2020-01-02 18:15:55 <nexgen2> (alpine 3.7
2020-01-02 18:15:57 <nexgen2> )
2020-01-02 18:16:57 <nexgen2> udhcpc
2020-01-02 18:16:59 <nexgen2> udhcpc: started, v1.31.1
2020-01-02 18:17:03 <nexgen2> udhcpc: socket(AF_PACKET,2,8): Address family not supported by protocol
2020-01-02 18:25:51 <AinNero> im having trouble even interpreting those numbers
2020-01-02 18:29:07 <nexgen2> how to disable ipv6 in udhcpc ?
2020-01-02 18:30:40 <AinNero> im surprised udhcpc does ipv6 in the first place
2020-01-02 18:32:34 <nexgen2> are there any alternatives to udhcpc in Alpine?
2020-01-02 18:32:39 <nexgen2> can I switch it please?
2020-01-02 18:33:06 <AinNero> yes?
2020-01-02 18:33:15 <nexgen2> though may be it is not ipv6 problem
2020-01-02 18:34:39 <AinNero> udhcpc6 is a separate program
2020-01-02 18:35:03 <AinNero> so i think the "normal" udhcpc does not have IPv6 support to begin with
2020-01-02 18:36:09 <nexgen2> > udhcpc: socket(AF_PACKET,2,8): Address family not supported by protocol
2020-01-02 18:36:29 <AinNero> IPv6 would be AF_INET6
2020-01-02 18:37:55 <mps> /etc/network/if-up.d/dad
2020-01-02 18:40:04 <AinNero> for reference, the both numbers are SOCK_DGRAM and ETH_P_IP
2020-01-02 18:41:02 <AinNero> nexgen2: what kind of kernel do you run?
2020-01-02 18:41:25 <nexgen2> Linux kurgan_alpine 5.4.0-1-virt #2-Alpine SMP Wed, 27 Nov 2019 12:15:36 UTC x86_64 Linux
2020-01-02 18:41:47 <nexgen2> it worked before upgrade a few weeks ago
2020-01-02 18:42:13 <nexgen2> 5.4.0-1-virt
2020-01-02 18:45:25 <nexgen2> going to install dhcpcd-8.1.2-r0
2020-01-02 18:50:08 <AinNero> do you have af_packet in lsmod?
2020-01-02 18:51:44 <AinNero> i suspect the autoloading of the af_packet kernel module could have been accidentally removed
2020-01-02 18:52:26 <nexgen2> no
2020-01-02 18:52:26 <nexgen2> modprobe af_packet
2020-01-02 18:52:27 <nexgen2> modprobe: FATAL: Module af_packet not found in directory /lib/modules/5.4.0-1-virt
2020-01-02 18:53:52 <AinNero> does '/lib/modules/5.4.0-1-virt' exist?
2020-01-02 18:54:00 <nexgen2> sure not :(
2020-01-02 18:54:07 <nexgen2> upgrade replaced the kernel
2020-01-02 18:54:12 <AinNero> you need to reboot after upgrading kernel
2020-01-02 18:54:13 <nexgen2> fixing
2020-01-02 18:54:52 <nexgen2> I used an external kernel supplied to qemu command line
2020-01-02 18:55:01 <nexgen2> there is an old copy
2020-01-02 18:55:53 <nexgen2> there were rumors about instability of kernel v5.4.x
2020-01-02 18:56:07 <nexgen2> is it better to downgrade to v5.3.x?
2020-01-02 18:56:21 <AinNero> ah yes. i have code in my bash history to warn me when the kernel modules mismatch the kernel
2020-01-02 18:56:25 <AinNero> *bashrc
2020-01-02 18:56:54 <nexgen2> how can I fix a package version?
2020-01-02 18:57:04 <nexgen2> like apt-mark hold kernel-xxx
2020-01-02 18:57:15 <mps> my personal experience is that the 5.4 is more problematic than 5.3
2020-01-02 18:57:31 <nexgen2> I mean fixation at a specific version to avoid further upgrades
2020-01-02 18:57:51 <nexgen2> going to downgrade to v5.3.x latest
2020-01-02 18:58:01 <AinNero> install it with linux-lts==...
2020-01-02 18:58:04 <nexgen2> and fixate it from upgrade
2020-01-02 18:58:11 <mps> but 5.4.7 fixed some issues and it is uploaded today to edge
2020-01-02 18:58:12 <AinNero> or edit its entry in /etc/apk/world
2020-01-02 19:02:25 <monado> Seems like a vpn fixes it
2020-01-02 19:03:49 <nexgen2> https://pkgs.alpinelinux.org/packages?name=*linux-virt*&branch=v3.11
2020-01-02 19:04:08 <nexgen2> please let me know, where can I get older kernel like 5.3.x ?
2020-01-02 19:04:24 <nexgen2> and preferably Libre without obfusctaed blobs
2020-01-02 19:04:46 <qiu3344> There is no linux-libre package on Alpine
2020-01-02 19:05:08 <Cogitri> And no older kernel packages.
2020-01-02 19:05:09 <qiu3344> If you want to use it you need to compile it from source
2020-01-02 19:05:11 <nexgen2> most likely I will be able to copy it from Debian later
2020-01-02 19:05:26 <nexgen2> for now a general kernel would suit
2020-01-02 19:05:29 <mps> what is linux-libre
2020-01-02 19:05:42 <nexgen2> it is a deblobbed kernel
2020-01-02 19:06:13 <mps> aha, thanks
2020-01-02 19:06:17 <nexgen2> cat libre-kernel.list
2020-01-02 19:06:17 <nexgen2> deb [arch=amd64] mirror://linux-libre.fsfla.org/pub/linux-libre/freesh/mirrors.txt freesh main
2020-01-02 19:06:17 <nexgen2> deb [arch=armhf] mirror://linux-libre.fsfla.org/pub/linux-libre/freesh/mirrors.txt freesh main
2020-01-02 19:06:25 <armin> that involves that the default kernel has blobs, is that correct?
2020-01-02 19:06:35 <nexgen2> yes :(
2020-01-02 19:06:45 <armin> wow, i didn't know
2020-01-02 19:06:46 <nexgen2> these blobs are "open source"
2020-01-02 19:06:53 <nexgen2> but obfuscated
2020-01-02 19:06:57 <qiu3344> I mean you can always don't include them in your build
2020-01-02 19:07:21 <nexgen2> what about older kernel please?
2020-01-02 19:07:30 <qiu3344> just don't select them in "make nconfig"
2020-01-02 19:07:38 <nexgen2> is there an archive of earlier packages like in Debian?
2020-01-02 19:07:52 <_ikke_> No, we don't have the storage space to keep all packages
2020-01-02 19:07:59 <_ikke_> We tried
2020-01-02 19:08:16 <qiu3344> Maybe try 5.4.7?
2020-01-02 19:08:32 <nexgen2> no other fast ways anyway :)
2020-01-02 19:09:14 <nexgen2> was there anything critical in 5.4.6 ?
2020-01-02 19:09:46 <nexgen2> linux-lts-5.4.6-r0 x86_64 {linux-lts} (GPL-2.0) [installed]
2020-01-02 19:12:43 <mps> nothing critical, but some people have issues with i915 driver
2020-01-02 19:14:20 <nexgen2> oh, rebooted with new kernel and dhcp works, thank you very much
2020-01-02 19:14:36 <nexgen2> you make the fastest package manager
2020-01-02 19:14:52 <nexgen2> is it possible somehow to verify its packages?
2020-01-02 19:15:06 <nexgen2> like debsums
2020-01-02 19:15:14 <mps> apk verify
2020-01-02 19:15:24 <mps> not like debsums
2020-01-02 19:16:40 <AinNero> how is debsums useful when the packages are signed anyways?
2020-01-02 19:17:24 <nexgen2> if hacked may be
2020-01-02 19:17:44 <AinNero> if hacked the signatures wont match
2020-01-02 19:18:00 <nexgen2> that is what I was looking for, to verify them
2020-01-02 19:18:01 <AinNero> both dpkg and apk would refuse to even install it then
2020-01-02 19:18:18 <nexgen2> 1) install good package
2020-01-02 19:18:53 <nexgen2> 2) someone connects to the server unauthorized and modifies something
2020-01-02 19:19:05 <nexgen2> 3) I need to verify there are no changes
2020-01-02 19:19:57 <qiu3344> Wat
2020-01-02 19:22:21 <mps> 'someone connects to the server unauthorized and modifies something' and modifies it in such way that the signature stays same as original
2020-01-02 19:22:25 <nexgen2> I did not understand how to fixate a package at a specific level
2020-01-02 19:22:25 <mps> :)
2020-01-02 19:22:50 <nexgen2> googling though
2020-01-02 19:22:52 <qiu3344> (:
2020-01-02 19:23:05 <AinNero> mps: why bother if you can hack apk to always act like the signature matches
2020-01-02 19:23:31 <mps> AinNero: you noticed my ':)'
2020-01-02 19:24:32 <nexgen2> ah, verify
2020-01-02 19:24:36 <nexgen2> man missing
2020-01-02 19:24:37 <AinNero> somehow apk verify does not check anything
2020-01-02 19:24:45 <mps> btw, did you reported your issue with vt52 (or similar) to ncurses upstream
2020-01-02 19:24:46 <AinNero> judging from the strace log
2020-01-02 19:25:10 <AinNero> mps: no, i found even more differences
2020-01-02 19:25:26 <AinNero> i think the ncurses vt520 is a different kind of device than i physically posess
2020-01-02 19:25:39 <AinNero> F keys all generate different codes
2020-01-02 19:25:46 <nexgen2> apk verify
2020-01-02 19:25:57 <mps> ah, you have 'real' terminal
2020-01-02 19:26:00 <nexgen2> exits almost at one
2020-01-02 19:26:11 <nexgen2> *once
2020-01-02 19:26:38 <nexgen2> for example another tool like:
2020-01-02 19:26:46 <nexgen2> wajig integrity
2020-01-02 19:26:51 <AinNero> nexgen2: apk verify does not check the extracted file contents
2020-01-02 19:26:52 <nexgen2> works for a long time
2020-01-02 19:27:13 <AinNero> just that the pkgs in the cache and repo stuff matches
2020-01-02 19:27:16 <nexgen2> then there is no a tool for verification of installed packages?
2020-01-02 19:27:30 <AinNero> dunno
2020-01-02 19:27:44 <qiu3344> Never heard of something like that
2020-01-02 19:27:59 <nexgen2> RPM yum has one?
2020-01-02 19:29:02 <nexgen2> may be later apk will get it
2020-01-02 19:29:20 <nexgen2> it is not safe without verification
2020-01-02 19:30:03 <nexgen2> Though I have a good knowledge of Debian and Devuan without systemD
2020-01-02 19:30:14 <nexgen2> I like Alpine very much for its fast apk
2020-01-02 19:30:20 <nexgen2> almost instant often
2020-01-02 19:30:36 <AinNero> nexgen2: packages are signed and verified on install time
2020-01-02 19:30:43 <nexgen2> it is clear
2020-01-02 19:30:50 <AlpineDiscordBri> [discord] <monado> excuse my ignorance but what exactly are you verifying against
2020-01-02 19:30:51 <AinNero> if an attacker gets in and modifies stuff, your apk is compromised as well
2020-01-02 19:31:01 <nexgen2> the problem is about how to find later changes
2020-01-02 19:31:26 <nexgen2> I can reinstall a few packages like apk again
2020-01-02 19:31:34 <AlpineDiscordBri> [discord] <monado> I see
2020-01-02 19:31:46 <nexgen2> can chroot from another kernel
2020-01-02 19:31:50 <nexgen2> from livecd
2020-01-02 19:32:13 <AinNero> ... and run the compromised shell?
2020-01-02 19:32:19 <monado> Does alpine have a chroot install option?
2020-01-02 19:32:35 <AinNero> monado: there are chroot tarballs on the official page
2020-01-02 19:32:41 <nexgen2> AinNero, there are many thing to be compromissed
2020-01-02 19:32:49 <monado> oh, maybe I failed to see them. Thanks a lot
2020-01-02 19:33:03 <nexgen2> shall can be reinstalled too
2020-01-02 19:33:19 <AinNero> apk verify checking the installed files would be interesting, possibly
2020-01-02 19:33:27 <AinNero> lbu diff already does a similar thing for /etc
2020-01-02 19:33:41 <nexgen2> deb and rpm have integrity checks
2020-01-02 19:33:53 <nexgen2> IMHO, it is a must for production
2020-01-02 19:34:10 <AinNero> feel free to patch up apk
2020-01-02 19:34:34 <nexgen2> I would be glad, but I am very tight on my resources like time
2020-01-02 19:34:45 <Cogitri> monado: You can just get a static copy of apk-tools from the mirrors and bootstrap a chroot from that
2020-01-02 19:34:55 <Cogitri> That's what I usually do to install on ZFS
2020-01-02 19:34:59 <AinNero> nexgen2: how is that different from "work for me for free" ?
2020-01-02 19:35:21 <AinNero> afaik most ppl here have a dayjob or even family
2020-01-02 19:35:26 <nexgen2> actually I would like to write a single interface manager for Devuan, Alpine and OpenBSD
2020-01-02 19:35:39 <nexgen2> the same syntax, same commands
2020-01-02 19:35:52 <monado> I think I might need to use alpine a little more, and on bare metal
2020-01-02 19:35:58 <nexgen2> say: ./apm.sh verify
2020-01-02 19:36:10 <nexgen2> I have it for dev and rpm already
2020-01-02 19:36:19 <nexgen2> it is just a wrapper of course
2020-01-02 19:36:28 <nexgen2> syntax unification
2020-01-02 19:37:36 <nexgen2> AinNero, I am not clear about your last phrase "work for free"
2020-01-02 19:38:31 <nexgen2> you mentioned to try to improve apk (development), it is a long task
2020-01-02 19:38:43 <nexgen2> not affordable for me right now
2020-01-02 19:38:52 <nexgen2> that is what I wrote earlier
2020-01-02 19:39:06 <nexgen2> I do not ask anyone to code for me free of charge
2020-01-02 19:40:01 <nexgen2> It would be good if some big company would sponsor Alpine for features like: integrity check
2020-01-02 19:40:46 <AinNero> usually they would implement it themselves and submit it
2020-01-02 19:40:56 <nexgen2> good too
2020-01-02 19:41:09 <AinNero> like IBM payed that one guy to improve s390 support
2020-01-02 19:41:13 <nexgen2> did someone try to force you to integrate systemD, btw?
2020-01-02 19:41:38 <AinNero> not without pay
2020-01-02 19:41:58 <nexgen2> and you would do it for money?
2020-01-02 19:42:01 <nexgen2> good money
2020-01-02 19:42:12 <AinNero> no, im outbooked
2020-01-02 19:42:21 <nexgen2> we found systemD is an NSA trojan
2020-01-02 19:42:38 <nexgen2> there are so few distros without it now
2020-01-02 19:42:48 <monado> how is it a trojan
2020-01-02 19:42:50 <nexgen2> Devuan, Alpine, GUIX
2020-01-02 19:42:50 <qiu3344> Now that is a wild statement
2020-01-02 19:42:54 <AinNero> dunno. im not in the mood for tinfoil-hatting
2020-01-02 19:42:55 <monado> not trying to incite anything
2020-01-02 19:43:06 <monado> I’d just like to be educated
2020-01-02 19:43:07 <north1> nexgen2 (IRC): now that is the kind of comment i would see in Phoronix
2020-01-02 19:43:14 <nexgen2> systemD on the one hand is a unification of many APIs
2020-01-02 19:43:48 <AinNero> we go our alpine way. the systemd distros go their way
2020-01-02 19:43:49 <AinNero> done
2020-01-02 19:44:03 <nexgen2> very good, even excellent
2020-01-02 19:44:11 <nexgen2> systemD is why I am here :)
2020-01-02 19:44:17 <monado> is it that systemd try’s to do too much
2020-01-02 19:44:19 <qiu3344> I mean, I don't like systemd either but I wouldn't call it a trojan lol
2020-01-02 19:44:24 <nexgen2> I mean I need a systemD free distro
2020-01-02 19:44:54 <monado> I like that alpines just fast. I use void on my desktop, but runit wasn’t the draw for me
2020-01-02 19:45:00 <nexgen2> systemD is an interface between hardware backdoors and programs
2020-01-02 19:45:15 <monado> what about other init systems?
2020-01-02 19:45:28 <nexgen2> it allows to make a planetaty unified grid of computers controlled by NSA
2020-01-02 19:45:36 <qiu3344> monado: What were you missing in runit? Just curious
2020-01-02 19:45:44 <nexgen2> in more convenient way than they control them rigth now
2020-01-02 19:46:01 <monado> qiu3344: no, I’m a relatively new Linux user in the grand scheme of things
2020-01-02 19:46:26 <Cogitri> qiu3344: Proper oneshots, proper dependency management of services...
2020-01-02 19:46:43 <monado> Other parts of void inticed me, i run a Fedora Silverblue system on my other laptop
2020-01-02 19:47:24 <qiu3344> Do you run alpine on any PCs?
2020-01-02 19:47:44 <monado> not yet
2020-01-02 19:48:18 <monado> I’ve got one virtual machine so I could get a taste
2020-01-02 19:48:35 <qiu3344> I recently switched my laptop it and it's really working smoothly
2020-01-02 19:48:48 <benharri> anyone tried to run gnome lately?
2020-01-02 19:48:50 <monado> Alpine on your laptop?
2020-01-02 19:48:57 <qiu3344> Yup
2020-01-02 19:49:00 <monado> I run gnome on my silverblue system
2020-01-02 19:49:05 <Cogitri> benharri: Yes, I've been using GNOME ever since
2020-01-02 19:49:09 <nexgen2> VOID linux seems to have a few of maintainers?
2020-01-02 19:49:15 <nexgen2> the same about Slackware
2020-01-02 19:49:19 <monado> some are here
2020-01-02 19:49:34 <benharri> i've been considering running alpine on my laptop as well, been using fedora and debian sid lately
2020-01-02 19:49:40 <monado> duncaen and maldridge are the ones I can name
2020-01-02 19:49:57 <benharri> dnf is impressively slow, especially compared to apk
2020-01-02 19:50:22 <monado> Fedora Silverblue seems like an interesting paradigm, which is why I want to hear criticisms of systemd. Also, almost anything is faster than DNF
2020-01-02 19:50:26 <nexgen2> even apt is very slow compared to apk on HDDs
2020-01-02 19:50:30 <qiu3344> Alpine is a really good choise. Especially if you laptop has only 32Gbs of storage ;)
2020-01-02 19:50:36 <nexgen2> about 10 times slower
2020-01-02 19:50:42 <benharri> apk is incredibly fast
2020-01-02 19:51:01 <nexgen2> is it possible to rebuild Alpine for i486?
2020-01-02 19:51:06 <AinNero> one think i love alpine is that you dont have annoying gimmicks that try to be helpful and get into your way instead
2020-01-02 19:51:08 <benharri> i've only run alpine in VMs so far, so i guess that's the next step
2020-01-02 19:51:14 <monado> apk and xbps are great
2020-01-02 19:51:37 <nexgen2> Alpine might be good for a secure console on i486 computer
2020-01-02 19:51:42 <AinNero> Ubuntu had modem manager per default, it opened and wrote to every new tty per default
2020-01-02 19:51:51 <nexgen2> with a less amount of hardware backdoors
2020-01-02 19:51:59 <AinNero> ... even when it was just a USB adapter into a routers root shell
2020-01-02 19:52:01 <qiu3344> monado: but xbps-src is abit annoying to use
2020-01-02 19:52:22 <monado> I find it alright once I read the docs
2020-01-02 19:52:43 <monado> and I’d be considered noob level in a place like this
2020-01-02 19:52:54 <qiu3344> Yeah but downloading the whole git repo is a waste of resources
2020-01-02 19:53:27 <monado> I guess, but I can’t come up with better solutions myself
2020-01-02 19:54:00 <qiu3344> apkbuilds?
2020-01-02 19:54:45 <monado> is this an arch thing?
2020-01-02 19:54:58 <monado> or that’s pkgbuilds
2020-01-02 19:55:05 <qiu3344> No those were pkgbuilds
2020-01-02 19:55:21 <qiu3344> apkbuilds are an Alpine thing
2020-01-02 19:55:27 <monado> I haven’t used arch, and I still need to use more of alpine
2020-01-02 19:55:39 <monado> I came to void from Fedora
2020-01-02 19:56:18 <qiu3344> After using Alpine for a while arch is unusable ;)
2020-01-02 19:56:43 <qiu3344> Just so much heavier in comparison
2020-01-02 19:57:39 <monado> Fedora Workstation is looking quite big too, but Silverblue is an interesting take on the Desktop Linux paradigm
2020-01-02 20:01:23 <nexgen2> I found usable only two ones for now: Devuan and Alpine
2020-01-02 20:01:43 <nexgen2> both are excellent
2020-01-02 20:02:12 <nexgen2> Devuan has a very popular package format, more powerful package manager
2020-01-02 20:02:29 <nexgen2> Alpine advantage is only apk performance
2020-01-02 20:02:43 <AinNero> + binary size + ram usage
2020-01-02 20:02:56 <nexgen2> and existence of Alpine as an alternative is important too
2020-01-02 20:03:06 <AinNero> Alpine is small enough that diskless becomes a realistic thing
2020-01-02 20:03:12 <nexgen2> yes, Alpine is very light
2020-01-02 20:03:22 <AinNero> several companies use it as netbooted live system
2020-01-02 20:03:37 <AinNero> essential for provisioning workflows
2020-01-02 20:04:53 <monado> I like my rolling release, so no devuan for me
2020-01-02 20:05:37 <monado> nexgen2 can you point me to somewhere that isn’t suckless so I can learn more about systemd and it’s implications?
2020-01-02 20:06:09 <nexgen2> I hate systemD
2020-01-02 20:06:26 <monado> yes, why? I want to read
2020-01-02 20:06:39 <nexgen2> unification might be good if its implementation would not be so buggy
2020-01-02 20:06:43 <monado> Not trying to incite anything, just want to educate myself
2020-01-02 20:07:26 <nexgen2> OpenRC is 100 times more stable than systemD because it is just RC
2020-01-02 20:07:38 <nexgen2> while systemD is a kernel number TWO
2020-01-02 20:07:51 <benharri> it's not a kernel though
2020-01-02 20:08:06 <nexgen2> in term of CPU rings it is not a kernel
2020-01-02 20:08:27 <nexgen2> but in terms of how many things systemD tries to do now it is a second kernel
2020-01-02 20:08:53 <nexgen2> APIs are accumulated in systemD now
2020-01-02 20:09:12 <nexgen2> systemD even manages kernel parameters already
2020-01-02 20:09:30 <monado> by the way, isn’t it systemd
2020-01-02 20:09:39 <nexgen2> corporations could not pawn kernel number 1 for some reasons
2020-01-02 20:09:42 <monado> (sans the capital D)
2020-01-02 20:09:43 <benharri> i generally don't have a problem with it, it's a reasonably consistent model/interface
2020-01-02 20:09:50 <nexgen2> and they have created kernel number two
2020-01-02 20:10:15 <nexgen2> for server it is awful
2020-01-02 20:10:32 <nexgen2> it makes very hard to recover a server from a problem
2020-01-02 20:10:37 <nexgen2> it is too complex
2020-01-02 20:10:45 <nexgen2> it may be good for a desktop
2020-01-02 20:10:50 <benharri> it's been pretty great for me on my servers
2020-01-02 20:10:52 <nexgen2> not for 24x7 server
2020-01-02 20:11:02 <monado> I use my Mac as a Minecraft server
2020-01-02 20:11:03 <nexgen2> great until broken
2020-01-02 20:11:13 <benharri> ymmv but i haven't had any major issues
2020-01-02 20:11:15 <monado> so I don’t think I’m the prime user
2020-01-02 20:11:47 <nexgen2> systemD breaks main principle of reliable systems
2020-01-02 20:11:48 <nexgen2> KISS
2020-01-02 20:13:00 <opal> systemd breaks*
2020-01-02 20:13:51 <opal> i'd have less of a problem with it if it was written competently
2020-01-02 20:14:12 <opal> any time it breaks, though, its too complex for me to look at it and immediately figure out
2020-01-02 20:14:36 <opal> i hold the belief that complex software is not free
2020-01-02 20:14:56 <opal> or at least less free
2020-01-02 20:15:06 <nexgen2> systemD complexity is good for their sponsors for many reasons
2020-01-02 20:15:53 <nexgen2> it is like an obfuscated trojan which cannot be coded by anyone except their own coder just because of mess they produced
2020-01-02 20:16:09 <nexgen2> they crippled Debian
2020-01-02 20:16:23 <nexgen2> people run from Debian to Centos and Redhat
2020-01-02 20:16:28 <nexgen2> profit ! :)
2020-01-02 20:16:43 <AlpineDiscordBri> [discord] <monado> interesting take
2020-01-02 20:17:10 <nexgen2> systemD is present in many distros only like a school for a future Redhater
2020-01-02 20:17:13 <AlpineDiscordBri> [discord] <monado> but what ruined Debian? If it was systemd, wouldn’t CentOS and RHEL face the same issues
2020-01-02 20:17:29 <nexgen2> systemD is buggy even in Centos
2020-01-02 20:17:43 <nexgen2> it is sometime unstable even in REDHAT!
2020-01-02 20:17:54 <nexgen2> it is UNUSABLE in Debian
2020-01-02 20:17:58 <AinNero> IMHO the chat history is useless if its just the monthly systemD banter
2020-01-02 20:18:00 <AlpineDiscordBri> [discord] <monado> but Cent is just compiled RHEL
2020-01-02 20:18:11 <AinNero> like, we had this so many times already...
2020-01-02 20:18:25 <AinNero> isn't there an dedicated channel for systemD hating?
2020-01-02 20:18:27 <AlpineDiscordBri> [discord] <monado> AinNero: id like to know the other side of the coin here
2020-01-02 20:18:37 <AlpineDiscordBri> [discord] <monado> Oh, thats pretty cool
2020-01-02 20:18:49 <nexgen2> not sure why Centos has less stable systemD according to mentions on the forums
2020-01-02 20:19:01 <nexgen2> they compare exactly two things: Centos and Redhat
2020-01-02 20:19:19 <nexgen2> and Centos is behind in release like Devuan now
2020-01-02 20:19:41 <qiu3344> AinNero: Yup there is
2020-01-02 20:19:50 <AinNero> name?
2020-01-02 20:20:12 <benharri> #devuan
2020-01-02 20:20:18 <nexgen2> systemD is integrated in Debian much warse than in Redhat
2020-01-02 20:20:36 <benharri> at least it was all systemd-hate last time i stopped by there
2020-01-02 20:20:44 <AlpineDiscordBri> [discord] <monado> lmao benharri
2020-01-02 20:20:46 <nexgen2> most likely it is done intentionally
2020-01-02 20:20:57 <benharri> imo hating on systemd is unproductive
2020-01-02 20:21:06 <nexgen2> for a small package there is almost no integration, it works the same way in any distro
2020-01-02 20:21:10 <nexgen2> but not systemD
2020-01-02 20:21:11 <benharri> it's good that alternatives exist, don't use it if you don't want to
2020-01-02 20:21:30 <qiu3344> AinNero: Gimmie a sec. W will find it
2020-01-02 20:21:40 <nexgen2> I migrated to Devuan from Debian
2020-01-02 20:22:07 <nexgen2> and now enjoy some virtual guests with Alpine
2020-01-02 20:22:10 <AlpineDiscordBri> [discord] <monado> Isn’t Debian voting on systemd
2020-01-02 20:22:18 <AlpineDiscordBri> [discord] <monado> Virtual guests?
2020-01-02 20:22:34 <AinNero> how do physical guests work?
2020-01-02 20:22:40 <nexgen2> :)
2020-01-02 20:22:43 <AinNero> can there be virtual hosts?
2020-01-02 20:22:57 <AlpineDiscordBri> [discord] <monado> Oh, like vms
2020-01-02 20:22:59 <nexgen2> I mean I use bare metal Devuan and Alpine guest
2020-01-02 20:23:18 <AinNero> ?
2020-01-02 20:23:20 <AlpineDiscordBri> [discord] <monado> I used bare metal macOS and vm alpine
2020-01-02 20:23:29 <AlpineDiscordBri> [discord] <monado> Why does any of this matter
2020-01-02 20:23:30 <AinNero> what would be a non-bare metal guest?
2020-01-02 20:23:44 <nexgen2> virtual hosts can exist too btw when nested
2020-01-02 20:26:15 <nexgen2> nice reading about systemD as a backdoor API
2020-01-02 20:26:17 <nexgen2> https://translate.google.com/translate?hl=&sl=ru&tl=en&u=https%3A%2F%2Fweb.archive.org%2Fweb%2F20191231133414%2Fhttps%3A%2F%2Fwww.linux.org.ru%2Fnews%2Fdebian%2F15429763%2Fpage12%3Flastmod%3D1577796404020&sandbox=1
2020-01-02 20:26:20 <qiu3344> AinNero: #systemdsucks
2020-01-02 20:26:47 <AinNero> This channel name sounds like hate speech
2020-01-02 20:27:31 <AlpineDiscordBri> [discord] <monado> What ended up winning?
2020-01-02 20:27:44 <qiu3344> I mean it is. Most people don't have legit reasons for not using systemd
2020-01-02 20:28:25 <AlpineDiscordBri> [discord] <monado> Is Debian continuing to use sysd?
2020-01-02 20:28:32 <qiu3344> Personally I don't like the service file syntax and binary logs
2020-01-02 20:28:48 <nexgen2> binary logs are awful
2020-01-02 20:29:18 <nexgen2> they can hide some things if your host is targeted
2020-01-02 20:29:35 <nexgen2> it is Windows like
2020-01-02 20:29:50 <Cogitri> You do know that you can sign them cryptographically, right?
2020-01-02 20:29:56 <dalias> ainnero, i think you don't know what hate speech means
2020-01-02 20:30:00 <qiu3344> I mean they have thier benefits
2020-01-02 20:30:12 <Cogitri> And binary logs allow for easier rotating, embedding other files (e.g. coredumps) and so on
2020-01-02 20:30:14 <nexgen2> systemD has no way for critical server from a technical point of view
2020-01-02 20:30:27 <AinNero> dalias: its bait
2020-01-02 20:30:56 <Cogitri> One could argue that SysV which can accidentally kill the wrong process if the PID file id wrong has no place in a server :^)
2020-01-02 20:31:04 <nexgen2> it is just for third party control of the hosts
2020-01-02 20:31:33 <nexgen2> it is better to compare systemD to OpenRC
2020-01-02 20:32:08 <qiu3344> I think that the only sane option for an init system to use in production is s6 guys ;)
2020-01-02 20:33:19 <qiu3344> You can read it fairly quick and then you can know for sure that your init system is not spying on you
2020-01-02 20:33:51 <mps> I think this discussion is better for #alpine-offtopic
2020-01-02 20:35:30 <qiu3344> Mps is right. This is not the right place for a discussion like this
2020-01-02 20:35:38 <qiu3344> Let's move to #alpine-devel
2020-01-02 20:35:50 <mps> :)
2020-01-02 20:35:58 <qiu3344> (:
2020-01-02 20:36:10 <mps> #alpine-offtopic
2020-01-02 20:53:25 <AinNero> dalias: grow some skin pls
2020-01-02 20:53:44 <dalias> fuck off and stop wasting people's time spewing terf shit in #musl-offtopic
2020-01-02 20:54:25 <dalias> (which you joined just to be abusive)
2020-01-02 20:54:26 <AinNero> that i say its a slur does not imply that i share their position
2020-01-02 20:54:45 <AinNero> ... well ok
2020-01-02 20:54:54 <dalias> "terf is a slur" is transphobic speech and is not tolerated in #musl & #musl-offtopic
2020-01-02 20:54:58 <tehcloud> Keep Technical Projects Technical
2020-01-02 20:55:18 <AinNero> dalias: this is really black-white and immature
2020-01-02 20:55:20 <dalias> and i would hope the same is true here but it's not mu channel
2020-01-02 20:55:24 <_ikke_> Stop!
2020-01-02 20:55:25 <dalias> my*
2020-01-02 22:21:44 <MY-R> <AlpineDiscordBri> [discord] <monado> Is Debian continuing to use sysd?
2020-01-02 22:22:47 <MY-R> ^ that bridge configuratio is kind of unredable, cant replace bot name with something shorter without that [discord] prefix? :\
2020-01-02 22:26:41 <north1> Can't it be done like matrix?
2020-01-02 22:27:01 <MY-R> ^ ye
2020-01-02 22:29:34 <ichianko> I used the same bridge on other IRC channel and it can be configured
2020-01-02 22:29:41 <clandmeter> does matrix now use an official way to bridge to irc?
2020-01-02 22:30:10 <ichianko> Never used Matrix so I have no idea about it
2020-01-02 22:30:41 <MY-R> with matrix was nice and transparent just some sufix at the end of nick [m]
2020-01-02 22:31:09 <ichianko> I can contact the hoster and ask for some config changes
2020-01-02 22:38:36 <MY-R> ps what is the diffrence between "merged" and "close" on Alpine GL? :/
2020-01-02 22:39:02 <_ikke_> MY-R: whether gitlab think the MR is actually merged or not
2020-01-02 22:39:40 <MY-R> uhmm
2020-01-02 22:40:28 <MY-R> ahh ok, got it!
2020-01-02 22:42:26 <MY-R> kind of... but found answer here: https://stackoverflow.com/questions/43340029/closing-gitlab-merge-request/43340415
2020-01-02 22:42:42 <MY-R> whatever! :>
2020-01-02 22:47:39 <_ikke_> it has to do with our workflow
2020-01-02 22:48:06 <_ikke_> we rebase the commits and push to another repo which gets synchronized to gitlab
2020-01-02 22:48:13 <_ikke_> and gitlab does not recognize that
2020-01-02 22:48:54 <MY-R> oh ok, thanks for explanation _ikke_
2020-01-02 22:52:22 <jlo_ampard> Hi, I was inactive for a very long while here. I was at the time present with another pseudo (jch) and was active on the wiki.
2020-01-02 22:52:50 <jlo_ampard> Unfortunately I lost my password there and my email to enroll is obsolete.
2020-01-02 22:54:00 <jlo_ampard> Who can I contact to try to be able tu updates my wiki pages again as I'm back to Alpine work and I would like to update some old pages (from 2015)
2020-01-02 22:54:23 <jlo_ampard> ?
2020-01-02 22:57:53 <MY-R> jlo_ampard, I'm prety sure that somebody from AL team will read that sooner or later so be patient and stay on IRC for a while or ask again another time
2020-01-02 22:59:21 <jlo_ampard> ;)
2020-01-02 22:59:50 <MY-R> 
ACTION the bot :D

2020-01-02 23:01:15 <jlo_ampard> I'm happily running 230 AL boxes in run-from-ram tonight :)
2020-01-02 23:01:48 <MY-R> damn, wooooot!? :D
2020-01-02 23:02:21 <MY-R> bot net? ;)
2020-01-02 23:02:34 <jlo_ampard> (in production, not counting lab and test units)
2020-01-02 23:02:42 <jlo_ampard> no, in IoT project
2020-01-02 23:03:24 <MY-R> sounds nice!
2020-01-02 23:03:26 <jlo_ampard> driving green energy production across Europe
2020-01-02 23:04:03 <jlo_ampard> very nice indeed, so I would like to update a bit my wiki pages there ;)
2020-01-02 23:04:34 <MY-R> ah europe time zone, so be here after morning and ask again :)
2020-01-02 23:05:23 <jlo_ampard> I stepped in when they had 150 units running debian and starting to fail and I chenged all boxes remotely to Alpine. New ones being natively Alpine.
2020-01-02 23:05:46 <MY-R> and now all happy?
2020-01-02 23:05:52 <jlo_ampard> better
2020-01-02 23:05:59 <MY-R> great!
2020-01-02 23:06:07 <jlo_ampard> but still some room for enhancement
2020-01-02 23:06:32 <jlo_ampard> for instance porting every bit of company software to native apk
2020-01-02 23:07:05 <jlo_ampard> but at least no more disk corruption from those debian boxes
2020-01-02 23:07:35 <MY-R> it sound rly great and they have to trust you for sure :D
2020-01-02 23:07:57 <mps> jlo_ampard: which arch?
2020-01-02 23:08:21 <jlo_ampard> it was quite of a chalenge to make them jump from debian to alpine but they are happy now
2020-01-02 23:08:31 <jlo_ampard> x86_64
2020-01-02 23:09:06 <jlo_ampard> with a jav+karaf software stack
2020-01-02 23:09:09 <mps> thanks for info
2020-01-02 23:09:19 <jlo_ampard> (and some third part needing glibc)
2020-01-02 23:10:05 <jlo_ampard> I use a couple of custom made ISO that I push OTA to units
2020-01-02 23:10:15 <jlo_ampard> and I add custom repo to them
2020-01-02 23:11:03 <jlo_ampard> At manufacture time units are booting in PXE to fetch custom ISO and apkovl
2020-01-02 23:11:27 <jlo_ampard> after that evetything is OTA
2020-01-02 23:12:03 <MY-R> Alpine ftw! :>
2020-01-02 23:12:36 <jlo_ampard> We are going to go from prototype to industrial grade with nice sales
2020-01-02 23:13:55 <jlo_ampard> One the most fun was to redo the initramfs we use I think ;)
2020-01-02 23:14:26 <micro4lpha> hey guys
2020-01-02 23:14:31 <micro4lpha> I'm trying to build an alpine nginx with a custom website/config
2020-01-02 23:14:35 <micro4lpha> should I do a Dockerfile with docker pull amd64/alpine, going on EC2 ECS AMI
2020-01-02 23:17:19 <jlo_ampard> I had to do a lot of system customization and use heavily pip to add missing lib that where lacking in Alpine repo. I now plan to add all those to repo (better for run-from-ram mode than post-install scripts) if my management let me do it...
2020-01-02 23:18:16 <jlo_ampard> I asked 40 days to do that...
2020-01-02 23:19:15 <MY-R> jlo_ampard, if they let you change distribution then they dont have much choice now, right? :D
2020-01-02 23:19:28 <jlo_ampard> I guess so...
2020-01-02 23:22:58 <jlo_ampard> I need arguments to explain to management it is a bad idea to develop a custom configuration tool in Java instead of writing ACF modules. anyone ?
2020-01-02 23:23:38 <jlo_ampard> (Iwe already use 4 custom ACF I did wrote)
2020-01-02 23:27:51 <artok> micro4lpha: are you going to run it under docker?
2020-01-02 23:29:25 <artok> I've got Dockerfile, slightly outdated for that
2020-01-02 23:29:41 <jlo_ampard> And did someone tried Zulu Java with Alpine? worth it?
2020-01-02 23:30:17 <artok> if it is for x86_64, FROM alpine:3.11.2 should do when building the image
2020-01-02 23:31:32 <artok> actually I'm in middle of rewriting that, hold on for few mins and I'll do the build
2020-01-02 23:34:46 <clandmeter> micro4lpha: you can also use the official nginx image which is based on alpine
2020-01-02 23:35:04 <clandmeter> just mount the config you want enabled
2020-01-02 23:35:13 <artok> that too works
2020-01-02 23:35:35 <artok> just depends on what different plugins you want to enable
2020-01-02 23:35:59 <clandmeter> not sure which plugins the officla docker image supports
2020-01-02 23:36:05 <clandmeter> i guess the basic ones
2020-01-02 23:37:13 <bwen> Anyone got Netflix working under alpine?
2020-01-02 23:38:16 <bwen> in firefox
2020-01-02 23:39:29 <artok> oh that was fast build
2020-01-02 23:41:02 <micro4lpha> thanks "clandmeter
2020-01-02 23:41:08 <micro4lpha> thanks @clandmeter
2020-01-02 23:41:38 <artok> for the dockerfile example : https://hastebin.com/ifuhalavoj.sql
2020-01-02 23:43:42 <micro4lpha> yes artok, going to run it on docker
2020-01-02 23:43:56 <micro4lpha> it'll run an angular website
2020-01-02 23:44:01 <micro4lpha> with a proxy from nginx
2020-01-02 23:44:08 <micro4lpha> forgot to mention that
2020-01-02 23:44:35 <micro4lpha> I should probably do a multistage build, correct? Build the angular app and then send to /usr/share/nginx/html
2020-01-02 23:45:06 <artok> that dockerfile that I did is multi-stage build
2020-01-02 23:45:29 <bwen> @micro4alpha Web apps are basically static assets. Why not just host it on a cdn (AWS/GCLOUD bucket) which a domain points to?
2020-01-02 23:46:04 <artok> with those options, and with 1.17.7 nginx, image size is 22.9M, nginx:alpine official is 21.5MB
2020-01-02 23:47:13 <artok> just get the nginx:alpine for 1.17.6 nginx and have your html dir as persistent storage (named volume, k8s volumeclaim, what ever your infra is on)
2020-01-02 23:47:58 <artok> as my dockerfile uses rather simple entrypoint, but works
2020-01-02 23:48:41 <artok> it has not been designed for example to act (yet) as ssl termination and so on, as I use traefik to do that on my infrastructure
2020-01-02 23:53:31 <monado> Does anyone know who runs the twitter account
2020-01-02 23:56:28 <artok> I run my own
2020-01-03 00:03:41 <monado> Well, the alpine twitter account
2020-01-03 00:05:37 <AlpineDiscordBri> <UtopicUnicorn> Can't you just DM them?
2020-01-03 00:05:45 <micro4lpha> @bwen, we are building a cryptocurrency exchange.
2020-01-03 00:06:04 <micro4lpha> any website update we want it to be dockerized so it is inmutable and we can rollback/update with a new container
2020-01-03 00:06:07 <micro4lpha> rather than updating files
2020-01-03 00:06:09 <micro4lpha> manually
2020-01-03 00:06:38 <AlpineDiscordBri> <Synth1943> https://twitter.com/alpinelinux?lang=en
2020-01-03 00:06:45 <AlpineDiscordBri> <Synth1943> They don't seem to be accepting DMs
2020-01-03 00:06:54 <AlpineDiscordBri> <Synth1943> I suggest a public tweet in this case
2020-01-03 00:07:31 <AlpineDiscordBri> <UtopicUnicorn> Mint would never accept me, if I had to reach out I would have to create weird twitter/facebook accounts haha
2020-01-03 00:07:33 <eggbean> Does anybody know what I can do about this error?  https://ctxt.io/2/AABA1qSgEw
2020-01-03 00:07:52 <AlpineDiscordBri> <UtopicUnicorn> How long did you leave it run?
2020-01-03 00:08:00 <AlpineDiscordBri> <Synth1943> That's a warning
2020-01-03 00:08:25 <artok> eggbean: is it an error or warning?
2020-01-03 00:08:30 <AlpineDiscordBri> <Synth1943> Warning fwics
2020-01-03 00:08:39 <AlpineDiscordBri> <Synth1943> Caused by the `-Wunused-parameter` flag
2020-01-03 00:08:44 <AlpineDiscordBri> <UtopicUnicorn> Else it would not have gone past 66%
2020-01-03 00:08:46 <AlpineDiscordBri> <Synth1943> Notice the `-W` bit
2020-01-03 00:08:59 <artok> "just" warning
2020-01-03 00:09:08 <AlpineDiscordBri> <Synth1943> Yep, it's "just" a warning 😛
2020-01-03 00:09:14 <eggbean> artok: I don't know.  The image does get built thoough
2020-01-03 00:09:21 <AlpineDiscordBri> <Synth1943> So it's fine :D
2020-01-03 00:09:28 <artok> then there is some error, search more
2020-01-03 00:09:29 <AlpineDiscordBri> <UtopicUnicorn> also
2020-01-03 00:09:33 <artok> it stops at error
2020-01-03 00:09:38 <AlpineDiscordBri> <UtopicUnicorn> >  warning: unused parameter 'allow_transient' [-Wunused-parameter]
2020-01-03 00:10:10 <AlpineDiscordBri> <UtopicUnicorn> Basically this means that they called the parameter but never used it
2020-01-03 00:10:19 <AlpineDiscordBri> <Synth1943> In short it's fine
2020-01-03 00:10:25 <AlpineDiscordBri> <UtopicUnicorn> Sloppy programming
2020-01-03 00:10:25 <artok> micro4lpha: consider having your infra CI/CD thingie as:
2020-01-03 00:10:30 <AlpineDiscordBri> <Synth1943> It depends imho
2020-01-03 00:10:37 <AlpineDiscordBri> <Synth1943> You don't know the context of the code
2020-01-03 00:10:46 <eggbean> artok: I'm not sure what you mean.  I'm asking about how do I start looking for the problem
2020-01-03 00:11:00 <AlpineDiscordBri> <Synth1943> Change the weechat codebase?
2020-01-03 00:11:03 <AlpineDiscordBri> <UtopicUnicorn> True, but if you use it for a final build you would want it to use the least amount of resources
2020-01-03 00:11:15 <AlpineDiscordBri> <Synth1943> I doubt it would make a difference
2020-01-03 00:11:21 <artok> if site needs just updates, not server update, update the html dir on volume
2020-01-03 00:11:27 <AlpineDiscordBri> <Synth1943> I'm trusting the developers here
2020-01-03 00:11:31 <artok> eggbean: show more of your logs
2020-01-03 00:11:41 <AlpineDiscordBri> <UtopicUnicorn> Weechat is that thing the Chinese use right?
2020-01-03 00:11:47 <AlpineDiscordBri> <Synth1943> No.
2020-01-03 00:11:53 <AlpineDiscordBri> <Synth1943> It's an IRC client.
2020-01-03 00:11:54 <jn__> that's WeChat
2020-01-03 00:12:00 <AlpineDiscordBri> <UtopicUnicorn> OH
2020-01-03 00:12:03 <eggbean> artok: That the only error.  How much more do you want to see, as it will not be easy copying the whole thing
2020-01-03 00:12:12 <AlpineDiscordBri> <UtopicUnicorn> Haven;t used IRC in decades
2020-01-03 00:12:29 <artok> oh god we have some awful bridge to irc
2020-01-03 00:12:40 <artok> irc is the only real thing! =D
2020-01-03 00:13:08 <AlpineDiscordBri> <monado> Yeah
2020-01-03 00:13:11 <artok> eggbean: truly you have something else there?
2020-01-03 00:13:13 <monado> Yeah
2020-01-03 00:13:17 <jn__> IMHO, bridges should use NOTICE to deliver messages. looks better.
2020-01-03 00:13:27 <AlpineDiscordBri> <Synth1943> Testing
2020-01-03 00:13:31 <eggbean> artok: What do you mean?
2020-01-03 00:13:36 <AlpineDiscordBri> <Synth1943> :GWbruhShrekThonk: 😄
2020-01-03 00:13:44 <dalias> the reason for notice instead of privmsg is not "looks better". it's to prevent bot loops
2020-01-03 00:13:53 <artok> eggbean: after that last line, you have more
2020-01-03 00:13:57 <artok> log
2020-01-03 00:14:02 <eggbean> ok, one minute
2020-01-03 00:14:03 <monado> I don’t think anyone’s going to abuse the bridge as it stands
2020-01-03 00:14:03 <dalias> generally by convention automated things can auto-reply to msg but not to notice
2020-01-03 00:14:04 <AlpineDiscordBri> <Synth1943> ```
2020-01-03 00:14:05 <AlpineDiscordBri> <Synth1943> can this handle markdown?
2020-01-03 00:14:06 <AlpineDiscordBri> <Synth1943> ````
2020-01-03 00:14:10 <monado> no
2020-01-03 00:14:10 <resynth1943> And it can not
2020-01-03 00:14:15 <jn__> dalias: ah, TIL
2020-01-03 00:14:16 <CompanionCube> how would that even work
2020-01-03 00:14:22 <monado> no frickin clue
2020-01-03 00:14:24 <resynth1943> Ideally it wouldn't send three messages
2020-01-03 00:14:35 <monado> Newlines aren’t allows
2020-01-03 00:15:04 <artok> ideally there wouldn't be any bridges, if people want to use browser for irc, there is clients for them
2020-01-03 00:15:09 <resynth1943> ^
2020-01-03 00:15:21 <resynth1943> Also, some IRC clients are capable of Markdown
2020-01-03 00:15:29 <resynth1943> I was testing the Freenode web client
2020-01-03 00:15:40 <artok> if ircII can't handle, it shouldn't be there =)
2020-01-03 00:15:52 <artok> ok, I'm using irssi, but still
2020-01-03 00:16:07 <monado> it’s fine, I’ll add rules to the channel
2020-01-03 00:16:13 <monado> pastebins are preferred
2020-01-03 00:16:30 <artok> as in irc
2020-01-03 00:16:34 <dalias> well markdown is plaintext; fancy formatting of it is just a presentation matter
2020-01-03 00:16:50 <monado> Yeah
2020-01-03 00:16:57 <AlpineDiscordBri> <UtopicUnicorn> I mean reading code without syntag higlighting sucks
2020-01-03 00:16:59 <dalias> so i don't see how it's inappropriate to use
2020-01-03 00:17:02 <AlpineDiscordBri> <UtopicUnicorn> Which is what we use MD for mostly
2020-01-03 00:17:03 <dalias> everyone can read it as text
2020-01-03 00:17:13 <dalias> i don't mind if someone talks about `ls(1)`
2020-01-03 00:17:32 <artok> that is not problem, the problem is flooding
2020-01-03 00:17:50 <artok> multi-line stuff when pasting stuff would be enough
2020-01-03 00:18:05 <artok> and ones that don't belong to that conversation can ignore
2020-01-03 00:18:14 <AlpineDiscordBri> <Synth1943> I wouldn't mind using a terminal IRC client while we're talking about this
2020-01-03 00:18:32 <monado> again, id rather just pastebin
2020-01-03 00:18:32 <resynth1943> I'm actually not sure why I'm typing in both channels lol
2020-01-03 00:19:01 <resynth1943> I like the non-bloated interface of IRC, too
2020-01-03 00:19:06 <resynth1943> No reactions are nice
2020-01-03 00:19:14 <artok> that
2020-01-03 00:19:28 <monado> irc and discord have their own uses
2020-01-03 00:19:29 <resynth1943> I'm guessing that's the niche of IRC, which I can totally understand
2020-01-03 00:19:45 <resynth1943> I wonder what Discord would have been like had they chose the IRC route
2020-01-03 00:19:47 <AlpineDiscordBri> <UtopicUnicorn> Personally I like Discord
2020-01-03 00:20:45 <eggbean> artok: https://ctxt.io/2/AABA6js-Fg
2020-01-03 00:20:55 <AlpineDiscordBri> <Synth1943> Theoretically they could've just made an IRC client and added voice functionality, although that would be lacking in some areas. IRC is a well-established platform though.
2020-01-03 00:21:10 <artok> eggbean: 100%, so you're fine
2020-01-03 00:21:16 <AlpineDiscordBri> <Synth1943> resynth1943: testing
2020-01-03 00:21:20 <resynth1943> Ahh, nice
2020-01-03 00:21:26 <AlpineDiscordBri> <UtopicUnicorn> I mean they would just be another IRC client
2020-01-03 00:21:38 <monado> don’t flood their channel
2020-01-03 00:21:43 <eggbean> artok: You made me paste all that text to tell me that?
2020-01-03 00:21:50 <artok> eggbean: running it from dockerfile and it stops ?
2020-01-03 00:22:02 <resynth1943> Who's flooding wha tnow?
2020-01-03 00:22:07 <resynth1943> Oops
2020-01-03 00:22:48 <eggbean> I said it builds, but I get this warning/error and I am enquiring about how I can find out what is happening there
2020-01-03 00:22:53 <artok> eggbean: let's start again with question: something stops and throws error?
2020-01-03 00:23:30 <eggbean> I get the error, it build finished, as you can see
2020-01-03 00:23:53 <eggbean> the build finishes, I meant
2020-01-03 00:24:45 <eggbean> But the resultant container id buggy.  This is not the only error I get, but I am adding different components one at a time,
2020-01-03 00:24:47 <artok> that is no error
2020-01-03 00:24:53 <artok> that is warning
2020-01-03 00:25:21 <artok> so there must be other bugs or errors somewhere else
2020-01-03 00:25:33 <eggbean> I still want to attend to the warning and do whatever I need to do to stop the warning from appearing
2020-01-03 00:26:30 <artok> well you can add cflags to the build not to warn about that
2020-01-03 00:26:47 <artok> or tell the developers that it is annoying you
2020-01-03 01:03:18 <bwen> so no widevine support for musl? >_< (Still on my firefox Netflix problem)
2020-01-03 01:55:28 <dalias> bwen, the lib is there aiui but it doesn't seem to work. i have no idea why
2020-01-03 02:16:21 <bwen> @dalias moved on to chromium which has a non-free chromum-widewine. Just have to figure out how to compile/install, hopefully that'll work
2020-01-03 02:19:04 <dalias> ;/
2020-01-03 02:19:17 <dalias> i don't recall it working either but maybe i didnt install the nonfree package for it
2020-01-03 02:19:49 <dalias> i want to find the right way to setup an android emulator and just run the android apps for stuff like that, sandboxed
2020-01-03 02:44:41 <bwen> nope didnt work... I think I'm gonna give up :/
2020-01-03 06:26:11 <ichianko> Anyone have issue with using NetworkManager? I couldn't make it manage any interfaces including ethernet and wlan. For example using `nmcli dev set wlan0 managed yes` can't make wlan0 managed by networkmanager
2020-01-03 07:49:10 <zenny> Good morning, I am trying lxc in alpine 3.11.2 (ref. https://wiki.alpinelinux.org/wiki/LXC), but there are two issues. Even after installing lxc-templates package, it reports that alpine template is not available (lxc-create: alpine: lxccontainer.c: do_lxcapi_create: 1817 Unknown template "alpine"). Second, when I tried to use the dummy kernel module inside /etc/modprobe.d/dummy.conf, I am getting, "libkmod: kmod_config_parse: /etc/mo
2020-01-03 10:08:17 <zenny> I am stuck with the connection to internet from lxc container after following this (https://wiki.alpinelinux.org/wiki/LXC#Creating_a_LXC_container_without_modifying_your_network_interfaces)! Any clue?
2020-01-03 14:16:19 <DCiple> Morning everyone! Are there any ZFS folks out there? I'm getting between 1.4 - 1.7 MB writes every 4  - 6 seconds in my array via zpool iostat. This has been going on for about a week and zpool status shows everything is normal. My first thought was something like ransomware but I can't find any date modified updates in any of my file's metadata. I'm looking to try and trackdown which program or process is making these writes just
2020-01-03 14:16:19 <DCiple> to validate that everything is normal. Is there a way to check which process is writing data to a harddrive? Google has not been to terribly helpful with my case.
2020-01-03 14:18:38 <programmerq> sysdig open source can help figure that out.
2020-01-03 14:19:07 <DCiple> It literally just stopped lol
2020-01-03 14:19:32 <DCiple> I think it just needs me to get on freenode and ask a question.
2020-01-03 14:20:29 <programmerq> hmmm, I remembered wrong. I thought there was a package for sysdig for alpine. guess not.
2020-01-03 14:25:04 <programmerq> I may be biased since I just started working for sysdig the company, but the open source tool is pretty powerful and can help figure out exactly what you're talking about. It does require a kernel module. Maybe I should see if I can get an apkbuild working for it.
2020-01-03 14:54:25 <admicos> hey all! what would be the best way to automatically mount (and keep mounted) an unreliable NFS4 share (can randomly go offline)?
2020-01-03 15:08:32 <AinNero> the NFS server goes offline?
2020-01-03 15:08:41 <admicos> yes
2020-01-03 15:28:35 <micro4lpha> @artok, for the NGINX Dockerfile you sent yesterday what are you using as your entrypoint.sh?
2020-01-03 15:38:49 <DCiple> @admicos There's a logic problem there in your question. Is there a specific symptom resulting from the share leaving that you are trying to overcome?
2020-01-03 15:41:16 <admicos> DCiple: Well, the share itself is hosted on my home network, which is VPN'd over to an Alpine server (NOT on my home net). I'm just looking for something so I won't have to remount the NFS share if the connection cuts for some reason
2020-01-03 15:46:08 <DCiple> admicos: so you have your home network, and a remote alpine server with NFS exposed. Where is the VPN terminating on both ends?
2020-01-03 15:46:35 <admicos> terminating?
2020-01-03 15:47:10 <DCiple> are you vpn'nd between two firewalls - like a site to site vpn? is it just a straight client to client vpn?
2020-01-03 15:48:09 <admicos> it should be a client to client vpn
2020-01-03 15:48:15 <admicos> though the vpn connection isn't an issue, that works well
2020-01-03 15:48:28 <DCiple> It always stays active, never times out?
2020-01-03 15:48:54 <admicos> the vpn itself is solid, but my home connection can get flaky every now and then
2020-01-03 15:49:33 <DCiple> so the flakeyness is what is causing the disconnect, does the vpn auto-recover from the flakeyness?
2020-01-03 15:50:15 <admicos> it should
2020-01-03 15:51:11 <DCiple> well the first idea that comes to my head is script a re-mount upon detection of the circuit coming back online.
2020-01-03 15:52:33 <DCiple> You're question seems to boil down to: "I have an unreliable connection, how do I get NFS to stop behaving like it has an unreliable connection"? So I think you want to mitigate the unreliable connection somehow
2020-01-03 15:53:15 <admicos> well, not exactly. i'm fine with the server not being able to access the share if it can't. i'm just looking for a way to auto mount it whenever it can
2020-01-03 15:53:51 <DCiple> I assume the mount is listed in /etc/fstab?
2020-01-03 15:53:56 <admicos> yes
2020-01-03 15:54:24 <DCiple> write a program to mount -a upon detection criteria
2020-01-03 15:56:11 <DCiple> the problem is if the line is flakey and you try the mount -a, you'll have to wait for a timeout which can be lengthy.
2020-01-03 16:06:44 <bwen> anyone got Netflix/Primevideo to work in Firefox/Chromium on Alpine? :\
2020-01-03 16:15:32 <bl4ckb0ne> is there a way to specify an optional dependency in an apkbuild file ?
2020-01-03 16:17:34 <_ikke_> bl4ckb0ne: no
2020-01-03 16:18:52 <bl4ckb0ne> so i should just skip it
2020-01-03 16:40:56 <qiu3344> bwen: The easiest way to get Netflix running on Alpine is using the google-chrome flatpak
2020-01-03 16:41:42 <bwen> @qiu3344 aight i'll look into it, thx
2020-01-03 16:51:33 <qiu3344> It works fine on my laptop so it should also for you
2020-01-03 17:32:14 <bwen> @qiu3344 which remote did you add? it does not find "google-chrome" ...
