't find a network card by default. :(
2020-12-16 15:01:24 <tehcloud> might try netconsole when I have time to mess with this again, for now it's 5.9 for me
2020-12-16 15:01:45 <ggg> ikke Should I do something or just start the Extended iso?
2020-12-16 15:02:40 <mps> ggg: modinfo r8169 | grep 8125
2020-12-16 15:03:07 <mps> and does lspci list ethernet card
2020-12-16 15:09:58 <MY-R> 8169 need to be reloaded or need do modprobe realtek or something like that
2020-12-16 15:10:26 <MY-R> I got same when install Alpine, but forgot what was exactly doing
2020-12-16 15:15:10 <MY-R> ggg: can you try do "modprobe realtek; modprobe -r r8169; modprobe r8169"
2020-12-16 15:23:51 <szgy> # chattr +i /home/username/.weechat/*.conf
2020-12-16 15:27:06 <ggg> mps I tried and got the following results. https://ibb.co/PM87pCC
2020-12-16 15:28:16 <ggg> MY-R I tried this too, but unfortunately it didn't work either.
2020-12-16 15:31:31 <MY-R> ggg: 10ec:8125
2020-12-16 15:31:36 <mps> ggg: uhm, this is output from busybox lspci, not much helpful
2020-12-16 15:31:59 <mps> MY-R: thanks
2020-12-16 15:32:17 <MY-R> mps: hmm?
2020-12-16 15:32:24 <ggg> mps MY-R I do it with a diskless setup
2020-12-16 15:32:25 <MY-R> so it is newer NIC 2.5 gbit
2020-12-16 15:33:01 <mps> maybe module is not included in initramfs or modloop
2020-12-16 15:33:11 <mps> no, it is
2020-12-16 15:33:14 <mps> huh
2020-12-16 15:33:29 <MY-R> what is exact module name to reload it?
2020-12-16 15:33:49 <mps> r8169
2020-12-16 15:33:51 <MY-R> like I said, got same with different realtek nic
2020-12-16 15:34:10 <ggg> What you asked: modprobe realtek; modprobe -r r8169; modprobe r8169
2020-12-16 15:34:49 <mps> 'modprope -r r8169' should work but I don't know for sure because I don't have such cards to test
2020-12-16 15:35:10 <MY-R> try "modprobe -r r8169; modprobe realtek; modprobe r8169"
2020-12-16 15:35:57 <MY-R> and then "ifup eth0"
2020-12-16 15:36:14 <MY-R> ggg: ^
2020-12-16 15:37:13 <ggg> MY-R I tried: ifup: ignoring unknown interface eth0
2020-12-16 15:38:37 <MY-R> :\
2020-12-16 15:39:09 <ggg> MY-R and i found the next in dmesg: r8169 0000:02:00.0: unknown chip XID 641
2020-12-16 15:40:36 <mps> ggg: version of your kernel?
2020-12-16 15:40:57 <MY-R> he using iso so 5.4
2020-12-16 15:41:17 <mps> ah, then no luck, afaik
2020-12-16 15:41:34 <ggg> 5.4.83-0-lts :(
2020-12-16 15:41:36 <MY-R> he got B version which is in 5.9+ ?
2020-12-16 15:41:55 <mps> yes, iirc from head
2020-12-16 15:43:05 <ggg> MY-R mps I downloaded it from the official site and uploaded it to USB. Unfortunately, I have nothing else.
2020-12-16 15:43:53 <mps> blame realtek why they didn't upstreamed driver addition earlier
2020-12-16 15:44:50 <MY-R> https://linuxreviews.org/Realtek_RTL_8125
2020-12-16 15:46:36 <ggg> MY-R I see :( . Is there anything I can do to make it work in diskless mode?
2020-12-16 15:47:11 <mps> yes, attach usb ethernet or wifi
2020-12-16 15:47:22 <MY-R> ggg: wait for Alpine 3.13 which will have kernel 5.10 or manualy create image based on Edge
2020-12-16 15:47:26 <mps> one which is supported
2020-12-16 15:48:10 <ggg> MY-R And when will 3.13 come out?
2020-12-16 15:48:30 <MY-R> mps is good with predicting dates :>
2020-12-16 15:48:41 <mps> :)
2020-12-16 15:49:16 <MY-R> mps: any chance 3.13.0 will be before new year? :D
2020-12-16 15:49:29 <mps> I doubt
2020-12-16 15:50:04 <mps> but before next equinox, I predict ;)
2020-12-16 15:50:56 <mps> seriously, we are waiting for dalias to release stable musl
2020-12-16 15:51:58 <ggg> Thanks and so in order of magnitude? Or how can I get another kernel to go in diskless mode?
2020-12-16 15:52:07 <MY-R> oh ye "musl" issues :\
2020-12-16 15:55:17 <mps> ggg: as I told above, attach usb ethernet or wifi which works
2020-12-16 15:57:04 <ggg> mps it won't work very much because it should be a router and so should both NICs.
2020-12-16 15:58:05 <mps> after you install you can upgrade to new kernel and use it without external adapters
2020-12-16 15:59:42 <ggg> mps This sounds good. Can I have a link to this? And will it work in diskless mode?
2020-12-16 16:01:14 <mps> ah diskless, not sure it is easy to setup
2020-12-16 16:06:08 <ggg> mps Yes, i thought :(
2020-12-16 16:08:35 <ggg> Sorry if I seem confused, but I'm really a beginner here yet. Do I understand that then before 3.13 I have no chance to use these cards in diskless mode?
2020-12-16 16:11:35 <COVID-1984> anyone using alpine on a VPS with FS encryption?
2020-12-16 16:12:10 <COVID-1984> Wondering about the best install method... maybe a terraform? maybe taking the VPS' stock alpine image and then adding the FS encryption? Thoughts?
2020-12-16 16:13:46 <dysfun> i'm trying to build a custom kernel and getting this from abuild: ERROR: No such package: .makedepends-linux-lts
2020-12-16 16:14:18 <ikke> dysfun: that usually means something went wrong with installing the dependencies (and now it tries to remove them again)
2020-12-16 16:14:34 <dysfun> if i don't supply -r it complains about ccache missing
2020-12-16 16:14:46 <dysfun> (it's installed)
2020-12-16 16:14:54 <ikke> dysfun: apk fix
2020-12-16 16:15:35 <ialokin> ikke, are you aware that your nick means "don't" in norwegian?
2020-12-16 16:15:47 <ikke> yes
2020-12-16 16:15:57 <dysfun> i get an error about ttf-linux-libertine in apk fix
2020-12-16 16:16:06 <dysfun> not sure why it would depend on that
2020-12-16 16:16:26 <ikke> It's unrelated
2020-12-16 16:16:38 <ikke> It just prevents dependencies from getting installed
2020-12-16 16:16:54 <ikke> What's the error you get?
2020-12-16 16:16:55 <dysfun> okay, it's executing two triggers: fontconfig and mkfontscale and it doesn't say what the error is
2020-12-16 16:17:41 <dysfun> i just uninstalled it
2020-12-16 16:17:57 <dysfun> ah, now we're getting somewhere
2020-12-16 16:18:01 <dysfun> thanks
2020-12-16 16:18:03 <ikke> np
2020-12-16 16:23:37 <mps> ggg: there are chances ofc, but you have to do that 'by hand'
2020-12-16 16:29:17 <mps> and 'ikke' in Dutch ;)
2020-12-16 16:36:16 <ggg> mps And can you help me? :)
2020-12-16 16:37:19 <mps> ggg: my help can be only one sentence: install in sys mode
2020-12-16 16:39:54 <ggg> mps Thank you very much and have nice day ;)
2020-12-16 16:41:12 <mps> I run routers in sys mode with fs on mmc cards for years without problems, except usual problems with computers
2020-12-16 16:41:51 <mps> ggg: thank you, same to you :)
2020-12-16 16:42:38 <ggg> mps And what exactly does that mean? Does it boot from the SD card and then run from RAM?
2020-12-16 16:44:05 <ikke> No, it runs from some persistent media
2020-12-16 16:44:26 <mps> no, boot from sdcard and root fs is on card, and also data partitions
2020-12-16 16:44:52 <mps> like all normal servers work
2020-12-16 16:45:15 <ikke> ggg: with diskless installs, you can only replace the kernel by updating the boot medium
2020-12-16 16:45:19 <mps> a lot easier and safe to upgrade and maintain
2020-12-16 16:46:07 <dysfun> 
ACTION hopes you RAID over SD cards

2020-12-16 16:46:36 <ikke> if there is not a lot of writes, then they should last quite a bit
2020-12-16 16:47:37 <mps> my mail server is on mmc more than 4 years
2020-12-16 16:48:02 <mps> never had problem
2020-12-16 16:48:10 <dysfun> is there an easy way to change the default boot line you get generated for you? it keeps putting nomodeset there and i have x using modesetting
2020-12-16 16:48:53 <mps> dysfun: yes
2020-12-16 16:48:57 <ggg> Really? My USB and SD cards always die.
2020-12-16 16:49:34 <mps> dysfun: /etc/update-extlinux.conf
2020-12-16 16:49:39 <dysfun> ah, ta
2020-12-16 16:50:29 <mps> dysfun: for syslinux, if you use grub it in /etc/grub/default or somewhere like this
2020-12-16 16:50:44 <dysfun> i just use syslinux cause it was the default
2020-12-16 16:52:29 <Nisargadatta> anyone here use palemoon ?
2020-12-16 16:55:00 <Nisargadatta> I'm reading that it use one thread, and consider if it can be good for old cpu
2020-12-16 17:03:47 <ggg> Thanks again, mates! mps and MY-R and ikke !!! Bye!
2020-12-16 17:05:54 <MY-R> ggg: no problem, see you in 3.13 I guess! :)
2020-12-16 17:23:41 <alu20> hi how often do packages on the non edge release get updated?
2020-12-16 17:24:02 <ikke> alu20: mostly when there are bug / security fixes
2020-12-16 17:24:43 <alu20> oh ok, otherwise just wait for new release/use edge?
2020-12-16 17:26:20 <TBK[m]> backports can also be requested
2020-12-16 17:31:48 <alu20> thanks. let me check with the maintainer of a couple things.
2020-12-16 17:51:34 <guix-vits> s390x is listed twice on https://wiki.alpinelinux.org/wiki/Alpine_Linux:Overview#Technical_overview
2020-12-16 18:19:22 <sa0> Hello
2020-12-16 18:19:42 <sa0> Someone available to help with an Xorg issue ?
2020-12-16 18:20:30 <sa0> Trying the distro as VBox guest but unable to get X11 working (I wanted to test only sddm + bspwm) ...
2020-12-16 18:23:08 <TBK[m]> guix-vits: thanks for reporting
2020-12-16 18:37:38 <guix-vits> sa0: do sddm starts?
2020-12-16 19:21:46 <Hello71> sa0: remove nomodeset from kernel command line
2020-12-16 19:48:51 <sa0> OK thanks, trying
2020-12-16 19:54:20 <sa0> indeed removing nomodeset from cmdline seems to go further ... at least Xorg starts without blaming on drm not found like before ...
2020-12-16 19:54:27 <sa0> thX Hello71
2020-12-16 20:03:37 <Hello71> also consider dropping vbox
2020-12-16 20:05:08 <sa0> what you mean ?
2020-12-16 20:05:34 <Hello71> virtualbox is not good
2020-12-16 20:06:53 <sa0> sure, it is only to test Alpine before going further
2020-12-16 20:15:29 <sa0> strange, after reboot, xorg doesn't start anymoie
2020-12-16 20:15:34 <sa0> *anymore
2020-12-16 20:18:21 <j33s> I seem to have borked my alpine install. i had a 3.12 install and accidentally set my repo to edge and ran an update. then cancelled the update halfway through
2020-12-16 20:18:51 <j33s> not i get this when i try and install go: https://trash.j3s.sh/alpine-troubleshooting.txt
2020-12-16 20:18:59 <j33s> any idea how to revert to 3.12 from edge?
2020-12-16 20:19:25 <ikke> j33s: what does your /etc/apk/repositories look like?
2020-12-16 20:20:45 <j33s> https://trash.j3s.sh/etc-apk-repos.txt
2020-12-16 20:21:10 <j33s> I didn't have the @edge tag previously, so my upgrade started installing the newest versions of musl and such :(
2020-12-16 20:21:50 <ikke> try apk upgrade --available
2020-12-16 20:21:54 <j33s> for example, here's musl: https://trash.j3s.sh/musl-stable.txt it's on the edge version
2020-12-16 20:22:13 <j33s> apk upgrade --available results in OK: 275 MiB in 105 packages
2020-12-16 20:24:42 <ikke> I think easiest for now is to temporarily remove the pinned repositories, and run apk upgrade --update --available
2020-12-16 20:28:13 <j33s> okay gimme a sec
2020-12-16 20:28:29 <Hello71> sa0: at a guess, it is vbox bug. make sure you have set vboxsvga, then power off and power on
2020-12-16 20:28:44 <sa0> ok
2020-12-16 20:31:32 <j33s> ikke that worked like a charm! ty!!
2020-12-16 20:36:58 <sa0> following the wiki to install, after having created .cshrc file, it seems the directive unsetenv does not work with ash ... is it normal ?
2020-12-16 21:29:02 <mps> sa0: .cshrc is for csh or tcsh not for ash
2020-12-16 21:29:16 <sa0> ok
2020-12-16 22:54:11 <bitdefect> Hi there, I've seen that the release cycle for major versions is roughly 6 months. Are there plans for 3.13 already?
2020-12-16 23:04:37 <TBK[m]> bitdefect: yes, we are working on it.
2020-12-16 23:07:04 <bitdefect> TBK[m]: I'm asking, since I'd love to see a version of suricata in stable, which supports hyperscan.
2020-12-16 23:07:14 <bitdefect> TBK[m]: Is there anything one could do to help?
2020-12-16 23:09:13 <TBK[m]> bitdefect: One could provide MRs (gitlab.alpinelinux.org) for aports that fails to build :) - #alpine-commits | https://build.alpinelinux.org
2020-12-16 23:58:38 <bz2S> i had apk fail during an upgrade of the kernel due to kernel faults. system was in a broken state
2020-12-16 23:59:04 <bz2S> i got backup and running by restoring an old kernel and init in boot as well as modules to lib
2020-12-16 23:59:19 <bz2S> but apk upgrade thinks it completed successfully... any ways to force it to reupgrade the kernel?
2020-12-17 00:00:59 <bz2S> technically, what failed was the initramfs. but this was due to kernel faults before i ran the upgrade
2020-12-17 00:01:40 <bz2S> uh, so i think i need to rerun the initramfs generation. lets try that
2020-12-17 00:30:44 <hendursaga> Anyone know of any SoundFont packages for Alpine? Like for FluidSynth, etc. I can't find the equivalent of, say, Arch's freepats-general-midi or soundfont-fluid.
2020-12-17 00:37:28 <Hello71> https://pkgs.alpinelinux.org/contents?file=*.sf2&path=&name=&branch=edge
2020-12-17 00:39:07 <bz2S> that worked. so if you have kernel faults, don't apk upgrade or it may fail halfway
2020-12-17 01:00:52 <hendursaga> Hello71: Thanks but I was looking more for orchestral/piano SoundFonts. I don't see any packaged but that's OK I guess.
2020-12-17 01:02:54 <TBK[m]> hendursaga: MR are always welcome (gitlab.alpinelinux.org/alpine/aports). It is normally pretty easy to port a Arch PKGBUILD file to APKBUILD.
2020-12-17 01:04:51 <hendursaga> TBK[m]: I'll look into it. First I'll have to find a good SF
2020-12-17 01:51:30 <guix-vits> https://wiki.alpinelinux.org/wiki/Pine64_A64_LTS : should link "download the generic ARM image" point to latest-stable instead?
2020-12-17 03:06:11 <TBK[m]> guix-vits: I think that would be a good idea.
2020-12-17 06:58:36 <midao> hi :) I'm still figuring this irc thing out so let me know if I do anything wrong here.
2020-12-17 07:02:56 <midao> I have a pretty wierd bug and am a bit lost so I thought I'd ask for help here. I'm on Alpine edge (with testing enabled, but from my understanding this should not matter) and basically updating the system every other day. As a window manager, I'm using dwm and since yesterday morning, it does not refresh the displayed windows unless I move the
2020-12-17 07:02:57 <midao> mouse. Sometimes even that does not help, but most of the times it works that way. I did recompile dwm before that but without changing anything (and the default build without my config and patches has the same problem) so I'm not quite sure what to do here. I already checked the Xorg log but could not find anything and dwm itself does not produce
2020-12-17 07:02:57 <midao> any error messages as well. Any help would be appreciated :)
2020-12-17 07:04:00 <midao> I also tried to get a list of recently updated packages to try and find out whether something changed there but I could not figure out how to do this with apk
2020-12-17 07:09:09 <MY-R> midao: not sure what is going on but since you are on edge, did you have any not upgraded packages on? check it with: apk update; apk version
2020-12-17 07:11:37 <midao> no it does not show anything, I also installed a linux-lts update afterwards in the hope that it would fix it (unlikely, but I was desperate ^^) and this also did not help. basically, all packages installed are on the newest version
2020-12-17 07:12:21 <MY-R> midao: and how did you install dwm? from package or compiling manualy?
2020-12-17 07:14:16 <MY-R> if you compiling and didnt install dwm from repo then be sure that got basic dependencies like: "libxft-dev libx11-dev libxinerama-dev"
2020-12-17 07:14:56 <MY-R> so ye if all that not helping then have to figure out which package upgrade started cause the problem with dwm :\
2020-12-17 07:36:23 <midao> @MY-R i installed from source, so the suckless git repo and have all the dependencies installed. it also worked for two weeks without any problems but I'm not sure whether my recompilation triggered something or whether something in the alpine environment changed. I also just skimmed through the past few days of changelog in the alpine aports repo
2020-12-17 07:36:24 <midao> but nothing caught my eye there either
2020-12-17 07:38:04 <midao> I'm currently still at work but I will later try to install dwm in a live image and see whether I have the same problem there
2020-12-17 07:39:34 <MY-R> midao: damn :\
2020-12-17 07:40:09 <MY-R> midao: did you try build last stable release of dwm or master?
2020-12-17 07:40:23 <midao> yeah tried master as well
2020-12-17 07:40:38 <midao> and both clang and gcc ^^
2020-12-17 07:41:38 <MY-R> if it isnt something specific to Alpine then some bug reports should appear around in other distros
2020-12-17 07:41:59 <midao> but as I said, I did not change anything compared to the build that was working, I just recompiled after playing around with the gcc and clang packages
2020-12-17 07:42:23 <MY-R> ye and edge is moving fast
2020-12-17 07:46:12 <midao> true ^^ I'm currently reinstalling (almost) all packages that I installed manually, maybe that helps... I created a text file with all of them so I can easily install them on a new setup
2020-12-17 07:48:05 <midao> that also did not work :(
2020-12-17 07:50:07 <MY-R> most probably some lately upgraded package doesnt like dwm
2020-12-17 07:54:54 <midao> I sadly don't exactly know anymore when I last compiled the previous working version, it was somewhere around the first week of December
2020-12-17 08:28:52 <guix-vits> midao:  idk the matter.  did u recompiled dwm itself after updates?
2020-12-17 08:41:39 <fridim> hi.   wiki is down
2020-12-17 08:41:57 <ikke> not for me
2020-12-17 08:42:13 <ikke> https://wiki.alpinelinux.org/ is working for me
2020-12-17 08:43:16 <fridim> https://imgur.com/a/ScrYEI7   ok weird
2020-12-17 08:44:06 <ikke> fridim: what IP does wiki.alpinelinux.org resolve to for you?
2020-12-17 08:44:20 <fridim> 147.75.101.119
2020-12-17 08:44:28 <ikke> ok, that's correct
2020-12-17 08:45:05 <fridim> OK, it's just firefox
2020-12-17 08:45:10 <fridim> nevermind :)
2020-12-17 08:48:19 <midao> @guix-vits yes I recompiled two days ago after the previous install from around the first week of december (but not changing anything in dwm)
2020-12-17 08:51:17 <gjabell> did some of alpine's networking stack change in the last couple weeks? I updated yesterday (on edge) and suddently I have no /etc/init.d/networking
2020-12-17 08:52:33 <ikke> apk fix busybox
2020-12-17 08:53:10 <ikke> gjabell: there was a separate ifupdown-ng-openrc pacakge, but that gave issues and is now unnecessary, so it was removed again
2020-12-17 08:54:01 <gjabell> hmm, still not seeing it in init.d after apk fix busybox, does it reappear on a reboot?
2020-12-17 08:56:13 <TBK[m]> apk fix openrc
2020-12-17 08:57:41 <gjabell> that did it, thanks :)
2020-12-17 08:57:53 <ikke> TBK[m]: thanks, forgot what the correct package was :)
2020-12-17 08:57:59 <gjabell> is there a way to see all the broken packages? plain apk fix doesn't seem to fix everything
2020-12-17 09:00:49 <midao> gjabell: i assume fix does not do anything for non-broken packages so this should work (but might take some time): for p in $(apk list --installed | awk '{ print $1 }'); do apk fix $p; done
2020-12-17 09:01:09 <midao> you might have to add a sudo/doas
2020-12-17 09:02:35 <gjabell> hmm yeah
2020-12-17 09:02:44 <gjabell> I think it'd need to cut off the versioning info too
2020-12-17 09:02:55 <ikke> also apk audit --system --packages
2020-12-17 09:03:42 <ikke> though, that does not seem to catch this
2020-12-17 09:05:07 <midao> is the version information always the last two parts of the package name? then a tr | awk | tr afterwards could work (or some sed regex but i'm not good with sed)
2020-12-17 09:08:50 <midao> apk list --installed | awk '{ print $1 }' | tr '-' ' ' | awk 'NF{NF-=2};1' | tr ' ' '-'
2020-12-17 09:09:09 <midao> ^ but only if the last two columns are always the version
2020-12-17 09:11:27 <midao> apparently this reinstalls everything so be careful with it
2020-12-17 09:11:42 <midao> maybe it fixes my bug
2020-12-17 09:11:42 <ikke> yes, that's what apk fix <pkgname> does
2020-12-17 09:12:10 <midao> ah I did not know that was the default, thanks
2020-12-17 09:14:13 <guix-vits> if `fix` accepts (idk) many arguments: source manifest.sh;  apk fix $manifest
2020-12-17 09:17:19 <guix-vits> but how to compose it right?  some packages are installed by default.  ok.
2020-12-17 09:52:31 <abenz_> ah firefox 83 got stuck for the first time! just when I was starting to get comfortable thinking mozilla fixed it!
2020-12-17 10:04:04 <mr-robot> Hi Everyone! Does it possible to install alpine linux offline mode?
2020-12-17 10:30:06 <mr-robot> Hi Everyone! Does it possible to install alpine linux offline mode?
2020-12-17 10:39:40 <ikke> mr-robot: I don't think we have an image that contains everything for an offline install
2020-12-17 10:57:31 <guix-vits> > apk fix openrc
2020-12-17 10:57:31 <guix-vits> was that recent trouble with 'networking' file happened due to partial updates?
2020-12-17 12:43:13 <lupo_on_freenode> I'm trying to install alpine-standard-3.12.3 on an older laptop. Problem: setup-alpine doesn't find any network device. "ifconfig -a" shows eth0, but "ls /dev" shows no eth*. Suggestions?
2020-12-17 12:47:44 <MY-R> lupo_on_freenode: paste output of "lspci -k"
2020-12-17 12:48:42 <MY-R> you wont find eth devices in /dev/ but if it showing up in ifconfig then it have to exist
2020-12-17 12:50:27 <MY-R> maybe it is realtek driver which need reload or load and then will work
2020-12-17 13:03:56 <lupo_on_freenode> MY-R, No net -- no pastebin =:) This may be the important line: "02:00.0 Class 0200: 10ec:8168 r8169"
2020-12-17 13:06:04 <lupo_on_freenode> lsmod shows "libphy 94208 2 realtek,r8169"
2020-12-17 13:06:48 <MY-R> lupo_on_freenode: ye got same card and same issue but forgot what exactly done, but will try what I remember
2020-12-17 13:08:19 <MY-R> lupo_on_freenode: do modprobe -r r8169; modprobe -r realtek; modprobe r8169
2020-12-17 13:08:39 <MY-R> and then try: ifup eth0
2020-12-17 13:14:30 <lupo_on_freenode> MY-R, I had to make an appropriate "/etc/network/interfaces", the "ifup eth0" succeeded. Is there a way to automate that?
2020-12-17 13:14:50 <lupo_on_freenode> s/the /then/
2020-12-17 13:15:18 <MY-R> lupo_on_freenode: during installation process?
2020-12-17 13:16:20 <MY-R> that file is screated during "setup-alpine" (setup-interfaces)
2020-12-17 13:18:46 <lupo_on_freenode> MY-R, as far as I see, setup-interaces does _not_ create an "/etc/network/interfaces". Could that be?
2020-12-17 13:19:04 <lupo_on_freenode> ^setup-interaces^setup-interfaces
2020-12-17 13:27:48 <mps> MY-R: isn't that solved last night :)
2020-12-17 13:28:20 <MY-R> mps: nah, his driver working just fine :)
2020-12-17 13:28:44 <MY-R> lupo_on_freenode: Im not sure when you already making changes
2020-12-17 13:32:29 <lupo_on_freenode> MY-R, I did a reboot meanwhile to start clean again, but "setup-interfaces" did not create an "/etc/network/interfaces".
2020-12-17 13:38:11 <Hello71> lupo_on_freenode: did you verify your image
2020-12-17 13:39:14 <lupo_on_freenode> Hello71, I did.
2020-12-17 13:40:28 <ikke>  /\\\
2020-12-17 13:42:50 <Hello71> reboot, ip a, sh -x setup-interfaces 2>&1 | tee setup-interfaces.log
2020-12-17 13:43:06 <Hello71> then either connect manually or save it to another drive for us
2020-12-17 13:45:09 <MY-R> lupo_on_freenode: reboot to what, live Alpine image? or to already installed Alpine (sys mode)?
2020-12-17 13:53:25 <lupo_on_freenode> MY-R, live image "alpine-standard-3.12.3-x86_64.iso".
2020-12-17 13:56:19 <lupo_on_freenode> Hello71, "ip a" => https://paste2.org/HG0UYAJm
2020-12-17 13:57:12 <MY-R> lupo_on_freenode: AFTER installation you will have on HARD DISK that file
2020-12-17 13:57:36 <MY-R> lupo_on_freenode: do ifup eth0 and if wont show any error then install Alpine
2020-12-17 13:58:18 <lupo_on_freenode> Hello71, "sh -x /sbin/setup-interfaces 2>&1 | tee setup-interfaces.log" => https://paste2.org/h0FKgh58
2020-12-17 13:58:41 <MY-R> you can run setup-interfaces to be sure that your network is up and running if you want but that should already cover setup-alpine
2020-12-17 13:58:49 <Hello71> lupo_on_freenode: seems working fine?
2020-12-17 14:00:59 <COVID-1984> anyone using alpine on a VPS with FS encryption? Wondering about the best install method... maybe a terraform? maybe taking the VPS' stock alpine image and then adding the FS encryption? Thoughts?
2020-12-17 14:03:02 <lupo_on_freenode> MY-R, OK. But still (in this case) setup-alpine (setup-interfaces) does not succeed in establishing a network connection, which enforces to create an "/etc/network/interfaces" manually.
2020-12-17 14:03:31 <guix-vits> 
ACTION aaaa covid

2020-12-17 14:05:30 <MY-R> lupo_on_freenode: if you already testing stuff then reboot, do ifup eth0 and then run setup-interfaces
2020-12-17 14:08:56 <lupo_on_freenode> MY-R, using the live image?
2020-12-17 14:12:00 <MY-R> lupo_on_freenode: ye, that is your issue with live image, right?
2020-12-17 14:13:05 <MY-R> after when Alpine is on disk it doesnt matter if you will have that file or not because can just create by sel for by running "alpine-setup" again and skip partitioning
2020-12-17 14:14:46 <lupo_on_freenode> Hello71, at least eth0 gets no dhcp lease ...
2020-12-17 14:15:04 <Hello71> but did you configure it
2020-12-17 14:45:47 <lupo_on_freenode> It seems I'm through it now. Because of installing on a laptop, there are two network devices setup-alpine offers to configure: eth0 and wlan0. I had success when quitting the "network configuration part" by choosing "done" when being asked whether "wlan0" should be configured.
2020-12-17 14:49:20 <lupo_on_freenode> When continuing with <enter> instead, as far as I see, the network configuration inevitably ends with configuring the loopback interface twice in "/etc/network/interfaces", which makes setting up the network fail.
2020-12-17 14:59:40 <lupo_on_freenode> I tried again with an USB network adapter which gets eth1 -- Success when answering "done" for being asked whether to configure wlan0, fail continuing at that point. It catches my eye that (only) in the latter case I'm being asked whether to configure the loopback interface.
2020-12-17 15:01:02 <wsinatra> lupo_on_freenode: you can choose to manually edit the /etc/network/interfaces file afterwards if needed. Even if it's just to verify things.
2020-12-17 15:01:35 <wsinatra> If you're just running setup-alpine once though it really should only configure the lo and whatever interface you pass it, so it's a bit strange you're seeing two lo configurations get added
2020-12-17 15:02:59 <xerz> Heya! I'm considering installing Alpine on a new computer, and I noticed my network cards aren't compatible with the provided Linux 5.4
2020-12-17 15:03:29 <xerz> Is there any way I can get an edge ISO with an up-to-date kernel?
2020-12-17 15:03:39 <xerz> Thanks in advance!
2020-12-17 15:04:03 <lupo_on_freenode> wsinatra, Of course creating / editing is an option =:) On the other hand I did not expect to be forced to do it after installing alpine under qemu without doing it.
2020-12-17 15:04:52 <lupo_on_freenode> s|creating / editing |& /etc/network/interfaces|
2020-12-17 15:04:54 <Hello71> xerz: you can make your own with aports/scripts/mkimage.sh
2020-12-17 15:05:29 <Hello71> alternatively, 3.13 will probably be out in a week or two
2020-12-17 15:05:50 <ikke> (very optmistic :P)
2020-12-17 15:07:25 <wsinatra> lupo_on_freenode: That's really truly bizarre. I spin alpine images up and down on qemu all day nearly every day at work. Is that the latest ISO you're using?
2020-12-17 15:07:28 <lupo_on_freenode> I have the suspicion that the "logic" of setup-alpine / setup-interfaces does not take account of having more then one network interface present.
2020-12-17 15:07:31 <mps> much too :)
2020-12-17 15:08:16 <xerz> <Hello71 "xerz: you can make your own with"> I need it by now so, guess I'll have to build my own
2020-12-17 15:08:24 <lupo_on_freenode> Perhaps one could test this case by adding an USB network adapter?
2020-12-17 15:08:40 <xerz> How is it that there is no daily builds or anything like that?
2020-12-17 15:08:42 <Hello71> 3.13 will not be out for a week or two
2020-12-17 15:08:45 <lupo_on_freenode> wsinatra, alpine-standard-3.12.3-x86_64.iso
2020-12-17 15:08:53 <Hello71> xerz: arch doesn't either
2020-12-17 15:09:05 <Hello71> you can also install alpine using a non alpine iso
2020-12-17 15:09:15 <xerz> Arch has a recent kernel on its image though
2020-12-17 15:09:31 <Hello71> yes because arch is rolling release. it's still one month old though
2020-12-17 15:09:31 <xerz> Yeah the latter sounds best tbh
2020-12-17 15:09:49 <xerz> Enough for my computer to work, I checked
2020-12-17 15:10:00 <Hello71> https://www.archlinux.org/download/ says 5.9.11
2020-12-17 15:10:06 <xerz> Yep
2020-12-17 15:10:18 <wsinatra> you could compile the kernel yourself, I have to do that for my Droid4, but otherwise it works excellently
2020-12-17 15:10:41 <Hello71> yeah but how do you get it on the image
2020-12-17 15:10:49 <Hello71> you still need to use mkimage.sh
2020-12-17 15:11:03 <Hello71> they don't really need to compile a kernel, just use a newer package
2020-12-17 15:11:11 <wsinatra> lupo_on_freenode: it seems I'm still using 3.12.0 ISOs, let me see if I see the same thing in qemu using the .3 iso
2020-12-17 15:11:41 <Hello71> releases always stay on same major (minor) kernel version
2020-12-17 15:12:03 <Hello71> if point releases jumped kernel version there wouldn't be much point in releases at all, it would just be rolling
2020-12-17 15:14:31 <xerz> OK got the Arch ISO
2020-12-17 15:14:48 <xerz> How do I get `setup-alpine`? I see there's no tutorial on the wiki
2020-12-17 15:15:10 <vulpine> .... why do you have an archiso
2020-12-17 15:15:27 <xerz> Have you seen the last messages?
2020-12-17 15:15:38 <xerz> Basically because the Alpine ISO is too old
2020-12-17 15:15:53 <artok> =D
2020-12-17 15:16:31 <artok> too old to do what?
2020-12-17 15:16:45 <xerz> Install on my B550 Gaming Edge Wifi
2020-12-17 15:17:35 <artok> and that is ?
2020-12-17 15:17:46 <xerz> A year old motherboard
2020-12-17 15:18:07 <xerz> That has a Realtek LAN and an Intel Wifi card
2020-12-17 15:18:23 <xerz> Neither of which are supported on Linux 5.4
2020-12-17 15:19:18 <xerz> So Arch ISO it is
2020-12-17 15:19:45 <xerz> And Arch install if I don't figure out how to get Alpine edge
2020-12-17 15:20:49 <artok> mkimage could work running it on docker container
2020-12-17 15:21:13 <artok> or spin up virtualbox vm with vagrant or...
2020-12-17 15:21:17 <xerz> How long could that take on a Core 2 Duo?
2020-12-17 15:21:43 <artok> no idea, never actually done it
2020-12-17 15:21:57 <xerz> Sounds like a bad plan then tbh
2020-12-17 15:23:29 <xerz> Booted Arch, what next?
2020-12-17 15:23:50 <artok> get apk static
2020-12-17 15:24:41 <xerz> Actuall nvm, it's having trouble with the graphics card for whatever reason
2020-12-17 15:24:48 <xerz> * Actually nvm, it's having trouble with the graphics card for whatever reason
2020-12-17 15:25:47 <xerz> So uhhh, only distro I know works is Mageia 8 Beta
2020-12-17 15:25:50 <xerz> I'm not installing that lmao
2020-12-17 15:26:35 <xerz> Maybe I can use that live ISO for apk?
2020-12-17 15:27:54 <Hello71> add or remove nomodeset from kernel command line
2020-12-17 15:29:07 <xerz> Ah, how do I do that with an ISO with systemd boot?
2020-12-17 15:29:08 <xerz> Holdup
2020-12-17 15:30:30 <lupo_on_freenode> Thanks for help =:)
2020-12-17 15:33:26 <xerz> F1 key, got it
2020-12-17 15:34:29 <Hello71> f1?
2020-12-17 15:34:50 <xerz> Yeah, that shows a menu with an edit entry option
2020-12-17 15:34:56 <xerz> And it worked!
2020-12-17 15:35:31 <xerz> Pings!
2020-12-17 15:35:35 <xerz> Time to apk
2020-12-17 15:44:13 <xerz> Installed, now I need to learn how to bootstrap thia
2020-12-17 15:44:20 <xerz> * Installed, now I need to learn how to bootstrap this
2020-12-17 15:47:16 <xerz> Actually what about this? https://github.com/alpinelinux/alpine-chroot-install
2020-12-17 15:48:35 <Hello71> if you read it, it basically downloads apk-tools-static and runs apk -p . --initdb add alpine-base
2020-12-17 15:49:32 <xerz> Oh thanks, I need to setup repos first tho
2020-12-17 15:52:13 <xerz> https://wiki.alpinelinux.org/wiki/Alpine_Linux_in_a_chroot
2020-12-17 15:52:18 <xerz> Finally, an useful wiki page
2020-12-17 15:56:17 <xerz> Got it, `setup-alpine` works!
2020-12-17 16:13:02 <Hello71> ok :)
2020-12-17 17:08:54 <xerz> OK, it's been a while trying to learn everything, but it boots!
2020-12-17 17:09:05 <xerz> Thanks a lot for the help!
2020-12-17 17:09:15 <xerz> I think I'll make a tutorial on this
2020-12-17 17:09:24 <xerz> btrfs too, neat
2020-12-17 17:09:54 <xerz> …wait
2020-12-17 17:10:22 <xerz> Network doesn't work, but it's Linux 5.10
2020-12-17 17:10:30 <xerz> Why
2020-12-17 17:11:43 <xerz> Missing firmware package maybe?
2020-12-17 17:11:44 <c705> is the cable unplugged
2020-12-17 17:11:55 <xerz> No duh
2020-12-17 17:12:03 <xerz> I only rebooted
2020-12-17 17:12:47 <xerz> It gave a device missing error
2020-12-17 18:00:12 <xerz> Ah, I think I found the issue, I didn't properly setup udev
2020-12-17 18:05:10 <minimal> udev or mdev?
2020-12-17 18:05:29 <minimal> Alpine normally uses mdev
2020-12-17 18:05:42 <xerz> udev, I want to use GNOME
2020-12-17 18:05:53 <xerz> also now boot freezes instead, yay?
2020-12-17 18:06:03 <xerz> And networking ain't solved
2020-12-17 18:06:19 <minimal> I believe you don't need udev for Gnome on Alpine
2020-12-17 18:06:36 <xerz> That's what the wiki says though
2020-12-17 18:07:07 <xerz> The error I get is this btw https://gitlab.alpinelinux.org/alpine/aports/-/issues/9079
2020-12-17 18:07:22 <xerz> New open ticket at the end
2020-12-17 18:09:06 <xerz> `nomodeset` lets me go to the shell at least
2020-12-17 18:11:39 <xerz> I'm also getting cannot start errors "as udev-trigger would not start"
2020-12-17 18:14:11 <xerz> Must be some service startup issue, seems like these work when started manually
2020-12-17 18:16:30 <xerz> `echo “rc_need=udev-settle” >/etc/conf.d/networking` didn't do it
2020-12-17 18:17:43 <xerz> At this point though, it's taking too long and I'm better off with some other distro
2020-12-17 18:31:32 <minimal> I use udev on Alpine (for cloud-init), never seen that error
2020-12-17 18:38:29 <MY-R> I'm wondering if he noticed that “ ” arent " " ...
2020-12-17 18:42:34 <minimal> could be. I've never needed to put a udev reference in the networking conf.d file though
2020-12-17 18:42:44 <MY-R> xerz: check what you really have now in "/etc/conf.d/networking" there should be rc_need=udev-settle and not “rc_need=udev-settle”
2020-12-17 18:46:06 <mps> I use libudev-zero instead of eudev
2020-12-17 18:46:11 <mps> works fine
2020-12-17 19:13:49 <Newbyte> Is there any way to check what packages would be upgraded by an upgrade without performing an upgrade?
2020-12-17 19:14:18 <maxice8> apk version ?
2020-12-17 19:14:27 <caskd> Newbyte: apk upgrade -s
2020-12-17 19:14:35 <caskd> -s stands for simulate
2020-12-17 19:14:40 <Newbyte> 2.12
2020-12-17 19:15:02 <Newbyte> thanks caskd
2020-12-17 19:18:54 <mps> apk upgrade --simulate
2020-12-17 19:21:40 <halosghost> mm, that'd be a cool cli tool: intercepts all syscalls to simulate the result of running a given command without actually doing it all
2020-12-17 19:22:01 <halosghost> would remove the need for such options, and allow for a safe way of trying things on the cli
2020-12-17 19:22:12 <halosghost> I dub thee “dryrun”
2020-12-17 19:22:20 <halosghost> mps: go forth and bring it to fruition ☺
2020-12-17 19:22:57 <mps> halosghost: yes, but I don't have time to play with forth in last years ;)
2020-12-17 19:23:09 <halosghost> lol
2020-12-17 19:23:15 <halosghost> mps: C is preferred anyhow
2020-12-17 19:23:18 <halosghost> please carry on ☺
2020-12-17 19:23:53 <mps> yes, but also my day job is not near C, which miss much too to me
2020-12-17 19:25:38 <xerz> <MY-R "xerz: check what you really have"> I used the right commas thx
2020-12-17 19:25:45 <xerz> Anyway I'm on Arch now
2020-12-17 19:25:58 <xerz> nomodeset might have been to blame for GNOME issues
2020-12-17 19:26:16 <xerz> but it couldn't boot up without it so what was the point
2020-12-17 21:56:46 <lupo_on_freenode> I have installed Alpine with alpine-standard-3.12.3-x86_64.iso, did "setup-xorg-base" with the packages needed (I think), then installed i3wm and terminus-font, added "font pango:Terminus 12px" to the config. Problem: i3 does not use the font, shows boxes with hex codes (?) instead. Any suggestions?
2020-12-18 05:15:11 <guix-vits> it is a make dependency of linux-lts, but the package does not exist:  https://pkgs.alpinelinux.org/packages?name=linux-firmware-any&branch=edge .  please, help.
2020-12-18 05:39:04 <ikke> guix-vits: it's a virtual package
2020-12-18 05:40:04 <ikke> https://gitlab.alpinelinux.org/alpine/aports/-/blob/master/main/linux-firmware/APKBUILD#L76
2020-12-18 05:42:18 <guix-vits> ikke: thank you.
2020-12-18 05:42:52 <ikke> there is even a package called linux-firmare-none that satisfies that dependency
2020-12-18 11:27:57 <pinsetonl> Apk upgrade with new lts kernel bricked my setup.
2020-12-18 11:29:09 <pinsetonl> now boot ends on exit_boot() failed and efi_main() failed
2020-12-18 11:29:30 <pinsetonl> How do I fix?
2020-12-18 11:41:00 <guix-vits> pinsetonl: idk.  try in chroot, in meantime: https://wiki.alpinelinux.org/wiki/Alpine_Linux_package_management#Holding_a_specific_package_back
2020-12-18 11:57:06 <midao> hmm apparently I forgot to pack a USB and cannot try dwm in a live image, can I downgrade from edge to a specific stable release? change the repo URLs, update the apk cache and reinstall everything?
2020-12-18 12:10:34 <ikke> apk upgrade --available
2020-12-18 12:12:16 <midao> ikke: thanks
2020-12-18 12:33:31 <pinsetonl> guix-vits: how do I downgrade kernel from chroot exactly?
2020-12-18 12:33:56 <pinsetonl> 5.10->5.4 for example
2020-12-18 12:35:27 <guix-vits> pinsetonl:  idk, honestly.  apk add linux=5.4~ maybe.
2020-12-18 12:37:28 <caskd> does anyone know if apk supports repos with basic auth?
2020-12-18 12:37:51 <ikke> I don't think so
2020-12-18 12:39:03 <caskd> apk uses libfetch, right?
2020-12-18 12:39:42 <ikke> Yes
2020-12-18 12:40:15 <caskd> hmm, i'm going to test it and hope for the best because it would take less than reading the code tbh
2020-12-18 12:40:21 <guix-vits> pinsetonl: linux=~5.4
2020-12-18 12:41:03 <guix-vits> in single quotes, heh.
2020-12-18 12:45:42 <pinsetonl> linux-lts-5.10.1-r0: breaks: world[linux-lts~5.4]
2020-12-18 12:46:30 <pinsetonl> guix-vits: I guess I have to install kernel version from stable then, right?
2020-12-18 12:48:06 <guix-vits> maybe remove 5.10, then add 5.4?
2020-12-18 12:49:34 <guix-vits> idk Alpine, but looks like conflict between installed 5.10 and the 5.4 that is going to be instslled.
2020-12-18 12:50:32 <midao> latest stable and recompile fixed it, thanks for everyone helping me :)
2020-12-18 12:53:23 <pinsetonl> guix-vits: apk add -X http://lateststablerep -U apk add 'linux-lts==~5.4' did the job
2020-12-18 12:53:31 <pinsetonl> let's see if it helps
2020-12-18 12:54:47 <pinsetonl> guix-vits: I should've installed guix for this kind of stuff...
2020-12-18 12:56:57 <pinsetonl> guix-vits: yeah downgrading kernel worked
2020-12-18 12:57:32 <pinsetonl> ncopa: I guess 5.10.1 is broken, at least for me
2020-12-18 12:59:20 <ncopa> pinsetonl: what bootloader?
2020-12-18 12:59:31 <pinsetonl> refind
2020-12-18 12:59:48 <ncopa> i had to add a boot option
2020-12-18 13:00:12 <ncopa> efi=no_disable_early_pci_dma
2020-12-18 13:00:22 <ncopa> due to this: https://cateee.net/lkddb/web-lkddb/EFI_DISABLE_PCI_DMA.html
2020-12-18 13:01:04 <ncopa> pinsetonl: i'd be interested to hear if that solves it for you
2020-12-18 13:01:08 <guix-vits> pinsetonl:  on x86_64 guix s fairly good, yes..
2020-12-18 13:01:22 <ncopa> it did boot with grub though
2020-12-18 13:01:47 <ncopa> but not when i used the kernel as "bootmanager" directly in uefi
2020-12-18 13:01:54 <guix-vits> btw:  aren't btrfs and zfs snapshots can do the job?
2020-12-18 13:02:17 <guix-vits> with "downgrading" et all.  idk
2020-12-18 13:02:46 <pinsetonl> guix-vits: don't want to compile kernel every time...
2020-12-18 13:02:57 <pinsetonl> ncopa: gonna try it
2020-12-18 13:03:30 <guix-vits> btrfs is in mainline?
2020-12-18 13:04:20 <pinsetonl> yes, it is
2020-12-18 13:04:33 <guix-vits> cool
2020-12-18 13:23:30 <pinsetonl> ncopa: It worked. How do I unpin package now?
2020-12-18 13:30:49 <ncopa> um...
2020-12-18 13:31:06 <ncopa> i think i normally just edit /etc/apk/world file
2020-12-18 13:31:34 <ncopa> i think there is a apk <something> way to do it but i dont remember how
2020-12-18 13:32:22 <ncopa> so i wonder if we should fix the CONFIG_EFI_DISABLE_PCI_DMA to the default
2020-12-18 13:32:45 <ncopa> since it has been a problem for atleast 2 ppl so far
2020-12-18 13:58:58 <oliv3r[m]> Hi, I'm trying to bootstrap alpine (as I have done thousands of times) and i have a package (my own) that's causing problems (I'll have to deal with that :p) The problem is, if I run a http server and host the files there, everything is fine, but if i run the repo locally (from a folder) and I do --repository /path/to/internal
2020-12-18 14:01:06 <oliv3r[m]> apk add complains in that case that it cannot find the package (missing) but I cannot figure out'why'
2020-12-18 14:02:59 <oliv3r[m]> specifically it fails on the dependency off, though I bet if i ask it to install that directly, it fails too
2020-12-18 14:38:34 <guix-vits> oliv3r[m]:  file:///pathto ?
2020-12-18 14:38:40 <guix-vits> maybe
2020-12-18 14:38:44 <guix-vits> idk
2020-12-18 14:38:54 <oliv3r[m]> well i've been using this for ages; but let me try that, maybe it helps the internal parser
2020-12-18 14:39:49 <ncopa> no, i dont think file:// will help
2020-12-18 14:40:34 <guix-vits> 
ACTION *drums*

2020-12-18 14:41:17 <oliv3r[m]> nope
2020-12-18 14:41:54 <guix-vits> 
ACTION ta-da-dum-ts

2020-12-18 14:42:06 <oliv3r[m]> it does not work indeed :)
2020-12-18 14:43:07 <caskd> cool news: apk does indeed support basic auth!
2020-12-18 14:43:17 <caskd> fetch http://user:*@localhost:7574/data-packages/x86_64/APKINDEX.tar.gz
2020-12-18 14:43:28 <caskd> managed to install a package with this
2020-12-18 14:43:32 <oliv3r[m]> ncopa any tricks on debugging this? :)
2020-12-18 14:43:43 <oliv3r[m]> it does seemt o find the APKINDEX as its not complaining
2020-12-18 14:45:27 <guix-vits> maybe local repo treated another way.  is it a git tree?  all commited?
2020-12-18 14:46:58 <ncopa> oliv3r[m]: I'd do soemthing like: apk search --repositories-file /dev/null --repository /path/to somepackage
2020-12-18 14:47:16 <ncopa> maybe with strace to find out if it actually reads the APKINDEX.tar.gz
2020-12-18 14:47:44 <oliv3r[m]> good one
2020-12-18 14:55:39 <oliv3r[m]> ncopa apk search finds it just fine ... apk search --allow-untrusted --root /workdir/.build/armv7/testboard/initramfs/ --arch armv7 --repositories-file /dev/null --repository /workdir/release/internal --repository /workdir/release/upstream mypkg
2020-12-18 15:00:24 <ncopa> so maybe it does not find some of the dependencies?
2020-12-18 15:01:31 <oliv3r[m]> yeah but i have no idea why not; if i pull it online
2020-12-18 15:03:19 <oliv3r[m]> i know one dependency is 'broken' and it can't find it :)
2020-12-18 15:03:28 <oliv3r[m]> but wouldn't it list the broken dependency in apk add?
2020-12-18 15:04:22 <ncopa> i would expect it to
2020-12-18 15:04:46 <oliv3r[m]> sadly, searching for the exact thing works, but fails on apk add :(
2020-12-18 15:05:01 <ncopa> try strace the apk add
2020-12-18 15:19:34 <oliv3r[m]> heh; loads of mmap/unmap :( i'll try to figure it :)
2020-12-18 15:26:17 <oliv3r[m]> openat(5, "/workdir/release/upstream/armv7/APKINDEX.tar.gz", O_RDONLY|O_CLOEXEC) = 7 #define	E2BIG		 7	/* Argument list too long */
2020-12-18 15:26:38 <oliv3r[m]> ironically; it's the wrong folder that's failing :p (upstream is where my apk copies are)
2020-12-18 15:27:49 <oliv3r[m]> heh, it actually fails on both though i'll double check if i didn't copy/paste the same file to both location in my quest to figure it out
2020-12-18 15:29:11 <oliv3r[m]> na different fies, 32k .gz and 160k .gz; which isn't that big
2020-12-18 15:31:38 <oliv3r[m]> hmm, openat of the 'normal' file also returns '7' so I don't think it's an error :)
2020-12-18 15:39:59 <guix-vits> oliv3r[m]:  maybe paste the strace? (https://paste.debian.net)
2020-12-18 15:49:39 <oliv3r[m]> i would; but Length of code is not allowed to exceed 150kB :p
2020-12-18 16:33:19 <Hello71> ix.io
2020-12-18 16:33:29 <Hello71> also that 7 is totally unrelated
2020-12-18 16:33:40 <Hello71> openat returns an fd
2020-12-18 16:35:34 <guix-vits> > The codebase for ix.io is intended to be free and open-source. It is not published at the moment because the author doesn't want to publish code that isn't pretty and pleasant (and is also deeply lazy about getting it there).
2020-12-18 16:35:51 <guix-vits> arr
2020-12-18 16:36:12 <ikke> "The codebase is open-source in spirit only"
2020-12-18 16:36:20 <guix-vits> arr
2020-12-18 16:36:58 <guix-vits> author is refined troll
2020-12-18 16:37:21 <guix-vits> elite
2020-12-18 16:37:38 <maxice8> spiritual open source
2020-12-18 16:37:48 <guix-vits> in vacuum
2020-12-18 16:58:01 <artok> ehh
2020-12-18 16:58:23 <artok> termbin works with netcat, so no other needed
2020-12-18 17:05:49 <halosghost> at some point, I'll get my pastebin running
2020-12-18 19:55:39 <omni> I noticed that I can't have IPv6 addresses in /etc/network/interfaces for ifaces set to auto, since the ipv6 module is loaded after networking is started and not included in initramfs-virt
2020-12-18 19:55:53 <omni> perhaps I should open a ticket for its inclusion..
2020-12-18 20:25:54 <honza_> hallo
2020-12-18 20:26:29 <honza_> I have problem with compiling minisatip
2020-12-18 20:26:50 <honza_> after ./configure I have
2020-12-18 20:27:15 <honza_> checking for gcc... gcc
2020-12-18 20:27:15 <honza_> checking whether the C compiler works... no
2020-12-18 20:27:15 <honza_> configure: error: in `/prac/minisatip':
2020-12-18 20:27:15 <honza_> configure: error: C compiler cannot create executables
2020-12-18 20:27:15 <honza_> See `config.log' for more details
2020-12-18 20:28:25 <honza_> on internet I read - install libc6-dev (on ubuntu) , I install libc6-compat , but not work
2020-12-18 20:29:06 <honza_> how I make bad?
2020-12-18 20:29:39 <honza_> I dont understant - checking whether the C compiler works... no
2020-12-18 20:36:27 <yjftsjthsd> honza_ try installing alpine-sdk and see if that makes it work
2020-12-18 20:37:35 <honza_> thanks, it is ok
2020-12-18 20:38:54 <yjftsjthsd> libc6-compat shouldn't be needed; that's for glibc binary compatibility, which should be irrelevant if you're building from source
2020-12-18 20:39:20 <yjftsjthsd> (unless the software is actually glibc-only, I suppose)
2020-12-18 20:39:28 <honza_> sorry , thank you very much
2020-12-18 20:39:57 <yjftsjthsd> No need to be sorry; we all have to learn these things somehow:) You're quite welcome
2020-12-18 20:40:00 <honza_>  alpine-sdk is pacgages, what I dont have
2020-12-18 20:40:33 <yjftsjthsd> What?
2020-12-18 20:40:41 <honza_> my problem is solved
2020-12-18 20:41:04 <honza_> thank You (sorry for my English)
2020-12-18 20:41:14 <yjftsjthsd> Good:) It's okay
2020-12-18 20:53:23 <yjftsjthsd> Okay, now my turn to ask a question: Is it possible to make an account in Alpine with no password which may be accessed by SSH? If I just `useradd -m foo` and populate /home/foo/.ssh, sshd blocks the login because the account is locked. If I run `passwd -u foo` to unlock the account, it says, "passwd: unlocking the password would result in a
2020-12-18 20:53:23 <yjftsjthsd> passwordless account." But I don't *want* to set a password, I want an account that's only accessible via SSH with a key
2020-12-18 20:57:53 <wsinatra> then configured /etc/ssh/sshd_config to only allow ssh via key
2020-12-18 20:58:31 <wsinatra> and set the password to some asinine string using a credential manager, like pass or lastpass. say 48 characters of jumbled nonsense
2020-12-18 21:00:23 <yjftsjthsd> The thought had crossed my mind to just ... hand it 48 chars from /dev/random and forget about it, but that seems dumb (why provide a password that doesn't actually do anything), and it's not needed on any other distro I've tried
2020-12-18 21:00:43 <yjftsjthsd> like, I will if I have to, but I was hoping there was a nice way to tell passwd to just deal with it
2020-12-18 21:04:27 <yjftsjthsd> Actually, I'm curious why this happens in Alpine and not ex. Arch; are they using different shadow implementations?
2020-12-18 21:05:09 <maxice8> I just have /etc/shadow use :*: instead of :!:
2020-12-18 21:08:05 <yjftsjthsd> Aha, that seems to work: `usermod -p \* foo && usermod -U foo` seems to work! Thanks:)
2020-12-18 22:33:51 <tomleb> Did anyone get podman working on alpine ?
2020-12-18 23:31:15 <yjftsjthsd> What error are you getting?
2020-12-18 23:39:21 <yjftsjthsd> Oh wait no I don't have it working on alpine. Sorry, confused it with Void.
2020-12-18 23:49:20 <ipcloud> tomleb: not yet ; something  I interested in
2020-12-18 23:49:28 <ipcloud> s/I/I am/
2020-12-18 23:49:29 <maxice8> tomleb: there are people using podman as a docker replacement in Alpine, ask in #alpine-devel
2020-12-18 23:59:37 <feuerteufel> Hi @ all
2020-12-19 00:01:44 <feuerteufel> I yust graded my alpine edge up to the latest versin, now the Network is not coming up automaticly on reboot. I have to start it by "ipup eth0" ...
2020-12-19 00:02:33 <feuerteufel> Anyone an idea how to fix that one?
2020-12-19 00:06:55 <tomleb> maxice8: thanks, I will ask
2020-12-19 00:10:13 <tehcloud> feuerteufel, have not had that problem
2020-12-19 00:13:56 <feuerteufel> I also had it before, it's the first time. But Idon't have an Idea how to fix it ... The networking is running when it is started by hand ...
2020-12-19 00:27:56 <feuerteufel> Sorry, "I also didn't had it before" ;)
2020-12-19 00:29:16 <Hello71> apk fix openrc
2020-12-19 00:32:45 <feuerteufel> Hello71: Yea, that fixed it Thank you verry mutch!!
2020-12-19 08:30:46 <steve24> I just installed Alpine. I can log in via ssh as root but I get permission denied for a new user just created. I know the password is correct.
2020-12-19 08:33:29 <Cogitri> I assume your ssh server allows password auth?
2020-12-19 08:33:38 <Cogitri> Could check your logs
2020-12-19 08:46:44 <steve24> Yes, uncommented pass auth and restarted. Messages log says Failed password for invalid user steve from <my ip addresss> port 55812 ssh2. steve is in passwd. passwd passwd
2020-12-19 08:55:03 <detha> 'invalid user' says it doesn't know the user. Can you su from root to that account?
2020-12-19 09:19:42 <steve24> Ok. Figured it out. I thought bash was installed by default with busybox and that's the shell I assigned to the new user. Changed it to "ash" (whatever that is) and things work now. Thanks.
2020-12-19 09:21:44 <steve24> The tip off was when I couldn't su from root to steve (Thx detha)
2020-12-19 09:24:15 <mistnim> hi, I installed alpine on disk, but all the modules that were loaded on the installation media are not available now (modoprobe says not found). lsmod gives me only one loaded.
2020-12-19 09:24:43 <mistnim> should I install them or do something to get them?
2020-12-19 09:31:02 <mistnim> /lib/modules dir seems to be the same of the installation media
2020-12-19 09:34:28 <ikke> they should come with the kernel
2020-12-19 09:37:21 <mistnim> ikke: yes, and it seems that they are there in /lib/modules
2020-12-19 09:37:33 <ikke> what does uname -r return?
2020-12-19 09:37:38 <ikke> and ls /lib/modules
2020-12-19 09:38:56 <mistnim> ikke: consider I'm on the installation media, uname-r: 5.4.83-0-rpi4
2020-12-19 09:39:28 <mistnim> /nnt/lib/modules says the same thing
2020-12-19 09:40:57 <mistnim> I would have to reboot, but I don't imagine it would say something different to uname -r?
2020-12-19 09:43:34 <mistnim> ok I try it
2020-12-19 09:56:01 <mistnim> ikke: thanks, that was it, it installed a new version of the kernel, but I needed to copy it in /boot.
2020-12-19 09:56:19 <mistnim> I mean in the actual boot partition
2020-12-19 10:19:39 <oliv3r[m]> managed to capture a strace by removing mmapa nd mumap :) https://paste.debian.net/1177576/ stil baffles me though
2020-12-19 11:19:27 <guix-vits> oliv3r[m]: /workdir, not ./workdir?
2020-12-19 11:24:46 <guix-vits> mount("proc", "/workdir/.build/armv7/testboard/initramfs//proc", "proc", 0, NULL) = -1 EPERM (Operation not permitted)
2020-12-19 11:26:31 <guix-vits> ok, maybe this.  but but as i remember you can do all just fine if running http server.  strange.
2020-12-19 12:12:34 <EF> Any advice, new drive, worked, rebooted got LBA-48 error in BIOS, clanky clanky head when booting now doesn't show up in Alpine. Likely dead and need to return it?
2020-12-19 12:13:21 <ikke> If a harddrive makes clanky noises, then, I would certainly think so
2020-12-19 12:14:40 <EF> Ok thanks.
2020-12-19 12:20:19 <EF> Enjoying Apline btw, it is a replacement for a Synology box as hopefully CoTS kit with Alpine has a far long EoL/support than Synology.
2020-12-19 12:25:41 <EF> *Alpine
2020-12-19 13:18:52 <Cogitri> Speaking of NAS, does one of you happen to know a webUI that works with Alpine for managing a NAS? Would be neato if there was something like FreeNAS-ish but without the FreeBSD core
2020-12-19 13:19:29 <Cogitri> Mainly the automatic email in case things go wrong (e.g. harddrive dies) stuff is kind of a hassle to setup manually
2020-12-19 14:50:41 <TBK[m]> I have planned to port the Cockpit project when I migrate my server to Alpine in the beginning of the new year.
2020-12-19 14:54:13 <maxice8> Doesn't it heavily rely on systemd?
2020-12-19 14:56:56 <TBK[m]> That is also my impression, but I have not taken a deeper look as of yet.
2020-12-19 15:07:45 <TBK[m]> Another option is Ajenti, but it will also require some work.
2020-12-19 15:19:14 <oliv3r[m]> guix-vits nah; its a docker container where this is run in, where /workdir is -w :p so the path is fine; and that is visible too as the apindex file can be read just fine
2020-12-19 21:52:53 <xerz> hmm has anyone tried to run a desktop like Sway on QEMU?