2020-01-03 17:33:05 <danieli> it won't be in alpine repos because it is proprietary
2020-01-03 17:33:14 <danieli> watching netflix would require widevine
2020-01-03 17:34:12 <bwen> @danieli chromium-widevine is in the non-free, but does not work.  I was talking about flatpak repos, sorry for the confusion
2020-01-03 17:34:51 <qiu3344> google chrome is in https://ostree.endlessm.com/ostree/eos-apps
2020-01-03 17:35:13 <qiu3344> flatpak remote-add --user eos-apps --no-gpg-verify https://ostree.endlessm.com/ostree/eos-apps
2020-01-03 17:35:15 <qiu3344> flatpak install --user eos-apps com.google.Chrome
2020-01-03 17:36:36 <qiu3344> Sorry, it looks like that one isn't working anymore. Try flatpak remote-add --user --no-gpg-verify chrome http://blog.kukuh.syafaat.id/google-chrome-flatpak-repo-testing/repo
2020-01-03 17:37:00 <qiu3344> and then install the com.google.Chrome package
2020-01-03 17:37:03 <bwen> where do you find those remotes? :P
2020-01-03 17:37:29 <qiu3344> https://gist.github.com/cho2/a9c20556e0585943624208c3ca1a5af7 ;)
2020-01-03 17:43:34 <ajar> hi all. i have not been able to find any instructions on getting plymouth (splash screen) to work with Alpine. I've look on the wiki and the search the mail lists. Any thoughts? I've tried various things i've found for arch linux, etc.KMS and DRM appear to be working.
2020-01-03 17:44:22 <ajar> and I've tried on RPI4 and x86_64 builds so obviously something i'm just not understanding
2020-01-03 17:59:08 <ncopa> artok: did you have issues with alpine 3.11 aarch64 on rpi4?
2020-01-03 18:01:33 <ajar> and for the record i have installed the plymouth packages, selected a theme, tried mkinitfs (does it not support plymouth?)
2020-01-03 19:01:08 <eggbean> Does anybody know why I am getting this error when compiling for a docker image?  http://sprunge.us/YFs8cK
2020-01-03 19:01:49 <_ikke_> seems like you are missing dependencies?
2020-01-03 19:01:58 <Cogitri> Well, I suppose you're either missing the -dev packages for those or the build system of the package you're building hasn't set the right library path
2020-01-03 19:02:15 <eggbean> What can I do?
2020-01-03 19:03:32 <eggbean> I already have added the php-dev and php-embed apks
2020-01-03 19:03:40 <Cogitri> Install the packages providing those libs
2020-01-03 19:03:45 <Cogitri> So e.g. libxml2-dev
2020-01-03 19:03:46 <eggbean> Not sure what else I need?
2020-01-03 19:04:01 <eggbean> Oh, okay.  Thanks, I'll try that.
2020-01-03 19:04:07 <_ikke_> eggbean: protip: check the php7 aport
2020-01-03 19:04:15 <Cogitri> You can search for the lib with `apk search so:lib<what it complains about>`
2020-01-03 19:04:35 <_ikke_> eggbean: https://git.alpinelinux.org/aports/tree/community/php7/APKBUILD
2020-01-03 19:04:41 <eggbean> _ikke_: aport?  What's that?
2020-01-03 19:04:43 <_ikke_> ^^
2020-01-03 19:05:02 <_ikke_> eggbean: it's the repository that contains all the build recipes that alpine uses to build its packages
2020-01-03 19:05:22 <eggbean> _ikke_: Thanks, I'll read up on that
2020-01-03 19:05:31 <_ikke_> oh, you're building weechat, not php
2020-01-03 19:05:40 <_ikke_> any reason your are building it yourself?
2020-01-03 19:05:43 <eggbean> Cogitri: What did you search for?
2020-01-03 19:06:04 <eggbean> _ikke_: For the latest release from github
2020-01-03 19:06:24 <Cogitri> eggbean: You can search what package provides a lib with that command
2020-01-03 19:06:30 <eggbean> Basically, I'm learning for to make images and using this as a learning project
2020-01-03 19:06:35 <_ikke_> this is the weechat APKBUILD file: https://git.alpinelinux.org/aports/tree/community/weechat/APKBUILD
2020-01-03 19:06:53 <eggbean> Ah, thanks
2020-01-03 19:07:15 <_ikke_> depends_dev contains all the libraries you need
2020-01-03 19:07:31 <eggbean> I see
2020-01-03 19:07:48 <eggbean> Okay thanks guys, I will go through this
2020-01-03 19:10:07 <micro4lpha> test
2020-01-03 19:22:09 <Cogitri> Pong I guess
2020-01-03 21:05:26 <porcellis> alpine is great <3
2020-01-03 21:16:08 <blas-fange> hi
2020-01-03 21:16:18 <blas-fange> happy new year
2020-01-03 21:21:58 <AlpineDiscordBri> <monado> happy new year
2020-01-03 23:12:10 <artok> ncopa: yah, rpi4 doesn't boot if kernel and initramfs aren't on sd root
2020-01-03 23:12:27 <artok> I'll doublecheck tomorrow if I've got latest firmware
2020-01-04 00:19:32 <ajar> artok - if it helps at all, I've followed the directions here: https://wiki.alpinelinux.org/wiki/Classic_install_or_sys_mode_on_Raspberry_Pi and it works great for my rpi4
2020-01-04 00:19:42 <ajar> i've done it a couple of times
2020-01-04 00:19:45 <ajar> consistent results
2020-01-04 00:20:15 <mps> ajar: thanks for report
2020-01-04 00:20:24 <ajar> the diskless image works great too of course, but i assume you're doing something like the classic/sys install
2020-01-04 00:21:19 <mps> he told he even cannot boot without moving files
2020-01-04 00:22:02 <AinNero> > if kernel and initramfs aren't on sd root
2020-01-04 00:22:11 <AinNero> where else could you put them?
2020-01-04 00:23:33 <AinNero> mkinitfs didn't have plymouth support last time i checked
2020-01-04 00:23:44 <mps> AinNero: kernel and initramfs are under /boot
2020-01-04 00:24:10 <mps> dtbs are under /
2020-01-04 00:24:20 <mps> for rpis only
2020-01-04 00:25:41 <AinNero> ah i get it
2020-01-04 00:26:00 <ajar> yes forgive me, i'm not as expert as a lot of you. the RPI class install puts boot (and the kernel and intitramfs) in a boot partition and the root gets mounted in a secondary partition
2020-01-04 00:26:11 <AinNero> last time we discussed it we were against adding plymouth to mkinitfs
2020-01-04 00:27:27 <mps> i'm still against plymouth in initramfs by default
2020-01-04 00:27:51 <mps> useless bloat, imo
2020-01-04 00:27:56 <ajar> AinNero: Thanks. Was hoping since there were Plymouth packages available I could.  Is there a work around? or does it mean code changes
2020-01-04 00:28:05 <AinNero> just the logic to conditionally add it is bloat of similar size
2020-01-04 00:28:35 <AinNero> either in the initramfs-init script (which is already a complex mess) or in mkinitfs itself
2020-01-04 00:29:57 <mps> plymouth-0.9.4-r0 installed size: 835584
2020-01-04 00:31:22 <mps> good night all :)
2020-01-04 00:33:38 <rapha> hi
2020-01-04 00:33:42 <rapha> i'm trying to install via PXE
2020-01-04 00:34:11 <rapha> but https://wiki.alpinelinux.org/wiki/PXE_boot mentions a file called gpxe.kpxe which i can't find anywhere ... where do you get that from?
2020-01-04 00:34:35 <AinNero> is it not on boot.alpinelinux.org ?
2020-01-04 00:34:50 <rapha> AinNero:
2020-01-04 00:34:58 <rapha> no, not that i could find
2020-01-04 00:35:35 <rapha> also not sure if that's the right thing ... trying to boot using a Dell server's PXE bios, since it doesn't boot from USB
2020-01-04 00:36:31 <rapha> there's an interesting arch wiki page mentioning how to set up dnsmasq: https://wiki.archlinux.org/index.php/PXE#Installation ... might that be helpful?
2020-01-04 00:37:08 <rapha> (https://wiki.alpinelinux.org/wiki/PXE_boot doesn't even mention boot.alpine...)
2020-01-04 00:38:17 <AinNero> not if you wanna do gpxe
2020-01-04 00:38:45 <AinNero> but if you dont know exactly what you need, you might be using the boot.ipxe file from boot.a.o as well
2020-01-04 00:39:08 <AinNero> or setup an own pxelinux boot config as described in the link you posted
2020-01-04 00:41:04 <rapha> AinNero: so i would download boot.ipxe and tell dnsmasq to instruct the machine to download that and then b.a.o will take it from there?
2020-01-04 00:41:53 <AinNero> tbh i dont know how to deliver ipxe files via dnsmasq
2020-01-04 00:42:09 <AinNero> always did it via the pxelinux.0 method
2020-01-04 00:54:28 <rapha> the problem seems to be much before that
2020-01-04 00:54:45 <rapha> dnsmasq running in debug mode keeps saying "no address range available for DHCP request" ... it seems that Dell server expects to get an IP in a specific range
2020-01-04 00:56:40 <dalias> ?
2020-01-04 00:57:24 <dalias> that doesn't sound like what that error means
2020-01-04 00:58:12 <danieli> that means the DHCP server is getting a request for an address on a network where it is not configured with a pool
2020-01-04 00:58:31 <rapha> hmm
2020-01-04 00:58:33 <danieli> iirc the full output message is "no address range available for DHCP request via <interface>
2020-01-04 00:58:40 <danieli> so you've probably misconfigured it
2020-01-04 00:59:06 <rapha> but i gave 192.168.0.1 to that interface (ip add add ...) and the DHCP range configued for dnsmasq is 192.168.0.2,192.168.254.254
2020-01-04 00:59:14 <rapha> doesn't get any wider than that, does it?
2020-01-04 01:00:00 <rapha> okay, also tried with .2 until .10 ... same message still
2020-01-04 01:00:08 <danieli> the largest size for 192.168.x.x is 192.168.0.0/16
2020-01-04 01:00:12 <danieli> i.e. 192.168.0.0-192.168.255.255
2020-01-04 01:01:36 <danieli> anyway, dhcp-range would have to be "192.168.0.2,192.168.255.254,255.255.0.0,12h" or something like that
2020-01-04 01:01:40 <rapha> danieli: https://gist.github.com/sixtyfive/78d75a3810cadbccc622003c28f93c89
2020-01-04 01:03:57 <danieli> are you running multiple instances of dnsmasq?
2020-01-04 01:07:05 <rapha> danieli: not according to ps ax
2020-01-04 01:07:30 <danieli> then the only mistake i can see in that config is 'bind-interfaces' and the lack of a subnet mask
2020-01-04 01:07:50 <danieli> i might just be mistaken and missing something, but other than that, it looks right to me
2020-01-04 01:08:04 <rapha> both of these are from the arch wiki page on PXE ... i'll change them
2020-01-04 01:09:21 <danieli> then again you are using a "non-standard" subnet size of /16 rather than /24
2020-01-04 01:09:28 <danieli> i don't know if dnsmasq has a default there
2020-01-04 01:10:14 <danieli> you did say you tried "dhcp-range=192.168.0.2,192.168.0.10,24h" though, which should have worked
2020-01-04 01:10:51 <rapha> aha!
2020-01-04 01:10:57 <rapha> okay, with your fixes it now worked
2020-01-04 01:11:33 <rapha> also, dnsmasq-tftp: sent /srv/dnsmasq/boot.ipxe to 192.168.0.107
2020-01-04 01:11:47 <rapha> but the Dell doesn't seem to know what to do with it
2020-01-04 01:11:59 <rapha> just sitting there showing the IP config it got
2020-01-04 01:15:40 <rapha> (would boot.ipxe be showing any text output if it were being loaded?)
2020-01-04 01:19:34 <rapha> tried to let it send ipxe.lkrn instead. also no reaction.
2020-01-04 01:19:53 <rapha> wondering if these Dell servers are meant to only PXE boot some kind of Dell images
2020-01-04 01:20:15 <rapha> shame, was looking forward to playing with a rack server for the first time :/
2020-01-04 01:22:07 <danieli> i recommend you ask ##networking, this isn't really alpine-specific at all
2020-01-04 01:22:56 <rapha> oh okay
2020-01-04 01:23:03 <rapha> will do, thank you danieli
2020-01-04 01:23:05 <danieli> anyway, there isn't much to rack servers vs tower servers other than the form factor and some implementation details
2020-01-04 01:23:12 <rapha> hmm
2020-01-04 01:23:12 <danieli> they'll functionally be pretty much equivalent
2020-01-04 01:23:19 <danieli> except 1U/2U servers are super loud
2020-01-04 01:23:23 <rapha> i find the hardware to be really different
2020-01-04 01:23:35 <rapha> and F*** yes it's LOUD
2020-01-04 01:23:40 <rapha> (it's a 1U)
2020-01-04 01:23:43 <danieli> :)
2020-01-04 01:23:57 <rapha> hah! got it to get to a grub rescue prompt now :)
2020-01-04 01:24:16 <rapha> but "NBP is too big to fit in free base memory" :P
2020-01-04 01:24:45 <rapha> it seems boot.a.o is meant to be accessed from the USB drive image provided there; not from a BIOS-based client
2020-01-04 01:29:27 <rapha> danieli: offering ipxe.pxe seems to be the right file to let dnsmasq offer to the client. that then tells me the address of boot.a.o/boot.ipxe (interestinly, https://, not http://?!) ... then Error 0x23e11623b. i'm assuming that has to do with no internet being available atm.
2020-01-04 01:40:38 <rapha> cool, "connection timed out", that's better already!°
2020-01-04 02:08:30 <rapha> danieli: got towards it doenloading vmlinuz-vanilla which is taking a long time, but iseems to be going on :)
2020-01-04 02:09:06 <rapha> (bit weird because loads of "transmit timed out" in between percentages)
2020-01-04 02:18:10 <rapha> oh :-(
2020-01-04 02:18:13 <rapha> "PowerEdge 2950  BNX2 driver fails to download from HTTP. Errors out with "Transmit timed out" "
2020-01-04 06:24:53 <monado> hi
2020-01-04 06:38:43 <danieli> hello
2020-01-04 12:09:49 <Greyfn> Anybody have any idea what package I need to work with network namespaces in Alpine?
2020-01-04 12:20:09 <Greyfn> Ah, I'm stupid - for future reference it was the util-linux package I needed.
2020-01-04 12:43:16 <okias[m]> Cogitri hey, this will be probably easier. I noted on Gentoo bugzilla it should work for them with latest patchset, but not found patch which is neededed yet: https://dev.gentoo.org/~anarchy/mozilla/patchsets/firefox-71.0-patches-04.tar.xz
2020-01-04 12:45:10 <okias[m]> sadly in the list I see only that patch I mentioned before, which seems to affect something
2020-01-04 13:00:49 <mps> okias[m]: do you refer for some bug report or merge request?
2020-01-04 13:03:51 <okias[m]> mps ?
2020-01-04 13:04:02 <okias[m]> sorry, I read "prefer" instead refer
2020-01-04 13:04:27 <okias[m]> I refer to firefox not building on ARMHF
2020-01-04 13:05:00 <mps> ah, it is built few days ago, I took patch from void linux
2020-01-04 13:05:19 <mps> oh, no
2020-01-04 13:05:24 <mps> armhf?
2020-01-04 13:05:51 <mps> didn't tried because don't have enoguh good armhf builder
2020-01-04 13:06:32 <mps> my armhf lxc can build normal sized pkgs, not such big as firefox
2020-01-04 13:07:04 <okias[m]> oh, well https://git.alpinelinux.org/aports/commit/testing/firefox/APKBUILD?id=0b6cd5c4ee48b8d8ce1f8c039afedddb834627da by it's broken
2020-01-04 13:07:10 <mps> but I built it on armv7, so probably same can work on armhf
2020-01-04 13:08:08 <okias[m]> sadly mozilla didn't give a much love to linux, nor arch like arm :(
2020-01-04 13:08:52 <mps> I didn't tried armhf after successful build on armv7, someone with good (read BIG) armhf builder could try
2020-01-04 13:09:24 <okias[m]> if I had something better than old BananaPi, I would try :D
2020-01-04 13:09:32 <mps> :)
2020-01-04 13:09:50 <okias[m]> also my tablet with pmOS doesn't seems to be best place with 1G ram and 32G nand :D
2020-01-04 13:11:12 <mps> last time building some of firefox in 6x series took about 14 hours on my 8core cortex with 4GB ram, though on micro sd filesystem
2020-01-04 13:14:35 <okias[m]> 4G is pretty low, you need some big ARM :D
2020-01-04 13:17:11 <mps> remember time when run mosaic (firefox ancestor) on 8MB (mega, not error) RAM
2020-01-04 13:18:18 <mps> or it was 4MB, time pass and I forgot exact amount
2020-01-04 13:24:28 <AinNero> i dont wanna be the "the old times were much better"-guy
2020-01-04 13:24:46 <AinNero> but IMHO the resource usage of modern software is largely inacceptable
2020-01-04 13:26:14 <AinNero> base systems of several gigabytes? idle ram usage 500 MB ?
2020-01-04 13:26:33 <AinNero> what the hell do they even sue the 500 MB for
2020-01-04 13:26:43 <AinNero> s/sue/use/
2020-01-04 13:27:06 <mps> paid by memory/cpu manufacturers
2020-01-04 13:29:46 <AinNero> when computing got cheaper, we made software more demanding
2020-01-04 13:45:37 <Cogitri> I quite enjoy the new features that have been added, so imho it's worth it
2020-01-04 13:47:34 <Cogitri> okias[m]: Well, it's not that FF doesn't care about armhf, it's just that it can't support webasm on it without loads of work (and then with terrible performance)
2020-01-04 13:47:57 <Cogitri> And about nobody uses it anymore so the effort isn't worth it
2020-01-04 13:48:57 <okias[m]> well, don't forget about many phones and tables on armhf
2020-01-04 13:49:55 <Cogitri> If people using armhf are interested in having FF on it I don't mind them adding support for it, but to be honest I don't feel like investing the effort to fix a legacy arch that's being phased out
2020-01-04 13:50:37 <Cogitri> And I don't want to keep back security upgrades on FF for the majority of the users to fix armhf
2020-01-04 14:42:51 <mps> Cogitri: re features in new hardware: I had fully comfortable machines with 33-66MHz and 4-8MB ram and worked quite fine for software and hardware development
2020-01-04 14:43:34 <mps> now I have 2GHZ (xCores) with 4-8GB ram
2020-01-04 14:45:03 <mps> granted, they are better and maybe faster but not even 10 times with CPUs faster 100 times and RAM amount 1000 times bigger
2020-01-04 14:45:54 <mps> my personal perception I got 2 times better boxes, nothing more
2020-01-04 14:53:38 <Cogitri> Yup, I guess there are diminishing returns
2020-01-04 14:54:22 <Cogitri> But I still wouldn't want to use slower machines w/ faster software that has less features usually (although you do need a balance between those two)
2020-01-04 14:56:55 <mps> at these (old :) ) times I didn't had git but rcs, primitive syntax highlight admit, not plugins for vi etc, but things worked and job was done
2020-01-04 15:01:37 <olegb> I have just installed Alpine on an old 686 laptop - how do I enable screen blanking after an amount of time (inactive)?
2020-01-04 15:02:05 <mps> olegb: on console or X
2020-01-04 15:02:09 <olegb> console
2020-01-04 15:03:09 <mps> probably in /etc/conf.d/kbd
2020-01-04 15:04:13 <olegb> mps: thanks i'll have a look =)
2020-01-04 15:05:05 <mps> kbd-vlock apk, I think
2020-01-04 15:09:45 <olegb> mps: thank - alpine is new to me, but I am quite impressed!
2020-01-04 15:11:27 <mps> olegb: setterm -blank nn
2020-01-04 15:15:08 <olegb> thanks
2020-01-04 15:15:31 <mps> np
2020-01-04 15:42:18 <MY-R> olegb, add to kernel command: consoleblank=600
2020-01-04 15:42:24 <MY-R> it is in seconds
2020-01-04 15:54:07 <olegb> cool
2020-01-04 17:14:54 <olegb> thanks MY-R and mps ended up adding consoleblank=600 to /boot/extlinux.conf
2020-01-04 17:34:38 <mps> olegb: yes, this is good, but you can always change this timeout for particular VT with setterm
2020-01-04 17:44:40 <QableAnon> I just upgraded to Edge but a message came back saying "Internal error: Unreleased memory pool(s) dtree [0x7ff33d24cf00]
2020-01-04 17:45:06 <QableAnon> Is this something to be concerned about?
2020-01-04 17:45:55 <Cogitri> Well, where did that message pop up?
2020-01-04 17:45:59 <Cogitri> Did apk throw that?
2020-01-04 17:49:57 <QableAnon> yes after using apk upgrade --update-cache --available
2020-01-04 18:03:49 <olegb> mps: great - thanks again!
2020-01-04 18:09:31 <mps> olegb: you are welcome
2020-01-04 21:51:47 <cosha> hello all, i've used the alpine-chroot-install to create a raspberry pi image, but having a few issues. specifically, for some reason, the image doesn't mount in rw on boot, but also, the /boot partition doesn't seem to mount either (even though i get a tty shell)
2020-01-04 21:52:12 <cosha> i can post the script that i run to create the image if you need
2020-01-04 21:59:58 <Cogitri> Well, that sounds like you've forgotten to fill your /etc/fstab to me
2020-01-04 22:00:47 <mps> cosha: you want to run rasperry pi in chroot?
2020-01-04 22:08:30 <cosha> mps: no, i can run it in chroot, and i set it all up. i package it up as an img, and burn it to an sdcard that i then boot up on my raspberry pi.. there it won't boot with the filesystem in write mode
2020-01-04 22:10:03 <cosha> Cogitri: looking at fstab looks okay to me, but i don't know enough about it.. maybe a second pair of would help: entry 1: /dev/mmcblk0p1   /boot   vfat   defaults    0    2
2020-01-04 22:10:06 <cosha> entry 2: /dev/mmcblk0p2   /   ext4   defaults,noatime   0   1
2020-01-04 22:10:49 <mps> add 'rw' for root
2020-01-04 22:11:05 <cosha> okay, let me try
2020-01-04 22:11:05 <mps> example: /dev/mmcblk0p2   /   ext4   rw,defaults,noatime   0   1
2020-01-04 22:17:30 <cosha> mps: yep, so i edited fstab, and it still mounted as ro, doing a manual remount,rw works though. in addition, /boot is still not showing up as mounted
2020-01-04 22:19:59 <mps> hm, my /boot looks like this 'LABEL=r1.bfs /boot    ext4    auto'
2020-01-04 22:20:22 <Cogitri> Hm, is there a `fstab` service for openrc you have to start?
2020-01-04 22:20:29 <Cogitri> Not on my laptop rn so can't check
2020-01-04 22:21:07 <mps> I use u-boot and specify 'root=/dev/mmcblk0p2 rw rootwait' in its config
2020-01-04 22:35:07 <cosha> Cogitri: humm checking, according gentoo wiki there's several services that read fstab.. but also, there's a service running called "mount-ro".. suspect
2020-01-05 00:28:39 <Naib> evening, is there anyway of removing mariadb-test (600meg) when mariadb is installed?
2020-01-05 00:30:06 <Naib> well 370meg...
2020-01-05 00:58:24 <Greyfn> can anybody enlighten me why as to why floppy support is include in alpine? Specifically lbu
2020-01-05 00:58:49 <Greyfn> just curious.
2020-01-05 01:10:13 <AinNero> i think its what the default entries in /etc/fstab are
2020-01-05 01:10:31 <AinNero> minus cdrom bc its read-only
2020-01-05 02:38:28 <bl4ckb0ne> is there a way to specify the name of the downloaded source in an apkbuild file?
2020-01-05 02:38:37 <bl4ckb0ne> https://github.com/sunaku/tamzen-font/archive/Tamzen-1.11.4.tar.gz the file downloaded doesnt have the same name
2020-01-05 02:38:45 <bl4ckb0ne> and that messes up the sha512sum
2020-01-05 03:20:45 <bl4ckb0ne> nevermind i got it, needed `abuild checksum` before
2020-01-05 06:41:43 <AlpineDiscordBri> <PibePlayer> Hello!
2020-01-05 07:02:24 <Cogitri> cosha: You call mount-ro during shutdown usually so you don't shutdown while something still writes to /
2020-01-05 10:48:19 <mcm_> it seems that the default `lftp` package is built w/o openssl support. is there any intention to change that?
2020-01-05 10:49:30 <mcm_> also how can i create a feature request for this package?
2020-01-05 10:55:55 <Cogitri> Open an issue on https://gitlab.alpinelinux.org/alpine/aports/issues
2020-01-05 11:02:22 <mcm_> Cogitri, thanks :-)
2020-01-05 11:03:14 <idnc_sk> hi
2020-01-05 11:03:20 <idnc_sk> quick one
2020-01-05 11:03:28 <idnc_sk> ID 2357:010c TP-Link TL-WN722N v2
2020-01-05 11:03:39 <idnc_sk> usbcore: registered new interface driver r8188eu
2020-01-05 11:03:48 <idnc_sk> do you know which firmware package this is in alpine?
2020-01-05 11:03:59 <idnc_sk> its a wireless usb dongle
2020-01-05 11:13:08 <mcm_> how can i find out the running alpine version? since the system is missing /etc/lsb-release ?
2020-01-05 11:13:47 <bwen> @mcm_ /etc/apk/respositories ?
2020-01-05 11:14:41 <bwen> or /etc/os-release
2020-01-05 11:15:36 <mcm_> bwen, thank you :-)
2020-01-05 14:04:56 <Naib> is there any reason why the mariadb package (inc deps) has increase from ~100meg to ~600meg?
2020-01-05 14:26:38 <eggbean> Has anybody got any idea what is going here?  https://ctxt.io/2/AABAjveqFg
2020-01-05 14:27:21 <_ikke_> What specifically are you referring to?
2020-01-05 14:29:03 <eggbean> The red lines?
2020-01-05 14:30:10 <eggbean> I had added a package that stopped that happening before, so I have added all sorts of previous packages, hoping to eliminate one by one.
2020-01-05 14:30:28 <eggbean> I am using that aport uild file as a reference, which you pointed out to ne
2020-01-05 14:30:32 <eggbean> me
2020-01-05 14:30:47 <_ikke_> That just seems extra information about the translation process
2020-01-05 14:31:37 <eggbean> It's not a warning?
2020-01-05 14:32:04 <eggbean> It is something that I had happen before, but I seemed to have done something which stopped it from happening at some stage
2020-01-05 14:32:32 <_ikke_> Maybe it's warning about the untranslated messages
2020-01-05 14:32:50 <_ikke_> eggbean: maybe you see it because you installed gettext
2020-01-05 14:32:54 <_ikke_> I think it's innocent
2020-01-05 14:33:35 <eggbean> gettext is something I had added at some point.  It's part if guile or something?
2020-01-05 14:33:44 <_ikke_> it's a separate project
2020-01-05 14:33:48 <eggbean> Thanks, I'll play around with that
2020-01-05 22:45:07 <Greyfn> I'm getting a curious error. Same as mentioned here: https://github.com/gliderlabs/docker-alpine/issues/334
2020-01-05 22:45:23 <Greyfn> thing is; lots of talk there about docker containers or DNS issues
2020-01-05 22:45:59 <Greyfn> but I'm building from scratch using the mkimg scripts, using all of my own infrastructure - my own DNS, my own package mirrors on my local network
2020-01-05 22:48:29 <Greyfn> the package I'm specifically trying to bring in is always openssh - no other packages fail
2020-01-05 22:48:59 <Greyfn> I'm not sure if thats relevant - but all other packages install just fine, and right in the middle it's always openssh failing
2020-01-05 22:49:14 <Greyfn> and it's repeatable too
2020-01-05 22:50:50 <Greyfn> even been so far as to hardcode the mirror hostname/ip in /etc/hosts and I still get the same issue
2020-01-05 22:52:34 <Greyfn> (and in case it wasn't clear, docker isn't in the equation at all)
2020-01-05 23:16:44 <Greyfn> I reckon it's an issue in libfetch.
2020-01-06 00:28:26 <AinNero> mps: the vt520 terminfo situation had a weird plot twist
2020-01-06 00:28:44 <AinNero> the terminfo upstream uses the definitions as pre-written in the datasheet
2020-01-06 00:29:38 <AinNero> the difference is, that im using a PC keyboard, and the terminfo definition was made for the original keyboard
2020-01-06 05:18:54 <winny> hi - how do i force reinstallation of a package, or alternately, force removal of a package with rdepends
2020-01-06 05:19:19 <north1> apk fix -r
2020-01-06 05:20:29 <bleb> to install alpine on a usb stick with persistent storage, i made a fat32 partition in /dev/sdc1, then did setup-bootable alpine-extended-3.11.2-x86_64.iso /dev/sdc1
2020-01-06 05:21:05 <bleb> but to do this i had to install alpine-extended-3.11.2-x86_64.iso twice: once to make the installation media, and again while booted to the installation media in order to be able to run the setup-bootable command
2020-01-06 05:21:14 <bleb> would there have been a less dumb way to do that
2020-01-06 05:21:31 <winny> north1: thanks - I like how it's a separate command to discourage silly behavior
2020-01-06 12:33:19 <mps> AinNero: no one of pcvt* can't help you
2020-01-06 12:47:02 <AinNero> there are vt*pc variants, but the vt520 doesnt have that variant
2020-01-06 13:23:08 <mcm_> where can i find the ssh fingerprint for gitlab.alpinelinux.org ?
2020-01-06 13:26:08 <mps> AinNero: afaik, only option is to create your own terminfo, though not sure is it worth hassle
2020-01-06 13:27:02 <AinNero> already shipping one with my dotfiles https://github.com/nero/etc/blob/master/terminfo/vt520.info
2020-01-06 13:27:22 <AinNero> one can export the data with infocmp, and then edit and re-compile it with tic
2020-01-06 13:29:02 <AinNero> some thing that bothers me:
2020-01-06 13:29:19 <AinNero> on alpine, the terminfo for urxvt-unicode ships separately from the rest of the terminfo data
2020-01-06 13:30:03 <mps> yes, and I don't know reason for that
2020-01-06 13:30:07 <AinNero> me neither
2020-01-06 13:30:53 <mps> maybe maintainer personal preference
2020-01-06 13:32:18 <mps> for my personal use I changed st terminfo somewhat but don't think to push it anywhere, it is for my personal tweaks
2020-01-06 13:34:42 <AinNero> do you use a customized st?
2020-01-06 13:35:28 <mps> yes, with some patches
2020-01-06 13:37:19 <mps> in my personal repo, scrollback, xresources, anysize, hidecursor, delkey and some tweaks for hhk like keyboard
2020-01-06 13:38:34 <north1> HHK?
2020-01-06 13:39:39 <mps> hapy hacker keyboard
2020-01-06 13:40:22 <mps> no PG{UP,DOWN}, home, end, insert, del and similar keys
2020-01-06 13:40:53 <mps> similar to those on chromebooks and some macs
2020-01-06 13:42:19 <AinNero> ah, i remember why i didnt stick to st
2020-01-06 13:42:26 <AinNero> mediocre unicode support
2020-01-06 13:42:53 <mps> hah, really ?
2020-01-06 13:43:03 <AinNero> when i buy crap on ebay, i get emoji's in mutt
2020-01-06 13:44:40 <AinNero> they still cause desyncs on quite some terminal emulators, https://w1r3.net/paste/KXWBTc.png
2020-01-06 13:44:44 <mps> I'm getting Chinese spam and see it quite fine in mutt under st, though I don't understand what they want to gift me :)
2020-01-06 13:45:27 <AinNero> the vt520 is only usable because tmux replaces all unicode with "_" when i set LANG correctly
2020-01-06 13:45:37 <north1> Oh, almost bought a chromebook for my dad
2020-01-06 13:45:39 <Cogitri> mps: It's HHKB
2020-01-06 13:45:59 <AinNero> im glad we got utf8 to escape the charset hell
2020-01-06 13:46:09 <AinNero> but unicode turned into a whole different kind of hell
2020-01-06 13:47:14 <Cogitri> HHKB's layout seems great and I really like the Topre switches, but having to learn the layout is annoying
2020-01-06 13:47:26 <Cogitri> Especially when you have to switch back and forth between that and QWERTZ
2020-01-06 13:47:38 <Cogitri> Switching between QWERTY and QWERTZ is annoying already
2020-01-06 13:50:17 <mps> Cogitri: it in name hhk in xkb pkg
2020-01-06 13:51:52 <mps> AinNero: most of my writing/reading is in st and in latin and cyrillic scripts, works fine
2020-01-06 13:52:24 <mps> vim, mutt and less
2020-01-06 13:52:27 <Cogitri> mps: Ah, thought we're talking about the hardware, not the software :)
2020-01-06 13:52:48 <AinNero> mps: cyrillic!
2020-01-06 13:53:12 <mps> Cogitri: heh, at the it is nearly same thing
2020-01-06 13:53:12 <AinNero> im a bit of a slavaboo but i didn't ass myself to actually learn it
2020-01-06 13:53:58 <mps> AinNero: well, half of Slovene use only latin and not cyrillic
2020-01-06 13:54:19 <mps> about half*
2020-01-06 13:54:34 <mps> I don;t know exact percentages
2020-01-06 13:56:20 <AinNero> i got some bulgarian friends, they are my source for slavish stuff
2020-01-06 13:58:21 <mps> if you managed to learn english I think you can learn any other lang
2020-01-06 14:01:17 <AinNero> requires alot of effort to be able to speak it on a acceptable level
2020-01-06 14:01:25 <AinNero> and just for the memes, i dont think thats worth it
2020-01-06 14:01:28 <mps> german lang was more 'natural' to learn for me (aber ich vorgesehen)
2020-01-06 14:02:00 <Cogitri> English was a pretty easy lang to learn or me since German and English are somewhat similar at times
2020-01-06 14:02:13 <Cogitri> But languages like Spanish are pretty tough
2020-01-06 14:02:28 <mps> why not, learning foreign langs is quite good, imo
2020-01-06 14:02:54 <Cogitri> Sure, it's good but I have enough other stuff to do :)
2020-01-06 14:03:08 <Cogitri> And Spanish isn't a language I ever see myself using
2020-01-06 14:03:32 <mps> well, we all have but we have to enjoy sometimes :)
2020-01-06 14:04:59 <mps> not that I think I'm smart, but learning foreign langs helped me to understand my native lang better (and not only lang)
2020-01-06 14:06:28 <mps> when I learned italian to some degree one new world was opened to me, and same happened with russian
2020-01-06 14:12:44 <mps> back to HHK, this is generic term and there are more manufacturers (and much more those who make it DIY)
2020-01-06 14:13:16 <mps> they give them different names
2020-01-06 14:40:27 <PhoenixMage> Hey guys, I am trying to run setup-disk on a new deivce and it seems linux-vanilla doesnt exist?
2020-01-06 14:40:48 <PhoenixMage> It did in 3.10 but not in 3.11 or edge
2020-01-06 14:41:17 <_ikke_> It's replaced with linux-lts
2020-01-06 14:41:48 <PhoenixMage> hmmm, then the setup scripts may need to be updated?
2020-01-06 14:42:55 <_ikke_> if it's hardcoded to use linux-vanilla, then yes
2020-01-06 14:44:06 <_ikke_> It apparently is
2020-01-06 14:44:10 <_ikke_> you can override it with -k
2020-01-06 14:44:55 <_ikke_> PhoenixMage: Can you open an issue here? https://gitlab.alpinelinux.org/alpine/alpine-conf/issues
2020-01-06 14:45:17 <PhoenixMage> Sure, I will do it in the morning
2020-01-06 14:45:24 <PhoenixMage> Sorry its nearly 2am
2020-01-06 14:45:29 <_ikke_> np
2020-01-06 14:57:32 <amolith> I'm trying to modify the startup parameters of lxd but don't know how to do so. I started it with service lxd start so I assume it's running with OpenRC but can't find any docs about editing services
2020-01-06 14:58:07 <clandmeter>  /etc/init.d/ and /etc/conf.d
2020-01-06 14:58:26 <clandmeter> those are the openrc location for init scripts
2020-01-06 15:08:44 <amolith> Looking at the config files doesn't seem to be helping 🤔 When trying to launch an image, I get an error saying no file descriptors available. The Arch wiki says to increase the limit in the systemd service file but I don't see a way to do that with OpenRC. Do you have any idea how I would fix the error?
2020-01-06 15:10:41 <amolith> I'm wanting to use Alpine as both the host OS and the contained OS
2020-01-06 15:10:42 <_ikke_> There is an rc_ulimit variable
2020-01-06 15:15:39 <PhoenixMage> _ikke_: Logged the issue
2020-01-06 15:15:45 <PhoenixMage> I'm off cheers
2020-01-06 15:16:10 <_ikke_> o/
2020-01-06 15:16:13 <_ikke_> sleep tight
2020-01-06 15:16:54 <amolith> _ikke_: what would the syntax for that look like?
2020-01-06 15:17:14 <_ikke_> I have no idea to be honest
2020-01-06 15:17:30 <amolith> I'll continue looking around then lol. Thank you
2020-01-06 15:18:00 <_ikke_> I found it here: https://github.com/OpenRC/openrc/blob/master/user-guide.md
2020-01-06 15:18:05 <_ikke_> but no explenation on how to use it
2020-01-06 15:19:28 <amolith> I found exactly that in Funtoo's wiki
2020-01-06 15:19:28 <amolith> https://www.funtoo.org/Package:OpenRC
2020-01-06 15:23:57 <amolith> So it looks like it would be rc_ulimit="
2020-01-06 15:23:57 <amolith> nproc infinity
2020-01-06 15:23:57 <amolith> "
2020-01-06 15:23:57 <amolith> nofile infinity
2020-01-06 15:26:01 <MY-R> amolith, here are examples from docker: https://github.com/docker/engine/tree/master/contrib/init/openrc
2020-01-06 15:26:13 <MY-R> Alpine using same both files
2020-01-06 15:38:29 <amolith> MY-R: that was perfect but now I'm running into another error 😆 Thank you lol
2020-01-06 16:16:52 <amolith> I got everything fixed. Thanks for the help!
2020-01-06 18:42:54 <gjabell> anyone know if the -R abuild option got removed?
2020-01-06 18:43:04 <gjabell> it's listen in the wiki but not abuild -h
2020-01-06 18:44:35 <_ikke_> https://git.alpinelinux.org/abuild/tree/abuild.in#n2519
2020-01-06 18:44:43 <_ikke_> apparently it's still in the option list, but never checked
2020-01-06 18:45:30 <gjabell> hmm, odd
2020-01-06 18:45:34 <gjabell> wonder why it was changed
2020-01-06 18:46:10 <gjabell> ah
2020-01-06 18:46:21 <gjabell> https://git.alpinelinux.org/abuild/commit/abuild.in?id=9a398eac0c0ac854e5cc05f1b58c54a0f40ebc5f
2020-01-06 18:48:17 <gjabell> I guess -u got similarly removed
2020-01-06 18:49:32 <gjabell> on a somewhat unrelated note, is there a pre-defined way to build packages in a "clean" environment?
2020-01-06 18:49:59 <gjabell> otherwise I sometimes end up with dependencies already installed locally but not remotely, and forget to add them to the depends vars
2020-01-06 18:50:52 <_ikke_> rootbuild
2020-01-06 18:51:32 <gjabell> wow, straight up missed that
2020-01-06 18:51:33 <gjabell> thanks
2020-01-06 19:57:50 <tiredsysadmin> Hi, quick question. I've installed the aarch64 version of Alpine on a raspberry pi, it boots, I can set it up, and enable community repos. However wireguard and zfs (while in the repos) do not seem to have their respective kernel modules packaged.
2020-01-06 19:58:11 <tiredsysadmin> I get an error saying the modules are not there, and looking through the modules directories, I can verify this.
2020-01-06 19:58:37 <tiredsysadmin> I did also try the edge repos, same result.
2020-01-06 20:00:14 <tiredsysadmin> Is this a bug, or am I missing something obvious?
2020-01-06 20:00:44 <artok> might be that they arent built with kernel
2020-01-06 20:02:07 <tiredsysadmin> I would have expected the modules to be packaged in the repos. If you go to wireguard.io, they include installation steps for Alpine. Installing the tools should be enough according to that. On the RPi, the tools and deps install, but no wireguard kernel module.
2020-01-06 20:03:26 <tiredsysadmin> I find it weird that two things that rely on a kernel module are having the same issue.
2020-01-06 20:03:45 <artok> https://www.wireguard.com/compilation/
2020-01-06 20:04:18 <artok> and there you need linux-rpi-dev package
2020-01-06 20:05:05 <tiredsysadmin> Yeah, I can compile a kernel module, but figured the tools and libraries wouldn't be packaged for Alpine without a kernel module binary as well. Or it was a bug/error and the packages are missing.
2020-01-06 20:07:05 <mps> tiredsysadmin: linux-vanilla or linux-lts ?
2020-01-06 20:07:36 <artok> might need another package, as rpi kernel is it's own thing still
2020-01-06 20:07:57 <tiredsysadmin> I tried installing the wireguard-lts and zfs-lts versions, and the former generated a bunch of errors, and the latter just didn't install anything.
2020-01-06 20:08:13 <tiredsysadmin> mps
2020-01-06 20:09:27 <winny> TIL about /etc/apk/repositories.d/{name}.list
2020-01-06 20:09:27 <winny> (had to visit the apk-tools sources to figure that out)
2020-01-06 20:09:31 <tiredsysadmin> artok: hmm, my only exposure to Alpine has been through docker containers, where I obviously don't need a kernel. But even a RPi kernel should be able to load separate binary modules
2020-01-06 20:09:37 <mps> are they for same version as linux-lts
2020-01-06 20:09:56 <tiredsysadmin> mps: I'm not sure and I don't have access to the pi right now
2020-01-06 20:10:30 <mps> afaik, with every kernel upgrade WG is rebuilt
2020-01-06 20:13:20 <tiredsysadmin> Hmm, maybe I should fire it up on an x86_64 VM and see if it behaves differently on that platform
2020-01-06 20:16:20 <tiredsysadmin> Thanks mps and artok, you've given me something to look into.
2020-01-06 20:19:15 <mps> yes, WG is built for rpi http://dl-cdn.alpinelinux.org/alpine/v3.11/community/aarch64/wireguard-rpi-5.4.6-r0.apk
2020-01-06 20:39:52 <Naib> anyone know if mariadb is going to be updated to not be 6x as largE?
2020-01-06 23:32:49 <shunter> When compiling go, I often get the warning "loadinternal: cannot find runtime/cgo". I don't need cgo for what I'm doing, but I'd rather not receive this warning. Does anyone else share this issue?
2020-01-07 10:33:12 <MSHughes> Morning, a package we are using exists on 3.10 but not 3.11. What's the correct thing to do in order to get this supported? I was looking at the aports git repo for the package, and I could try copy it over, but into which branch would it need to exist if this is the right thing to do?
2020-01-07 10:34:17 <AinNero> which package?
2020-01-07 10:35:01 <MSHughes> ospd-netstat
2020-01-07 10:37:11 <MSHughes> The dependant packages are there, just not this :(
2020-01-07 10:37:44 <interpreter> Hello everyone, can somebody help me launch Alpine in QEMU? I'm on x86-64, but want to run AArch64. I'm following https://drewdevault.com/2018/09/10/Getting-started-with-qemu.html except everything is aarch64 and it doesn't boot. Stuck in the QEMU monitor
2020-01-07 10:38:43 <AinNero> fcolista: ping for ospd-netstat
2020-01-07 10:39:22 <AinNero> its still in testing
2020-01-07 10:39:49 <MSHughes> How does that usually work then? How long would it usually sit there?
2020-01-07 10:41:35 <interpreter> Full command I'm using: qemu-system-aarch64 -M virt -m 512 -cpu cortex-a57 -drive file=alpine.qcow2,media=disk,if=virtio -cdrom alpine-virt-3.11.2-aarch64.iso -nic user,model=virtio
2020-01-07 10:41:50 <interpreter> What's missing?
2020-01-07 10:42:05 <AinNero> MSHughes: i just pinged the maintainer, he will have more background than i do
2020-01-07 10:42:33 <MSHughes> Okay, well thanks AinNero, much appriciated
2020-01-07 10:47:01 <AinNero> interpreter already left..
2020-01-07 10:53:45 <fcolista> MSHughes, does ospd-netstat work for you?
2020-01-07 10:54:33 <MSHughes> hey fcolista, I have not been able to test yet. How would I be able to verify?
2020-01-07 10:54:46 <fcolista> MSHughes, that's the reason why is on testing :)
2020-01-07 10:54:54 <fcolista> add the edge/testing repository
2020-01-07 10:54:59 <fcolista> in /etc/apk/repositories
2020-01-07 10:55:10 <MSHughes> fcolista: haha sure, will give it a try
2020-01-07 10:55:16 <fcolista> http://rsync.alpinelinux.org/alpine/edge/testing
2020-01-07 10:56:37 <MSHughes> fcolista: first impression is that it works
2020-01-07 11:11:18 <fcolista> MSHughes, how did you test?
2020-01-07 11:16:19 <newber2020> Good afternoon
2020-01-07 11:21:38 <newber2020> First time trying out alpine, I'm attempting to set it up on my Thinkpad X1 Carbon with Sway wm. So far it went smooth but now I can't get my touchpad to work, trackpoint works fine and the mousebuttons on the touchpad work too (L/R/M). The touchpad also does not show up on `libinput list-devices` or `swaymsg -t get_inputs`.
2020-01-07 11:26:43 <Cogitri> Hm, odd. Can you provide us a `lspci -nnk`? (With lspci from pciutils)
2020-01-07 11:35:52 <newber2020> Will do, just a moment
2020-01-07 11:51:32 <newber2020> Here's my output from `lspci -nnk`: https://pastebin.com/rVRs1wYK
2020-01-07 12:29:29 <MSHughes> fcolista: just calling the command line binary but I’ll test against our stuff shortly
2020-01-07 12:43:38 <MSHughes> fcolista: It looks like it all returns as expected
2020-01-07 12:44:57 <MSHughes> (as with the package in 3.10)
2020-01-07 12:56:56 <fcolista> MSHughes, ok
2020-01-07 12:56:57 <fcolista> algitbot> aports:master |Francesco Colista| community/ospd-nestat: moved from testing | http://dup.pw/aports/c445c7ed
2020-01-07 13:10:21 <MSHughes> Thanks fcolista <3
2020-01-07 14:35:50 <newber2020> If it's any help, I also checked the output of `libinput list-devices` via Fedora live for comparison, there the touchpad does show up as `SYNA8004:00 06CB:CD8B Touchpad`.
2020-01-07 14:37:51 <newber2020> So I guess it's not a libinput problem, but I'm rather missing a necessary module?
2020-01-07 15:01:44 <catkiller> Hi! I've been struggling a lot trying to implement PAM authentication backed by LDAP in alpine.
2020-01-07 15:03:07 <catkiller> Most other distros do it via nsswitch, but that's not there in Alpine, so I've been trying to enable it in /etc/pam.d/. I've got the /lib/security/pam_ldap.so library installed and did a crude attempt at configuring various /etc/pam.d files to use the library
2020-01-07 15:03:40 <catkiller> but I've got no visibility or ways to easily test my configuration, and it doesn't work at all. I can't seem to find anybody who has resolved that issue. Any ideas?
2020-01-07 15:13:14 <_ikke_> catkiller: https://github.com/pikhq/musl-nscd
2020-01-07 15:13:22 <_ikke_> https://pkgs.alpinelinux.org/package/edge/main/x86_64/musl-nscd
2020-01-07 15:29:02 <PICCORO> hy why xfce4 was moved to community repo?
2020-01-07 15:32:11 <_ikke_> probably because it needs dependencies in community, and also because it
2020-01-07 15:32:24 <_ikke_> it's more appropriate for desktop environments anywy
2020-01-07 16:25:19 <catkiller> _ikke_: Thank you! That looks promising!
2020-01-07 17:56:25 <porcellis> anyone here uses alpine as a daily driver for desktop usage?
2020-01-07 17:56:39 <porcellis> if so, how's been your experience with it?
2020-01-07 17:56:50 <KatolaZ> porcellis: define "desktop usage"
2020-01-07 17:56:53 <KatolaZ> :)
2020-01-07 17:58:12 <porcellis> *to do all desktop basic usage, browsing web, watching/playing media, coding, etc*
2020-01-07 18:04:18 <gjabell> I have for about 2 months
2020-01-07 18:04:42 <dalias> porcellis, yes, works fine for me, far better than other dists just because it's not slow and swapping to hell :-p
2020-01-07 18:05:08 <dalias> but there is stuff you can't easily get
2020-01-07 18:05:19 <gjabell> so far really been liking it, although I've ended up having to package some stuff that was missing
2020-01-07 18:05:58 <dalias> i ought to package the things i miss rather than just building them myself
2020-01-07 18:06:55 <shunter> I've been using Alpine for nearly a month here. Definitely more configging and wiki-reading than I normally do since it's also my first Linux experience without Poetteringware.
2020-01-07 18:08:12 <shunter> I'm really comfortable with it where I'm at. It was also my first time contributing when I added a missing package to aports.
2020-01-07 18:09:54 <AinNero> porcellis: most stuff i do is in the terminal anyways, so if i get along with alpine mostly depends on how well firefox works
2020-01-07 18:10:18 <Cogitri> Alpine works fine on my laptop
2020-01-07 18:10:30 <AinNero> i got mobile data working with pppd
2020-01-07 18:10:59 <porcellis> nice!!!
2020-01-07 18:12:09 <porcellis> I've been considering to use as a daily driver as well. So far has used on some of my servers, and been a very pleasant experience.
2020-01-07 18:12:41 <porcellis> I'm using Arch for a year I guess
2020-01-07 18:13:44 <gjabell> arch vs alpine as desktop doesn't feel much different other than the package selection imo
2020-01-07 18:13:56 <gjabell> there might be some things that need heavy patching to run on musl as well
2020-01-07 18:40:25 <mps> anyone upgraded to linux-lts 5.4.8? is it better then previous versions?
2020-01-07 20:21:22 <zava> https://ibb.co/KVHHfY8 hey networking does this. the screenshot is quite self-explanatory. basically I have to sudo setup-interfaces before wlan is able to connect to my wifi. starting or restarting networking produces the error you can see  in the screenshot. is there a solution for this?
2020-01-07 20:59:58 <zava> Hey it appears that setup-interfaces only starts the wpa-supplicant service instead of adding it to runlevel boot.
2020-01-07 21:01:25 <zava> so that when rebooting subsequently a dhcp lease isnt possible and the networking service is forked to the background
2020-01-07 21:02:02 <zava> at least in my case and in that of some others as a search revealed
2020-01-08 01:24:41 <halosghost> o/
2020-01-08 01:25:09 <halosghost> so, if I have an aarch64 machine, can I install the armhf version of alpine?
2020-01-08 01:25:43 <halosghost> and are there any significant pros/cons of doing so (other than the obvious, some things are only written for one or the other)
2020-01-08 01:25:46 <halosghost> ? :)
2020-01-08 01:25:52 <dalias> probably depends. i think not all aarch64 machines can run 32-bit arm
2020-01-08 01:25:57 <halosghost> mm
2020-01-08 01:26:02 <halosghost> that's frustrating
2020-01-08 01:26:09 <dalias> so you have to check
2020-01-08 01:26:17 <dalias> if it can, the advantage is saving a lot of memory
2020-01-08 01:26:30 <dalias> but the disadvantage is not being able to use huge amounts of memory in a single process
2020-01-08 01:26:44 <halosghost> well, to be fair, the machine will have 4GB of RAM on the SoC
2020-01-08 01:26:52 <halosghost> so, 64-bit address space doesn't exactly matter ot me
2020-01-08 01:26:59 <dalias> it still can matter
2020-01-08 01:27:15 <dalias> swap, mapped files, etc
2020-01-08 01:27:22 <halosghost> my typical RAM usage is 1.5GB on a 64-bit x64 system
2020-01-08 01:27:25 <halosghost> :)
2020-01-08 01:27:29 <halosghost> I'm really not worried about RAM :)
2020-01-08 01:28:14 <halosghost> 
ACTION just reread “64-bit x64”

2020-01-08 01:28:19 <halosghost> it's late and I've had a long day
2020-01-08 01:28:22 <halosghost> :)
2020-01-08 01:31:07 <artok> Cavium ThunderX SoC's that are used on many cloud systems can't run 32bit stuff
2020-01-08 01:31:31 <halosghost> anyone familiar enough with pine64 to know if the pinebook pro can run armhf?
2020-01-08 01:31:38 <halosghost> because, if it can, that might be really lovely :)
2020-01-08 01:31:49 <halosghost> the performance improvement and greater compatibility would be lovely
2020-01-08 02:26:27 <maldridge> is something going on with dl-cdn right now?  I can't get dnsmasq to install
2020-01-08 02:27:49 <artok> seems to work for me quite ok
2020-01-08 02:30:25 <maldridge> hmm
2020-01-08 02:31:22 <maldridge> https://hastebin.com/efokorirok.txt
2020-01-08 02:35:33 <halosghost> oh hastebin, for when you need js to share text
2020-01-08 02:56:40 <AlpineDiscordBri> <monado> hastebin doesn’t save your content either
2020-01-08 02:58:15 <halosghost> monado: no idea what that means
2020-01-08 02:59:47 <AlpineDiscordBri> <monado> I have hastebins from a while ago that aren’t saved on their servers
2020-01-08 02:59:53 <AlpineDiscordBri> <monado> so they just net blanks
2020-01-08 03:00:25 <halosghost> useful
2020-01-08 05:24:55 <newber2020> Is it possible that support for newer thinkpad touchpads/synaptics i2c is not included in the alpine kernel yet?
2020-01-08 05:28:19 <tehcloud> newber2020, I had issues when I first started using Alpine
2020-01-08 05:28:33 <tehcloud> it turned out someone sent a patch in to enable some options in the default kernel build
2020-01-08 05:28:42 <tehcloud> that fixed it for me
2020-01-08 05:28:57 <tehcloud> it was related to pinctrl or something...
2020-01-08 05:30:15 <tehcloud> https://gitlab.alpinelinux.org/alpine/aports/issues/11036
2020-01-08 05:30:44 <newber2020> tehcloud: I see, since it's basically the only thing so far not working for me here, trackpoint and touchpad buttons work and libinput can't find it. So this is the only thing left for me to assume
2020-01-08 05:31:06 <tehcloud> yep, it sounds like the exact problem I was having
2020-01-08 05:33:26 <newber2020> So guess I'd have to try compiling a customized kernel to try and find the culprit
2020-01-08 05:33:50 <tehcloud> I'd start with options like the ones in that issue
2020-01-08 05:34:08 <tehcloud> and take a look at dmesg
2020-01-08 05:34:18 <tehcloud> and maybe some boot options can be added to your grub config
2020-01-08 05:36:34 <newber2020> Thanks for the pointers
2020-01-08 07:31:32 <danieli> wonder if I can make alpine run in vmware workstation pro on my new computer - on my work laptop it refused to boot when the VM was configured as 64-bit linux (kernel panic'ed or no bootloader)
2020-01-08 07:32:06 <Cogitri> Why not Qemu?
2020-01-08 07:32:49 <danieli> "I'm having an issue with X" "why not use Y instead?"
2020-01-08 07:32:55 <danieli> it's not an option in this case unfortunately
2020-01-08 07:32:56 <maldridge> I have had a lot of trouble out of vmware danieli
2020-01-08 07:33:20 <danieli> maldridge: it's usually been one of the better hosted hypervisors for me, especially their more enterprise-oriented ones
2020-01-08 07:33:21 <maldridge> I couldn't get stable behavior under fusion, or exsi
2020-01-08 07:33:37 <danieli> NUMA tuning is however a whole different beast to tackle
2020-01-08 07:33:54 <maldridge> yeah I don't even try for NUMA tuning
2020-01-08 07:34:01 <maldridge> I just wanted it to boot reliably
2020-01-08 07:34:28 <Cogitri> danieli: that's unfortunate indeed
2020-01-08 07:34:41 <danieli> the host OS is windows at the moment
2020-01-08 07:35:07 <maldridge> if you have access to it, Hyper-V is shockingly usable
2020-01-08 07:35:09 <danieli> anyway - networking just works, the performance and 'comfort' far outweighs that of virtualbox et al., the more annoying ACPI functionality tends to be more waterproof
2020-01-08 07:35:19 <artok> so far virtualbox has been running alpine quite ok
2020-01-08 07:35:21 <danieli> well, you do have access to it on win10 pro
2020-01-08 07:35:28 <danieli> but it's been worse than ass for me
2020-01-08 07:35:50 <danieli> at least on my work laptop - couldn't for the life of me get any sort of device passthrough /or/ networking to work well
2020-01-08 07:36:08 <maldridge> oh I didn't try for device passthrough, and the network performance on my windows test machine was crap to begin with
2020-01-08 07:36:19 <maldridge> but it did boot things, and it hasn't corrupted the alpine image I have there yet
2020-01-08 07:36:46 <danieli> don't get me wrong, I work with hyper-v a lot at work, so I'm highly familiar with it
2020-01-08 07:37:01 <danieli> but I still prefer vmware, xen, qemu, whatever
2020-01-08 07:40:33 <danieli> maldridge: by having hyper-v disabled and vmware workstation pro active, there is one thing I really want that I'm missing out on - namely Application Guard
2020-01-08 07:40:38 <danieli> it's pretty cool tech
2020-01-08 07:52:11 <danieli> ah, darn it, still kernel panic in 64-bit mode
2020-01-08 08:34:34 <MSHughes> How long does it usually take from something being merged into master on aports for it being accessible on the apk public repo?
2020-01-08 08:42:59 <_ikke_> 15 to 30 minutes
2020-01-08 08:43:23 <_ikke_> But if the builder is stuck on something, it can take longer
2020-01-08 08:45:44 <MSHughes> _ikke_: okay thanks, it has almost been a day; maybe I am missing something
2020-01-08 08:46:13 <_ikke_> MSHughes: what package?
2020-01-08 08:46:28 <MSHughes> ospd-netstat
2020-01-08 08:46:47 <_ikke_> https://pkgs.alpinelinux.org/packages?name=ospd-netstat&branch=edge
2020-01-08 08:46:50 <_ikke_> a specific version?
2020-01-08 08:48:02 <MSHughes> 3.11
2020-01-08 08:49:17 <_ikke_> aports only contains 1.0b1
2020-01-08 08:49:22 <_ikke_> ah
2020-01-08 08:49:32 <_ikke_> yeah, it's only in edge
2020-01-08 08:49:34 <_ikke_> not in 3.11
2020-01-08 08:51:11 <MSHughes> _ikke_: That's what I am trying to work out :D sorry, what does it take now?
2020-01-08 08:51:34 <danieli> stuff doesn't get backported
2020-01-08 08:51:41 <danieli> well, not unless it's some sort of critical fix
2020-01-08 08:51:59 <_ikke_> MSHughes: You would need to use edge to get that package
2020-01-08 08:52:12 <_ikke_> The next stable release (3.12) would contain it
2020-01-08 08:52:16 <danieli> that package will be included in the next stable re--
2020-01-08 08:52:17 <danieli> what he said
2020-01-08 08:52:28 <MSHughes> Oh I see!
2020-01-08 08:52:52 <danieli> you could however do package pinning to pull that specific package from edge into stable, but that's risky business because cross-dependencies can break due to versioning
2020-01-08 08:54:18 <MSHughes> The dependant packages are fine in 3.11
2020-01-08 08:54:48 <_ikke_> for now
2020-01-08 08:56:24 <MSHughes> Sure but the existing packages don't change on a version, right?
2020-01-08 08:56:43 <MSHughes> Or do they? It's just new packages that don't come?
2020-01-08 08:57:05 <_ikke_> MSHughes: it can be that the package is rebuilt against new version in edge that 3.11 does not have
2020-01-08 08:57:33 <MSHughes> Okay that makes sense
2020-01-08 08:57:53 <MSHughes> I think our testing would pick that up actuaally. How would we be able to pin to a version in edge?
2020-01-08 08:58:14 <_ikke_> add @edge <url> to /etc/apk/repositories
2020-01-08 08:58:27 <_ikke_> and then apk add ospd-netstat@edge
2020-01-08 08:58:51 <MSHughes> Thanks a million, and sorry for the dumb questions <3 I'll give it a go
2020-01-08 08:59:06 <_ikke_> Happy to help
2020-01-08 09:12:58 <markand> I'm considering a switch to Alpine on my main laptop, I usually use efibootmgr/luks rather than full bootloaders, anyone did that or know if it's possible?
2020-01-08 09:13:14 <markand> my idea is to simply partitionate as usual and use the minimal rootfs then install just what I need
2020-01-08 09:13:32 <markand> but I've never tried luks on openrc based systems
2020-01-08 09:23:40 <AinNero> its not a matter of the init system, its a matter of the initramfs
2020-01-08 10:00:05 <markand> ah okay, because on systemd based systems there is a special service for that that break the bootup until the passphrase is put
2020-01-08 10:09:43 <mps> anyone have amdgpu driver issue on linux-lts?
2020-01-08 10:09:51 <Cogitri> Nop, works for me
2020-01-08 10:09:54 <Cogitri> On a 5700XT
2020-01-08 10:10:04 <Cogitri> And my VegaM in my laptop
2020-01-08 10:10:41 <mps> ah, you are assignee for #11081
2020-01-08 10:12:22 <mps> could you reply there again and tell OP that it works
2020-01-08 10:12:58 <mps> I don't have amd gpu so not authoritative to answer
2020-01-08 10:17:11 <Cogitri> I replied already: https://gitlab.alpinelinux.org/alpine/aports/issues/11081#note_58809
2020-01-08 10:17:17 <Cogitri> Didn't see the notifications though
2020-01-08 10:17:32 <Cogitri> It really is annoying that Gitlab doesn't automatically subscribe me to Issues I replied on
2020-01-08 10:21:23 <mps> yes, somewhat annoying. and I would like if it send me even my replies to have full thread in mailbox
2020-01-08 13:21:40 <trickos> Hi, I've permission problem on /tmp folder with standard user. I tried to mount tmpfs as /tmp with rw, without success.
2020-01-08 13:22:48 <trickos> ( in my case, with NVT sync attempt - Openvas)
2020-01-08 13:24:03 <trickos> someone have a tips to resolv this problem ?
2020-01-08 13:25:00 <mps> dir, partition or tmpfs
2020-01-08 13:27:44 <trickos> "can't make dir ...:permission denied; can't create directory: permission denied" on /tmp
2020-01-08 13:27:45 <trickos> I tried to add "tmpfs /tmp tmpfs rw,nosuid,nodev" in fstab file, change permission, without success
2020-01-08 13:28:26 <trickos> So I rollback the fstab file and reboot to get the 'default install scheme'
2020-01-08 13:29:39 <mps> mode=1777
2020-01-08 13:30:11 <mps> tmpfs           /tmp            tmpfs   nodev,nosuid,size=10%,mode=1777    0 0
2020-01-08 13:31:49 <trickos> Same result
2020-01-08 13:39:05 <mps> do you see /tmp is mounted
2020-01-08 13:42:09 <trickos> yes
2020-01-08 13:42:26 <trickos> tmpfs on /tmp type tmpfs ...
2020-01-08 13:44:05 <mps> ls -ld /tmp
2020-01-08 13:45:26 <trickos> drwxrwxrwt root root ... /tmp
2020-01-08 13:47:17 <mps> hmm, touch /tmp/tmpfile (as user)
2020-01-08 13:47:34 <mps> not root
2020-01-08 13:48:43 <trickos> success
2020-01-08 13:49:38 <trickos> but with greenbone-nvt-sync, untar the bz2 in /tmp failed with permission message
2020-01-08 13:51:27 <trickos> very strange
2020-01-08 13:51:33 <mps> heh, don't know what is this, but looks suspicious
2020-01-08 13:51:44 <trickos> yes T_T
2020-01-08 13:53:26 <trickos> If the author of https://wiki.alpinelinux.org/wiki/Setting_up_GVM10 was here (:
2020-01-08 14:13:22 <fcolista> trickos, i'm here
2020-01-08 14:13:42 <fcolista> what's up?
2020-01-08 14:16:30 <trickos> fcolista I've an issue with greenbone nvt sync
2020-01-08 14:16:58 <fcolista> writing permission on /tmp is not a greenbone nvt sync i believe
2020-01-08 14:16:59 <trickos> With permission denied after wget the tar.bz2
2020-01-08 14:17:49 <trickos> ""can't make dir ...:permission denied; can't create directory: permission denied" on /tmp"
2020-01-08 21:52:52 <bl4ckb0ne> is there a way to mesure boot time?
2020-01-08 22:03:31 <AinNero> bootchartd is a popular choice
2020-01-08 22:04:17 <AinNero> but if you only need a single duration measurement, reading the uptime after boot is usually fine
2020-01-09 00:47:55 <dsmster> just installed alpine 3.11.2 on esx 6.5, boots up, ran setup, but network not working; eth0 is there, vmxnet3 device, configured up with correct ip/mask/broadcast, but it's not passing traffic, any ideas?  other linux os working fine ?
2020-01-09 00:57:44 <dsmster> interesting, the -virt version would not talk to network but the -standard version did.  so - going with standard
2020-01-09 02:37:09 <PhoenixMage> hmmmm, How can I stop abuild stripping delimited quotes inside of env vars?
2020-01-09 02:38:33 <PhoenixMage> https://pastebin.com/bxaiv62r
2020-01-09 02:39:37 <PhoenixMage> echo returns the correct things with the " intact but cmake complains it the -latomic dir doesnt exist
2020-01-09 06:36:57 <atkka> hello, anyone have any issues with the latest pi/arm images?
2020-01-09 06:38:39 <atkka> nvm: https://alpinelinux.org/posts/Alpine-3.11.2-released.html
2020-01-09 06:40:16 <PhoenixMage> atkka: What issues?
2020-01-09 06:40:51 <PhoenixMage> I am running it on RPi3, Lime 2 and Rock Pi 4
2020-01-09 06:40:54 <atkka> 3.11.0 wouldn't boot on pi
2020-01-09 06:41:09 <atkka> 3.11.2 fixed the issues aparently, see above link
2020-01-09 06:41:58 <PhoenixMage> So what is the issue you are seeing?
2020-01-09 06:42:07 <PhoenixMage> Or you havent tried yet?
2020-01-09 06:43:51 <atkka> my problem was 3.11.0 images won't boot
2020-01-09 06:44:00 <atkka> downloading new images and trying again
2020-01-09 06:45:54 <PhoenixMage> ok, so general question, not specific issue
2020-01-09 07:00:32 <atkka> PhoenixMage: all sorted now
2020-01-09 07:00:37 <atkka> just bad images is all
2020-01-09 07:08:07 <trickos> Hi, how to install a specific apkbuild since https://github.com/faradayio/aports/tree/master/testing?
2020-01-09 07:13:55 <KatolaZ> trickos: why do you need to install 4-years-old packages?
2020-01-09 07:14:21 <KatolaZ> (and unofficial ones, apparently...)
2020-01-09 07:21:22 <trickos> To try a setup with fcolista
2020-01-09 07:22:00 <trickos> By the main and community v3.11, one daemon of an install doesn't exit
2020-01-09 07:22:03 <trickos> exist
2020-01-09 07:52:49 <PhoenixMage> trickos: What daemon?
2020-01-09 07:55:25 <trickos> openvassd PhoenixMage
2020-01-09 07:57:03 <PhoenixMage> There is a community package for that
2020-01-09 07:57:13 <PhoenixMage> apk add openvas
2020-01-09 07:57:36 <PhoenixMage> Make sure you have a community repo in /etc/apk/repositories
2020-01-09 08:08:16 <trickos> yes PhoenixMage
2020-01-09 09:33:37 <mps> hm, reading debian sec announce about firefox 'which could potentially result in the execution of arbitrary code, data exfiltration or cross-site scripting.'
2020-01-09 09:34:17 <mps> isn't that default for browsers :)
2020-01-09 09:44:58 <Cogitri> Well, the vuln certainly is pretty bad
2020-01-09 09:45:13 <Cogitri> Will look at it once home if no one beats me to it
2020-01-09 09:46:31 <mps> I'm trying to look from funny side of life on these 'sec fixes/theater'
2020-01-09 09:47:31 <mps> right now building 72.0.1 on x86_64 and armv7. will see will the build work
2020-01-09 09:48:05 <mps> oh, while typing armv7 finishing :)
2020-01-09 11:12:47 <PhoenixMage> So, anyone have any idea how to delimit quotes in a variable and get abuild to use it without stripping them
2020-01-09 11:58:58 <_ikke_> PhoenixMage: it's just shell scripting
2020-01-09 11:59:04 <_ikke_> PhoenixMage: can you show an example?
2020-01-09 12:00:48 <PhoenixMage> https://pastebin.com/bxaiv62r
2020-01-09 12:01:04 <PhoenixMage> The echo prints it correcly
2020-01-09 12:01:28 <PhoenixMage> When the cmake is run its like the quotes arent there
2020-01-09 12:01:45 <PhoenixMage> Thats from build()
2020-01-09 14:03:55 <bl4ckb0ne> can I have a review for this https://lists.alpinelinux.org/~alpine/aports/patches/3204 please?
2020-01-09 14:06:20 <_ikke_> At first glance it looks ok
2020-01-09 14:39:30 <telmich> good morning
2020-01-09 14:39:58 <telmich> did the rofi package vanish? I have it on an older installation, but searching on a new installation, I don't see the package anymore
2020-01-09 14:40:37 <telmich> hmm, it's in aports/testing
2020-01-09 14:40:44 <_ikke_> 16abf166cd8900e26e72cace334c284e9476c5df
2020-01-09 14:41:22 <_ikke_> That's rofi-pass
2020-01-09 14:41:27 <bl4ckb0ne> > depends="xdg-utils rofi pass gawk pwgen xdotool xclip"
2020-01-09 14:41:31 <_ikke_> 2fd96ce013b3eaafaa0f7a3088b98fe75d6ac273
2020-01-09 14:41:48 <bl4ckb0ne> ah its a bot
2020-01-09 14:41:57 <bl4ckb0ne> 
ACTION goes get another coffee

2020-01-09 14:44:53 <telmich> hmm, similar issue with ghc I assume
2020-01-09 14:45:34 <telmich> background: I was trying to setup a new notebook with cdist so a lot of packages are pulled in; maybe I can have a loot at rofi later.
2020-01-09 15:01:58 <kapil_> hello, how to install new alpine linux?
2020-01-09 15:02:58 <bl4ckb0ne> https://wiki.alpinelinux.org/wiki/Installation
2020-01-09 15:05:14 <kapil_> thanks
2020-01-09 15:07:48 <kapil_> can i install plazma desktop?
2020-01-09 15:30:43 <markand> kapil_, yes https://wiki.alpinelinux.org/wiki/KDE
2020-01-09 15:30:53 <markand> there is even GNOME these days
2020-01-09 15:31:23 <kapil_> thank you for replying
2020-01-09 15:31:40 <kapil_> is it faster than arch linux?
2020-01-09 15:32:10 <markand> since they use the Linux kernel no there are no much chanecs
2020-01-09 15:32:49 <markand> the "big" difference between Arch and Alpine is the musl (the C library) which is the core of everything
2020-01-09 15:33:11 <markand> and obviously few other things but they don't matters (busybox, mandoc, etc)
2020-01-09 15:33:17 <kapil_> what is benifit of musl?
2020-01-09 15:33:35 <markand> clean and correct
2020-01-09 15:33:39 <markand> glibc is total bloatware
2020-01-09 15:33:59 <kapil_> oh thank you so much.
2020-01-09 15:34:00 <bl4ckb0ne> musl is smaller and with no gnu crap
2020-01-09 15:34:10 <kapil_> oh great
2020-01-09 15:34:48 <kapil_> i will try it. but packages are too less
2020-01-09 15:35:08 <markand> indeed, Alpine has not the same amount of contributors and is much more recent
2020-01-09 15:35:34 <markand> we waited for you kapil_ to bring new packages :))
2020-01-09 15:36:04 <bl4ckb0ne> i did more package in 1 week of alpine than several years of arch
2020-01-09 15:36:09 <kapil_> thanks `apk add plasma-desktop` gives error : unsatisfiable constrains
2020-01-09 15:36:26 <markand> I think you may need to enable community repo
2020-01-09 15:36:41 <kapil_> thanks
2020-01-09 15:36:44 <markand> https://wiki.alpinelinux.org/wiki/Enable_Community_Repository
2020-01-09 15:37:13 <markand> make sure to use the same version (the wiki page mention edge)
2020-01-09 15:38:18 <kapil_> thanks, now its working
2020-01-09 15:39:06 <kapil_> this is the first linux i installed without user manual
2020-01-09 15:42:21 <PureTryOut[m]> Oh someone installing Plasma on Alpine, nice 😃
2020-01-09 15:42:54 <markand> I'm gonna try GNOME this weekend
2020-01-09 15:46:25 <kapil_> after following guide , when i do xinit it dont start kde
2020-01-09 15:46:35 <kapil_> kde crash
2020-01-09 15:47:16 <kapil_> Errors from xkbcomp are not fatal to the X server
2020-01-09 15:48:45 <PureTryOut[m]> That's just a warning that you can ignore
2020-01-09 15:49:09 <PureTryOut[m]> You did run `setup-xorg-base` first right?
2020-01-09 17:12:18 <micro4lpha> what's the naming convention people use to throw in a default config which does not replace the one you are actually running?
2020-01-09 17:12:21 <micro4lpha> such as default.conf
2020-01-09 17:12:26 <micro4lpha> default.conf_default ?
2020-01-09 17:17:36 <dalias> i use .dist
2020-01-09 17:17:46 <dalias> (for "as distributed"/"for distribution")
2020-01-09 17:17:53 <dalias> e.g. config.mak.dist
2020-01-09 18:39:12 <Cogitri> name.config.default is common too
2020-01-10 00:52:10 <PhoenixMage> So if I am making my own packages, is there a standard place to keep them?
2020-01-10 01:20:30 <atkka> hello
2020-01-10 01:20:55 <atkka> I'm having some raspberry pi related issues
2020-01-10 01:22:03 <atkka> trying to issue lbu commit -e, aes-256-cbc is not supported
2020-01-10 01:25:20 <atkka> openssl is up to date
2020-01-10 01:30:06 <atkka> can anyone recommend another cipher?
2020-01-10 04:35:10 <kaey> atkka: encryption looks broken to me
2020-01-10 04:35:17 <kaey> specifically this check https://git.alpinelinux.org/alpine-conf/tree/lbu.in#n129
2020-01-10 04:35:39 <kaey> openssl enc -ciphers returns columnated list
2020-01-10 04:38:02 <kaey> furthemore it doesn't look like it uses proper kdf (like pbkdf or scrypt) and authentification (like hmac)
2020-01-10 04:38:52 <kaey> i think lbu should switch to using scrypt instead of openssl directly
2020-01-10 04:51:05 <atkka> kaey: thank you very much
2020-01-10 04:54:05 <atkka> I will poke around the script some and see if I can make it work
2020-01-10 04:54:35 <atkka> I noticed a couple minor issues at first glance
2020-01-10 04:56:57 <atkka> broken encryption makes it a no go for my current project
2020-01-10 04:57:06 <atkka> and day to day use in general
2020-01-10 05:08:37 <kaey> you can use lbu pkg somefile.tgz and then encrypt it separately, with scrypt for example
2020-01-10 05:11:33 <atkka> do you know how lbu commit -e worked in the past?
2020-01-10 05:11:58 <atkka> how did it decrypt at boot to load your commits?
2020-01-10 05:21:28 <kaey> https://git.alpinelinux.org/mkinitfs/tree/initramfs-init.in#n46
2020-01-10 05:21:40 <kaey> atkka: ^
2020-01-10 05:21:59 <atkka> kaey: thanks, checking
2020-01-10 05:23:29 <atkka> doesn't it need a password?
2020-01-10 05:24:48 <kaey> it asks for it
2020-01-10 05:25:05 <atkka> ok, I couldn't parse that
2020-01-10 05:25:14 <atkka> still learning posix sh
2020-01-10 05:26:12 <atkka> so, you said earlier it looked broken, do you think it ever worked?
2020-01-10 05:27:12 <atkka> if I'm reading it right those lines from lbu.in were commited in 2007
2020-01-10 05:33:49 <iggy> PhoenixMage: not that I've heard of, but if you find something, let us know
2020-01-10 07:56:54 <AlphaHan> Im trying to install the package coturn on a clean system. But it seems broken. Any info on that ?
2020-01-10 08:04:19 <Cogitri> Define broken
2020-01-10 08:17:04 <AlphaHan> Cogitri: A package you are not able to install because it cannot sort its own dependencies and you cant either
2020-01-10 08:17:38 <AlphaHan> A package thats in the sytem and available, but cannot be used.
2020-01-10 08:19:08 <_ikke_> AlphaHan: He is asking for details, not a general description
2020-01-10 08:19:14 <_ikke_> What is the error you are getting
2020-01-10 08:19:33 <AlphaHan> Ikkr: Working on it
2020-01-10 08:20:05 <AlphaHan> https://i.postimg.cc/sXXZp6Cs/coturn.png
2020-01-10 08:23:29 <_ikke_> It might need to be rebuilt
2020-01-10 08:23:43 <AlphaHan> looks like it
2020-01-10 08:24:18 <AlphaHan> Would just like to report it. Not sure where I should do it
2020-01-10 08:26:45 <chromebooksla> How to install 'info' program on alpine linux?
2020-01-10 08:42:33 <_ikke_> AlphaHan: gitlab.a.o
2020-01-10 09:15:46 <markand> I'm considering installing Alpine on my VPS (using the miniroot fs) and then install all packages etc
2020-01-10 09:15:58 <markand> any recommendations on it?
2020-01-10 09:17:35 <markand> the painful thing is the bootloader, because the remote shell they provide will present the disk as /dev/sdb and afterwards when you reboot to the normal system it is /dev/sda
2020-01-10 09:18:02 <Cogitri> Why is that a pain?
2020-01-10 09:18:12 <Cogitri> Just use UUIDs for your rootfs and it won't matter
2020-01-10 09:18:32 <markand> yes that seems to be a good idea
2020-01-10 09:41:40 <markand> what's the difference between armhf and armv7 images? is armhf the armv6 synonym?
2020-01-10 09:42:10 <atkka> any consensus on why lbu commit -e is broken?
2020-01-10 09:42:27 <atkka> markand: armv7
2020-01-10 09:42:41 <atkka> is for multi core pis, 2 & 3
2020-01-10 09:43:09 <atkka> armhf works on single core pis, zero etc
2020-01-10 09:43:25 <markand> okay, I guess armv7 is hard-float as well?
2020-01-10 09:43:35 <atkka> I believe so
2020-01-10 09:43:39 <markand> understood, thanks
2020-01-10 09:44:23 <atkka> np, you can use aarch64 for 3 and 4 as well
2020-01-10 09:45:00 <atkka> basically, armhf for original and zeros, armv7 for 2, aarch64 for 3 and 4
2020-01-10 09:45:28 <mps> VFPv3, NEON and Thumb
2020-01-10 09:45:44 <mps> differences, I mean
2020-01-10 09:46:15 <mps> i.e. armv7 uses features from newer CPUs
2020-01-10 09:47:54 <mps> so, armhf is for older boards while armv7 is for 'newer' ones
2020-01-10 09:48:20 <atkka> and aarch64 is for newerer ones :)
2020-01-10 09:48:21 <mps> while the 'line' is more zig-zag
2020-01-10 09:48:46 <mps> aarch64 is for 64bit arm cpus
2020-01-10 09:48:58 <atkka> 3 and 4 pi have 64bit arms
2020-01-10 09:49:17 <atkka> and I belive pi2 ver. 1.2 as well
2020-01-10 09:49:21 <markand> but AFAIK there are some video issues on the pi
2020-01-10 09:49:37 <markand> can't remember well, like one has opensource driver while the opposite not
2020-01-10 09:49:47 <mps> rpi4 and rpi3 is 64bit, though not sure if there is also 32bit rpi3 variant
2020-01-10 09:50:26 <atkka> markand: I'm not sure what video issues you speak op
2020-01-10 09:50:31 <atkka> s/op/of
2020-01-10 09:51:07 <atkka> if you need hardware acceleration for videos and such you can use omxplayer
2020-01-10 09:51:17 <mps> though armv7 alpine can be used on some 64bit boards
2020-01-10 09:52:33 <atkka> mps: do you use diskless mode on any of your installs?
2020-01-10 09:53:23 <mps> only for testing RPi zero
2020-01-10 09:53:36 <atkka> have you tried lbu commit -e?
2020-01-10 09:53:48 <atkka> it is broken from my understanding
2020-01-10 09:54:44 <mps> no, I didn't, but noticed your issue (yesterday?) and thought to test when find some time
2020-01-10 09:55:04 <atkka> thanks, I haven't done thorough testing yet
2020-01-10 09:55:21 <atkka> was going to file a bug report tomorrow if I can't find a solution
2020-01-10 09:57:29 <atkka> when using a diskless install all important/user info is archived in a apkvol.tar.gz correct
2020-01-10 10:00:25 <mps> I'm trying to make it bootable with u-boot, so not ready for test right now
2020-01-10 10:01:29 <atkka> ok, no problem
2020-01-10 10:03:28 <mps> it booting, what I should try 'lbu -e' only?
2020-01-10 10:03:53 <atkka> lbu commit -e
2020-01-10 10:04:03 <mps> yes, this
2020-01-10 10:04:36 <mps> Cipher aes-256-cbc is not supported
2020-01-10 10:05:13 <atkka> yep
2020-01-10 10:05:17 <atkka> same here
2020-01-10 10:05:33 <atkka> doesn't matter what I change it to
2020-01-10 10:05:53 <atkka> checking with openssl ciphers -v
2020-01-10 10:06:19 <atkka> pretty sure that's the command that lists available ciphers
2020-01-10 10:08:42 <markand> someone can share its `rc-update show -v` output?
2020-01-10 10:09:11 <atkka> markand: sure one sec
2020-01-10 10:10:53 <atkka> https://p.teknik.io/SoRAr
2020-01-10 10:11:11 <markand> thanks
2020-01-10 10:11:25 <atkka> sure thing
2020-01-10 10:11:57 <mps> atkka: looks like a bug
2020-01-10 10:14:51 <atkka> thanks, I am working on it but its above my pay grade :)
2020-01-10 10:15:38 <atkka> bug from 2007?
2020-01-10 10:16:21 <mps> I don't have much free time these days
2020-01-10 10:18:45 <atkka> I have free time but a lot to learn before I'm able to contribute mush
2020-01-10 10:18:51 <atkka> s/mush/much
2020-01-10 10:23:59 <AinNero> recently we discussed some method for reproducible network interface names
2020-01-10 10:24:05 <AinNero> afair it was a new aport
2020-01-10 10:24:11 <AinNero> anyone remembers the name of it?
2020-01-10 10:26:15 <AinNero> ah, grepping irclogs helped, its macifrename
2020-01-10 10:27:02 <mps> macifrename
2020-01-10 10:27:17 <mps> eh, same idea :)
2020-01-10 10:28:55 <markand> mounting /dev/vda1 on /sysroot failed: No such file or directory
2020-01-10 10:28:59 <markand> Oo
2020-01-10 10:29:12 <markand> of course /dev/vda1 is listed and seen in fdisk -l
2020-01-10 10:29:54 <AinNero> does /sysroot exist?
2020-01-10 10:29:56 <markand> I suppose I forgot to install e2fsprogs maybe
2020-01-10 10:29:57 <markand> yes
2020-01-10 10:31:02 <AinNero> you dont need e2fsprogs for mounting ext, usuall
2020-01-10 10:31:16 <AinNero> whats your setup?
2020-01-10 10:31:18 <markand> yes obviously its in the kernel
2020-01-10 10:31:54 <markand> a custom installation (mostly based on the https://wiki.alpinelinux.org/wiki/Alpine_Linux_in_a_chroot page)
2020-01-10 10:32:28 <markand> I guess my initramfs is probably too minimalistic
2020-01-10 10:56:21 <dwmon> Hi all, does anyone know if uwsgi-python3 would have been removed from the latest version? My builds that have been running fine for 12 months have just started failing. I manually pulled down the Alpine main/community repos and there's only a uwsgi-python (no 3)
2020-01-10 11:02:20 <dwmon> This does appear to be the case, as I've just pinned the build back to v3.10 and all is fixed. Would anyone be able to clarify why uwsgi-python3 has been binned off?
2020-01-10 11:07:21 <MY-R> dwmon, uwsgi-python is python3
2020-01-10 11:07:53 <dwmon> hmm, I swapped to that as a first test, but my build then 'succeeded' but didn't actually work
2020-01-10 11:09:21 <dwmon> but if that's what's happened, at least that solves that part of the puzzle, thanks MY-R
2020-01-10 11:09:27 <MY-R> 3.11.0: "Python 2 is deprecated. Majority of Python 2 packages was removed and will be completely removed in next release."
2020-01-10 11:09:31 <dwmon> Now to solve: !!! UNABLE to load uWSGI plugin: Error loading shared library /usr/lib/uwsgi/python3_plugin.so: No such file or directory !!!
2020-01-10 11:09:49 <Cogitri> Odd, that works fine for me
2020-01-10 11:09:55 <dwmon> Yeah I read that in the release notes, but I didn't think that would affect the package name
2020-01-10 11:10:03 <dwmon> presumed the python3 one would stay as python3
2020-01-10 11:10:09 <dwmon> or 'assumed' should I say
2020-01-10 11:10:34 <AinNero> probably 3ee2dbe9737e86f9432a72e9cf49f41de7e5804b
2020-01-10 11:13:18 <MY-R> dwmon, probably there is no point to keep that naming since py3 stuff should be default after drop py2
2020-01-10 11:13:46 <dwmon> MY-R, I can see why that makes things cleaner in the long-term, just slightly tricky in the short-term
2020-01-10 11:13:54 <dwmon> but obviously it's the way things should head
2020-01-10 11:13:55 <MY-R> yep
2020-01-10 11:14:11 <AinNero> dwmon: did you run CI on edge?
2020-01-10 11:14:20 <dwmon> I was just pulling latest
2020-01-10 11:15:01 <AinNero> if you need stability, consider staying on stable branches
2020-01-10 11:15:37 <dwmon> yeah this isn't a live product yet (despite the enormous dev time), things in live are on stable, but this is still in dev so it just pulls the latest
2020-01-10 11:15:42 <dwmon> thank you though
2020-01-10 11:20:54 <dwmon> I'll move back to 3.11 when I get more time and then figure out how to fix the shared library load issue, thanks for the help all
2020-01-10 11:26:31 <dwmon> back to 3.10*
2020-01-10 11:58:53 <jirib> hi, all
2020-01-10 11:59:09 <jirib> no pandoc on alpine? nobody uses it or is there any technical problem?
2020-01-10 12:03:20 <Cogitri> Probably Haskell
2020-01-10 12:28:17 <AinNero> yes, haskell
2020-01-10 12:28:31 <AinNero> ghc is a huge pita for maintenance
2020-01-10 12:29:16 <AinNero> s/maintenance/packaging/
2020-01-10 12:30:05 <Cogitri> Yes it is
2020-01-10 12:31:19 <Cogitri> Huh, `du -h` and `ls -lah` print different sizes for a file, that's fun
2020-01-10 12:32:19 <AinNero> i think sparse files may cause such a difference
2020-01-10 12:32:23 <AinNero> the ls one is larger?
2020-01-10 12:32:51 <Cogitri> Yes, way larger
2020-01-10 12:32:56 <Cogitri> 17M vs 4M
2020-01-10 12:33:24 <Cogitri> Ans 587M for the entire dir vs 2.6GB for the entire dir with du vs. ls
2020-01-10 12:33:28 <Cogitri> s/Ans/And/
2020-01-10 12:34:05 <Cogitri> The dir is full of coredumps
2020-01-10 12:35:40 <MY-R> Cogitri, heh, type in google "du ls different size" and enjoy :D
2020-01-10 12:35:49 <Cogitri> Ah, `du -bh` does the trick, thanks for the pointer AinNero
2020-01-10 12:37:05 <AinNero> ah yes, if -b fixed it, it were sparse files
2020-01-10 12:51:14 <markand> okay I think I lack kvm_(intel) modules to appropriately mount /dev/vda1 device in my initramfs
2020-01-10 12:51:48 <AinNero> the kvm modules usually dont mount stuff...
2020-01-10 12:52:31 <markand> hmmm, It's the only module I see missing between my other vps though (both have virtio)
2020-01-10 13:02:28 <markand> I have features="ata base ide scsi usb virtio ext4 keymap" in my mkinitfs.conf for now
2020-01-10 13:11:59 <markand> okay with a rootfstype=ext4 it boots
2020-01-10 13:12:11 <jirib> funny voidlinux has pandoc
2020-01-10 13:12:31 <markand> 
ACTION makes angry look at busybox's mount

2020-01-10 13:12:41 <markand> what a stupid error to use ENOENT to indicate missing -t option
2020-01-10 13:13:19 <mps> markand: iirc, this is because busybox mount
2020-01-10 13:13:49 <MY-R> same story about using -t so it will load module?
2020-01-10 13:14:13 <markand> btw, just before initramfs starts I have “sh: zfs: unknown operand”
2020-01-10 13:14:33 <mps> MY-R: it should, but who knows :)
2020-01-10 13:15:26 <markand> hurray I have my VPS alpine :)
2020-01-10 13:15:36 <MY-R> more like people always assume that for example ext4 module is loaded, in most of Alpine tutorials I see -t
2020-01-10 13:17:54 <markand> is there a way to prevent apk from updating my /boot/grub/grub.cfg on update btw?
2020-01-10 13:18:18 <mps> I always ad 'rootfstype' in boot loader, if it isn't there already ofc
2020-01-10 13:19:34 <MY-R> markand, what have you got in grub.cfg that cant put in dunno... /etc/default/grub ?
2020-01-10 13:20:04 <markand> well, my actual grub.cfg with 6 lines written by hand are sufficent
2020-01-10 13:21:15 <MY-R> mps, but all those problem with missing ext4 are when running Alpine live to install Alpine cus after installation rootfstype= is added automaticly
2020-01-10 13:22:20 <mps> MY-R: for x86* yes, but not sure for arms
2020-01-10 13:22:33 <mps> ARMs*
2020-01-10 13:23:06 <MY-R> ye ARM need little more to play with boot stuff
2020-01-10 13:23:51 <mps> I'm trying to move all my boxes to ARM
2020-01-10 13:26:06 <MY-R> markand, grub.cfg is auto generated file so editing it from hand is just "wrong" and files is generated everytime you run grub-mkconfig
2020-01-10 13:28:00 <markand> I usually don't use grub-mkconfig since it generates bloat config... but yeah I'll use /etc/default/grub since it's not possible to disable autogen
2020-01-10 13:28:21 <MY-R> "bloat config"
2020-01-10 13:28:30 <MY-R> so dont use grub
2020-01-10 13:28:55 <markand> unfortunately on MBR based systems it still the least worse IMHO (I usually stick with efibootmgr when I can)
2020-01-10 13:29:28 <MY-R> Alpine default bootloader is syslinux
2020-01-10 13:30:49 <MY-R> mps, what about compiling stuff, any powerfull arm machine?
2020-01-10 13:32:20 <mps> I have one armv7 chromebook with 4GB ram and 8 CPUs
2020-01-10 13:32:52 <mps> and usb ssd for FS intensive tasks
2020-01-10 13:33:11 <markand> 
ACTION is in hurry to get hands on RISC-V based desktops

2020-01-10 13:33:53 <MY-R> Alpine got efibootmgr too
2020-01-10 13:33:54 <MY-R> ah
2020-01-10 13:34:08 <MY-R> mps :)
2020-01-10 13:34:37 <mps> ofc, I have aarch64 chromebook also
2020-01-10 13:35:14 <MY-R> maniac! ;)
2020-01-10 13:36:11 <mps> no, they are a lot better choices than x86*
2020-01-10 13:36:46 <markand> the problem with ARM is the high number of variants
2020-01-10 13:36:46 <mps> especially with alpine :)
2020-01-10 13:38:36 <mps> 'especially with alpine :)' I mean, it is better choice
2020-01-10 13:40:46 <markand> I need to contrib some packages to get it on my desktop as well
2020-01-10 13:51:53 <bl4ckb0ne> why are the dri drivers in /usr/lib/xorg/modules/dri ?
2020-01-10 13:58:14 <telmich> is there any database/kv store that you can recommend that easily works in a cluster and has decent python client support?
2020-01-10 13:58:41 <telmich> I started using etcd (python-etcd3) some time ago, but the library is pretty bare and reconnects/cluster changes are not well handled
2020-01-10 13:59:05 <telmich> I used to run some mongodb clusters years ago, but I found mongodb quite unstable by default
2020-01-10 14:05:47 <patrick2> Hi. I want to move my home NAS from freebsd (xigmanas) to alpine linux. I do know my way around in Linux and I am using it as my main OS for the past 15 years.
2020-01-10 14:06:35 <patrick2> The difficulty I have with the switch is ZFS. All my data is on two ZFS pools and I do not want to change that.
2020-01-10 14:07:42 <Cogitri> So what's your problem?
2020-01-10 14:07:48 <Cogitri> I happen to use ZFS on Alpine
2020-01-10 14:08:11 <patrick2> I am testing if this works and how I would do it via virtualbox (on an ArchLinux laptop)
2020-01-10 14:08:32 <patrick2> So, I am trying to build an Alpine install ISO/USB with ZFS support.
2020-01-10 14:08:48 <patrick2> I have already tried to follow these: https://wiki.alpinelinux.org/wiki/How_to_make_a_custom_ISO_image_with_mkimage
2020-01-10 14:08:55 <patrick2> and these: https://wiki.alpinelinux.org/wiki/How_to_make_a_custom_ISO_image
2020-01-10 14:09:03 <patrick2> But I cannot get these to work.
2020-01-10 14:09:41 <patrick2> My preference is to build the ISO inside a docker container (I am a docker user, but not very profficient at it)
2020-01-10 14:10:25 <patrick2> For this I used input from: https://stackoverflow.com/questions/55324152/create-alpine-linux-iso-from-docker-libburn-permission-denied
2020-01-10 14:11:12 <patrick2> So,  how to continue? How can I get a reliable installation medium with zfs support?
2020-01-10 14:15:04 <AinNero> i just checked the mkimg scripts in the aports
2020-01-10 14:15:10 <AinNero> if there is zfs support already
2020-01-10 14:15:42 <AinNero> and it seems the extended iso has zfs per default
2020-01-10 14:17:49 <patrick2> I am running the extended iso in virtualbox right now, and the command zpool is not available.
2020-01-10 14:18:38 <AinNero> apk add zfs
2020-01-10 14:19:46 <AinNero> its on the iso, even if its not installed into the in-RAM system per default
2020-01-10 14:20:13 <patrick2> oke, i did not realise this. So installing it and lbu-ing it should mean that it works from there onwards
2020-01-10 14:20:31 <patrick2> sounds good!
2020-01-10 14:20:33 <AinNero> yes, the fact that its installed is recorded in the apkovl (what lbu saves to disk)
2020-01-10 14:21:02 <patrick2> hmm... oke.
2020-01-10 14:21:09 <patrick2> sounds good.
2020-01-10 14:21:15 <telmich> After I removed the @testing tag from the repos, I get the following error from apk:
2020-01-10 14:21:18 <telmich> WARNING: The repository tag for world dependency 'corkscrew@testing' does not exist
2020-01-10 14:21:21 <telmich> How can I fix that?
2020-01-10 14:21:38 <AinNero> did you just remove the '@testing' prefix?
2020-01-10 14:21:43 <telmich> yes
2020-01-10 14:21:51 <AinNero> if yes, you can remove the '@testing' suffix from /etc/apk/world
2020-01-10 14:22:05 <telmich> ahhhhh
2020-01-10 14:22:08 <telmich> omg, that is easy
2020-01-10 14:22:14 <AinNero> and run `apk fix` to check if it complains
2020-01-10 14:22:32 <telmich> omg... that was soooooo easy
2020-01-10 14:22:38 <telmich> thanks a lot, AinNero
2020-01-10 14:22:39 <AinNero> btw, corkscrew is largely unmaintained it seems
2020-01-10 14:23:02 <telmich> it's in my somewhat-old-package list, I haven't actually used it for some years
2020-01-10 14:23:38 <AinNero> ah okay. im using BSD netcat as replacement
2020-01-10 14:24:05 <patrick2> so what do you recommend. Run alpine from USB and using lbu for persistency) or install it to a disk and run it from there? I am planning to run most services from a docker container (samba, nextcloud, torrent, backup scripts)
2020-01-10 14:24:26 <telmich> Hm, now I get ERROR: unsatisfiable constraints: jool-modules-virt-5.4.5-r0 (missing): required by: world[jool-modules-virt-5.4.5-r0]
2020-01-10 14:24:37 <telmich> And trying to delete it keeps the same error message
2020-01-10 14:25:17 <telmich> ah, it works when I add the version at the end
2020-01-10 14:26:31 <AinNero> whacky way to work around: edit the world file manually and run apk fix
2020-01-10 14:27:19 <AinNero> personally i prefer to edit my package list in a editor instead of researching which apk command would be appropiate
2020-01-10 14:30:20 <Cogitri> patrick2: I like using a proper disk since I deeply mistrust USBs
2020-01-10 14:31:35 <programmerq> I only use a usb stick for os (not data) configured to minimize writes, and only when I have two of the usb devices set up in a mirror. usb drives DO go bad.
2020-01-10 14:32:19 <programmerq> my NAS has that configuration-- two USB drives in a mirror so I can keep the sata interfaces free for actual disks.
2020-01-10 14:37:00 <mps> atkka: are you still here? remove '$' from the end from /sbin/lbu on line around 124,  '$OPENSSL enc -ciphers | grep "^$ENCRYPTION$" > /dev/null'
2020-01-10 14:37:32 <mps> i.e. "^$ENCRYPTION" not "^$ENCRYPTION$" ($ at the on var)
2020-01-10 14:42:36 <AinNero> mps: could you elaborate that? that looks like it needs to be fixed in mkinitfs
2020-01-10 14:43:32 <mps> yes, it need fix
2020-01-10 14:43:52 <mps>  /sbin/lbu script is in alpine-conf pkg
2020-01-10 14:44:04 <AinNero> ah yes
2020-01-10 14:44:57 <mps> looks line 'openssl enc -ciphers' changed format from one cipher on one line to wide
2020-01-10 14:45:43 <mps> looks like*
2020-01-10 14:45:44 <AinNero> nice find tho
2020-01-10 14:50:47 <patrick2> I have a xigmanas running now. It runs from usb and it works great... as in... when the usb failes, I get another USB, dd the iso image onto it, put it back, boot it up, copy the 1 config file... and the whole system is up and running again. The config file is king in that setup.
2020-01-10 14:51:30 <patrick2> I would love a similar setup with a (alpine) linux backend instead of freebsd. Mainly because of containers and me being more at home in a linux environment.
2020-01-10 14:52:35 <patrick2> That said. I might go and install alpine on an SSD. Do you recommend to "format" the ssd with zfs as well?
2020-01-10 14:53:20 <patrick2> otherwise I would probably use xfs...
2020-01-10 14:53:42 <patrick2> any other good advice for running a setup like I described with alpine?
2020-01-10 14:56:52 <Cogitri> Well, not much to do wrong there I suppose
2020-01-10 14:57:08 <Cogitri> I just used ZFS for me / as well, but using XFS or smth would be fine too I suppose
2020-01-10 14:58:06 <patrick2> a lovely function of xigmanas was getting an e-mail when SMART disk data indicated that it disk might fail sometime in the near future. How do you implement this in alpine?
2020-01-10 14:58:54 <Cogitri> I did it like this: https://wiki.archlinux.org/index.php/S.M.A.R.T.#Notifying_potential_problems
2020-01-10 14:59:33 <patrick2> yes, the arch wiki is incredible (arch is my daily driver).
2020-01-10 15:06:06 <patrick2> when creating partitions (boot, root, other) and formatting the root (/dev/sda2) as ZFS, do I still need to add the ashift=12 option?
2020-01-10 15:09:50 <Cogitri> Not 100% sure, I just do it
2020-01-10 15:10:19 <patrick2> I will ask at #zfs if it exists
2020-01-10 15:10:25 <Cogitri> I usually just follow the ZFS native encryptio thingie I did a while ago
2020-01-10 15:10:30 <Cogitri> It still exists for sure
2020-01-10 15:10:35 <Cogitri> And it won't do harm
2020-01-10 15:10:49 <patrick2> will it not? I am not super sure what it does really
2020-01-10 15:11:01 <Cogitri> It sets the block size
2020-01-10 15:11:26 <Cogitri> Some drives have 512 Byte blocks but emulate 4KByte blocks (or was it the other way around?)
2020-01-10 15:11:40 <Cogitri> And if you use the emulated size you get subpar performance
2020-01-10 15:13:19 <patrick2> naah, it has to do with alighnment... I am sure of that... I must look into it
2020-01-10 15:14:53 <patrick2> and then you are right anyway
2020-01-10 15:15:05 <patrick2> hahaha... yes it is confusing (to me at least)
2020-01-10 15:19:30 <patrick2> Oke, so I followed this tutorial (in virtualbox): https://wiki.alpinelinux.org/wiki/Setting_up_ZFS_on_LUKS
2020-01-10 15:19:50 <patrick2> and I skipped everythong related to LUKS. I did this twice by now to check for faults of my side...
2020-01-10 15:20:12 <patrick2> Still, I end up with an error: Missing operating system.
2020-01-10 15:20:49 <patrick2> The only part of that tutorial which does not work is the umounting of zfs, since it tells me it cannot do that.
2020-01-10 15:20:58 <patrick2> why pointers?
2020-01-10 15:21:04 <markand> uwsgi fails to start but does not say why, furthermore, a process uwsgi is still running anyway
2020-01-10 15:21:07 <markand> ¯\_(ツ)_/¯
2020-01-10 15:21:46 <patrick2> I mean: any pointers?
2020-01-10 15:22:51 <patrick2> ow and the update-extlinux gives errors (but the wiki tells me to expect those and ignore them, and yes... I did this in the chroot env)
2020-01-10 15:40:39 <Cogitri> I just use rEFInd
2020-01-10 15:40:48 <patrick2> again no success... does anybody know a good resource for installing alpine on zfs?
2020-01-10 15:42:09 <patrick2> I would love to be able to use snapshots on my os.
2020-01-10 15:42:27 <patrick2> what is rEFInd
2020-01-10 15:43:49 <mps> patrick2: you are chatting with one who know very well zfs on alpine, i.e. Cogitri
2020-01-10 15:43:53 <patrick2> oke, but I do not use efi
2020-01-10 15:53:45 <minus> i'm using alpine in diskless mode on a raspi, but i want to install more packages than RAM can fit; is there any way that's possible?
2020-01-10 16:04:00 <Cogitri> Well, I haven't ever used the default Alpine bootloader so I can't help you with that, patrick2
2020-01-10 16:04:03 <tehcloud> ramdisk?
2020-01-10 16:04:23 <tehcloud> lol, yeah... you're not gonna fit more in there than what you have available :p
2020-01-10 16:04:23 <Cogitri> I'd personally either use rEFInd on EFI or GRUB on BIOS I guess
2020-01-10 16:05:00 <Cogitri> I follow https://wiki.alpinelinux.org/wiki/Setting_up_ZFS_with_native_encryption since I wrote that
2020-01-10 16:05:05 <Cogitri> Actually, let me update that
2020-01-10 16:06:38 <patrick2> I followed the instructions on that link as well... but it does not boot.
2020-01-10 16:06:59 <patrick2> I will check later for your updates.... I got to go cook some food for the kids :-)
2020-01-10 16:07:07 <patrick2> Thanks fror now!!!!
2020-01-10 16:11:50 <Cogitri> Updated that. Define not booting please, the bootloader doesn't show up?
2020-01-10 16:11:57 <Cogitri> Glad to be of help :)
2020-01-10 16:12:02 <Cogitri> I guess I should eat something too
2020-01-10 17:49:09 <Kamilion> howdy folks, I just downloaded 3.11.2 amd64 standard ISO and launched it in virtualbox. "apk add python3" says "unsatisfyable constraints" now, and I don't recall seeing this error before.
2020-01-10 17:49:58 <paper_> apk update?
2020-01-10 17:50:11 <Kamilion> i did, says 65 distinct packages available.
2020-01-10 17:50:31 <Kamilion> tried apk upgrade too. didn't do anything.
2020-01-10 17:50:51 <Kamilion> i'm usually a debian/ubuntu guy, so this is a total headscratcher.
2020-01-10 17:51:31 <_ikke_> did you add repositories to /etc/apk/repositories?
2020-01-10 17:51:40 <_ikke_> What is the actual error message?
2020-01-10 17:51:59 <_ikke_> Tried a 3.11 docker container, and it works there
2020-01-10 17:52:07 <Kamilion> http://puu.sh/EYqoY/49a5f93f09.png
2020-01-10 17:52:22 <_ikke_> ah
2020-01-10 17:52:23 <Kamilion> add repositories? I don't recall ever doing that in the past.
2020-01-10 17:52:26 <_ikke_> You are missing repositories
2020-01-10 17:52:28 <Kamilion> is that something new?
2020-01-10 17:52:31 <_ikke_> no
2020-01-10 17:52:35 <_ikke_> did you run setup-alpine?
2020-01-10 17:52:44 <Kamilion> no, this is just the iso in vbox, as I said
2020-01-10 17:53:03 <tehcloud> it looks a lot like your only source is the ISO
2020-01-10 17:53:10 <_ikke_> yes
2020-01-10 17:53:41 <Kamilion> aha, didn't realize I had to run setup-alpine even though I did not want to install
2020-01-10 17:53:45 <Kamilion> thanks.
2020-01-10 17:54:06 <_ikke_> well, the goal of setup-alpine is to install
2020-01-10 17:54:07 <Kamilion> yeah, that got it. Just hit enter a bunch of times
2020-01-10 17:54:52 <Kamilion> therre, finally got what I needed.
2020-01-10 17:54:53 <Kamilion> http://puu.sh/EYqs2/c81893c419.png
2020-01-10 17:54:54 <Kamilion> Thank you.
2020-01-10 17:55:14 <_ikke_> np
2020-01-10 17:55:21 <Kamilion> no libdl.
2020-01-10 17:55:34 <_ikke_> setup-apkrepos if you just want to setup the mirrors
2020-01-10 17:55:38 <Kamilion> can this python3.8 load C extensions through ffi or not?
2020-01-10 17:55:58 <_ikke_> I have no idea myself
2020-01-10 17:56:29 <Kamilion> Even 'I don't know' is a worthwhile response at the moment. Appreciate it. :)
2020-01-10 17:57:24 <telmich> Kamilion: what is ffi?
2020-01-10 17:57:34 <_ikke_> foreign function interface
2020-01-10 17:57:34 <Kamilion> foriegn function interface
2020-01-10 17:58:04 <Kamilion> i'm trying to get the scrypt C extension working.
2020-01-10 18:00:44 <telmich> thanks guys!
2020-01-10 18:44:20 <markand> it looks like perl scripting for irssi does not work: Irssi: Error loading module perl/fe: Error relocating /usr/lib/irssi/modules/libfe_perl.so: perl_scripts_autorun: symbol not found
2020-01-10 18:46:10 <markand> oh someone already reported it: https://gitlab.alpinelinux.org/alpine/aports/issues/5681
2020-01-10 18:47:04 <mps> markand: /load perl
2020-01-10 18:47:49 <markand> mps, yes it fixed it :) is there something that changed in upstream irssi? I've never needed to do that before (12 years I'm using irssi)
2020-01-10 18:48:41 <mps> not sure is that changed, but iirc you need to tell irssi to load modules
2020-01-10 18:49:28 <mps> on alpine perl module is separate
2020-01-10 18:50:08 <markand> ah yes I've seen that in the APKBUILD
2020-01-10 18:51:36 <mps> I think you know that it should go in startup
2020-01-10 18:52:11 <markand> yes, did it
2020-01-10 18:52:33 <markand> time to migrate my laptop to alpine as well!
2020-01-10 19:12:38 <Cogitri> Good :)
2020-01-10 19:12:40 <Cogitri> Have been using Alpine on my laptop for some time now
2020-01-10 19:13:07 <Cogitri> The only thing annoying me right now is that Kernel 5.4 broke hibernation for me, I think
2020-01-10 19:48:01 <mps> atkka: just found that 'lbu commit -e' error is already reported and there is a solution to what I come from my test
2020-01-10 19:48:05 <mps> https://gitlab.alpinelinux.org/alpine/alpine-conf/issues/10451
2020-01-10 19:58:54 <atkka> mps: thanks a ton!
2020-01-10 20:00:06 <mps> this should be fixed, but I'm not sure if I have right to push in alpine-conf
2020-01-10 20:29:59 <bl4ckb0ne> could I have a review on this please? https://lists.alpinelinux.org/~alpine/aports/patches/3204
2020-01-10 20:30:41 <_ikke_> bl4ckb0ne: let me take a look
2020-01-10 20:31:02 <_ikke_> clandmeter: ^
2020-01-10 20:31:10 <mps> :)
2020-01-10 20:31:11 <_ikke_> happens everytime :)
2020-01-10 20:31:43 <mps> yes, noticed it and is quite annoying
2020-01-10 20:31:48 <clandmeter> ?
2020-01-10 20:32:14 <_ikke_> Name matching a trigger
2020-01-10 20:32:28 <_ikke_> kb0
2020-01-10 20:32:35 <bl4ckb0ne> my name matched a trigger?
2020-01-10 20:32:39 <_ikke_> yes
2020-01-10 20:32:40 <bl4ckb0ne> ah yes
2020-01-10 20:33:04 <clandmeter> heh
2020-01-10 20:33:13 <clandmeter> its a bug in your nickname
2020-01-10 20:34:06 <_ikke_> "You're holding it wrong' :P
2020-01-10 20:34:24 <bl4ckb0ne> D:
2020-01-10 20:34:24 <_ikke_> s/g'/g"
2020-01-10 20:34:51 <clandmeter> _ikke_: do we have an issue for: sudo: setrlimit(RLIMIT_CORE): Operation not permitted?
2020-01-10 20:35:47 <_ikke_> I don't think so
2020-01-10 20:36:25 <clandmeter> i guess im not the only getting that?
2020-01-10 20:36:31 <_ikke_> nope
2020-01-10 20:36:36 <_ikke_> it's happening everywhere
2020-01-10 20:36:37 <mps> nope
2020-01-10 20:37:00 <mps> yes, I see it regularly
2020-01-10 20:37:18 <clandmeter> heh
2020-01-10 20:37:20 <_ikke_> https://ask.fedoraproject.org/t/sudo-setrlimit-rlimit-core-operation-not-permitted/4223
2020-01-10 20:37:31 <clandmeter> gitlab search is really smart
2020-01-10 20:37:34 <clandmeter> toooo smart
2020-01-10 20:37:38 <_ikke_> https://bugzilla.redhat.com/show_bug.cgi?id=1773148
2020-01-10 20:38:04 <_ikke_> Set disable_coredump false
2020-01-10 20:38:14 <_ikke_> (as a workaround)
2020-01-10 20:38:30 <clandmeter> right, but i have too many systems to set that.
2020-01-10 20:38:39 <_ikke_> yes
2020-01-10 20:38:53 <clandmeter> needs to come from upstream aka ncopa ;-)
2020-01-10 20:38:54 <_ikke_> Shall I create an issue about it?
2020-01-10 20:39:00 <mps> i'm accustomed already, and don't care for fix :)
2020-01-10 20:39:08 <clandmeter> yes please
2020-01-10 20:39:09 <_ikke_> I would care for a fix
2020-01-10 20:39:11 <_ikke_> less noise
2020-01-10 20:39:25 <clandmeter> nod
2020-01-10 20:44:51 <_ikke_> https://gitlab.alpinelinux.org/alpine/aports/issues/11122
2020-01-10 21:14:46 <clandmeter> thx!
2020-01-10 21:36:42 <coredumb> Good evening
2020-01-10 21:36:47 <coredumb> And happy new year
2020-01-10 21:38:12 <coredumb> I'm trying to install Alpine on a RPI4
2020-01-10 21:38:41 <coredumb> is there a way to make it have network and SSH by default so that I could install it without a screen ?
2020-01-10 21:40:29 <_ikke_> coredumb: 2 ways: 1) create an APKOVL with the correct settings. 2) create a custom image with the correct settings)
2020-01-10 21:41:05 <coredumb> mmmmh
2020-01-10 21:41:22 <coredumb> I don't even know what device should be configured X_X
2020-01-10 22:33:31 <clandmeter> coredumb: you can
2020-01-10 22:34:09 <clandmeter> there is a much more simple way
2020-01-10 22:34:21 <clandmeter> ssh_key= kernel parameter
2020-01-10 22:50:48 <coredumb> clandmeter: well apparently network isn't even up
2020-01-10 22:51:18 <coredumb> at least DHCP wise
2020-01-10 22:51:31 <coredumb> can it be configured with kernel parameter as well ?
2020-01-10 22:55:58 <coredumb> https://wiki.alpinelinux.org/wiki/Initramfs_init oh I see
2020-01-10 22:56:11 <coredumb> it's supposed to go in cmdline.txt right ?
2020-01-10 22:57:24 <atkka> coredumb: do you happen to have a serial cable that does 3.3v tx/rx?
2020-01-10 22:57:45 <coredumb> atkka: indeed not
2020-01-10 22:57:46 <atkka> that's how I set up pis headless with no network connection
2020-01-10 22:58:15 <atkka> doing it as we speak with a buspirate
2020-01-10 22:58:36 <coredumb> yeah would be so much easier
2020-01-10 22:59:26 <atkka> yeah, 3 wires, tx,rx,gnd to the uart pins is all you need
2020-01-10 22:59:58 <atkka> I am curious how to do it by creating a custom apkvol though
2020-01-10 23:03:50 <coredumb> oh yeah network up \o/
2020-01-10 23:04:15 <coredumb> now have to make an http server somewhere to give the ssh key
2020-01-10 23:04:32 <coredumb> atkka: isn't lbu saving an apkvol to be reused ?
2020-01-10 23:06:29 <coredumb> clandmeter: thx!
2020-01-10 23:07:07 <atkka> coredumb: https://wiki.alpinelinux.org/wiki/Manually_editing_a_existing_apkovl
2020-01-10 23:07:32 <coredumb> yeah need at least one :P
2020-01-10 23:07:49 <atkka> haha true
2020-01-10 23:08:06 <atkka> did you get network and ssh working headless?
2020-01-10 23:08:25 <coredumb> atkka: network is up yes
2020-01-10 23:08:54 <atkka> ethernet, dhcp? what steps did you take?
2020-01-10 23:09:02 <coredumb> ethernet static
2020-01-10 23:09:07 <coredumb> https://wiki.alpinelinux.org/wiki/Initramfs_init
2020-01-10 23:09:18 <coredumb> edited cmdline.txt on the SD card
2020-01-10 23:09:32 <atkka> thanks
2020-01-10 23:09:32 <coredumb> set ip=xxxxx:xxx:xxxx and alpine_repo
2020-01-10 23:09:41 <coredumb> now I need a URL
2020-01-10 23:09:47 <coredumb> for the SSH key
2020-01-10 23:10:42 <atkka> thanks for the info
2020-01-10 23:10:54 <atkka> may come in handy sometime
2020-01-10 23:13:43 <coredumb> ok let's try
2020-01-10 23:14:33 <coredumb> so now got network up
2020-01-10 23:14:51 <coredumb> still ssh refused
2020-01-10 23:15:39 <coredumb> well
2020-01-10 23:15:48 <coredumb> now no idea what's happening >_<
2020-01-10 23:17:41 <coredumb> isn't there a squashfs image I can chroot inside to modify directly?
2020-01-10 23:19:45 <crockwork_> What's the best way to see what processes are using what resources on the network?
2020-01-10 23:20:13 <atkka> crockwork_: what resources are you referring to?
2020-01-10 23:20:56 <atkka> https://github.com/raboof/nethogs
2020-01-10 23:21:08 <atkka> nethogs might be what you're looking for
2020-01-10 23:21:11 <crockwork_> Good point. I just want to know information about the network activity currently happening, such as what applications have an open socket, what port, and if there's activity on there.
2020-01-10 23:21:26 <crockwork_> Oh, it does bandwidth, which is neat to know.
2020-01-10 23:22:08 <alxsdf> how do I submit a patch to the alpine linux aports repositories
2020-01-10 23:22:34 <alxsdf> a package has a missing dependency and I'd like to try and contribute more
2020-01-10 23:22:47 <atkka> crockwork_: nethogs is in community repos
2020-01-10 23:23:30 <crockwork_> yeeeeee.
2020-01-10 23:23:55 <crockwork_> How do people manage their personally-installed stuff, by the way? I feel like just downloading, building, and running make install a shitload will eventually lead to bad juju.
2020-01-10 23:25:21 <alxsdf> I have a folder g/ (git) which stores all the source distributions I've downloaded and installed
2020-01-10 23:26:17 <alxsdf> some things don't write the 'make uninstall' thing so it's difficult to uninstall them :<
2020-01-10 23:27:09 <coredumb> ansible here
2020-01-10 23:27:18 <coredumb> under git
2020-01-10 23:30:33 <alxsdf> you could probably use something like gnu stow to properly take care of and version source distributions that you compile and install
2020-01-10 23:31:09 <alxsdf> I just send a patch to the mailing list or is there another work flow?