2020-12-19 21:54:03 <xerz> I can't seem to get it to work and `dmesg | grep drm` always comes up empty, no matter if I install all  `linux-firmware` and regardless of kernel or VM settings choice
2020-12-19 21:59:02 <Hello71> remove nomodeset from kernel command line
2020-12-20 05:31:51 <guix-vits> i had a look in generic tar for aarch64.  seems it cannot boot a / on btrfs as is?  u-boot can read this fs.
2020-12-20 07:28:18 <guix-vits> as i understood, the module loop is not used by initramfs.  also it has no /sbin/btrfs.  but the snapshots before updates.
2020-12-20 07:34:12 <guix-vits> i mean, btrfs is just up to 3 mib.  doing like mkinitfs, or unpack-repack, is not so plug-n-play.
2020-12-20 10:26:00 <telmich> I'm trying to find out why users created by `useradd` are disabled by default in alpine and while doing this I am looking for the `which packages owns useradd` command and while doing this I noticed that apk does not have apk-doc nor a manpage nor seems the --help indicate what to query. I'd be thankful for a pointer in either correct direction
2020-12-20 10:28:02 <telmich> Or in other words: how do I create a user with `useradd` that does not have a password setup, but which I can ssh into, if a config tool like cdist has already placed the correct ssh key into ~/.ssh/authorized_keys?
2020-12-20 10:28:24 <guix-vits> telmich:  maybe it gives bash as default?  today was mentioned smth like that.
2020-12-20 10:28:53 <telmich> The problem is that there is an `!` in the shadow entry, which practically disables login for the user
2020-12-20 10:29:27 <telmich> ahhh ... found it! /etc/default/useradd has an INACTIVE line
2020-12-20 10:29:29 <guix-vits> someone have suggested !/*
2020-12-20 10:29:48 <guix-vits> earlier
2020-12-20 10:31:18 <telmich> So... how do I get around this, as this breaks `__user` from cdist on almost all alpine systems
2020-12-20 10:32:02 <telmich> hmm.. no, INACTIVE=-1 probably refers to the --inactive parameter for expiration
2020-12-20 10:33:29 <telmich> `usermod -U app` complains that `usermod: unlocking the user's password would result in a passwordless account.` - which is exactly what I want.
2020-12-20 10:34:07 <telmich> `usermod -p '' app` seems to do what I want
2020-12-20 13:01:11 <nihilowy> hi
2020-12-20 13:01:52 <nihilowy> anyone use epiphany here? in mine facebook site doest work
2020-12-20 13:03:53 <Apachez> epi what?
2020-12-20 13:04:04 <nihilowy> epiphany web browser
2020-12-20 13:04:18 <nihilowy> like surf /lukait same engine
2020-12-20 13:05:27 <nihilowy> or maybe its sway problem
2020-12-20 13:47:54 <q66> Ariadne: around?
2020-12-20 14:54:10 <Ariadne> q66: whats up
2020-12-20 14:57:07 <q66> Ariadne: was https://github.com/void-linux/void-packages/commit/be4cdeb83a33b9fc4a2b258f9041843624c5553a meant to happen?
2020-12-20 14:57:19 <q66> just curious since it means i'll have to rebuild things using libucontext
2020-12-20 14:58:15 <Ariadne> hmm
2020-12-20 14:58:32 <Ariadne> no, the ABI should remain the same
2020-12-20 14:59:25 <q66> thought so
2020-12-20 14:59:37 <q66> which is why i wanted to ping you about it to see what should be done next
2020-12-20 15:00:09 <Ariadne> the desired ABI is one where [get|set|make|swap]context is aliased to their libucontext_ counterpart
2020-12-20 15:00:16 <Ariadne> that way ABI is not broken
2020-12-20 15:00:30 <Ariadne> i think theres some funky stuff going on with ppc though
2020-12-20 15:00:37 <q66> there might be
2020-12-20 15:00:58 <q66> it would seem there is direct references to __getcontext and so on in libraries
2020-12-20 15:01:13 <Ariadne> hmm
2020-12-20 15:01:23 <Ariadne> thats strange
2020-12-20 15:01:29 <Ariadne> not what i intended
2020-12-20 15:02:00 <q66> well, that's what the linker suggests anyway
2020-12-20 15:02:00 <Ariadne> __getcontext was the original internal function name
2020-12-20 15:02:08 <q66> yeah i understood that much
2020-12-20 15:02:14 <Ariadne> that became libucontext_getcontext
2020-12-20 15:02:20 <Ariadne> and getcontext is supposed to be a weak_alias
2020-12-20 15:02:29 <Ariadne> but apparently it is not on ppc
2020-12-20 15:03:02 <Ariadne> wonder what happened
2020-12-20 15:03:26 <Ariadne> the intricacies of ppc linking are...not really a familiar topic for me
2020-12-20 15:03:32 <Ariadne> maybe koorogi knows?
2020-12-20 15:04:46 <q66> it means this is quite bad because the internals of libucontext were already leaked inside the userland
2020-12-20 15:04:47 <q66> hmm
2020-12-20 15:05:25 <Ariadne> we can always bump SOVERSION
2020-12-20 15:05:44 <Ariadne> but i thinkit is important to make sure the library is linking as intended first
2020-12-20 15:05:46 <q66> yeah that would fi it
2020-12-20 15:05:47 <q66> yeah
2020-12-20 15:05:50 <q66> *fix
2020-12-20 15:05:50 <Ariadne> so we dont have to bump again
2020-12-20 15:06:07 <q66> bumping soversion would at least mean i know what is affected and what to rebuild
2020-12-20 15:06:50 <Ariadne> hmm, does void not generate soname dependencies?
2020-12-20 15:06:55 <q66> it does
2020-12-20 15:06:59 <q66> which is how i will know
2020-12-20 15:07:05 <Ariadne> in alpine you can just lookfor what packages depend on so:libucontext.so.0
2020-12-20 15:07:07 <q66> if soname changes it will stage the repo
2020-12-20 15:07:15 <q66> and tell me what needs to be rebuilt
2020-12-20 15:07:38 <q66> yeah void packages have shlib-requires metadata
2020-12-20 15:07:40 <Ariadne> well i prefer to wait on SOVERSION bump until we know what is going on :)
2020-12-20 15:07:56 <q66> (and shlib-provides)
2020-12-20 15:08:02 <Ariadne> i have a bunch of stuff to push out anyway
2020-12-20 15:08:08 <Ariadne> i basically rewrote the library
2020-12-20 15:08:13 <q66> alright, cool
2020-12-20 15:08:30 <Ariadne> but i dont want to bump SOVERSION *until* we know the actual issue is fixed in the lib
2020-12-20 15:08:42 <Ariadne> programs should link directly against the weak aliases
2020-12-20 15:10:26 <q66> yup
2020-12-20 15:10:28 <q66> makes sense
2020-12-20 16:22:31 <xpacio> :)
2020-12-20 16:22:32 <xpacio> Hola a todos, soy jose.alvarez desde mexico.
2020-12-20 16:22:32 <xpacio> he encontrado estos comandos pero no logro identificar si existe algun apk que los contenga.
2020-12-20 16:22:32 <xpacio> - - - - - - - - - - - - - - - - - -
2020-12-20 16:22:32 <xpacio> Tengo alpine linux en una laptop que ocupo por ssh, existe alguna forma de apagar la pantalla despues de 1 minuto de inactividad o  despues del arranque?
2020-12-20 16:22:32 <xpacio> Hi everyone, I'm jose.alvarez from Mexico.
2020-12-20 16:22:32 <xpacio> I have alpine linux on a laptop that I use by ssh, is there a way to turn off the screen after 1 minute of inactivity or after booting?
2020-12-20 16:22:32 <xpacio> I have found these commands but I cannot identify if there is an apk that contains them.
2020-12-20 16:22:33 <xpacio> [commands]
2020-12-20 16:22:33 <xpacio> vbetool dpms off
2020-12-20 16:22:34 <xpacio> setterm --blank 1 --powerdown 2
2020-12-20 16:22:34 <xpacio> Note:
2020-12-20 16:22:35 <xpacio> /sys/module/kernel/parameters/consoleblank = 0
2020-12-20 16:22:35 <xpacio> Thanks in advance.
2020-12-20 16:23:23 <ikke> https://pkgs.alpinelinux.org/contents?file=setterm&path=&name=&branch=edge&arch=x86_64
2020-12-20 16:27:47 <ikke> xpacio: util-linux contains setterm
2020-12-20 16:27:59 <ikke> I also read that you can provide consoleblank as kernem parameter
2020-12-20 16:28:22 <xpacio> how fast! thanks . :)
2020-12-20 19:03:45 <PureTryOut[m]2> Where does apk place a lock file to determine it's already running somewhere?
2020-12-20 19:04:34 <ikke> /lib/apk/db/lock
2020-12-20 19:05:06 <ikke> note that it uses flock, so the file existing does not mean the lock is in use
2020-12-20 19:06:26 <PureTryOut[m]2> Oh, interesting
2020-12-20 19:07:05 <PureTryOut[m]2> Is there some Bash command I can run then to check if it's in use?
2020-12-20 19:07:51 <ikke> there is an flock command
2020-12-20 19:08:04 <ikke> in util-linux
2020-12-20 19:08:28 <ikke> oh, bb also has flock
2020-12-20 19:08:47 <PureTryOut[m]2> Yeah found it, but I get a bad file descriptor when I point it to it? 🤔
2020-12-20 19:10:44 <ikke> I think you need to provide a command
2020-12-20 19:11:06 <ikke> `flock /lib/apk/db/flock echo test` works
2020-12-20 19:11:20 <ikke> even true/false work
2020-12-20 19:12:43 <PureTryOut[m]2> Hmm seems it needs root permissions 😕
2020-12-20 19:12:58 <ikke> right
2020-12-20 19:13:05 <ikke> I think that's on purpose
2020-12-20 19:14:04 <PureTryOut[m]2> Hmm well should be alright, need sudo to invoke apk anyway
2020-12-20 19:14:06 <ikke> well, I'm almost certain that it's on purpose
2020-12-20 19:14:09 <ikke> yes
2020-12-20 19:14:59 <ikke> that's basically what causes this error: 'ERROR: Unable to lock database: Permission denied'
2020-12-20 19:15:43 <PureTryOut[m]2> Can I use flock to tell me if it's locked, rather than waiting till it's unlocked?
2020-12-20 19:16:04 <ikke> -n, --nonblock           fail rather than wait
2020-12-20 19:19:06 <PureTryOut[m]2> Thanks, that does the trick!
2020-12-20 20:06:59 <hechos> hiu guys.
2020-12-20 20:08:12 <hechos> ash shell can made a color prompt from .profile right?
2020-12-20 20:08:26 <ikke> yes
2020-12-20 20:08:33 <ikke> there is a script in /etc/profile.d/
2020-12-20 20:08:37 <ikke> color_prompt
2020-12-20 20:08:43 <ikke> rename it to color_prompt.sh
2020-12-20 20:09:30 <hechos> ohh thanks ikke
2020-12-20 20:09:37 <hechos> lot of work on console need some colors.
2020-12-20 21:26:13 <c705> my terminal doesn't show utf-8 characters. where do I configure this? locale?
2020-12-20 21:55:26 <mps> fonts
2020-12-21 00:13:33 <guix-vits> c705: maybe  https://pkgs.alpinelinux.org/packages?name=*ttf-*&branch=edge
2020-12-21 03:36:07 <NeedHalpPlz> hey guys
2020-12-21 03:36:11 <NeedHalpPlz> how are we doing tonight
2020-12-21 04:21:29 <tomleb> Hi, I am running alpine 3.12 on a raspberry pi. So far so good, right now I'm trying to install wireguard. I've added modules=overlay to be able to install wireguard-rpi4. The install ran fine (afaik), but I keep booting to the old kernel, not the new one. Any idea where I can look to debug this?
2020-12-21 04:24:19 <libredude> hello
2020-12-21 04:30:04 <andeyo> hello
2020-12-21 05:07:29 <HalpMePlz23> I'm getting some issues setting up eth) during install on OpenRC
2020-12-21 05:07:38 <HalpMePlz23> can anyone help me out? or have an idea why?
2020-12-21 05:20:32 <HalpMePlz> Sorry, my pc keeps crashing
2020-12-21 05:21:18 <HalpMePlz> I'm having issues configuring network in OpenRC during install
2020-12-21 05:21:47 <HalpMePlz> my eth0 and wlan does not work
2020-12-21 13:26:00 <tomleb> Seems like I had to make a symbolic link of /media/mmcblk0p1/boot to /boot and then install wireguard-rpi4. Is this the typical way of installing it?
2020-12-21 13:34:26 <minimal> wireguard-rpi4 installs module inside /lib so don't see why the /boot sym link would be needed
2020-12-21 13:35:17 <minimal> however a kernel update would be a different matter as the kernel does install files into /boot
2020-12-21 13:35:51 <tomleb> Yeah, kernel is .84 with wireguard-rpi4 but .83 currently.
2020-12-21 13:37:17 <minimal> there's an ongoing "issue" around the RPI kernel packaging regarding /boot
2020-12-21 13:38:38 <tomleb> Ok, is the workaround to do like I did? Is that "issue" documented anywhere?
2020-12-21 13:40:55 <minimal> issue is around /boot/dtbs-rpi (and dtbs-rpi2 / rpi4) containing files that really should be in /boot
2020-12-21 13:41:06 <tomleb> ah ok
2020-12-21 13:41:59 <minimal> I currently fudge around it by bind-mounting /boot onto /boout/dtbs-rpi (or rpi2 / rpi4 - directory name differs depending if armv7 or aarch64 and what type of RPI you have)
2020-12-21 13:42:48 <minimal> oops, sorry I *meant* to say bind-mount /boot/dtbs-rpi onto /boot
2020-12-21 13:43:26 <minimal> the idea is that when kernel package update is installed it write files to /boot/dtbs-rpi which (due to bind-mount) end up in /boot
2020-12-21 13:48:57 <minimal> tomleb: I'm hoping to fix the situation
2020-12-21 13:58:18 <markand> is there a bug with the /etc/init.d/networking script on edge?
2020-12-21 13:58:27 <markand> I have a /etc/network/interfaces like this:
2020-12-21 13:58:28 <markand> auto lo
2020-12-21 13:58:33 <markand> iface lo inet loopback
2020-12-21 13:58:45 <markand> and networking script fails with:
2020-12-21 13:59:00 <markand> ifquery: could not parse /etc/network/interfaces
2020-12-21 14:05:11 <ikke> apk info -W $(which ifquery)
2020-12-21 14:18:51 <ikke> markand: ^
2020-12-21 14:30:36 <EF> Is it good practice to have apps, like Nextcloud, stored outside of webroot and then soft linked into it?
2020-12-21 14:31:28 <ikke> Generally it's only necessary to have static assets in your webroot
2020-12-21 14:32:13 <ikke> And the entrypoint
2020-12-21 14:34:46 <EF> Ok thanks. Trying to overcome the insanity that is Synology and do things by hand (big learning curve) more correctly.
2020-12-21 15:50:23 <hechos> hi guys.
2020-12-21 15:52:25 <hechos> mpv is closed unexpectedly
2020-12-21 15:53:58 <Apachez> EF: depends on taste
2020-12-21 15:54:10 <Apachez> EF: good thing of symlinking is easier versioning
2020-12-21 15:54:31 <Apachez> like /opt/nextcloud_v1  and /opt/nextcloud_v2 and then you just symlink to current version
2020-12-21 15:54:44 <Apachez> bad thing with symlinks are additional hdd and cpu cycles to do the same thing
2020-12-21 15:54:54 <Apachez> at least if you scale up to gazillions of requests
2020-12-21 15:59:04 <pmjohann> Hello guys
2020-12-21 15:59:18 <ikke> There are also gals here :)
2020-12-21 15:59:31 <pmjohann> Even better! :)
2020-12-21 15:59:45 <pmjohann> I love the project you are doing, my fave linux distro for getting work done!
2020-12-21 16:01:42 <mps> 'project' implies there are (or was) plan but I don't feel we have nothing of that :)
2020-12-21 16:02:15 <ikke> Loose collaboration of people with somewhat similar goals :)
2020-12-21 16:02:36 <mps> evolutive Loose ...
2020-12-21 16:03:00 <mps> and that sounds good to me
2020-12-21 16:03:18 <hechos> anyone have full deps to compile mpv?
2020-12-21 16:03:32 <pmjohann> Often I also use it on raspberry Pis
2020-12-21 16:03:36 <hechos> waf check a bunch of deps.
2020-12-21 16:03:48 <pmjohann> Something I really didn't find any docs for is how to manually set the size of the root partition which gets allocated from RAM
2020-12-21 16:03:55 <pmjohann> by default it is half of the ram
2020-12-21 16:04:05 <mps> hechos: they are listed in mpv APKBUILD
2020-12-21 16:04:05 <pmjohann> is there a way to manually set it?
2020-12-21 16:04:31 <ikke> pmjohann: It's a kernel peramater
2020-12-21 16:05:45 <ikke> rootflags=size=<n>
2020-12-21 16:05:51 <ikke> parameter*
2020-12-21 16:06:41 <hechos> mps: is last version
2020-12-21 16:06:59 <mps> ?
2020-12-21 16:07:08 <hechos> mps: my bad.
2020-12-21 16:07:23 <hechos> mps: got deps and try to compile last version from git.
2020-12-21 16:07:46 <mps> aports git?
2020-12-21 16:08:33 <hechos> no git from mpv.
2020-12-21 16:08:47 <hechos> im cloning aports right now.
2020-12-21 16:08:49 <ikke> How are you building it?
2020-12-21 16:09:33 <hechos> as a bunch of deps from github mpv.io
2020-12-21 16:10:16 <mps> hechos: there is #mpv channel on freenode
2020-12-21 16:11:57 <hechos> mps: okey.
2020-12-21 16:13:19 <markand> ikke, it's in ifupdown-ng
2020-12-21 16:26:32 <AL3X> Hi all, I am new to ZFS, but not new to Alpine. I would like to ask for a little help with setting up alpine with native zfs encryption according to https://wiki.alpinelinux.org/wiki/Alpine_Linux_with_root_on_ZFS_with_native_encryption. The physical disk has a capacity of 5TB. Since the tutorial uses a MBR I can only use 2TB. Setting it up with a GPT I can get the 5TB but I can't get it to boot. In what way would one need to differ from
2020-12-21 16:26:32 <AL3X> the wiki article to get just the same setup working on a disk with >2TB? THANKS <3
2020-12-21 16:27:32 <pmjohann> ohh @ikke thank you!
2020-12-21 16:28:59 <AL3X> I tried 'sfdisk --quiet --label gpt /dev/sda' instead of 'sfdisk --quiet --label dos /dev/sda' and in the end 'dd if=/usr/share/syslinux/gptmbr.bin of=/dev/sda" instead of 'dd if=/usr/share/syslinux/mbr.bin of=/dev/sda', to no avail.
2020-12-21 17:14:37 <xordspar0> Does anyone here automate their updates? I was thinking it would be pretty easy to set up cron to run `apk update && apk upgrade` or something like that, but I wonder if there would be corner cases (required reboots, manual intervention) that would make it more trouble than it's worth.
2020-12-21 17:16:46 <mps> apk search apk-cron
2020-12-21 18:32:32 <EF> Apachez: thanks for the tips. Server is low volume but security is the No.1 requirement
2020-12-21 18:36:35 <xordspar0> mps: Cool, you can't get any simpler than that. I guess I'll give it a shot and see how it works out for me.
2020-12-21 19:51:37 <l1x> hi folks, how can I create tap0 at boot time?
2020-12-21 19:51:57 <l1x> do i need to run commands from the interfaces file?
2020-12-21 19:54:52 <mps> l1x: first you need put tun module in /etc/modules
2020-12-21 19:55:12 <l1x> yep that i figured out
2020-12-21 19:56:47 <mps> l1x: rest is on your use case, f.e. openvpn do that automatically
2020-12-21 20:05:12 <l1x> mps: i need to have tap0 after boot because I would like to bridge that interface with eth0
2020-12-21 20:05:29 <l1x> i guess ineed to add the creation code to the pre if-up hook
2020-12-21 20:42:59 <zxz> Good day, Everyone -- I have created a pSeries/POWER9 guest running Alpine 3.12.3 on a POWER9 host (QEMU 5.1, Linux 5.9) and am seeing a minute delay between the openfirmware quiesce/Linux load, and Alpine login prompt.  Could anyone please tell me if this is expected behavior?  If not, do you have any suggestions to improve the speed of the boot?
2020-12-21 20:44:25 <ikke> zxz: is 'quiet' being passed as kernel argument?
2020-12-21 20:44:31 <Apachez> how do one happen to have a power9 as a spare? ;)
2020-12-21 20:44:59 <Apachez> not even a rpi3 takes a minute to boot
2020-12-21 20:45:06 <Apachez> but yeah look into kerneloptions
2020-12-21 20:45:09 <Apachez> or bootoptions
2020-12-21 20:45:35 <mps> RDRND?
2020-12-21 20:46:24 <zxz> ikke: I am booting from the 3.12.3 ISO to perform the install, and have not adjusted the kernel parameters.  This behavior is also present after an install.  I'll give that a look -- thanks!
2020-12-21 20:46:45 <zxz> Apachez: Much appreciated -- I'll take a look!
2020-12-21 20:47:20 <zxz> mps: Should I add the spapr-rnd device to the guest as well?
2020-12-21 20:47:26 <ikke> zxz: do you see anything happening in between?
2020-12-21 20:48:05 <crosbymichael> so whats the best way to connect to iscsi targets on boot for a system? this has alreays confused me
2020-12-21 20:48:37 <mps> zxz: I don't know for power machines but for other it was issue earlier
2020-12-21 20:49:14 <zxz> ikke: I don't see anything, either on the guest's framebuffer, or guest's serial out for that minute; immediately afterwards, I see the OpenRC output that I'd expect, and then am presented with the login.  The time from the OpenRC startup to the login prompt takes less than 2 seconds.
2020-12-21 20:49:15 <crosbymichael> some way to integrate the iscsi targest with fstab
2020-12-21 20:49:23 <mps> I have to wait about 15 minutes on some arm32 boards till I patched kernel
2020-12-21 20:49:34 <ikke> crosbymichael: apparently there is an iscsid
2020-12-21 20:49:42 <zxz> mps: I appreciate the suggestion -- I'll check that out too!  Thank you!
2020-12-21 20:49:49 <ikke> open-iscsi-openrc-2.1.0-r2
2020-12-21 20:50:02 <ikke> crosbymichael: do you use open-iscsi?
2020-12-21 20:50:08 <crosbymichael> ya, i do
2020-12-21 20:51:10 <mps> zxz: here is how I install and boot aarch64 in qemu http://arvanta.net/alpine/install-aarch64-qemu/ maybe can help
2020-12-21 20:51:20 <crosbymichael> that looks like it will do.  now i just have to figure what type of config it was :)
2020-12-21 20:51:41 <zxz> mps: I'm grateful for the help, thank you!
2020-12-21 20:51:51 <mps> zxz: np
2020-12-21 21:10:16 <AL3X> hello everyone, any clue how to install alpine on a single drive > 2TB? When using GPT part table and syslinx' gptmbr.bin I keep getting "Missing OS" upon booting.
2020-12-21 21:12:08 <mps> AL3X: you need boot partition, 512MB is quite ok
2020-12-21 21:12:45 <AL3X> I'll give it a try right away
2020-12-21 21:13:05 <mps> and mark boot partition as 'bootable' with fdisk/parted
2020-12-21 21:15:19 <Hello71> AL3X: i recommend using grub for any non-trivial configuration
2020-12-21 21:15:30 <Hello71> additionally use efi if your firmware supports it
2020-12-21 21:16:09 <AL3X> didn't try with grub yet, will go that path also
2020-12-21 21:18:05 <crosbymichael> the setup-alpine script to install works everytime for me
2020-12-21 21:18:19 <tomleb> crosbymichael: I'm using this for raspberry pi installation: https://wiki.alpinelinux.org/wiki/Raspberry_Pi_-_Headless_Installation. Not sure if that works for alpine on other systems. Can easily be modified to install iscsi with target configuration.
2020-12-21 21:46:53 <AL3X> mps, thanks got it working. i mixed these two wikis: "Setting up LVM on GPT-labeled disks" and "Alpine Linux with root on ZFS with native encryption". Now I got a 5TB zfs fs with native encryption. nice. good night (day) everyone!
2020-12-21 22:32:12 <daftaupe> is there some maintenance on https://bugs.alpinelinux.org ?
2020-12-22 01:56:02 <COVID-1984> just checked  ntpctl -s all
2020-12-22 01:56:15 <COVID-1984> after a fresh install... they are all unsynced, even after restart of teh service
2020-12-22 01:56:21 <COVID-1984> figured the setup script would handle that
2020-12-22 01:56:25 <COVID-1984> it acted like it did
2020-12-22 01:57:56 <COVID-1984> maybe I shouldn't have used openntpd
2020-12-22 01:59:43 <Apachez> any of you who have been playing around with isc-dhcp-server?
2020-12-22 02:01:19 <COVID-1984> https://gitlab.alpinelinux.org/alpine/aports/-/issues/9635 indeed
2020-12-22 02:01:23 <COVID-1984> same issue
2020-12-22 02:01:30 <COVID-1984> haha wow just open unresolved eh
2020-12-22 02:03:16 <tehcloud> that's n awesome nick you've got there
2020-12-22 02:03:59 <COVID-1984> tehcloud: thanks :)
2020-12-22 02:04:27 <COVID-1984> I get lots of compliments, but I don't dare talk about it because I just end up getting banned for it.
2020-12-22 02:06:13 <nacelle> Apachez: what do you want to know?
2020-12-22 02:06:32 <Apachez> are there a way to run it in strict option82 mode?
2020-12-22 02:06:38 <Apachez> like without a lease file sort of speak?
2020-12-22 02:08:18 <steve24> Anyone Alpine php users know why the unix socket file for PHP7 (php-fpm.sock) is not install or created for the php-fpm7 service? Is the TCP socket only supported?
2020-12-22 02:08:55 <nacelle> Apachez: just static assignments, no actual leasing? hrmm
2020-12-22 02:13:59 <Thermi> steve24: Works for me. Check logs.
2020-12-22 02:14:11 <Thermi> steve24: you probably need to add a config.
2020-12-22 02:14:40 <Thermi> steve2425: you probably need to add a config. It works for me. I added configs for my sites. :)
2020-12-22 02:15:59 <Apachez> nacelle: yeah, the leases will be based on where the client is physically located in the network
2020-12-22 02:16:33 <Apachez> so technically I dont need any lease because any dhcp query with circuit.id=SW1_INT4 should always get the same ip handed over
2020-12-22 02:17:26 <steve2425> That's good to know Thermi. I set up php-fpm using unix sockets on FreeBSD, but all the file were installed automatically.
2020-12-22 02:17:49 <Thermi> steve2425: the default config is enabled, but without any site configs, there probably won't be any unix sockets to use
2020-12-22 02:19:27 <steve2425> what site config file are you referring to?
2020-12-22 02:20:51 <Thermi> steve2425: Pools. That what I meant.
2020-12-22 02:20:56 <Thermi> steve2425: you need to define pools.
2020-12-22 02:29:46 <steve2425> Ok, thanks. I'll need to look at my php-fpm.conf and www.conf files from FreeBSD setup because I think some things were setup automatically.
2020-12-22 05:14:24 <FestiveCup> /*28
2020-12-22 06:11:36 <iggy> is there a way to manually trigger the mkinitfs trigger that normally gets run after updates?
2020-12-22 06:16:01 <ikke> I'm not aware of a way to manually execute a trigger
2020-12-22 06:16:36 <iggy> I'll just uninstall and reinstall something, lol
2020-12-22 06:36:14 <infinite_recursi> https://wiki.alpinelinux.org/wiki/ISP_Mail_Server_3.x_HowTo
2020-12-22 06:36:47 <infinite_recursi> This is a fantastic guide. However, can you also add some documentation on how to setup a send-only email server too. Thanks
2020-12-22 13:03:09 <Hello71> apk fix
2020-12-22 13:37:14 <l1x> hi guys if i still experience this on 3.12 what should i do? https://gitlab.alpinelinux.org/alpine/aports/-/issues/3506
2020-12-22 13:42:03 <Hello71> wait for 3.13 :p
2020-12-22 13:42:30 <Hello71> did you install bridge
2020-12-22 13:43:18 <l1x> yep
2020-12-22 13:44:35 <Hello71> try 3.13/edge then
2020-12-22 13:44:59 <Hello71> the default networking manager is being changed to ifupdown-ng
2020-12-22 13:47:15 <MY-R> bridge was working fine for me in 3.12.x, maybe is matter of configuration in "interfaces" file
2020-12-22 13:49:21 <l1x> MY-R Hello71 thanks, i think the problem is that apk add bridge does not work
2020-12-22 13:49:29 <l1x> sudo apk add bridge
2020-12-22 13:49:29 <l1x> ...ERROR: unsatisfiable constraints:
2020-12-22 13:49:29 <l1x>     provided by: samurai
2020-12-22 13:49:29 <l1x>   ninja (virtual):
2020-12-22 13:49:29 <l1x>     required by: world[ninja]
2020-12-22 13:50:02 <MY-R> so bridge pkg isntinstalled
2020-12-22 13:51:24 <l1x> hmm, does this mean that?
2020-12-22 13:51:56 <MY-R> it is just single file script but you should better fix your issue
2020-12-22 13:52:06 <kmmndr> Hi all :-)
2020-12-22 13:52:41 <kmmndr> Is anyone here using alpine on arm(v7) ?
2020-12-22 13:52:52 <MY-R> l1x: do apk update and then try
2020-12-22 13:53:32 <l1x> kmmndr: rpi4 is armv7?
2020-12-22 13:53:33 <ikke> Why does it need to install ninja?
2020-12-22 13:53:42 <l1x> ikke: not sure
2020-12-22 13:53:43 <ikke> rpi4 is aarch64 afaik
2020-12-22 13:54:12 <kmmndr> Date in an alpine docker container is good for alpine:3.12, wrong for alpine:edge
2020-12-22 13:54:13 <MY-R> ye have no idea what it showing ninja etc, maybe some pinned repository with same name dunno
2020-12-22 13:54:29 <kmmndr> l1x: are you using docker too ? :-)
2020-12-22 13:54:35 <l1x> kmmndr: yes :)
2020-12-22 13:54:40 <l1x> MY-R: i just removed it
2020-12-22 13:54:53 <kmmndr> l1x: Could you try "docker run --rm -it alpine:edge date" ?
2020-12-22 13:55:02 <l1x> it was the problem!