2020-01-10 23:33:44 <crockwork_> coredumb: How do you use ansible for that?
2020-01-10 23:34:26 <coredumb> atkka: apparently ip=dhcp doesn't work
2020-01-10 23:34:47 <coredumb> crockwork_: yeah was more talking about "managing" stuff
2020-01-10 23:34:53 <coredumb> make packages for the binaries
2020-01-10 23:34:59 <coredumb> or use docker
2020-01-10 23:35:25 <crockwork_> Huh. Interesting.
2020-01-10 23:36:04 <coredumb> actually my OS has bare minimum installed
2020-01-10 23:36:08 <coredumb> X and WM
2020-01-10 23:36:22 <coredumb> and all applications are in docker images
2020-01-10 23:36:29 <coredumb> on various different distribs
2020-01-10 23:36:33 <crockwork_> No way? That's a lot of overhead, isn't it?
2020-01-10 23:36:37 <crockwork_> Is that how Qubes does it?
2020-01-10 23:37:33 <coredumb> Qubes uses Xen underneath
2020-01-10 23:37:46 <coredumb> there's basically no overehad
2020-01-10 23:37:54 <crockwork_> Are you using qubes?
2020-01-10 23:37:58 <coredumb> no
2020-01-10 23:38:07 <crockwork_> Ah, okay. Did you make it yourself?
2020-01-10 23:38:11 <coredumb> yep
2020-01-10 23:38:21 <crockwork_> That's kinda nifty
2020-01-10 23:38:38 <coredumb> it works
2020-01-10 23:38:58 <coredumb> can make different subnets, to isolate applications and the likes
2020-01-10 23:40:25 <coredumb> so you can use Pulseaudio server from Ubuntu on alpine by sharing the audio device to it
2020-01-10 23:40:36 <coredumb> and things like that
2020-01-10 23:42:12 <coredumb> atkka: any link to the serial cable you use?
2020-01-11 00:08:29 <mps> alxsdf: best option to contribute to alpine is to create merge request on gitlab.alpinelinux.org
2020-01-11 00:09:25 <mps> though you can send patch by 'git send-email' to alpine aports mailing list
2020-01-11 00:10:29 <mps> but gitlab have CI which checks merge request and build it, so it is faster and most developers use it
2020-01-11 00:17:48 <alxsdf> mps: thanks :)
2020-01-11 00:18:23 <mps> np
2020-01-11 00:18:28 <alxsdf> coredumb: do you have documentation on how to set something like that up? :)
2020-01-11 00:50:07 <coredumb> alxsdf: certainly not :D
2020-01-11 00:51:27 <coredumb> alxsdf: still you can find details here: https://git.mauras.ch/salt/void-desktop https://git.mauras.ch/docker/alpinelinux https://git.mauras.ch/docker/voidlinux
2020-01-11 00:53:21 <coredumb> alxsdf: there's actually pretty much everything of my setup there
2020-01-11 01:04:54 <alxsdf> thanks :)
2020-01-11 02:05:17 <atkka> coredumb:
2020-01-11 02:05:33 <atkka> coredumb: https://www.adafruit.com/product/237
2020-01-11 02:05:56 <atkka> I use that personally, but this would probably be better and cheaper
2020-01-11 02:06:10 <atkka> https://www.adafruit.com/product/954
2020-01-11 06:12:25 <reidrankin> Good evening. I've been doing some work adapting alpine for use in high-assurance applications, and I'd like to sound the community out about a proposed change to the kernel config.
2020-01-11 06:17:51 <reidrankin> I'm targeting small appliance platforms, and from where I sit a significant feather in Alpine's cap is the simplicity of its init process. A single engineer can follow the process all the way through from the launch of GRUB to initrd to openrc initializing whatever service they care about, and understand exactly what state the platform is in at any given time and where it's going next.
2020-01-11 06:19:55 <reidrankin> Simplicity is the key to a secure system, so Alpine fits really well with what I'm trying to build.
2020-01-11 06:28:26 <reidrankin> I especially love the nothing-up-my-sleeve factor: any given Alpine system can be represented as a kernel, an initrd, an apkovl and a set of apk packages. Each of those except the apkovl can be hashed and compared to the public known-good values, and the apkovl can be inspected pretty easily. I'm even working on tooling that will produce deterministic tarballs, which can be verified to contain no hidden data.
2020-01-11 06:30:06 <reidrankin> My work does require some small initrd modifications, but the beauty of the Linux initrd format is that I can just tack my changes on the end of the (hash-checkable) file. (I've got some nice deterministic tooling for that already done.)
2020-01-11 06:36:23 <reidrankin> For example, I can add dm-verity at initrd time pretty easily by pulling the appropriate modules out of the modloop file and tacking it on the end of the initrd, and adding the appropriate boot options to get the init script to load them. The resulting initrd can be run through the verifier, which flags anything that doesn't come from upstream, which is all minimal, easily human-verifiable glue code.
2020-01-11 06:38:54 <reidrankin> In this manner, I empower my users to make their own decisions about what software to trust: instead of programming my appliances with a public key and making myself a single-point-of-failure who could compromise any of my customers' devices, the customer signs their own updates... after using the verification tools I'm developing to confirm that the update contains exactly what they think it does and nothing else.
2020-01-11 06:40:07 <reidrankin> Anyhow, that long intro is to explain why I really don't want to just compile my own kernel. There'd be no way for a customer to know I hadn't hidden all sorts of nasty surprises in there.
2020-01-11 06:44:40 <reidrankin> I'm using UEFI/TPM 2.0 platforms, and the whole boot process is intended to get measured into the TPM. I've even come up with a handful-of-lines shell script that lets the init process measure things like the apkovl and apks during the initrd process.
2020-01-11 06:48:16 <reidrankin> Unfortunately, to maintain the chain of trust after handing over to a running system, you need Integrity Measurement Architecture enabled -- which is the ONE THING (I know of) that you can't load as a module. You have to have CONFIG_IMA=y in the kernel or you're out of luck, because of the chicken-and-egg problem: IMA has to be loaded before loading any modules, because part of its job is to measure them.
2020-01-11 06:51:58 <reidrankin> So, while Alpine would be a really nice platform for high-assurance, TPM-enabled appliances, without CONFIG_IMA=y its kernel it really isn't usable in that role. And building a custom kernel totally ruins the whole "nothing-up-my-sleeve" idea.
2020-01-11 06:55:17 <reidrankin> I'll note that multiple other distros, like Ubuntu (12.04.4+: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1244627) and Fedora (28+: https://bugzilla.redhat.com/show_bug.cgi?id=790008) have compiled in IMA (though, obviously, not enabled it by default) for the purpose of supporting these types of scenarios.
2020-01-11 07:01:42 <reidrankin> Also, this wouldn't only be useful for the appliance crowd: my work on this approach started with trying to find uses for Hyper-V's virtual TPM, and I understand Xen has similar capabilities. The ability to use TPM-backed security to continuously check for breaches has been very interesting for the cloud crowd as well, many of which already use Alpine.
2020-01-11 07:03:34 <iggy> open an issue, if it doesn't have a negative impact, I'm sure they will add it
2020-01-11 07:04:39 <markand> hello
2020-01-11 07:04:54 <Cogitri> Hello there
2020-01-11 07:05:10 <markand> the initramfs asks password for my LUKS device but is not listed in /dev/mapper after, failing to boot. any ideas?
2020-01-11 07:05:52 <markand> I have dmcrypt=cryptroot cryptroot=UUID=10619912-5aa0-4698-b058-fdc3477a797c in the kernel command line though
2020-01-11 07:11:35 <reidrankin> What does it say when it asks for the passphrase?
2020-01-11 07:13:01 <markand> Enter passphrase for /dev/nvme0n1p3: :)
2020-01-11 07:14:07 <reidrankin> OK, by that point the dm-crypt module will have been loaded correctly, so devicemapper should be working OK.
2020-01-11 07:14:12 <markand> I have added more information on the mailing list question : https://lists.alpinelinux.org/~alpine/users/%3C6c4e65a2-5d6f-41f5-d7fe-21f3444ceaf8%40malikania.fr%3E
2020-01-11 07:14:24 <markand> what I don't understand is why /dev/mapper is left empty
2020-01-11 07:17:13 <reidrankin> huh
2020-01-11 07:17:43 <reidrankin> sorry, I misunderstood. gimme a another sec
2020-01-11 07:18:45 <markand> after initramfs asked for opening my luks device, it is unable to mount / (No such file or directory) and I guess because it is not listed in /dev/mapper (so something is wrong there)
2020-01-11 07:19:07 <markand> manually calling cryptsetup /dev/nvme0n1p3 cryptroot works though and the partition is listed as /dev/mapper/cryptroot
2020-01-11 07:22:46 <reidrankin> it's supposed to be cryptdm=cryptoroot, not dmcrypt=cryptoroot
2020-01-11 07:22:49 <reidrankin> :P
2020-01-11 07:23:59 <reidrankin> this is what I love about alpine's init system: I can *start* debugging by looking at the source code, because it's small enough I can be sure I'll find the answer
2020-01-11 07:25:43 <reidrankin> in case you're wondering, the mapper name is the second parameter to crypt_activate_by_passphrase(), which is called by nlplug-findfs, which takes the name using its -m option.
2020-01-11 07:26:39 <reidrankin> the init script looks for the "cryptdm" option and does cryptopts="$cryptopts -m ${KOPT_cryptdm}"
2020-01-11 07:27:03 <markand> ah
2020-01-11 07:27:05 <markand> good catch!
2020-01-11 07:27:26 <reidrankin> which will be later used in nlplug-findfs $cryptopts -p /sbin/mdev ${KOPT_debug_init:+-d} $KOPT_root
2020-01-11 07:27:33 <markand> let me try that immediately
2020-01-11 07:28:22 <reidrankin> that said, if it can't find your root partition, I'd expect the UUID in the root option to be wrong
2020-01-11 07:33:03 <coredumb> atkka | https://www.adafruit.com/product/954   < ah looks great
2020-01-11 07:36:30 <reidrankin> beware the lack of a DTR line on that
2020-01-11 07:37:02 <reidrankin> if you want to use it with an Arduino too, it won't reset the device on opening the port like the built-in USB does
2020-01-11 07:37:38 <reidrankin> so it can't enter the bootloader to flash the device
2020-01-11 07:44:11 <markand> reidrankin, thanks it worked, now few things to fix and the install is okay
2020-01-11 07:44:22 <reidrankin> glad I could help!
2020-01-11 07:44:33 <markand> for some reasons / is still mounted ro once openrc finished booting
2020-01-11 08:06:04 <mps> reidrankin: such small and cheap serial cable is quite fine for ARM boards, they don't need DTR (and rest) for serial consoles
2020-01-11 10:36:40 <PhoenixMage> Do subpackages have an implicit dependency on the main package?
2020-01-11 10:52:26 <north1> No
2020-01-11 11:19:58 <markand> now running GNOME on Alpine, great work guys :)
2020-01-11 11:24:56 <armin> markand: is suspend/hibernate working for you?
2020-01-11 11:26:01 <markand> yes, it's thinkpad x1 carbon
2020-01-11 11:26:06 <markand> quite good linux support ;)
2020-01-11 11:26:12 <markand> I just have gdm taking years to start though
2020-01-11 11:26:39 <armin> markand: that's an interesting one to debug i guess, did you check what it tries to do meanwhile?
2020-01-11 11:27:23 <markand> last time I had this was when I used GNOME 2 back in mandrake 10. I remember it was something related to the hostname/dns (yes, weird but it was popular one)
2020-01-11 11:33:41 <Cogitri> Yeah, at some point GDM started taking rather long to start with some recent upgrafe
2020-01-11 11:33:56 <Cogitri> I'll go debug this after exam week
2020-01-11 11:34:08 <Cogitri> Does Xorg work for you?
2020-01-11 11:36:12 <markand> I didn't test yet as I usually prefer wayland
2020-01-11 11:45:19 <markand> quick question regarding the merge of edge in releases branches, does it happend for all packages or at the discretion of maintainers? (I'm looking at autofs for example)
2020-01-11 11:53:31 <Cogitri> I use Wayland too, I think Xorg is broke
2020-01-11 17:19:58 <mehlon> heyo, I'm trying to decrypt a file with gnupg: gpg -d file.gpg but it says gpg: problem with the agent: Permission denied
2020-01-11 17:27:46 <mehlon> I figured it out  https://askubuntu.com/questions/1080204/gpg-problem-with-the-agent-permission-denied
2020-01-11 17:38:34 <paradigm> Does apk have a way to check that a package install is in good shape?  Equivalents for what I'm looking for with other package managers include `dkpg -V`, `rpm -V`, `pacman -Qk`, and `xbps-pkgdb`.
2020-01-11 17:43:03 <_ikke_> paradigm: apk verify --help
2020-01-11 17:43:33 <_ikke_> no, sorry, that verifies an APK
2020-01-11 17:43:54 <paradigm> I made the same mistake when exploring the --help
2020-01-11 17:44:39 <mps> what you mean by 'good shape'
2020-01-11 17:45:26 <paradigm> Query if the installed files have been altered from their expected installed state.  Has anyone `rm`'d or `chmod`'d a file that belongs to the package.
2020-01-11 17:45:43 <mps> ah, understand
2020-01-11 17:45:50 <paradigm> I probably could have phrased it better the first time
2020-01-11 17:51:53 <mps> long time ago I used tripwire and aide for such checks
2020-01-11 17:52:19 <paradigm> I'll give those a look, thanks!
2020-01-11 17:52:56 <mps> used them from RO media, and generate signatures in 'clean' environment
2020-01-11 17:53:21 <mps> not sure if any of integrity pkg is in alpine
2020-01-11 17:53:37 <_ikke_> I think apk fix does it, but fix it at the same time
2020-01-11 17:53:49 <_ikke_> (when you specify a package)
2020-01-11 17:54:38 <mps> I think 'apk fix' don't touch conffiles and init scripts
2020-01-11 17:55:05 <mps> though didn't checked
2020-01-11 17:55:09 <paradigm> Combine with --simulate might do what I want
2020-01-11 17:55:27 <_ikke_> but the package manager does not track changes to config files
2020-01-11 17:56:41 <MY-R> apk manifest <pkg name> will give sha1 sums of files included in package so can use that
2020-01-11 17:56:47 <paradigm> I tried `mv /bin/ls /bin/ls~; apk fix -s busybox` and it just said it was reinstalling busybox.  No notice that /bin/ls wasn't as expected.
2020-01-11 17:58:39 <paradigm> Creating my own with `apk manifest` is certainly an optin
2020-01-11 17:58:57 <mps> paradigm: do you want it for security checking or just consistency checking
2020-01-11 17:59:19 <paradigm> Consistency checking
2020-01-11 18:01:34 <paradigm> `apk manifest` seems to leave out things like symlinks and permissions.  Still, better than nothing.
2020-01-11 18:02:08 <mps> yes
2020-01-11 18:02:27 <mps> maybe new apk will be better at this
2020-01-11 18:05:21 <paradigm> I missed the news that there's work on a new one.  That's quite exciting.  This consistency checking and a way to search for which package provides a file (equivalent to `apt-file`, `dnf provides`, `pacman -F`, etc) are really the only things I am interested in that apk lacks.  Other than that it's been superb in my experience.
2020-01-11 18:06:58 <mps> apk-file ?
2020-01-11 18:06:58 <_ikke_> paradigm: https://lists.alpinelinux.org/~alpine/devel/%3C20191203180717.0016af8a%40vostro%3E
2020-01-11 18:07:13 <_ikke_> mps: tool to tell what package contains a specific file
2020-01-11 18:07:27 <mps> apk add apk-file, I mean
2020-01-11 18:07:58 <paradigm> ERROR: unsatisfiable constraints: apk-file (missing)
2020-01-11 18:08:08 <mps> testing repo
2020-01-11 18:08:43 <mps> no, community, sorry
2020-01-11 18:09:13 <mps> 'apk policy apk-file' => http://dl-cdn.alpinelinux.org/alpine/v3.11/community
2020-01-11 18:10:59 <paradigm> I have no idea how I missed this >.< Thank you!
2020-01-11 18:11:56 <mps> np :)
2020-01-11 18:12:17 <mps> there are a lot of pkgs in alpine for which I don't know
2020-01-11 18:13:13 <mps> few times I even worked to package some thing to alpine and discovered that they are already packaged :)
2020-01-11 18:15:57 <paradigm> The new alpine mailing list item mentions `apk audit` which I wasn't familiar with.  Sure enough, `mv /bin/busybox{,~} && apk audit --system` does what I wanted, although it misses symlink changes.
2020-01-11 18:16:51 <_ikke_> paradigm: ah, I looked at audit and I didn't think it did what you want
2020-01-11 18:17:35 <_ikke_> paradigm: if you talk about busybox symlinks, it's because the symlinks are not part of the package, they are created by a trigger
2020-01-11 18:17:55 <paradigm> I also passed it over when exploring the --help before asking you fine folks.  Happily, `apk audit --system --check-permissions` seems to be spot on for what I wanted.
2020-01-11 18:18:04 <paradigm> That'd explain it
2020-01-11 18:21:40 <MY-R> ah ye audit and I thought I saw it in different distribution :D
2020-01-11 19:24:20 <Invader_Bork> hello how does one create the xdg user folders?
2020-01-11 19:25:04 <Invader_Bork> i mean the xdg-user-dirs-update command
2020-01-11 20:48:02 <gabin_> @Invader_Bork  what do you have in your ~/.config/user-dirs.dirs ?
2020-01-11 20:49:55 <Invader_Bork> gabin_, nothing
2020-01-11 20:53:18 <gabin_> XDG_DESKTOP_DIR="$HOME/Desktop"
2020-01-11 20:53:19 <gabin_> XDG_DOCUMENTS_DIR="$HOME/Documents"
2020-01-11 20:53:21 <gabin_> XDG_DOWNLOAD_DIR="$HOME/Downloads"
2020-01-11 20:53:23 <gabin_> XDG_MUSIC_DIR="$HOME/Music"
2020-01-11 20:53:25 <gabin_> XDG_PICTURES_DIR="$HOME/Pictures"
2020-01-11 20:53:27 <gabin_> XDG_PUBLICSHARE_DIR="$HOME/Public"
2020-01-11 20:53:29 <gabin_> XDG_TEMPLATES_DIR="$HOME/Templates"
2020-01-11 20:53:31 <gabin_> XDG_VIDEOS_DIR="$HOME/Videos"y
2020-01-11 20:53:33 <gabin_> you should put such thing in it and reissue a xdg-users-dirs
2020-01-11 20:54:06 <gabin_> if you don't need eg: PUBLICSHARE, just put ""
2020-01-11 20:56:19 <gabin_> not sure that you don't need something else. I don't use it on alpine :)
2020-01-11 21:10:56 <Invader_Bork> what package contains xdg-users-dirs?
2020-01-11 21:31:54 <mps> how can see gcc default flags
2020-01-12 03:01:35 <alxsdf> Invader_Bork: https://pkgs.alpinelinux.org/contents?file=user-dirs.dirs&path=&name=&branch=edge
2020-01-12 03:01:49 <alxsdf> Invader_Bork: It's not given in a package, see: https://wiki.archlinux.org/index.php/XDG_user_directories
2020-01-12 03:02:38 <alxsdf> no other packages contain the binary xdg-user-dirs either
2020-01-12 03:05:51 <alxsdf> but if you stick 'export' in front of them and source the script in your .mksrc then it'll work fine
2020-01-12 03:06:28 <alxsdf> bear in mind you need uhh XDG_RUNTIME_DIR for some applications which tbh tripped me up
2020-01-12 05:17:22 <doggone> Is there a way to query which repo a package was installed from? For example, show if the packages in /etc/apk/world are from main or community?
2020-01-12 06:39:01 <KatolaZ> doggone: apk info -W FILE
2020-01-12 06:39:09 <KatolaZ> will tell you which package own a file
2020-01-12 06:39:39 <KatolaZ> oh you want to know the repo, sorry
2020-01-12 06:39:41 <KatolaZ> dunno
2020-01-12 08:05:57 <mps> doggone: apk policy apkname
2020-01-12 08:12:56 <doggone> mps: thx!
2020-01-12 08:17:11 <mps> :)
2020-01-12 08:21:13 <doggone> now i've discovered i've got several packages installed that aren't in the repos anymore
2020-01-12 09:01:49 <doggone> Any idea what happened to the php7-pear-* packages since Alpine 3.11?
2020-01-12 09:07:20 <doggone> Seems that the roundcubemail package, which depended on those, also disappeared
2020-01-12 09:12:15 <doggone> Ah, since I wasn't using the repo version of Roundcube anyway, but the upstream php files, I didn't need those packages anyway. They're included in a dir called "vendor", so no biggie.
2020-01-12 09:16:19 <patrick2> Cogitri; Thanks for sharing and updating https://wiki.alpinelinux.org/wiki/Setting_up_ZFS_with_native_encryption
2020-01-12 09:16:42 <patrick2> It does contain parts not in https://wiki.alpinelinux.org/wiki/Setting_up_ZFS_on_LUKS
2020-01-12 09:17:48 <patrick2> I willI will test the instructions you wrote and report back on them. I have a username at the alpine wiki now, and will use the discussion page to make comments and/or suggestions.
2020-01-12 09:22:11 <doggone> Not the most elegant script, but posting it here in case anyone is interested: https://pastebin.com/9NGC2vLT
2020-01-12 09:32:26 <mps> does rustc have switch which is equivalent to -mno-unaligned-access in gcc
2020-01-12 10:28:12 <patrick2> Cogitri. I am following your guide, but things seem to go wrong when installing alpine in a chroot. That description does not work out for me.
2020-01-12 10:28:18 <patrick2> I will look into more later
2020-01-12 10:31:01 <Cogitri> Well, what exactly goes wrong?
2020-01-12 10:31:39 <Cogitri> mps: Don't think so
2020-01-12 10:34:37 <mps> Cogitri: hm, how then build firefox for arm32. right now it faults with unaligned access error
2020-01-12 11:14:38 <patrick2> Cogitri; typing all those commands is a rather complicating factor. Should the alpine-chroot-install script work as well?
2020-01-12 11:19:03 <patrick2> and what is the reason for the ubuntu requirement? Is it really not possible from the alpine-extended iso?
2020-01-12 13:09:33 <patrick2> install guide for zfs ROOT. It said: Install the ZoL and linux-vanilla package: apk add linux-vanilla zfs
2020-01-12 13:09:49 <patrick2> However, there is no linux-vanilla (missing)
2020-01-12 13:10:00 <patrick2> what should this be?
2020-01-12 13:11:25 <patrick2> also, mkinitfs is not installed...
2020-01-12 13:15:55 <MY-R> patrick2, linux-lts
2020-01-12 13:20:16 <patrick2> I figured it out indeed... thanks!
2020-01-12 13:22:03 <mr_ab> Okay, heh, .. I gotta ask, .. Where the heck do these man pages come from such as "man man" using the man-pages apk?  It says they are "IEEE/The Open Group" "POSIX Programmer's Manual"
2020-01-12 13:22:25 <patrick2> and how to I get extlinux installed ?
2020-01-12 13:23:01 <patrick2> oke, it is syslinux
2020-01-12 13:23:16 <mr_ab> I can't find the original copy of them from the posix manual or anything though.  It's sort of weird.
2020-01-12 14:00:04 <misterunknown> Hi guys. I try to get docker working in a LXC container and have some problems. First of all, docker says that 'overlayfs
2020-01-12 14:00:11 <misterunknown> is not available in the kernel.
2020-01-12 14:01:01 <misterunknown> Indeed there is no such module loaded, but it's available in /lib/modules/4.19.76-0-vanilla/kernel/overlayfs
2020-01-12 14:01:23 <_ikke_> misterunknown: you might need to load that on the host
2020-01-12 14:01:33 <misterunknown> But when I try to load the module by "modprobe overlayfs", it says "modprobe: FATAL: Module overlayfs not found in directory /lib/modules/4.19.76-0-vanilla"
2020-01-12 14:02:18 <misterunknown> _ikke_: Do I have to add the module name to one of the files under /lib/modules/4.19.76-0-vanilla/module* first?
2020-01-12 14:02:30 <misterunknown> And if yes, in which one?
2020-01-12 14:03:14 <_ikke_> No, you should not have to modify anything in /lib/modules/*
2020-01-12 14:04:04 <misterunknown> Hm... Weird...
2020-01-12 14:06:21 <misterunknown> _ikke_: Oh, maybe I simply should reboot. I obviously missed a kernel update and didn
2020-01-12 14:06:26 <misterunknown> 't reboot yet.
2020-01-12 14:10:05 <mps> yes, you need reboot if you upgraded kernel
2020-01-12 14:12:48 <Cogitri> patrick2: I guess it might be possible to use alpine-extended (and the guide mentions that it might be possible to use other Isos), it's just that Ubuntu was simplest for me
2020-01-12 14:12:54 <Cogitri> Since it comes with everything included
2020-01-12 14:19:56 <misterunknown> _ikke_: Ok, reboot did it. I was able to load the overlay module.
2020-01-12 14:20:32 <misterunknown> Now I'm getting another error when starting dockerd in my lxc container: failed to start daemon: Devices cgroup isn't mounted
2020-01-12 14:20:55 <mps>  /etc/init.d/cgrops start
2020-01-12 14:21:04 <mps>  /etc/init.d/cgroups start
2020-01-12 14:22:40 <misterunknown> mps: This is already started. I also see the cgroups:
2020-01-12 14:23:23 <misterunknown> https://termbin.com/788j
2020-01-12 14:23:31 <misterunknown> That's the view in the hostsystem
2020-01-12 14:24:24 <misterunknown> mps: Within the container I get this: https://termbin.com/tz60
2020-01-12 14:24:32 <misterunknown> (within the LXC container)
2020-01-12 14:24:56 <misterunknown> I also already set lxc.cgroup.devices.allow = a
2020-01-12 14:25:04 <misterunknown> and lxc.cap.drop =
2020-01-12 14:25:48 <_ikke_> You'll probably need to forward the mountpoint
2020-01-12 14:25:55 <mps> I'm not sure is it possible to run docker in lxc without some tweaks
2020-01-12 14:26:46 <misterunknown> -.- Nevermind... I just had to start /etc/init.d/cgroups within the lxc container, which brings me another stop further...
2020-01-12 14:26:57 <misterunknown> Now I
2020-01-12 14:27:35 <misterunknown> Now I'm getting an error related to initialization of the network controller.
2020-01-12 14:27:57 <mps> does it run (/etc/init.d/cgroups) out of the box in lxc
2020-01-12 14:28:39 <misterunknown> mps: Yes, at last it didn't show any errors, and I see the cgroup mounts with "findmnt" within the container now
2020-01-12 14:29:44 <misterunknown> I know that it is not trivial to run nested containers, especially with docker containers in lxc, but I read some article on the internet which claim it should be possible, and I want to try it myself.
2020-01-12 14:31:34 <mps> I thought to try this but never found strong reason to play with it
2020-01-12 14:54:28 <Guest56620> I just want to save everything with lbu. Which folders should I include to do that?
2020-01-12 14:54:51 <_ikke_> Guest56620: What is everything?
2020-01-12 14:55:08 <Guest56620> Everything that would be stored on the disk usually
2020-01-12 14:55:28 <_ikke_> Why not do a disk install?
2020-01-12 14:55:29 <Guest56620> I just want my system to be loaded into RAM because of performance
2020-01-12 14:55:43 <Guest56620> It's running on a USB sticm
2020-01-12 14:55:47 <Guest56620> stick
2020-01-12 14:55:48 <_ikke_> There are files that are updated regularly, they don't make sense to save with lbu
2020-01-12 14:56:11 <Guest56620> Ok
2020-01-12 14:56:13 <_ikke_> and there are a lot of files that are part of packages
2020-01-12 14:56:32 <_ikke_> Those make no sense to save with lbu
2020-01-12 14:56:55 <Guest56620> Is there any other way to load the whole system into RAM and save it only manually to disk?
2020-01-12 14:57:13 <_ikke_> Guest56620: There used to be a data install target
2020-01-12 14:57:32 <_ikke_> meaning the root filesystem was still in ram, but certain directories (typically /var) was mounted on disk
2020-01-12 14:57:42 <_ikke_> I think that would make more sense
2020-01-12 14:58:21 <Guest56620> Can I install packages to /var?
2020-01-12 14:58:37 <_ikke_> packages are installed to all kinds of directories
2020-01-12 14:58:53 <_ikke_> /usr/share /usr/lib /etc /var and more
2020-01-12 14:59:20 <_ikke_> But in alpine, the list of installed packages is stored in /etc/apk/world, so on boot, it will just install those again
2020-01-12 14:59:40 <_ikke_> so persisting all the programs/application files is redundant
2020-01-12 14:59:45 <Guest56620> But I don't want to install all packages each time I boot
2020-01-12 14:59:55 <Guest56620> Also, dwm is not a package.
2020-01-12 15:00:04 <Guest56620> So it would get lost all the time.
2020-01-12 15:00:34 <_ikke_> Guest56620: apk is really fast, and the package files are cached on the USB, so you barely notice that it's happening
2020-01-12 15:00:51 <_ikke_> Guest56620: https://pkgs.alpinelinux.org/package/edge/main/x86_64/dwm
2020-01-12 15:00:57 <_ikke_> It's available as a package
2020-01-12 15:01:05 <Guest56620> Ok
2020-01-12 15:16:54 <timonp> I switched from mobile to desktop, the guy who wants to run his USB system from RAM but have a way to save all changes manually.
2020-01-12 15:18:11 <timonp> This probably won't work out the way you suggested it. I need Minecraft world saves and stuff.
2020-01-12 15:18:47 <timonp> Is there some other way to back up the entire system from tempfs to USB? Doesn't need to be lbu.
2020-01-12 15:20:33 <chrisb> i think tiny core linux is a good model for running in RAM and saving persistence to USB thumb drive
2020-01-12 15:20:53 <timonp> Yeah but I don't like it, tried it.
2020-01-12 15:21:11 <timonp> Then I tried SliTaz, that just didn't work on my device.
2020-01-12 15:21:51 <chrisb> timonp: ok, but you can see how that distro manages RAM filesystems/persistent filesystems and implement it for apline
2020-01-12 15:21:58 <chrisb> s/apline/alpine/
2020-01-12 15:22:23 <timonp> Good idea, thank you
2020-01-12 15:23:06 <timonp> Do you mean tiny core or alpine?
2020-01-12 15:23:14 <timonp> Sorry, slitaz
2020-01-12 15:23:19 <timonp> not alpine of course
2020-01-12 15:32:22 <timonp> tinycore has filetools, which is like lbu.
2020-01-12 15:34:37 <timonp> slitaz has what i need
2020-01-12 15:34:44 <timonp> tazusb writefs
2020-01-12 15:42:05 <patrick2> Cogitri: I have several issues with the wiki descriptions. I tried to fix most of them, but I end up with a system that is not booting. I tried several times, and wrote down what I did. Could you give it a look and check what I did wrong? The description: https://pastebin.com/8GqLnMij
2020-01-12 15:47:34 <timonp> Ok, this looks hard but doable. I'll do a disk install first and implement the RAM loading later on.
2020-01-12 15:48:00 <timonp> chrisb: thanks
2020-01-12 15:48:31 <okias[m]> Firefox -> I can well reproduce ARMv7 crash. Just by backspacing one letter from URL bar, but sadly no debug to see.
2020-01-12 15:53:43 <mps> okias[m]: which version?
2020-01-12 15:54:23 <okias[m]> latest in aports
2020-01-12 15:54:38 <okias[m]> mps 72.0.1
2020-01-12 15:54:39 <mps> 72.0.1 ?
2020-01-12 15:54:56 <okias[m]> 71.something worked OK on tegra30
2020-01-12 15:55:10 <mps> in my test it crashes during start
2020-01-12 15:55:25 <Cogitri> patrick2: oh
2020-01-12 15:55:27 <mps> 71.0 works fine on exynos
2020-01-12 15:55:35 <okias[m]> https://paste.sr.ht/%7Eokias/a2845604b956fcd02576073b2400188189fa4ca5
2020-01-12 15:55:42 <okias[m]> it seems to be using invalid instruction?
2020-01-12 15:55:57 <okias[m]> since these crashes are repeated in exact same position
2020-01-12 15:55:58 <Cogitri> Could you try w/ root=ZFS=<root volume> on your kernel cmdline?
2020-01-12 15:56:01 <patrick2> Cogitri: wait a second. I rewrote that text to something more clear
2020-01-12 15:56:43 <Cogitri> Oh and add the zfs feature to mkinitfs.conf I guess
2020-01-12 15:57:06 <patrick2> https://pastebin.com/7ADUSnnB
2020-01-12 15:57:14 <mps> okias[m]: yes, aligment error, I rebuilt it with -mno-unaligned-access for gcc/g++ but didn't helped, only got little better dmesg
2020-01-12 15:57:53 <okias[m]> for me it's triggered by writing sezznam.cz into URL list and then going to zz and backspacing z
2020-01-12 15:58:02 <okias[m]> s/URL list/URL bar/
2020-01-12 15:59:08 <mps> did you started from new and empty ~/.mozilla dir or older one
2020-01-12 16:01:50 <okias[m]> older, I can do wipe
2020-01-12 16:01:53 <okias[m]> if you want
2020-01-12 16:04:12 <patrick2> Cogitri: I changed the update-extlinux.conf, the mkinitfs.conf already contained zfs in the features list (that was on the wiki).
2020-01-12 16:05:34 <mps> okias[m]: I don't think we have 71.0 in edge repo
2020-01-12 16:06:39 <patrick2> Cogitri: the root=ZFS=<root volumne> was the trick. It is booting now. I will update my notes on the pastebin
2020-01-12 16:08:08 <mps> okias[m]: and, no, you don't need to wipe your dir, I just asked could be older dir related
2020-01-12 16:09:03 <mps> I wiped it and tried but that didn't helped
2020-01-12 16:09:59 <okias[m]> mps I have development device, so it's from 71. :P
2020-01-12 16:10:01 <Cogitri> patrick2: Nice
2020-01-12 16:10:49 <patrick2> Thanks Cogitri. I updated my description on pastebin and will put a link to it in the wiki discussion section (so others might use it as well).
2020-01-12 16:11:20 <mps> well, I did 'tar cf old-mozilla.tar .mozilla' so can revert to 71.0
2020-01-12 16:11:31 <Cogitri> Ah, wouldn't incorporating it into the Wiki page be best, patrick2?
2020-01-12 16:12:53 <patrick2> well, I prefer it if somebody would give it a good look. The issue is that it touches also the Alpine_Linux_in_a_chroot wiki and I find it rather strange that there are faults in that part. So I start with leaving the link around.
2020-01-12 16:15:28 <patrick2> it is time to make some food for the kids again... Thanks again for your help. Tonight I will check if my booting system manages to run as intended as well.
2020-01-12 16:20:50 <patrick2> edited /etc/network/interfaces, restart networking, getting the error: ioctl 0x8913 failed: No such device.
2020-01-12 16:21:17 <patrick2> It is in the exact same VM in which I just performed the installation and there I had network without a problem
2020-01-12 16:21:20 <patrick2> Any honts?
2020-01-12 16:22:19 <patrick2> hints
2020-01-12 16:22:28 <okias[m]> mps: (feeling playful) would you consider build Firefox 73 (next) with this patch -> https://phabricator.services.mozilla.com/D58594 ? It's only 8 years old, but it's prepared for 74 :P
2020-01-12 16:23:56 <mps> okias[m]: why not :)
2020-01-12 16:24:10 <okias[m]> it should drop dependency on startup-notifications :)
2020-01-12 16:24:22 <okias[m]> also it adding one line and removing 100 :D
2020-01-12 16:26:09 <mps> ah, they started to make better firefox :D
2020-01-12 16:30:08 <patrick2> any idea on why networking does not detect my eth0?
2020-01-12 16:30:19 <_ikke_> Driver / module missing?
2020-01-12 16:30:23 <patrick2> which perfectly worked during installation from the livecd?
2020-01-12 16:32:29 <_ikke_> patrick2: does ip link show it?
2020-01-12 16:32:44 <patrick2> no, it does not show it
2020-01-12 16:33:17 <patrick2> it is the default virtualbox network card. I think those are supported quite well in the kernel.
2020-01-12 16:33:44 <patrick2> I also added a different network card (from virtualbox), and that does also not show up
2020-01-12 17:07:45 <MY-R> patrick2, all should work just fine but from your paste it all looks too messy
2020-01-12 17:09:04 <MY-R> patrick2, boot alpine image/iso and after login as root just do: setup-interfaces; ifup eth0
2020-01-12 17:10:24 <MY-R> or do that with running alpine
2020-01-12 17:23:41 <_ikke_> setup-interfaces is not going to work if the interface does not exist
2020-01-12 17:47:07 <MY-R> patrick2, maybe try run this command and reboot "rc-update add hwdrivers sysinit"
2020-01-12 18:07:13 <timonp> I rebooted and all of a sudden, the login prompt was weird, like this: https://media.discordapp.net/attachments/403741067335434240/665973984923222036/IMG_20200112_184152_compress66.jpg?width=436&height=551
2020-01-12 18:08:05 <timonp> Also, I couldn't connect to WiFi anymore although I followed this guide exactly and it did work before: https://wiki.alpinelinux.org/wiki/Connecting_to_a_wireless_access_point
2020-01-12 18:08:58 <timonp> Do you know how to fix that? Because I googled both problems and didn't find anything.
2020-01-12 18:48:04 <markand> someone knows why an SDL fails to create a window with "GLX is not supported"?
2020-01-12 18:48:12 <markand> it opens correctly if I use SDL_VIDEODRIVER=wayland though
2020-01-12 18:53:06 <markand> in fact glxgears also fails to open
2020-01-12 18:53:24 <markand> Error: couldn't get an RGB, Double-buffered visual
2020-01-12 18:53:53 <north1> Using wayland DE/sway ?
2020-01-12 18:55:25 <Cogitri> I had that on GNOME Wayland too but after doing something it magically fixed itself and I have no clue how I did that
2020-01-12 18:55:58 <markand> north1, yes GNOME
2020-01-12 18:56:02 <markand> under wayland
2020-01-12 20:54:05 <kubast2> Hey, I have noticed that "dhcp" option in setup-alpine, at least with a rpi1 doesn't work when running over usb0 interface with usb tethering, I set the interface up and used busybox udhcp -i usb0 to get the address from usb0
2020-01-12 20:55:13 <kubast2> Well I got it working, but I am not sure whether the interface shouldn't be set up during the setup-alpine process?
2020-01-12 21:05:50 <kubast2> cool
2020-01-12 22:20:40 <misterunknown> mps, _ikke_ Thank for your inputs, I finally got docker to run in my LXC container.
2020-01-12 22:20:40 <misterunknown> The reason for that is that I use a hostsystem which only funcions as a LXC host. All services run in LXC containers, and I also have a "terminal" container which I use for management issues. Within this container I
2020-01-12 22:20:40 <misterunknown> want to have the ability to develop and test docker images.
2020-01-12 22:34:01 <mps> misterunknown: nice to hear that you managed to make it working :)
2020-01-12 22:53:51 <mehlon> so I'm trying to install stuff from testing (on alpine armv7l) but all I'm getting is stuff like ERROR: fakeroot-1.24-r0: BAD signature
2020-01-12 22:58:18 <kubast2> I see I am trying to mold the rpi b+ image into a sys installation
2020-01-12 22:58:22 <kubast2> via voodoo magic
2020-01-12 22:59:25 <kubast2> so far I just install alpine-base onto my new rootfs
2020-01-12 23:00:02 <kubast2> I will need to examine the initramfs
2020-01-12 23:01:31 <kubast2> let me check if there is a rpi kernel package first tbh
2020-01-12 23:10:02 <mehlon> ah, incidentally, I recently installed alpine on an sd card (sys install). but then I moved the whole thing to a large usb drive instead
2020-01-12 23:18:52 <kubast2>  ./sbin/apk.static -X http://dl-cdn.alpinelinux.org/alpine/latest-stable/main --arch armhf -U --allow-untrusted --root ../test/ add linux-rpi
2020-01-12 23:18:57 <kubast2> ERROR: unsatisfiable constraints:
2020-01-12 23:19:08 <kubast2> Impossible perhaps the archives are incomplete
2020-01-12 23:19:18 <kubast2> http://dl-cdn.alpinelinux.org/alpine/latest-stable/main/armhf/ linux-rpi-5.4.8-r0.apk                             07-Jan-2020 13:42            24383960
2020-01-12 23:19:32 <kubast2>   linux-rpi (missing):     required by: world[linux-rpi]
2020-01-12 23:20:05 <kubast2> well I can also unpack the kernel for the time being but
2020-01-12 23:20:09 <kubast2> as in initramfs
2020-01-12 23:20:13 <kubast2> and copy over the lib/modules
2020-01-12 23:22:35 <kubast2> cp -r modprobe.d/ ~/test/etc/ && cp -r modules/ ~/test/lib/ && cp -r firmware/ ~/test/lib/
2020-01-12 23:22:37 <kubast2> I guess I am done
2020-01-12 23:23:06 <kubast2> I just need to set root=UUID=mypartitiontable or root=/dev/mmcblk0p2
2020-01-12 23:24:07 <kubast2> quiet console=tty1 root=/dev/mmcblk0p1
2020-01-12 23:24:23 <kubast2> let's see how it works I have removed the apk folder moved the files from boot/boot to boot
2020-01-12 23:24:31 <kubast2> disabled initramfs
2020-01-12 23:24:37 <kubast2> I think I am ready to go
2020-01-12 23:24:54 <kubast2> ah right
2020-01-12 23:24:58 <kubast2> fstab mount boot
2020-01-12 23:26:49 <kubast2> it won't boot
2020-01-12 23:27:33 <kubast2> and ot lerme; [amocs
2020-01-12 23:27:39 <kubast2> kernel panics
2020-01-12 23:27:53 <kubast2> I need to supply it with init for sure
2020-01-12 23:28:08 <kubast2> and I will need to make sure it has getty spawning
2020-01-12 23:29:29 <kubast2> lrwxrwxrwx 1 root root 22 12-15 20:33 /sbin/init -> ../lib/systemd/systemd
2020-01-12 23:29:30 <kubast2> huh
2020-01-12 23:29:37 <kubast2> ah right
2020-01-12 23:29:50 <kubast2> ls sbin/init -la I am stupid
2020-01-12 23:30:07 <kubast2> busybox init is a different init though
2020-01-12 23:30:25 <kubast2> I thought busybox implements runnit a like thingy not openrc for sure
2020-01-13 01:15:59 <ghg> Hey, how can I run post install scripts for a foreign architecture?
2020-01-13 01:16:12 <ghg> and post install triggers
2020-01-13 01:16:31 <ghg> I am using static apk-tool-static for my host architecture
2020-01-13 01:17:24 <ghg> and then use the --arch armhf and --root rootfs-armhf/ --initdb add alpine-base switch
2020-01-13 01:18:03 <ghg> I suppose I can conduct a list of failed .triggers and post installs and run them seperately?
2020-01-13 01:18:28 <ghg> woth certain amount of caution of course
2020-01-13 04:21:43 <dngray[m]> i'm trying to make the sockd service only start if openvpn is started which brings up tun0
2020-01-13 04:21:50 <dngray[m]> otherwise i get this error https://termbin.com/1kmos
2020-01-13 04:22:42 <dngray[m]> i tried doing this https://wiki.gentoo.org/wiki/OpenRC#Dependency_behavior
2020-01-13 04:25:17 <dngray[m]> echo 'rc_need="!net net.tun0"' >> /etc/conf.d/sockd
2020-01-13 04:25:19 <dngray[m]> ln -s sockd /etc/init.d/sockd.tun0
2020-01-13 04:26:38 <dngray[m]> but now i get this other error instead https://termbin.com/qi5s
2020-01-13 04:35:28 <reidrankin> dngray: I think your main problem is that the virtual service net.tun0 doesn't exist because tun0 isn't an interface yet
2020-01-13 04:36:25 <reidrankin> but I did notice that the added rc_need line in /etc/conf.d/sockd is from a different example than the extra symlink was
2020-01-13 04:37:15 <reidrankin> so I'm pretty sure that symlink shouldn't be there
2020-01-13 04:59:21 <reidrankin> dngray: you need to do ln -s networking /etc/init.d/net.tun0
2020-01-13 05:00:58 <reidrankin> looks like alpine's networking script supports that style of adapter-specific service, but you have to create the symlinks for them yourself
2020-01-13 05:01:21 <reidrankin> I gather that's not true on Gentoo, but I've never run it myself so IDK
2020-01-13 05:02:35 <reidrankin> also, you should use wireguard instead :P
2020-01-13 05:04:12 <dngray[m]> right
2020-01-13 05:04:37 <reidrankin> way better performance, even better security, and it's a stateless design - no central daemon that can drop the connection and cause your traffic to get exposed
2020-01-13 05:05:18 <dngray[m]> reidrankin: in this case i have nftables configured so that no trafic is exposed if it goes down
2020-01-13 05:05:45 <reidrankin> yeah, but the point is that you've gotta bother with stuff like that
2020-01-13 05:05:48 <dngray[m]> i think my VPN provider was in the process of setting it up
2020-01-13 05:06:30 <reidrankin> openvpn is fine, I used it for many years, but it's just strictly worse now
2020-01-13 05:06:30 <dngray[m]> i'll probably switch over when they announce it as stable
2020-01-13 05:07:10 <dngray[m]> reidrankin: do i still need: echo 'rc_need="!net net.tun0"' >> /etc/conf.d/sockd
2020-01-13 05:07:35 <reidrankin> I've been following that, they're looking to get it mainlined into the kernel, which will be when they call it stable
2020-01-13 05:08:15 <dngray[m]> yeah no doubt
2020-01-13 05:08:24 <dngray[m]> i do like wireguard though
2020-01-13 05:08:34 <reidrankin> still, I've personally audited the code (one of their goals is to have a small enough codebase that a single dev can do that) and I'd trust a snapshot build of wireguard over a stable build of anything else any day
2020-01-13 05:09:26 <reidrankin> Jason's writes some of the highest-quality code I've ever seen
2020-01-13 05:10:02 <reidrankin> you need rc_need="net.tun0" at least
2020-01-13 05:10:22 <reidrankin> the '!net' bit was used in that example to remove the dependency on the general 'net' service
2020-01-13 05:10:34 <dngray[m]> okay ill oh
2020-01-13 05:10:41 <dngray[m]> so i better make sure to do
2020-01-13 05:10:45 <dngray[m]> 'rc_need="net net.tun0"' >> /etc/conf.d/sockd
2020-01-13 05:12:02 <reidrankin> the !net bit was specific to the sshd example on the webpage, I think; i'd leave it off. you don't need a dependency on net, net.tun0 will take care of that.
2020-01-13 05:12:27 <dngray[m]> oh okay
2020-01-13 05:12:40 <dngray[m]> echo 'rc_need="net.tun0"' >> /etc/conf.d/sockd
2020-01-13 05:12:46 <dngray[m]> then
2020-01-13 05:12:48 <dngray[m]> :P
2020-01-13 05:12:57 <reidrankin> yep
2020-01-13 05:13:04 <dngray[m]> that should work nicely then.
2020-01-13 05:13:30 <dngray[m]> basically i use sockd, because i send all my dnscrypt-proxy traffic into it
2020-01-13 05:13:49 <dngray[m]> which then goes out over the VPN as opposed to going out directly to my ISP
2020-01-13 05:14:22 <dngray[m]> https://wiki.alpinelinux.org/wiki/Linux_Router_with_VPN_on_a_Raspberry_Pi#.2Fetc.2Fsockd.conf
2020-01-13 05:14:25 <reidrankin> why use dnscrypt if you're sending it over a VPN anyway?
2020-01-13 05:14:31 <dngray[m]> at some point i need to clean up that wiki article
2020-01-13 05:14:48 <reidrankin> or, rather, why send it over the VPN if it's already run through dnscrypt?
2020-01-13 05:14:53 <dngray[m]> well dnscrypt provides authentication
2020-01-13 05:15:07 <dngray[m]> well why not :P
2020-01-13 05:15:23 <dngray[m]> also all adresses that are looked up are cached by unbound
2020-01-13 05:15:37 <dngray[m]> so that means fast lookups for things that have been looked up before
2020-01-13 05:16:23 <reidrankin> you've probably got to resolve the address of the VPN server before connecting openvpn
2020-01-13 05:16:52 <dngray[m]> nah they're hardcoded IPs in the config
2020-01-13 05:17:47 <reidrankin> nice
2020-01-13 05:18:40 <dngray[m]> what is annoying is sites that use extremely low TTL on DNS
2020-01-13 05:19:45 <reidrankin> are you redirecting everything else via the VPN too?
2020-01-13 05:21:55 <dngray[m]> https://00f.net/2019/11/03/stop-using-low-dns-ttls/
2020-01-13 05:22:01 <dngray[m]> reidrankin: yes depending on the source VLAN
2020-01-13 05:22:33 <dngray[m]> if it's a host on VLAN2 then yes it goes directly out my WAN interface, if it's a host on VLAN3 then it will go directly out the VPN
2020-01-13 05:22:44 <dngray[m]> https://wiki.alpinelinux.org/wiki/Linux_Router_with_VPN_on_a_Raspberry_Pi_(IPv6) is how its setup
2020-01-13 05:23:11 <dngray[m]> except i am not using iptables/ip6tables, i went over to nftables some time ago
2020-01-13 05:24:18 <reidrankin> I've got a similar setup, in that I my guest wifi is open but runs through the VPN (so their potentially nefarious behavior doesn't get my home IP on all the blacklists)
2020-01-13 05:24:30 <reidrankin> nftables is way better, yeah
2020-01-13 05:24:43 <reidrankin> especially with the 'inet' address type that does v4/v6 all in one
2020-01-13 05:25:38 <reidrankin> do you have a native v6 address?
2020-01-13 05:27:12 <reidrankin> i use AzireVPN, they have wireguard service, but almost as importantly to me the wireguard connection gets a publically-routable IPv6 address
2020-01-13 05:28:37 <reidrankin> well, a /64, so with a bit of OpenVPN hackery to make it stop complaining about handing out a /72 to each client I've now got native v6 on all my machines
2020-01-13 05:28:49 <dngray[m]> AzireVPN i was keen to add to privacytools.io after they meet our requirements (must have external auditing)
2020-01-13 05:29:21 <dngray[m]> apparently its going to happen in the future when they raise the money to do it
2020-01-13 05:35:26 <reidrankin> neat site. I'll take a look at Mullvad as well.
2020-01-13 05:36:22 <dngray[m]> heh
2020-01-13 05:36:25 <dngray[m]> > If you use a local DNS cache such as dnscrypt-proxy that allows minimum TTLs to be set, use that feature. This is okay. Nothing bad will happen. Set that minimum TTL to something between 40 minutes (2400 seconds) and 1 hour. This is a perfectly reasonable range.
2020-01-13 05:37:04 <dngray[m]> was mentioned in: https://00f.net/2019/11/03/stop-using-low-dns-ttls/
2020-01-13 05:42:26 <dngray[m]> i wonder if it's possible to run a one-shot telnet command
2020-01-13 05:44:37 <dngray[m]> i looked at getting line stats via SNMP from my modem, unfortunately it doesn't seem to expose that
2020-01-13 05:44:48 <dngray[m]> but if i telnet into it and run "adsl show info"
2020-01-13 05:45:02 <dngray[m]> i can get those stats in a format i could regex
2020-01-13 05:45:25 <reidrankin> i think you'll need something like expect-lite
2020-01-13 05:46:42 <dngray[m]> mm you might be right
2020-01-13 05:47:20 <reidrankin> expect-lite is a joy to use, but obviously it has to be installed first
2020-01-13 05:47:55 <reidrankin> looks like alpine only has the original expect as a package
2020-01-13 05:48:28 <reidrankin> it's been a while since I used it, but it'll work too, just less intuitively
2020-01-13 05:49:21 <dngray[m]> yeah i might just do it on my archlinux workstation, and then package it up for alpinelinux
2020-01-13 05:52:20 <dngray[m]> is on ubuntu though https://packages.ubuntu.com/eoan/expect-lite https://packages.ubuntu.com/eoan/expect
2020-01-13 10:00:49 <kubast2> Hey, what does alpine do(related to openrc) during instllation on a target rootfs?
2020-01-13 10:01:21 <kubast2> As in what steps should I partake to set a default mode/add agetty to default boot mode ?
2020-01-13 10:02:22 <kubast2> do I just link a service /etc/init,d/agetty to /etc/runlevels/default/agetty.tty1
2020-01-13 10:04:07 <mps>  /etc/inittab
2020-01-13 10:04:39 <kubast2> https://termbin.com/ibq9
2020-01-13 10:06:17 <mps> hm, 'ttyS0::respawn:/sbin/agetty -L 115200 ttyS0 vt100 --login-pause --autologin root --noclear'
2020-01-13 10:06:33 <mps> sample
2020-01-13 10:35:54 <misterunknown> Hi. Short question: What exactly does "-S" with adduser?
2020-01-13 10:36:01 <kubast2> mps enabled ttyS0 ( I have a disabled serial port right now) and added to tty1-6  --login-pause --autologin root --noclear from what I can see openrc boots "Starting boot runlevel" then "Starting default runlevel" all I see is a blinking "-" , the tty shows the inputs I make but nothing more
2020-01-13 10:36:07 <kubast2> adduser --help
2020-01-13 10:36:33 <misterunknown> kubast2: Yes, it says "Create system user", but what difference is between a system user and a "normal" user?
2020-01-13 10:36:40 <misterunknown> Any differences in /etc/passwd?
2020-01-13 10:36:55 <mps> kubast2: this was just example, you should adapt it to your needs, and not just copy it
2020-01-13 10:37:00 <kubast2> ah
2020-01-13 10:37:48 <kubast2> iirc system users don't get home dirs by default and are tied to a different UID and GID range misterunknown
2020-01-13 10:39:44 <mps> 'adduser -h' => -H     Don't create home directory
2020-01-13 11:03:50 <misterunknown> mps: Ok, so -S only affects the UID/GID, as long as it is not explicitly given
2020-01-13 11:06:36 <mps> misterunknown: yes, -S is for creating system users
2020-01-13 13:23:43 <ebelair> Hi everyone, I'm facing issue while installing a library from dockerFile Debian based image on alpine image
2020-01-13 13:24:34 <ebelair> To avoid flooding the channel, I've created a stack overflow question here: https://stackoverflow.com/questions/59717155/installing-firebird-client-and-php-library-on-alpine-docker-image
2020-01-13 13:24:50 <Cogitri> So what's the problem?
2020-01-13 13:25:40 <ebelair> After copied the firebird client files into /usr/lib the system does not find the firebird lib
2020-01-13 13:26:21 <ebelair> I'm new to alpine so I don't really know how to migrate from debian to alpine
2020-01-13 13:27:15 <ebelair> So the command `docker-php-ext-install pdo_firebird` return the following error: configure: error: libfbclient, libgds or libib_util not found!
2020-01-13 13:27:45 <ebelair> I don't understand why because I've installed the *.so in the /usr/lib dir
2020-01-13 13:35:05 <ebelair> I'm focusing on libfbclient but I'm wondering if  libgds or libib_util exists on alpine
2020-01-13 13:37:11 <Cogitri> You're probably trying to use binaries/libs compiled against glibc on Alpine
2020-01-13 13:37:16 <Cogitri> Which you can't since we use musl
2020-01-13 13:38:13 <afontain[m]> 
ACTION sent a long message:  < https://matrix.org/_matrix/media/r0/download/matrix.org/pLJqWixlqxvVQSsTIHJuqgyw >

2020-01-13 13:38:31 <afontain[m]> firebird is not packaged
2020-01-13 13:39:29 <afontain[m]> I'm not sure if it also needs firebird to work for you, but in that case, you would need to package firebird
2020-01-13 13:39:46 <ebelair> ha ok ... bloody work though
2020-01-13 13:40:20 <afontain[m]> I think packaging only the pdo extension would be simple
2020-01-13 13:41:11 <ebelair> I'll try to find a way to do that
2020-01-13 13:42:06 <ebelair> It's behind my knowledge unfortunately ^^"
2020-01-13 13:42:17 <Cogitri> Good luck, I suppose
2020-01-13 13:42:30 <afontain[m]> I was surprised how simple Alpine packages are ^.^
2020-01-13 13:42:32 <ebelair> lol thanks
2020-01-13 13:43:17 <afontain[m]> you could try compiling php7 with `--with-pdo-firebird=shared` or something
2020-01-13 13:46:11 <ebelair> I'll try all of your purposes
2020-01-13 13:48:31 <afontain[m]> the 101 to compile a package is 1. clone aports https://gitlab.alpinelinux.org/alpine/aports  2. `cd community/php7`  3. `abuild -rd`
2020-01-13 13:52:06 <afontain[m]> Something that bugs me with that is that I have to do semthing like `chgrp $(groups|cut -d' ' -f1) /var/cache/distfiles; chmod g+w /var/cache/distfiles` so that abuild can fetch the sources :-/
2020-01-13 13:52:42 <afontain[m]> Cogitri: do you think I should make a bug report for that?
2020-01-13 13:57:51 <ebelair> The problem with  compiling php7 with `--with-pdo-firebird=shared` is that it would try to install the firebird 3 libs that is incompatible with fb 2.5
2020-01-13 13:58:47 <Cogitri> afontain[m]: No, as mentioned on the wiki page you're supposed tondo that
2020-01-13 13:59:00 <Cogitri> You don't want any random user messing with your distfiles
2020-01-13 13:59:15 <afontain[m]> ok, then that's fine I guess
2020-01-13 13:59:43 <afontain[m]> well, it could be owned by group abuild
2020-01-13 13:59:55 <afontain[m]> that's already less random
2020-01-13 14:02:27 <Cogitri> I suppose that could be done if you trust your abuild group
2020-01-13 14:17:09 <ebelair> The lazy solution would be to switch on php-fpm image :p
2020-01-13 14:21:58 <mps> afontain[m]: yes, alpine pkgs are simple but I have fear they will become more complicated when the more users from other distros pressure for features from their previous distro :)
2020-01-13 14:55:40 <paper_> has someone been successful getting adb to recognize devices?
2020-01-13 15:04:39 <afontain[m]> yes
2020-01-13 15:05:07 <afontain[m]> both a real phone, and (though the device itself didn't work) anbox
2020-01-13 15:05:23 <paper_> really? what's the trick?