2020-12-22 13:55:09 <l1x> somehow ninja / samurai got stuck
2020-12-22 13:55:26 <l1x> and after removing those (i do not need them) bridge is installed
2020-12-22 13:55:43 <ikke> kmmndr: there have been issues with tzdata, but they should've been fixed
2020-12-22 13:56:21 <l1x> kmmndr: docker run --rm -it alpine:edge date ; docker: Error response from daemon: failed to create endpoint trusting_bouman on network bridge: failed to add the host (veth5b248a2) <=> sandbox (veth7bec953) pair interfaces: operation not supported.
2020-12-22 13:56:22 <kmmndr> tzdata is a package
2020-12-22 13:56:31 <ikke> kmmndr: yes
2020-12-22 13:56:52 <kmmndr> It is not in the base image
2020-12-22 13:57:27 <kmmndr> I mean the bug is in the base alpine image
2020-12-22 13:57:44 <ikke> kmmndr: testing it on armv7, but it shows the correct date for me
2020-12-22 13:58:13 <ikke> https://tpaste.us/DM8e
2020-12-22 13:58:17 <kmmndr> I did a "docker pull alpine:edge" to be sure to have the latest one
2020-12-22 13:58:33 <ikke> kmmndr: what is wrong about it?
2020-12-22 13:58:42 <kmmndr> ikke: That's not in a docker container :-)
2020-12-22 13:59:02 <ikke> it is
2020-12-22 14:00:15 <l1x> kmmndr: docker run --rm -it alpine:edge date; Tue Dec 22 14:00:01 UTC 2020
2020-12-22 14:00:19 <ikke> https://tpaste.us/7gwg
2020-12-22 14:00:29 <l1x> thanks guys, bridge networking is functional again
2020-12-22 14:02:52 <kmmndr> https://tpaste.us/prBn
2020-12-22 14:03:10 <kmmndr> What could be wrong ?
2020-12-22 14:05:24 <COVID-1984> kmmndr: hardware clock?
2020-12-22 14:05:34 <COVID-1984> cyberattack apocayplse? :)
2020-12-22 14:06:08 <kmmndr> COVID-1984: Nice nick :-D
2020-12-22 14:06:13 <l1x> kmmndr: are you running this on a box that has RTC?
2020-12-22 14:06:37 <l1x> kmmndr: the call is implemented in the libc that returns the time
2020-12-22 14:06:41 <l1x> you can trace it
2020-12-22 14:07:47 <l1x> https://gist.github.com/l1x/87f08870fef5d7db86dd010fa536d8e9
2020-12-22 14:07:53 <l1x> what does this print?
2020-12-22 14:07:57 <l1x> for both
2020-12-22 14:12:25 <kmmndr> l1x: here are the results https://tpaste.us/g6yb
2020-12-22 14:15:00 <kmmndr> l1x: And here the same with `--privileged` to fix the operation not permitted error https://tpaste.us/ejNO
2020-12-22 14:17:43 <l1x> clock_gettime(CLOCK_REALTIME, {tv_sec=1608646405, tv_nsec=329329253}) = 0 interesting
2020-12-22 14:21:00 <l1x> this is correct
2020-12-22 14:24:58 <kmmndr> l1x: hum, does this means that I have a working/non-working RTC ?
2020-12-22 14:33:50 <Hello71> time64 is broken due to a docker bug
2020-12-22 14:38:16 <ikke> ah, seccomp related
2020-12-22 14:38:32 <Hello71> kmmndr: your host is out of date
2020-12-22 14:41:15 <ikke> kmmndr: the issue is that docker+seccomp intercept a syscall and return EPERM instead of ENOSYS
2020-12-22 14:47:40 <Hello71> it was fixed in docker and backported in ubuntu
2020-12-22 14:47:45 <Hello71> but you have to actually apply updates
2020-12-22 14:47:58 <kmmndr> Nice catch Hello71, ikke ! Thank you very much :-)
2020-12-22 14:48:56 <kmmndr> I'm using debian/armbian on the host ... I'll have to wait for update
2020-12-22 14:49:31 <kmmndr> Hello71: Why alpine:3.12 isn't affected ?
2020-12-22 14:50:21 <ikke> kmmndr: in the mean time, you should be able to load a custom updated seccomp profile
2020-12-22 14:59:40 <kmmndr> I think I would better build u-boot and kernel packages for my platform and install alpine on host
2020-12-22 15:03:37 <l1x> kolev_: the later, you need to be able to get to that syscall
2020-12-22 15:03:44 <l1x> i think
2020-12-22 15:03:53 <mps> kmmndr: what is your SBC
2020-12-22 15:23:13 <l1x> is it possible to have the bridge mac address the same as the eth0 mac address? if the bridge is only 1 member interface (eth0) it is the case. but when i add an extra device it changes.
2020-12-22 15:24:19 <AinNero> what for?
2020-12-22 15:30:16 <kmmndr> mps: this is an Helios4 (https://kobol.io/helios4/)
2020-12-22 15:30:34 <l1x> AinNero: i would like to set the br0 mac to the same as eth0 so it gets the static IP via DHCP i want
2020-12-22 15:30:39 <l1x> post-up ip link set br0 address 00:0a:e7:2c:44:2a
2020-12-22 15:31:02 <l1x> one way to get it done
2020-12-22 15:31:09 <l1x> kmmndr: no shit, you have one!
2020-12-22 15:31:19 <l1x> I wanted to order but the delay...
2020-12-22 15:32:18 <kmmndr> mps: I've already built u-boot and kernel packages for helios4, but it is missing optimizations (fans is working permanently for example)
2020-12-22 15:32:55 <kmmndr> l1x: I did bought 4 of them to spread my one datacenter / replication :-D
2020-12-22 15:33:09 <l1x> kmmndr: :D
2020-12-22 15:33:24 <l1x> have you had a chance to look at helios64?
2020-12-22 15:34:14 <kmmndr> As I have helios4, I haven't touched any helios64. But it looks good as well !
2020-12-22 15:43:31 <mps> kmmndr: aha, Marvell ARMADA 388
2020-12-22 15:45:18 <mps> not enabled in alpine kernels
2020-12-22 16:03:21 <kmmndr> mps: I know, but I did build my own packages for kernel and u-boot ... til official support
2020-12-22 16:04:55 <mps> kmmndr: can you look at linux-edge config-edge.armv7 and tell/post what needs to be enabled for this board
2020-12-22 16:09:44 <kmmndr> mps: oh nice, I'll do that ! Thanks for the tip
2020-12-22 16:23:11 <mps> kmmndr: np, also if you mind about u-boot please tell us
2020-12-22 16:42:43 <kpcyrd> I have a dumb question: https://wiki.alpinelinux.org/wiki/Configure_Networking#IPv4_Static_Address_Configuration says that starting with 3.13 the network needs to be configured differently, was that in a recent update? I prepared a system that runs edge because of some packages and is configured to do dhcp, but it doesn't show up in the network and I forgot to bring a serial cable to debug it :/
2020-12-22 16:43:29 <kpcyrd> (therefore a bit of a dumb question)
2020-12-22 16:47:07 <kpcyrd> (if the old syntax is still supported that shouldn't be the issue)
2020-12-22 16:59:59 <kpcyrd> ah, figured it out, if a usb stick is attached this causes boot issues somehow
2020-12-22 17:00:06 <kpcyrd> never mind, sorry :)
2020-12-22 17:09:09 <kpcyrd> is there a way to write a boot loader to the usb stick but mount /boot from the internal drive?
2020-12-22 17:11:30 <Cogitri> Sure, but the installer won't do that for you
2020-12-22 17:11:51 <Cogitri> But bootloaders can detect OSes on other disks too
2020-12-22 17:13:02 <kpcyrd> Cogitri: I'm a bit confused how the code in the mbr knows which partition to mount as /boot :)
2020-12-22 17:14:02 <kpcyrd> or, not mount as /boot, but which partition contains extlinux.conf
2020-12-22 17:16:59 <Cogitri> Ah, no clue how one would do it with syslinux, I only ever use rEFInd :)
2020-12-22 17:18:51 <kpcyrd> ah well, seems I need to order a serial cable unless I want to move /boot over to the usb stick. anyway, thanks! :)
2020-12-22 17:29:19 <Hello71> kpcyrd: mbr boot flag
2020-12-22 17:30:01 <Hello71> again, just use grub
2020-12-22 17:32:46 <mps> yes, mark boot partiotion as bootable
2020-12-22 17:33:23 <mps> and no, syslinux is quite good, I use it instead of grub wherever possible
2020-12-22 17:40:27 <kpcyrd> syslinux seems to search for a boot flag on the drive it loaded the bootloader from
2020-12-22 17:41:11 <mps> yes
2020-12-22 17:48:21 <l1x> hmm what could block incoming traffic on alpine using a bridged interface? outgoing traffic is working, only incoming is not
2020-12-22 17:50:19 <ikke> a bridge is layer 2
2020-12-22 17:50:28 <ikke> usually there is nothing blocking there
2020-12-22 17:53:12 <l1x> this is what i thought
2020-12-22 17:53:18 <MY-R> l1x: dhcp issue?
2020-12-22 17:53:29 <MY-R> give more context
2020-12-22 17:53:45 <MY-R> maybe isnt about bridge at all
2020-12-22 17:53:53 <l1x> so i have 2 interfaces, eth0 tap0 eth0 is dhcp ethernet, tap0 is for VMs
2020-12-22 17:53:58 <ikke> is there an interface accepting the traffic?
2020-12-22 17:54:09 <l1x> i want to bridge eth0 and tap0 into br0
2020-12-22 17:54:26 <l1x> if I have only eth0 it works as expected
2020-12-22 17:54:46 <l1x> once i add tap0 it starts to block incoming traffic
2020-12-22 17:58:08 <l1x> https://gist.github.com/l1x/0eb1a1b47cad27397fd476b572e21d59#what-doesnt
2020-12-22 17:58:30 <MY-R> if all those looks like one bridge then cant use br0 for VMs?
2020-12-22 18:00:52 <l1x> MY-R: good question. i am using a microvm. i have to do this -> --iface-id eth0 --guest-mac "AA:FC:00:00:00:01" --host-dev-name tap0
2020-12-22 18:01:15 <l1x> maybe it works with br0, but i thought seeing qemu  examples that people to it with tap0
2020-12-22 18:01:53 <MY-R> Im usually telling my qemu to use br0 so dont have to add anything in interfaces file
2020-12-22 18:02:18 <l1x> nice, let me try that
2020-12-22 18:05:27 <EF> Has anyone any suggestions to pull mail from a remote server and store/serve it locally (or via Wireguard)?
2020-12-22 18:06:13 <programmerq> if the mail is accessible via pop or imap, there are plenty of applications that can synchronize those messages to some other location.
2020-12-22 18:06:16 <MY-R> l1x: and you need load some modules before and echo 0 to arp stuff to get it work: https://wiki.alpinelinux.org/wiki/Bridge#Bridging_for_QEMU
2020-12-22 18:06:35 <l1x> yep did all that
2020-12-22 18:07:27 <MY-R> l1x: add to auto load modules this too: br_netfilter
2020-12-22 18:07:30 <EF> programmerq: could you make a couple of suggestions please? It is low volume with low number of trusted users. I can go start reading then. TIA
2020-12-22 18:07:49 <MY-R> l1x: so then your bridge-nf-call-iptables changes will be active
2020-12-22 18:08:19 <Hello71> ikke: ebtables is a thing but if you're using it you're probably doing something wrong
2020-12-22 18:08:30 <l1x> thanks MY-R
2020-12-22 18:08:46 <ikke> Hello71: hence the 'usually' :)
2020-12-22 18:08:59 <MY-R> l1x: cus without previously loaded br_netfilter that wont work: echo 0 > /proc/sys/net/bridge/bridge-nf-call-iptables
2020-12-22 18:09:45 <MY-R> and Im prety sure you need do: echo 0 > /proc/sys/net/bridge/bridge-nf-call-ip6tables AND echo 0 > /proc/sys/net/bridge/bridge-nf-call-arptables
2020-12-22 18:09:56 <MY-R> but that you can check if are disabled after boot
2020-12-22 18:13:44 <l1x> MY-R: that did the trick!
2020-12-22 18:14:03 <l1x> now i just need to figure out how to force the bride to have the same mac as the eth0
2020-12-22 18:14:09 <l1x> if that is possible
2020-12-22 18:14:55 <ikke> a bridge does not have a mac..
2020-12-22 18:15:08 <l1x> no?
2020-12-22 18:15:09 <ikke> individual ports on the bridge do
2020-12-22 18:15:26 <ikke> a bridge is like a switch
2020-12-22 18:15:27 <l1x> yes but there is a mac that is visible to the outside world
2020-12-22 18:15:38 <l1x> you can set that different ways
2020-12-22 18:15:59 <l1x> bridge_hw 00:22:aa:44:bb:cc
2020-12-22 18:16:04 <l1x> one way (not on alpine)
2020-12-22 18:16:35 <ikke> oh, apparently the bridge is an interface device itself
2020-12-22 18:16:43 <ikke> https://unix.stackexchange.com/a/319984/10643
2020-12-22 18:21:24 <ikke> "The bridge device listed with your other networking devices does not represent the virtual bridge, it represents a virtual NIC that is connected to the bridge."
2020-12-22 18:21:28 <l1x> yes
2020-12-22 18:21:32 <l1x> it has a mac
2020-12-22 18:21:47 <l1x> that you can influence many ways
2020-12-22 18:21:57 <MY-R> you running dhcp client on that bridge so it took mac from nic
2020-12-22 18:22:42 <l1x> yes but since i have 2 interfaces it uses the second (wrong) interface's mac
2020-12-22 18:22:44 <l1x> l1x@alpine ~ [124]> ifconfig eth0 | egrep -o 'HW.*'
2020-12-22 18:22:44 <l1x> l1x@alpine ~> ifconfig tap0 | egrep -o 'HW.*'
2020-12-22 18:22:44 <l1x> HWaddr 82:F3:EA:44:5E:04
2020-12-22 18:22:44 <l1x> HWaddr DC:A6:32:75:B5:66
2020-12-22 18:22:44 <l1x> l1x@alpine ~> ifconfig br0 | egrep -o 'HW.*'
2020-12-22 18:22:44 <l1x> HWaddr 82:F3:EA:44:5E:04
2020-12-22 18:23:52 <l1x> i am trying to get it to use eth0's mac
2020-12-22 18:23:57 <MY-R> just set good/correct one on interfaces which want
2020-12-22 18:24:10 <MY-R> then you will be sure that wont change in some random way
2020-12-22 18:25:00 <l1x> yeah, but here is the question, how do you do that on Alpine?
2020-12-22 18:25:07 <MY-R> hwaddress ether xx:xx:xx:xx:xx:xx
2020-12-22 18:25:26 <MY-R> I think that should do the work
2020-12-22 18:25:47 <MY-R> in your br0 section where got dhcp client running
2020-12-22 18:25:49 <MY-R> dont touch eth0
2020-12-22 18:26:25 <l1x> lets try
2020-12-22 18:27:11 <l1x> woo, yes that works! thanks MY-R!
2020-12-22 18:27:22 <MY-R> sweet, enjoy :)
2020-12-22 23:59:45 <kreios> I have a Raspberry Pi 4 that is acting as a NAT gateway between multiple vlans which is working but I am seeing duplicate packets that have the internal source address showing up on the external interface.  This is with edge.  Has anybody seen this before?
2020-12-23 00:01:36 <kreios> It is not all packets, it just a portion of them.
2020-12-23 08:22:05 <pmis> Hello everybody!
2020-12-23 08:22:29 <pmis> I have setup a virtual box with alpine linux.
2020-12-23 08:23:49 <pmis_> Therefore connection to the internet (e.g. to install packages
2020-12-23 08:23:55 <pmis_> does not work
2020-12-23 08:24:19 <pmis_> How can I permanently remove the default route for the host-only interface?
2020-12-23 08:57:57 <blubshep> So is the `apk search` command deprecated or something, it does not appear in the `apk -hv` output or the new apk manpages
2020-12-23 08:58:27 <ikke> blubshep: it's not deprecated
2020-12-23 08:59:14 <blubshep> any reason it does not appear, it's I found that change a little strange
2020-12-23 09:00:20 <ikke> Has to do with the introduction of man pages, and the automatic generation of help information
2020-12-23 09:00:44 <ikke> Best to open an issue here: https://gitlab.alpinelinux.org/alpine/apk-tools
2020-12-23 09:01:47 <blubshep> ah okay that makes sense
2020-12-23 09:07:28 <pmis> How can I permanently remove the default route for the host-only interface?
2020-12-23 09:08:57 <ikke> pmis: I think you need to prevent that route from being set in the first place
2020-12-23 09:09:12 <pmis> Ok, how do I do that?
2020-12-23 09:19:22 <pmis> @ikke: How do I prevent this route from beeing created?
2020-12-23 09:20:13 <ikke> I'm not sure what is responsible for adding it in the first place
2020-12-23 09:20:26 <ikke> I assume it's virtualbox that does this
2020-12-23 09:33:22 <pmis> I am not so sure this is a virtual box issue. I configured the interface with a static IP adress. However, I keep searching.
2020-12-23 09:33:56 <ikke> The only other thing that adds default routes is dhcp
2020-12-23 09:45:34 <pmis> eth0 is dhcp
2020-12-23 09:45:39 <pmis> eth1 is statuc
2020-12-23 09:46:02 <avih> Hi, I haven't updated my install for about two months, and now this command: `sudo apk upgrade -U -a' errors with: WARNING: Failed to perform initial self-upgrade, continuing with full upgrade.
2020-12-23 09:46:02 <avih> ERROR: unable to select packages:
2020-12-23 09:46:02 <avih>   ninja (virtual):
2020-12-23 09:46:02 <avih>     provided by: samurai
2020-12-23 09:46:02 <avih>     required by: world[ninja]
2020-12-23 09:46:56 <avih> i've been maintaining the same install (edge) for some years now, and it's the first time i'm seeing such failure. any ideas?
2020-12-23 09:47:55 <avih> (rarely adding/removing packages, mostly just issuing this upgrade command once in a while)
2020-12-23 09:51:36 <MY-R> avih: yesterday was somebody with same error and he couldnt install anything and not sure if he removed some weird pinned repository or something else
2020-12-23 09:51:56 <avih> i don't think i have pinned things. how do i check?
2020-12-23 09:52:05 <ikke> cat /etc/apk/repositories
2020-12-23 09:52:42 <avih> the only uncommented ones are main, community, testing
2020-12-23 09:52:51 <ikke> avih: does apk upgrade --available work?
2020-12-23 09:52:53 <avih> (on edge)
2020-12-23 09:52:58 <avih> sec
2020-12-23 09:53:27 <MY-R> "...continuing with full upgrade"
2020-12-23 09:53:29 <avih> isn't that exactly what my command does? (apk upgrade -U -a) ?
2020-12-23 09:53:48 <MY-R> what is output of "apk version"
2020-12-23 09:54:13 <avih> pretty long list, you want me to upload it somewhere?
2020-12-23 09:54:30 <MY-R> well, sounds like you didnt update for a long time
2020-12-23 09:54:40 <MY-R> what isnt nice idea with being on edge :/
2020-12-23 09:54:48 <avih> ikke: same error with your command
2020-12-23 09:55:05 <avih> i said i didn't upgrade for about two months. maybe 3, but not much more
2020-12-23 09:55:27 <MY-R> edge changing very fast
2020-12-23 09:56:02 <avih> more like two. linux firmware is 20201022
2020-12-23 09:56:15 <avih> ok, so what now?
2020-12-23 09:56:17 <MY-R> check if dont have some version of package enforced in /etc/apk/world
2020-12-23 09:56:44 <MY-R> normaly there should be just package names without version, check if got any with added version
2020-12-23 09:56:52 <avih> k, sec
2020-12-23 09:57:40 <avih> doesn't look to me like any of them has version
2020-12-23 09:58:29 <MY-R> and your mirror looks fine and updated? or using default one "cdn" ?
2020-12-23 09:59:11 <avih> where should i look at that?
2020-12-23 09:59:29 <MY-R> in /etc/apk/repositories
2020-12-23 10:00:06 <avih> i have this http://nl.alpinelinux.org/alpine/edge/main  (and same for community and testing)
2020-12-23 10:00:11 <b1d0x> Anyone have a clue why I am having this issue while attempting to make lrzip -> http://paste.debian.net/plain/1178046 ?
2020-12-23 10:01:01 <MY-R> avih: change to other mirror, looks like nl. isnt even on mirrors.alpinelinux.org anymore dunno why
2020-12-23 10:01:04 <ikke> b1d0x: are you trying to build a static binary?
2020-12-23 10:01:11 <avih> and that's /etc/apk/world: https://0x0.st/iC4_.txt
2020-12-23 10:01:14 <b1d0x> yeah
2020-12-23 10:01:31 <ikke> b1d0x: then you need the -static versions of those libs as well
2020-12-23 10:01:45 <ikke> MY-R: It's now just an alias for nl3.a.o
2020-12-23 10:01:51 <MY-R> avih: ninja package doesnt exist anymore too and it blocking upgrade
2020-12-23 10:02:02 <MY-R> ikke: ah ok, so should still work
2020-12-23 10:02:14 <ikke> yes
2020-12-23 10:02:40 <avih> MY-R: hmm.. thanks. strange, 1. it did work two months ago (though obviously it could have been removed later), but 2. shouldn't it printed some error which hints at the issue? the error seems unrelated...
2020-12-23 10:02:53 <MY-R> avih, remove ninja and use samurai
2020-12-23 10:03:09 <ikke> grep ninja /etc/apk/world
2020-12-23 10:03:29 <MY-R> avih: well it printed it in error message :)
2020-12-23 10:03:31 <avih> just remove it from /etc/apk/world?
2020-12-23 10:03:47 <MY-R> no
2020-12-23 10:03:53 <MY-R> remove it by apk del
2020-12-23 10:03:58 <avih> so there are two issues? one that the mirror is no more, and 2 that ninja/samurai thing?
2020-12-23 10:04:21 <MY-R> avih: mirror working fine, you could just update domain or change it to something else just in case
2020-12-23 10:04:31 <MY-R> problem is with package which was replaced by different one
2020-12-23 10:04:51 <MY-R> remove ninja, do upgrade and install samurai
2020-12-23 10:05:05 <MY-R> it will probably remove some deps but that will fix with installing samurai after
2020-12-23 10:05:24 <avih> MY-R: ikke: thanks, removing ninja (and i also added samurai) seems to have fixed that.
2020-12-23 10:05:33 <avih> is that something which can be expected when a package is removed?
2020-12-23 10:05:35 <ikke> Why do you have samurai installed?
2020-12-23 10:05:48 <MY-R> ikke: he didnt update edge since 2-3 months
2020-12-23 10:05:50 <ikke> (or ninja before)
2020-12-23 10:05:57 <ikke> just wondering
2020-12-23 10:06:00 <avih> i don't recall, i guess i needed it to build something?
2020-12-23 10:06:02 <MY-R> avih: https://gitlab.alpinelinux.org/alpine/aports/-/commit/2c7891f0cd17a3010261698b53e60bc4f2bbb468
2020-12-23 10:06:13 <b1d0x> ikke: it worked indeed! thanks
2020-12-23 10:06:29 <MY-R> avih: with edge need to be prepared and from time to time follow commits probably to avoid such a suprises
2020-12-23 10:06:54 <ikke> MY-R: apk should have been able to handle that itself though
2020-12-23 10:06:59 <ikke> not sure why it couldn't
2020-12-23 10:07:35 <MY-R> ikke: ye but when package got other deps then it block whole process, maybe he could manualy upgrade package but it never worked for me too well either with full system upgrades
2020-12-23 10:08:10 <avih> MY-R: even if i have follows commits (which i did use to do at some stage) would not make me think that upgrade would get broken
2020-12-23 10:08:40 <avih> but otherwise, no, it's not practical for me to follow commits.
2020-12-23 10:09:28 <MY-R> avih: well, yeah :) I had same problems from time to time when package was replaced or dropped or worst when back to older version, it happens and hard to debug it after to create decent issue on gitlab to try resolve it :\
2020-12-23 10:09:32 <ikke> The reason for the provides/replaces is to facilitate these upgrades
2020-12-23 10:10:23 <MY-R> but for sure warning/error messages could be wroten in much more readable way :P
2020-12-23 10:10:29 <MY-R> they can confuse form time to time
2020-12-23 10:11:29 <MY-R> ninja, virtual, samurai, sounds like some hide and seek game ;))
2020-12-23 10:12:55 <avih> ikke: ninja is required to build https://github.com/google/shaderc
2020-12-23 10:14:08 <avih> so, i did full upgrade, got this error (shows at the summary as one error):
2020-12-23 10:14:08 <avih> ERROR: libfm-extra-1.3.1-r0: trying to overwrite usr/lib/libfm-extra.so.4.1.2 owned by libfm-1.3.1-r1.
2020-12-23 10:14:08 <avih> ERROR: libfm-extra-1.3.1-r0: trying to overwrite usr/lib/libfm-extra.so.4 owned by libfm-1.3.1-r1.
2020-12-23 10:14:25 <avih> how should i resolve it?
2020-12-23 10:15:35 <ikke> samurai is a drop-in replacement
2020-12-23 10:16:05 <avih> did apk fix, ended up with "Reinstalling libfm-extra (1.3.1-r0)" (was it reverted ro -r0?), which fixed the issue.
2020-12-23 10:16:24 <avih> ikke: yeah, i presumed so, but you asked why i needed it, so i let you know
2020-12-23 10:16:33 <ikke> nod
2020-12-23 10:17:33 <avih> so what's the issue with libfm-extra? was it reverted from 1.3.1-r1 to 1.3.1-r0, and apk doesn't know how to handle it without using "fix"?
2020-12-23 10:18:07 <pmis> specified the interface eth1 without gateway and the default route is gone.
2020-12-23 10:19:04 <ikke> pmis: that makes sense, so you set it yourself
2020-12-23 10:21:13 <MY-R> avih: was some problem with circual dependencies so package -extra was created, check commits: https://gitlab.alpinelinux.org/alpine/aports/-/commits/master?utf8=%E2%9C%93&search=libfm
2020-12-23 10:21:53 <MY-R> avih: when you see some problem with package then is good to search it in commits and after in issues or pull requests
2020-12-23 10:23:10 <avih> so, on these cases (the libfm one, and the fact that ninja was replaced by samurai), i think apk could do better with the errors, maybe add some suggestion, like "try apk fix" (for libfm) or "seems you have package `ninja' installed, but it's no longer supported. try `apk del ninja' before you continue" or some such
2020-12-23 10:24:00 <avih> as it stands, while i'm sure the error message is correct, it doesn't really help fixing the issue.
2020-12-23 10:25:11 <avih> for ninja/samurai, maybe even add a field at the dpk db that ninja was replaced by samurai (either by keeping the ninja entry in some form, or by adding to the samurai package some notion that it replaces ninja)
2020-12-23 10:25:17 <avih> apk*
2020-12-23 10:26:53 <avih> actually, seems it already somehow knows that ninja is provided by samurai (judging by the initial error message i posted here).
2020-12-23 10:27:30 <avih> so it's only missing either automatic handling, or a recommendation on how to fix it
2020-12-23 10:35:24 <pmis> ikke: thanks for you help
2020-12-23 11:18:13 <mrsenshi> Hi, should we be expecting the first v3.13 release candidates to drop in the next month or so?
2020-12-23 11:20:38 <avih> MY-R: ikke: anyway, thanks again for your help. later.
2020-12-23 12:24:04 <gjabell> has anyone else had trouble upgrading linux-lts to 5.10.2?
2020-12-23 12:24:39 <gjabell> didn't seem to install properly the first  time so I did apk fix but it says the apk is unavailable
2020-12-23 12:28:49 <ikke> hmm, it's still available https://dl-cdn.alpinelinux.org/alpine/edge/main/x86_64/linux-lts-5.10.2-r0.apk
2020-12-23 12:30:24 <gjabell> hmm weird
2020-12-23 12:30:36 <gjabell> tried a couple of different mirrors but they all give the same message
2020-12-23 12:30:59 <gjabell> what's odd is if I install the linux-lts-dev package I get the right version
2020-12-23 14:24:14 <dsm42> gjabell: i just upgraded one of my boxes and it installed 5.10.2
2020-12-23 14:24:52 <gjabell> hmm that's bizarre
2020-12-23 14:24:55 <Hello71> gjabell: maybe corrupted cache
2020-12-23 14:25:01 <Hello71> rm /etc/apk/cache/*
2020-12-23 14:26:23 <gjabell> doesn't seem to change anything
2020-12-23 14:27:01 <gjabell> I wonder if it has something to do with also needing the zfs module; that also has an upgrade for 5.10.2 but also isn't installed
2020-12-23 15:37:34 <MY-R> gjabell: anything else in "apk version" output
2020-12-23 15:38:33 <gjabell> yeah both linux and zfs recognize that they are behind the repo versions (no additional packages)
2020-12-23 15:38:33 <MY-R> somebody before updated first "linux-lts-dev" package to have 5.10.2 kernel
2020-12-23 15:39:06 <gjabell> if I install the -dev package then I get the updated version, just not with the base packages
2020-12-23 15:40:51 <MY-R> some other package probably blocking, need check what depend on those both...
2020-12-23 15:41:06 <MY-R> gjabell: doing apk -i -a upgrade?
2020-12-23 15:41:59 <gjabell> just shows the usual OK message
2020-12-23 15:42:45 <gjabell> maybe something to do with the firmware?
2020-12-23 15:44:25 <MY-R> why, firmware cant upgrade too?
2020-12-23 15:46:07 <gjabell> well I just uninstalled linux-firmware-amdgpu and it installed a bunch of other packages from the linux-firmware group, which seems a bit odd
2020-12-23 15:46:19 <gjabell> reinstalling also removes them again
2020-12-23 15:49:06 <MY-R> ye it could be
2020-12-23 15:50:33 <MY-R> gjabell: be sure oyu have installed "linux-firmware" package
2020-12-23 15:52:36 <gjabell> yeah should be installed, doesn't seem to make a difference though
2020-12-23 15:53:17 <gjabell> the weirdest thing is that it can't seem to find the apks in the repo, even though it recognizes them in the manifest
2020-12-23 15:53:34 <MY-R> hmm?
2020-12-23 15:53:43 <MY-R> apk search something doesnt work?
2020-12-23 15:54:10 <gjabell> no that's all fine, but when I run apk fix linux-lts it says "APK unavailable, skipped"
2020-12-23 15:54:19 <gjabell> same for zfs-lts
2020-12-23 15:56:06 <MY-R> what arch you running and how installed?
2020-12-23 15:56:48 <gjabell> x86_64, installed a while back from a usb installer (desktop setup on edge)
2020-12-23 15:57:58 <MY-R> previous kernel updates did work? which kernel version you got now installed?
2020-12-23 15:58:50 <gjabell> I think so but I'm on 5.4.81 so maybe I just haven't noticed until now
2020-12-23 15:59:04 <gjabell> not sure how much the kernal lts version jumps in versions
2020-12-23 15:59:50 <MY-R> 5.4 sounds like you didnt upgrade to EDGE kernel at all, not even once
2020-12-23 16:00:39 <MY-R> so question is why it couldnt be done before when you done it with whole system :\
2020-12-23 16:01:09 <gjabell> hmm was under the impression that you could run lts kernel with edge repos
2020-12-23 16:01:41 <gjabell> ie adding the edge repos doesn't automatically change the kernel
2020-12-23 16:02:28 <MY-R> well, 5.10.x is new LTS now too
2020-12-23 16:03:01 <gjabell> does it matter if I'm running lts kernel on edge?
2020-12-23 16:06:21 <MY-R> ah well you got 5.81 and it was in EDGE repo so upgrade was fine but now something blocking
2020-12-23 16:06:28 <MY-R> 5.4.81*
2020-12-23 16:10:16 <MY-R> gjabell: do you have installed "exfat-utils" ?
2020-12-23 16:11:22 <gjabell> nope :/
2020-12-23 16:12:00 <MY-R> just checking so you also dont have it "exfatprogs" right? :)
2020-12-23 16:12:27 <gjabell> no, do have e2fsprogs tho
2020-12-23 16:15:34 <Hello71> apk add linux-lts=5.10.2-r0
2020-12-23 16:15:44 <Hello71> -s i guess
2020-12-23 16:16:08 <gjabell> hmm might be blocked by wireguard-lts
2020-12-23 16:16:33 <Hello71> why don't you ask apk to tell you instead of guessing
2020-12-23 16:17:19 <gjabell> well it lists wireguard-lts as a blocker, but I'm not sure whether that's the cause :P
2020-12-23 16:17:30 <gjabell> although it seems wireguard-lts isn't actually a package anymore
2020-12-23 16:17:37 <gjabell> so would make sense if it was related to that
2020-12-23 16:18:27 <MY-R> gjabell: omg
2020-12-23 16:18:33 <MY-R> remove that
2020-12-23 16:18:56 <MY-R> wireguard is built in in 5.10.x kernels, dont need external module anymore
2020-12-23 16:20:07 <MY-R> damn, next time I will be asking about full copy/paste logs :\
2020-12-23 16:21:57 <MY-R> thought he got no other informational output like me few times when packages didnt want upgrade
2020-12-23 16:22:04 <gjabell> alright that + uninstalling zfs and linux and reinstalling seems to have fixed it
2020-12-23 16:22:23 <gjabell> fwiw I didn't get any helpful information until the version-pinned install
2020-12-23 16:22:34 <MY-R> ah ok
2020-12-23 16:22:44 <gjabell> apk upgrade / add / etc only gave the APK unavailable message
2020-12-23 16:22:50 <MY-R> so ye, apk suck with that at all :)
2020-12-23 16:22:55 <MY-R> thanks Hello71!
2020-12-23 16:22:55 <gjabell> thanks for the help though y'all :)
2020-12-23 16:24:14 <MY-R> gjabell: but dont forget to unpin that version or you wont automaticly upgrade to next version... again... ;)
2020-12-23 16:24:55 <gjabell> thanks, it worked without the pin once I uninstalled zfs and linux-lts ;)
2020-12-23 16:29:25 <MY-R> gjabell: you did run that command once then better check if dont see any version here: grep linux-lts /etc/apk/world
2020-12-23 16:30:02 <gjabell> ah good point
2020-12-23 16:30:29 <gjabell> looks alright, thanks for the heads up tho
2020-12-23 16:32:55 <MY-R> cool! be glad for Hello71 trick :) I would already give up and start making idiot from myself :D
2020-12-23 16:40:26 <Hello71> I think probably linux-lts should have provides=wireguard-lts or something, like arch
2020-12-23 16:40:40 <ikke> yes, and replaces as well I guess
2020-12-23 16:40:50 <ikke> although, there are no files to take over
2020-12-23 16:41:07 <Hello71> no, the files are under kernel/ instead of extra/
2020-12-23 16:41:36 <Hello71> idk, still confused about apk provides/replaces vs pacman
2020-12-23 16:42:00 <ikke> I'm not fully certain myself how things work exactly
2020-12-23 17:17:03 <eydaimon> where do I see the output for my cronjobs ?
2020-12-23 17:17:57 <wsinatra> /var/log/messages
2020-12-23 17:18:18 <eydaimon> I don't see anything there tho. any keyword I need to grp for ?
2020-12-23 17:18:41 <eydaimon> nevermind
2020-12-23 17:18:45 <wsinatra> just grepping cron should work
2020-12-23 17:19:05 <wsinatra> if you want more verbose logging you can also install the syslog package, and then rc-update add syslog boot
2020-12-23 17:22:37 <eydaimon> thank you
2020-12-23 17:23:01 <wsinatra> don't mention it :)
2020-12-23 17:24:54 <eydaimon> any idea why I get an error when trying to save?  Can't write to backup file (add ! to override)
2020-12-23 17:40:09 <eydaimon> wsinatra: do you mean syslog-ng ?
2020-12-23 17:46:11 <eydaimon> I don' tsee anymore logging
2020-12-23 17:46:36 <eydaimon> wish it was easy to just use mail like other distros
2020-12-23 17:49:04 <wsinatra> ah yeah that's the name of the package my bad
2020-12-23 17:49:16 <mps> eydaimon: alpine don't strive to be like other distros
2020-12-23 17:49:36 <mps> or sysklogd
2020-12-23 17:49:56 <mps> but busybox applets also works
2020-12-23 17:59:58 <wsinatra> eydaimon: what exactly are you trying to get setup?
2020-12-23 18:09:35 <xordspar0> The avahi-daemon service depends on the dbus service, but the avahi package only depends on dbus-libs. dbus-openrc contains the service. Shouldn't avahi also depend on dbus-openrc?
2020-12-23 18:11:41 <ikke> -openrc packages are automatically installed if both the main and openrc packages are present
2020-12-23 18:15:17 <xordspar0> I see. Should avahi depend on dbus then?
2020-12-23 18:28:56 <Hello71> i think you can run avahi in client mode
2020-12-23 19:04:20 <maldridge> in the boot process is rc.local processed after mounts are complete?
2020-12-23 19:29:19 <mps> maldridge: rc.local is after all in default runlevel
2020-12-23 19:29:56 <mps> and you mean local.d
2020-12-23 19:29:58 <mps> ?
2020-12-23 19:30:51 <mps> '  after *'
2020-12-23 19:33:39 <maldridge> to un-XY this, I need to --make-shared on a filesystem, and I need that to happen after it mounts
2020-12-23 19:33:46 <maldridge> so I'm trying to figure out where to add tihs
2020-12-23 19:38:48 <bokomaru> Instead, could you use the "shared" option in your fstab?
2020-12-23 19:39:21 <bokomaru> It would require util-linux. I've never done this, but I see it as an "experimental" feature in the util-linux mount man page
2020-12-23 19:46:55 <eydaimon> wsinatra: trying to get so I know if a cronjob fails or not
2020-12-23 19:47:18 <eydaimon> mps: that's fine as long as the goal is to be easier and better than other distros
2020-12-23 19:47:54 <eydaimon> mps: just different for the sake of different and running into the unexpected consumes a lot of time for the admin
2020-12-23 19:50:35 <maldridge> bokomaru: I don't think that applies to /dev
2020-12-23 20:15:23 <mps> eydaimon: alpine moto is 'small, simple, secure'. and it is easy as consequence of that, but 'easy' could mean different things to different people
2020-12-23 20:21:00 <bokomaru> maldridge: Hmm I don't think devtmpfs is otherwise a special case, it's just not in Alpine's fstab. Two ideas I have: 1) Add it to the mountopts="exec,nosuid,mode=0755" in /etc/init.d/devfs to make it mountopts="exec,nosuid,mode=0755,shared", or 2) something like  "none /dev devtmpfs remount,shared 0 0"  in fstab
2020-12-23 20:21:46 <maldridge> I'll give those a try
2020-12-23 20:24:02 <bokomaru> I _think_ (2) is safe, because the dependencies would go something like localmount -> fsck -> dev -> dev-mount. So /dev should already be mounted before fstab is used, so the remount should succeed
2020-12-23 20:24:31 <mps> hm, /dev          devtmpfs devtmpfs rw,nosuid,relatime,size=10240k,nr_inodes=501059,mode=755
2020-12-23 20:24:49 <mps> it is mounted automatically
2020-12-23 20:25:03 <maldridge> right, but not with the options I want
2020-12-23 20:25:13 <ikke> probably already in initramfs?
2020-12-23 20:25:21 <mps> ofc, you can add it to /etc/fstab
2020-12-23 20:25:40 <mps> ikke: no, I don't use initramfs on this machine
2020-12-23 20:25:48 <ikke> ok
2020-12-23 20:30:06 <bokomaru> Regardless of initramfs or not, automatic or not, I'm pretty sure that /etc/init.d/devfs tries to mount or remount devtmpfs on /dev anyways
2020-12-23 20:32:03 <ikke> hmm, mountopts is hardcoded in there
2020-12-23 20:32:23 <bokomaru> yep
2020-12-23 20:33:33 <ikke> but is it actually being usedA
2020-12-23 20:34:43 <ikke> I feel like it's not actually being looded
2020-12-23 20:35:04 <bokomaru> Hmm don't know
2020-12-23 20:37:01 <MY-R> from "/etc/conf.d/devfs": "1. If there is an entry for /dev in fstab, it will be used."
2020-12-23 20:37:52 <ikke> right, the mountopts is a fallback
2020-12-23 20:39:35 <bokomaru> oh there ya go
2020-12-23 20:39:37 <maldridge> so the reigning theory is to configure it in fstab and that should work
2020-12-23 20:41:25 <MY-R> # 2. If devtmpfs is defined in the kernel, it will be used.
2020-12-23 20:41:26 <MY-R> # 3. If tmpfs is defined in the kernel, it will be used.yyesY
2020-12-23 20:42:55 <steinbach> any chance to get digikam on alpine linux?
2020-12-23 20:43:13 <steinbach> into the repos
2020-12-23 20:59:39 <hendursaga> I can't wait for the next release, regardless..
2020-12-23 21:06:05 <maldridge> wll, putting it in fstab I now have no /dev anymore
2020-12-23 21:10:26 <Hello71> without initramfs the kernel is supposed to mount devtmpfs
2020-12-23 21:10:31 <Hello71> https://git.alpinelinux.org/aports/tree/main/linux-lts/config-lts.x86_64#n2055
2020-12-23 21:11:23 <bokomaru> I just tried "none /dev devtmpfs shared 0 0" in my fstab
2020-12-23 21:12:07 <bokomaru> I get /dev mounted, and findmnt -o+PROPAGATION shows "shared"
2020-12-23 21:16:00 <bokomaru> It did print an error message "mount: /dev: none already mounted or mount point busy". When I added "remount" to the fstab options, that error message went away, and I still get a good /dev mount with shared enabled.
2020-12-23 21:19:01 <maldridge> maybe I did something wrong, I'll reboot
2020-12-23 21:19:44 <maldridge> I have none /dev devtmpfs rw,nodev,relatime,size399096k,mode=755,shared,remount 0 0 as the line in my fstab
2020-12-23 21:19:48 <MY-R> maybe need regenerate initramfs
2020-12-23 21:20:12 <MY-R>  /dev nodev
2020-12-23 21:21:06 <bokomaru> I didn't have to regenerate my initrfams. Maybe try it without those extra options like "size399096k"
2020-12-23 21:21:10 <maldridge> hmm, I'm trying to do a lot of this inside of ansible, I suppose I can use a trigger to redo the initramfs
2020-12-23 21:21:21 <maldridge> I mainly just pulled in the options that were there when it was already mounted
2020-12-23 21:22:21 <MY-R> but is "nodev" option in /dev correct?
2020-12-23 21:22:56 <bokomaru> No I don't think so :-)
2020-12-23 21:23:22 <MY-R> I thought nodev reseting other settings
2020-12-23 21:24:22 <maldridge> wow just realized I was looking at totally the wrong filesystem, I need run to be shared, not dev
2020-12-23 21:24:59 <bokomaru> Yeah so when I add "size399096k", I get a "bad option" error. That's one's bad too
2020-12-23 21:25:06 <bokomaru> unsurprisingly :-)
2020-12-23 21:25:10 <maldridge> alright, going to change that to be just shared,remount
2020-12-23 21:25:18 <maldridge> and fix this to be just a tmpfs and for /run
2020-12-23 21:26:44 <MY-R> :D
2020-12-23 21:27:14 <MY-R> ahh those dirty glasses or bubbles in beer ;)
2020-12-23 21:27:31 <maldridge> or being distracted by my sandwich being delivered
2020-12-23 21:27:44 <MY-R> ah ye, hunger is worst
2020-12-23 21:28:36 <maldridge> ok now I have:
2020-12-23 21:28:40 <MY-R> bon appetit maldridge
2020-12-23 21:28:49 <maldridge> none /run tmpfs shared,remount 0 0
2020-12-23 21:28:58 <maldridge> and it is shown as still being private
2020-12-23 21:29:23 <maldridge> eh I'll eat in a minute
2020-12-23 21:31:54 <bokomaru> Maybe the problem there is /run is a bind mount?
2020-12-23 21:32:13 <MY-R> shoudlnt be: tmpfs /run tmpfs rw,nodev,... 0 0 ?
2020-12-23 21:32:15 <maldridge> well, I am trying to bind it into a container, and this is failing because its not shared
2020-12-23 21:32:38 <MY-R> and then bind it
2020-12-23 21:33:23 <maldridge> alright, going to go quickly eat a sandwich and then I'll poke at this again
2020-12-23 21:33:34 <bokomaru> So if your only purpose is to make a directory that you can use as a root for namespace mounts or whatever
2020-12-23 21:33:58 <MY-R> wait, what container, docker?
2020-12-23 21:34:36 <maldridge> yes, I'm trying to mount /run into a docker container so I can see the docker control socket from inside that container
2020-12-23 21:35:28 <MY-R> shouldnt be that set up when running container? with --mount
2020-12-23 21:36:02 <MY-R> or doing like that and scream about "shared"?
2020-12-23 21:36:28 <bokomaru> Are you trying to control the host docker daemon from inside one of its containers? Maybe not a good idea?
2020-12-23 21:36:38 <MY-R> https://docs.docker.com/storage/bind-mounts/  is kind of confusing
2020-12-23 21:37:54 <MY-R> bokomaru: maybe isnt so bad because there is plenty of options to set up like: shared, slave, rshared, rslave etc
2020-12-23 21:38:03 <MY-R> there are*
2020-12-23 21:42:53 <maldridge> MY-R: I was not aware that the mount syntax would make it shared, I can look into that
2020-12-23 21:43:27 <maldridge> bokomaru: well this entire thing is a sacrificial architecture, but the express purpose is to deploy a container so that version number is decoupled from alpine, and then to deploy more containers with it
2020-12-23 21:44:57 <MY-R> maldridge: I think docker will do own "magic" when you will set up those stuff when running container (--mount)
2020-12-23 21:45:08 <maldridge> hmm, I'll give that a try
2020-12-23 21:45:59 <maldridge> presently I'm using a volume mount with :shared, but I will make /run a mount instead
2020-12-23 21:46:40 <bokomaru> Might not matter now, but ironically, I tried your "none /run tmpfs shared,remount 0 0" in my fstab. I get a good /run mount as "shared". So your problem is either that something else is different, or that whatever error you're seeing is actually caused by something else
2020-12-23 21:46:54 <MY-R> ye there are few ways to do it and every little different, there is even --tmpfs
2020-12-23 21:47:12 <MY-R> bokomaru: or his docker already overwriting stuff right after start
2020-12-23 21:49:51 <MY-R> or maybe like last time with ikke (sorry for highlight) when he was set up network stuff but linode got own scripts to do it and was overwriting his changes
2020-12-23 21:50:43 <maldridge> well, there are no other scripts on startup, but I'm not entirely sure what's wrong
2020-12-23 21:50:49 <maldridge> I'll check further once I finish my sandwich
2020-12-23 21:51:51 <MY-R> imho changes in host should be minimal and mostly docker doing such a things (and usually do) so should start from there
2020-12-23 21:54:42 <bokomaru> MY-R: Yeah I pretty much agree with that, shouldn't usually need to be messing about w/ the host
2020-12-23 21:55:10 <bokomaru> I'm pretty sure that one of the Linux namespace technologies that docker uses requires a mountpoint to be shared. For fun, I installed docker, did "service start docker" and "service stop docker"
2020-12-23 21:55:40 <bokomaru> Verified w/ findmount -o+PROPAGATION, docker actually does remount /run with "shared" on start. Then it restores it to "private" on stop.
2020-12-23 21:56:39 <MY-R> so, dont bother with host, focus on docker, right? :P
2020-12-23 21:57:29 <bokomaru> Yeah. I'm assuming the problem is with "so I can see the docker control socket from inside that container"
2020-12-23 21:57:53 <maldridge> yeah without run as shared I get bad fs magic
2020-12-23 22:00:18 <bokomaru> Not sure there's much more I can say, I'll prob drop off soon. This might be a better problem/question to ask on a docker chatroom or mailing list though
2020-12-23 22:00:47 <maldridge> sure, my original assumption was to just fix this as a mount at boot with alpine
2020-12-23 22:01:12 <maldridge> its interesting that docker makes it shared on startup, I wonder why this isn't working
2020-12-23 22:03:49 <MY-R> well, like bokomaru mentioned before that wouldnt be wise to give container full access to host /run etc so probably need take extra special steps to make it work
2020-12-23 22:04:06 <MY-R> and docker wanna be very smart
2020-12-23 22:04:18 <maldridge> the intent really is to give the container access to everything
2020-12-23 22:04:37 <maldridge> its already privileged pid=host net=host
2020-12-23 22:05:04 <MY-R> that for sure need extra work to do
2020-12-23 22:05:26 <maldridge> a container that starts more containers is sufficient for my purposes
2020-12-23 22:05:38 <maldridge> I now just need it to survive reboots, which depends on the shared mounts all coming back
2020-12-23 22:07:28 <MY-R> in general "host" should be unbreakable by containers so every time when give them more power then is hard to keep that rule on :\
2020-12-23 22:07:59 <maldridge> given that my alternative is taking the obsolete version of nomad that alpine packages, I'll take the tradeoff
2020-12-23 22:12:33 <MY-R> nomad is in edge so would be enough to send MR
2020-12-23 22:13:25 <maldridge> correct, but unless alpine is planning to keep multiple versions of packages in the mirrors, I'm then beholden to keeping my cluster in lockstep with whatever version is packaged at the moment
2020-12-23 22:13:36 <maldridge> I'm sure you can see why that would be stressful and undesireable
2020-12-23 22:13:45 <MY-R> ye can imagine
2020-12-23 22:14:53 <MY-R> there were some talk about keeping packages in few versions but that for sure is in early stage and need more people to work on it
2020-12-23 22:14:59 <MY-R> talks*
2020-12-23 22:16:03 <maldridge> realistically the next stage for this cluster is to move off of alpine
2020-12-23 22:16:26 <maldridge> but alpine provided nice flexibility when I was developing
2020-12-23 22:17:40 <steinbach> The difference between standart, except for the packages... says that extended runs from ram. Is there any performance improvement, and if so... what is the difference of ram usage?
2020-12-23 22:21:12 <steinbach> How much more ram does it use? I'm willing to sacrifice some ram if the performance boost is worth it.
2020-12-23 22:21:28 <eydaimon> mps: so how can I see errors from cronjobs ?
2020-12-23 22:21:29 <MY-R> maldridge: Im pretty sure there is some simple thing to fix that shared "issue" just need spend more time to figure it out and also can open "issue" in Alpine gitlab or ML about supporting multiple major versions of software and try to cooperate
2020-12-23 22:23:48 <MY-R> steinbach: are you thinking to run whole Alpine system in ram? the only boost which get will be less time to wait when program will run for first time
2020-12-23 22:25:32 <steinbach> MY-R: Not the whole system, I do not know what componentes are running from ram in the extended version. It just says "runs from ram" on the extended version download "field". I think I might just stay standard.
2020-12-23 22:25:38 <maldridge> MY-R: I can, but realistically alpine is the wrong choice for what I'm trying to do.  Alpine is a great self-hosting OS that works really well at that, but what I need here is a kernel and docker, nothing else not even a package manager.
2020-12-23 22:28:32 <MY-R> steinbach: there is diskless mode which put/install all packages you need in early boot and stays in your ram (all componentes, so you can remove your disk) but that is good for devices which dont make many changes in running system etc
2020-12-23 22:29:35 <MY-R> steinbach: but you probably wanted just install Alpine on disk and it doesnt matter what you pickup before installation, if you select to run from ram then you can basicly use same media from which boot to instal something on it since all stuff will be already in your memory
2020-12-23 22:30:41 <MY-R> maldridge: yeah, so need focus probably more on docker and their "magic" :)
2020-12-23 22:33:35 <xordspar0> As someone who is not an Alpine developer, how can I get something into aports? Can I just send a patch? Do I need to convince someone to take on maintainership in a mailing list?
2020-12-23 22:36:55 <MY-R> xordspar0: https://gitlab.alpinelinux.org/alpine/aports you can create correct MR - Merge Requests (preferable) or you can create ISSUE and ask to create package for you
2020-12-23 22:37:24 <mps> eydaimon: I cannot debug your problems. I only can tell that syslog works and cron works, I'm sure because I use them on many machines
2020-12-23 22:37:44 <xordspar0> MY-R: Thanks
2020-12-23 22:38:16 <MY-R> xordspar0: then sooner or later it can appear in Alpine Edge in testing repository, after testing you can ask to move it to community so it could be a part of next release
2020-12-23 22:44:51 <MY-R> xordspar0: but ye if your package wont have any "stable" maintainer who will take care about updates then will end up as "unmaintained" and wont even appear in releases, so better way is to just create own local repository but patch/MR wont hurt to send :)
2020-12-23 22:46:04 <xordspar0> Yeah, I'll probably send a patch and meanwhile start using it from my local repo.
2020-12-23 22:48:54 <MY-R> xordspar0: ye, dont afraid to send, there are always nice people who will try help to create nice APKBUILD :)
2020-12-23 22:55:34 <steinbach> MY-R: im going the traditional way
2020-12-23 22:56:35 <MY-R> steinbach: sounds good :)
2020-12-23 23:32:13 <yjftsjthsd> Is there any documentation around variables that apk notices? I was poking at telling it to use a proxy and `grep -a PROXY /sbin/apk` shows a few hits, but https://gitlab.alpinelinux.org/alpine/apk-tools/-/blob/master/doc/apk-add.8.scd / https://gitlab.alpinelinux.org/alpine/apk-tools/-/blob/master/doc/apk.8.scd does not
2020-12-23 23:54:27 <eydaimon> mps: I'd write the wiki on how to set it up if I got it working
2020-12-24 00:57:18 <eydaimon> mps: iow, I just want to know what the errors are from the output. all it's currently telling me is that it was executed. no stdout or stderr is reported
2020-12-24 01:20:37 <xordspar0> Can you use mkimage to make an install iso that boots with sshd running? I figured out how to install ssh, but I don't see how to enable services.
2020-12-24 01:29:00 <xordspar0> I'm following the "How to make a custom ISO image" wiki page, which helps a lot but it doesn't say how to have a service start at boot or if that's possible.
2020-12-24 01:31:38 <yjftsjthsd> Can you add a apkovl with that change?
2020-12-24 01:31:48 <yjftsjthsd> (wildly guessing; I don't actually know)
2020-12-24 03:17:03 <Hello71> either add an ovl, change the initramfs, or add a package owning /etc/runlevels
2020-12-24 03:17:13 <Hello71> in approximate order of technical soundness
2020-12-24 05:28:12 <xordspar0> yjftsjthsd, Hello71: Well, I gave apkovl a shot and it is much simpler than trying to make my own image from scratch. It really is the right solution for this problem. Thanks! It works great in qemu, but when I try it on my headless server I don't see port 22 open up. Oh well, I guess I'll need to put a graphics card in it and see what's going on after all.
2020-12-24 06:59:18 <mps> eydaimon: try to run crond in foreground and log to stderr to see first does it actually invoke scripts
2020-12-24 10:33:28 <Apachez> merry xmas!
2020-12-24 10:34:33 <caskd> merry xmas all!
2020-12-24 10:34:46 <mps> thanks
2020-12-24 10:35:27 <mps> every day is holiday, btw
2020-12-24 10:39:44 <artok> yes, thanks to that chinese bat-eater!
2020-12-24 10:40:42 <mps> artok: you believe in fables :)
2020-12-24 10:41:06 <artok> umm.. no =)
2020-12-24 11:36:39 <vits-test> thank you for nomodeset being default.  it is so so cool on tty + qemu.
2020-12-24 11:38:17 <ikke> ..
2020-12-24 12:53:15 <fys> Hola!
2020-12-24 12:53:17 <fys> :)
2020-12-24 12:53:51 <fys> Just popping in after playing with alpine on a pcengines apu2 board
2020-12-24 12:54:04 <fys> I gotta say, really impressive stuff
2020-12-24 12:54:15 <fys> ain't seen a "new" distro i liked in a loooong while lol
2020-12-24 13:02:51 <Hello71> xordspar0: did you remember to configure the networking? it's not sufficient to just enable sshd
2020-12-24 14:14:48 <hechos> I want to thank the entire channel for the help they provide throughout this year and wish you a very merry Christmas, peace love with your loved ones
2020-12-24 14:14:58 <fys> \o/
2020-12-24 14:15:48 <hechos> and please dev update qutebrowser to lastest version and mpv work very buggy on musl port.
2020-12-24 14:16:32 <hechos> i know. im using alpine to workstation os.
2020-12-24 14:17:40 <hechos> may it be not good idea, but on server side work awesome.
2020-12-24 14:18:42 <tehcloud> I use it as a workstation
2020-12-24 14:24:42 <tehcloud> I think that use case is less common for the distro, but it works for me and brings back the joys of using *nix in the late 90s and early 2000s without the painful parts of that
2020-12-24 14:28:35 <hechos> tehcloud: for work fine, but qutebrowser render crashes on musl ports, im user of voidlinux too and musl ports of void render crashed.
2020-12-24 14:29:02 <hechos> but last version v1.14.1 work much more stable.
2020-12-24 14:29:06 <tehcloud> I use this distro partially because of that, and I fix these problems
2020-12-24 14:29:14 <tehcloud> then hope upstream will take the fixes
2020-12-24 14:29:34 <tehcloud> more of us using musl distros will lead to higher quality code
2020-12-24 14:29:59 <fys> i am sticking with devuan on my desktops for now
2020-12-24 14:30:10 <fys> but alpine is perfect so far for these little apu boards
2020-12-24 14:30:18 <hechos> fys: im tired about debian.
2020-12-24 14:30:38 <hechos> i started used slackware.
2020-12-24 14:31:14 <hechos> on beginning but switch to debian almost a 1 year
2020-12-24 14:31:27 <fys> i was using gentoo before that but i got tired of compiing firefox and by the time i switched to using a binary copy of Chrome my system was so cluttered and janky i just wiped it
2020-12-24 14:31:38 <hechos> and then study how things work and jump to archlinux
2020-12-24 14:32:05 <tehcloud> I tried Arch when I realized I had to stop using Debian, but only because I had an nvidia card at that time
2020-12-24 14:32:11 <hechos> all gone fine, but systemd is comming and finally switch to void, and freebsd
2020-12-24 14:32:24 <tehcloud> now I can use open source drivers for my GPU so it's Alpine
2020-12-24 14:32:24 <fys> that machine is now an esxi host and now im on devuan for desktop, alpine for server/embedded stuff
2020-12-24 14:32:47 <hechos> im got old hardware nvidia ones, on my house.
2020-12-24 14:32:55 <fys> that's why i switched to devuan from debian
2020-12-24 14:33:04 <fys> i will used systemd at work but i won't at home
2020-12-24 14:33:08 <fys> use*
2020-12-24 14:33:28 <hechos> nvidia drivers are mature but nvidia is so bad to opensource. but recently amd go turns... too
2020-12-24 14:34:01 <hechos> runit work nicely too.
2020-12-24 14:34:37 <fys> i found a happy medium in my life a long time ago.. i use as open source/space saving gear as i can for my work/hobbyist stuff and for my media/enjoyment .. i have a pc that boots into Stram big picture
2020-12-24 14:34:41 <fys> lol
2020-12-24 14:34:42 <hechos> well i like but not part have to compiled.
2020-12-24 14:34:53 <hechos> im talk about gentoo
2020-12-24 14:35:02 <fys> yeah
2020-12-24 14:35:07 <tehcloud> I did the Gentoo thing for about 3 years
2020-12-24 14:35:18 <hechos> a lot of work.
2020-12-24 14:35:21 <tehcloud> then I stopped having time to wait for everything to build
2020-12-24 14:35:23 <fys> alpine is exactly what i was looking for
2020-12-24 14:35:31 <hechos> but still remember 1980 for amiga times...
2020-12-24 14:35:38 <fys> pfsense -> alpine happened very seamlessly
2020-12-24 14:35:47 <fys> 1980 was Commodore times
2020-12-24 14:35:56 <fys> 1990s was Amiga times
2020-12-24 14:36:03 <hechos> yeah, but play games on falcon ataris
2020-12-24 14:36:27 <fys> well im american so we didn't relly get the micro market like europe
2020-12-24 14:36:28 <hechos> two company is a shame to died, like BeOS
2020-12-24 14:36:35 <tehcloud> <fys> well im american so we didn't relly get the micro market like europe
2020-12-24 14:36:37 <tehcloud> thank God
2020-12-24 14:36:42 <fys> why thank god
2020-12-24 14:36:54 <tehcloud> because it's less for me to have to emulate
2020-12-24 14:36:58 <fys> lol
2020-12-24 14:37:02 <fys> dummy
2020-12-24 14:37:18 <hechos> lol.
2020-12-24 14:37:21 <tehcloud> oi bruv, ye 'aven't lived til' you played the ZX Speccy!
2020-12-24 14:37:24 <fys> i could probably use alpine on my desktop quite easily
2020-12-24 14:37:32 <fys> and i might try it this weekend
2020-12-24 14:37:33 <hechos> well for a kid star playing with colecovision.
2020-12-24 14:37:49 <tehcloud> hechos, I wrote a ColecoVision emulator :)
2020-12-24 14:37:53 <hechos> lol wooh nice times. to be a kid.
2020-12-24 14:38:05 <fys> as long as i can compile my terminal emulator, dwm and qt5
2020-12-24 14:38:05 <hechos> tehcloud: name?
2020-12-24 14:38:20 <hechos> fys: here all suckless.