2020-01-13 15:05:36 <paper_> I have a udev rules file which used to work on void
2020-01-13 15:05:42 <afontain[m]> I did not have to do anything
2020-01-13 15:06:19 <afontain[m]> I added an udev rule for fastboot though. Maybe adb benifited for it
2020-01-13 15:06:23 <afontain[m]> *from
2020-01-13 15:06:37 <paper_> ok... that means the problem is between the seat and the keyboard
2020-01-13 15:06:49 <paper_> can I have the rule? maybe it will help
2020-01-13 15:09:40 <afontain[m]> SUBSYSTEM=="usb", ATTR{idVendor}=="0bb4", ATTR{idProduct}=="0c01", MODE="0660", GROUP="plugdev", SYMLINK+="android%n"
2020-01-13 15:09:56 <afontain[m]> in /etc/udev/rules.d/51-android.rules
2020-01-13 15:10:21 <paper_> will try, thanks
2020-01-13 15:10:23 <afontain[m]> (I'm in the group plugdev, and the id correspond to what `lsusb` says)
2020-01-13 15:16:50 <paper_> it works <3
2020-01-13 15:17:34 <paper_> thanks a lot
2020-01-13 15:18:32 <paper_> and now it doesn't wtf?
2020-01-13 15:19:01 <paper_> it asked for authorization, so I clicked yes and now it doesn't even recognise it
2020-01-13 15:22:03 <afontain[m]> :-/
2020-01-13 15:24:11 <MY-R> paper_, in other windows/distributions was working fine?
2020-01-13 15:24:33 <paper_> yes
2020-01-13 15:25:39 <MY-R> some of lineageOS versions I had got broken adb...
2020-01-13 15:25:50 <MY-R> paper_, try: adb kill-server; adb usb
2020-01-13 15:26:37 <paper_> no devices/emulators found
2020-01-13 15:26:52 <paper_> I have a Mi A2 Lite - almost stock android
2020-01-13 15:28:50 <MY-R> paper_, and "ADB_TRACE=all adb usb" doent show anything intresting?
2020-01-13 15:29:52 <MY-R> or remove ~/.adb dir to start fresh, reboot phone etc
2020-01-13 15:31:30 <paper_> MY-R: before running the command I plugged my phone in the computer and it asked for verification... but nothing happened after it
2020-01-13 15:31:53 <MY-R> ye and should allow it/remember or something like that
2020-01-13 15:32:04 <paper_> I did
2020-01-13 15:32:34 <paper_> in the command output I didn't find anything interesting
2020-01-13 15:32:35 <MY-R> you can try run it at root to check if that is the issue
2020-01-13 15:32:49 <paper_> will try rebooting
2020-01-13 15:32:55 <paper_> no, I already did that
2020-01-13 15:35:50 <paper_> still nothing after reboot
2020-01-13 15:36:19 <paper_> and there is no such thing as ~/.adb on my system
2020-01-13 15:36:57 <mps> do you have udev running
2020-01-13 15:37:25 <paper_> yes
2020-01-13 15:39:02 <mps> don't know, last time I tried it it worked but that was more than 3 months ago
2020-01-13 15:39:54 <afontain[m]> I could try again if you want
2020-01-13 15:40:23 <paper_> it isn't urgent, but if you have time
2020-01-13 15:42:14 <afontain[m]> 
ACTION sent a long message:  < https://matrix.org/_matrix/media/r0/download/matrix.org/RRrfzskWdjsrZLjBbiWMEwax >

2020-01-13 15:42:21 <afontain[m]> it works for me
2020-01-13 15:43:57 <paper_> ok, I will keep trying
2020-01-13 15:44:09 <paper_> thanks for help
2020-01-13 15:45:34 <afontain[m]> hey, adb shell now works with anbox!
2020-01-13 15:45:35 <afontain[m]> progress!
2020-01-13 15:46:53 <PureTryOut[m]> This is on pmOS?
2020-01-13 15:47:31 <PureTryOut[m]> Damn you're getting far with Anbox, nice work. You should probably take up maintainership once we have it in our repo lol
2020-01-13 15:47:42 <PureTryOut[m]> Also, upstreaming it to Alpine eventually would be nice
2020-01-13 15:49:16 <trasz> how can I run upgrade script (in this case busybox-1.31.1-r8.post-upgrade) by hand?
2020-01-13 15:49:47 <afontain[m]> > This is on pmOS?
2020-01-13 15:49:47 <afontain[m]> I'd consider it on Alpine, that's stock alpine-lts kernel, and everything was built using abuild
2020-01-13 15:50:29 <afontain[m]> trasz: you could get it from the pmaports repository
2020-01-13 15:52:18 <trasz> afontain[m]: isn't pmaports for postmarketOS?
2020-01-13 15:52:19 <PureTryOut[m]> Or just the Alpine repos...
2020-01-13 15:52:41 <afontain[m]> I wanted to say aports, sorry
2020-01-13 15:52:47 <PureTryOut[m]> Lol
2020-01-13 15:52:49 <afontain[m]> https://gitlab.alpinelinux.org/alpine/aports
2020-01-13 15:53:11 <afontain[m]> I nearly did the same mistake earlier :P
2020-01-13 15:54:23 <trasz> afontain[m]: that's it, thanks!
2020-01-13 15:57:37 <trasz> huh, when run by hand it works just fine :-/
2020-01-13 16:00:05 <afontain[m]> (paper_: I have a shell on my phone too)
2020-01-13 16:52:52 <Shared404> I was reading throughthe dual boot install guide on the wiki, I noticed that it recommended ext3. Is this still recommended, or can I use ext4?
2020-01-13 16:54:15 <markand> Shared404, yes no problem so far
2020-01-13 16:55:26 <Shared404> Cool, just wanted to double check before going foward with it. Thanks for the help!
2020-01-13 18:06:31 <programmerq> I want to make an apkbuild for a kernel module. I'm not 100% sure what the best practices are around this. This module upstream utilized dkms, so I think I'll need to first get an apkbuild for dkms as a build dependency. I'm not 100% sure how I should package the module since the build will be kernel specific.
2020-01-13 18:07:30 <Cogitri> You can take a look at other kernel modules we package like ZFS
2020-01-13 18:08:10 <programmerq> looking now
2020-01-13 18:12:55 <programmerq> looks like it simply follows the release version of the kernel, and uses depends/install_if to tie it to that kernel.
2020-01-13 18:14:17 <programmerq> is the alpine build server scripts somewhere public as a reference so I can imitate how alpine builds and publishes packages?
2020-01-13 20:29:44 <nwe> good evening, has anyone get spotify to run on alpine-linux
2020-01-13 20:30:06 <_ikke_> nwe: I don't think that's possible
2020-01-13 20:30:15 <_ikke_> spotify is built against glibc
2020-01-13 20:30:23 <jn__> there's gcompat
2020-01-13 20:30:37 <jn__> nwe: you might want to ask awilfox in #musl
2020-01-13 20:31:08 <nwe> jn__: oh okey I will ask him and see..
2020-01-13 20:32:04 <_ikke_> https://code.foxkit.us/adelie/gcompat
2020-01-13 20:36:18 <Cogitri> Or run it in a Flatpak
2020-01-13 20:38:02 <nwe> flatpak ?
2020-01-13 20:44:56 <Cogitri> https://flathub.org/apps/details/com.spotify.Client
2020-01-13 21:43:21 <velix> There's no package with memtest on Alpine?
2020-01-13 21:43:49 <_ikke_> https://pkgs.alpinelinux.org/contents?file=memtest&path=&name=&branch=edge
2020-01-13 21:44:17 <velix> Crazy
2020-01-13 21:44:24 <velix> That's a damn important tool :)
2020-01-13 21:44:43 <velix> Ah, I could use this: http://dl-8.alpinelinux.org/alpine/v3.0/testing/x86/memtest86%2B-4.20-r3.apk
2020-01-13 21:45:26 <_ikke_> Probably not
2020-01-13 21:45:57 <velix> Yeah, workes.
2020-01-13 21:46:32 <velix> memtester is also no there. I'm giving up with apk and start compiling ANYTHING, like Gentoo in the past :)
2020-01-13 21:47:32 <velix> ah! alpine/edge/testing/x86_64/memtester-4.3.0-r0.apk
2020-01-13 21:48:41 <_ikke_> It should even be easy enough to package things yourself
2020-01-13 21:49:23 <velix> I did, but I cannot find it.
2020-01-13 21:49:59 <velix> The search is missing an "all" on branch
2020-01-13 21:50:10 <velix> It's annoying to look in each branch manually
2020-01-13 21:50:12 <_ikke_> yeah, that was done on purpose
2020-01-13 21:50:17 <_ikke_> for performance reasons
2020-01-13 21:50:21 <velix> branch + repository.
2020-01-13 21:50:37 <velix> Hmm... Okay, next time, I've got a spare service, I will donate it.
2020-01-13 21:50:58 <velix> I'm doing realtime auto complete on ~ 22 million german road names in real time.
2020-01-13 21:51:06 <velix> I think, hosting Alpine search would fit on those machines.
2020-01-13 21:51:24 <_ikke_> It's not due to lack of hardware
2020-01-13 23:02:09 <gjabell> Is it possible to downgrade from edge by setting repos to a stable branch and upgrading?
2020-01-13 23:20:14 <danieli> gjabell: no - it will stick to the highest versions it knows of, which would be the ones currently installed
2020-01-13 23:25:21 <MY-R> danieli, no if will use "apk -i -a upgrade"
2020-01-13 23:25:45 <MY-R> but still reinstalling like that all packages is asking for troubles probably
2020-01-13 23:27:49 <gjabell> Yeah that's kinda what I figured haha
2020-01-14 00:25:19 <okias[m]> Hello guys. "Geary" email client needs gnome-keyring to run
2020-01-14 00:25:38 <okias[m]> otherwise ends with: ![err] 01:22:17.0549 geary[NET|SER|REP|CNV|PER]: application-controller.vala:348: Error opening libsecret: The name org.freedesktop.secrets was not provided by any .service files
2020-01-14 00:25:56 <okias[m]> should I fill bugzilla entry or is someone up to fill dep?
2020-01-14 03:31:12 <iggy> okias[m]: a gitlab issue would probably be the best way to get it fixed
2020-01-14 08:58:58 <misterunknown> Hi guys. I want to create a docker container based on php:7-cli-alpine. To enable the user to change the UID/GID of the php process I use environment variables. Currently I have a script docker-startup.sh which checks if the user entered a UID/GID which already exists, possibly remove the user/group, and then create it with "www-data". I really don't like that solution, so I wondered, if I
2020-01-14 08:59:01 <misterunknown> simply could remove _all_ users except root and www-data from the /etc/passwd file. The image contains only files which are owned by root, and there are no other processes than php running (as www-data).
2020-01-14 10:23:30 <okias[m]> iggy thanks, I send pull directly -> https://gitlab.alpinelinux.org/alpine/aports/merge_requests/3037
2020-01-14 10:26:51 <mps> okias[m]: firefox 72.0.1 still faults even with a patch from url you posted
2020-01-14 10:27:37 <okias[m]> mps it's funny, I just looked at aports bugs to find if there is issue about that
2020-01-14 10:28:02 <okias[m]> mps oh, nope, that patch was unrelated.. It just drops one dependency for Firefox and improve loading
2020-01-14 11:20:05 <opal> Sircbot is case-sensitive when looking for the channel directory for a script to execute. For example, if you have a channel named "#Alpine", sircbot will look for /etc/sircbot.d/#Alpine/*.
2020-01-14 11:20:05 <opal> heh
2020-01-14 11:20:28 <opal> im guessing sircbot "works for alpine's irc purposes" but still that's kind of... not how irc works
2020-01-14 11:20:47 <opal> im sure you all dont need to be told though, just found it amusing
2020-01-14 12:25:50 <ncopa> in retrospective, there are lots of things that sircbot could have done differently
2020-01-14 13:09:20 <Xano> Good afternoon! I'm trying to set up a cron job (the container primarily serves a PHP web app through nginx). I installed busybox-initscripts to have a standard crond running, but I'm not sure how to check the status. All `rc-*` commands, as well as `/etc/init.d/crond status` give me "You are attempting to run an openrc service on a system which openrc did not boot."
2020-01-14 13:10:16 <Xano> The documentation (https://wiki.alpinelinux.org/wiki/Alpine_Linux_Init_System, from 2017) states OpenRC is Alpine's init system yet it's not installed by default
2020-01-14 13:10:32 <_ikke_> Xano: a docker container is not a virtual machine
2020-01-14 13:10:41 <_ikke_> it's just a process that is isolated
2020-01-14 13:10:47 <_ikke_> so there is not init system running
2020-01-14 13:11:07 <_ikke_> just installing busybox-initscripts does not change that
2020-01-14 13:11:41 <Xano> Hmmz, that makes sense.
2020-01-14 13:11:56 <_ikke_> either you need to run some kind of services manager, or you split those up in separate containers (docker-compose)
2020-01-14 13:13:45 <Xano> This was supposed to be a temporary hack, but a separate container feels less hacky now.
2020-01-14 13:14:00 <Xano> _ikke_, Thanks for the explanation and suggestions :)
2020-01-14 13:14:34 <_ikke_> you're welcome
2020-01-14 13:57:49 <markand> any alternative to autofs? it segfaults here :(
2020-01-14 13:58:40 <Cogitri> systemd-automount :^)
2020-01-14 13:58:48 <Cogitri> So fixing autofs is your best choice I think
2020-01-14 14:00:00 <markand> that's probably why it's still in testing/
2020-01-14 14:36:18 <programmerq> https://travis-ci.org/alpinelinux/aports looks like it is no longer utilized perhaps? There's also https://build.alpinelinux.org/ but I get a 404 whenever I click on a link to a build log from pkgs.alpinelinux.org.
2020-01-14 14:36:45 <programmerq> oh, nevermind, I can see some of the build logs.
2020-01-14 14:37:42 <programmerq> I'm still trying to figure out the way that the packages are built and published as a learning exercise.
2020-01-14 14:37:57 <MY-R> programmerq, older logs probably erased (winter cleaning ;))
2020-01-14 15:10:10 <Cogitri> How do I check why a package isn't upgraded again?
2020-01-14 15:11:08 <north1> try to upgrade it forcefully
2020-01-14 15:11:11 <north1> It should error our
2020-01-14 15:15:27 <Cogitri> It just says that it breaks world :/
2020-01-14 15:15:58 <Cogitri> Ah, apparently it just wasn't on the mirror yet
2020-01-14 17:59:39 <Guest30996> How can I install Minecraft?
2020-01-14 18:00:53 <Guest30996> What I did so far is: apk add openjdk8-jre, download and unzip the official Minecraft.tar.gz, try to run the minecraft-launcher binary in it
2020-01-14 18:01:15 <afontain[m]> tl;dr: either there is a flatpak or you'll have trouble
2020-01-14 18:01:27 <Cogitri> Well, what happens if you do that?
2020-01-14 18:01:36 <Cogitri> Probably the easiest option, yes
2020-01-14 18:01:50 <afontain[m]> It's probably an executable linked against the glibc
2020-01-14 18:02:41 <Guest30996> Or is there an old Minecraft.jar?
2020-01-14 18:03:03 <Cogitri> afontain[m]: Considering it's a Java program I doubt that
2020-01-14 18:03:05 <Guest30996> What happened is: /bin/ash: minecraft-launcher: not found
2020-01-14 18:03:27 <afontain[m]> executables that run with the glibc usually don't run on Alpine
2020-01-14 18:03:30 <Guest30996> The Minecraft.jar can still run new versions afaik
2020-01-14 18:03:39 <Guest30996> but it's not on the official website
2020-01-14 18:04:22 <afontain[m]> `ldd minecraft-launcher`
2020-01-14 18:05:43 <Guest30996> Error loading shared library ld-linux-x86-64.so.2: No such file or directory (needed by minecraft-launcher)
2020-01-14 18:06:14 <_ikke_> try java -jar Minecraft.jar
2020-01-14 18:06:25 <programmerq> Guest30996▸ https://pkgs.alpinelinux.org/contents?file=ld-linux-x86-64.so.2&path=&name=&branch=edge&arch=x86_64
2020-01-14 18:06:28 <_ikke_> (you might need to install openjdk8-jre for that)
2020-01-14 18:06:49 <programmerq> (package exists with that file)
2020-01-14 18:07:00 <dalias> you could possibly use gcompat
2020-01-14 18:07:13 <_ikke_> that's what the link of programmerq shows
2020-01-14 18:08:08 <Guest30996> Reinstalled openjdk8-jre, same error
2020-01-14 18:08:23 <_ikke_> Guest30996: did you try java -jar Minecraft.jar?
2020-01-14 18:08:53 <Guest30996> There is no Minecraft.jar, they changed it
2020-01-14 18:09:01 <Guest30996> It's Minecraft.tar.gz now
2020-01-14 18:09:14 <Guest30996> And inside that archive there's a minecraft-launcher binary
2020-01-14 18:09:31 <_ikke_> I guess that's what downloads the actual jar
2020-01-14 18:09:55 <_ikke_> then you need gcompat like programmerq / dalias mentioned
2020-01-14 18:11:03 <Guest30996> Now ldd minecraft-launcher is successful, do you want the output?
2020-01-14 18:11:24 <Guest30996> with gcompat
2020-01-14 18:11:39 <_ikke_> Just try to launch it
2020-01-14 18:11:52 <Guest30996> still not found
2020-01-14 18:12:06 <Guest30996> I just type in minecraft-launcher and press enter, right?
2020-01-14 18:12:51 <Guest30996> Error is /bin/ash: minecraft-launcher: not found
2020-01-14 18:13:08 <Guest30996> Same with ./minecraft-launcher
2020-01-14 18:14:04 <Guest30996> And yes, I'm in the correct directory
2020-01-14 18:14:24 <afontain[m]> `ls ./minecraft-launcher` ?
2020-01-14 18:14:39 <_ikke_> yes, if it's not in PATH you need to use that
2020-01-14 18:15:02 <Guest30996> Output is `./minecraft-launcher`
2020-01-14 18:15:09 <afontain[m]> maybe it hasn't got the executable bit set
2020-01-14 18:15:22 <afontain[m]> chmod +x minecraft-launcher
2020-01-14 18:15:22 <Guest30996> With chmod, right?
2020-01-14 18:15:23 <_ikke_> chmod +x minecraft-launcher
2020-01-14 18:15:25 <Guest30996> thx
2020-01-14 18:15:41 <afontain[m]> then `./…` works
2020-01-14 18:15:51 <Guest30996> still not found
2020-01-14 18:17:05 <Guest30996> `/bin/ash: ./minecraft-launcher: not found`
2020-01-14 18:18:31 <afontain[m]> what does ldd mincraft-launcher says?
2020-01-14 18:21:18 <Guest30996> https://ibb.co/TKnRCS0
2020-01-14 18:21:35 <Guest30996> Didn't want to copy the whole output
2020-01-14 18:25:16 <Guest30996> don't have an irc client over there so I need to do it by hand
2020-01-14 18:25:58 <_ikke_> It's alright
2020-01-14 18:26:06 <_ikke_> What architecture are you on?\
2020-01-14 18:26:45 <Guest30996> x86_64
2020-01-14 18:27:35 <programmerq> brainstorming a bit here-- there are alternate launcher projects out in the wild. You may have more success with one of those. Alternately, you could run the official launcher on a glibc system once, grab the minecraft.jar that it downloads, and then copy that over.
2020-01-14 18:31:06 <Guest30996> Minecraft is installed on this device and I'm doing a find in / now, it may take a while.
2020-01-14 18:32:15 <Guest30996> There is no Minecraft.jar
2020-01-14 18:32:34 <Guest30996> Oh
2020-01-14 18:32:38 <Guest30996> The regex was fucked up
2020-01-14 18:32:41 <Guest30996> Now I found it
2020-01-14 18:35:52 <afontain[m]> great!
2020-01-14 18:52:31 <Guest30996> I got the launcher to run but it just keeps on saying "Preparing...".
2020-01-14 18:53:05 <bl4ckb0ne> did you tried to untar the tar.gz ?
2020-01-14 18:53:20 <Guest30996> Which tar.gz?
2020-01-14 18:53:25 <Guest30996> The one from above?
2020-01-14 18:53:32 <Guest30996> It was always untared
2020-01-14 18:53:51 <Guest30996> Minecraft.tar.gz?
2020-01-14 18:54:23 <bl4ckb0ne> yes
2020-01-14 18:54:35 <Guest30996> Yes, it was unpacked
2020-01-14 18:55:22 <Guest30996> Oh, I just see it threw an exception:
2020-01-14 18:55:48 <Guest30996> There's a duplicate key in some JSON, it's null
2020-01-14 18:56:57 <Guest30996> Maybe the old launcher isn't supported anymore?
2020-01-14 18:57:23 <Guest30996> Because I think this was the old launcher that I had before I downloaded the official one, not a .jar downloaded by the official one
2020-01-14 19:01:44 <bl4ckb0ne> in the arch aur there's two minecraft packages
2020-01-14 19:01:47 <bl4ckb0ne> https://aur.archlinux.org/cgit/aur.git/tree/?h=minecraft
2020-01-14 19:01:51 <bl4ckb0ne> look at this one
2020-01-14 19:10:48 <iggy> misterunknown: there's no reason you couldn't... the existing users are mostly just a convenience (so different users/groups don't end up with wildly different UID/GIDs on every system you have)
2020-01-14 19:16:34 <Guest30996> Ok, I grabbed the up to date Minecraft.jar from the AUR package.
2020-01-14 19:16:38 <Guest30996> Testing it now
2020-01-14 19:18:23 <Guest30996> threw the same exception
2020-01-14 19:18:30 <Guest30996> probably the same file
2020-01-14 19:20:04 <Guest30996> Is it a problem that I run this inside an st shell in dwm?
2020-01-14 19:20:14 <_ikke_> should not matter
2020-01-14 19:21:39 <Guest30996> The bad JSON was parsed when trying to get a complete version list btw
2020-01-14 20:00:05 <Guest30996> So I can't run the official launcher, I can't run Minecraft.jar, I'm officially fucked.
2020-01-14 20:00:51 <Guest30996> maybe the flatpak?
2020-01-14 20:01:03 <Guest30996> wanted to avoid flatpak but seems like i have no other option
2020-01-14 20:09:03 <minus> the wiki is being mean: >New users are not allowed to add ip addresses and phone numbers within 5 hrs of account creation
2020-01-14 20:09:16 <minus> not even adding/editing something with ip addresses/phone numbers
2020-01-14 20:16:17 <_ikke_> something something spam something something
2020-01-14 20:31:09 <Shared404> Hello There. I was reading https://wiki.alpinelinux.org/wiki/Create_a_Bootable_USB and I noticed that it mentions that it is possible to use dd to create a bootable usb. Is this instead of or as well as the instructions lower down the page? Thanks in advance
2020-01-14 20:32:01 <_ikke_> Shared404: if you dd the ISO, you get a read-only image
2020-01-14 20:32:17 <_ikke_> if you just want to install from it, that's fine
2020-01-14 20:32:31 <_ikke_> if you want a run-from-ram with LBU, you need something writable
2020-01-14 20:32:43 <Shared404> Ok, that makes sense.
2020-01-14 20:32:52 <Shared404> Sorry for my bad reading comprehension
2020-01-14 20:33:18 <_ikke_> That page is a bit confusing
2020-01-14 20:34:16 <_ikke_> Shared404: This might be a bit clearer: https://docs.alpinelinux.org/user-handbook/0.1a/Installing/medium.html
2020-01-14 20:36:12 <Shared404> Cool! that page looks like exactly what I needed.
2020-01-14 20:36:22 <Shared404> Many thanks
2020-01-14 21:09:38 <minus> is there a convenient way to work through .apk-new files like pacdiff on arch?
2020-01-14 21:11:03 <_ikke_> update-conf
2020-01-14 21:11:24 <_ikke_> part of alpine-conf
2020-01-14 21:11:48 <minus> lovely, thanks
2020-01-14 21:17:01 <minus> it doesn't offer opening vimdiff <old> <new> though. this does the job: for apknew in $(ag --one-device -g .apk-new /); do vim -d "${apknew%.*}" "$apknew"; done
2020-01-14 21:18:52 <_ikke_> right, that part is missing
2020-01-14 21:44:35 <Shared404> When using setup-bootable can I just use the iso file as the source or do I need an actual cd-rom or usb running alpine already
2020-01-14 21:52:27 <Shared404> Nevermind. I'm retarded.
2020-01-14 21:52:55 <Shared404> For anyone who was wondering, it is fine to use the iso.
2020-01-15 00:49:29 <misterunknown> iggy: Thanks for your answer!
2020-01-15 12:54:42 <apangona> are there any resources how to package php webapps?
2020-01-15 12:55:17 <Cogitri> The existing APKBUILDs for such apps (e.g. Nextcloud)
2020-01-15 12:55:51 <clandmeter> or phpmyadmin
2020-01-15 12:56:38 <apangona> Thanks! I already found them, but was unsure if there were some guidelines available
2020-01-15 13:00:01 <ocin> I have a alpine USB stick (baked with mkimg scripts + 1 ovl file) and a problem. It boots and everything works fine on a fresh install but when I use it on a device where a different apkovl is present (sda1 in that case) the stick tries to use that one, thus modloop fails and the apkovl from the USB is not loaded at all. I tried loads of alpine_dev, ovl_dev, apkovl boot parameters (usbdisk, uuid,
2020-01-15 13:00:01 <ocin> combinations, ..) but I can just not get the alpine stick to load the apkovl on it - I am out if ideas now what to try
2020-01-15 14:11:11 <markand> there is a netiquette for mailing lists right? I feel mails from Lenz a bit rude and non-constructive TBH (regarding my question about GNOME + Wayland)
2020-01-15 14:23:58 <zfoo_> when running abuild it exits with an error (1, no child process), any ideas?
2020-01-15 14:26:13 <markand> zfoo_, you should ask in #alpine-devel with a detailed paste of the rror
2020-01-15 15:20:24 <mps> dalias: https://gitlab.alpinelinux.org/alpine/aports/issues/11128
2020-01-15 15:20:59 <mps> sorry for annoyance, but would like to hear your opinion
2020-01-15 15:21:35 <mps> when (and if) you found time, ofc
2020-01-15 15:22:43 <dalias> i will look
2020-01-15 15:22:56 <mps> thanks
2020-01-15 15:23:00 <dalias> i thought someone patched memcpy.S to work with thumb
2020-01-15 15:23:18 <dalias> maybe patch was dropped or left in limbo
2020-01-15 15:28:48 <mps> I remember that you worked few weeks ago on new malloc, so thought maybe memcpy could be related somehow
2020-01-15 15:36:08 <dalias> its not related
2020-01-15 15:38:17 <mps> aha
2020-01-15 16:07:45 <pltrz> folks running Alpine on their desktop/laptop - how do you do sound? do you run plain ALSA, or use a sound server to control audio in/out and mixing?
2020-01-15 16:09:30 <north1> Pulseaudio with alsa output plug-in
2020-01-15 16:09:57 <mps> I prefer plain ALSA
2020-01-15 16:10:54 <mps> I don't listen same time more sound sources
2020-01-15 16:11:57 <pltrz> if I do want multiple sound sources mixing, and controlling audio direction in/out different devices, is pulseaudio really my only option?
2020-01-15 16:12:45 <mps> jack?
2020-01-15 16:13:07 <mps> though I don't have experience with it
2020-01-15 16:15:47 <pltrz> might just go with pulse then, need just a bit more than plain alsa provides, but pulse seems like it does a milion things I don't need but oh well \o/
2020-01-15 16:16:11 <dozn> You'd want dmix, but it's not packaged for Alpine.
2020-01-15 16:16:16 <dozn> https://wiki.archlinux.org/index.php/Advanced_Linux_Sound_Architecture#Dmix
2020-01-15 16:16:16 <pltrz> oh, what is algitbot?
2020-01-15 16:16:29 <pltrz> dozn: thanks, ill check it out
2020-01-15 16:17:35 <dozn> Oh, maybe it's just built into ALSA, it's been many years since I've played with it.
2020-01-15 16:22:00 <MY-R> plain ALSA, didnt touch anything and can listen mpv, firefox, game etc in same time
2020-01-15 16:28:57 <dalias> mps, i found the patch. https://www.openwall.com/lists/musl/2019/09/13/8
2020-01-15 16:35:47 <mps> dalias: thanks, will look later if it can be applied and test it. now I'm out on moving computer
2020-01-15 16:36:37 <mps> you are awesome, btw. I thought to look at totally different sites
2020-01-15 16:36:50 <dalias> ?
2020-01-15 16:37:14 <mps> newlib, openwrt and similar
2020-01-15 16:37:37 <mps> and you found it on openwall.com
2020-01-15 16:38:18 <dalias> well i found it in my mailbox then just looked up the url from the message date :)
2020-01-15 16:42:55 <mps> nice, your mailbox must be big :)
2020-01-15 16:51:10 <bl4ckb0ne> is there a license on https://gitlab.alpinelinux.org/alpine/alpine-conf/tree/master ?
2020-01-15 17:10:45 <dalias> mps, entire mailing list archive is smaller than a month of typical bloated html email so ¯\_(ツ)_/¯
2020-01-15 17:15:14 <mps> eh, I thought to make script which will scan my mailbox and remove html and images from mail which have them
2020-01-15 17:16:05 <mps> but never had time, though sometimes look in mutt for big mails and do that manually
2020-01-15 17:16:28 <dalias> :)
2020-01-15 17:16:56 <dalias> my point was just that keeping around a complete archive isn't really a big deal in terms of space cost and it's kinda a must for your own projects
2020-01-15 17:17:10 <dalias> since third party archives are not necessarily reliable
2020-01-15 17:18:10 <mps> I keep mails from the 1994 for a lot of 'things'
2020-01-15 17:19:40 <mps> 3.5GB
2020-01-15 18:48:35 <ziggys> hi, everyone
2020-01-15 18:48:52 <ziggys> I have a problem with xorg and glamor-egl
2020-01-15 18:49:13 <ziggys> with no glamor-egl installed, xorg fails to start
2020-01-15 18:49:52 <ziggys> installing glamor-egl solves the problem, but when installing I have an error
2020-01-15 18:50:02 <ziggys> ERROR: glamor-egl-0.6.0-r4: trying to overwrite usr/lib/xorg/modules/libglamoregl.so owned by xorg-server-1.20.6-r0.
2020-01-15 18:50:56 <ziggys> https://bocados.p4g.club/qtkuy3vo
2020-01-15 18:52:47 <ziggys> any ideas?
2020-01-15 18:55:42 <mps> ziggys: do you mixed releases?
2020-01-15 18:56:06 <ziggys> no, fresh install
2020-01-15 18:56:35 <mps> I see
2020-01-15 18:56:44 <mps> it is bug
2020-01-15 18:57:12 <mps> could you fill bug report on gitlab.a.o
2020-01-15 18:58:30 <ziggys> oh! ok, will report
2020-01-15 18:58:33 <ziggys> thanks
2020-01-15 19:02:35 <mps> np
2020-01-15 22:08:31 <Shared404> I created a usb using the setup-bootable tool. When attempting to boot it has decided to send me to the grub CLI. ...
2020-01-15 22:08:59 <Shared404> I've been using this page: https://unix.stackexchange.com/questions/418401/grub-error-you-need-to-load-kernel-first to try to trouble shoot, obviously changing device names and file paths.
2020-01-15 22:09:17 <_ikke_> hmm, I would expect it to use syslinux by default
2020-01-15 22:10:12 <Shared404> Where should I head from here?
2020-01-15 22:10:43 <Shared404> If I try running the top code block from the accepted answer on the linked page it boots into a recovery shell,
2020-01-15 22:11:33 <Shared404> If I try to run "linux /boot/vmlinuz-lts root=/dev/sda1" I get error: invalid magic number.
2020-01-15 22:24:41 <clandmeter> mps: i think glamor-egl is now part of xorg.
2020-01-15 22:24:56 <clandmeter> its probably missing a config file
2020-01-15 22:28:26 <Shared404> Grub its self appears to be working fine. The recovery shell I was mentioning before is inside of linux. It launches with the error message "/sbin/init not found in new root. Launching emergency recovery shell"
2020-01-15 22:30:35 <AinNero> Shared404: what device is the grub on?
2020-01-15 22:30:39 <AinNero> sda2 ?
2020-01-15 22:32:18 <Shared404> sda1 IIRC, but I'm not 100% sure.
2020-01-15 22:32:38 <Shared404> Is there a good way to double check from inside grub?
2020-01-15 22:32:42 <mps> clandmeter: yes, it is xorg, but also in glamor-egl with same path and filename
2020-01-15 22:33:00 <AinNero> Shared404: is sda1 both your boot partition and your root file system?
2020-01-15 22:34:06 <AinNero> from the recovery shell you can run blkid to see what file systems are visible
2020-01-15 22:34:30 <Shared404> Ok, just one sec
2020-01-15 22:36:44 <Shared404> It looks like I might just be retarded. It appears to have recognized the computers internal hard drive as sda and my flash drive as sdb.
2020-01-15 22:36:56 <mps> so, glamor-egl should probably be removed from archive and aports
2020-01-15 22:37:06 <Shared404> GRUB should be on sdb1
2020-01-15 22:37:47 <AinNero> usually sda1 and sda1 are boot partitions for linux distros
2020-01-15 22:38:07 <AinNero> root should be sda2 or sda3
2020-01-15 22:38:31 <AinNero> that you get init not found might be because it tries to mount the boot partition instead of the root partition
2020-01-15 22:41:00 <Shared404> Ok. This makes sense.
2020-01-15 22:41:24 <Shared404> Sorry I'm taking so long to respond, I'm going back over docs trying to find my screw up.
2020-01-15 22:42:30 <Shared404> Is there a better place for documentation on setup-bootable than https://wiki.alpinelinux.org/wiki/Create_a_Bootable_USB ?
2020-01-15 22:42:49 <Shared404> currently looking at docs.alpinelinux.org , but not seeing much
2020-01-15 22:47:00 <mps> hm, glamor-egl have one file which is not xorg-server, usr/share/X11/xorg.conf.d/glamor.conf, not sure if this is needed
2020-01-15 23:06:23 <attos> I'm running Alpine Version 3.11.2, x86 (VIA C7 processor), kernel  5.4.8-0-lts. It runs really fine. I have installed the lirc sources abd compiled them successfully. I want to use my homebrew serial ir receiver, but there is no serial_ir kernel module under /lib/modules/5.4.8-0-lts/kernel/drivers/media/rc (or elsewhere).
2020-01-15 23:09:57 <mps> attos: it is not enabled in kernels
2020-01-15 23:10:46 <attos> mps: So I have to compile my own kernel ?
2020-01-15 23:12:02 <attos> That would be suboptimal.
2020-01-15 23:12:18 <mps> that's best option :) you can optimize it yo your usage/workload
2020-01-15 23:12:52 <mps> other option is to fill bug/issue request at gitlab.a.o
2020-01-15 23:13:23 <mps> s/yo /to /
2020-01-15 23:14:24 <attos> Is there a way to build only the serial_ir kernel module compatible to the distro kernel ?
2020-01-15 23:17:16 <mps> it could be but that would be more hassle than rebuild  current kernel from aports and enable modules/options you need
2020-01-15 23:24:42 <mps> and you will have to build it again with every alpine kernel upgrade
2020-01-15 23:25:31 <mps> as I told other option is to fill issue and ask that that driver be enabled
2020-01-15 23:25:38 <attos> This would have been my next question ...
2020-01-15 23:26:12 <mps> usually we enable most of drivers or options on users requests
2020-01-15 23:27:05 <attos> O.K. I will try to fill an issue.
2020-01-15 23:27:31 <mps> (don't ask to enable BPF, though :) )
2020-01-15 23:56:10 <attos> So it's done: https://gitlab.alpinelinux.org/alpine/aports/issues/11135
2020-01-15 23:57:08 <attos> May I ask until when I can hope to see this module in a new kernel build ?
2020-01-16 00:00:36 <mps> attos: hope in next kernel upgrade for 3.11, but not sure (I'm not kernel maintainer)
2020-01-16 00:01:11 <mps> and I think we should upgrade kernel for 3.11 soon
2020-01-16 00:03:44 <attos> That would be great. Thanks.
2020-01-16 00:05:32 <mps> I asked maintainer about this in issue
2020-01-16 00:07:12 <misterunknown> Hi. Does anyone know something like a "task queue daemon"? I look for a daemon in which I can queue shell commands, which are in turn executed sequentially (order is important). I already looked at "cron" and "at" but both of them are not really suitable for my needs. I don't want the jobs run at a specific time, but rather in a specific order, and one-by-one.
2020-01-16 00:32:57 <codebam> how can I install gnupg with smartcard support on alpine?
2020-01-16 00:35:57 <Shared404_> I don't know what I screwed up before, but reformating the flash drive and redoing setup-bootable fixed my issue.
2020-01-16 00:36:07 <Shared404_> Thanks for the help, sorry for my idiocy.
2020-01-16 00:51:03 <codebam> I can't find scdaemon
2020-01-16 00:51:13 <codebam> I just want to use my yubikey :(
2020-01-16 00:55:40 <MY-R> codebam, gnupg-scdaemon
2020-01-16 00:56:13 <codebam> MY-R: just found that like 10 seconds ago actually. it's working now but the key doesn't seem to be getting detected?
2020-01-16 00:58:26 <codebam> it's prompting for the key now though, which is good
2020-01-16 01:00:00 <codebam> got it working
2020-01-16 01:00:03 <codebam> thanks!
2020-01-16 01:00:14 <codebam> restarted pcscd and it worked
2020-01-16 01:01:56 <codebam> does firefox drm work on alpine?
2020-01-16 01:02:12 <MY-R> codebam, you can add yourself to group "gnupg"
2020-01-16 01:02:16 <codebam> using the firefox-esr package that is
2020-01-16 01:02:23 <codebam> MY-R: why would I need to do that?
2020-01-16 01:02:30 <codebam> it was working without
2020-01-16 01:02:37 <MY-R> dunno, if all working then no need probably
2020-01-16 01:02:49 <codebam> oh okay, cool
2020-01-16 01:04:08 <MY-R> kind of work I guess but forget about using netflix/spotify
2020-01-16 01:05:28 <codebam> hmm? so netflix/spotify does not work then?
2020-01-16 01:05:45 <MY-R> last time I checked - nope
2020-01-16 01:05:50 <codebam> ah okay
2020-01-16 01:06:03 <codebam> I guess if I really want it I can set up a chroot
2020-01-16 01:06:25 <MY-R> yep
2020-01-16 01:09:42 <codebam> thanks!
2020-01-16 01:09:49 <codebam> might be back in a bit
2020-01-16 01:10:29 <codebam> opened in tmux :)
2020-01-16 01:38:49 <codebam> how do I set up elogind?
2020-01-16 02:18:23 <codebam> how do I add to the PATH for ash? I put this inside ~/.ashrc but it doesn't seem to be working. export PATH="$HOME/.local/bin:$PATH"
2020-01-16 02:18:28 <codebam> however it works fine if I run it from a shell
2020-01-16 02:21:20 <codebam> it doesn't seem to be loaded when I start ash
2020-01-16 02:37:23 <pltrz[m]> ash doesn't load .ashrc
2020-01-16 02:37:36 <pltrz[m]> it just loads .profile
2020-01-16 02:38:43 <pltrz[m]> and if the ENV environment variable is set to a path to a file, it will read that
2020-01-16 02:39:44 <pltrz[m]> so to have a .ashrc just add `ENV=$HOME/.ashrc` to .profile
2020-01-16 02:40:06 <uplime> or . ~/.ashrc
2020-01-16 02:43:46 <pltrz[m]> codebam: but in the case of just adding a PATH, assuming you don't need to reload that every time you open a shell, you could just throw it into .profile instead, you don't need a .ashrc
2020-01-16 02:54:57 <codebam> pltrz[m]: adding that to .profile didn't work
2020-01-16 02:55:08 <codebam> my path still doesn't include what I added to it in .ashrc
2020-01-16 02:58:00 <pltrz[m]> codebam: you need to log out/in for the new .profile to be loaded
2020-01-16 02:58:51 <codebam> ohh, okay
2020-01-16 02:59:18 <codebam> that worked, thanks pltrz[m] :)
2020-01-16 02:59:40 <pltrz[m]> no problemo!
2020-01-16 03:01:12 <codebam> what would be the best idea if I want to work on some python software that requires I compile a pip module that doesn't compile on musl?
2020-01-16 03:01:23 <codebam> should I just use the arch chroot when I'm developing it or?
2020-01-16 03:01:27 <codebam> s/the/an
2020-01-16 03:03:25 <codebam> also, is there a way I can run graphical applications from in a chroot inside wayland?
2020-01-16 03:03:31 <codebam> specifically, sway
2020-01-16 09:32:01 <ocin> I have a alpine USB stick (baked with mkimg scripts + 1 ovl file) and a problem: On a fresh install all is fine but when I boot the stick on a pc where a different apkovl is present on a disk partition (sda1 in that case) the stick tries to use that one, thus modloop verify fails, basically everything is broken and the apkovl from the USB is not loaded at all. I tried loads of alpine_dev, ovl_dev,
2020-01-16 09:32:01 <ocin> apkovl boot parameters (usbdisk, uuid, combinations, ..) but I can just not get the alpine stick to load the apkovl on it - I am out if ideas now what to try, any ideas?
2020-01-16 09:38:19 <coopzone> Hello, can anyone tell me how to update a minor release, ie from 3.11.0 to 3.11.2
2020-01-16 09:39:00 <_ikke_> apk uprade -U --available
2020-01-16 09:39:07 <_ikke_> No need to explicitly target it
2020-01-16 09:41:38 <coopzone> ok, thanks. I take it that does not update the /etc/alpine-release file? As that still shows 3.11.0 ?
2020-01-16 09:42:00 <_ikke_> I think it should
2020-01-16 09:42:35 <_ikke_> what does apk version alpine-base return?
2020-01-16 09:44:00 <coopzone> it returns
2020-01-16 09:44:03 <coopzone> Installed:                                Available:
2020-01-16 09:44:03 <coopzone> alpine-base-3.11.2-r0                   = 3.11.2-r0
2020-01-16 09:44:31 <coopzone> this is good! But the /etc/alpine-release file does not change.
2020-01-16 09:45:14 <_ikke_> can you try apk info -L alpine-base?
2020-01-16 09:45:56 <MY-R> wasnt like ncopa forgot to update that last time? :P
2020-01-16 09:46:47 <_ikke_> not for 3.11.2
2020-01-16 09:47:05 <MY-R> ahh, so it was with .1
2020-01-16 09:47:17 <_ikke_> yes
2020-01-16 09:48:08 <coopzone> it returns
2020-01-16 09:48:17 <coopzone> Installed:                                Available:
2020-01-16 09:48:17 <coopzone> alpine-base-3.11.2-r0                   = 3.11.2-r0
2020-01-16 09:48:28 <coopzone> ignore that, hang on
2020-01-16 09:48:51 <coopzone> alpine-base-3.11.2-r0 contains:
2020-01-16 09:48:51 <coopzone> etc/os-release
2020-01-16 09:48:51 <coopzone> etc/alpine-release
2020-01-16 09:48:51 <coopzone> etc/issue
2020-01-16 09:49:09 <coopzone> there we go, finger problems!
2020-01-16 10:03:58 <coopzone> thanks everyone, i am happy it's at the latest version, even if the release file does not say so!. Bye for now.
2020-01-16 12:47:59 <mckaygerhard> https://wiki.alpinelinux.org/wiki/Eight_Tips_Of_Skin_Care_For_Netizen  <--spam
2020-01-16 12:49:25 <_ikke_> goen now
2020-01-16 13:24:47 <AinNero> i heard avocados and fish have oils that are good for the skin
2020-01-16 13:31:07 <mps> I would check such marketing with quantum RNG and gnupg with 8192 key length :D
2020-01-16 14:57:30 <dngray[m]> hmm
2020-01-16 14:57:49 <dngray[m]> seems i cannot do this: rc_need="net net.tun0"
2020-01-16 14:59:13 <dngray[m]> for sockd, which needs the tun0 interface when openvpn starts
2020-01-16 14:59:16 <dngray[m]> or it won't start
2020-01-16 15:00:04 <dngray[m]> and dnscrypt-proxy requires sockd to be started or it crashes
2020-01-16 15:00:15 <dngray[m]> because i am using a socks proxy
2020-01-16 15:02:06 <dngray[m]> https://termbin.com/0p9x is the error i get
2020-01-16 15:10:00 <gjabell> what's the process for switching maintainers of a package?
2020-01-16 15:11:19 <_ikke_> Creating a MR that does it and getting approval from the old maintainer
2020-01-16 15:11:57 <_ikke_> Or stating that the current maintainer is not active anymo
2020-01-16 15:12:00 <_ikke_> anymore
2020-01-16 15:18:21 <gjabell> ok
2020-01-16 15:27:46 <mps> gjabell: I would first ask over IRC or mail rather
2020-01-16 15:29:02 <_ikke_> yes, indeed
2020-01-16 15:31:51 <gjabell> ah, I just submitted a patch to the mailing list cc'ing the original maintainer
2020-01-16 15:31:56 <gjabell> I can ping them here if they're around tho
2020-01-16 15:32:43 <gjabell> clandmeter: do you mind if I adopt the riot-web package?
2020-01-16 15:33:12 <clandmeter> gjabell: go ahead
2020-01-16 15:34:01 <gjabell> cool, thanks :)
2020-01-16 15:35:04 <dngray[m]> mm, openrc is always so hard :P
2020-01-16 15:35:25 <dngray[m]> i suppose i could do rc_need openvpn
2020-01-16 15:35:33 <dngray[m]> instead of net.tun0
2020-01-16 15:36:43 <Cogitri> gjabell: Since you're taking over that package: You don't happen to know if I can use a self-hosted Riot.im in the desktop client?
2020-01-16 15:37:03 <Cogitri> I currently use the default riot thingie right now since I use the web client
2020-01-16 15:38:28 <gjabell> Cogitri: you should be able to use self-hosted on both of them, since the desktop is just an electron app with basically the same stuff as the web app
2020-01-16 15:38:46 <gjabell> Cogitri: not sure if the desktop one is packaged for alpine atm tho
2020-01-16 15:39:30 <Cogitri> No since electron doesn't run on musl
2020-01-16 15:39:57 <Cogitri> I use the Flatpak right now but I think Riot.im is hardcoded in the Electron client
2020-01-16 15:40:46 <gjabell> Ah, interesting
2020-01-16 15:40:50 <gjabell> I know I've used the desktop one with a selfhosted server in the past, but maybe the flatpak is different?
2020-01-16 15:41:23 <gjabell> It should theoretically show up once you click "log in", like the web app
2020-01-16 15:41:54 <gjabell> With a message saying "your logging into <server>" or something, and then you can pick "change server" and use a selfhosted one
2020-01-16 15:49:02 <C0nundrum> Hello i wrote small string split function which works fine on bash 3 on osx but it only returns the first element on alpine bash 5. Not sure what's wrong
2020-01-16 15:49:04 <C0nundrum> https://pastebin.com/6JVKxBD5
2020-01-16 16:00:32 <Shared404> Is there any reason not to tell lbu to track everything on a usb?
2020-01-16 16:11:22 <Cogitri> gjabell: Yes, but doesn't that only change what matrix server you use as backend and not where you pull riot.im from?
2020-01-16 16:11:34 <Cogitri> Or is that bundled in the desktop client already? 🤔
2020-01-16 16:12:43 <gjabell> Cogitri: oh I misunderstood your original question haha
2020-01-16 16:13:05 <gjabell> I don't think there's a way to use a remote riot.im "website" with the desktop app, but there's no real reason you'd need to anways
2020-01-16 16:13:41 <gjabell> unless you were running a modified version of riot.im, but you'd need to build your own desktop app for that anyways
2020-01-16 16:14:02 <gjabell> since it's not actually loading the riot.im webpage inside the desktop app, it's just using the same assets when it's compiled
2020-01-16 16:32:40 <Cogitri> Ah good
2020-01-16 17:48:59 <Barcelona> Moin moin
2020-01-16 17:49:07 <Barcelona> Anyone know what has happened to linux-vanilla in the main repo?
2020-01-16 17:49:27 <Barcelona> https://git.alpinelinux.org/aports/tree/main/linux-vanilla
2020-01-16 17:50:27 <mps> it is replaced by linux-lts
2020-01-16 17:50:42 <Barcelona> No way to get non-lts?
2020-01-16 17:50:49 <Barcelona> Seems weird
2020-01-16 17:51:05 <Barcelona> Also, this page needs updating https://wiki.alpinelinux.org/wiki/Kernels
2020-01-16 17:51:12 <mps> you can download it from 3.10 release
2020-01-16 17:51:30 <Barcelona> I just want the config
2020-01-16 17:52:31 <mps> you can get it from git repo, searching proper commit
2020-01-16 17:52:47 <Barcelona> Is this no longer going to be updated?
2020-01-16 17:52:55 <Barcelona> I use it in a script
2020-01-16 17:53:14 <mps> ?
2020-01-16 17:53:23 <Barcelona> I use the config in a script to auto-build kernels
2020-01-16 17:54:42 <mps> if you have running linux-vanilla you can get it from /proc/configs.gz
2020-01-16 17:55:18 <Barcelona> Right
2020-01-16 17:55:21 <Barcelona> Let me rephrase
2020-01-16 17:55:42 <feuerteufel> Hi @ all
2020-01-16 17:55:44 <Barcelona> Will a /new/ config ever be created by upstream that is updated to the newest version of the vanilla kernel
2020-01-16 17:55:57 <Barcelona> I have the old config already, I have had this script running for yeras
2020-01-16 17:56:03 <Barcelona> years*
2020-01-16 17:56:17 <Barcelona> But if the config won't be updated, I guess I'd rather use the lts version
2020-01-16 17:56:26 <mps> Barcelona: no, eba4f554cc53325bb37f0cca97a5dfd024582271
2020-01-16 17:56:31 <Barcelona> Ok
2020-01-16 17:56:43 <Barcelona> I will use linux-lts's config I guess, it probably should work
2020-01-16 17:57:01 <Barcelona> Thanks for letting me know! mps++
2020-01-16 17:57:16 <mps> you can checkout repo at this commit -1 and found latest linux-vanilla configs
2020-01-16 17:57:18 <feuerteufel> I'm yust runing in almost the same question, are you guys reading my mind? :-))
2020-01-16 17:57:46 <mps> :)
2020-01-16 17:58:23 <feuerteufel> So, is there a "default" kernel to "normal" use?
2020-01-16 17:58:38 <mps> material world is just manifestation of higher 'spiritual' world
2020-01-16 17:58:56 <Barcelona> Ha nice
2020-01-16 17:58:58 <Barcelona> I guess use lts
2020-01-16 17:59:08 <Barcelona> I am compiling against the lts config rn
2020-01-16 17:59:15 <mps> feuerteufel: default kernel for 3.11 and up is linux-lts
2020-01-16 17:59:50 <mps> linux-vanilla was 4.19.x and linux-lts is 5.4.x
2020-01-16 17:59:50 <Barcelona> Okay well I should be off, see you guys later
2020-01-16 17:59:55 <Barcelona> o/
2020-01-16 18:00:02 <mps> \o
2020-01-16 18:02:12 <feuerteufel> So, if I upgrade my 3.10.3 to 3.11.x. The kernel also will upgraded to Linux-lts?
2020-01-16 18:03:00 <mps> feuerteufel: I think no
2020-01-16 18:03:17 <mps> you will have to do this 'by hand'
2020-01-16 18:03:24 <mps> apk add linux-lts
2020-01-16 18:03:42 <mps> but not sure
2020-01-16 18:10:44 <programmerq> I've found that when moving away from an old kernel, I have to delete the old files in /boot otherwise the bootloader can still use them.
2020-01-16 18:10:55 <programmerq> like when linux-grsec was dropped, just removing the package wasn't enough.
2020-01-16 18:11:42 <mps> programmerq: yes, and good to have it as rescue kernel so I didn't deleted it
2020-01-16 18:12:44 <programmerq> my bootloader would boot from it as the default kernel still, so I deleted it
2020-01-16 18:15:24 <feuerteufel> programmerq: Same problem I had, I remeber that there was a diffrent name for the kernel but I don't remeber the name
2020-01-16 18:26:55 <okias[m]> I need force delete package... and then install same version, how to do that>?
2020-01-16 18:27:01 <okias[m]> any --force* doesn't work for me
2020-01-16 18:27:20 <okias[m]> and it has dependency, but I don't care, since I'll install it again right away
2020-01-16 18:27:42 <mps> apk fix --reinstall
2020-01-16 18:29:48 <okias[m]> mps does it redownload package?
2020-01-16 18:29:56 <okias[m]> since I need it from local repo
2020-01-16 18:30:35 <mps> not sure, never tried in that way
2020-01-16 18:30:43 <okias[m]> I don't see it touches my http server
2020-01-16 18:30:50 <okias[m]> except downloading index
2020-01-16 18:31:35 <mps> for local pkg I use 'apk add --allow-untrusted /path/pkgfilename' and always works
2020-01-16 18:32:54 <okias[m]> asus-grouper:~# apk fix iio-sensor-proxy --force --allow-untrusted  --no-cache --force-refresh
2020-01-16 18:33:15 <okias[m]> well, it's in my computer (this is the tablet)
2020-01-16 18:34:37 <okias[m]> oh I see.. I accidentally build for different arch
2020-01-16 18:34:56 <mps> not sure, didn't had such needs
2020-01-16 18:36:10 <okias[m]> (1/1) [APK unavailable, skipped] Reinstalling iio-sensor-proxy (2.8-r2)
2020-01-16 18:40:12 <okias[m]> why I can't just remove that package, why?
2020-01-16 18:40:31 <okias[m]> I don't know what it reinstalls, I removed all cache
2020-01-16 18:45:24 <okias[m]> at least some --force-upgrade to redowload+reinstall?
2020-01-16 18:50:54 <okias[m]> mps in the end I sent it by ssh and did by your way, but it's kinda frustrating when you running mirror :D
2020-01-16 19:02:49 <codebam> I followed the installation guide for LVM on LUKS but when I try to boot grub it's dropping me to a recovery shell?
2020-01-16 19:03:47 <codebam> my /etc/default/grub came mostly empty, there was no GRUB_CMDLINE_LINUX_DEFAULT so I had to create one. is that normal?
2020-01-16 19:04:18 <mps> okias[m]: maybe pinned version install could help in you case, if you have pkg with different version number
2020-01-16 19:05:43 <codebam> my /etc/default/grub http://ix.io/27z9
2020-01-16 19:06:13 <codebam> my /boot/grub/grub.cfg http://ix.io/27za
2020-01-16 19:17:12 <codebam> am I doing something wrong? I've gone over this wiki like 5 or 6 times and I can't find anything I've done wrong
2020-01-16 19:17:23 <codebam> but grub still isn't booting
2020-01-16 19:17:40 <codebam> the /etc/default/grub looks wrong
2020-01-16 19:18:32 <codebam> I used the correct UUID in the crypttab and the /etc/default/grub though
2020-01-16 19:20:42 <Cogitri> "grub still isn't booting" == it doesn't show up at all or after selecting the kernel it doesn't boot?
2020-01-16 19:22:36 <codebam> Cogitri: it drops me to a recovery shell after saying it can't find lvmid/wtKD...
2020-01-16 19:23:20 <Cogitri> So grub does in fact work
2020-01-16 19:23:56 <Cogitri> Your kernel doesn't boot though. I suppose you might have forgotten to include the relevant modules in your mkinitfs.conf ?
2020-01-16 19:24:59 <codebam> I added lvm and cryptsetup to the end of features=""
2020-01-16 19:28:35 <codebam> just regenerated it to be sure. rebooting
2020-01-16 19:32:58 <codebam> that didn't work
2020-01-16 19:33:04 <codebam> same error
2020-01-16 19:38:15 <codebam> I'm not sure what to do
2020-01-16 19:38:26 <codebam> I booted and chroot again
2020-01-16 19:38:35 <codebam> why isn't it booting though?
2020-01-16 20:08:48 <codebam> anyone have any ideas?
2020-01-16 20:08:55 <codebam> or would be able to help me out
2020-01-16 20:27:35 <MY-R> codebam, is your root=/dev/mapper/vg0-root partition /boot or / ?
2020-01-16 20:28:08 <codebam> MY-R: it's /
2020-01-16 20:28:23 <MY-R> so I think there is wrong path in grub.cfg
2020-01-16 20:28:24 <codebam> I mounted the boot partition under /boot
2020-01-16 20:28:58 <codebam> do I need to set root= in the /etc/default/grub?
2020-01-16 20:29:29 <MY-R> or not sure already, can you try hit "e" and edit path from "/vmlinuz-lts root=/dev/mapper...." to "/boot/vmlinuz-lts root=/dev/mapper...."
2020-01-16 20:29:42 <MY-R> in grub menu when booting
2020-01-16 20:30:06 <codebam> I don't have a grub menu when booting, I just get dropped to a grub recovery shell
2020-01-16 20:30:58 <MY-R> BIOS or UEFI?
2020-01-16 20:31:25 <codebam> UEFI
2020-01-16 20:31:52 <codebam> I noticed in the wiki it also puts /boot in the LVM + LUKS
2020-01-16 20:32:02 <codebam> so I don't know if that might be causing issues
2020-01-16 20:32:16 <codebam> the ESP is outside though
2020-01-16 20:36:40 <Cogitri> Putting everything in the ESP won't hurt usually
2020-01-16 20:40:42 <codebam> oh, okay
2020-01-16 20:43:50 <codebam> I'll start from the beginning
2020-01-16 20:45:02 <MY-R> codebam, when you manualy mount everything and type "lsblk -f" then be sure that partition marked "crypto_LUKS" is the one which UUID need in /etc/default/grub in cryptroot=
2020-01-16 20:47:08 <codebam> MY-R: yes that is the one I set for the UUID
2020-01-16 20:47:44 <codebam> I'm going to try with /boot outside the LVM and just keeping the ESP inside /boot
2020-01-16 20:48:23 <MY-R> codebam, and the name with LVM2_member is your cryptdm=lvmcrypt
2020-01-16 20:49:14 <MY-R> if all that is right then was some problem during install grub or ye with uefi
2020-01-16 20:49:23 <codebam> MY-R: is that what the UUID in /etc/crypttab should be?
2020-01-16 20:49:43 <codebam> the LVM2_member?