2020-12-24 14:38:30 <fys> im mostly suckless
2020-12-24 14:38:40 <fys> surf doesn't fit my needs
2020-12-24 14:38:51 <fys> and i like my fork of termit over st
2020-12-24 14:39:16 <fys> but i agree with the suckless mindset
2020-12-24 14:39:21 <hechos> aha.
2020-12-24 14:44:46 <fys> dangit tehcloud
2020-12-24 14:44:55 <fys> now i want to try to put alpine on my desktop
2020-12-24 14:45:00 <fys> >:(
2020-12-24 14:45:06 <tehcloud> do it, it's the only way to live
2020-12-24 14:45:18 <tehcloud> and you can help expose bugs to be fixed
2020-12-24 14:45:21 <fys> ive got a 30gb msata drive attached to a usb adapter
2020-12-24 14:45:29 <fys> i could do a quick install to it
2020-12-24 14:46:14 <fys> i just need to read more documentation before i feel comfortable doing that
2020-12-24 14:46:37 <fys> been studying it and making notes throughout my process of putting it on this apu2 board
2020-12-24 14:47:00 <fys> Luckily some saint did most of the leg work recently
2020-12-24 14:47:01 <fys> https://wiki.alpinelinux.org/wiki/Alpine_Install:_from_a_disc_to_PC_Engines_APU
2020-12-24 14:50:35 <fys> I was actually concerned with how long I could could reasonably use these boards as pfsense was a little heavy for them.
2020-12-24 14:50:52 <fys> Now with alpine on them, I feel a lot more comfortable.
2020-12-24 14:50:53 <fys> lol
2020-12-24 15:04:34 <fys> setting up an alpine dev envrionment on my esxi host now..
2020-12-24 15:04:46 <fys> i hate not being able to let things go
2020-12-24 15:04:50 <fys> :|
2020-12-24 15:07:39 <tehcloud> be happy, you are going deeper down the rabbit hole of what is currently the best distro
2020-12-24 15:08:08 <fys> well that's why im setting up a dev envrionment
2020-12-24 15:08:16 <fys> that's where things reveal themselves
2020-12-24 15:08:30 <fys> things are always rosey until you need to compile something!
2020-12-24 15:08:34 <fys> :P
2020-12-24 15:32:42 <fys> welp
2020-12-24 15:32:54 <fys> dwm compiled fine, let's see if my termit fork does as well
2020-12-24 15:37:03 <fys> aw fudge
2020-12-24 15:37:08 <fys> no vte package?
2020-12-24 15:37:20 <fys> oh
2020-12-24 15:37:22 <fys> nm
2020-12-24 15:37:32 <fys> i just noticed i only have one repository enabled. (:
2020-12-24 16:01:46 <fys> hm
2020-12-24 16:01:55 <fys> attempting to remove packages seems to do nothing, weird
2020-12-24 16:02:08 <tehcloud> it's just fast
2020-12-24 16:02:13 <tehcloud> check the filesystem
2020-12-24 16:02:21 <fys> lotus:/home/fys/termit/build# apk del --force dconf
2020-12-24 16:02:21 <fys> OK: 1932 MiB in 344 packages
2020-12-24 16:02:21 <fys> /usr/bin/dconf
2020-12-24 16:02:21 <fys> lotus:/home/fys/termit/build# which dconf
2020-12-24 16:02:21 <tehcloud> probably all gone
2020-12-24 16:02:34 <tehcloud> hmmm... is that a symlink maybe?
2020-12-24 16:02:37 <fys> nope
2020-12-24 16:02:39 <fys> binary works
2020-12-24 16:03:42 <fys> oh boy
2020-12-24 16:03:55 <fys> i think i might have this figured out. ignore my comment for now.
2020-12-24 16:17:20 <mps> fys: apk fix dconf
2020-12-24 16:17:40 <fys> thanks!
2020-12-24 16:18:23 <mps> doesn't work always
2020-12-24 16:19:08 <fys> worked
2020-12-24 16:19:10 <fys> thanks :)
2020-12-24 16:30:07 <xordspar0> Hello71: I did remember networking. This issue turned out to be the BIOS was picking the wrong disk to boot. You can make the fanciest, highest-tech rescue OS, but it doesn't help you get past the BIOS. :)
2020-12-24 16:31:54 <xordspar0> Pro tip: If you are working on a headless machine and it boots in the the wrong OS and you *do* have ssh access, kexec is awesome for getting you into the right OS.
2020-12-24 16:33:04 <xordspar0> hechos: Merry Christmas to you too!
2020-12-24 16:42:30 <xordspar0> romanos: And you as well!
2020-12-24 20:35:25 <toogley> hey, i am unsure whether i want a BIOS boot or UEFI. i tend to favor BIOS boot because the UEFI spec and everything seems to be more complicated but i don't know.
2020-12-24 20:35:51 <toogley> i would prefer a stable, minimalistic and secure way to boot, its for a laptop
2020-12-24 20:38:09 <mps> bios
2020-12-24 20:39:02 <toogley> mps: can you elaborate why?
2020-12-24 20:39:21 <mps> you already told: simple
2020-12-24 20:39:49 <mps> stable, minimalistic and secure => simple
2020-12-24 20:40:33 <toogley> ah, but how is bios boot more simple? i understand  a few things but not much of it.
2020-12-24 20:45:08 <nacelle> "secure" and bios booting? no.
2020-12-24 20:45:28 <nacelle> (secure and uefi booting? questionable.)
2020-12-24 20:45:53 <toogley> nacelle: what is the alternative?
2020-12-24 20:46:27 <nacelle> question your reality?
2020-12-24 20:47:05 <toogley> hm, with secure i mean reasonable secure, like most i can get out of this laptop without coreboot
2020-12-24 20:48:21 <nacelle> so i would drop the secure part of your context.  its like saying "I want a secure house."
2020-12-24 20:48:27 <nacelle> they're all that way.
2020-12-24 20:49:24 <toogley> hm, right
2020-12-24 20:49:54 <nacelle> then it becomes a difference of how uefi establishes a hold on devices at boot vs bios
2020-12-24 20:52:32 <toogley> nacelle: what do you mean by "a hold" ?
2020-12-24 21:02:24 <nephele> Hey, I am looking to install alpine linux on an orange pi zero, according to the linux-sunxi wiki the device tree should be sun50i-h5-orangepi-zero-plus.dtb , but this isnt listed as known in the wiki/Alpine_on_ARM wiki page, is that page perhaps out of date? and if not how could i build an alpine linux image myself for installation?
2020-12-24 21:08:06 <Hello71> as a practical matter, csm has more bugs, both because it's layered on top of efi anyways, and because crappy vendors don't test it
2020-12-24 21:12:30 <mps> nephele: I'm preparing orangepi pc these days
2020-12-24 21:12:56 <mps> waiting for power connector
2020-12-24 21:13:18 <mps> I prepared u-boot and sd mmc already
2020-12-24 21:19:01 <nephele> is u-boot all that is needed?
2020-12-24 21:19:11 <nephele> As in a costum u-boot build for the target
2020-12-24 21:20:06 <mps> yes, all these boards require specific u-boot
2020-12-24 21:20:34 <mps> is sun50i-h5 64bit?
2020-12-24 21:21:30 <nephele> okay, is alpine linux on rpi3 a suficient host to compile this?
2020-12-24 21:21:35 <nephele> yes, i think it is armv8
2020-12-24 21:22:18 <mps> then making u-boot is more complicated
2020-12-24 21:23:01 <mps> but you probably already read this on sunxi site
2020-12-24 21:25:04 <nephele> it sais u-boot has support since 2018, more complicated than what do you mean?
2020-12-24 21:26:44 <mps> it need Arm Trusted Firmware to be added
2020-12-24 21:27:34 <mps> sunxi site have page about this
2020-12-24 21:29:40 <nephele> Thanks
2020-12-24 21:33:45 <nephele> if i have u-boot compiled, is it sufficient to modify the alpine linux aarch64 image by installing it there?
2020-12-24 21:38:51 <mps> nephele: depends if you want 'run from ram' or sys install
2020-12-24 21:39:32 <nephele> I would prefer the sys install, the board is supposed to run java and has 512mb ram, so memory is a bit sparce
2020-12-24 21:41:11 <mps> hmm, I have unfinished draft notes about that, not sure is good to 'publish' it yet
2020-12-24 21:41:53 <mps> I can paste draft but it should be used as hints and not proper guide
2020-12-24 21:46:30 <nephele> is the normal image the run from ram type then?
2020-12-24 21:47:42 <mps> you mean tarball?
2020-12-24 21:48:23 <nephele> they are both a tarball, i mean the one that has u-boot included as "arm generic" on the download page
2020-12-24 21:49:33 <mps> yes, generic is intended to run from ram
2020-12-24 21:50:03 <mps> though it could be 'tweaked' to run in sys mode
2020-12-24 21:50:30 <nephele> Alright, i'll use the ram mode for now then, and revisist that if it becomes a problem
2020-12-24 21:51:16 <mps> I prefer to use apk to populate all needed pkgs directly to mmc and set needed 'things' to boot
2020-12-24 21:53:05 <nephele> that is basically a sys install then?
2020-12-24 21:53:28 <mps> maybe I'll finish notes this weekend if I found time to make it for orangepi pc, which should be 'generic' method to setup alpine on arm boards
2020-12-24 21:53:43 <mps> yes, sys mode
2020-12-24 21:54:50 <mps> here is draft https://tpaste.us/x1Qv so you can 'get' basic idea
2020-12-24 21:57:10 <nephele> for what is the fat32 partition there?
2020-12-24 21:58:39 <mps> rpi zero w
2020-12-24 21:58:56 <mps> on sunxis I use ext4
2020-12-24 22:00:42 <nephele> hm, i suppose i am not quite wise how it starts to boot then, does it use the mbr boot code, or is u-boot in a partiton that the firmware reads?
2020-12-24 22:05:31 <mps> I didn't yet added installing u-boot part
2020-12-24 22:06:16 <nephele> okay, i'll check what the arm generic image does now :)
2020-12-24 22:07:29 <mps> here is my shell script to install alpine on allwinner bananapi https://tpaste.us/yNK0
2020-12-24 22:07:53 <mps> but this one is in 'run from ram' mode
2020-12-24 22:11:29 <mps> i'm waiting delivery of few mmc cards, ssd disk and some other things, hope will not wait till next week
2020-12-24 22:12:00 <mps> then I will test complete installation and make notes
2020-12-24 22:15:53 <nephele> alright
2020-12-24 22:16:15 <nephele> i'll try to compile u-boot on my rpi and report back if i got the orange pi zero to boot
2020-12-25 03:29:03 <chovy> https://stackoverflow.com/questions/65444794/how-do-i-install-firefox-esr-on-node14-alpine-docker-container?noredirect=1#comment115703922_65444794
2020-12-25 03:29:07 <chovy> any ideas?
2020-12-25 10:01:43 <EF```> Happy Christmas one and all.
2020-12-25 10:08:26 <mps> EF```: also to you :)
2020-12-25 11:30:19 <afontain_> nico: "which packages owns useradd" `apk search cmd:useradd`, or, if installed `apk info -W $(which useradd)`
2020-12-25 11:30:58 <telmich> afontain: gracias! Now I remember seeing it before, but it's a bit hard to spot in the docs
2020-12-25 11:57:01 <ikke> chovy: alpine 3.11 does not have firefox-esr
2020-12-25 11:58:37 <omni> am I the only on running syslinux and zfs root on edge?
2020-12-25 11:59:16 <omni> I'm trying to figure out if I'm doing something wrong or if, perhaps, this should also be tagged v3.13 https://gitlab.alpinelinux.org/alpine/aports/-/issues/12176
2020-12-25 12:06:28 <WeblordPepe> happy christmas fellow humans and tcp/ip enabled artificial intelligence virtualized hive minds
2020-12-25 12:06:39 <WeblordPepe> from new zealand
2020-12-25 12:07:18 <ikke> #udpenabledartificualintellginecevirtualizedhivemindslivesmatterstoo
2020-12-25 12:12:07 <mps> poor icmp is forgotten :)
2020-12-25 12:12:41 <ikke> I try to send icmp packets, but they always come back :P
2020-12-25 12:13:06 <mps> don't send echo ones
2020-12-25 12:13:50 <mps> and also echo ones don't come back a lot of times :)
2020-12-25 12:14:21 <ikke> yeah, but that ruines the joke :)
2020-12-25 12:14:32 <ikke> ruins*
2020-12-25 12:15:10 <mps> or we don't yet see jokes there :)
2020-12-25 12:20:52 <WeblordPepe> icmp is always neglected
2020-12-25 12:20:53 <mps> my new orangepi pc 'passed' in smoke test
2020-12-25 12:21:58 <WeblordPepe> now it can smoke the whole carton
2020-12-25 12:22:19 <mps> nephele: so, couldn't make install guide as I promised
2020-12-25 12:42:37 <nephele> no wories :)
2020-12-25 12:45:11 <mps> huhm..
2020-12-25 12:46:01 <mps> I had hope that this thing will be usable for some things
2020-12-25 12:46:59 <mps> from now I should buy only military grade devices
2020-12-25 17:07:24 <sr20det> hi. is there any graphical interface for apk?
2020-12-25 17:07:42 <ikke> nothing official
2020-12-25 17:08:03 <ikke> I think the gnome app manager has support for it though
2020-12-25 17:09:15 <sr20det> ikke: thank you
2020-12-25 17:15:39 <Cogitri> Yes, GNOME Software supports managing packages
2020-12-25 17:15:49 <Cogitri> I think KDE Discover also does so
2020-12-25 17:59:38 <hechos> okey guys using aports build qutebrowser mpv youtube-dl
2020-12-25 18:00:16 <hechos> and this appear more stable for a day os using.
2020-12-25 18:01:44 <hechos> how works process of update packages from repos or when is due, after release of new branch?
2020-12-25 18:05:07 <hechos> always as to install with --allow-untrusted?
2020-12-25 18:30:55 <mps> I have proof-of-concept perl tui program which browses APKINDEX files and shows details about apks which are selected by scrolling over apks list in one tui window
2020-12-25 19:02:39 <omni> neat
2020-12-25 21:51:36 <eydaimon>  
2020-12-25 21:51:46 <eydaimon> oops
2020-12-25 22:09:37 <chovy> ikke: what is the latest version of alpine?
2020-12-25 22:09:46 <ikke> 3.12
2020-12-25 22:27:29 <cim209> chovy: https://hastebin.com/vifuduguva.makefile
2020-12-25 22:28:09 <chovy> does the node:14-alpine docker not support 3.12?
2020-12-25 22:28:15 <chovy> and does 3.12 have firefox?
2020-12-25 22:28:24 <ikke> 3.12 has it
2020-12-25 22:28:29 <chovy> ok
2020-12-25 22:28:34 <ikke> the first question, I don't know
2020-12-25 22:28:45 <ikke> what do you want to achieve with installing firefox in docker?
2020-12-25 22:31:34 <chovy> i want to use firefox with puppeteer
2020-12-25 22:33:08 <ikke> headless?
2020-12-25 22:35:05 <chovy> yeah
2020-12-25 23:27:03 <chovy> i get this error when trying node:lts-alpine3.12
2020-12-25 23:27:04 <chovy> http://sprunge.us/iP1uN0
2020-12-25 23:27:08 <chovy> something about python
2020-12-25 23:27:20 <chovy> RUN apk add vim python python-dev py-pip build-base curl chromium htop ack
2020-12-25 23:30:39 <TBK[m]> chovy: apk add vim python3 python3-dev py-pip build-base curl chromium htop ack
2020-12-25 23:32:16 <chovy> py3-pip or just py-pip
2020-12-25 23:35:08 <TBK[m]> use py3-pip, py-pip "redirects" to py3-pip.
2020-12-25 23:35:51 <chovy> ok
2020-12-25 23:36:51 <TBK[m]> https://git.alpinelinux.org/aports/tree/community/py3-pip/APKBUILD#n47
2020-12-25 23:38:38 <chovy> thanks
2020-12-26 04:08:36 <fys> crap i gotta break my internet for a few minutes to convert my home network from unbound to bind
2020-12-26 04:41:33 <fys> that wasn't too bad
2020-12-26 04:41:48 <fys> (:
2020-12-26 09:36:52 <guix-vits> hello.  why in alpine-ext 3.12.3 x86_64 there is no /dev/disk/* ?  mine /dev/sda(1 .. 7)s are looks same w/o labels.
2020-12-26 09:37:27 <maxice8> Guess those are made by udev which Alpine does not use by default
2020-12-26 09:38:43 <mps> yes
2020-12-26 09:43:43 <guix-vits> ok, thank you.  (tries blkid thingy then)
2020-12-26 09:46:08 <guix-vits> (yay it works)
2020-12-26 11:33:25 <gjabell> does anyone know what the story is behind Erlang packages being broken on Erlang 23?
2020-12-26 11:39:43 <Cogitri> gjabell: You mean on Alpine? I think we merged all the subpackages back into the mainpackage since installing a subpackage pulled in everything else anyway
2020-12-26 11:39:54 <gjabell> yeah
2020-12-26 11:40:07 <gjabell> hmm alright
2020-12-26 11:41:46 <gjabell> I guess ejabberd being broken on 23 is not directly related then?
2020-12-26 14:39:24 <Hello71> guix-vits: for scripting, invoke findfs
2020-12-26 14:42:10 <guix-vits> thank you.
2020-12-26 15:14:15 <gray-_-wolf> Hello Alpine people! IIRC, some time back someone mentioned some lua script, that is able to tell which packages in repository need to be rebuilt. Is that available somewhere? I would like to avoid writing mine own.
2020-12-26 15:14:46 <ikke> gray-_-wolf: lua-aports
2020-12-26 15:15:59 <gray-_-wolf> Thanks :)
2020-12-26 15:21:34 <gray-_-wolf> Hm it cannot be legally used :/ Luckily I do not care, but it imho should get *some* license
2020-12-26 15:22:23 <ikke> well, the license is about distribution, not usage
2020-12-26 15:22:45 <Cogitri> The license situation for some Alpine stuff is...less than optimal in general though
2020-12-26 15:22:51 <ikke> agreed
2020-12-26 15:22:53 <Cogitri> E.g. aports is all rights reserved I suppose?
2020-12-26 15:23:01 <Cogitri> And we just happen to distribute that ourselves
2020-12-26 15:23:03 <gray-_-wolf> I do not think that is true. IANAL but AFAIK I need a license to use the program.
2020-12-26 15:24:08 <ikke> https://pkgs.alpinelinux.org/package/edge/main/x86_64/lua-aports mentions MIT
2020-12-26 15:24:09 <Cogitri> And changing the license on aports now would be hell I suppose because we'd need CLA from all past contributors to change the thing (or rewrite some aports, if they're actually significant enough to be subject to intellectual property)
2020-12-26 15:24:10 <mps> we should 'fight' against licenses and for them
2020-12-26 15:24:29 <Cogitri> mps: Well, you can 'fight' against licenses by having a liberal license
2020-12-26 15:24:36 <Cogitri> But having no license at all is the worst thing you can do
2020-12-26 15:24:54 <mps> I mean 'no any license'
2020-12-26 15:25:07 <Cogitri> At that point it's better to use the WTFPL instead :)
2020-12-26 15:25:24 <mps> what does that mean?
2020-12-26 15:25:39 <gray-_-wolf> https://en.wikipedia.org/wiki/WTFPL
2020-12-26 15:25:41 <ikke> kinda like CC0
2020-12-26 15:25:46 <ikke> but for software
2020-12-26 15:25:50 <gray-_-wolf> O WHAT THE FUCK YOU WANT TO PUBLIC LICENSE
2020-12-26 15:25:57 <Cogitri> Yeah, the wording is a bit strong :)
2020-12-26 15:26:10 <Cogitri> But something like WTFPL, CC0 or Public-Domain is pretty close to not having a license
2020-12-26 15:26:11 <tehcloud> Unlicense is the best of the public domain licenses
2020-12-26 15:26:11 <mps> but it is still license?
2020-12-26 15:26:17 <tehcloud> CC0 is very wordy for the same shit
2020-12-26 15:26:27 <Cogitri> tehcloud: But lawyers love words :D
2020-12-26 15:26:35 <Cogitri> mps: Well, it's a license that tells you to do whatever you want
2020-12-26 15:26:43 <ikke> https://spdx.org/licenses/WTFPL.html
2020-12-26 15:26:47 <Cogitri> You just need some kind of agreement that says "Yes, you can really do what you want"
2020-12-26 15:26:53 <Cogitri> Because nothing is worse than uncertainty
2020-12-26 15:27:03 <tehcloud> WTFPL doesn't limit liability
2020-12-26 15:27:06 <tehcloud> that's a problem
2020-12-26 15:27:19 <mps> I dream a world without license, or any license enforcement at least
2020-12-26 15:27:44 <tehcloud> I personally use MPL-2.0 for my own projects
2020-12-26 15:27:45 <mps> ah, world without lawyers and judges
2020-12-26 15:27:48 <tehcloud> nice balance
2020-12-26 15:27:57 <Cogitri> mps: That sounds slightly scary :)
2020-12-26 15:27:59 <gray-_-wolf> I just stuck gpl-2.0 on my personal stuff, but I get it is not perfect for everyone
2020-12-26 15:28:00 <Hello71> mps: hell, we can't even stop copyright from indefinitely expanding
2020-12-26 15:28:11 <Cogitri> I just slap GPL-3.0-or-later on everything I do or AGPL if it's for the web
2020-12-26 15:28:34 <gray-_-wolf> I never understood the -or-later. If you want the code back (why I use gpl), it does not quarantee that
2020-12-26 15:28:44 <mps> Hello71: yes, we live in worst of worlds known to today
2020-12-26 15:28:57 <tehcloud> GPL reads like a diatribe to me
2020-12-26 15:29:02 <tehcloud> like someone's angry drunken rant
2020-12-26 15:29:33 <tehcloud> but it does do the job it is designed to do
2020-12-26 15:29:46 <Cogitri> gray-_-wolf: Well, if a GPL-4.0 is released that fixes some problem that's in GPL-3.0 projects that have GPL-3.0-or-later can be used under the terms of GPL-4.0 too
2020-12-26 15:30:11 <Cogitri> Yup, it's kind of unfortunate how companies tend to avoid GPL (not LGPL) projects though
2020-12-26 15:30:13 <gray-_-wolf> Well yeah but everyone is free to make changes under gpl-3.0-only and you would not be able to merge them beck
2020-12-26 15:30:38 <tehcloud> LGPL's wording seems really specific to libraries, otherwise I would probably have used it
2020-12-26 15:30:58 <tehcloud> yes, I actually read all of these in full :p
2020-12-26 15:31:47 <mps> there is no solution for licensing, it is intentionally created to be 'such'
2020-12-26 15:33:00 <tehcloud> this is the software licensing conversation
2020-12-26 15:33:05 <tehcloud> the font licensing conversation is far worse
2020-12-26 15:37:58 <mps> copyrights, licenses and patents are invented in middle of 19 century to exploit people
2020-12-26 15:39:11 <mps> and I don't want to assign any of these to my public works
2020-12-26 15:39:36 <ikke> in many jurisdictions, copyright is the default
2020-12-26 15:39:48 <mps> yes, I know :)
2020-12-26 15:39:59 <mps> s/:)/:(/
2020-12-26 15:40:44 <mps> soon people will need license to be born
2020-12-26 15:41:42 <gray-_-wolf> one way to solve overpopulation...
2020-12-26 15:41:44 <tehcloud> you got a loicense for that mate?
2020-12-26 15:42:17 <mps> gray-_-wolf: war is better for population issue :)
2020-12-26 15:42:50 <maxice8> tehcloud: the British can't catch a break.
2020-12-26 15:43:46 <mps> imagine one war now as described in Mahabharata, all problem solved :)
2020-12-26 16:13:43 <guix-vits> mps: but survivors will start over with digging-sticks level tools.  and problem will reappear later?
2020-12-26 16:15:21 <mps> guix-vits: yes, but that will last for long and people will have many millennia of good life
2020-12-26 16:15:44 <guix-vits> good life?  like midewal?
2020-12-26 16:17:23 <mps> no, I mean millennia before our civilisation started
2020-12-26 16:17:56 <tehcloud> the Brahma weapon is basically a nuke
2020-12-26 16:18:21 <mps> tehcloud: yes :)
2020-12-26 16:19:28 <mps> interesting fact is that the Kerala still have high radiation
2020-12-26 18:00:22 <ulf_> Hello everyone. I've installed Alpine and XFCE on my laptop. I'm trying to boot into lightdm, however it just gives me a blinking cursor.
2020-12-26 18:00:31 <ulf_> I've followed the instructions under Install XFCE on the wiki. I've got the xf86-video-amdgpu driver installed for my graphics.
2020-12-26 18:00:41 <ulf_> Any ideas? I reckon it's related to the graphics chip driver, but I'm not sure what to change. Where are the relevant logs stored?
2020-12-26 18:00:48 <ulf_> Any help is much appreciated.
2020-12-26 18:06:01 <tehcloud> can you use startx or startxfce4 if you log in via terminal?
2020-12-26 18:06:28 <mps> ulf_: this is how tested in qemu http://arvanta.net/alpine/install-x/
2020-12-26 18:06:29 <tehcloud> diagnose any problems starting X first, if you can start X then you know it is something related to lightdm
2020-12-26 18:35:02 <ulf_> mps: I'll do a fresh install and follow those commands instead of the wiki. I missed out some of that stuff. Thanks.
2020-12-26 18:36:03 <mps> ulf_: does it works now?
2020-12-26 18:44:59 <ulf_> mps: gonna try in a bit. I'll report back.
2020-12-26 18:48:33 <ulf_> Any idea about installing GNOME instead of XFCE. That's what I was aiming for but I faced the same issue there and figured it didn't work out of the box or something.
2020-12-26 18:52:21 <mps> hmm, using small and simple distro and then install gnome, huh
2020-12-26 18:52:45 <ulf_> Yeah I know ;)
2020-12-26 18:53:02 <mps> you lost 'alpine' then :)
2020-12-26 18:53:04 <ulf_> Goes against the ethos a bit. But I'm used to GNOME, so it'd be great if I could get it installed.
2020-12-26 18:53:22 <ulf_> You'll be glad to know I've never used Ubuntu.
2020-12-26 18:53:34 <mps> ulf_: yes, some people here can help you, just wait some time
2020-12-26 18:53:54 <mps> also I never used ubuntu
2020-12-26 18:54:58 <ulf_> Yeah, no worries, it is boxing day after all.
2020-12-26 18:55:17 <ulf_> I've used practically every other distro.
2020-12-26 19:00:10 <maxice8> GNOME is very good though
2020-12-26 19:00:46 <mps> windows also :P
2020-12-26 19:01:41 <mps> but yes, gnome is good tbh
2020-12-26 19:02:15 <exception> only with 999 extensions
2020-12-26 19:02:24 <mps> and xfce is better
2020-12-26 19:02:37 <exception> xfce > kde > gnome
2020-12-26 19:02:51 <mps> though awesome wm is best for me
2020-12-26 19:04:13 <maxice8> I have 2 extensions enabled, works pretty well
2020-12-26 19:26:52 <maringuu> Hello. Has anyone problems with firefox recentyl? It always gives me "Gah. Your tab just crashed" and the logs indicate something about pango and out of memory.
2020-12-26 19:27:22 <maringuu> I'm ofc not out of memory. Neither in disk space or ram..
2020-12-26 19:27:28 <Cogitri> I think we had someone with similar problems, but the log also mentioned something about seccomp as well
2020-12-26 19:27:33 <tehcloud> maringuu, that has been a longstanding problem
2020-12-26 19:27:48 <tehcloud> I've had unstable Firefox on Alpine the entire time I've used it, basically :p
2020-12-26 19:28:02 <tehcloud> although on the most recent install it has not been bad
2020-12-26 19:28:10 <Cogitri> Just works for me 🤔
2020-12-26 19:28:13 <maringuu> Cogitri: Yeah it mentions that too
2020-12-26 19:28:24 <tehcloud> might have something to do with plugins/addons you're using too
2020-12-26 19:28:51 <maringuu> It just worked for me just until today :(
2020-12-26 19:29:06 <Cogitri> maringuu: Then that's your problem, seccomp violations instantly kill the thing
2020-12-26 19:29:41 <maringuu> Hmm interesting
2020-12-26 19:29:56 <Cogitri> We probably have to punch another hole in the seccomp sandbox for that, but it's hard to do that & test it without a reproducer
2020-12-26 19:32:32 <mps> I run FF with ~/.mozilla mounted on tmpfs of 128MB size. when it reaches about 70% of usage I got these messages, i.e. low disk space
2020-12-26 19:37:24 <maringuu> Thanks for the information. Should I report this but at firefox or at alpine?
2020-12-26 19:38:20 <maringuu> mps: And the messages also were about pango?
2020-12-26 19:38:35 <mps> no, just disk low
2020-12-26 19:49:53 <maringuu> A woraround I just found for seccomp errors is to set "MOZ_DISABLE_CONTENT_SANDBOX=1" which like the name says disables seccomp sandboxing.
2020-12-26 21:52:28 <ulf_> mps: here's the error I received when running startx:
2020-12-26 21:52:38 <ulf_> Server terminates with error (1).
2020-12-26 21:52:52 <ulf_> Xinit: unable to connect to X server: connection refused
2020-12-26 21:52:57 <ulf_> Xinit: server error
2020-12-26 21:53:03 <ulf_> Gonna crack open the logs now.
2020-12-26 21:54:00 <ulf_> I followed those instructions you linked.
2020-12-26 22:04:02 <mps> ulf_: guide is tested and intended for qemu, I just posted it as example
2020-12-26 22:04:35 <mps> some people told that it helped them
2020-12-26 22:10:18 <ulf_> Ah, OK. Thanks anyway.
2020-12-26 22:11:18 <mps> np
2020-12-26 22:11:32 <ulf_> I installed those drivers that were missing (ati etc.) however the error still happens.  https://usercontent.irccloud-cdn.com/file/P0oZJAym/IMG_20201226_215624.jpg
2020-12-26 22:13:35 <mps> for such reports use text paste service, install 'apk add tpaste', then run 'tpaste < /var/log/Xorg.0.log' and post resulting url here
2020-12-26 22:16:13 <ulf_> Thanks - the URL is https://tpaste.us/yN8p
2020-12-26 22:17:17 <ulf_> Seems like installing those drivers made progress in fixing it, but it still doesn't work for some reason.
2020-12-26 22:19:07 <mps> what says 'cat /proc/cmdline'
2020-12-26 22:28:37 <ulf_>  BOOT_IMAGE=/boot/vmlinuz-lts root=/dev/mapper/vg0-lv_root ro modules=sd-mod,usb-storage, ext4 nomodset quiet rootfstype=ext4
2020-12-26 22:29:15 <ikke> remove nomodeset
2020-12-26 22:29:27 <mps> ikke: hmm :)
2020-12-26 22:29:33 <ikke> :P
2020-12-26 22:30:24 <ulf_> Sure, I'll give that a try.