2020-01-16 20:50:58 <MY-R> ye
2020-01-16 20:51:15 <MY-R> oh wait
2020-01-16 20:51:18 <codebam> oh, that was the mistake then. I'll fix that. I think the wiki page is wrong
2020-01-16 20:51:21 <codebam> oh
2020-01-16 20:51:29 <MY-R> one sec
2020-01-16 20:51:44 <codebam> okay
2020-01-16 20:52:42 <MY-R> codebam, the UUID which you got from "blkid -s UUID -o value /dev/sda2" should be in grub "cryptroot=UUID="
2020-01-16 20:53:31 <codebam> yes that's the crypto_LUKS though
2020-01-16 20:53:57 <MY-R> and cryptdm=lvmcrypt should be a LVM2_member
2020-01-16 20:54:05 <codebam> lvmcrypt is the LVM2_member which has a different UUID
2020-01-16 20:54:15 <MY-R> yep so all that is corret
2020-01-16 20:54:16 <codebam> it's called lvmcrypt though
2020-01-16 20:54:29 <codebam> okay but it's not booting
2020-01-16 20:55:08 <MY-R> when you chroot then be sure you have all your boot etc stuff mounted
2020-01-16 20:55:38 <MY-R> and watch at the grub "errors" if any and ye do like they wrote with efi stuff
2020-01-16 20:56:07 <codebam> there were no grub errors when I installed. I mounted everything it said in the wiki page
2020-01-16 20:56:28 <codebam> but let me try again
2020-01-16 21:16:23 <g1eb> Hi Alpine! I need help installing python 3.7 in my alpine:edge docker contaienr
2020-01-16 21:17:05 <g1eb> it keeps installing python 3.8 for some reason, idk how to specify 3.7
2020-01-16 21:17:41 <Cogitri> You can't unless you use a alpine:3.10 container
2020-01-16 21:17:52 <Cogitri> Edge and 3.11 only come with 3.8
2020-01-16 21:19:18 <g1eb> ahh, I see, let me try 3.10 on second
2020-01-16 21:21:02 <g1eb> checks out, I'm getting python 3.7.5 with alpine:3.10 container, thanks!
2020-01-16 21:40:29 <codebam> okay so I installed now with /boot and /boot/efi in their own partitions. I now get a menu but it's not properly decrypting the luks volume
2020-01-16 21:41:08 <codebam>  /dev/mapper/vg0-root on /sysroot failed: no such file or directory
2020-01-16 21:41:23 <codebam> then I get dropped into an initramfs recovery shell
2020-01-16 21:41:57 <codebam> I've updated the initramfs to contain cryptsetup, but I didn't get a prompt to enter the password
2020-01-16 21:45:04 <codebam> I'm not sure what to do now. as far as I can tell this should be working
2020-01-16 21:46:40 <MY-R> codebam, at least some progress because now it read initrd :D
2020-01-16 21:47:16 <codebam> yeah :)
2020-01-16 21:50:04 <MY-R> codebam, paste your /etc/default/grub and /boot/grub/grub.cfg and /etc/fstab and we will see
2020-01-16 21:50:12 <codebam> okay
2020-01-16 22:00:19 <codebam_> grub.cfg: https://ix.io/27Af
2020-01-16 22:00:56 <codebam_> etc/default/grub: https://ix.io/27Ag
2020-01-16 22:01:29 <MY-R> codebam_, Error code: SSL_ERROR_RX_RECORD_TOO_LONG
2020-01-16 22:01:30 <codebam_> etc/fstab: https://ix.io/27Ah
2020-01-16 22:01:38 <MY-R> ix.io looks like "broken"
2020-01-16 22:01:39 <codebam_> oh sorry they're not https
2020-01-16 22:01:44 <MY-R> ah
2020-01-16 22:04:35 <MY-R> codebam_, first what noticed, no lvm module in grub.cfg
2020-01-16 22:05:07 <codebam_> MY-R: why would lvm be in grub, it's in the initramfs
2020-01-16 22:05:38 <MY-R> your root / partition is LVM one after all
2020-01-16 22:05:49 <MY-R> your previous grub.cfg got all fine
2020-01-16 22:06:01 <codebam_> how do I add lvm to grub?
2020-01-16 22:06:23 <AinNero> you only need lvm for grub if your boot partition is in lvm
2020-01-16 22:06:35 <AinNero> which on alpine, isnt if you use the installer scripts
2020-01-16 22:06:38 <codebam_> my /boot is it's own ext4 partition
2020-01-16 22:06:52 <AinNero> then grub only needs the ext4 stuff, which is there per default
2020-01-16 22:07:34 <codebam_> the root is lvm + crypto-luks
2020-01-16 22:07:39 <codebam_> but it's not booting
2020-01-16 22:09:13 <AinNero> can you try to mount your root partition manually from the initramfs recovery shell?
2020-01-16 22:09:29 <codebam_> how would I do that?
2020-01-16 22:09:51 <AinNero> some cryptsetup and some lvm commands, but i dont know them in detail
2020-01-16 22:10:04 <AinNero> just to look at which step it fails
2020-01-16 22:10:20 <codebam_> oh, right I should have cryptsetup from in that shell I guess
2020-01-16 22:10:32 <codebam_> okay, I'll reboot
2020-01-16 22:11:58 <MY-R> ah ye I forgot that I have got just single partition
2020-01-16 22:12:01 <codebam> the initramfs shell doesn't see my nvme drive or it's partitions
2020-01-16 22:12:34 <MY-R> add nvme to /etc/mkinitfs/mkinitfs.conf
2020-01-16 22:12:44 <codebam> oh okay
2020-01-16 22:13:40 <MY-R> then generate new initrd
2020-01-16 22:13:51 <codebam> yeah
2020-01-16 22:28:48 <alpurn> Good day friends.  I'm looking for some guidance with the following script (part of our CI system) : https://git.riboflav.in/snippets/4
2020-01-16 22:29:55 <alpurn> I'm compiling some go,  creating an APK with the binary artifact, signing it, then creating a filesystem and trying to install the dependant packages yet for some reason when I boot the OS many packages aren't included yet some are.
2020-01-16 22:31:23 <alpurn> It's obviously rather hacky and missing some sanity - I just need to figure how to get the packages in /media/mmcblk0/apks/armhf/ (including my homebrew APK) and sign the APKINDEX.tar.gz
2020-01-16 22:34:22 <alpurn> Note: this is armhf and is building on a pi-3 (I'm not using armv7 because require support for pi-zero and pi2
2020-01-16 22:37:37 <alpurn> I think I may be missing a signature on the index or something - unsure
2020-01-17 01:20:05 <codebam_> sorry for asking this again, but how do I load my .ashrc from inside .profile?
2020-01-17 01:21:14 <dalias> . .ashrc
2020-01-17 01:21:44 <codebam_> oh, thank you
2020-01-17 01:21:46 <dalias> i guess you could do ~/.ashrc but i dont think there are cases where .profile is sourced where working dir isn't home
2020-01-17 01:21:48 <codebam_> didn't realize that's all it was
2020-01-17 01:21:56 <codebam_> right okay
2020-01-17 01:24:52 <codebam_> how do I configure pulseaudio? is there a package for pavucontrol?
2020-01-17 01:25:31 <codebam_> oh I see it's in edge
2020-01-17 01:35:49 <pltrz[m]> codebam: note that .profile is only loaded on login shells, so that would be when you login to the tty usually
2020-01-17 01:36:25 <pltrz[m]> which means if you add . .ashrc to .profile, it will also only be loaded that first time. new shells won't load .ashrc
2020-01-17 01:37:36 <pltrz[m]> if you want an .ashrc that is loaded each time, the way .bashrc works, for example then you should add ENV=$HOME/.ashrc to .profile
2020-01-17 01:41:30 <pltrz[m]> read here:
2020-01-17 01:41:31 <pltrz[m]> https://linux.die.net/man/1/ash
2020-01-17 01:41:38 <pltrz[m]> on the "Invocation" section
2020-01-17 05:39:30 <drakon> Hi! I'm facing some issues with my Alpine system and came here looking for help. I have an old laptop that I use as a WiFi repeater, previously, it was running Arch Linux without any problems. Some days ago I decided to switch to Alpine as I wanted to try it and I started having bandwidth problems on all of my AP clients. I can download stuff on full speed directly on that old laptop, but I have nearly half of the bandwidth on all of the AP clients.
2020-01-17 05:39:30 <drakon> It's running with the physical interface wlan0 and a virtual interface ap0 created with iw tool. My network card is an Atheros AR242x / AR542x
2020-01-17 05:40:05 <drakon> Already tried turning off power saving and nothing happened
2020-01-17 05:54:06 <drakon> actually my card is Atheros AR2425
2020-01-17 09:15:34 <crockwork_> wireguard's not in the current alpine kernel, right?
2020-01-17 09:19:40 <jn__> crockwork_: it's not in any of the kernel-{virt,lts,vanilla} kernels, because the newest one is 5.4.12 as released on kernel.org, which doesn't contain wireguard
2020-01-17 09:19:53 <jn__> but there's a separate wireguard kernel module package
2020-01-17 09:20:01 <crockwork_> Oh, interesting.
2020-01-17 09:20:09 <crockwork_> Is there a good tutorial for setting it up?
2020-01-17 09:20:19 <jn__> yes, on wireguard.org
2020-01-17 09:20:27 <jn__> errr, .com
2020-01-17 09:20:28 <crockwork_> neato.
2020-01-17 09:22:06 <jn__> so, step one: install the wireguard package that matches the kernel that you're running (if available)
2020-01-17 09:22:25 <jn__> step two: configure wireguard as you wish, e.g. with https://www.wireguard.com/quickstart/
2020-01-17 09:22:37 <jn__> step three: join #wireguard for further questions
2020-01-17 09:29:00 <clandmeter> crockwork_: we have an entry on our wiki
2020-01-17 09:31:23 <clandmeter> the simplistic approach without using wg-quick.
2020-01-17 09:31:28 <clandmeter> https://wiki.alpinelinux.org/wiki/Configure_a_Wireguard_interface_(wg)
2020-01-17 12:07:43 <pltrz[m]> how can I update the manpage cache? as in I manually copied some manpages to /usr/local/share/man/man8/foo.8 and they are not man-able.
2020-01-17 12:11:48 <AinNero> i thought there is no manpage cache on alpine
2020-01-17 12:11:58 <AinNero> can you strace the man command to see where its looking?
2020-01-17 12:17:43 <pltrz[m]> AinNero: https://pastebin.com/raw/5r2K8TXp
2020-01-17 12:18:25 <pltrz[m]> strace man apk-add
2020-01-17 12:18:56 <mps> pltrz[m]: set env MANPATH
2020-01-17 12:19:52 <AinNero> > open("/usr/local/man", O_RDONLY|O_NONBLOCK|O_CLOEXEC|O_PATH) = -1 ENOENT (No such file or directory)
2020-01-17 12:19:59 <AinNero> ah i see
2020-01-17 12:20:16 <AinNero> its indeed that man looks in other places than were you placed the manpage
2020-01-17 12:20:24 <pltrz[m]> mps: ah yes! that'll do it
2020-01-17 12:20:30 <AinNero> so what mps says or place it elsewhere
2020-01-17 12:21:16 <mps> to mandoc should be added conf file
2020-01-17 12:21:42 <pltrz[m]> mps: ah yes this is what I was gonna ask
2020-01-17 12:21:54 <pltrz[m]> thx, will try it out
2020-01-17 12:22:05 <mps> I thought month ago when fixed segfault but forgot to add conffile
2020-01-17 12:25:17 <pltrz[m]> mps: do you mean /etc/man.conf as described in man(1) ?
2020-01-17 12:25:43 <pltrz[m]> because I don't see any reference to a conf file in mandoc(1)
2020-01-17 12:26:02 <mps> pltrz[m]: yes
2020-01-17 12:26:48 <pltrz[m]> mps: gotcha
2020-01-17 12:27:34 <mps> 'gotcha' is some slang word I don't understand. what it means
2020-01-17 12:28:01 <_ikke_> 'got it'
2020-01-17 12:28:07 <_ikke_> I understand you
2020-01-17 12:28:13 <mps> ah, thanks
2020-01-17 12:29:23 <pltrz[m]> mps: sorry about that - we are an international community! :D
2020-01-17 12:30:40 <mps> yes, and we should use 'normal' (book) languages
2020-01-17 12:31:02 <_ikke_> mps: This is informal chatting, it
2020-01-17 12:31:23 <mps> not if user asks for help
2020-01-17 12:31:35 <_ikke_> it's hard to avoid qolloquialisms :)
2020-01-17 12:32:18 <mps> yes, I understand this, and I'm not criticizing anyone
2020-01-17 12:33:21 <mps> but we should have in mind that there a lot people reading this and not all of them are proficient in 'slang'
2020-01-17 12:36:10 <_ikke_> yes, that's certainly good to keep in mind
2020-01-17 12:37:10 <mps> :)
2020-01-17 12:37:42 <pltrz[m]> mps: it's great community we have here.
2020-01-17 12:38:14 <pltrz[m]> any of you guys planning on going to FOSDEM?
2020-01-17 12:40:21 <AinNero> mps: for slang words, you can use https://www.urbandictionary.com/ for explanation, also available as !u with duckduckgo search
2020-01-17 12:41:59 <mps> AinNero: thanks. yes, I know, but I'm not always on desktop and don't have always fast internet link.
2020-01-17 12:43:55 <pltrz[m]> I think extremely common ones that have made their way into staples of casual conversation - without being cryptic - are ok. In fact I'd even say that these are borderline not slang anymore. The vast majority of slang terms do not fall into that category though as they usually need more cultural context.
2020-01-17 12:44:31 <AinNero> most of us are not native english speakers and barely know basic idioms
2020-01-17 12:49:46 <pltrz[m]> it's true. I'm glad that we are all understanding of this fact :)
2020-01-17 14:03:01 <PICCORO> hi _ikke_ and ncopa ihave a long old alpine linux group and also are part of others in telegram.. that already are linked in community.. may could we anunced here and at mail list as official telegram group of alpine linux?
2020-01-17 14:03:30 <PICCORO> those are : * @alpine_linux * @alpinelinuxru * @alpine_linux_espanol
2020-01-17 14:07:08 <north1> I think better not
2020-01-17 14:13:13 <mps> telegram? service which require phone number to register?
2020-01-17 14:13:55 <_ikke_> yes
2020-01-17 14:14:23 <mps> my memory works
2020-01-17 14:18:20 <mps> no, it doesn't. last friday someone asked for fix on pkg which I can't remember now :|
2020-01-17 14:21:19 <daftaupe> hello, I'm having issue booting on a btrfs root partition, I tried adding btrfs to /etc/mkinitfs/mkinitfs.conf, but I'm dropped in a shell and can't even type in there :/ anyone having a thought about that ?
2020-01-17 14:21:50 <pltrz[m]> daftaupe: did you rebuild the initramfs after?
2020-01-17 14:22:15 <pltrz[m]> mkinitfs(1)
2020-01-17 14:23:56 <daftaupe> yup, and I notice the file timestamp is at the time of the rebuild
2020-01-17 14:24:36 <pltrz[m]> i _think_ it would be mkinitfs -c /etc/mkinitfs/mkinitfs.conf
2020-01-17 14:24:46 <pltrz[m]> oh, I see
2020-01-17 14:26:20 <pltrz[m]> hmm really not sure :/ sorry
2020-01-17 14:26:39 <daftaupe> No worries I'm here for fun :)
2020-01-17 14:28:48 <PICCORO> telegram yes _ikke_
2020-01-17 14:29:29 <PICCORO> russian grou and english group has now 22 membeers.. rules anre well stable and only are specific for alpine issues and some related linux topics
2020-01-17 14:30:21 <mps> PICCORO: you want to add gateway irc<->telegram?
2020-01-17 14:33:33 <pltrz[m]> has there ever been consideration for Matrix? then you could easily bridge to whatever service people prefer
2020-01-17 14:34:09 <_ikke_> We used to have our own matrix server, but we decided to not continue it
2020-01-17 14:34:15 <mps> pltrz[m]: aren't you on matrix?
2020-01-17 14:35:05 <pltrz[m]> yes, I am!
2020-01-17 14:35:14 <pltrz[m]> you can tell because of the *[m], right?
2020-01-17 14:35:29 <mps> yes
2020-01-17 14:35:53 <pltrz[m]> there's a way to disable that and use my registered IRC nick but I haven't bothered yet
2020-01-17 14:36:28 <pltrz[m]> _ikke_: what were some of the reasons?
2020-01-17 14:36:58 <_ikke_> pltrz[m]: performance issues with the server software
2020-01-17 14:38:11 <alpurn> Where am I failing here? https://git.riboflav.in/snippets/4
2020-01-17 14:38:27 <alpurn> I've a feeling it has to do with the signature / key signing step
2020-01-17 14:38:32 <_ikke_> How should we know?
2020-01-17 14:38:41 <_ikke_> I mean, we don't even know what is failing
2020-01-17 14:38:44 <pltrz[m]> ah I see. yeah the reference implementation of the matrix server (synapse) is written in python, which is faar from ideal. I'm looking forward to some of the other implementations to start gaining maturity
2020-01-17 14:39:30 <alpurn> _ikke_: I'm trying to include some default apk's in an armhf (raspi) build but none of the packages actually install
2020-01-17 14:41:16 <marc|gonzalez> Hello everyone. ncopa, IIUC alpine 3.11 is based on kernel 5.4 ... yet in an alpine:3.11 docker container, uname -r says 4.15.0-70-generic. I'm confused. Did I do something wrong?
2020-01-17 14:41:34 <alpurn> marc|gonzalez: docker uses your host kernel.
2020-01-17 14:41:41 <mps> pltrz[m]: also I'm looking forward for small and stable matrix server and good small client for some time but didn't found anything which looks ok for my use case
2020-01-17 14:41:54 <marc|gonzalez> alpurn: /me face palm (of course) thanks!
2020-01-17 14:41:55 <alpurn> marc|gonzalez: docker != VM
2020-01-17 14:42:09 <_ikke_> That can't be repeated often enough :)
2020-01-17 14:43:30 <marc|gonzalez> if I use a docker container of alpine, should I specify an exact version, like 3.11.2 or does it make sense to use the latest "point releases"?
2020-01-17 14:43:57 <alpurn> Pin a release if your dependent package version matters - or just pin in apk
2020-01-17 14:44:12 <_ikke_> using something like 3.12 should be pretty safe
2020-01-17 14:44:20 <_ikke_> 3.12 should still include point releases
2020-01-17 14:44:27 <alpurn> For reproducable errors lock the version - if you're a cowboy like me :latest is most fun.
2020-01-17 14:44:53 <_ikke_> sorry, 3.11 :)
2020-01-17 14:45:05 <_ikke_> :latest is still stable for alpine
2020-01-17 14:45:16 <_ikke_> Cowboys want to live on the :edge
2020-01-17 14:45:28 <marc|gonzalez> I dunno if I'm a cowboy ^_^
2020-01-17 14:45:31 <alpurn> ^^ I stand corrected - half-cocked, shootin from the hip.
2020-01-17 14:47:01 <pltrz[m]> mps: at least for clients, there is a half-decent TUI client written in Go, called gomuks. been using it for a week or so now
2020-01-17 14:47:03 <pltrz[m]> https://github.com/tulir/gomuks
2020-01-17 14:47:31 <alpurn> I've searched high and low for someone else's packaging script that uses the v2 signing requirements - doesn't seem to exist.  Perhaps when I finally figure it out I'll contribute to the docs.
2020-01-17 14:48:01 <pltrz[m]> still missing a couple features, and needs a little bit of polish but sort-of does the trick for what I look for at least
2020-01-17 14:48:37 <mps> pltrz[m]: thanks for info, will look
2020-01-17 14:51:32 <mps> looking screenshot it looks similar like my irssi setup
2020-01-17 14:53:15 <pltrz[m]> you can also toggle different parts of the ui, can leave it just with the text in the room if you like, for example. disabling the rooms list on the left, and users on right
2020-01-17 14:56:08 <mps> thanks for info, will put it notes
2020-01-17 14:56:41 <pltrz[m]> no problem :)
2020-01-17 14:56:55 <medard> Hello. Anybody here running NFS server on alpine?
2020-01-17 14:57:24 <medard> I got my hands on free dell T340 so I want to convert my current home server into pure storage unit.
2020-01-17 15:04:31 <PICCORO> mps nop.. but i'll doi later a bot that will pass all irc messages to a specific channel for devs in telegram, and wil ask here if wants feedback from..
2020-01-17 15:04:31 <PICCORO> mps: my ask are for make the already made and now long made telegram alpine groups official.. are 3 groups one engish, oher russian and other in spanish that lasted are that i manage
2020-01-17 15:06:21 <alpurn> medard: I run smb on a zfs (I know, don't do that) raidz1, NFS has produced too many complexities.
2020-01-17 15:07:38 <alpurn> My use case was specific (docker volumes over nfs) - wont do that again.
2020-01-17 15:07:55 <medard> alpurn, that setup you're using was my first idea but I never really worked with zfs
2020-01-17 15:08:27 <medard> it will be used by docker, too, besides other stuff (smb share and such)
2020-01-17 15:08:59 <alpurn> I've moved docker over to an old 3rd gen X1 Carbon - the "Linux ISO's" are stored temporarily until urrar'd and copied over smb.
2020-01-17 15:09:51 <medard> Well what would you suggest for me?
2020-01-17 15:09:52 <alpurn> pool             18T   18T  393G  98% /pool
2020-01-17 15:10:32 <alpurn> medard: you building a nas? If so don't use zfs - btr is a better choice (imo)
2020-01-17 15:10:37 <medard> I want to use this computer as NAS basically. But some of the space will be used for docker, some for smb share(or other - that is accessible on windows machine)
2020-01-17 15:10:42 <alpurn> Or UFS / XFS - whatever, just stay away from zfs :P
2020-01-17 15:11:16 <medard> what about things like glusterfs?
2020-01-17 15:11:22 <medard> (i don't know anything about that at all)
2020-01-17 15:11:57 <alpurn> It's brilliant tech but I'm old and jaded so I use what I know - smb + !zfs + docker - you're all good.
2020-01-17 15:13:46 <alpurn> When the parity checking on my nntp grabber did it's thing the machine would crawl, like I mean almost unresponsive.  Switched to CIFS, urrar'd + parity-check on a local disk, then transfer over SMB.
2020-01-17 15:17:39 <alpurn> _ikke_: if .apk's exist in cache and not in apks/armhf will they be included on boot?
2020-01-17 15:18:35 <_ikke_> You mean automatically installed?
2020-01-17 15:18:44 <_ikke_> For that, the would need to be added to /etc/apk/world
2020-01-17 15:20:10 <alpurn> Brilliant, so if I apk fetch in /media/mmcblk0/cache and update /etc/apk/world with the package names it should do -- do I also need an APKINDEX.tar.gz and / or .boot_repository in /media/mmcblk0/apks/armhf ?
2020-01-17 15:21:08 <_ikke_> I don't believe so, but I'm not sure
2020-01-17 15:22:42 <alpurn> Agh really wish I didn't have to read source for this - once I solve I'll contribute my findings.
2020-01-17 15:25:29 <medard> fhgfh
2020-01-17 15:35:46 <alpurn> I'm debugging this by swapping sd cards, running setup-alpine and `diff -r` against my manual configuration vs my automated build configuration, rather tedious.
2020-01-17 16:06:43 <ludovicchabant> hey there, it looks like Alpine recently removed abuild's -R option. What's the way to recursively build dependencies nowadays?
2020-01-17 16:23:39 <pltrz[m]> ludovicchabant: abuild -r
2020-01-17 16:25:06 <ludovicchabant> pltrz[m]: yep that's what I did but it doesn't seem to find dependencies that are other packages (to be found in sibling directories to the current one).
2020-01-17 16:25:34 <ludovicchabant> am I missing some env variable or other way to tell abuild that the parent directory contains all the other packages, or something?
2020-01-17 16:27:54 <pltrz[m]> build the dependencies with abuild, have them installed in a local repository, like $HOME/packages, add the path to the local repository to /etc/apk/repositories, then retry abuild -r on the package, maybe?
2020-01-17 16:34:59 <ludovicchabant> pltrz[m]: but each dependency might have its own dependencies, and I would end up having to manually build 20 packages in the correct order through trial and error... or maybe build each directory with -d to skip dependency checking...
2020-01-17 16:35:13 <ludovicchabant> so I guess there's no obvious replacement for -R then?
2020-01-17 16:38:15 <pltrz[m]> don't understand, are we dealing with circular dependencies?
2020-01-17 16:39:49 <_ikke_> ludovicchabant: check lua-aports
2020-01-17 16:40:00 <ludovicchabant> nope, just a local directory containing ~30 packages, and I want to build one of them, which depends on a bunch of other ones, which depend on other ones, etc. no circular dependencies.
2020-01-17 16:40:01 <_ikke_> It can list packages in build-order
2020-01-17 16:40:32 <_ikke_> So you can use that as input to decide in what order to build packages
2020-01-17 16:41:55 <_ikke_> ap builddirs dir1 dir2 dir3
2020-01-17 16:43:04 <ludovicchabant> _ikke_: thanks! is there a doc/man page somewhere?
2020-01-17 16:45:28 <_ikke_> ludovicchabant: no, not really afaik
2020-01-17 16:45:52 <ludovicchabant> ok, thanks, I'll play around with it
2020-01-17 16:46:49 <_ikke_> You should execute it in the dir with all the packages, it's kind of picky about that
2020-01-17 16:47:45 <_ikke_> Something like this should work:  ap builddirs $(ls -1)
2020-01-17 16:47:57 <_ikke_> (asuming there are no dirs with spaces in them)
2020-01-17 16:57:57 <_ikke_> (that would just generate the list in order
2020-01-17 16:57:59 <_ikke_> )
2020-01-17 17:00:17 <_ikke_> Something like this: for DIR in $( ap builddirs $(ls -1)); do ( cd "$DIR"; abuild -r; ) || break; done
2020-01-17 17:03:08 <ludovicchabant> w00t yeah that's what I ended up doing, more or less
2020-01-17 17:03:22 <ludovicchabant> it worked, thanks!
2020-01-17 18:49:08 <codebam> what fonts should I install for character support? for some reason firefox is using fontawesome to display text in sans right now
2020-01-17 18:49:24 <codebam> it's replacing "linux" with the icon for linux and etc. for all the different icon names
2020-01-17 19:20:21 <karol-evaluates-> Hi, best regards to all alpine users. I am using Alpine 3.10 updated ("virtual" version) with PostgreSQL installed with apk. I am having trouble with psql history. Basically it does not work. .psql_history id filled with _HiStOrY_V2_.
2020-01-17 19:20:38 <karol-evaluates-> Hi, best regards to all alpine users. I am using Alpine 3.10 updated ("virtual" version) with PostgreSQL installed using apk. I am having trouble with psql history. Basically it does not work. .psql_history id filled with _HiStOrY_V2_.
2020-01-17 19:24:05 <mps> karol-evaluates-: issue with libedit
2020-01-17 19:26:06 <karol-evaluates-> so i have read. Should i compile psql client myself?
2020-01-17 19:26:37 <karol-evaluates-> against libreadline?
2020-01-17 19:27:27 <mps> you can add libreadline and use LD_PRELOAD also
2020-01-17 19:27:36 <mps> forgot exact syntax
2020-01-17 19:28:18 <karol-evaluates-> ok, i will search for details. thanks
2020-01-17 19:28:59 <mps> np
2020-01-17 19:39:37 <mps> karol-evaluates-: found in my alias, export LD_PRELOAD=/usr/lib/libreadline.so.8.0
2020-01-17 19:41:42 <karol-evaluates-> @mps thanks.
2020-01-17 19:45:24 <karol-evaluates-> works like charm. Best regards
2020-01-17 19:47:22 <mps> karol-evaluates-: :)
2020-01-17 20:05:26 <okias[m]> mps any progress about Firefox crash? Just asking if.. :)
2020-01-17 20:06:18 <okias[m]> btw. thanks for that patch posted in armv7 performance discussion, I got also 40% performance increase in test-case after applying
2020-01-17 20:08:42 <mps> okias[m]: no, I didn't had time, worked on fixes, upgrades and testing for yesterday release
2020-01-17 20:09:42 <mps> but I see that firefix-esr appeared in armv7 edge
2020-01-17 20:09:52 <mps> maybe it will work
2020-01-17 20:10:01 <okias[m]> that's fine. I'll try look at it
2020-01-17 20:10:09 <okias[m]> oh, that's nice
2020-01-17 20:10:22 <mps> simply tell, time is short these days
2020-01-17 20:11:33 <mps> to rephrase Krishna from Bhagavad Gita: I'm comming as time, destroyer of humans :)
2020-01-17 20:19:55 <mps> okias[m]: re memcpy optimization for thumb, Felker is here :)
2020-01-17 20:35:07 <okias[m]> mps meanwhile I did https://gitlab.alpinelinux.org/alpine/aports/merge_requests/3130 to not lose progress
2020-01-17 20:37:14 <mps> nice, you saved me time to testing this, which I planed for this weekend :)
2020-01-17 20:42:44 <daftaupe> When running setup-alpine, can I give /mnt as the disk I want to install on ?
2020-01-17 20:43:09 <daftaupe> After having mounted my partition there
2020-01-17 20:43:43 <mps> daftaupe: you can set some env variable as dest but I forgot exact name
2020-01-17 20:44:03 <mps> on 3.11 and edge releases
2020-01-17 20:45:00 <mps> ROOTFS iirc
2020-01-17 20:45:36 <clandmeter> you can use setup-disk which can point to a mount point
2020-01-17 20:48:04 <daftaupe> But isn't setup-alpine calling setup-disk ?
2020-01-17 20:48:15 <mps> yes
2020-01-17 20:48:46 <daftaupe> Hmm it's looking for a block device in setup-alpine
2020-01-17 20:57:30 <daftaupe> Alright did use the setup-disk without issue
2020-01-17 21:19:53 <mps> daftaupe: yes, I always use setup-disk when installing. this way I prepare partitions how I prefer and not defaults
2020-01-17 21:24:23 <daftaupe> Yep sounds that's the way to do. But I thought I could prepare them and still use setup-alpine
2020-01-17 21:26:48 <mps> that should be also possible by setting some env vars but I didn't tried, so not 100% sure
2020-01-17 21:28:59 <mps> for example set SWAP_SIZE=0 will skip swap create
2020-01-17 21:34:43 <daftaupe> Alright
2020-01-17 21:35:11 <daftaupe> I see that envs variables are talked about in the setup-disk help
2020-01-17 23:34:32 <codebam> would someone be able to help me figure out why firefox is using fontawesome instead of monospace for any websites that use "monospace"?
2020-01-17 23:34:52 <codebam> fc-match "monospace" returns a proper monospace font so I'm not sure what's causing it
2020-01-17 23:35:32 <codebam> but it's really annoying. any code that has company names or fontawesome names in it gets replaced by that icon
2020-01-17 23:35:50 <codebam> for example: https://0x0.st/zFKW.png
2020-01-17 23:47:13 <MY-R> codebam, did you UNcheck in firefox font configuration "Allow pages to choose their own fonts, instead of your selections above" ?
2020-01-17 23:47:48 <codebam> MY-R yeah
2020-01-17 23:49:34 <codebam> even with it checked it does the same
2020-01-17 23:49:45 <MY-R> codebam, and your monospace font isnt bitmap one?
2020-01-17 23:50:01 <codebam> no it's like the screenshot I sent above
2020-01-17 23:51:19 <fangs2> Did anyone get iodine to work?
2020-01-17 23:51:41 <fangs2> for me iodined stops sending data after the initial negotiation
2020-01-17 23:51:45 <codebam> fangs2 I got it to work on another distro at one point
2020-01-17 23:52:20 <fangs2> codebam: thanks but I suspect its not working fine on alpine
2020-01-17 23:52:37 <fangs2> have to check on other distros later
2020-01-17 23:52:44 <codebam> oh okay
2020-01-17 23:53:14 <fangs2> you know that test page? it doesnt get detected there
2020-01-17 23:53:46 <fangs2> although it does the negotiations, chooses mtu size, checks version, password and so on
2020-01-17 23:53:54 <codebam> oh, okay
2020-01-17 23:54:24 <fangs2> but then i cant ping the server or anything after dns0 comes
2020-01-17 23:54:26 <Schoentoon> I have a diskless install on a pi4 and I want to add a custom init script to init.d what would be the best method to include this in the lbu commits? Or is there some other location to put your own custom init scripts into?
2020-01-17 23:54:48 <MY-R> codebam, did you try install "ttf-dejavu" and check if that will change anything? dejavu is default font which firefox should use as fallback but still weird that by uncheck that option in fonts you cant enforce own font...
2020-01-17 23:55:02 <codebam> installing that now
2020-01-17 23:56:09 <codebam> oh that fixed it
2020-01-17 23:56:11 <codebam> thanks!
2020-01-17 23:56:50 <fangs2> so, did the iodine server work for anyone here?
2020-01-17 23:57:02 <MY-R> codebam, did it install only dejavu or some extra package?
2020-01-17 23:57:13 <codebam> MY-R only ttf-dejavu
2020-01-17 23:57:15 <fangs2> I see its on the testing repo
2020-01-17 23:57:46 <fangs2> and maybe this means its not fully working
2020-01-18 00:00:49 <codebam> do you guys sudo or doas?
2020-01-18 00:01:08 <MY-R> su :D
2020-01-18 00:01:17 <codebam> oh lol nice
2020-01-18 00:03:06 <codebam> what does this mean? "Then you go into the src/linux-VER and edit the config file. Copy the .config file overriding the config-NAME.ARCH in the srcdir." https://wiki.alpinelinux.org/wiki/Custom_Kernel#Configuring_kernel
2020-01-18 00:03:27 <codebam> I cd'd into the folder, now what am I supposed to copy where?
2020-01-18 00:04:53 <MY-R> codebam, use alpine kernel config and copy it to that dir as .config
2020-01-18 00:05:07 <codebam> oh okay cool
2020-01-18 00:06:24 <codebam> is this the right config? the config-lts.x86_64 didn't have 64 bit kernel checked off when I ran `make menuconfig`?
2020-01-18 00:06:41 <codebam> oh, never mind
2020-01-18 00:07:25 <fangs2> so, nobody knows about iodine on alpine?
2020-01-18 00:07:51 <MY-R> in aports/main/linux-lts after run for a moment "abuild -rK" then ctrl + c you copy "cp -f config-lts.x86_64 src/linux-5.4/.config" and "cd src/linux-5.4/" and "make menuconfig"
2020-01-18 00:07:56 <MY-R> codebam, ^
2020-01-18 00:08:10 <codebam> yeah I did that. got it now. thank you though! :)
2020-01-18 00:08:36 <mps> MY-R: src/build-5.4
2020-01-18 00:09:28 <MY-R> then copy back of course to mv -f .config ../../config-vanilla.x86_64 , bumping up revision to 100 (so will know that is your kernel ;) )
2020-01-18 00:10:15 <MY-R> mps, dir name changed in 5.4? didnt compile any custom kernel yet on it
2020-01-18 00:10:28 <mps> src/build-lts.x86_64/
2020-01-18 00:10:56 <mps> vanilla is replaced by lts
2020-01-18 00:11:30 <MY-R> ye ye sorry, was copy/changing that in fly
2020-01-18 00:11:32 <mps> but src/build-xxx is where the kernel is built
2020-01-18 00:11:33 <codebam> has anyone got podman working on alpine? when I start a container I get this error:
2020-01-18 00:11:35 <codebam> ERRO[0000] "container_linux.go:346: starting container process caused \"process_linux.go:297: applying cgroup configuration for process caused \\\"mountpoint for devices not found\\\"\": OCI runtime error"
2020-01-18 00:11:43 <mps> MY-R: np :)
2020-01-18 00:12:10 <mps> for example, src/build-virt.x86_64/
2020-01-18 00:13:04 <attos2> I have a problem with the kernel module serial_ir which was enabled in the new kernel 5.4.12-1-lts recently. For further context please have a look at https://gitlab.alpinelinux.org/alpine/aports/issues/11135. When loading the module there is a "port 03f8 already in use" error (complete error message see "https://paste.debian.net/1126480/"). "setserial /dev/ttyS0 uart none" doesn't work, because of
2020-01-18 00:13:06 <attos2> "setserial: illegal UART type: none" error. Unloading the serial module already using this port doesn't work either, because the serial module seems to be compiled into the kernel (CONFIG_SERIAL_8250=y).
2020-01-18 00:13:34 <codebam> does alpine have cgroups?
2020-01-18 00:13:46 <codebam> is that why I'm getting that error
2020-01-18 00:15:22 <MY-R> mps, nah, I'm using /linux-5.4/ dir to create there .config and after use it to build package
2020-01-18 00:15:59 <codebam> yeah you just copy it back after you're done modifying it
2020-01-18 00:16:02 <codebam> :)
2020-01-18 00:16:19 <mps> MY-R: that's ok if you building kernel for you, but not ok for alpine
2020-01-18 00:16:57 <MY-R> mps, I dont care about anyone else if doing custom one just for self :D
2020-01-18 00:17:27 <mps> for some of my boxes I build on totally different environment
2020-01-18 00:18:05 <mps> strait from upstream, and tweaked for particular use cases
2020-01-18 00:18:27 <mps> and adding patches I need
2020-01-18 00:19:11 <MY-R> ye but we talking just about modification current kernel config to just change/add few things
2020-01-18 00:20:11 <mps> then the good option would be: cd main/linux-lts ; abuild deps unpack prepare
2020-01-18 00:20:32 <fangs2> how one can save money on amazon ec2 using alpine?
2020-01-18 00:20:33 <MY-R> and using default Alpine config so the one in build-lts.x86_64 is same as default
2020-01-18 00:20:37 <mps> cd src/build-lts.x86_64 ; make menuconfig
2020-01-18 00:20:40 <mps> cd -
2020-01-18 00:20:56 <mps> abuild build rootpkg index
2020-01-18 00:21:31 <mps> good idea is to bump pkgrel
2020-01-18 00:22:39 <mps> src/build-lts.x86_64 is example, it can be something else if build on other arches or flavour
2020-01-18 00:24:14 <codebam> what provides libcgroup on alpine?
2020-01-18 00:24:40 <MY-R> I'm bumping rev to 100 just for self and removing from APKBUILD virt.x86_64 stuff because dont need it
2020-01-18 00:24:47 <mps> :)
2020-01-18 00:25:19 <mps> codebam: /etc/init.d/cgroups start
2020-01-18 00:26:27 <codebam> hmm "mount: /sys/fs/cgroup/unified: none already mounted or mount point busy."
2020-01-18 00:26:33 <codebam> going to try enabling it and rebooting
2020-01-18 00:26:38 <MY-R> mps, ye that Ctrl + C HACK on wiki somebodu could replace with "abuild deps unpack prepare" ...
2020-01-18 00:27:33 <mps> I noticed it but didn't changed, don't want to delete someone efforts :)
2020-01-18 00:28:00 <MY-R> just adjust!
2020-01-18 00:28:10 <mps> is it on wiki or gitlab, I forgot
2020-01-18 00:28:17 <MY-R> wiki
2020-01-18 00:28:50 <mps> I don't know wiki syntax so never wrote anything there, though I created account there
2020-01-18 00:29:34 <mps> tried, but stepped back soon
2020-01-18 00:29:34 <MY-R> is already prety outdated howto, edited on 18 March 2018
2020-01-18 00:30:01 <mps> ah, so you are one which should fix it :)
2020-01-18 00:30:17 <MY-R> :D
2020-01-18 00:30:24 <MY-R> ye will try someday
2020-01-18 00:30:53 <codebam> that worked, thanks mps! podman works now
2020-01-18 00:31:07 <codebam> have to manually modprobe tun, but it works after I do that
2020-01-18 00:31:08 <mps> codebam: you are welcome
2020-01-18 00:31:31 <mps> put tun in /etc/modules, it will be loaded on boot
2020-01-18 00:32:36 <MY-R> already got text file with steps to install full disk encryption Alpine with GRUB with encrypted /boot (UEFI and MBR/BIOS) without multiple passphrase questions
2020-01-18 00:33:26 <MY-R> so would be cool to lower amount of steps needed to do it and publish
2020-01-18 00:34:43 <mps> cat guide-encrypted-grub-alpine.txt | tpaste :)
2020-01-18 00:35:40 <MY-R> if very pre alpha stage so is more like a notepad so hope tomorrow will be readable to put on tpaste :D
2020-01-18 00:51:33 <MY-R> mps, http://tpaste.us/V4md
2020-01-18 00:52:18 <codebam> do password fields crash firefox-esr?
2020-01-18 00:52:25 <codebam> or is it just my profile?
2020-01-18 00:52:29 <MY-R> with BIOS need replace all "sda2" to "sda1" or whatever there is but already tested it in qemu with UEFI and working
2020-01-18 00:53:03 <codebam> the flatpak works, but it's using a different profile so I can't tell
2020-01-18 00:53:04 <MY-R> codebam, you are unlucky one who got that too :\
2020-01-18 00:53:20 <codebam> so it doesn't work?
2020-01-18 00:55:13 <MY-R> not sure, on edge you could try "firefox" testing package
2020-01-18 01:01:11 <mckaygerhard> _ikke_ account registration on wiki have been deactivated?
2020-01-18 01:02:51 <mps> MY-R: nice :)
2020-01-18 01:04:10 <MY-R> is nice when working :D
2020-01-18 01:05:39 <mps> you took my joke about tpaste too seriously
2020-01-18 01:06:04 <MY-R> add some real notes, few examples and maybe will be useful and after update it to keep up changes :D
2020-01-18 01:06:24 <mps> uhm, just discovered that ctrl+shift+w quits firefox :/
2020-01-18 01:06:37 <MY-R> told you that got it messy but got :D
2020-01-18 01:07:11 <MY-R> hehe, for me is enough alt + q (i3wm bind)
2020-01-18 01:07:17 <mps> sure, good starting
2020-01-18 01:08:00 <mps> hehe, awesome wm is smarter :) ignores alt+q
2020-01-18 01:08:06 <MY-R> tbh such a pastes are more redable for me than wiki/docs :D
2020-01-18 01:08:40 <MY-R> too smart! ;)
2020-01-18 01:08:44 <mps> TBH, I really agree with you about this, I prefer text/plain
2020-01-18 01:09:10 <mps> maybe markdown sometimes
2020-01-18 01:09:39 <MY-R> yeah
2020-01-18 01:24:56 <codebam> how can I get the list of all the packages I have in world?
2020-01-18 01:25:08 <codebam> never mind
2020-01-18 01:25:15 <codebam> /etc/apk/world
2020-01-18 01:25:55 <codebam> what does alpine use for crond?
2020-01-18 01:26:19 <mps> busybox applet crond
2020-01-18 01:26:26 <MY-R> codebam, in world got packages which you installed manualy
2020-01-18 01:26:28 <codebam> oh cool
2020-01-18 01:26:54 <codebam> MY-R oh yeah right
2020-01-18 01:26:58 <MY-R> apk info will show all installed
2020-01-18 01:27:13 <codebam> oh cool okay
2020-01-18 06:32:41 <_ikke_> mckaygerhard: not that I'm aware of
2020-01-18 06:44:53 <codebam> what graphics package do we need for webrender to work in firefox?
2020-01-18 06:44:59 <codebam> it's saying unsupported hardware, not crashing
2020-01-18 06:45:09 <codebam> my hardware is supported though, it's intel
2020-01-18 06:45:17 <codebam> I've also used webrender on other distros
2020-01-18 09:20:58 <PhoenixMage> Anyone know where the Dockerfiles are kept for the alpine linux docker images for things like gitlab-runner?
2020-01-18 09:35:19 <russkel> PhoenixMage, https://gitlab.alpinelinux.org/alpine/infra
2020-01-18 09:35:21 <russkel> looks like
2020-01-18 09:38:30 <_ikke_> This is the source for the docker images: https://gitlab.alpinelinux.org/alpine/infra/docker
2020-01-18 09:39:13 <_ikke_> These are all the images: https://hub.docker.com/u/alpinelinux
2020-01-18 09:39:24 <_ikke_> PhoenixMage: ^
2020-01-18 09:59:00 <PhoenixMage> Thanks _ikke_
2020-01-18 20:04:12 <mrgreen> Running Alpine in VM but having trouble with fonts in firefox, could not find any thing on alpine wiki wondered if anyone could help?
2020-01-18 20:05:58 <tugrik> mrgreen: I can't help, but best to more specific on "having trouble with fonts" and see if you pique anyone's interest
2020-01-18 20:06:40 <mrgreen> Font rendering in firefox is really bad, loaded many different fonts with no effect
2020-01-18 20:07:51 <tehcloud> I installed basically every font package
2020-01-18 20:08:04 <tehcloud> now I'm not having too many issues there
2020-01-18 20:08:20 <tehcloud> but yes, fonts are somewhat painful in this distribution when you first install
2020-01-18 20:09:07 <mrgreen> Usually load bitstream and dejavu but I could not find anything on setting up locales etc...
2020-01-18 20:10:21 <Cogitri> No problems for me w/ Firefox standard settings
2020-01-18 20:11:48 <mrgreen> Might be Virtaulbox
2020-01-18 20:11:59 <mrgreen> *virtualbox
2020-01-18 20:25:11 <jburt_> I'm feeling pretty dumb right now. Anyone have any guidance on installing alpine to an sd card from the .tar.gz archives? None of the tools I've previously used will recognize the .tar.gz, and manually formatting to FAT32 and extracting the files fails to boot.
2020-01-18 20:25:22 <feuerteufel> Hi @ all
2020-01-18 20:26:00 <AinNero> feuerteufel: i like your nick
2020-01-18 20:27:33 <feuerteufel> i don't remeber ... someone said so many years ago, so I get stuck with that :-))
2020-01-18 20:29:34 <feuerteufel> I'm thinking about rebooting a alpine Server, for what reason I have to do it?
2020-01-18 20:29:45 <AinNero> kernel upgrade
2020-01-18 20:29:52 <feuerteufel> Ok on Kernel change
2020-01-18 20:30:09 <feuerteufel> But something else?
2020-01-18 20:30:39 <feuerteufel> e.g alpine-base or so?
2020-01-18 20:32:05 <AinNero> pid is is busybox, and doesnt have much functionality
2020-01-18 20:32:13 <AinNero> *pid 1
2020-01-18 20:36:45 <feuerteufel> Ok, one more  is when xen-hypervisor get's updated ...
2020-01-18 20:40:24 <gjabell> Anyone know of any existing tools for automatically keeping packages in a custom alpine repo in sync with upstream?
2020-01-18 20:40:46 <AinNero> feuerteufel: essentially, everything before PID 1 changing requires an reboot
2020-01-18 20:40:56 <gjabell> For instance, building a package in testing for a stable machine
2020-01-18 20:46:24 <feuerteufel> AinNero: But how do I find that out?
2020-01-18 20:50:33 <AinNero> grub and extlinux have update scripts
2020-01-18 20:51:20 <AinNero> they are usually called when stuff in /boot changes
2020-01-18 20:51:47 <AinNero> hooking into them or detecting whether they have been called might be an idea
2020-01-18 20:52:10 <feuerteufel> Yes, I know
2020-01-18 20:53:20 <feuerteufel> I was looking for somthing like "/var/run/reboot-required" on Debian
2020-01-18 20:54:49 <mps> jburt_: which arch
2020-01-18 21:25:48 <jburt_> armhf
2020-01-18 21:29:09 <mps> jburt_: I created small script for RPi armhf few days ago which formats SD card, make partiotions, unpack tarball and tweak boot parameters
2020-01-18 21:29:41 <mps> I needed it for repetitive rpi release tests
2020-01-18 22:25:07 <codebam> not sure what you guys did
2020-01-18 22:25:15 <codebam> but the new mesa updates fixed firefox webrender :)
2020-01-18 22:25:17 <codebam> thanks!
2020-01-18 22:28:58 <minecrell> codebam: on edge?
2020-01-18 22:29:06 <codebam> yeah edge
2020-01-18 22:29:08 <MY-R> codebam, maybe just some package was missing and since mesa package was changed in last commit then it added
2020-01-18 22:29:17 <MY-R> codebam, what about firefox crashes on password fields?
2020-01-18 22:29:19 <codebam> nah it was throwing glx errors
2020-01-18 22:29:25 <codebam> MY-R still broken sadly :(
2020-01-18 22:29:39 <codebam> I use the flatpak if I need to login to websites... lol
2020-01-18 22:30:04 <minecrell> the Mesa code/version is exactly the same, so what MY-R said seems most likely
2020-01-18 22:30:10 <kreal> Hey! Can I create an apkvol from a non-diskless, non-raspi system and then put it on an SD card to use as a from-ram system?
2020-01-18 22:30:14 <codebam> downloading the appimage now though in hopes that it will work with webrender. they disabled it somehow on the flatpak release
2020-01-18 22:30:30 <codebam> minecrell oh okay
2020-01-18 22:30:34 <codebam> well anyways, it works now
2020-01-18 22:31:02 <gjabell> hmm mine doesn't crash on the password fields anymore
2020-01-18 22:31:31 <codebam> do appimages work on alpine?
2020-01-18 22:32:04 <codebam> when I'm trying to execute it I get "no such file or directory" but I can't get it to execute
2020-01-18 22:33:01 <codebam> I thought it would work since it would package the runtime, no?
2020-01-18 22:33:01 <MY-R> codebam, what gou and xorg driver do you using?
2020-01-18 22:33:11 <MY-R> gpu*
2020-01-18 22:33:11 <codebam> I'm not using Xorg. using wayland with sway
2020-01-18 22:33:20 <codebam> I use intel
2020-01-18 22:33:23 <codebam> intel graphics
2020-01-18 22:33:32 <codebam> mesa-dri-intel or whatever it was called
2020-01-18 22:34:48 <MY-R> codebam, and which from those you do NOT have installed: mesa-glapi mesa-gl mesa-egl mesa-gles wayland-libs-eglwayland-libs-egl
2020-01-18 22:36:05 <codebam> I have all of those installed
2020-01-18 22:36:12 <codebam> checked using grep on apk list
2020-01-18 22:36:28 <codebam> how do I run this executable?
2020-01-18 22:36:55 <MY-R> and got "glu" package too?
2020-01-18 22:37:41 <codebam> no what's that one for?
2020-01-18 22:37:58 <MY-R> try maybe install and check if firefox crash
2020-01-18 22:38:10 <MY-R> glu-9.0.1-r0 - Mesa OpenGL Utility library
2020-01-18 22:38:13 <codebam> it doesn't crash, it was just disabling webrender
2020-01-18 22:38:15 <codebam> but now it works
2020-01-18 22:38:23 <codebam> well, webrender was crashing and it was falling back
2020-01-18 22:38:26 <codebam> I installed glu now though
2020-01-18 22:38:36 <MY-R> on password field?
2020-01-18 22:39:14 <codebam> yeah still crashing
2020-01-18 22:39:21 <MY-R> :\
2020-01-18 22:40:00 <codebam> huh wait it looks like it has something to do with the profile looking for stuff inside flatpak
2020-01-18 22:40:08 <codebam> and getting permission denied
2020-01-18 22:40:23 <codebam> not exactly sure why it's trying to do that...
2020-01-18 22:40:35 <MY-R> codebam, you checked those packages exactly how? apk list |grep <package> ?
2020-01-18 22:40:44 <codebam> yeah
2020-01-18 22:40:53 <codebam> or rg actually (ripgrep)
2020-01-18 22:41:06 <MY-R> but you know that apk list will show you all packages in repo and not just installed ones?
2020-01-18 22:41:44 <MY-R> use apk list -I or apk info
2020-01-18 22:42:48 <codebam> I don't really get why it's trying to open the wrong profile. I'm going to start fresh
2020-01-18 22:43:09 <codebam> it's crashing when I switch to the password field saying: (/usr/lib/firefox/firefox:22221): dconf-WARNING **: 17:42:23.743: Unable to open /var/lib/flatpak/exports/share/dconf/profile/user: Permission denid
2020-01-18 22:43:16 <codebam> obviously because it's not a flatpak package
2020-01-18 22:43:40 <codebam> I have firefox as a flatpak, but I don't know where it would be getting that from since this is a brand new profile
2020-01-18 22:44:40 <codebam> I don't get it...
2020-01-18 22:47:12 <codebam> I've completely removed the firefox flatpaks now but it's still giving that error
2020-01-18 22:47:34 <codebam> is there somewhere else firefox would be searching for config files other than ~/.mozilla and ~/.cache/mozilla ?
2020-01-18 22:54:30 <codebam> why is this the only solution... https://github.com/flatpak/flatpak/issues/1496
2020-01-18 22:54:38 <codebam> why does firefox even need access to those folders
2020-01-18 22:54:45 <codebam> I think this is a firefox bug
2020-01-18 22:59:32 <codebam> it looks like that's not an error but a warning
2020-01-18 22:59:38 <codebam> the real error is below
2020-01-18 22:59:43 <codebam> https://0x0.st/zFLX.txt
2020-01-19 00:22:47 <shunter> It just clicked that I can run proprietary apps (read: glibc-dependent binaries) through Docker.
2020-01-19 00:48:16 <atkka> hello
2020-01-19 02:17:46 <neneko> Can I install Alpine on a VPS running Debian? Like, by running setup-alpine from inside a chroot maybe?
2020-01-19 03:51:56 <codebam> does anyone else use alpine that has that problem with firefox with password fields? I'm very confused because I'm logged into some websites so obviously it worked at some point
2020-01-19 03:52:51 <codebam> huh if I right click on the password field it doesn't crash
2020-01-19 03:54:39 <codebam> until I type into it
2020-01-19 04:13:09 <zalt_> how do i automatically load a kernel module
2020-01-19 04:26:49 <neneko> zalt_: /etc/modules-load.d
2020-01-19 04:27:39 <zalt_> yeah, i figured that out, now i want to create a service
2020-01-19 04:28:20 <zalt_> i'm skimming through openrc's docs, does it support oneshot kind of services?
2020-01-19 04:31:56 <shunter> As in, running a service without it being enabled?
2020-01-19 04:35:17 <zalt_> i just need a script to run at startup
2020-01-19 04:35:40 <zalt_> and, it's not daemon that runs in the background, so, the idea of stop and restart doesn't make sense
2020-01-19 04:38:22 <shunter> This *might* be a hack, but if it's a one-time program that runs on boot, you could get away with tacking it onto /etc/inittab.
2020-01-19 04:39:57 <neneko> Or put "@reboot /path/to/script.sh" in your crontab.
2020-01-19 04:53:50 <shunter> I'd call using crontab better over inittab
2020-01-19 05:16:42 <zalt_> what about /etc/local.d ?
2020-01-19 05:43:08 <zalt_> i just used that, it seems to do what i want
2020-01-19 08:18:43 <mojca> I would like to ask for some help/advice. I've never used Alpine before. I tried to compile some software on 3.11 using clang, and it fails to find emmintrin.h
2020-01-19 08:18:53 <mojca> Here's the full log: https://build.contextgarden.net/#/builders/60/builds/1
2020-01-19 08:19:23 <_ikke_> mojca: https://pkgs.alpinelinux.org/contents?file=emmintrin.h&path=&name=&branch=edge&arch=x86_64
2020-01-19 08:19:33 <_ikke_> mojca: seems like you need zig-dev
2020-01-19 08:19:35 <mojca> I noticed that this file is (probably) present in clang-devel, but it seems strange that a file needed to compile some piece of software is only in devel package
2020-01-19 08:20:04 <_ikke_> oh, no, that's just zig related what I said
2020-01-19 08:20:16 <_ikke_> all header files are in -dev packages
2020-01-19 08:20:38 <_ikke_> But I understand what you mean
2020-01-19 08:20:41 <mojca> I do understand that header files for, say, X11 would be in X11-dev.
2020-01-19 08:21:04 <mojca> What I don't understand is why "math.h" would be in clang-dev, for example.
2020-01-19 08:21:08 <mojca> The compiler seems pretty useless if header files are missing.
2020-01-19 08:21:45 <mojca> I need to admit that I have no clue what clang-dev is. But I would imagine it was something for people who want to ... who knows, build libraries on top of compiler?
2020-01-19 08:22:13 <_ikke_> mojca: I don't know the details either, but it might be that these header files are dependent on the compiler you use?
2020-01-19 08:22:51 <mojca> Also, from the search above, math.h is in g++6, not in g++6-dev
2020-01-19 08:23:04 <mojca> Sorry, g++, not g++6
2020-01-19 08:23:38 <mojca> It simply feels wrong to me to put a header that's needed for compilation to compiler-dev.
2020-01-19 08:24:44 <mojca> again, emmintrin.h is present in gcc, it's present in mingw-w64-gcc, so it works out of the box there, it just doesn't work with clang.
2020-01-19 08:25:04 <mojca> And that part feels wrong.
2020-01-19 08:30:06 <mojca> What's the source for clang-dev? This one: https://gitlab.alpinelinux.org/alpine/aports/blob/master/main/clang/APKBUILD
2020-01-19 08:31:16 <mojca> Is the package manager the one that decides what to put in the -dev package?
2020-01-19 08:42:04 <mps> mojca: usually we follow compiler upstream 'idea' about this
2020-01-19 08:43:58 <mojca> However upstream doesn't have a git repository for clang-dev?
2020-01-19 08:44:36 <mojca> So how do you decide what goes into -dev?
2020-01-19 08:46:59 <mps> well, case by case usually. for example, as a packager of zig I followed upstream advice first and looked at some other distro how they do
2020-01-19 08:52:59 <Cogitri> mojca: Yes, the package manager decides what's in clang-dev. If there's no function overwriting the default function for the subpackage $pkgname-dev default_dev is invoked.
2020-01-19 08:53:31 <Cogitri> You can look at abuild's source for what exactly that does
2020-01-19 08:59:23 <Cogitri> I do agree that the clang-dev package doesn't make too much sense though
2020-01-19 09:24:03 <mps> Cogitri: yes, I had this dilemma when creating zig apk
2020-01-19 09:24:49 <mps> i.e. put all in 'zig' (only) or split
2020-01-19 09:33:44 <Cogitri> With clang we're having the problem that it _is_ possible to link to libclang which probably is why we have clang-dev
2020-01-19 09:36:49 <mps> anyway, I expect programmers (or developers, what a strange term) could know intricacies of that 'things'
2020-01-19 09:44:19 <mps> okias[m]: I tested yesterday !3130
2020-01-19 09:44:38 <mps> speed up is about 40%
2020-01-19 09:45:52 <mps> now my old armv7 exynos samsung chromebook is the most pleasant desktop box
2020-01-19 10:22:04 <crockwork> I have a webserver on host A that is on Network A, and I have a VPN server set up on Network A's gateway. I have a host B (which I have root on, running alpine) that is on network B, which has a route to network A. I am currently sitting on network C, which has a route to network B, and therefore host B. How do I, host C on network C, get to opening a browser on host C that will access host A's webserver?
2020-01-19 11:24:19 <coredumb> crockwork: ssh socks proxy or simple ssh port forward would work
2020-01-19 13:28:04 <Naib> https://pkgs.alpinelinux.org/package/edge/main/x86_64/mariadb is there any explanation as to to why mariadb now depends on mariadb-test. This changed from 3.10 going to 3.11
2020-01-19 13:32:53 <mps> Naib: I'm following issue for some time but don't know what to do about it. someone should ask current maintainer about this
2020-01-19 13:44:25 <Naib> mps: good to know, I am only starting to play with alpine for containers and my test builds suddently exploded when 3.11 was released. I havn't found any reason for the inclusion of the test and other larger packages. a docker build has an open issue on this as well
2020-01-19 13:47:45 <mps> Naib: I agree, this looks really strange but I don't use mariadb so don't know is it ok to remove this dependency (though I think it should be removed)
2020-01-19 14:09:14 <artok> for container, I usually recommend doing multi stage build for service that you're running
2020-01-19 14:10:06 <artok> -> keeping version updates of your service to yourself
2020-01-19 15:54:03 <sirtomato> hey
2020-01-19 15:59:31 <pltrz[m]> it seems to fail to serve the book locally
2020-01-19 15:59:31 <pltrz[m]> anybody using mdbook ?
2020-01-19 15:59:36 <pltrz[m]> "mdbook serve"
2020-01-19 15:59:41 <pltrz[m]> returns the error
2020-01-19 15:59:52 <pltrz[m]> Error: Unable to launch the server
2020-01-19 16:00:00 <pltrz[m]> Caused By: Address not available (os error 99)
2020-01-19 16:06:33 <pltrz[m]> ddevault: just noticed you happen to be the maintainer - does mdbook run without error on your setup?
2020-01-19 16:06:58 <ddevault> yes, it works
2020-01-19 16:07:04 <ddevault> that error is probably because you're using that port for something else
2020-01-19 16:19:59 <pltrz[m]> hmm throws the same msg even if I specify another port with "-p"
2020-01-19 16:20:38 <pltrz[m]> ex. "mdbook serve -p 3200"
2020-01-19 16:55:42 <MY-R> pltrz[m], did you try with: mdbook serve -n 127.0.0.1 ?
2020-01-19 18:41:22 <dsmster> is this still the correct command to update an sdcard/rpi alpine from 3.11.2 -> 3.11.3 ?  I have my LBU_MEDIA set to mmcblk0p1 - but this from the wiki: etup-bootable -u {{#latestalp:alpine|url}} /media/$LBU_MEDIA ; errors out with : /media/usb: special device /dev/usbdisk does not exist;   any ideas?
2020-01-19 19:29:53 <artok> exported LBU_MEDIA ?
2020-01-19 19:46:46 <markand> 
ACTION alpinized his raspberry

2020-01-19 19:50:14 <pltrz[m]> <MY-R "pltrz, did you try with: mdbook "> I tried -n localhost, which should be the same but I will try that too. as soon as I get home. on the street now
2020-01-19 19:51:06 <MY-R> ok!
2020-01-19 20:01:27 <markand> hmmm, I use crontab -e as root and add some lines, then reopening crontab -e shows the initial file (as if I forgot to save)
2020-01-19 20:08:03 <dsmster> artok: yes
2020-01-19 23:17:31 <pltrz[m]> MY-R: so mdbook serve -n 127.0.0.1 still no luck
2020-01-19 23:17:56 <pltrz[m]> and strangly enough, I also cannot seem to ping 127.0.0.1
2020-01-19 23:18:11 <MY-R> that is why wanted you to check that
2020-01-19 23:20:09 <pltrz[m]> super weird how can I not ping localhost? never encountered that
2020-01-19 23:20:10 <MY-R> you didnt set up network, dont you have "lo" stuff in /etc/network/interfaces?
2020-01-19 23:20:44 <pltrz[m]> i have entries for lo and wlan0 there
2020-01-19 23:20:54 <MY-R> I have no idea how you installed Alpine but you skipped some part :P
2020-01-19 23:21:11 <MY-R> auto lo
2020-01-19 23:21:12 <MY-R> iface lo inet loopback
2020-01-19 23:21:13 <MY-R> ?
2020-01-19 23:21:21 <pltrz[m]> I am using iwd, first of all
2020-01-19 23:21:53 <MY-R> and in "/etc/hosts" at least: 127.0.0.1	localhost
2020-01-19 23:22:14 <pltrz[m]> oh derp
2020-01-19 23:22:19 <pltrz[m]> ip link set lo up
2020-01-19 23:22:28 <pltrz[m]> does the trick
2020-01-19 23:22:54 <pltrz[m]> don't know why it's down to start off anyways though
2020-01-19 23:26:24 <pltrz[m]> oh, I know why
2020-01-19 23:26:38 <pltrz[m]> _facepalm_
2020-01-19 23:26:50 <pltrz[m]> I'm not using /etc/init.d/networking service
2020-01-19 23:26:58 <MY-R> ye that for sure
2020-01-19 23:27:11 <pltrz[m]> because I'm using iwd to manage wireless and also as dhcp client
2020-01-19 23:27:43 <pltrz[m]> so i turned it off, didn't think about the loopback interface though
2020-01-19 23:28:50 <MY-R> you enable it, iwd will still start before networking service
2020-01-19 23:29:00 <MY-R> just enable it*
2020-01-19 23:32:18 <MY-R> pltrz[m], still dont know why you didnt run before "setup-interfaces" or "setup-alpine" :)
2020-01-19 23:32:44 <pltrz[m]> well, actually I did
2020-01-19 23:33:21 <pltrz[m]> but then removed wpa_supplicant replaced it with iwd, disabled the networking service and replaced with iwd service, made the iwd service provide net
2020-01-19 23:33:57 <pltrz[m]> I don't like the look of init.d/networking cause it's a bunch of ifup ifdown
2020-01-19 23:34:09 <pltrz[m]> which is really not a great busybox applet
2020-01-19 23:34:45 <pltrz[m]> it's even in the docs, saying it's a C program doing the job of a shell script
2020-01-19 23:34:48 <pltrz[m]> https://github.com/mirror/busybox/blob/master/docs/ifupdown_design.txt
2020-01-19 23:34:52 <pltrz[m]> https://github.com/mirror/busybox/blob/master/networking/ifupdown.c
2020-01-19 23:35:15 <pltrz[m]> after actually spending some time reading ifupdown.c I gotta agree
2020-01-19 23:36:12 <pltrz[m]> I'd prefer it init.d/networking used some shell with individual "ip ..." commands and "udhcp ..." commands than using ifup/ifdown
2020-01-20 02:43:32 <dalias> i have this with 5.4.12 on alpine: https://bbs.archlinux.org/viewtopic.php?id=250765
2020-01-20 02:44:06 <dalias> prev with .6 and i hoped update to .12 would fix bit it didnt
2020-01-20 02:44:09 <dalias> but*
2020-01-20 06:54:46 <mps> dalias: yes, and looks like 5.4.13 also doesn't solved this
2020-01-20 09:17:51 <wrench> Has anyone successfully set up LDAP authenfication on alpine? I can't figure out how to work around the lack of NSS
2020-01-20 09:18:36 <maldridge> you could use nss
2020-01-20 09:18:47 <_ikke_> https://pkgs.alpinelinux.org/package/edge/main/x86_64/musl-nscd
2020-01-20 09:18:49 <maldridge> musl-nscd will provide the interface you need, with the exception of the shadow map
2020-01-20 09:18:52 <_ikke_> :)
2020-01-20 09:19:27 <wrench> this does not start up, as it does not find libnss_files.so
2020-01-20 09:19:29 <maldridge> not entirely sure though that pam-nss-ldap provides a correct initgroups_dyn symbol
2020-01-20 09:19:46 <maldridge> well, did you provide that file by anything?
2020-01-20 09:19:56 <maldridge> in musl files is provided by the c library directly
2020-01-20 09:33:50 <midasi> what's the reason 'linux-tools' is not available for aarch64? It's explicitely excluded in APKBUILD. https://git.alpinelinux.org/aports/tree/testing/linux-tools/APKBUILD
2020-01-20 09:35:12 <_ikke_> midasi:  c9cf21eb34fdcc7be0da6445eae47a1588dd5dfa
2020-01-20 09:35:19 <_ikke_> No details thoug wyhy
2020-01-20 09:35:22 <midasi> thx
2020-01-20 09:35:54 <_ikke_> 5aa3c8914a17b18d081bec2e4d2dedb3bbea2c32 this one disabled aarch64
2020-01-20 09:41:12 <midasi> _ikke_: thank you, I will try to build it and see what's wrong
2020-01-20 09:55:46 <wrench> turns out, the problem wtih libnss_files.so was because of an error in my configuration. Still I can't build nss-pam-ldapd with nss because netdb.h from musl lacks the NETDB_INTERNAL definition
2020-01-20 09:56:06 <wrench> *with
2020-01-20 10:12:25 <Rodal> I have a problem during virtualbix instalation. BAD SIGNATURE... :-(
2020-01-20 10:12:53 <Rodal> Iso download checksum is correct..
2020-01-20 10:14:00 <Rodal> Any idea why it doesn't work to install alpine on virtualbox? Why bad signature?
2020-01-20 10:19:42 <_ikke_> What is virtualbix?
2020-01-20 10:20:03 <clandmeter> the i and the o are close
2020-01-20 10:20:13 <_ikke_> Yes, I was expecting that
2020-01-20 10:20:32 <_ikke_> But we don't package virtualbox
2020-01-20 10:20:58 <clandmeter> Rodal: does your vm has working network/internet?
2020-01-20 10:21:12 <Rodal> I would like to install alpine on a virtualbix vm
2020-01-20 10:21:31 <Rodal> Yes it have internet
2020-01-20 10:22:46 <clandmeter> still virtualbix?
2020-01-20 10:22:51 <clandmeter> is that a thing?
2020-01-20 10:22:53 <Rodal> I downloaded iso for standard and virt and both have the same error.
2020-01-20 10:23:20 <Rodal> Sorry i am on my phone.. sorry for my bad writing
2020-01-20 10:24:03 <clandmeter> did you run apk update?
2020-01-20 10:28:03 <Rodal> Yes, gives me an ok
2020-01-20 10:29:03 <mps> midasi: https://gitlab.alpinelinux.org/alpine/aports/issues/11119
2020-01-20 10:29:22 <Rodal> 3.11.3 [/media/cdrom/apks], OK: 65 distinct pacckages availabe
2020-01-20 10:32:08 <clandmeter> Rodal: which iso did you use (filename) and what is in your repositories file?
2020-01-20 10:32:22 <mps> midasi: some tools are only for specific arches
2020-01-20 10:32:25 <clandmeter> please use pastbin or similar
2020-01-20 10:34:17 <clandmeter> Rodal: I guess you are using dl-cdn mirror?
2020-01-20 10:34:28 <clandmeter> if so can you try dl-4.alpinelinux.org
2020-01-20 10:36:06 <Rodal> I use the first mirror in the list during setup-alpine. And the list is too long for the screen to see the mirror..
2020-01-20 10:36:31 <clandmeter> edit /etc/apk/repositories
2020-01-20 10:36:45 <clandmeter> change dl-cdn to dl-4
2020-01-20 10:37:38 <Rodal> edit is not found..
2020-01-20 10:38:09 <clandmeter> do you know how vi works?
2020-01-20 10:40:05 <Rodal> Yes, ok in this config it is: /media/cdrom/apks now
2020-01-20 10:40:49 <clandmeter> did you run setup-alpine?
2020-01-20 10:46:42 <Rodal> Yes and now i see the dl-cdn mirror, i changed it to http://dl-4 alpinelinux.org, and how could i save bow in vi?
2020-01-20 10:47:49 <clandmeter>  :wq
2020-01-20 10:48:05 <_ikke_> /kick clandmeter
2020-01-20 10:48:31 <clandmeter> ?
2020-01-20 10:48:38 <_ikke_> ":wq"
2020-01-20 10:48:51 <clandmeter> i dont follow
2020-01-20 10:48:56 <_ikke_> was a joke
2020-01-20 10:49:03 <_ikke_> not funny when I have to explain
2020-01-20 10:49:38 <_ikke_> :q -> quit. I thought, lets help you :)
2020-01-20 10:50:27 <clandmeter> ok i get it. sorry that i killed your joke :p
2020-01-20 10:52:37 <clandmeter> we dont even ship virtualbox?
2020-01-20 10:53:14 <clandmeter> ... on virtualbox...
2020-01-20 10:53:31 <clandmeter> i should not do 2 things at the same time.
2020-01-20 10:54:09 <Rodal> Ok now i got another error
2020-01-20 10:55:02 <Rodal> Http://dl-4.alpinelinux.org: Bo such file or directory
2020-01-20 10:55:12 <Rodal> Maybe i did something wrong now
2020-01-20 10:55:59 <clandmeter> which command did you run initially when you got the BAD SIGNATURE error?
2020-01-20 10:57:36 <Rodal> In alpine-setup after i chose installation mode (sys)
2020-01-20 10:58:11 <clandmeter> you did not get an error before?
2020-01-20 10:58:30 <clandmeter> you said you choose the first mirror out of the list
2020-01-20 10:58:41 <clandmeter> which is one of the first steps
2020-01-20 10:58:59 <clandmeter> iirc
2020-01-20 11:00:56 <Rodal> Yes and then i got: Updating repository indexes... done
2020-01-20 11:01:21 <clandmeter> but you said your repositories file only lists cdrom line?
2020-01-20 11:02:01 <Rodal> Yes it was after a fresh new start from the vm before i run setup-alpine..
2020-01-20 11:03:00 <clandmeter> i asked you to edit the repositories file after you run setup-alpine
2020-01-20 11:04:05 <clandmeter> <Rodal> 3.11.3 [/media/cdrom/apks], OK: 65 distinct pacckages availabe
2020-01-20 11:04:25 <clandmeter> this indicates only a single line in your repositories file
2020-01-20 11:05:59 <Rodal> I restarted the vm, run the setup, choose the first repo, no error for this, than after i chose "sys" i got error: bad signature, that is the status now.
2020-01-20 11:06:18 <Rodal> Abd now i should change the repo for the new mirror right?
2020-01-20 11:06:30 <clandmeter> choose nr 4
2020-01-20 11:06:33 <clandmeter> not nr 1
2020-01-20 11:06:37 <clandmeter> and see if that helps
2020-01-20 11:06:48 <Rodal> Ok i will try it now
2020-01-20 11:07:04 <clandmeter> better to just restart the vm and rerun setup-alpine
2020-01-20 11:07:17 <clandmeter> i dont think setup-apkrepos is smart enough to run twice
2020-01-20 11:08:26 <Rodal> Ok thanks
2020-01-20 11:11:34 <Rodal> Yeah... Error: linux-virt-5.4.12-r1: BAD signature
2020-01-20 11:12:08 <clandmeter> this is with dl-4?
2020-01-20 11:12:14 <Rodal> Yes
2020-01-20 11:12:22 <clandmeter> then your network is broken somehow
2020-01-20 11:12:57 <Rodal> And how could i see it? I never heard about broken network..
2020-01-20 11:14:25 <clandmeter> which iso did you download?
2020-01-20 11:14:29 <clandmeter> x86_64?
2020-01-20 11:14:39 <clandmeter> can you give the full name
2020-01-20 11:14:48 <Rodal> Of course
2020-01-20 11:15:25 <Rodal> Alpine-virt-3.11.3-x86_64
2020-01-20 11:15:39 <Rodal> Gpg verifyied
2020-01-20 11:16:16 <Rodal> And in virtualbox i setup bridge networt..
2020-01-20 11:16:30 <clandmeter> wget http://dl-cdn.alpinelinux.org/alpine/v3.11/main/x86_64/linux-virt-5.4.12-r1.apk
2020-01-20 11:16:34 <clandmeter> can you try that?
2020-01-20 11:23:41 <Rodal> I will try it.. maybe it is a virtualbox problem..
2020-01-20 11:29:18 <Rodal> Wget: bad address 'dl-cdn.alpinelinux.org'
2020-01-20 11:30:44 <_ikke_> dns issue
2020-01-20 11:32:38 <Rodal> Oh yes now wget could download it and saved it..
2020-01-20 11:33:23 <Rodal> And now what does this mean?
2020-01-20 11:33:46 <_ikke_> what?
2020-01-20 11:38:30 <Rodal> I could download the package with wget but in alpine-setup it is still bad signature..
2020-01-20 11:54:46 <PICCORO> hellow all, TRINITY DESKTOP (a continuation of KDE3)  Added initial support for build with musl libc, so now alpine linux can build a complet4e no-systemd desktop without cut off features! i suggest make it default desktop and work aside with them
2020-01-20 11:57:01 <nepugia> Interesting indeed, but before alpine declares itself a desktop distro with tde, maybe it should be packages first? :3
2020-01-20 12:03:06 <Rodal> Clandmeter: any other suggestions?
2020-01-20 12:05:55 <clandmeter> Rodal: no other than check your local network
2020-01-20 12:06:02 <clandmeter> can you install the downloaded pkg?
2020-01-20 12:06:09 <clandmeter> apk add ./xxx.apk
2020-01-20 12:11:57 <okias[m]> pretty request to gitlab maintainer -> since it's not nice to be tracked everywhere, could you disable reCAPTCHA from Google?
2020-01-20 12:13:02 <okias[m]> it's annoying, since it looks like often when you paste URL, you're required to verify by reCAPTCHA, but since Firefox has privacy protection on, I must disable it for reCAPTCHA and re-enable afterwards. Is there any other solution which could be used? Something like answer question?
2020-01-20 12:13:34 <clandmeter> okias[m]: im afraid not
2020-01-20 12:13:39 <nepugia> the raison detre of google recaptcha is that google has broken normal captchas
2020-01-20 12:14:13 <nepugia> a whitelist could work though, but that is hard to maintain, and still only leaves some people without recaptcha :)
2020-01-20 12:14:18 <clandmeter> okias[m]: you have to use recaptcha when you paste links?
2020-01-20 12:14:58 <okias[m]> I just filled bugreport and when adding link, it asked me to be verified (so I realized that I need disable privacy protection to get recaptcha loaded and visible)
2020-01-20 12:15:34 <okias[m]> and it's not first time. gitlab itself does that, but I'm sure they'll not care, but at least smaller projects could care little bit about privacy of their devs/users
2020-01-20 12:16:03 <okias[m]> recaptcha doesn't charge money of it's use, but g. doesn't offer it for free...
2020-01-20 12:16:46 <MY-R> Rodal, maybe setup some public dns server in /etc/resolv.conf (after when you configure network during "setup-alpine" you can ALT+F2 and login on second tty and do: echo "8.8.8.8" > /etc/resolv.conf
2020-01-20 12:16:55 <Rodal> Yeah it is still the same with wrong signatures for other distributions..
2020-01-20 12:17:22 <clandmeter> its hard to find a balance between 100% privacy minded and preventing spam.
2020-01-20 12:20:36 <okias[m]> clandmeter I like postmarketOS wiki.. they have simple question - what is pmOS based on... I must say it's not very sofisticated solution, but it works. When you'll have like 5 simple question, spambots won't fill it right
2020-01-20 12:21:08 <clandmeter> okias[m]: if you can provide a plugin for gitlab that works, i will be glad to use it.
2020-01-20 12:21:35 <clandmeter> okias[m]: you can maybe develop one yourself :)
2020-01-20 12:21:57 <okias[m]> clandmeter not much my focus field, but i'll look for alternative :)
2020-01-20 12:24:27 <mps> okias[m]: you can open account on gitlab.a.o and skip google
2020-01-20 12:25:16 <okias[m]> https://framagit.org/Mindiell/captcha but I think it's in early stage :( (so not production yet)
2020-01-20 12:25:24 <okias[m]> LibreCaptcha sounds cool
2020-01-20 12:26:21 <mps> and if you use google to login in gitlab.a.o you are already google user, aren't you
2020-01-20 12:26:33 <okias[m]> mps I use gitlab connection, so if I registered directly, it would skip these recaptchas?
2020-01-20 12:27:06 <okias[m]> mps here, be sure, I don't use google.. not even on my phone LineageOS phone
2020-01-20 12:27:15 <mps> okias[m]: I've account on gl.a.a and never seen any captcha
2020-01-20 12:27:22 <clandmeter> okias[m]: do you get a recaptcha screen when that happens?
2020-01-20 12:27:31 <clandmeter> or can it be something else?
2020-01-20 12:27:52 <clandmeter> i dont think recaptcha does a check on contents
2020-01-20 12:28:04 <clandmeter> its probably askimet which does that.
2020-01-20 12:28:05 <mps> okias[m]: btw, FYI pmOS is alpine based
2020-01-20 12:28:15 <okias[m]> mps do you have strict privacy protection in Firefox enabled?
2020-01-20 12:28:27 <okias[m]> since otherwise recaptcha capture your mouse movement and analyze it and then you don't have to confirm
2020-01-20 12:28:29 <okias[m]> that's the funny part, free behavioral data to G. :D
2020-01-20 12:28:48 <mps> I think I have, and umantrix, ghostery and some other plugins
2020-01-20 12:28:51 <okias[m]> mps https://wiki.postmarketos.org/wiki/Google_Nexus_7_2012_(asus-grouper) I know, I know :P
2020-01-20 12:29:06 <clandmeter> okias[m]: can you give me a screenshot of what happens?
2020-01-20 12:29:27 <okias[m]> ok, give me a few sec
2020-01-20 12:31:06 <okias[m]> 
ACTION uploaded an image: Snímek z 2020-01-20 13-30-00.png (144KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/SJBAqLEtesxtazoSPAFRLXDU >

2020-01-20 12:32:07 <clandmeter> okias[m]: thx, can you tpaste or pastbin it?
2020-01-20 12:32:16 <clandmeter> your issue
2020-01-20 12:32:27 <clandmeter> maybe i can get it from the issue
2020-01-20 12:32:34 <okias[m]> you don't see link?
2020-01-20 12:32:49 <okias[m]> it's there (I pushed it) this is just edit
2020-01-20 12:33:04 <okias[m]> and screen what happen when I try add additional link to the message
2020-01-20 12:34:01 <clandmeter> heh
2020-01-20 12:34:06 <clandmeter> i also get a captcha
2020-01-20 12:34:10 <clandmeter> and im admin :)
2020-01-20 12:36:31 <okias[m]> your preciousss dataaa (in Glum voice, from Google director of reCaptcha development)
2020-01-20 12:36:52 <Rodal> Thanks for helping clandmaster.. maybe it is my hardware or virtualbox..
2020-01-20 12:37:16 <clandmeter> thats a nice name
2020-01-20 12:37:20 <clandmeter> thanks! :)
2020-01-20 12:39:29 <mps> hm, logged out, logged in and don't see any re/captcha
2020-01-20 12:39:54 <mps> checked strict security is on
2020-01-20 13:01:58 <MY-R> mps, and did you create issue with any link inside and tried save?
2020-01-20 13:02:33 <MY-R> look here: https://gitlab.com/gitlab-com/support-forum/issues/3604
2020-01-20 13:03:14 <mps> MY-R: can't remember if I create issue but MR for sure
2020-01-20 13:03:41 <mps> could be this because I activated 2FA
2020-01-20 13:03:44 <MY-R> and guy from support "Stan Hu" basically white listing them... wtf
2020-01-20 13:07:36 <MY-R> I know google recaptcha trying to detect if you are bot or not based on many things, like IP address, most probably every IP range which is used by data centers is for them treated as "suspicious"
2020-01-20 13:08:34 <MY-R> or ISPs known from having spam users
2020-01-20 13:08:50 <_ikke_> Apparently the DC ip range for the VPS I use is not classified (yet) as suspicious / non-domestic
2020-01-20 13:09:03 <MY-R> yet!
2020-01-20 13:10:46 <_ikke_> I like to keep it that way
2020-01-20 13:10:49 <_ikke_> it has been for years
2020-01-20 13:11:54 <MY-R> for example When I used HE free tunnel (ipv6) then on many pages with ipv6 I started to see recaptchas too
2020-01-20 13:12:55 <MY-R> and since it is free service/tunnel then obvious that many people abused it to the point that is automaticly blacklisted :P
2020-01-20 13:12:55 <_ikke_> I have been using that vps are vpn host for years, no extra recaptchas or anything
2020-01-20 13:13:05 <_ikke_> but I have a dedicated IP, so that's nice
2020-01-20 13:16:09 <MY-R> ye, I have my IP in home internet unchanged since years and dont have issues with recaptchas too but cant say same about any bigger telekom with dynamic IPs
2020-01-20 13:50:56 <markand> there are no manual page for apk at the moment?
2020-01-20 13:52:44 <_ikke_> There are some in the works, but I don't know the current status
2020-01-20 13:52:59 <_ikke_> https://gitlab.alpinelinux.org/alpine/apk-tools/merge_requests/9
2020-01-20 13:54:24 <markand> nice
2020-01-20 14:23:27 <Cogitri> Do I reckon correctly that there's no ASAN available on musl?
2020-01-20 14:33:22 <ddevault> Cogitri: correct
2020-01-20 14:40:36 <Cogitri> That's unfortunate
2020-01-20 15:42:54 <repo> hiya!
2020-01-20 15:43:12 <repo> i'm trying to capture audio from a pulse device in ffmpeg
2020-01-20 15:43:35 <repo> using -f pulse however raises the error: Unknown input format: 'pulse'
2020-01-20 15:43:57 <repo> i've installed the pulseaudio package
2020-01-20 15:44:10 <repo> any clues what i might be doing wrong?
2020-01-20 15:50:24 <Cogitri> Maybe our ffmpeg is compiled w/o support for that?
2020-01-20 15:50:31 <repo> probably
2020-01-20 15:50:37 <repo> trying -f alsa next
2020-01-20 20:24:03 <asimpson> I have an apk question. If I install a package like zfs or wireguard that is built against a specific kernel version what happens when an update comes out? Does apk use something like DKMS (if I'm understanding that right)?
2020-01-20 20:25:45 <Cogitri> No, we don't use APK
2020-01-20 20:25:54 <Cogitri> s/APK/DKMS/
2020-01-20 20:25:56 <Cogitri> (oops)
2020-01-20 20:26:19 <Cogitri> Apk will automatically upgrade the kernel module and the kernel at the same time
2020-01-20 20:26:36 <Cogitri> We package the already compiled module for you so you don't have to build it yourself
2020-01-20 20:27:54 <asimpson> Cogitri: so it _should_ "just work"? For example, a ZFS installation would upgrade automatically and on reboot be using the new kernel/new version of ZFS?
2020-01-20 20:28:37 <Cogitri> Yup
2020-01-20 20:28:42 <Cogitri> "just works" for me
2020-01-20 20:29:10 <Cogitri> (The only thing to keep in mind (that I frequently forget when setting up a machine) is to add the zfs feature to mkinitfs :)
2020-01-20 20:29:16 <asimpson> Nice, that's good to know! I appreciate the quick reply!
2020-01-20 20:29:27 <asimpson> Yep I forget that specific one as well
2020-01-20 20:40:42 <dalias> mps, do you know where to find good info on it? if the offending commit is known, can alpine include a patch to revert it until it's fixed upstream?
2020-01-20 20:41:03 <dalias> mps, it's stupid that this has been present for so many releases
2020-01-20 20:41:43 <dalias> i wonder if it's related to removal of enable_rc6 option
2020-01-20 20:41:50 <dalias> previously i had to turn that off but now it no longer exists
2020-01-20 20:41:54 <north1> @dalias intel hangup ?
2020-01-20 20:41:57 <dalias> yes
2020-01-20 20:41:59 <mps> dalias: ncopa added patch today for linux-lts 5.4.13
2020-01-20 20:42:06 <north1> linux-lts>=5.4.13-r1
2020-01-20 20:42:31 <north1> the -r1 is very important, if you have aports tree checked out you can look at main/linux-lts/0001-drm-i915-gt-Detect-if-we-miss-WaIdleLiteRestore.patch for the patch
2020-01-20 20:42:41 <dalias> ok that'sthe patch i saw on a few tracker threads
2020-01-20 20:42:45 <mps> patch is in main/linux-lts/0001-drm-i915-gt-Detect-if-we-miss-WaIdleLiteRestore.patch
2020-01-20 20:42:54 <dalias> so hopefully it actually works
2020-01-20 20:43:19 <dalias> is the package with the patch available yet?
2020-01-20 20:43:21 <mps> don't know, I don't have exact hardware where I can test it
2020-01-20 20:43:52 <mps> linux-lts 5.4.13 should be already on mirrors
2020-01-20 20:44:05 <dalias> north1, said the -r1 is needed
2020-01-20 20:44:08 <dalias> but it's there now :)
2020-01-20 20:44:17 <dalias> updating now
2020-01-20 20:44:21 <dalias> let's hope this works
2020-01-20 20:44:21 <mps> linux-lts 5.4.13-r1
2020-01-20 20:44:31 <dalias> it's really annoying
2020-01-20 20:44:38 <dalias> mouse keeps working but otherwise X is frozen
2020-01-20 20:44:41 <dalias> ssh in works fine
2020-01-20 20:44:56 <dalias> sometimes it unfreezes after a few minutes or after ssh in to pm-suspend or chvt
2020-01-20 20:45:07 <dalias> other times it requires reboot
2020-01-20 20:45:11 <mps> to me it freeze for some time but only with firefox
2020-01-20 20:45:23 <dalias> it only freezes when gpu is in use
2020-01-20 20:45:31 <dalias> for me it's with openscad
2020-01-20 20:45:36 <mps> yes, same here
2020-01-20 20:45:42 <dalias> (which i still should package....)
2020-01-20 20:45:47 <north1> i managed to sorta reproduce it by opening thousands of Alacritty windows by holding Meta+Enter on sway :D
2020-01-20 20:46:01 <dalias> north1, :-p
2020-01-20 20:46:21 <dalias> for me it always happens when interactive-rotating viewport in openscad with mouse
2020-01-20 20:46:27 <dalias> (rapid 3d rendering)
2020-01-20 20:46:32 <mps> no, not openscad, don't know what is this, but with firefox after some time I can fix it by login over ssh
2020-01-20 20:46:34 <north1> sometimes it freezes the gpu, sometimes sway straight up crashes
2020-01-20 20:46:52 <dalias> openscad is cad/modeling software
2020-01-20 20:47:14 <mps> sometimes it kill firefox with OOM
2020-01-20 20:47:54 <mps> hmm, maybe I have same issue as you all, but less badly
2020-01-20 20:48:39 <mps> I noticed that adding swap about 2GB size looks like solution
2020-01-20 20:48:49 <dalias> hm?
2020-01-20 20:48:56 <dalias> why is adding swap a solution?
2020-01-20 20:49:05 <north1> @mps on your aarch64 chromebook ?
2020-01-20 20:49:50 <mps> north1: no, here I'm speaking about x86_64
2020-01-20 20:50:13 <mps> on aarch64 I have only issues with emmc card
2020-01-20 20:50:51 <mps> dalias: have no idea, but when adding swap it doesn't locks display
2020-01-20 20:52:33 <mps> I don't know if ncopa tested 5.4.13-r1, maybe he can tell us more does the patch he added helps
2020-01-20 20:54:07 <dalias> i'll pin glxgears for a while and see if it freezes
2020-01-20 20:56:08 <mps> good luck :)
2020-01-20 21:06:39 <dalias> maybe i'm imagining things but it feels like 3d performance is vastly improved by this
2020-01-20 21:06:52 <dalias> any idea if that's plausible?
2020-01-20 21:09:36 <mps> glxgears => 120 frames in 5.0 seconds = 23.890 FPS
2020-01-20 21:10:00 <dalias> i meant interactive use in openscad
2020-01-20 21:11:08 <mps> maybe it depends on screen resolution
2020-01-20 21:58:52 <dalias> wooo
2020-01-20 21:58:56 <dalias> mps, pretty sur ethis patch fixed it
2020-01-20 21:59:06 <dalias> *and* the old reason i was using enable_rc6=0
2020-01-20 21:59:20 <dalias> quake used to freeze for like 4 seconds switching video mode
2020-01-20 21:59:28 <dalias> now it does it instantly
2020-01-20 22:06:18 <mps> dalias: didn't understood, patch fixed it or not
2020-01-20 22:47:25 <dalias> mps, fixed both problems (new one & old one)
2020-01-20 22:50:58 <mps> dalias: thanks for info, maybe we can close issue report.
2020-01-21 02:05:21 <pltrz[m]> can't seem to specify my monospace font in sway. it just shows characters as boxes
2020-01-21 02:08:19 <pltrz[m]> setting monospace or Terminus is not working
2020-01-21 02:41:08 <pltrz[m]> also, cannot seem to run qutebrowser
2020-01-21 03:10:38 <codebam> I just came to say that apk is really cool
2020-01-21 03:12:13 <codebam> I removed gnupg2 for gnupg1 by installing lxc-templates with force and it recursively removed gnupg2 and then when I removed lxc-containers it added everything back and fixed itself
2020-01-21 03:12:44 <dalias> :p
2020-01-21 03:13:58 <codebam> this was like 85 packages that it just automatically added back. was really cool to watch
2020-01-21 03:14:10 <codebam> mostly because I thought I broke everything
2020-01-21 03:24:35 <codebam> question, can I boot alpine to RAM and install to the drive I booted from?
2020-01-21 03:24:45 <codebam> if I put it on my SSD and just boot to it
2020-01-21 04:01:26 <pltrz[m]> yes, you can do that
2020-01-21 04:02:11 <pltrz[m]> just run 'setup-disk -m sys /some/dir'
2020-01-21 04:02:18 <codebam> oh okay cool
2020-01-21 04:02:24 <codebam> thanks!
2020-01-21 04:02:33 <pltrz[m]> and mount the block device to /some/dir
2020-01-21 04:02:34 <codebam> just backing up my SSD still
2020-01-21 04:02:46 <codebam> yep. I'll just follow the encryption guide
2020-01-21 04:02:57 <codebam> since I want full disk encryption anyways
2020-01-21 04:03:08 <codebam> and install manually that way
2020-01-21 04:03:13 <pltrz[m]> good stuff
2020-01-21 04:03:26 <codebam> :)
2020-01-21 04:04:31 <pltrz[m]> I am trying to figure out why (1) sway is not rendering the fonts for and (2) qutebrowser does not run
2020-01-21 04:06:25 <codebam> what fonts? did you install fonts?
2020-01-21 04:06:54 <codebam> sway is working for me. I installed noto, dejavu, and hack
2020-01-21 04:06:57 <codebam> for fonts
2020-01-21 04:07:02 <codebam> and they just worked
2020-01-21 04:07:03 <pltrz[m]> yes, I installed 'font-terminus',
2020-01-21 04:07:13 <pltrz[m]> let me try one of those
2020-01-21 04:07:22 <codebam> terminus is a console font, not usable with fontconfig
2020-01-21 04:07:31 <codebam> it's not a truetype or opentype font
2020-01-21 04:07:42 <pltrz[m]> oh, I see.
2020-01-21 04:07:43 <iggy> did you run qutebrowser from a terminal and check if there were any errors?
2020-01-21 04:08:36 <codebam> I like terminus in the console. I wish there was a 1.5x version for my 1440p screen. I just use the fonts at 1x and pretend I can read them
2020-01-21 04:08:45 <pltrz[m]> iggy: yes, and also if I run sway with xwayland enabled, then it seems to spawn a running process but does not show a window. i have to kill the process manually
2020-01-21 04:09:50 <pltrz[m]> is hack packaged?
2020-01-21 04:09:54 <codebam> yeah it is
2020-01-21 04:10:02 <codebam> ttf-hack
2020-01-21 04:10:22 <pltrz[m]> oh, missed it
2020-01-21 04:12:45 <pltrz[m]> ok, ttf-hack is working
2020-01-21 04:17:41 <codebam> I'm using sway right now with the old config I was using on fedora and essentially everything works except waybar segfaults when I add or remove a display (which is easy to fix by just reloading the sway config)
2020-01-21 04:19:05 <codebam> also if you use firefox under wayland you can pass MOZ_ENABLE_WAYLAND=1 as an environment variable to run it in wayland instead of in xwayland
2020-01-21 04:21:13 <pltrz[m]> I also wondered which package should be provided for sway. I realized that mesa-dri-intel was a dep of xf86-video-intel
2020-01-21 04:21:25 <pltrz[m]> and they both work for getting sway to run
2020-01-21 04:29:01 <codebam> you don't need xf86-video-intel for sway, only mesa-dri-intel
2020-01-21 04:29:09 <codebam> I think
2020-01-21 04:29:27 <codebam> but xf86 is for xorg
2020-01-21 05:27:58 <pltrz[m]> yeah, that's what I thought
2020-01-21 06:03:10 <codebam> can the standard image boot to ram or do I need the ram image for that?
2020-01-21 06:03:30 <codebam> I wanted to install to this drive and didn't realize there weren't boot options
2020-01-21 06:03:50 <codebam> I'll just put it on a flash drive...
2020-01-21 07:12:41 <^7heo> hi
2020-01-21 07:12:59 <^7heo> any reason hugo isn't built for i686?
2020-01-21 07:15:18 <north1> @^7heo tests fail on that arch, it is written in the APKBUILD
2020-01-21 07:21:26 <^7heo> north1: yeah sorry I am merely a sporadic of alpine now, I would not check the manifest as my first reflex.
2020-01-21 07:21:40 <^7heo> s/of/user &/
2020-01-21 07:22:03 <^7heo> also typing on a phone so typos galore v_v
2020-01-21 07:22:55 <^7heo> north1: but thanks for your answer.
2020-01-21 09:37:56 <mps> okias[m]: are you around?
2020-01-21 09:43:14 <okias[m]> mps yup, hi :)
2020-01-21 09:45:08 <mps> hi!
2020-01-21 09:45:40 <mps> tried firefox 72.0.2 on armv7, still not fixed upstream
2020-01-21 09:46:23 <mps> did you got any response from upstream on https://bugzilla.mozilla.org/show_bug.cgi?id=1610274
2020-01-21 10:01:32 <misterunknown> Hi guys. I have a problem with cgroups: I want to add a systemd cgroup by default (to be able to work with docker), but I don't find an appropriate way to do it.
2020-01-21 10:01:44 <misterunknown> At the moment I do the following, after starting the system:
2020-01-21 10:01:45 <misterunknown> mkdir /sys/fs/cgroup/systemd
2020-01-21 10:01:52 <misterunknown> mount -t cgroup -o none,name=systemd cgroup /sys/fs/cgroup/systemd/
2020-01-21 10:02:07 <misterunknown> I didn't find any configuration for /etc/init.d/cgroups
2020-01-21 10:05:04 <mps> misterunknown: /etc/init.d/cgroups start
2020-01-21 10:05:50 <mps> or to activate it on boot: rc-update add cgroups
2020-01-21 10:06:44 <misterunknown> mps: Yes, this will mount all cgroups, except systemd... Thats my problem.
2020-01-21 10:07:16 <mps> ah, sorry, missed your need
2020-01-21 10:07:59 <mps> you want to run nested dockers?
2020-01-21 10:08:40 <misterunknown> mps: Yes. But afaik this would also apply if I just want to run docker containers in the hostsystem.
2020-01-21 10:09:07 <misterunknown> Because my host system also lacks a systemd cgroup mount.
2020-01-21 10:12:00 <mps> someone asked similar question few days ago here or on #alpine-devel (can't remember exactly)
2020-01-21 10:12:53 <mps> iirc, solution is found. you can try search channel log
2020-01-21 10:22:04 <misterunknown> mps: Well, the only related messages are the ones I asked a few days ago^^ I indeed found this manual solution.
2020-01-21 10:22:27 <misterunknown> I'd like to mount the systemd cgroup by default. But I'm not sure which is the best solution for that.
2020-01-21 10:24:50 <mps> aha, I thought you were one who asked but wasn't sure :)
2020-01-21 10:31:10 <Cogitri> Putting it in fstab?
2020-01-21 10:39:39 <misterunknown> Cogitri: I tried that, but it does not work, probably because the mountpoint does not exist. Also, I didn't find a way to automatically create the mountpoint _before_ fstab gets evaluated.
2020-01-21 10:46:38 <attos1> How can I disable serial device /dev/ttyS0 ? "setserial /dev/ttyS0 uart none" doesn't work: "setserial: illegal UART type: none". I'm running Alpine "5.4.12-1-lts, x86".
2020-01-21 10:47:25 <AinNero> per default, the serial device is just sitting there doing nothing
2020-01-21 10:47:36 <AinNero> do you want to prevent a getty from running on it?
2020-01-21 10:49:36 <attos1> I try to load the module serial_ir: "modprobe serial_ir" and there is an error in dmesg: "port 03f8 already in use", "use 'setserial /dev/ttySX uart none'"
2020-01-21 10:51:47 <attos1> The module serial_ir is used in the LIRC project to load the driver for homebrew serial ir interface.
2020-01-21 10:52:52 <AinNero> this looks like serial_ir is attemptin to bypass the kernel SERIAL_8250 driver
2020-01-21 10:53:17 <AinNero> which seems unreasonable to me since you can interface all RS232 pins via ioctl, which would be more postable
2020-01-21 10:53:35 <AinNero> *portable
2020-01-21 10:54:28 <attos1> But serial_ir is a module shipped by the original kernel. It's not a module provided by LIRC.
2020-01-21 10:54:56 <AinNero> yeah, i just checked the source, it talks to the io space directly
2020-01-21 10:55:19 <AinNero> the 8250 driver is hard-compiled into the linux kernel
2020-01-21 10:55:22 <attos1> Can I just unload the  SERIAL_8250 driver
2020-01-21 10:55:25 <AinNero> no
2020-01-21 10:55:31 <AinNero> its compiled in
2020-01-21 10:55:37 <AinNero> but it might be possible to unbind it
2020-01-21 10:55:41 <attos1> Puhh, that bad.
2020-01-21 10:55:59 <attos1> How ?
2020-01-21 10:58:19 <AinNero> related: https://lwn.net/Articles/143397/
2020-01-21 10:59:26 <attos1> AinNero: Thanks. I'll read and try.
2020-01-21 11:01:50 <AinNero> attos1: i think its `/sys/class/tty/ttyS0/device/driver# echo serial8250 > unbind`
2020-01-21 11:06:56 <AinNero> attos1: also, the setserial command didnt work because of busybox
2020-01-21 11:07:12 <AinNero> you could install util-linux and try to use that setserial
2020-01-21 11:07:21 <AinNero> tbh first time i hear of setserial
2020-01-21 11:08:59 <mps> AinNero: can't believe that you don't know for setserial
2020-01-21 11:09:13 <attos1> Using "echo serial8250 > /sys/class/tty/ttyS0/device/driver/unbind" gives "ash: write error: No such device"
2020-01-21 11:11:10 <attos1> If I type "setserial --help" I get a list of uart types.
2020-01-21 11:11:34 <attos1> How can I test which one is the right for my hardware ?
2020-01-21 11:12:07 <AinNero> you want "none", which is an option the busybox option doesn't have
2020-01-21 11:12:51 <AinNero> *the busybox setserial
2020-01-21 11:16:06 <attos1> I've installed util-linux, but there is no setserial in the package.
2020-01-21 11:17:40 <AinNero> i was wrong with the fact that its in util-linux
2020-01-21 11:17:59 <AinNero> seems to be an own project..
2020-01-21 11:18:28 <AinNero> ... which is not package for alpine linux
2020-01-21 11:32:37 <AinNero> attos1: would you post your /proc/ioports ?
2020-01-21 11:37:45 <attos1> https://paste.debian.net/1126961/
2020-01-21 11:38:39 <AinNero> still shows '03f8-03ff : serial'
2020-01-21 11:40:17 <attos1> Yes of course it does. I've not found any solution for the problem.
2020-01-21 11:42:06 <AinNero> i got it removed via unbind
2020-01-21 11:44:41 <attos1> But your suggestion to use "echo serial8250 > unbind" gives an error, as I posted earlier ("ash: write error: No such device").
2020-01-21 11:45:40 <admicos> hey all! is there a way to build a single subpackage from a modified apkbuild?
2020-01-21 11:46:03 <AinNero> attos1: this is what worked with my hardware, it might be different on your hardware
2020-01-21 11:47:46 <north1> admicos: no
2020-01-21 11:47:58 <admicos> :(
2020-01-21 11:48:07 <admicos> i guess i'll have to rebuild the entirety of nginx just for a single module
2020-01-21 11:50:42 <attos1> AinNero: Heureka ! It works :) Many thanks to you. This command works: "echo -n '00:04' > /sys/bus/pnp/drivers/seria/unbind". I just had to use the correct PCI-ID.
2020-01-21 11:53:02 <attos1> This PCI-ID was shown in "/sys/bus/pnp/drivers/serial" directory.
2020-01-21 11:54:12 <AinNero> is it a RS232 PCI card?
2020-01-21 11:54:35 <attos1> No. It's on board.
2020-01-21 12:00:48 <attos1> Now I have only one last question: How can I drop this "echo -n '00:04' > /sys/bus/pnp/drivers/seria/unbind" command automatically while system start up but before my serial_ir module will be loaded from /etc/modules ?
2020-01-21 12:06:16 <attos1> O.K. I think I can make a startup script in /etc/init.d doing this.
2020-01-21 12:07:19 <attos1> echo ....; modprobe serial_ir
2020-01-21 12:21:04 <mps> attos1: DIY IR serial driver? how it is interfaced with main board? usb, spi or something else
2020-01-21 12:33:47 <attos1> mps: via rs232 port
2020-01-21 12:34:56 <mps> attos1: why then standard serial port can't be used
2020-01-21 12:35:33 <mps> but to not bother you much, do you have link for this
2020-01-21 12:36:45 <attos1> Because when I "modprobe serial_ir" there is an error in dmesg :"port 03f8 already in use, use 'setserial /dev/ttySX uart none'"
2020-01-21 12:37:19 <attos1> And this setserial command doesn't work in alpine linux.
2020-01-21 12:37:58 <mps> yes, noticed that, I will try to package it for alpine, maybe this night
2020-01-21 12:38:25 <attos1> mps: That would be great.
2020-01-21 12:42:03 <attos1> mps: https://wiki.archlinux.org/index.php/LIRC?#Serial_port
2020-01-21 12:44:10 <mps> if I don't have to go to meet friends in a few minutes I would make it in a hour
2020-01-21 12:44:57 <mps> actually, I didn't noticed it is not packaged because busybox setserial worked in my use cases
2020-01-21 12:46:49 <mps> I sometimes play with rs485 devices, when have free time or bored with 'normal' work
2020-01-21 12:56:31 <markand> alpine <3
2020-01-21 12:57:13 <mps> markand: ?
2020-01-21 12:58:03 <markand> mps, sorry, just spreading how much I appreciate the whole project
2020-01-21 12:58:04 <markand> :-)
2020-01-21 12:58:51 <mps> ah, ok
2020-01-21 13:01:17 <markand> is there a way to check which packages were manually installed? (not as dependency?) or the equivalent to autoremove (apt) /autoerase (dnf)
2020-01-21 13:01:38 <markand> I don't find my answer in the apk info/del commands
2020-01-21 13:01:58 <markand> ah apk del has -r, but the package is already deleted
2020-01-21 13:07:20 <clandmeter> markand: yes
2020-01-21 13:07:30 <north1> See /etc/apk/world
2020-01-21 13:07:30 <clandmeter> check /etc/apk/world
2020-01-21 13:07:39 <north1> :D
2020-01-21 13:08:25 <markand> :) thanks
2020-01-21 19:19:23 <blueeyes> I did `ln -s $(which apk-autoupdate) /etc/periodic/hourly/apk-autoupdate`. is that the correct way to run apk-autoupdate once every hour?
2020-01-21 19:20:03 <ddevault> once an _hour_?
2020-01-21 19:20:10 <ddevault> that's asking for trouble
2020-01-21 19:20:32 <blueeyes> oh, I thought it was fine if I'm not on edge. I can do weekly or something instead
2020-01-21 19:20:33 <ddevault> I reckon what I'd do is 0 0 * * * chronic apk upgrade -U in root's crontab, with a mailto set up so that it can forward errors to an email address for further study
2020-01-21 19:20:37 <ddevault> no more often than daily, though
2020-01-21 19:20:43 <ddevault> probably weekly'd be better
2020-01-21 19:20:58 <blueeyes> okay, cool. thanks ddevault
2020-01-21 20:19:02 <mps> attos1: there will be setserial on mirrors in testing repo, in about 15 minutes
2020-01-21 20:24:36 <blueeyes> how would I go about creating an alpine linux lxc container? I don't understand what's going on with there being no templates
2020-01-21 20:25:44 <_ikke_> blueeyes: https://pkgs.alpinelinux.org/package/edge/main/x86_64/lxc-templates-legacy
2020-01-21 20:26:43 <blueeyes> _ikke_: am I supposed to use that though? it's legacy
2020-01-21 20:27:27 <_ikke_> https://brauner.github.io/2018/02/27/lxc-removes-legacy-template-build-system.html
2020-01-21 20:27:37 <_ikke_> if you want to build using templates, you need to use the legacy tempaltes
2020-01-21 20:27:39 <_ikke_> templates
2020-01-21 20:28:36 <blueeyes> oh okay, thank you
2020-01-21 20:46:39 <attos1> mps: I will try it as soon as I can get it :) Many thanks.
2020-01-21 20:47:50 <mps> np
2020-01-21 20:48:40 <mps> please report back any issue you have so we can polish it to be better
2020-01-21 20:53:18 <attos1> Of course, I will do.
2020-01-21 21:48:29 <PICCORO> seems there's no 3.11.3 repository right?
2020-01-21 21:50:04 <attos1> mps: setserial works fine. After "setserial /dev/ttyS0 uart none" I was able to load the serial_ri module without errors :) dmesg shows that the serial_ri module has detected my homebrew LIRC device.
2020-01-21 21:52:40 <mps> attos1: good, and you mean serial_ir and not serial_ri?
2020-01-21 21:53:46 <codebam> I installed `samurai` which satisfies the dependency for `ninja` and purged the ninja package. how do I undo?
2020-01-21 21:53:54 <codebam> i.e. how do I get the ninja package back
2020-01-21 21:54:09 <codebam> I can't `del samurai` because there are dependencies on `ninja`
2020-01-21 21:54:33 <mps> PICCORO: what you mean by 3.11.3 repo?
2020-01-21 21:57:47 <codebam> I guess I'm just asking. how do I fix my installed packages based on world
2020-01-21 21:58:10 <codebam> because right now samurai is installed to provide ninja but it's not in world, and ninja is in world
2020-01-21 21:58:23 <codebam> but I can't swap back
2020-01-21 22:00:38 <attos1> mps: yes, serial_ir
2020-01-21 22:02:25 <mps> attos1: ofc :)
2020-01-21 22:04:05 <mps> I think we should add /etc/init.d/setserial script and /etc/serial.conf but wanted first to see if it works correctly. Hope I can ask you to test when I prepare these addition (if no one do that in meantime)
2020-01-21 23:58:48 <alxsdf> It looks like the octave package is somehow broken
2020-01-21 23:59:18 <alxsdf> I run octave --gui and it tells me "GUI features missing or disabled in this build"
2020-01-21 23:59:39 <alxsdf> but the aports APKBUILD file has --with-qt
2020-01-22 00:00:43 <alxsdf> oh, there's no listed maintainer? shouldn't it be in unmaintained/
2020-01-22 00:25:09 <MY-R> alxsdf, correct: Build Octave Qt GUI:                  no (missing:Qt5Help)
2020-01-22 12:46:25 <bl4ckb0ne> is there a way for `abuild rootbld` to have internet access?
2020-01-22 12:46:57 <bl4ckb0ne> or alternatively, for abuild to not use the system packages
2020-01-22 12:47:11 <_ikke_> bl4ckb0ne: add "net" to options
2020-01-22 12:48:12 <bl4ckb0ne> `abuild rootbld net` ?
2020-01-22 12:48:35 <_ikke_> no
2020-01-22 12:48:37 <_ikke_> in the APKBUILD
2020-01-22 12:51:15 <bl4ckb0ne> nice it works
2020-01-22 12:51:18 <bl4ckb0ne> thanks
2020-01-22 13:12:07 <beginnner> hello, i am beginner user
2020-01-22 13:12:32 <beginnner> i try to make new package, but i got an error
2020-01-22 13:12:36 <beginnner> https://gitlab.com/red4jg/alpine/blob/master/error_log
2020-01-22 13:12:42 <Cogitri> Hello there
2020-01-22 13:12:51 <beginnner> hello
2020-01-22 13:39:13 <mps> I need protocols headers for some pkgs, which are missing in musl. one option is to use and package glibc files other is bsd files (https://github.com/freebsd/freebsd/tree/master/include/protocols)
2020-01-22 13:39:44 <mps> which one would be better for alpine, anyone have experience or advice?
2020-01-22 13:40:25 <beginnner> hello, i am beginner. i am creating new package, but i got an issue.
2020-01-22 13:40:27 <beginnner> https://gitlab.com/red4jg/alpine/blob/master/error_log
2020-01-22 13:40:30 <mps> I tend to think that glibc files are more compatible with linux user apps but not sure
2020-01-22 13:53:08 <claudiub> Hello! I've been hitting a very strange issue with the Alpine docker images, issue that I haven't been able to reproduce anywhere else. Basically, it seems that Alpine bash doesn't always execute the commands in a pipe in the same order every time. For example, running this scipt https://paste.ubuntu.com/p/q4c2J3GM9j/ won't yield 200 successes, it will contain 1-5 failures.
2020-01-22 13:53:50 <claudiub> it's even worse when running with the -x option, the failure count varies between 4 and 25 per run.
2020-01-22 13:55:12 <claudiub> but looking at the logs after running with the -x option, it seems that grep is being executed before sed. I've tried this with both alpine:3.10 and alpine:3:11
2020-01-22 13:55:26 <claudiub> any ideas / thoughts?
2020-01-22 14:38:54 <claudiub> Also, in the script, there's a commented alternative that has 100% success rate.
2020-01-22 15:02:02 <crabfuel> Hi. I'm having trouble with my alpine machine becoming unresponsive. I have syslog running and I see a cron running every 15 minutes until 3:45am and then nothing in the log until I restarted it because it was unresponsive. I'm hoping you can point me to a better logging solution or perhaps have an idea of what other logs to check.
2020-01-22 15:02:59 <KatolaZ> claudiub: you might have a problem with your condition...
2020-01-22 15:03:23 <PICCORO> crabfuel: are you runnin 3.11.3 ? i have some problems with that version.. 3.10 give me no problems
2020-01-22 15:03:34 <KatolaZ> you should test '`if [ $? -eq 0 ]; then ` instead
2020-01-22 15:03:54 <PICCORO> sorry _ikke_ i think that those mirros are down.. or repos are not working.. that's why i posted in infra channel, can someone check it?.. i have two days teting and there's no doub!
2020-01-22 15:05:46 <repo> hi
2020-01-22 15:07:20 <repo> i want to include the curl binary into a scratch container alongside a statically linked binary. I tried installing curl-static but that didn't provide a binary. Can you tell me what package(s) i need and what files i need to copy into the scratch container?
2020-01-22 15:07:28 <repo> *scratch image
2020-01-22 15:08:09 <crabfuel> 3.10.3
2020-01-22 15:08:10 <repo> mh maybe just curl and curl-static and then copy both the bin and /usr/lib/libcurl.a
2020-01-22 15:08:55 <crabfuel> sorry forgot to tag you, PICCORO. I'm using 3.10.3.
2020-01-22 15:10:34 <crabfuel> PICCORO: I only recently started to have this issue and I'm not sure what has changed. I uninstalled all of the packages that I recall since this started happening. I have several alpine lxc containers (on this alpine host) that do have changes so I suspect it may be one of them but I don't see anything in their logs either.
2020-01-22 15:20:33 <ncopa> beginnner: there are no error message. not sure what goes wrong
2020-01-22 15:22:55 <north1> I'd just nuke procmail right away
2020-01-22 15:23:11 <north1> People are exponentially quicker to report something if it breaks
2020-01-22 15:24:56 <ncopa> claudiub: smells like a bug. can you please report it with the test case on https://gitlab.alpinelinux.org/alpine/aports/issues
2020-01-22 15:26:06 <claudiub> KatolaZ: you mean something like: LANG=C sed --help 2>&1 | grep -q "GNU\|BusyBox"; if [ $? -eq 0 ]; then ? Because of the set -o errexit, whenever the issue occurs, the script directly exists, it won't even get to the end.
2020-01-22 15:26:07 <ncopa> crabfuel: what kernel version?
2020-01-22 15:26:40 <claudiub> ncopa: thanks, will do, will link to the opened issue asap.
2020-01-22 15:27:33 <PICCORO> ok crabfuel are not same issue like me.. i have problems to get any repos for 3.11.3 the 3.10.3 works well for me..
2020-01-22 15:28:03 <_ikke_> Do you have curl installed?
2020-01-22 15:28:23 <KatolaZ> claudiub: the script exits if the pipeline returns an error...
2020-01-22 15:31:27 <crabfuel> ncopa: 4.19.80-0-vanilla
2020-01-22 15:31:48 <KatolaZ> claudiub: I mean, if you have set -o errexit, the script exits when any command returns a non-zero value
2020-01-22 15:34:41 <ncopa> crabfuel: could be kernel issue
2020-01-22 15:34:52 <claudiub> KatolaZ: it doesn't in that if statement. :)
2020-01-22 15:35:57 <crabfuel> ncopa: should I try 3.11?
2020-01-22 15:36:03 <crabfuel> or edge?
2020-01-22 15:36:22 <crabfuel> at this point there is no stability so edge can't be any worse, haha.
2020-01-22 15:38:43 <ncopa> i'd try 3.11 yes
2020-01-22 15:39:08 <crabfuel> ncopa: if it is the kernel it's not a panic because I'd see that in the log, correct?
2020-01-22 15:39:31 <ncopa> no, i dont think you'll see a panic in the log
2020-01-22 15:45:53 <crabfuel> ncopa: ok. Thanks. I'm going to reboot. I'll be right back.
2020-01-22 16:08:24 <beginner> @ncopa thank you. but there is package failed error
2020-01-22 16:10:15 <mps> beginner: could you create merge request on gitlab.a.o so we can see complete situation
2020-01-22 16:10:46 <crabfuel> ncopa: I'm on 3.11.3 now, but the kernel is still 4.. What have I done wrong?
2020-01-22 16:11:06 <mps> crabfuel: apk add linux-lts
2020-01-22 16:11:29 <crabfuel> mps: will do. Thank you.
2020-01-22 16:12:02 <mps> and reboot and select lts from boot menu, ofc
2020-01-22 16:15:46 <beginner> @mps thank you. i will try it.
2020-01-22 16:16:13 <mps> nice
2020-01-22 16:17:02 <CrabFuel> That worked. Thanks. We will see if stability is better.
2020-01-22 16:23:29 <beginner> @algitbot thank you.