2020-12-26 22:42:37 <ulf_> Nano gives me an error saying: Error writing cmdline I/O error.
2020-12-26 22:42:46 <ulf_> I'm running it with sudo.
2020-12-26 22:42:58 <ikke> sorry, you cannot directly remove it from there
2020-12-26 22:43:17 <ikke> if you use syslinux, it's /etc/syslinux.conf
2020-12-26 22:44:01 <ulf_> I don't think I do, it's UEFI/GPT. /etc/syslinux.conf is empty as well.
2020-12-26 22:44:21 <ikke> What bootloader do you use?
2020-12-26 22:44:27 <ikke> grub?
2020-12-26 22:44:33 <ulf_> Yeah.
2020-12-26 22:44:53 <ikke> Then you need to update the grub config
2020-12-26 22:45:08 <ulf_> Ah OK. /etc/default/grub?
2020-12-26 22:45:17 <ulf_> Apologies for my lack of knowledge.
2020-12-26 22:46:50 <ikke> I think so
2020-12-26 22:48:21 <ulf_> Seems like the same error unfortunately. One second and I'll post new logs. Thanks for your help so far.
2020-12-26 22:48:58 <ulf_> https://tpaste.us/Yoar
2020-12-26 22:49:31 <ikke> well, after that, you need to update the bootloader and reboot
2020-12-26 22:50:34 <ulf_> I did - I ran sudo grub-mkconfig -o /boot/grub/grub.cfg
2020-12-26 22:50:38 <ulf_> Then rebooted.
2020-12-26 22:51:03 <ulf_> After removing nomodeset from /etc/default/grub
2020-12-26 22:52:31 <ikke> hmm, signal 6 is SIGABRT
2020-12-26 22:53:50 <ikke> "mps     ikke: without nomodeset in kernel cmdline and qemu '-vga virtio' parameter VM got /dev/dri/card0 (and rest) but X dies with "
2020-12-26 22:53:53 <ikke> heh
2020-12-26 22:54:00 <ikke> dies with SIGABRT
2020-12-26 22:55:24 <ulf_> I'm extremely lost haha.
2020-12-26 22:55:38 <ulf_> Maybe I should just give up and install something else?
2020-12-26 22:56:02 <ulf_> I just ran neofetch and it has 331 packages installed. 331! Now that is what I call no bloat.
2020-12-26 22:56:05 <ikke> mps: how did you solve that?
2020-12-26 22:56:58 <maxice8> package count is not a good indicator of bloat
2020-12-26 22:57:23 <ulf_> Nah, I know. Just mental how low it is.
2020-12-26 22:57:47 <ulf_> Disk space used or RAM usage is a better indication.
2020-12-26 23:01:48 <mps> ikke: what?
2020-12-26 23:02:22 <ikke> mps: Apparently you ran into this before
2020-12-26 23:03:16 <mps> hmm, disk is offline (in drawer) so can't look now
2020-12-26 23:03:58 <mps> wait, maybe on my daughter computer, if she didn't shutowned it
2020-12-26 23:05:09 <mps> '-vga virtio -display gtk,gl=on'
2020-12-26 23:05:44 <mps> that is to run win10 in qemu
2020-12-26 23:06:26 <mps> yes, I know you all will laugh, but she had to use win when she was at university
2020-12-26 23:07:08 <mps> and we had to buy one win10 license :(
2020-12-26 23:07:59 <ulf_> oof
2020-12-26 23:49:05 <ulf_> Decided to move to Void Linux. It was still worthwhile installing Alpine, learned a load of new stuff. Thank you all for you help.
2020-12-26 23:55:38 <iv-so> are there any third party repositories?
2020-12-26 23:57:34 <maxice8> We don't track any
2020-12-27 00:35:57 <Hello71> mps: windows virtio drivers doesn't support virtio gpu until very recently
2020-12-27 00:36:40 <Hello71> also you can use windows free as long as you are ok with "ACTIVATE WINDOWS" constantly popping up
2020-12-27 00:36:51 <Hello71> but imo it's not much worse than constant forced reboots
2020-12-27 00:37:11 <Hello71> ikke: it's /etc/update-extlinux.conf
2020-12-27 00:38:26 <Hello71> [    37.149] (--) AMDGPU(0): Chipset: "Unknown AMD Radeon GPU" (ChipID = 0x15d8)
2020-12-27 00:38:30 <Hello71> probably you need edge
2020-12-27 00:39:29 <Hello71> probably also apk del xf86-*
2020-12-27 05:39:42 <cads2> hey all, I'm in WSL alpine and I do  ln -s /mnt/c/Users/user/src src to link my source directory to my wsl home. When I cd into the directory, the shell changes pwd to /mnt/c/Users/user/src, which is unexpected
2020-12-27 05:41:00 <cads2> then I can move a folder from one sub-directory into another in my source directory, and something utterly terrifying happens - the folder appear to go missing after rename!
2020-12-27 05:44:21 <cads2> I can even cd into the moved folder. here, ls tells us that it can't resolve /user/home/src/moved_dir, but the PWD says we're in /mnt/c/Users/user/src/moved_dir
2020-12-27 05:46:19 <cads2> I get the feeling that this could a WSL+Alpine issue that wouldn't ordinarily happen in either mainline WSL or pure alpine. But I've gotten so used to working in alpine that I don't want to go back to wsl ubuntu
2020-12-27 05:47:14 <luem[34]> I would like to know if Alpine is 100% free software? Cheers
2020-12-27 05:50:12 <hg> luem[34]: you can see the licenses of the packages in the repos
2020-12-27 05:50:32 <hg> afact, the vast majority of it is
2020-12-27 05:50:41 <cads2> my questions:   1)  What is going on with the pwd? does the pwd in busybox change like that whenever we follow a softlink?   2) Can I install another version of the ln or mv command that will do the right thing when encountering these links?
2020-12-27 05:51:29 <hg> cads2: what do you mean?
2020-12-27 05:52:55 <cads2> hg, in /home/user I have a link  src -> /mnt/c/Users/user/src.   When I do cd /home/user/src   I expect the new pwd to be /home/user/src,  but when I do it in wsl alpine, the pwd reads /mnt/c/Users/user/src
2020-12-27 05:53:10 <hg> luem[34]: https://gitlab.alpinelinux.org/alpine/aports/-/tree/master/non-free those are all the packages that are known non-free
2020-12-27 05:53:18 <cads2> hg - is that behavior you'd expect from sh/busybox on its own?
2020-12-27 05:54:27 <hg> so, it's the equivalent of `pwd -P`
2020-12-27 05:54:55 <hg> yeah, that's the default for busybox
2020-12-27 05:55:16 <hg> and no, busybox's pwd doesn't support `-L`
2020-12-27 05:55:46 <hg> cads2: though, `echo $PWD` may work for you
2020-12-27 05:56:44 <cads2> well, if it just dereferences the path whenever it takes a softline across a fs boundary, I'm okay with that
2020-12-27 05:57:23 <cads2> I was worried I could be seeing behavior caused by WSL's interaction with busybox, like my mysteriously disappearing folders
2020-12-27 05:57:28 <hg> as far as I know, busybox's pwd (pwd -P) will always dereference symlinks
2020-12-27 05:57:47 <hg> cads2: are you using WSL2 or WSL1?
2020-12-27 05:58:14 <cads2> the former, but I upgraded from 1 to 2, so that could add an extra layer of awful
2020-12-27 05:58:20 <hg> mm
2020-12-27 05:58:22 <hg> 🤷
2020-12-27 05:58:29 <cads2> also I just tried to replicate the issue, and I couldn't
2020-12-27 05:58:49 <hg> what do you mean?
2020-12-27 06:03:28 <cads2> the problem I'm having is that when I'm in my src directory and I move a directory while also renaming it, that directory will "go missing":   1) ls will not show the directory but will show a warning "ls: ./directory: No such file or directory"   2) it will be possible to cd into the directory, but this gives a warning "Error: could not resolve path: /home/m/src/director'"   3) viewing the directory from windows explorer shows the files
2020-12-27 06:03:28 <cads2> unharmed   4) renaming the directory (to directory_rename) in explorer allows you to see them in wsl again    5) But you can still cd to the old "directory" in wsl, as if it still has a faulty inode
2020-12-27 06:04:43 <hg> yeah, that sounds like a WSL nightmare to me ):
2020-12-27 06:05:55 <cads2> real shame, because I am really digging the alpine WSL experience otherwise -- who would you talk to who would hear about this problem and be excited?
2020-12-27 06:05:58 <luem[34]> thanks hg appreciated
2020-12-27 06:06:38 <cads2> thinking I'll track down the WSL alpine repo and file a report at least
2020-12-27 07:24:03 <cads> hg, the plot thickens - windows file access is mediated through 9P, a Plan 9 inspired file server responsible for both directions of WSL2's bidirectional file integrations wherein windows sees WSL as a network drive, and wsl sees windows fs as a mounted FS
2020-12-27 07:25:56 <cads> so what happens when alpine moves around files in the windows FS? Well, 9P move the file and then either fails to update some cache, or updates the cache with contents that alpine is not expecting
2020-12-27 08:59:44 <mps> Hello71: 'recently' is vague term
2020-12-27 09:01:12 <mps> I'm sure I used '-vga virtio' about two years ago for running win10 in qemu, with some drivers from redhat which I installed in win
2020-12-27 13:40:19 <Hello71> mps: about two weeks ago
2020-12-27 13:40:44 <Hello71> the official iso does not even contain it yet
2020-12-27 13:41:10 <Hello71> you can use it but it is just stdvga compatible
2020-12-27 13:42:11 <Hello71> if you go to device manager it will say "Microsoft Basic Display Adapter"
2020-12-27 14:17:43 <guix-vits> hello.  why no graphics: https://git.alpinelinux.org/aports/tree/main/links/APKBUILD ?
2020-12-27 14:54:08 <ikke> perhaps to limit dependencies
2020-12-27 15:00:01 <guix-vits> ok, but.. images?
2020-12-27 18:10:35 <mps> Hello71: hmm, does this means that ms added these drivers, or I misunderstood you.
2020-12-27 18:10:56 <Hello71> well microsoft barely provides drivers at all
2020-12-27 18:11:17 <Hello71> only since windows 10 and to a lesser extent windows 7 via windows update
2020-12-27 18:11:31 <mps> I remember that I downloaded some iso images from redhat and installed these driver on win under qemu
2020-12-27 18:11:32 <Hello71> https://github.com/virtio-win/kvm-guest-drivers-windows/commit/51ea0580480289eaae74c98409512e63f41b15fe
2020-12-27 18:12:03 <Hello71> it is open source but basically funded by red hat
2020-12-27 18:12:32 <mps> that was about two years ago and I forgot details. happily didn't needed them anymore after that
2020-12-27 18:13:41 <mps> but with them performance of disk and video went well up on win in qemu, I remember
2020-12-27 18:20:47 <Hello71> with windows guest, virtio vga is exactly same as std vga
2020-12-27 18:22:16 <Hello71> but before qemu 2.2 the default was cirrus which has poor performance and limited resolution
2020-12-27 18:23:57 <mps> I don't know much about win in qemu. only know that these RH driver helped to be 'acceptable' (bearable). btw, I used it with sdl
2020-12-27 18:23:59 <Hello71> disk also will not magically speed up, you need to set if=virtio/-device virtio-blk
2020-12-27 18:24:10 <Hello71> host display output is totally irrelevant
2020-12-27 18:24:29 <Hello71> except I think you need qxl for spice
2020-12-27 18:25:13 <mps> without going to details (which I don't know much about) I repeat, drivers helped significantly
2020-12-27 18:25:43 <mps> and I don't care anymore
2020-12-27 18:25:46 <Hello71> if you don't change vm config, any improvement is placebo
2020-12-27 18:27:05 <mps> ah, -drive file=win10.img,if=virtio,index=0
2020-12-27 18:28:12 <mps> '-display sdl -vga qxl'
2020-12-27 18:29:29 <Hello71> yes, but you have to set if=virtio after installing drivers, otherwise windows will not install/boot. you can add hdd drivers from ide cdrom though
2020-12-27 18:31:03 <mps> yes, that is what I did, though forgot details
2020-12-27 18:36:49 <mps> finally, kernel 5.11 will support SDXC mmc card. hope will test that tomorrow
2020-12-27 21:26:29 <monkwitdafunk> hello users of alpine-linux channel
2020-12-27 21:27:04 <monkwitdafunk> is IPfire, and alpine supposed to be the two top players in live router distributions?
2020-12-27 21:27:21 <ikke> I would not know
2020-12-27 21:32:15 <guix-vits> btw, is usb autosuspend disabled in alpinelinux?
2020-12-28 02:02:34 <guix-vits> https://git.alpinelinux.org/aports/tree/main/links/APKBUILD#n5
2020-12-28 02:02:35 <guix-vits> graa-phics.  gra-phiics. *bambam*. graphiiicks.
2020-12-28 02:03:47 <guix-vits> graphichal links matter.
2020-12-28 02:05:09 <Hello71> guix-vits: whatever the kernel default is
2020-12-28 02:05:41 <Hello71> also "running in both graphics and text mode" "--disable-graphics" -.-
2020-12-28 03:04:39 <xordspar0> I duplicated my sys install to another disk with dd. The copied disk won't boot. During boot, I see "mounting UUID=(my disk's UUID) on /sysroot failed: No such file or directory". In the emergency shell, `fdisk -l` shows no disks. However, when I boot Alpine from a USB drive, the disk shows up and has the correct UUID.
2020-12-28 03:05:00 <xordspar0> 1. Is fdisk -l a valid way to list disks in the syslinux emergency shell? If it is, why isn't my disk listed there? 2. What could have changed when coping the disk? If it worked on the old disk shouldn't it be the same on the new one?
2020-12-28 04:22:01 <Hello71> at a guess, you switched from sata to nvme and didn't put it in mkinitfs
2020-12-28 05:06:51 <xordspar0> Hello71: Hm, I did switch from sata to nvme. I also reinstalled the kernel in a chroot, however, when I run mkinitfs manually in my chroot it fails to find a /lib/modules directory that matches the running kernel. The patient system has a newer kernel than the install system.
2020-12-28 05:36:56 <xordspar0> Oh, the matching /lib/modules thing is a red herring. You can specify the kernel version you want as an argument to mkinitfs. So I have run mkinintfs as part of a kernel package reinstall and manually from a rescue OS, but my NVME disk is still invisible during boot.
2020-12-28 05:42:51 <xordspar0> Oh! You have to explicitly enable NVME in the mkinitfs config. There's a wiki page for it and everything. It doesn't turn up in Duck Duck Go search results, but it's the top result on Google. :(
2020-12-28 05:43:29 <xordspar0> Thanks for pointing me in the right direction, Hello71. My system is booting fine now.
2020-12-28 13:56:04 <Hello71> i wonder if we should turn nvme on by default now
2020-12-28 14:01:02 <Cogitri> Yup, super easy to forget that in mkinitfs' config
2020-12-28 14:01:10 <Cogitri> And it's a pretty common config by now
2020-12-28 14:02:05 <clandmeter> doesnt mkinitfs check loaded modules?
2020-12-28 14:02:21 <mps> so, put everything by default
2020-12-28 14:33:39 <guix-vits> +1.  Alpine anyway has near 900mb root by now.
2020-12-28 14:35:26 <mps> hmm, intention was irony
2020-12-28 14:44:53 <guix-vits> mps: sorry, but destroying functionality isn't building functionality.  former is way easier.
2020-12-28 14:45:08 <guix-vits> i remember gentoo on my box.
2020-12-28 14:45:21 <guix-vits> want X?  buuuiiild iiit
2020-12-28 14:45:26 <guix-vits> suuucks
2020-12-28 14:46:12 <guix-vits> * suxx
2020-12-28 14:46:25 <mps> all things ....
2020-12-28 14:48:11 <mps> alpine moto starts with word 'small' not 'everything out of the box'
2020-12-28 14:51:07 <guix-vits> i wonder if 800mb vs 1.2gb do difference?
2020-12-28 14:51:55 <guix-vits> they both aren't fit in smart watch, anyway (maybe?)
2020-12-28 14:52:04 <mps> base install is about 350MB
2020-12-28 14:52:11 <mps> or even less
2020-12-28 14:52:31 <guix-vits> ok, i used extended.
2020-12-28 14:52:53 <ikke> doesn't matter
2020-12-28 14:53:00 <ikke> extended is just a larger iso
2020-12-28 14:53:16 <mps> some use apk with chroot to install
2020-12-28 14:53:24 <guix-vits> ah
2020-12-28 15:15:50 <jvoisin> hello
2020-12-28 15:16:04 <jvoisin> is there an easy way to recompile a package with a different set of CLFAGS?
2020-12-28 15:16:25 <jvoisin> (something like `apt-get source <…>`)
2020-12-28 15:16:55 <ikke> git clone --depth=1 https://gitlab.alpinelinux.org/alpine/aports
2020-12-28 15:17:10 <mps> jvoisin: export CFLAGS="$CFLAGS ..."
2020-12-28 15:23:29 <cfengineuser> Hey I noticed that unbound is broken for me on edge
2020-12-28 15:23:36 <cfengineuser> test:~$ /usr/sbin/unbound-checkconf
2020-12-28 15:23:36 <cfengineuser> Error relocating /usr/sbin/unbound-checkconf: reallocarray: symbol not found
2020-12-28 15:23:43 <cfengineuser> any help?
2020-12-28 15:24:10 <cfengineuser> it was working
2020-12-28 15:24:20 <cfengineuser> but I can see if i restart now unboudn fails to restart
2020-12-28 15:24:43 <mps> cfengineuser: create bug report and assign it to maintainer
2020-12-28 15:24:57 <caskd> no that's because of the musl bug
2020-12-28 15:25:02 <caskd> upgrade musl
2020-12-28 15:25:07 <cfengineuser> how?
2020-12-28 15:25:22 <mps> ah, not full upgrade
2020-12-28 15:25:36 <ikke> mps: reallocarray should be provided by musl, and the program was apparently compiled with it
2020-12-28 15:25:47 <cfengineuser> yep
2020-12-28 15:25:48 <ikke> so not a lot the package maintainer can do about
2020-12-28 15:25:56 <caskd> yeah reallocarray didn't work on some release, it was fixed later
2020-12-28 15:26:04 <cfengineuser> it was working
2020-12-28 15:26:06 <mps> ikke: yes, I remember I asked for it to be added to latest musl :)
2020-12-28 15:26:15 <cfengineuser> the old version ov unbound is still running
2020-12-28 15:26:18 <jvoisin> ikke: mps thanks
2020-12-28 15:26:52 <cfengineuser> so?
2020-12-28 15:26:53 <mps> but i read only 'symbol not found' part of msg
2020-12-28 15:27:13 <ikke> mps: that means the issue is with a library
2020-12-28 15:27:15 <cfengineuser> will full update fix it?
2020-12-28 15:27:24 <mps> sure
2020-12-28 15:27:24 <ikke> at least musl
2020-12-28 15:27:35 <mps> cfengineuser: apk upgrade -a
2020-12-28 15:27:41 <cfengineuser> hmm
2020-12-28 15:40:16 <Hello71> cfengineuser: you have pinned packages or some nonsense
2020-12-28 15:40:41 <cfengineuser> well I might have updated some and not others
2020-12-28 15:42:16 <cfengineuser> gonna start with fresh upodated flash
2020-12-28 16:57:59 <jiggles> does anyone have any experience booting alpine under qemu on arm64? I can get as far as grub and the messages relating to the EFI stub but it just sits there after that
2020-12-28 17:02:32 <Hello71> probably your console parameter is set wrong
2020-12-28 17:03:11 <pernia> caskd
2020-12-28 17:03:20 <mps> jiggles: http://arvanta.net/alpine/install-aarch64-qemu/
2020-12-28 17:03:37 <caskd> oh hi pernia
2020-12-28 17:03:59 <pernia> *kisses u*
2020-12-28 17:04:16 <caskd> do this on fedi lmao, this is for on-topic stuff
2020-12-28 17:04:28 <pernia> oki~
2020-12-28 17:04:45 <pernia> anyways i wanted to test this irc client and it seems super sweet
2020-12-28 17:04:48 <hechos> hello guys
2020-12-28 17:04:52 <pernia> o/
2020-12-28 17:05:07 <hechos> pernia: name?
2020-12-28 17:05:12 <caskd> what client?
2020-12-28 17:05:17 <pernia> catgirl
2020-12-28 17:05:19 <mps>  /join #test
2020-12-28 17:05:21 <pernia> its smol
2020-12-28 17:05:24 <hechos> lol.
2020-12-28 17:05:27 <pernia> hechos what?
2020-12-28 17:05:43 <hechos> anyone using alpine over root f2fs?
2020-12-28 17:06:01 <mps> hechos: yes :)
2020-12-28 17:06:16 <pernia> well byby
2020-12-28 17:06:47 <echelon> hi, what's causing this.. https://paste.ee/r/OzmBy
2020-12-28 17:08:11 <caskd> echelon: have community repos enabled?
2020-12-28 17:08:29 <caskd> ah nvm dumb question
2020-12-28 17:08:39 <echelon> i don't know, how do i do that
2020-12-28 17:08:49 <caskd> it's already enabled, my bad
2020-12-28 17:08:55 <echelon> is it
2020-12-28 17:09:23 <echelon> what's the deal with this: Ignoring http://dl-cdn.alpinelinux.org/alpine/v3.12/main/x86_64/APKINDEX.tar.gz: Permission denied
2020-12-28 17:10:40 <caskd> it seems like it cannot write the cached APKINDEX
2020-12-28 17:10:57 <caskd> not sure why since i haven't used alpine on docker for a long time
2020-12-28 17:11:24 <caskd> is /etc/apk/cache linked to anything?
2020-12-28 17:13:03 <caskd> also may i ask why you delete the apk cache even though you don't use it?
2020-12-28 17:13:07 <echelon> there is no /etc/apk/cache
2020-12-28 17:13:15 <hechos> mps: you using grub
2020-12-28 17:13:24 <echelon> just habit
2020-12-28 17:13:50 <mps> hechos: no, syslinux and u-boot
2020-12-28 17:14:27 <jiggles> mps: thanks I will give it a try
2020-12-28 17:16:18 <hechos> mps: may be need to load modules manually
2020-12-28 17:16:57 <echelon> caskd: ok, getting rid of the rm -rf on the apk cache seems to have fix it
2020-12-28 17:17:00 <echelon> fixed*
2020-12-28 17:17:24 <caskd> that happens after it so that's interesting, but glad to know
2020-12-28 17:17:58 <caskd> apk cache doesn't hurt your docker builds by the way since it's anyways reset for each build
2020-12-28 17:18:10 <caskd> so the --no-cache flag doesn't hurt
2020-12-28 17:18:27 <caskd> help*
2020-12-28 17:18:54 <caskd> only in cases where the base image has just been built it might help, but other than that it's useless
2020-12-28 17:20:44 <mps> hechos: add 'f2fs' to /etc/mkinitfs/mkinitfs.conf and recreate initramfs image
2020-12-28 17:21:28 <echelon> good to know :)
2020-12-28 17:21:45 <mps> and set APEND parameters for f2fs
2020-12-28 17:23:30 <hechos> mps: yes, im modify my script to fit the needs to run f2fs. well see.
2020-12-28 17:23:46 <hechos> mps: how work cycle of repository updates?
2020-12-28 17:24:20 <mps> hechos: what?
2020-12-28 17:24:41 <hechos> mps: for example need to compile to latest version of these programs to work properly on my setup. qutebrowser mpv youtube-dl and get latest version of that
2020-12-28 17:27:17 <hechos> when repository is get updates. when is release alpine?
2020-12-28 17:30:07 <mps> hechos: if you are using edge repo is updated daily, even minutely
2020-12-28 17:30:33 <mps> stable repo is updated 'as needed'
2020-12-28 17:44:22 <hechos> mps: yes, edge is not to stable. as stable is.
2020-12-28 17:45:09 <tehcloud> you can live on the edge if the software you use day to day is generally stable upstream
2020-12-28 17:46:04 <hechos> tehcloud: aha.
2020-12-28 17:47:24 <hechos> mps: got error fs not mount on RO
2020-12-28 17:48:53 <mps> hechos: do you have something like that 'APPEND modules=sd-mod,usb-storage,ext4,f2fs,sunxi-mmc' in extlinux.conf
2020-12-28 17:49:36 <hechos> here is grub but need to check what you say.
2020-12-28 17:50:18 <hechos> leave on prompt (none) login:
2020-12-28 17:54:43 <jiggles> mps: I still get stuck on the messages relating to the EFI stub, I am not running a grpahical environment so I need the output to my current tty. Any suggestions?
2020-12-28 17:56:11 <mps> jiggles: link I posted describes using it from tty
2020-12-28 17:56:59 <mps> jiggles: read comments in script how to obtain QEMU_EFI.fd 'bios' file
2020-12-28 17:57:29 <jiggles> mps: I figured but still no joy, I have tried using named pipes as well as amending the kernel cmdline adding both console=ttyS0 and console=ttyAMA0
2020-12-28 17:57:56 <jiggles> mps: have retireved the QUME_EFI.fd and followed the relevant instructions for both methods
2020-12-28 18:00:35 <mps> jiggles: my disk with all this is in drawer now so can't check, but both scripts worked for me and many times (and not only for me)
2020-12-28 18:06:06 <jiggles> mps: thanks. I am running on a pinebookpro if that would make a difference? How long does it take to boot on your hardware?
2020-12-28 18:07:05 <mps> jiggles: ah, then you have to change qemu params
2020-12-28 18:07:35 <mps> I can look later on my aarch64 machine when I come back to home
2020-12-28 18:08:49 <mps> jiggles: to run qemu on rk3399 I had to use some 'tweak' parameters
2020-12-28 18:09:00 <jiggles> mps: thanks that would be great, been trying for a few days now and can get it to the same spot in a variety of ways, but no further
2020-12-28 19:08:30 <echelon> caskd: ugh, running into the same issue again: https://paste.ee/r/IdlRf
2020-12-28 19:08:51 <echelon> i'm not even deleting the cache dir now
2020-12-28 19:12:11 <mps> jiggles: your host is rk3399?
2020-12-28 19:18:34 <guix-vits> 
ACTION if pinebook pro, then yes

2020-12-28 19:19:09 <mps> guix-vits: I think so, but wanted to hear confirm
2020-12-28 19:20:20 <mps> anyhow, I have to use this on my rk3399 machine to start qemu: taskset 03 qemu-system-aarch64 ....
2020-12-28 19:24:04 <guix-vits> > echelon, http://dl-cdn.alpinelinux.org/...
2020-12-28 19:24:04 <guix-vits> why not https, btw?
2020-12-28 19:24:43 <echelon> i wasn't telling to use http
2020-12-28 19:25:11 <jiggles> mps: correct, working on both the pinebook pro and the rockpro64 at the moment
2020-12-28 19:25:40 <mps> https doesn't work on dl-cdn
2020-12-28 19:27:43 <mps> jiggles: here is old script I used on rk3399 (gru-kevin chromebook) https://tpaste.us/QWz0
2020-12-28 19:40:50 <echelon> guix-vits: it was a proxy issue
2020-12-28 19:45:11 <jiggles> mps: thanks a bunch, I will give this a try
2020-12-28 20:08:51 <sr20det> hi. is 'Add: grsec_sysfs_restrict=0 to the end of the of the APPEND line'  https://wiki.alpinelinux.org/wiki/Desktop-notes  needed on standart iso image?
2020-12-28 20:10:01 <sr20det> i.e. is standard iso kernel grsec?
2020-12-28 20:11:46 <Cogitri> We don't have grsec in the repos anymore
2020-12-28 20:11:46 <mps> sr20det: guide is outdated
2020-12-28 20:11:51 <Cogitri> We wiki is just outdated
2020-12-28 20:12:56 <sr20det> chromium chrome://gpu shows software everywhere, but there is normal gallium acceleration in Xorg.log
2020-12-28 20:13:03 <sr20det> radeon
2020-12-28 20:13:32 <sr20det> what should i do?
2020-12-28 20:15:14 <Cogitri> We don't have the vaapi patches for chromium so I think that's expected (?)
2020-12-28 20:15:34 <sr20det> it cant do vdpau?
2020-12-28 20:15:50 <sr20det> and no webgl..
2020-12-28 20:16:14 <sr20det> okay. also setup-disks fails if disk contains any partitions, so i should do dd bs=1M count=[1-4] if=/dev/zero of=/dev/sda{1,2,3}
2020-12-28 20:16:36 <sr20det> pops message that mkfs.any  requires -f
2020-12-28 21:06:11 <Hello71> did you install mesa
2020-12-28 21:06:27 <sr20det> yep
2020-12-28 21:06:52 <Hello71> wipefs is better but i think busybox does not have
2020-12-28 21:07:24 <sr20det> also busybox's fdisk cant delete partitions
2020-12-28 21:10:28 <sr20det> sorry. no logs for now
2020-12-28 22:25:10 <meowray> how do you detect glibc in CMake?
2020-12-28 22:42:14 <maxice8> Think that is a better question for cmake support channels
2020-12-29 00:46:23 <stee> What happened to the alpine-desktop package? It isn't in 3.12
2020-12-29 01:04:02 <Hello71> it was removed due to questionable value add
2020-12-29 01:04:39 <Hello71> https://git.alpinelinux.org/aports/commit/community/alpine-desktop?id=59ed4dcbaef9dec859eb014a003cb7a4c89780b0
2020-12-29 01:10:49 <stee> thanks, I was looking for that commit
2020-12-29 01:11:11 <stee> soo, I did setup-xorg-base
2020-12-29 01:11:25 <stee> just wasn't clear what I need after that
2020-12-29 01:11:40 <stee> I'm looking at this: https://docs.alpinelinux.org/user-handbook/0.1a/Working/post-install.html#_getting_a_graphical_environment
2020-12-29 01:29:07 <Guest39163> ?
2020-12-29 01:45:31 <Hello71> stee: it's outdated
2020-12-29 09:12:46 <morten> Hi! Does anyone know a way to install wireguard on a raspberry pi?
2020-12-29 09:41:06 <artok> morten: isn't it quite straightforward? build the module and tools and off you go?
2020-12-29 09:48:27 <morten> I run into the same error, handled on mailinglist post [https://lists.alpinelinux.org/~alpine/users/%3CCAEhkKgV-OdZ8y406_yynrH7tcxjgXkKzSc6dCSZ_a6CUPUfBiA%40mail.gmail.com%3E]
2020-12-29 09:49:57 <morten> Allan Riordan Boll workaround works for me but is not persistent and a little bit hacky
2020-12-29 09:52:40 <artok> using overlay on top of modloop ?
2020-12-29 09:53:51 <morten> yes with lbu
2020-12-29 09:56:27 <artok> well then it's about adding the module into modloop squasfs
2020-12-29 09:56:30 <artok> +h
2020-12-29 10:00:35 <morten> is there any documentation how to do that?
2020-12-29 10:02:26 <sam_son> hello everyone, quick question, is there a way to upgrade all my php7* packages to php8? I understand that 'apk add php8' just adds php8 in parallel, it does not replace php7 (I'm running edge).
2020-12-29 10:03:07 <ikke> sam_son: sed -i 's/^php7/php8/' /etc/apk/world && apk fix
2020-12-29 10:03:14 <ikke> (use at own risk, not tested, etc)
2020-12-29 10:03:30 <sam_son> ah I see, ok thanks I'll give it a try
2020-12-29 10:05:46 <sam_son> hmm
2020-12-29 10:05:49 <sam_son> sort of worked
2020-12-29 10:06:01 <sam_son> its missing the sym link /usr/bin/php -> php8
2020-12-29 10:06:08 <artok> morten: mkinifts might do that
2020-12-29 10:06:29 <ikke> sam_son: Right, I don't think php8 will provide that yet
2020-12-29 10:06:50 <sam_son> ikke, still, its better than I expected, thanks for the help
2020-12-29 10:14:02 <morten> will it survive a apk upgrade with new kernel on squashfs?
2020-12-29 10:15:32 <artok> new kernel calls for new module
2020-12-29 10:17:20 <sam_son> hmm
2020-12-29 10:19:49 <sam_son> ikke, now my apache won't start:
2020-12-29 10:19:51 <sam_son> Can't locate API module structure `php8_module' in file /var/www/modules/mod_php8.so: Symbol not found: php8_module
2020-12-29 10:20:27 <sam_son> mod_php8.so is there... package php8-apache2 installed fine
2020-12-29 10:20:37 <sam_son> maybe I'm missing something else
2020-12-29 10:21:10 <ikke> sam_son: I have no idea about that, sorry
2020-12-29 10:21:16 <sam_son> no problem
2020-12-29 10:23:32 <sam_son> ok fixed it
2020-12-29 10:23:39 <sam_son> its a mistake in the apk
2020-12-29 10:23:46 <sam_son> file conf.d/php8-module.conf
2020-12-29 10:24:06 <ikke> Could you poen an issue at htps://gitlab.alpinelinux.org/alpine/aports?
2020-12-29 10:24:12 <sam_son> the string php8_module must be renamed to php_module
2020-12-29 10:24:22 <ikke> open*
2020-12-29 10:25:34 <artok> morten: what rpi do you have?
2020-12-29 10:25:58 <morten> 3b+
2020-12-29 10:35:15 <sam_son> ikke, done, issue 12249.
2020-12-29 10:35:31 <ikke> sam_son: yes, noticed it :)
2020-12-29 11:03:18 <sam_son> there, another issue
2020-12-29 11:03:25 <sam_son> I'm on fire...
2020-12-29 11:03:46 <ikke> good to find / report it now :)
2020-12-29 11:03:52 <ikke> (I saw the issue)
2020-12-29 13:55:00 <tpanmajia> I'm using Alpine 3.12.3 with Intel UHD620 and MATE
2020-12-29 13:55:10 <tpanmajia> when I run glxgears the window stucks
2020-12-29 13:55:23 <tpanmajia> only when I resize the window the gears could move
2020-12-29 13:55:37 <tpanmajia> however the FPS displayed in the terminal is vsynced and approximately 60
2020-12-29 13:55:47 <tpanmajia> could anyone help me with what's wrong?
2020-12-29 14:53:20 <Hello71> not if you leave after nine minutes
2020-12-29 14:53:45 <Hello71> wait, actually it's four minutes
2020-12-29 15:09:47 <artok> ping timeout is because morpheus is calling
2020-12-30 09:10:15 <xaltsc> Hey, is there any inconvenient to installing Alpine in diskless mode on a raspberry pi ?
2020-12-30 09:10:56 <nacelle> not remembering to commit a change
2020-12-30 09:11:00 <nacelle> but thats the name of the game
2020-12-30 10:39:35 <eoli3n> Hi
2020-12-30 10:39:38 <eoli3n> https://git.alpinelinux.org/aports/tree/community/nx-libs/APKBUILD#n27
2020-12-30 10:39:46 <eoli3n> what does this do ? "default_prepare"
2020-12-30 10:40:14 <ikke> eoli3n: https://gitlab.alpinelinux.org/alpine/abuild/-/blob/master/abuild.in#L694
2020-12-30 10:40:32 <eoli3n> thanks ikke
2020-12-30 10:40:45 <ikke> Mostly automatically applying any *.patch file defined in $sources
2020-12-30 14:42:24 <EF> Is it sane to add normal users to the "dockergroup"?
2020-12-30 14:43:48 <ikke> EF: it's a security risk
2020-12-30 14:43:59 <ikke> they basicaly become root
2020-12-30 14:45:17 <EF> Is it better/safe to add "dockeruser" to a "user" group?
2020-12-30 14:48:55 <ikke> Not sure what you want to achieve with that
2020-12-30 14:52:51 <EF> Seem to be struggling with permissions between users adding data, docker and the filesystem.
2020-12-30 14:58:48 <ikke> Do you mean files created from docker containers on the host?
2020-12-30 15:02:22 <EF> Yes and user able to copy a file to a docker folder/volume.
2020-12-30 15:02:45 <programmerq> EF▸ if a docker user wants to copy a file to/from a container, 'docker cp' can do that.
2020-12-30 15:10:39 <EF> Think I'm explaining it wrong but thanks for the advice
2020-12-30 15:20:52 <ikke> EF: adding the docker user to a group does not help
2020-12-30 15:21:13 <ikke> What matters what uid / gid is used in the docker container
2020-12-30 15:26:02 <EF> So docker runs as say docker:docker but files on the host should be say user:group and that UID/GID specified in the docker-compose file?
2020-12-30 15:26:41 <ikke> The docker daemon runs as root
2020-12-30 15:27:05 <ikke> To have access to the docker socket (what docker cli uses) you must be either root or member of the docker group
2020-12-30 15:27:38 <ikke> docker containers can run processes with any GID/UID
2020-12-30 15:28:00 <ikke> these can either map directly to the same UIDs / GIDs as the host (default), or different
2020-12-30 15:28:49 <ikke> say you create a user in the docker container, which as UID 1000. When they create a file on a mounted volume, the file is owned by UID 1000
2020-12-30 15:29:08 <ikke> On the host, that is mapped to the username with UID 1000
2020-12-30 15:29:47 <ikke> if you run a process as root, it's also owned by root on the host
2020-12-30 15:33:30 <ikke> (unless you use user namespaces)
2020-12-30 15:52:04 <EF> I'm going to read that several times and try to digest. Thank you for the full explination
2020-12-30 16:02:16 <xordspar0> See also the USER directive in Dockerfile: https://docs.docker.com/engine/reference/builder/#user
2020-12-30 16:02:37 <ikke> EF: It's quite simple. (ignore usernamespaces). The kernel only cares abuit UIDs / GIDs. The names associated with those as defined in /etc/passwd
2020-12-30 16:02:54 <ikke> so a container has a different /etc/passwd, so can show different uersnames for the same ids
2020-12-30 16:03:41 <ikke> But a process running as UID 1000 in the container is the same user as a user with UID 1000 on the host for what the kernel cares about
2020-12-30 16:04:56 <xordspar0> You can create the same user outside and inside a container with adduser (and remember to use the same UID), or you can just look up your host user's UID and do `USER 1001` in your dockerfile.
2020-12-30 16:08:17 <wsinatra> what about programs like salt which need escalated access? Would you still map those to a lower privileged UID?
2020-12-30 16:23:18 <ikke> wsinatra: I don't understand your question
2020-12-30 16:23:54 <ikke> I would not run salt-minion in docker if you want to manage the host
2020-12-30 16:25:25 <wsinatra> I was thinking the other way round, running the salt master node as a docker container
2020-12-30 16:25:47 <wsinatra> My memory of the documentation is a little fuzzy, but I think it requires the salt master be root to actually operate
2020-12-30 16:27:32 <ikke> wsinatra: afaik that's only due to certain directories, you can work around it
2020-12-30 16:27:37 <ikke> ie, file permissions
2020-12-30 16:28:27 <ikke> but you can even use usernamespaces then to still run salt-master as root, but outside of the container it would be different uid
2020-12-30 16:28:54 <wsinatra> Hmmm that sounds really interesting, I don't think I've touched namespaces with docker yet
2020-12-30 16:29:06 <ikke> me neither, but I know it's possible
2020-12-30 16:29:17 <ikke> https://docs.docker.com/engine/security/userns-remap/
2020-12-30 16:30:48 <artok> I rather recommend using named volumes, no bind mounting
2020-12-30 16:30:52 <wsinatra> Glad I asked! This looks really interesting, definitely something I can dig into today
2020-12-30 16:31:54 <ikke> artok: depends on your usecase
2020-12-30 16:35:17 <artok> sure, just haven't seen any =)
2020-12-30 16:35:31 <artok> ...any that would be better without named volume, that is
2020-12-30 16:35:36 <ikke> working on your host, but using containers for certain tasks
2020-12-30 16:36:08 <ikke> not having data locked into docker
2020-12-30 16:41:04 <artok> some dev envs might be, but even then I'm using emacs tramp mode to reach files
2020-12-30 16:41:48 <ikke> somehow tramp is too slow for me
2020-12-30 16:42:03 <ikke> (even with perssitent ssh connection)
2020-12-30 16:42:42 <ikke> For me, working locally + (inotify+rsync) is way faster
2020-12-30 16:51:41 <artok> that's good way yes
2020-12-30 16:52:20 <artok> just haven't checked how to do inotify thingies on macos
2020-12-30 16:54:22 <artok> hah! people are doing those kind of tools with node.js =D
2020-12-30 16:54:29 <mps> hmm, did we got epub reader few months ago? I can't find it
2020-12-30 17:04:08 <ikke> 2019-11-28 10:40:51     mps     I'm contemplating make fbreader apk
2020-12-30 17:05:14 <mps> ikke: yes, and still have unfinished local branch
2020-12-30 17:06:08 <mps> but iirc when I found something other someone created MR or posted patch, and I can't remember pkgname
2020-12-30 17:06:54 <mps> time to dig on old disk
2020-12-30 17:07:39 <ikke> Only see this, but that's more of an extractor: https://gitlab.alpinelinux.org/alpine/aports/-/merge_requests/6005
2020-12-30 17:08:35 <mps> yes, found it
2020-12-30 17:09:07 <mps> ok, disk my come here :)
2020-12-30 17:10:28 <mps> apk info foliate
2020-12-30 17:10:42 <mps> Simple and modern eBook viewer
2020-12-30 17:10:43 <ikke> a, ebook
2020-12-30 17:10:48 <ikke> I searched for epub
2020-12-30 17:11:02 <mps> isn't ebook same as epub?
2020-12-30 17:11:10 <ikke> not if you grep :)
2020-12-30 17:11:28 <mps> right :)
2020-12-30 17:12:16 <mps> don't like to write acronyms but this time 'rofl'
2020-12-30 17:13:35 <mps> huh, doesn't work because I use libudev-zero and libgudev have a bug
2020-12-30 17:14:31 <mps> Failed to load plugin '/usr/lib/gstreamer-1.0/libcluttergst3.so': Error relocating /usr/lib/libgudev-1.0.so.0: udev_device_get_usec_since_initialized: symbol not found
2020-12-30 17:47:27 <yZ5vlALg86lP> vazzer a ccc kiepitett egy mediacsoportot, 25 studioval amit tovabbiakban is akarnak uzemeltetni
2020-12-30 17:47:40 <yZ5vlALg86lP> whooops EWRONGCHAN
2020-12-30 18:27:02 <EF> ikke, xordspar0, think I understand. Thanks again. Trying to learn too much in one go so brain addled.
2020-12-30 19:23:16 <waleee-cl> yZ5vlALg86lP:  out of curiosity, which language was that?
2020-12-30 19:24:50 <mps> 'sounds' like hungarian
2020-12-30 19:25:15 <waleee-cl> yeah, google thought so too
2020-12-30 19:25:24 <mps> heh
2020-12-30 19:26:02 <waleee-cl> I didn't see the telltale signs of moderate sz usage
2020-12-30 19:26:23 <mps> not hard for me because I have friends whose it is native language :)
2020-12-30 19:26:34 <waleee-cl> lots of sz -> polish. Moderate to low, hungarian
2020-12-30 19:27:10 <ikke> also ł is an instant tell-tale
2020-12-30 21:11:30 <unrznbl[m]> by default it seems that my alpine box doesn't send a hostname to the dhcp server, is this changeable? I am not using dhclient (as far as I can tell)
2020-12-30 21:11:53 <ikke> what version of alpine?
2020-12-30 21:12:24 <unrznbl[m]> wait... I did `apk del dhclient` and that fixed it I think, with a reboot :)
2020-12-30 21:12:40 <unrznbl[m]> I was looking at https://serverfault.com/questions/1031911/alpine-linux-is-not-sending-hostname-to-dhcp-server and wanted to avoid installing it but then saw it WAS installed. :p
2020-12-30 21:12:43 <Apachez> did you save the state also with ltu?
2020-12-30 21:12:58 <unrznbl[m]> ltu, not familiar.
2020-12-30 21:13:01 <ikke> lbu
2020-12-30 21:13:08 <Apachez> I meant lbu :)
2020-12-30 21:13:10 <ikke> depends if you use a diskless setup
2020-12-30 21:13:11 <unrznbl[m]> I am full-disk install so maybe not applicable right?
2020-12-30 21:13:32 <ikke> yes
2020-12-30 21:13:43 <unrznbl[m]> I am 3.13.0_alpha20201218 according to os-release. it works! :) So I'm good. Thanks for trying to help!
2020-12-30 21:13:43 <Apachez> then you should be fine
2020-12-30 21:13:55 <ikke> edge switched to ifupdown-ng, there is https://github.com/ifupdown-ng/ifupdown-ng/blob/master/doc/ADMIN-GUIDE.md#ifupdown-ng-configuration use_hostname_for_dhcp
2020-12-30 21:14:03 <unrznbl[m]> yep. thanks!
2020-12-30 21:14:15 <ikke> which apparently is enabled by default
2020-12-30 21:14:30 <unrznbl[m]> yeah, I'm on edge. I use this on my primary hobby laptop (lenovo x220) to work on postmarketos on phones :)
2020-12-30 21:15:02 <ikke> with busybox ifupdown, you had to add `hostname $(hostname)
2020-12-30 21:15:08 <ikke> but that's no longer necessary
2020-12-30 21:15:18 <unrznbl[m]> I really enjoy using alpine btw. Been debian for so long can't say but not interested in that much anymore. Alpine feels much better. Thank you.
2020-12-30 21:16:32 <Apachez> same here but sofar I have only crossedover for rpi :)
2020-12-30 21:17:53 <cim209> alpine is great, been using it for containerz
2020-12-30 21:18:21 <cim209> and also as my servers' host OS
2020-12-31 00:03:20 <whatisthematrix[> Does anyone know a regex for opensmtpd and fail2ban on alpine Linux. I am super struggling to get opensmtpd deployed.
2020-12-31 01:40:54 <Segfault0> Hey I was trying to cross compile Alpine for PPC which has been going surprisingly well so far although I've run into a roadblock with Rust
2020-12-31 01:41:44 <Segfault0> I know it's not officially supported but a lot of packages with architecture specific setup seem to reference PPC (not ppc64/ppc64el) in some way so I decided to give it a shot
2020-12-31 01:42:07 <Segfault0> I've found exactly *one* reference to the same issue I'm having and that seemed to be with trying to compile rust for alpine on s390x https://zulip-archive.rust-lang.org/182449tcompilerhelp/94495crosscompilationandmissingrdlallocandfriends.html
2020-12-31 01:42:46 <Segfault0> Would anyone happen to have any ideas or would it be better for me to ask somewhere else?
2020-12-31 02:12:29 <Hello71> there is ongoing work on cross-enabling rust because it is required for spidermonkey and librsvg and other things
2020-12-31 02:17:23 <Segfault0> Oh yeah I forgot to mention to get to the point where I ran into that issue I had to compile on a 32 bit x86 host, I first tried a 64 bit host and it failed with something along the lines of g++ treating intptr_t and uintptr_t as ints and uints respectively despite compiling code for a 64 bit system
2020-12-31 02:19:08 <Segfault0> Good to see that it's being worked on I guess
2020-12-31 04:50:05 <xordspar0> I have a hard drive that was previously plugged into a Debian server and I'm trying to mount it on my Alpine server. It's formatted with a GPT partition table. When I used lsblk or fdisk on it I can see its partitions, but the partitions don't show up as devices in /dev/sdaX. I only see /dev/sda.
2020-12-31 04:53:03 <xordspar0> Wait, while I was typing that message the partitions showed up. Well, that will remain a mystery.
2020-12-31 07:20:24 <lastchansen> Hi! I just fell in love with Alpine a few days ago and I'm contemplating to migrate from Arch. However, I'm a bit confused about the different versions and how it clashes with what I need on a daily basis. For instance pandoc only exists in edge, but edge is not recommended for anything else than testing? So, am I out of luck in regards to pandoc?
2020-12-31 07:21:13 <maxice8> In this specific instance no because pandoc is compiled statically
2020-12-31 07:21:23 <maxice8> so you can add the `edge` repo with a pin and install pandoc from it
2020-12-31 07:21:35 <maxice8> doesn't apply to packages that link to dynamic libraries (overwhelming majority)
2020-12-31 07:22:35 <lastchansen> Oh.. interesting. So software that compiles statically will always be in edge?
2020-12-31 07:22:46 <lastchansen> I guess this is the same with neomutt?
2020-12-31 07:23:15 <lastchansen> So, I could uncomment edge in /etc/apk/repositories and just install what I need, but note that it could break at all times?
2020-12-31 07:24:10 <maxice8> No, pandoc being compiled statically and being on edge is not related, there is statically compiled stuff on main and community
2020-12-31 07:24:18 <maxice8> it is just a coincidence that works out in your favour in this situation
2020-12-31 07:24:37 <lastchansen> Arh!
2020-12-31 07:25:21 <maxice8> neomutt is not the case
2020-12-31 07:25:46 <maxice8> packages in testing just need the maintainer to give OK to move it to community where they will be in the next Alpine Linux release
2020-12-31 07:25:57 <lastchansen> Okay, but there is no harm in enabling edge and installing them? Can I leave edge uncommented?
2020-12-31 07:26:19 <ikke> yes, there is an issue with that
2020-12-31 07:26:23 <maxice8> If a package you want is in testing and you want it in community then contact the Maintainer of the APKBUILD or make a merge request and tag the maintainer
2020-12-31 07:26:54 <maxice8> stuff you install from edge is prone to break when in an stable release and is also not supported so we will ask that you try it without testing enabled beforehand.
2020-12-31 07:27:46 <lastchansen> Okay, I'll see about requesting it being moved and see if I can make do without :) Thanks for the quick answers.
2020-12-31 11:09:52 <telmich> It seems that my alpine boxes generate a new ipv6 address on every reboot - does anyone have a pointer on which configuration that is triggering?
2020-12-31 11:10:12 <telmich> The only non-standard setting I found so far was addr_gen mode set to 1, but I did not locate the source of that one so far either
2020-12-31 11:10:56 <ikke> telmich: do you mean a different slaac address?
2020-12-31 11:11:05 <telmich> Yes
2020-12-31 11:13:16 <telmich> I don't exactly mind how an address is generated, but what I need on these boxes are stable addresses
2020-12-31 11:13:38 <ikke> That would mean it needs to store that address soemwhere. not?
2020-12-31 11:13:59 <ikke> I don't think the kernel itself has a way to persist that
2020-12-31 11:15:31 <telmich> Well, by default EUI-64 embedding would be used based on the mac address and that is rather stable
2020-12-31 11:16:03 <telmich> But even if the stable secret was used, the "os" needs to store the seed and re-seed on boot -> this also ensures there is a stable address
2020-12-31 11:25:13 <ikke> right
2020-12-31 11:36:07 <ikke> /proc/sys/net/ipv6/conf/eth0/stable_secret
2020-12-31 11:38:53 <ikke> telmich: fyi, for me the addr_gen_mode is 0
2020-12-31 12:37:30 <m99> Hello, where to ask to get 'm' in kernel config? ( asking for CONFIG_VIDEO_EM28XX_DVB )
2020-12-31 12:47:06 <ikke> m99: create an issue on https://gitlab.alpinelinux.org/alpine/aports
2020-12-31 12:49:29 <m99> tnx
2020-12-31 12:50:31 <miskatonic> does that gitlab thing require proprietary javascript?
2020-12-31 12:51:18 <ikke> depends on what you call proprietary js
2020-12-31 12:51:44 <ikke> But it does probably use minimized JS libraries
2020-12-31 13:07:10 <telmich> ikke: that is interesting
2020-12-31 14:29:19 <xaltsc> Hey, I'm trying to install alpine on a rpi4b using this tutorial https://wiki.alpinelinux.org/wiki/Classic_install_or_sys_mode_on_Raspberry_Pi, but the rpi4 doesn't even boot on it. So far, I've narrowed to problem to come either from the SD Card or the tutorial, but in the event that the tutorial is the pb (I'm checking it right now), what should I do ?
2020-12-31 14:34:29 <guix-vits> xaltsc:  u-boot should be installed to some specific sector of sd, maybe?
2020-12-31 14:35:41 <guix-vits> some encyclopedical thingy: https://wiki.alpinelinux.org/wiki/DIY_Fully_working_Alpine_Linux_for_Allwinner_and_Other_ARM_SOCs
2020-12-31 14:36:17 <xaltsc> There are so many rpi/ARM tutorials on the wiki..
2020-12-31 14:36:39 <guix-vits> RPi must be has something like this: https://github.com/sigmaris/u-boot/wiki/RockPro64-boot-sequence
2020-12-31 14:37:10 <guix-vits> idk RPi
2020-12-31 14:37:37 <guix-vits> ofc RPi will differ
2020-12-31 14:37:52 <xaltsc> Well, alpine provides images specifically for RPis, so I would guess they have that already ?
2020-12-31 14:38:26 <guix-vits> idk
2020-12-31 14:44:31 <guix-vits> xaltsc: (?) https://wiki.alpinelinux.org/wiki/Raspberry_Pi
2020-12-31 14:47:28 <mps> guix-vits: RPIs don't use u-boot as first stage loader
2020-12-31 14:47:44 <guix-vits> i seen ye
2020-12-31 14:48:19 <guix-vits> (wtf it not boots then ;)
2020-12-31 14:50:00 <mps> here is my short notes how I install alpine on rpi zero, http://arvanta.net/alpine/flash-rpi-zero/
2020-12-31 14:50:32 <xaltsc> Yeah but I dont want to run it in diskless mode...
2020-12-31 14:51:23 <xaltsc> mps: Does your installation result in a diskless install or not ?
2020-12-31 14:51:29 <mps> xaltsc: I have not yet finished notes for this
2020-12-31 14:52:09 <xaltsc> I just need to find a way to boot, I'll figure out what's next afterwards anyway
2020-12-31 14:52:16 <mps> xaltsc: above link is just how to flash image ready for setup-alpine
2020-12-31 14:53:21 <xaltsc> mps: Perhaps I should do that on a usb key then setup-alpine on the sdcard ? I don't know if rpis can actually boot on usb keys.
2020-12-31 14:53:41 <mps> here is how I install armv7/armhf in sys mode https://tpaste.us/x1B1
2020-12-31 14:54:19 <mps> using chroot to finish and set bootable mmc
2020-12-31 14:55:44 <mps> xaltsc: though I don't have much experience with rpis, I use different arm SBCs
2020-12-31 14:56:00 <xaltsc> Since you're using apk, I assume you're doing this from alpine already ? Would this work with a live usb ?
2020-12-31 14:56:44 <mps> yes, from alpine. but there is apk-static which can be used from other linux distros
2020-12-31 14:57:19 <xaltsc> Yeah, I just found out my distro provides apk
2020-12-31 15:14:54 <xaltsc> mps: So, using your instructions, I get errors 127 when installing packages. Is that normal ?
2020-12-31 15:15:07 <xaltsc> (the line with apkà
2020-12-31 15:15:14 <xaltsc> s/à/)/
2020-12-31 15:16:10 <xaltsc> http://pastebin.fr/79387
2020-12-31 15:27:28 <mps> xaltsc: you are not installing on aarch64/arm64 machine?
2020-12-31 15:35:21 <andre> Good Day guys
2020-12-31 15:35:41 <Guest41427> Site wiki offline ?
2020-12-31 15:36:01 <caskd> fine here
2020-12-31 15:36:14 <caskd> what page did you try to access?
2020-12-31 15:36:33 <Guest41427> I need install MATE alpine, your have how too ?
2020-12-31 15:37:09 <Guest41427> Wiki.alpinelinux.org
2020-12-31 15:37:59 <Guest41427> What error 404
2020-12-31 15:38:13 <Guest41427> Now access !!!
2020-12-31 15:38:52 <Guest41427> Cloud be my DNS instable
2020-12-31 15:39:14 <Guest41427> Thanks @caskd
2020-12-31 15:40:54 <Guest41427> My DNS is really in trouble
2020-12-31 15:55:10 <xaltsc> mps: I am, on a aarch64
2020-12-31 15:56:37 <mps> hmm, then it should work
2020-12-31 15:59:25 <xaltsc> I've changed the --arch argument to aarch64, pehaps I shouldn't have done that ?
2020-12-31 16:00:05 <mps> it is ok
2020-12-31 16:00:42 <mps> just copy-pasted your invoke from pastebin and it worked without problem here
2020-12-31 16:01:06 <xaltsc> Ok, perhaps it's due to voidlinux then...
2020-12-31 16:02:05 <mps> could you try with apk.static, maybe will help
2020-12-31 16:02:59 <mps> though wonder why it doesn't work on void out-of-the-box
2020-12-31 16:53:47 <aeroraptor> anyone familiar with any issues using a pl2303 based USB-UART interface recently?
2020-12-31 16:54:15 <aeroraptor> whenever I try and access ttyUSB0 I get a generic "stty: can't open '/dev/ttyUSB0': I/O error"
2020-12-31 16:54:19 <aeroraptor> dmesg is throwing "[ 3618.677675] pl2303 ttyUSB0: failed to submit interrupt urb: -28"
2020-12-31 16:54:38 <aeroraptor> googling the second error is turning up a good amount of old stuff, but nothing recent
2020-12-31 16:54:43 <Hello71> sounds like connection issue
2020-12-31 16:55:35 <aeroraptor> I'll try another USB port
2020-12-31 16:56:44 <aeroraptor> doh go figure that worked
2020-12-31 16:56:52 <aeroraptor> I rebooted like 4 times to no avail, but a new USB port did it
2020-12-31 16:56:58 <aeroraptor> now I'm suspicious of that USB port
2020-12-31 17:00:32 <mps> aeroraptor: also happened to me about 3 months ago, on one external usb hub
2020-12-31 17:00:47 <aeroraptor> oh interesting! so it's a generic USB error, not specific to this device
2020-12-31 17:00:49 <aeroraptor> curious!
2020-12-31 17:02:20 <mps> I have about 8-10 pl2303 around and they works fine except one which I damaged by wrong connetion
2020-12-31 17:06:13 <Hello71> if your adapter is not that good you may also damage the motherboard
2020-12-31 17:16:59 <aeroraptor> also, while I'm here, I have a noob OpenRC question
2020-12-31 17:17:11 <aeroraptor> what's the best way to make sure a service starts before another service
2020-12-31 17:17:23 <aeroraptor> (in this case I need to make sure gpsd starts *after* chronyd)
2020-12-31 17:17:45 <ikke> aeroraptor: https://github.com/OpenRC/openrc/blob/master/service-script-guide.md#the-depend-function
2020-12-31 17:17:56 <aeroraptor> thank you!
2020-12-31 17:34:48 <pernia> racism
2020-12-31 17:53:29 <mps> ikke: irony is that maintainer of zathura-pdf-mupdf forgot that this pkg read epub :)
2020-12-31 17:53:42 <ikke> heh
2020-12-31 18:26:58 <xaltsc> Hey, so since my troubles earlier, I've managed to install alpine, however I do not have dns resolution. I've tried to start dnsd but /usr/sbin/dnsd does not exist.
2020-12-31 18:27:51 <ikke> is there anything in /etc/resolv.conf?
2020-12-31 18:28:01 <ikke> you don't need a daemon for dns resolution
2020-12-31 18:28:03 <mps> apk add busybox-extra
2020-12-31 18:28:30 <xaltsc> ikke: nameserver 192.168.1.254
2020-12-31 18:28:56 <xaltsc> mps: I get unsatisfied constraints
2020-12-31 18:29:29 <xaltsc> The computuer I'm on has the same (plus the IPv6 address)
2020-12-31 18:29:49 <ikke> xaltsc: what does `getent hosts alpinelinux.org` return?
2020-12-31 18:29:56 <mps> busybox-extras not busybox-extra
2020-12-31 18:30:23 <mps> sorry for confusion
2020-12-31 18:30:55 <xaltsc> mps: I get temporary error, which may be due to the server itself ?
2020-12-31 18:31:13 <xaltsc> I'm using alpine.42.fr apparently
2020-12-31 18:31:50 <xaltsc> ikke: getent returns nothing
2020-12-31 18:33:05 <xaltsc> I've added google dns server and it works, but this is suboptimal
2020-12-31 18:34:01 <ikke> seems like 192.168.1.254 does not resolve dns
2020-12-31 18:34:49 <xaltsc> It does, that's the one I have on my main PC
2020-12-31 18:35:56 <mps> not that I want to tell you what to use but dnsmasq is quite fine for me for different things
2020-12-31 18:57:40 <stee> Quick one: how can I list the contents of a package?
2020-12-31 18:58:01 <mps> apk info -L pkgname
2020-12-31 18:58:10 <ikke> mps: nice, you are faster then me once ;-)
2020-12-31 18:58:20 <mps> hehe
2020-12-31 18:58:27 <stee> ha thanx
2020-12-31 18:59:27 <mps> ikke: maybe I'm just got lottery this time
2020-12-31 19:57:41 <meowray> https://reviews.llvm.org/D67516 is this clang driver patch for musl detection needed?
2020-12-31 20:21:15 <artok> meowray: as long as there would be all other archs also
2020-12-31 23:16:07 <Apachez> Happy New Year!
2020-12-31 23:19:27 <mps> Apachez: thanks, also happy new year
2020-12-31 23:27:10 <caskd> Happy new year!