2020-01-22 16:29:16 <claudiub> ncopa: created bug report here: https://gitlab.alpinelinux.org/alpine/aports/issues/11152
2020-01-22 16:56:15 <KatolaZ> claudiub: it seems like the issue might be in bash's -o pipefail
2020-01-22 16:56:25 <KatolaZ> it works as expected in you don't set it
2020-01-22 17:06:42 <KatolaZ> claudiub: the problem is probably in the different return valus of gnu sed vs busybox sed on failure
2020-01-22 17:06:58 <KatolaZ> if you call busybox sed with "--help" it returns a 1
2020-01-22 17:07:11 <KatolaZ> while if you call gnu sed with --help, it returns a 0
2020-01-22 17:07:33 <KatolaZ> now, depending on scheduling, in the first case the pipe might or might not fail before the redirection takes place
2020-01-22 17:07:52 <KatolaZ> while in the second case you always get the pipe executed
2020-01-22 17:07:59 <KatolaZ> (i.e., if you have gnu sed)
2020-01-22 17:08:08 <KatolaZ> it's not a bug
2020-01-22 17:08:13 <KatolaZ> it's a matter of scheduling
2020-01-22 17:08:46 <KatolaZ> and of whether the busybox sed process dies before or after some of its output gets piped to grep
2020-01-22 17:08:50 <KatolaZ> my2cents
2020-01-22 17:09:12 <_ikke_> KatolaZ: Seems about right to my understanding as well
2020-01-22 17:10:03 <KatolaZ> I have tried with 4 different shells, before checking the return value of the two sed implementations
2020-01-22 17:10:25 <KatolaZ> (the behavious is obviously the same in all the shells, if you set -o pipefail)
2020-01-22 17:14:40 <KatolaZ> _ikke_: does one need to be logged in in gitlab to reply to an issue?
2020-01-22 17:15:01 <_ikke_> yes
2020-01-22 17:15:15 <_ikke_> we do support github and gitlab.com sso
2020-01-22 17:16:09 <KatolaZ> just wanted to report the discussion above
2020-01-22 17:16:16 <_ikke_> right
2020-01-22 17:16:17 <KatolaZ> in #11152
2020-01-22 17:20:53 <KatolaZ> _ikke_: pasted the note above in 11152, HTH
2020-01-22 17:31:11 <PICCORO> hey busybox fdisk at install cannot create GPT partitions? right?
2020-01-22 19:03:17 <PICCORO> umm tested several times.. no w with virtual box and got error when try to install ..
2020-01-22 19:55:57 <neneko> When I try to boot 3.11 on rpi zero, it gets stuck on the color test screen. Is there anything I need to do other than extracting the armhf tar to the root of the fat partition?
2020-01-22 19:56:27 <neneko> (Other distros, like raspbian, boots just fine)
2020-01-22 20:00:49 <neneko> Is the armhf kernel compatible with zero (non-wireless mode) ?
2020-01-22 20:01:04 <neneko> s/mode/model
2020-01-22 20:04:02 <mps> neneko: didn't tried with real display, but on serial console it works fine
2020-01-22 20:04:44 <neneko> Ah. I want hdmi output. Is that something I need to enable?
2020-01-22 20:05:09 <mps> it should work by default
2020-01-22 20:05:42 <mps> I added usercfg.txt after unpacked tarball to sd card
2020-01-22 20:06:18 <neneko> Not for me. It works if I swap the kernel for the raspbian one, but then it doesn't load the initramfs.
2020-01-22 20:06:20 <mps> actually, I have shell script to do all these 'things'
2020-01-22 20:07:06 <mps> I can paste script if you want to look and adapt to your case
2020-01-22 21:39:08 <mps> is it possible to set up hot plugged net (eth) interfaces on alpine
2020-01-22 21:40:03 <mps> i.e. when plug usb ethernet adapter to automatically run ifup
2020-01-22 22:01:52 <MY-R> mps, try with 'dhcpcd' and start dhcpcd service, I remember it was detecting if I plugged in or out my ethernet cable so new interface probbaly will detect too
2020-01-22 22:07:36 <mps> MY-R: I want static ip
2020-01-22 22:08:01 <MY-R> hmmm
2020-01-22 22:08:33 <mps> but, if noting else helps maybe I can use dhcp and fix IP by MAC on dhcp server
2020-01-22 22:09:09 <mps> thanks for last resort idea
2020-01-22 22:10:33 <MY-R> dont be so mad at dhcp! :P
2020-01-22 22:12:36 <neneko> So I got the pi to boot, but now I can't umount the sd card so I can install to it. Says "Resource busy", but lsof doesn't show any open files in /media/mmcblk0p1
2020-01-22 22:14:17 <MY-R> neneko, I hate that message :D
2020-01-22 22:14:39 <MY-R> neneko, what else did you mount? /dev/ /sys/ /proc/ stuff?
2020-01-22 22:17:05 <mps> I think something must be unmounted first
2020-01-22 22:17:26 <neneko> Yeah just the usual stuff. And /proc/mounts says / is tmpfs
2020-01-22 22:17:28 <mps> maybe /media/mmcbll0p1
2020-01-22 22:17:54 <neneko> Yeah that's what I'm trying to do.
2020-01-22 22:18:04 <neneko> But "resource busy"
2020-01-22 22:23:08 <neneko> I guess I'll have to try this alb thing instead of doing a normal system install.
2020-01-22 22:23:13 <mps> I never tried that but remember someone told that also something else should be unmounted
2020-01-22 22:23:19 <mps> modloop maybe
2020-01-22 22:23:43 <mps> see what is mounted and guess what can be unmounted
2020-01-22 22:25:49 <neneko> modloop was the culprit. Thanks!
2020-01-22 22:27:05 <mps> hah, have to note this when someone asks next time :)
2020-01-22 22:27:12 <neneko> Actually, unmounting that broke the install script. It can't find the keymaps.
2020-01-22 22:27:13 <mps> thanks for feedback
2020-01-22 22:28:21 <mps> hmm, time to fire up again my rpi zero from scratch and test
2020-01-22 22:28:34 <MY-R> areny keymaps installed by network?
2020-01-22 22:28:48 <MY-R> s/areny/arent
2020-01-22 22:29:21 <mps> I think it is set before network
2020-01-22 22:29:52 <MY-R> I remember always got that message on alpine without configured network first
2020-01-22 22:41:33 <neneko> I think setup-alpine installs the apks from the sd card when you run it maybe? /usr/share/bkeymaps doesn't exist unless I run the script with the sd card mounted.
2020-01-22 22:42:42 <neneko> And there's a bunch of apks inside /media/mmcblk0p1/apks
2020-01-22 22:49:19 <MY-R> neneko, yep true!
2020-01-22 22:49:48 <MY-R> that is why when I was doing chroot install and done setup-alpine after reboot then it couldnt find keymaps :)
2020-01-22 23:01:32 <neneko> I tried mounting the sd again, installing xkb-bkeymap, tzdata, and e2fsprogs, and then unmounting and running setup-alpine again. That fixed the errors, but now it says I'm missing syslinux, which isn't included in the repo in /media/mmcblk0p1/apks.
2020-01-22 23:02:33 <mps> you can ignore this
2020-01-22 23:03:02 <mps> syslinux is not used on ARMs
2020-01-22 23:03:10 <lastebil> I've installed the Xen dom0 alpine setup, and all is working fine - but it seems that the qemu-img package is not available via apk, despite it showing in the web package repo.
2020-01-22 23:04:44 <lastebil> is there something perhaps blocking it, or is there a difference in the indexes on different mirrors, or - heck I have no idea. I'd just like to be able to use the qemu-img (more functionality) instead of qemu-img-xen - and mount qcow2 images in an emergency via qemu-nbd.
2020-01-22 23:06:05 <lastebil> (this is more a "ok what did I do strangely with my apk setup" than a xen, or qemu, or virtualization question.  I think my apk setup is a bit 'off' perhaps (though I have done nothing to it outside of the xen dom0 install (from the install image.)
2020-01-22 23:06:10 <lastebil> )
2020-01-22 23:08:11 <neneko> lastebil:  qemu-img is in the community repo, which is not enabled by default.
2020-01-22 23:08:46 <neneko> You need to add it in /etc/apk/repositories
2020-01-22 23:09:29 <lastebil> aha!  (yes, It's been a while since I was able to use alpine regularly (:  )
2020-01-22 23:10:29 <lastebil> and there it is, installed. Thanks!
2020-01-22 23:10:54 <neneko> Glad I could help :)
2020-01-22 23:24:40 <codebam> hey, getting this error with dbus. is our dbus implementation different than other distros?
2020-01-22 23:25:10 <codebam> https://termbin.com/a85r
2020-01-22 23:25:16 <codebam> it works if dbus isn't running
2020-01-22 23:25:48 <codebam> is there a way I can hide dbus from the program? maybe by setting an environment variable?
2020-01-22 23:26:15 <armin> codebam: that can happen on other distributions too, see this issue on github: https://github.com/the-cavalry/light-locker/issues/36
2020-01-22 23:26:27 <codebam> armin saw that issue. but what would I do to fix it?
2020-01-22 23:26:34 <codebam> I can't use dbus-launch, tried that already
2020-01-22 23:26:38 <armin> codebam: good question, i wish i could answer that.
2020-01-22 23:26:59 <codebam> I would just disable dbus, but mako (notification daemon) needs it
2020-01-22 23:28:03 <codebam> this is so stupid. https://getkiss.org/blog/20190828a
2020-01-22 23:28:21 <codebam> they removed the ability to run firefox without dbus. but firefox runs perfectly fine without dbus (if it's not running)
2020-01-22 23:29:08 <armin> firefox definitely shouldn't segfault on that, though
2020-01-22 23:29:23 <codebam> this is the flatpak for firefox dev edition
2020-01-22 23:29:25 <armin> yeah i currently run a freebsd laptop entirely without dbus without any issues
2020-01-22 23:29:31 <armin> saw that :)
2020-01-22 23:29:38 <codebam> oh okay cool
2020-01-22 23:30:14 <armin> unfortunately most developers still rely on dbus being "just there".
2020-01-22 23:30:18 <codebam> it looks like it called abort() no? so it didn't really segfault
2020-01-22 23:30:37 <codebam> it was just mozilla being mean :(
2020-01-22 23:30:54 <armin> i don't know, i think it segfaulted after redirecting the call to abort()
2020-01-22 23:30:58 <armin> for WHATEVER reason
2020-01-22 23:31:03 <codebam> oh okay
2020-01-22 23:31:31 <codebam> I just think it must be dbus related because it runs fine when dbus isn't running
2020-01-22 23:31:37 <armin> firefox definitely should never segfault, that's a bug anyways, question is if you could convince it to behave.
2020-01-22 23:31:59 <armin> it is dbus related, yes.
2020-01-22 23:32:12 <armin> i mean hey, it's dbus, of course that has to suck.
2020-01-22 23:32:24 <codebam> if I could hide dbus from it so it would start that would be enough for me. for now I guess I'll just kill dbus whenever I want to use this flatpak. all I use it for is widevine anyways
2020-01-22 23:33:06 <armin> it's poettering software, i don't think my views on his stuff are mentionable in public chats.
2020-01-22 23:34:17 <armin> well not quite, but since he wanted to do kdbus...
2020-01-22 23:34:53 <codebam> armin who are you referring to? the author of dbus?
2020-01-22 23:34:54 <Cogitri> I doubt killing dbus is a solution to that, fixing Firefox is
2020-01-22 23:35:03 <codebam> haha
2020-01-22 23:35:25 <Cogitri> codebam: He's trying to blame Poettering for some reason I guess :)
2020-01-22 23:35:36 <Cogitri> Let's not go down that path
2020-01-22 23:35:46 <armin> codebam: lennart poettering is´n't frm what i know, but considering that he wanted dbus in the kernel, i think he's safe to blame here, too.
2020-01-22 23:35:47 <codebam> oh okay. yeah agreed
2020-01-22 23:35:56 <armin> also yeah let's not go down that path.
2020-01-22 23:35:58 <codebam> ohh lennart haha
2020-01-22 23:36:15 <codebam> where are all the systemd haters at?
2020-01-22 23:36:24 <Cogitri> Hopefully not here.
2020-01-22 23:36:28 <codebam> :)
2020-01-22 23:37:08 <Cogitri> Like or dislike what you want but I don't understand how some people can hate on something like software so much
2020-01-22 23:37:23 <armin> exactly. just use something different. it's that simple.
2020-01-22 23:38:08 <mps> Cogitri: human nature
2020-01-22 23:39:03 <Cogitri> I guess so, yes :c
2020-01-22 23:39:54 <armin> codebam: you said you're not willing to go down that road.
2020-01-22 23:40:04 <armin> ah no.
2020-01-22 23:40:07 <armin> Cogitri:
2020-01-22 23:40:12 <armin> nick completion fail.
2020-01-22 23:41:00 <armin> anyways, there are perfectly fine reasons to avoid systemd and alpine is one, so it's likely that people annoyed by systemd might end up here earlier or later, too.
2020-01-22 23:41:16 <armin> so better get used to them.
2020-01-22 23:41:18 <codebam> armin what road?
2020-01-22 23:41:29 <codebam> blaming poettering?
2020-01-22 23:41:33 <armin> codebam: that systemd bashing road we're accidentally on.
2020-01-22 23:41:40 <armin> basically, yes.
2020-01-22 23:41:46 <codebam> oh. I mean if you don't like software don't use it
2020-01-22 23:41:50 <codebam> no one is forcing you to use systemd
2020-01-22 23:41:54 <armin> exactly.
2020-01-22 23:42:27 <armin> talking about software i don't use currently: i have 0 alpine systems at the moment i have to admit.
2020-01-22 23:42:40 <armin> i use it inside docker containers all the time though.
2020-01-22 23:43:10 <Cogitri> I use Alpine on all my stuff
2020-01-22 23:43:21 <armin> freebsd is what i grab first.
2020-01-22 23:43:35 <Cogitri> But I must admit the init part (!) of systemd really is a lot nicer than OpenRC
2020-01-22 23:44:14 <armin> dunno, it's still slower than runit, and maintaining it is a mess because man systemd<TAB> gives me more than 160 hits.
2020-01-22 23:44:26 <armin> i see no reason to use it anyways.
2020-01-22 23:44:40 <Cogitri> It certainly is more reliable than runit due to having service deps etc. though
2020-01-22 23:44:53 <armin> also the project seems to significantly suffer from omnipotent hyperactivity.
2020-01-22 23:45:37 <mps> as early adopter of systemd I can tell that after some time when it started to do everything I switched from it
2020-01-22 23:45:54 <Cogitri> And yes, there are perfectly fine reasons for avoiding systemd, but all the pointless hate against it really annoys me and is one of the things that drove me away from Void, there were *so* many people who only used Void so they didn't have to use systemd and could complain about it
2020-01-22 23:45:55 <mps> maybe one day we will see systemd-vim
2020-01-22 23:45:59 <Cogitri> That got old over time
2020-01-22 23:46:18 <armin> Cogitri: can't say so. nowadays systemd runs pretty good, but it took many many years of funny screens on beamers in the underground stations showing "[***   ] A stob job is running: " messages that hung infinitely to get there.
2020-01-22 23:46:31 <armin> i've seen them every day.
2020-01-22 23:46:42 <Cogitri> Probably due to services not terminating probably
2020-01-22 23:47:12 <armin> exactly, which in turn is a bug if there isn't a timeout to send kill -9 if a service doesn't behave.
2020-01-22 23:47:46 <Cogitri> That has been configurable since like systemd 200-ish though?
2020-01-22 23:47:53 <armin> if you even go as far as systemd goes about re-starting a service if it dies (why would it die? is it buggy?) why then would you stop rebooting your system if one does and hang forever?
2020-01-22 23:47:58 <armin> yeah no, that's just a bug.
2020-01-22 23:48:33 <mps> do we need to discuss systemd issues on #alpine-linux
2020-01-22 23:48:38 <Cogitri> Doesn't seem buggy to me, I don't want to kill my database server in the middle of a dump
2020-01-22 23:48:54 <armin> see, that's what i meant with going down that road...
2020-01-22 23:48:57 <Cogitri> No, but at this time I don't think we're annoying others, mps :)
2020-01-22 23:50:38 <mps> Cogitri: well, we never know for sure, there are people who want to help our users and can be disturbed or irritated by OT discussion
2020-01-22 23:51:15 <mps> although, have to admit that I sometimes do exactly this, i.e. OT talks
2020-01-22 23:51:34 <Cogitri> https://github.com/a-schaefers/systemE one last offtopic message since it fits so well :)
2020-01-22 23:52:21 <mps> Cogitri: :D
2020-01-22 23:52:55 <codebam> haha in elisp
2020-01-22 23:52:57 <codebam> that's great
2020-01-22 23:53:18 <mps> ok, good night all!
2020-01-22 23:53:39 <codebam> night mps! :)
2020-01-22 23:56:02 <Cogitri> Night :)
2020-01-23 00:04:42 <mckaygerhard> nah Cogitri systeme depends on sinit.. so prefers sinit!
2020-01-23 00:13:24 <codebam> I followed this to get lxc set up for unprivledged users but the bridge I'm using doesn't seem to have internet? (I can't ping)
2020-01-23 00:13:26 <codebam> https://linuxcontainers.org/lxc/getting-started/
2020-01-23 00:13:45 <codebam> I didn't create images using download, I used the oci template to pull the alpine oci image
2020-01-23 00:14:13 <codebam> how can I give my lxc containers (or br0 I guess?) internet access?
2020-01-23 00:17:38 <codebam> never mind. I should just look at our wiki
2020-01-23 00:26:14 <attos> Is there something like a ld.so.cache and a way of listing that cache like "ldconfig -p" in alpine linux ?
2020-01-23 00:42:50 <iXNyNe> What is involved in becoming a sponsor for example on this page? https://alpinelinux.org/sponsors/
2020-01-23 01:58:17 <danieli> iXNyNe: in those cases, providing infrastructure
2020-01-23 01:59:18 <danieli> rapidswitch/iomart provides a (pretty damn reliable) box, packet and linode provides various cloud-y infrastructure which most of the stuff runs, vpsfree i don't remember, gigabyte is where clandmeter works, they also have a site with some services there too iirc
2020-01-23 01:59:26 <danieli> fastly provides dl-cdn, basically a global cdn
2020-01-23 02:02:06 <iXNyNe> Gotcha. That's at a pretty big scale :) are there other donation options?
2020-01-23 02:17:50 <programmerq> iXNyNe▸ on https://wiki.alpinelinux.org/wiki/Alpine_Linux:Contribute it looks like the closest is a small donation/gift to specific developers (if they've listed some way to receive such)
2020-01-23 02:17:56 <programmerq> https://wiki.alpinelinux.org/wiki/Alpine_Linux:Developers
2020-01-23 02:32:25 <iXNyNe> Cool, thanks!
2020-01-23 03:22:24 <Amolith> I'm trying to run LXC containers on Alpine with LXD. It will launch Alpine images fine but fails with anything Debian-based and that's necessary for a service I'm looking to run. Does anyone have any idea how to fix this: https://bin.nixnet.services/?1de89d91da8dcfe0#6XkGmnsrmNcezM5aFegg3PYbPExut4goTLgpwtxtxZjW
2020-01-23 03:27:32 <Amolith> I assumed it would be better to ask here rather than #lxcontainers
2020-01-23 04:16:33 <codebam> why is the i/o scheduler set to "none"?
2020-01-23 04:18:02 <codebam> https://termbin.com/hy8r
2020-01-23 04:25:05 <codebam> never mind, it's because it's an ssd apparently
2020-01-23 12:24:29 <llnu> has anyone here managed to set up a ceph radosgw s3 like object storage with alpine?
2020-01-23 13:58:51 <llnu> i managed to get anwsers on ceph irc. the correct way to start radosgw is "/usr/bin/radosgw -f --cluster ceph --name client.rgw."client_name" -setuser ceph --setgroup ceph"
2020-01-23 14:01:35 <llnu> a more related question: after an upgrade i got a "boot error". i reinstalled syslinux, but didnt fix. where should i start looking? im on edge
2020-01-23 14:31:32 <tomz-alpine> he;;p
2020-01-23 14:31:36 <tomz-alpine> hello
2020-01-23 14:41:38 <tomz-alpine> is the BCM4360 802.11ac Wireless supported in alpine?
2020-01-23 14:42:21 <mps> tomz-alpine: do you know what kernel config option is needed for it? or module name?
2020-01-23 14:42:32 <mps> is it wl.ko by chance
2020-01-23 14:46:27 <tomz-alpine> bcma
2020-01-23 14:47:44 <mps> option CONFIG_B43_BCMA=y is set
2020-01-23 14:48:31 <mps> but I think firmware for it is not packaged
2020-01-23 14:50:41 <tomz-alpine> o
2020-01-23 14:50:44 <tomz-alpine> ok
2020-01-23 14:52:04 <mps> I remember I did once set up similar device on one old macbook, but had to find firmware on the net and add it manually
2020-01-23 14:59:55 <programmerq> modern linux recommendations for ramdisk is to use tmpfs. I want to try running docker out of a ramdisk, which means I want an ext4 filesystem backed by a block ramdisk.
2020-01-23 15:00:18 <dalias> why do you want to do that?
2020-01-23 15:00:27 <dalias> it seems like a really bad idea.
2020-01-23 15:00:39 <programmerq> I want a performant ephemeral build server
2020-01-23 15:00:55 <dalias> if you want performance why would you add an extra layer of overhead?
2020-01-23 15:01:20 <programmerq> extra layer of overhead? compared to what? I'm trying to cut out an actual disk in favor of running in ram.
2020-01-23 15:01:30 <dalias> vs tmpfs
2020-01-23 15:01:43 <programmerq> docker's storage drivers require either ext4 or xfs with d_type=on
2020-01-23 15:01:50 <programmerq> hence looking away from tmpfs
2020-01-23 15:02:19 <programmerq> well, actually-- let me just try it with tmpfs. it may still work
2020-01-23 15:02:40 <dalias> docker does not require any ext4 or *any* control over the underlying disk/filesystem
2020-01-23 15:03:51 <dalias> anyway in the abstract, the reason why i said extra layer of overhead
2020-01-23 15:04:28 <programmerq> I worked for docker for five years. we only list ext4 and xfs as the supported backing filesystem when using the overlay2 driver (which uses overlayfs in the kernel). It may still function with tmpfs, but it's certainly not tested with tmpfs.
2020-01-23 15:04:55 <programmerq> but non-supported-but-still-functions will work for this harebrained experiment.
2020-01-23 15:05:05 <dalias> normal topology with ext4 or some other filesystem is:
2020-01-23 15:05:23 <programmerq> I just have a big fat new homelab server with way more ram than I need so I want to play with it.
2020-01-23 15:05:29 <dalias> ... <---> fs cache layer <----> filesystem driver <----> block device
2020-01-23 15:05:33 <dalias> with tmpfs it's just
2020-01-23 15:05:36 <dalias> ... <---> fs cache layer
2020-01-23 15:05:46 <programmerq> ah, so that's why it's recommended to not use ext4 in a ramdisk-- you'll get double copies.
2020-01-23 15:05:48 <dalias> all tmpfs means is using the fs cache layer with no backing behind it
2020-01-23 15:06:13 <dalias> well actually fs cache layer can, in some cases, share pages with the block buffer layer for block devices
2020-01-23 15:06:17 <dalias> but it's still more overhead
2020-01-23 15:07:24 <dalias> using tmpfs is just cutting out a huge amount of code/layers involved
2020-01-23 15:07:34 <programmerq> indeed.
2020-01-23 15:07:59 <programmerq> that makes a lot of sense. thanks for the explanation!
2020-01-23 15:08:05 <dalias> np
2020-01-23 16:50:53 <telmich> re
2020-01-23 16:51:31 <telmich> Does anyone have experience with touch screens that don't work (i.e. nothing happens), but happily show up in the xinput list? I checked also the list-props and the device is enabled
2020-01-23 16:51:36 <telmich> however no events on xev
2020-01-23 18:07:40 <PICCORO> there's somebody that can tell me how to manually make bootable the installed system?, i perform "setup-disk -m sys -s 4096 /dev/sda" and choose none for apk cache and others.. so my intentions are created only one partition install with grub /boot in same partition.. since setup-disk just copy the system, i manually edited the syslinux cfg file to use "/boot" envent "/" but still not works
2020-01-23 18:11:04 <mps> PICCORO: when using setup-disk this way you have to create partition and mkfs on it, then mount it and after that run setup-disk with parameter
2020-01-23 18:22:20 <PICCORO> hey my intention are to use only one partiton but setup disk forced me to created boot then.. i made all manually but if setup to use only onepartition (no second for boot) system dont boot, setup disk seems always assume that boot are in a individual partition and syslinux wrote as is mps
2020-01-23 18:32:02 <mps> I'm not sure will it work without separate boot partition
2020-01-23 18:36:21 <PICCORO> seems in old days works.. but now.. i must wiath to install the another linux to make it boot able.. pufff what a shame dissapoinment
2020-01-23 18:37:08 <mps> no, you don't need another linux if you follow wiki page about manual install
2020-01-23 18:37:34 <mps> I did that few times from alpine boot media (usb)
2020-01-23 18:40:40 <dafu> hi, is there a trick to installing modules on a ram install system? zfs works but wireguard does not show up in modloop.
2020-01-23 18:42:05 <mps> dafu, it there are no alreade prepared file in /etc/mkinitfs/features.d/ dir you can create one and add it to /etc/mkinitfs/mkinitfs.conf
2020-01-23 18:42:16 <mps> and run mkinitfs, ofc
2020-01-23 18:45:08 <dafu> thanks, i was wondering about that. i'll give it a try
2020-01-23 18:47:50 <PICCORO> mps the manual install of the wiki page are several days outdated.. that: https://wiki.alpinelinux.org/wiki/Setting_up_disks_manually#Custom_partitioning and that https://wiki.alpinelinux.org/wiki/Installing_Alpine_on_HDD_dualbooting
2020-01-23 18:48:34 <PICCORO> in fact i want to keep up to date those pages but seems process are not workinig well
2020-01-23 18:49:12 <PICCORO> the problem are that need to make also for one partition only with swap file event swap partition
2020-01-23 18:49:26 <PICCORO> others linux can do that! withou mayor process
2020-01-23 18:50:44 <mps> PICCORO: sorry, I don't have time to write even short guide how I did that, but I think you have enough knowledge to solve it by 'try, error, fix' method
2020-01-23 18:51:48 <PICCORO> hey here said https://wiki.alpinelinux.org/wiki/Classic_install_or_sys_mode_on_Raspberry_Pi#Installation that an update in root fresh copy system must be done after that.. what update? apk files? boot? what? mps
2020-01-23 18:56:27 <mps> RPis are somewhat special and not good to follow if one wants to install on PC
2020-01-23 19:12:40 <PICCORO> mps i uses that only as reference..
2020-01-23 19:13:09 <PICCORO> i follow the two prevous.. install sucess and works but boot does not.. i must install recue boot with other live linux to
2020-01-23 19:23:50 <mps> maybe you have to do 'dd bs=440 count=1 conv=notrunc if=/usr/share/syslinux/mbr.bin of=/dev/sdX' or something similar
2020-01-23 19:57:47 <bl4ckb0ne> do I need more than cups and simple-scan to recognize my printer as a scanner with simple-scan?
2020-01-23 20:02:21 <PICCORO> bl4ckb0ne: you alo need the pps and cups-sdk i never tried in alpine
2020-01-23 20:02:27 <PICCORO> *also
2020-01-23 20:16:32 <bl4ckb0ne> cant find that pps package
2020-01-23 20:24:14 <bl4ckb0ne> i managed to use my camera as a scanner, but still no printer
2020-01-23 20:24:20 <mps> dalias: regarding #11145 issue, iirc you told on #musl channel that this happens when you run firefox from terminal
2020-01-23 20:25:36 <mps> I run it on armv7 from terminal and it doesn't give any error (except missconfigured gtk theme)
2020-01-23 20:34:31 <attos> When loading the kernel module parport_pc shouldn't be there a device like '/dev/parport0' available ?
2020-01-23 20:40:08 <attos> I'm running lcd4linux on my alpine pc (alpine 3.11) and have a homebrew LCD-Display connected to the parallel port.
2020-01-23 20:40:16 <_ikke_> Not sure, shouldn't something like udev or mdev create such nodes?
2020-01-23 20:41:11 <attos> And have to provide "/dev/parport0" to the display driver.
2020-01-23 20:42:31 <attos> _ikke_: Maybe. I just loaded the kernel module parport_pc and can't find any device under "/dev".
2020-01-23 20:43:03 <attos> The module loaded correctly (looking at lsmod).
2020-01-23 20:43:59 <_ikke_> https://www.kernel.org/doc/html/v4.11/admin-guide/parport.html
2020-01-23 20:45:02 <_ikke_> seems you need specific drivers as well
2020-01-23 20:51:00 <attos> If I load the lp module, there will be  /dev/lp0 created. But this is for line printers, because when now running lcd4linux providing /dev/lp0 there is error: "HD44780: ioctl(/dev/lp0, PPCLAIM) failed: 22 Invalid argument", "HD44780: could not initialize parallel port!". I think I have to provide the raw parallel port device which was /dev/parport0 in the past.
2020-01-23 20:52:55 <_ikke_> attos: try modprobe ppdev
2020-01-23 20:55:56 <attos> _ikke_: Now it works :) module ppdev creates /dev/parport0 and my homebrew parallel LCD-Display for lcd4linux works. Many thanks.
2020-01-23 20:56:47 <AinNero> attos: nice thing to make use of these ports
2020-01-23 20:56:55 <AinNero> how is it wired up?
2020-01-23 20:56:58 <Hund> I just installed Alpine on my Raspberry Pi, but I can't figure out how to get any network up and running?
2020-01-23 20:57:31 <Hund> I checked the wiki, but it didn't help.
2020-01-23 20:57:34 <AinNero> Hund: did you (or as past of the installer) run setup-interfaces ?
2020-01-23 20:57:38 <AinNero> s/past/part/
2020-01-23 20:57:47 <Hund> No. I didn't know of that.
2020-01-23 20:58:06 <AinNero> are you running it diskless?
2020-01-23 20:58:13 <_ikke_> attos: just searched around :-)
2020-01-23 20:59:01 <Hund> AinNero: Yes.
2020-01-23 20:59:06 <Hund> It's default.
2020-01-23 21:01:42 <AinNero> ah yes, if you just boot, you dont have network until you run setup-interfaces
2020-01-23 21:01:55 <AinNero> or run the installer in general, even if you dont install to disk
2020-01-23 21:02:28 <AinNero> i use lbu commit to make the network config persistent
2020-01-23 21:04:07 <Hund> I tried running lbu commit with and without -d as suggested in the wiki, but it just prints out a lot of information about how to use it.
2020-01-23 21:04:11 <Hund> What am I doing wrong?
2020-01-23 21:04:41 <_ikke_> I think the first step you need to do is setup-alpine
2020-01-23 21:04:46 <attos> AinNero: Have a look at https://lcd4linux.bulix.org/wiki/HD44780
2020-01-23 21:07:05 <AinNero> attos: nice
2020-01-23 21:07:13 <Hund> _ikke_: Thanks!
2020-01-23 21:17:10 <dalias> mps, i don't think the terminal part is important; you just need the output going somewhere you can see it
2020-01-23 21:17:29 <dalias> mps, probably what's happening is some component is performing fopen() or isatty() in the sandbox
2020-01-23 21:17:37 <dalias> maybe libavcodec or something
2020-01-23 21:18:18 <mps> dalias: sorry, have incident with cloud server now
2020-01-23 21:21:36 <mps> again some network node admins plays with routing
2020-01-23 21:43:39 <daftaupe> hello, new user here, I'd like to know if packages can be added between 3.11.3 and 3.11.4 for example ? Or are new packages to be expected when 3.12 will be released ?
2020-01-23 21:44:15 <_ikke_> stable releases will typically not see new packages
2020-01-23 21:44:41 <_ikke_> they are added to edge first, and when they are in main or community, will make it in the next release
2020-01-23 21:45:01 <daftaupe> that makes sense, I just wanted that confirmed, thanks _ikke_ :)
2020-01-23 21:45:27 <daftaupe> alright ! I'm waiting for Thunderbird hehe
2020-01-23 21:46:29 <_ikke_> It's in testing atm
2020-01-23 21:46:52 <daftaupe> that's what I've noticed
2020-01-23 21:46:59 <daftaupe> so I guess it needs more testing :)
2020-01-23 21:47:02 <_ikke_> So if you have the opportunity, please test it in edge and let the maintainer know everything is working
2020-01-23 21:47:45 <daftaupe> well I have one question, it seems rather easy to go from release to edge, but how can one switch back to a release when using edge ?
2020-01-23 21:47:59 <_ikke_> apk upgrade -U --available
2020-01-23 21:48:13 <_ikke_> the last argument is important
2020-01-23 21:48:20 <_ikke_> (option)
2020-01-23 21:48:30 <daftaupe> ah so it's the same process ?
2020-01-23 21:48:34 <_ikke_> yes
2020-01-23 21:49:25 <daftaupe> ok, well I might try to switch to edge, now that I know I'll be able to easily switch back
2020-01-23 21:50:39 <MY-R> daftaupe, after when you taste the EDGE you wont back to release :P
2020-01-23 21:51:52 <daftaupe> hehe, well I'm just basically missing Thunderbird, what's in release I like to far :)
2020-01-23 21:53:03 <daftaupe> thanks for your time _ikke_ :)
2020-01-23 21:53:08 <daftaupe> gotta go now
2020-01-23 21:55:32 <hohum> I installed nfs-utils but I can't mount a remote NFS on my alpine box
2020-01-23 21:55:37 <hohum> https://pastebin.com/xsezBhWq
2020-01-23 21:55:40 <hohum> what am I doing wrong?
2020-01-23 21:57:33 <clandmeter> can you try install util-linux?
2020-01-23 21:58:16 <hohum> yeah that got rid of the flock error message
2020-01-23 21:58:19 <hohum> but now I'm getting:
2020-01-23 21:58:25 <hohum> mount.nfs: Protocol not supported
2020-01-23 22:07:32 <mps> dalias: I'm back. didn't noticed anything that could be used to reproduce your issue
2020-01-23 22:09:55 <dalias> let me see if i can find what triggers it
2020-01-23 22:11:12 <dalias> hmm not youtube
2020-01-23 22:11:34 <mps> I remember your msg on #musl but have no idea why it works fine on my arm32
2020-01-23 22:14:23 <mps> ah, 'while testing mallocng and noticed this', to quote you
2020-01-23 22:18:58 <hohum> I'm guessing my problem is an alpine linux box can't be both an nfs server and an nfs client because this is the only machine on my network behaving badly
2020-01-23 22:19:20 <hohum> nfsmount won't start
2020-01-24 01:30:04 <tomz-alpine> I have install wine in alpine but cant find it only wine64 is their
2020-01-24 01:30:07 <tomz-alpine> ?
2020-01-24 01:30:54 <pltrz> i believe wine needs 32-bit libraries
2020-01-24 01:31:15 <tomz-alpine> how an i add a 32 bit library in alpine?
2020-01-24 01:31:29 <pltrz> as I understand it, you cannot run 32-bit programs using a musl-libc
2020-01-24 01:31:41 <pltrz> since it only supports 64-bit
2020-01-24 01:32:03 <tomz-alpine> ok
2020-01-24 01:32:05 <tomz-alpine> thanks
2020-01-24 01:32:16 <pltrz> no problem
2020-01-24 08:46:08 <markand> hello
2020-01-24 08:46:43 <markand> is that normal that mails sent to the mailing list aren't sent back to the original poster?
2020-01-24 08:47:07 <markand> I mean, if I send a mail to a usual mailing list, I then receive my own mail back (so I can answer/resolve it myself or correct if needed)
2020-01-24 08:48:09 <mps> :)
2020-01-24 08:50:03 <mps> it is not normal (imo) but that is how it works on alpine mailing lists
2020-01-24 10:03:11 <markand> for some reasons, sometimes List- headers disappear too :(
2020-01-24 10:04:22 <markand> but that's probably because some people use “answer all” rather than “answer list”
2020-01-24 15:09:07 <dalias> uhg, intel gpu hang happened again even with patch
2020-01-24 15:09:13 <dalias> [262230.211351] Asynchronous wait on fence i915:Xorg[11227]:34a866 timed out (hint:intel_atomic_commit_ready+0x0/0x50 [i915])
2020-01-24 15:09:18 <dalias> that seems like a new error tho
2020-01-24 15:09:39 <mps> yes, this is new
2020-01-24 15:09:59 <dalias> http://ix.io/28cP
2020-01-24 15:10:05 <dalias> full log
2020-01-24 15:10:25 <dalias> system is still up, i moved to console to save debug info
2020-01-24 15:10:33 <mps> rcs0 issue again
2020-01-24 15:11:00 <dalias> *nod*
2020-01-24 15:11:08 <mps> did I told you that adding some swap mitigate this in my case
2020-01-24 15:11:12 <dalias> yes
2020-01-24 15:11:20 <dalias> i have 2GB swap
2020-01-24 15:11:30 <mps> also I have same size
2020-01-24 15:11:46 <dalias> i wonder if i should update my Xorg/drivers too
2020-01-24 15:11:49 <dalias> could that help?
2020-01-24 15:12:07 <mps> didn't helped in my case
2020-01-24 15:12:12 <mps> no differences
2020-01-24 15:12:29 <mps> but ofc, you can try
2020-01-24 15:12:36 <dalias> heh lovely version number:
2020-01-24 15:12:37 <dalias> xf86-video-intel-2.99.917_git20170325-r4
2020-01-24 15:13:16 <dalias> is that latest? i don't see anything later
2020-01-24 15:13:29 <mps> yes, I have this version
2020-01-24 15:14:04 <dalias> ok it recovered and gl is working again
2020-01-24 15:14:19 <dalias> killed and restarted the 3d app just to be safe
2020-01-24 15:14:51 <dalias> hmm maybe there's a userspace driver package too i should update..?
2020-01-24 15:15:20 <mps> I plan to try 5.5-rc7 over the weekend to see if that is fixed
2020-01-24 15:15:22 <dalias> mesa-dri-intel? or something else?
2020-01-24 15:15:45 <dalias> yes there's a newer version of it
2020-01-24 15:15:58 <dalias> 18.3.5-r2 installed, 19.3.2-r2 available
2020-01-24 15:15:59 <mps> I updated to the latest alpine version but without luck
2020-01-24 15:16:03 <dalias> :/
2020-01-24 15:16:15 <dalias> the kernel patch seems to have improved it a lot anyway
2020-01-24 15:16:24 <dalias> i went for several days with no hangs instead of one or more a day
2020-01-24 15:16:38 <dalias> and it wasn't a hard hang but recovered after a few min
2020-01-24 15:16:41 <mps> my is mesa-dri-intel-19.2.7-r0
2020-01-24 15:16:53 <dalias> (no idea why timeout is in minutes rather than milliseconds...)
2020-01-24 15:17:46 <mps> but nowadays I work more and more on ARM workstations
2020-01-24 15:17:47 <dalias> if this happens it really ought to behave like a video card hotplug event, disconnect everything from it, re-init from clean state
2020-01-24 15:18:57 <mps> that shouldn't happen, imo
2020-01-24 15:24:38 <dalias> isn't it the best possible ux?
2020-01-24 15:24:58 <dalias> X should recover fine
2020-01-24 15:25:25 <dalias> i'd rather screen flash out for a sec and come back than screen stay frozen for 3 minutes
2020-01-24 15:25:56 <dalias> this is similar to the hell i fought with cdrom drives ~2000-2005
2020-01-24 15:26:18 <dalias> they'd get in permanently hung state but would reset fine if the system had bothered to reset them
2020-01-24 15:26:28 <dalias> so i hotplugged IDE cables and that was ... fun :-p
2020-01-24 15:26:30 <mps> ofc, but that didn't happened for me on 5.2 (and some early 5.3) versions
2020-01-24 15:26:37 <ncopa> i think the recommendation is to use modesetting driver instead of xf86-video-intel nowdays
2020-01-24 15:26:44 <ncopa> hi dalias, btw :)
2020-01-24 15:26:59 <dalias> after burning out a couple drives that way i finally decided never to buy non-usb optical drives again
2020-01-24 15:27:13 <ncopa> :D
2020-01-24 15:27:17 <dalias> hi ncopa
2020-01-24 15:27:29 <ncopa> dalias: sorru for not help you with more testing of time64
2020-01-24 15:27:33 <ncopa> sorry*
2020-01-24 15:27:43 <dalias> i'm pretty sure i am using the modesetting driver; it's nonclashing with fbcon in any case
2020-01-24 15:28:24 <ncopa> what do i need to help you with testing it? is it enough to LD_PRELOAD current binaries or do i need to rebuild lots of packages?
2020-01-24 15:29:21 <dalias> it's using intel driver, but intel says:
2020-01-24 15:29:21 <dalias> [ 34862.067] (II) intel(0): Using Kernel Mode Setting driver: i915, version 1.6.0 20190822
2020-01-24 15:29:39 <ncopa> ok. good. thats the correct one
2020-01-24 15:29:55 <dalias> ok
2020-01-24 15:30:18 <dalias> ok, so for time64
2020-01-24 15:31:11 <dalias> adelie (awilfox) and yocto/oe (khem) have done most of the hard work of testing packages with time64 and patching things that broke
2020-01-24 15:31:31 <dalias> adelie has a wiki on it: https://wiki.adelielinux.org/wiki/Project:Time64
2020-01-24 15:33:46 <dalias> dropping in a musl 1.2 libc (not LD_PRELOAD but same idea) will work with existing packages built for time32, but there's not much value in testing it; it's unlikely anything is still broken there
2020-01-24 15:34:00 <dalias> the two nontrivial tasks are
2020-01-24 15:34:19 <dalias> 1. patching packages that aren't time64-ready to be built with new musl, and
2020-01-24 15:34:33 <dalias> 2. determining how you want to make the upgrade path smooth (or how smooth you want to make it)
2020-01-24 15:35:52 <dalias> an inconsistent upgrade with a mix of libs built against time32 and time64 is going to have a small, but sufficient to make things break here and there), number of abi mismatches between libs/apps
2020-01-24 15:36:40 <ncopa> so, basically, it would be good if we could build world with musl 1.2
2020-01-24 15:36:52 <dalias> either you can do a bunch of analysis and put some depends/conflicts relationships in the packages affected to prevent this, or you can tell users this is a case where they want to just do a whole dist-upgrade at once
2020-01-24 15:36:57 <dalias> yeah
2020-01-24 15:37:18 <dalias> but before you do that, get the patches for known-broken stuff from adelie and yoe
2020-01-24 15:37:26 <ncopa> *nod*
2020-01-24 15:37:33 <ncopa> and fix the rest while doing it
2020-01-24 15:37:34 <dalias> so you don't end up hitting their same issues again and spending time trying to figure out how to fix
2020-01-24 15:37:41 <ncopa> ofc
2020-01-24 15:39:03 <ncopa> i guess we could also try find all references to time32 to find the exact packages that needs to be rebuild
2020-01-24 15:39:38 <ncopa> which symbol(s) is it?
2020-01-24 15:40:22 <ncopa> or is it anything that has `time_t` in sources
2020-01-24 15:41:03 <dalias> well there are two issues
2020-01-24 15:41:27 <dalias> to be Y2038-ready, anything using time needs to be rebuilt. but that's not the biggest concern right now as long as you *can* rebuild it later
2020-01-24 15:41:42 <dalias> the bigger issue is ABI boundaries between libraries
2020-01-24 15:42:17 <dalias> wherever you have an ABI boundary with a third-party library involving time_t-derived types, the library and its consumers need to be upgraded in sync
2020-01-24 15:44:32 <ncopa> right, there are not any specific libc functions/symbols we could look for, beause other libs would as well use time_t
2020-01-24 15:44:45 <ncopa> and there is not way to detect those in compiled binaries
2020-01-24 15:45:09 <ncopa> i guess that means that when we do the switch, we will need to rebuild world
2020-01-24 15:45:41 <ncopa> we do that for every stable release (every 6 month), so i guess thats doable
2020-01-24 15:46:01 <ncopa> hum
2020-01-24 15:46:09 <ncopa> its gonna be tricky
2020-01-24 15:46:58 <ncopa> but the good news is that we have another project that may require full rebuild
2020-01-24 15:47:07 <ncopa> we are talking about chaining the package format of .apk
2020-01-24 15:47:31 <ncopa> one that is in place we will probably want to rebuild edge from scratch
2020-01-24 15:48:13 <dalias> more importantly it's not the libc symbol that matters when it comes to upgrading libs/consumers in sync
2020-01-24 15:48:18 <dalias> it's the  types in their public interfaces
2020-01-24 15:48:30 <dalias> musl 1.2 does not break any of the abi between libc and libc consumer
2020-01-24 15:48:31 <ncopa> yup, i realized that
2020-01-24 15:48:49 <dalias> but the types defined by libc headers can also participate in defining ABI between third-party libs and their consumers
2020-01-24 15:49:40 <dalias> if you want to follow musl 1.2 but not switch to time64 immediately, there is a hack you can do
2020-01-24 15:49:57 <dalias> on 32-bit archs, build musl-dev headers from musl 1.1.x and install those
2020-01-24 15:50:48 <ncopa> right, so the problem is mainly the 32 bit arches
2020-01-24 15:51:03 <dalias> on 64-bit archs nothing changes
2020-01-24 15:51:13 <ncopa> make sense
2020-01-24 15:51:38 <ncopa> btw, this is something we should have discussed in #alpine-devel channel, not here
2020-01-24 15:51:42 <dalias> sorry
2020-01-24 15:51:58 <ncopa> you are warm welcome in there :)
2020-01-24 18:17:22 <m_wynn> Hello!  I was recently informed by a coworker that Python 3 is not stable for use running on alpine linux.  As 2.7 is losing support quickly, do you know when Python 3 will be considered stable?  Thanks
2020-01-24 18:27:55 <_ikke_> m_wynn: I don't see how python3 would not be stable for use in Alpine
2020-01-24 18:28:20 <Cogitri> I don't know where your coworker picked that ip
2020-01-24 18:28:28 <_ikke_> Most packed / applications have already been updated to only use python3
2020-01-24 18:28:36 <m_wynn> Cool thanks
2020-01-24 18:28:38 <m_wynn> lmao
2020-01-24 18:28:54 <_ikke_> The next stable release will most likely drop python2 completely
2020-01-24 20:17:46 <codebam> hey I'm trying to use virt-manager on alpine to remotely connect to libvirtd but the error it's giving me is that I don't have ssh_askpass? I installed openssh-askpass but it doesn't provide the file that virt-manager wants so I'm not sure what to do
2020-01-24 20:17:50 <codebam> this is the error I got
2020-01-24 20:18:15 <codebam> https://0x0.st/irp8.txt
2020-01-24 20:19:55 <codebam> alternatively I'd gladly use ssh keys, but I'm not sure how to do that from within virt-manager. it doesn't give me an option for it
2020-01-24 20:21:18 <topjor> If I am correct it just uses the default ssh configuration, so if you set it up for ssh, and you use virt-manager, it will work via the ssh configuration
2020-01-24 20:22:10 <codebam> yeah but on the client (my laptop) that's connecting to libvirtd on the server I don't have ssh-askpass
2020-01-24 20:22:17 <codebam> and openssh-askpass doesn't provide it
2020-01-24 20:22:26 <jaz> hello peeps o/  i think the description for persistent installation in the wiki is a little wrong. from what i've understood it's based on bazooka8's post (which is linked at the end of the paragraph). the difference is that he copies over the contens of the new /boot/ to the old boot dir which isn't happening in the instructions in the wiki.
2020-01-24 20:22:34 <jaz> sorry for wall of text ^^
2020-01-24 20:23:53 <jaz> uhm.. for raspberryPI that is
2020-01-24 20:25:19 <topjor> codebam, I am not that familiar to the alpine linux version of openssh-askpass, but if you want to setup ssh keys to authenticate, you can follow any tutorial for setting up ssh remote access with keys as virt-manager uses the same way as you would normally use ssh to connect over ssh to a server
2020-01-24 20:25:45 <codebam> topjor oh awesome. I didn't realize it would just call ssh directly, but that does make sense actually
2020-01-24 20:25:53 <codebam> so yeah I'll just set up ssh keys
2020-01-24 20:25:54 <codebam> that works
2020-01-24 20:28:54 <topjor> codebam, looking at the package, it seems to only provide a gtk verson and some profile configuration so I dont know whats up with that
2020-01-24 20:30:35 <codebam> topjor there's virsh for a non-gtk version which is provided by libvirt-client
2020-01-24 20:31:58 <jaz> anyway, i just wanted to let you guys know. have a nice day/evening/whatever o/
2020-01-24 20:32:01 <topjor> codebam, I meant the openssh-askpass package
2020-01-24 20:36:21 <codebam> topjor ohh yeah
2020-01-24 20:36:52 <codebam> yeah it seems like that's a mistake that it doesn't provide the ssh-askpass command
2020-01-24 20:40:05 <codebam> okay so. I got it to connect but now it's asking for `virtqemud-sock` which I guess allows it to network qemu hosts
2020-01-24 20:40:12 <codebam> not sure what provides that...
2020-01-24 20:42:17 <codebam> do I need libvirt to be running on the client as well as the host?
2020-01-24 20:43:32 <topjor> on other distro's I only needed virt-manager and its dep's for remote connections if I am correct, but it could be that I also had gnome boxes installed which requires the whole libvirt stack for the vms
2020-01-24 20:45:12 <codebam> well um... this is the error I get when I try to connect anyways https://termbin.com/0btf
2020-01-24 20:45:24 <codebam> I tried installing and starting libvirtd on the client but that didn't help
2020-01-24 20:46:35 <topjor> does `/var/run/libvirt/virtqemud-sock` exist on the remote host and does your user have permissions to it?
2020-01-24 20:47:43 <codebam> let me check
2020-01-24 20:49:01 <codebam> oh. I think I had to install libvirt-qemu
2020-01-24 20:49:05 <codebam> let me restart the daemon now
2020-01-24 20:50:12 <codebam> that worked
2020-01-24 20:50:16 <codebam> awesome!
2020-01-24 20:50:19 <topjor> libvirt is just a connection and management tool for hypervisors qemu
2020-01-24 20:50:35 <topjor> *like qemu
2020-01-24 20:51:00 <codebam> yeah and for xen and lxc
2020-01-24 20:51:18 <codebam> I would use lxc but it's an alpine host so I don't think I'd be able to use systemd distros
2020-01-24 20:51:25 <codebam> and I want to use this for docker
2020-01-24 20:52:01 <topjor> depends on its containerisation and startup system for those
2020-01-24 20:52:03 <codebam> lxc for anything that doesn't require me to run abnormal containers though
2020-01-24 20:52:18 <codebam> well yeah I just don't think it's a good idea to run docker inside lxc
2020-01-24 20:53:37 <topjor> I run CoreOS  (container linux) for my docker hosts, its a nice distro for just running containers
2020-01-24 20:54:16 <codebam> oh cool. I might try that. this is just running on a cheap desktop I bought the other day and put an SSD in
2020-01-24 20:55:09 <codebam> got a really good deal on it. was like $75 for 1TB, 12 GB of RAM, and an i5-3470
2020-01-24 20:55:33 <topjor> sounds not too bad
2020-01-24 20:55:45 <topjor> yea, I just use my pc for my home dev and lab stuff with some qemu/kvm vms through libvirt and coreos
2020-01-24 20:56:11 <alxsdf> codebam: where did you pick up that? :o
2020-01-24 20:56:19 <codebam> alxsdf kijiji
2020-01-24 20:56:21 <codebam> :)
2020-01-24 20:56:55 <alxsdf> ah, fair
2020-01-24 22:08:31 <dalias> opencolorio package was broken by version bump of yaml-cpp :(
2020-01-24 22:08:35 <dalias> see https://github.com/jbeder/yaml-cpp/issues/789
2020-01-24 22:08:48 <dalias> can someone rebuild a fixed version?
2020-01-24 22:09:17 <maldridge> _ikke_: Cogitri I'm the coworker
2020-01-24 22:09:43 <maldridge> I've observed some problems with pycrypto when building C wheels with pip
2020-01-24 22:10:03 <maldridge> ofc the alpine answer would be to just use the version from apk, but this means that I now need to synchronize my upgrade cycle with alpine's
2020-01-24 22:10:58 <_ikke_> maldridge: when did you notice those issues?
2020-01-24 22:11:10 <_ikke_> pycrypto is old
2020-01-24 22:11:22 <maldridge> around the time that void updated to py3, I wanted to see if alpine was having the same problems
2020-01-24 22:11:38 <maldridge> I think it was ~4 months ago but I can check git history to be sure
2020-01-24 22:11:42 <_ikke_> We've decided to replace pycrypto with cryptodome
2020-01-24 22:11:48 <_ikke_> that was not that long ago
2020-01-24 22:11:50 <_ikke_> o
2020-01-24 22:12:20 <maldridge> well I don't use any of the primitives anyway, but I still need them for ansible to buidl
2020-01-24 22:12:22 <maldridge> *build
2020-01-24 22:12:45 <_ikke_> maldridge: cryptodome presents itself as pycrypto
2020-01-24 22:13:00 <_ikke_> 852eb6ae7cd4c663659830dc4e192fbfcab3cdef
2020-01-24 22:13:27 <maldridge> hmm, so I should be able to pip install it prior to ansible and pip will pick it up as a replacement?
2020-01-24 22:14:30 <_ikke_> you should be able to install https://pypi.org/project/pycryptodome/
2020-01-24 22:14:40 <_ikke_> "All modules are installed under the Crypto package."
2020-01-24 22:15:01 <maldridge> hmm, I'll take a look
2020-01-24 22:15:14 <maldridge> that would save me a C compiler on every host and make python a little more palletable
2020-01-24 22:15:44 <_ikke_> I have no idea what kind of issues you were facing, but pycryptodome was built to replace pycrypto
2020-01-24 22:16:43 <maldridge> I'll take a look into it
2020-01-24 22:17:03 <maldridge> I've been kicking along a python virtualenv for a long time now for this, some of the packages may also have become stale
2020-01-24 23:21:56 <ddevault> does mkinitfs automatically bring in any shared libraries necessary for executables listed in $feature.file?
2020-01-24 23:25:54 <ddevault> yes
2020-01-25 00:04:08 <diegommm> Hi! I'm working on a small home-project and really need to use the '-T' option of command 'mount' within 'busybox-suid' package. I checked the source and all that has to be done to enable it is change to '=Y' in the busyboxconfig 'CONFIG_FEATURE_MOUNT_OTHERTAB'
2020-01-25 00:04:24 <diegommm> Should I make a path and ask to merge it?
2020-01-25 00:04:45 <diegommm> Is there any consequence that I should be aware of if this were to be done?
2020-01-25 00:06:28 <diegommm> ~
2020-01-25 00:09:44 <mps> diegommm: you can fill issue (feature) request on gitlab.a.o
2020-01-25 00:18:17 <russkel> diegommm, alternatively you could compile your own package
2020-01-25 00:24:25 <diegommm> russkel, I did so, but I wanted to know if there were any reason why wouldn't be this recommended. And maybe it might be useful to someone else.
2020-01-25 00:24:45 <diegommm> mps, that's it
2020-01-25 00:24:52 <diegommm> Thanks all you guys!
2020-01-25 00:25:14 <russkel> I wonder if because it's disabled by default there is some thing
2020-01-25 00:34:45 <diegommm> russkel, I think it's just a rather unused option. I'm using it just to make some wrapper around overlayfs, squashfs and chroot to have some desktop programs like chromium in alpine as desktop
2020-01-25 00:35:34 <diegommm> I'm using LD_INTERPRETER and LD_LIBRARY_PATH set inside that chroot and execute the program and all is good. Chromium works great like that
2020-01-25 00:35:39 <diegommm> (In Alpine)
2020-01-25 00:36:45 <russkel> sounds cool, what is the motivation behind it?
2020-01-25 01:01:51 <diegommm> I just wanted to use chromium on alpine but chromium is a hell of a building process. And doesn't support anything but glibc. So, I arranged a directory structure with rootfs of different distros and use them to run glibx-only programs
2020-01-25 01:03:37 <diegommm> I make branches of the main images (compressed with squashfs) and setup branches for each app I want to run. Chroot in that branch, install the app as usual for that distro
2020-01-25 01:04:57 <diegommm> And to run it on the "host" I just run it with a corresponding LD_INTERPRETER an a reasonable LD_LIBRARY_PATH so that the executable finds all its dependencies in the chrooted system, without being chrooted.
2020-01-25 01:17:42 <russkel> oh I see, pretty cool
2020-01-25 01:18:07 <russkel> kind of like containerisation
2020-01-25 01:19:33 <diegommm> Yeah, but with no containerization at all. It's just like running programs of different distros in a single distro.
2020-01-25 01:20:17 <diegommm> It means I have to "trust" the other distros. That's a burden - I have to trust BOTH Alpine and Debina (in this case)
2020-01-25 01:20:36 <diegommm> So it's just a quick and dirty hack at most
2020-01-25 01:21:04 <dalias> diegommm, alpine has a chromium package
2020-01-25 01:21:20 <diegommm> fuck me in the face
2020-01-25 01:21:37 <diegommm> you're right
2020-01-25 01:21:51 <diegommm> I might have miss-typed it twice
2020-01-25 01:22:01 <dalias> it's in community or testing
2020-01-25 01:22:09 <dalias> so it wouldn't be exposed in main-only setup
2020-01-25 01:22:28 <diegommm> Sure, I'm getting it now. So dumb
2020-01-25 01:23:19 <diegommm> Anyways, having 'mount -T' is still a hack that would be cool to have. I still use it to run skype
2020-01-25 01:23:25 <mps> :)
2020-01-25 01:23:54 <mps> sorry, couldn't resist to smile
2020-01-25 01:24:58 <diegommm> lol
2020-01-25 01:25:35 <mps> and yes, building chromium is nightmare, ime
2020-01-25 01:31:49 <russkel> haha oh dear
2020-01-25 02:02:06 <codebam> how do I do dynamic port forwarding (ssh -D) in alpine?
2020-01-25 02:02:20 <codebam> when I try to I get errors that look like "channel 3: open failed: administratively prohibited: open failed"
2020-01-25 02:04:51 <dalias> the server you're trying to ssh into has disabled it
2020-01-25 02:05:01 <dalias> see administratively prohibited
2020-01-25 02:05:15 <dalias> if it's your server, find that option in sshd_config and remove it
2020-01-25 02:07:15 <codebam> dalias yeah I'm on the server haha
2020-01-25 02:07:20 <codebam> ohh
2020-01-25 02:07:32 <codebam> okay, I'll enable it in the sshd_config. didn't realize that was a default. thanks
2020-01-25 02:07:42 <dalias> it might be a hardened setting alpine added
2020-01-25 02:07:53 <codebam> idk this was just the regular install
2020-01-25 02:08:09 <codebam> nothing changed actually. I installed it to this "server"
2020-01-25 02:08:19 <codebam> default settings, /dev/sda
2020-01-25 02:09:31 <codebam> haha with a nice comment left there
2020-01-25 02:09:41 <codebam> "feel free to enable these if your use case requires them"
2020-01-25 09:27:15 <chgangesd3jifsdn> Hey, what is the root password for the WSL version?
2020-01-25 09:27:22 <chgangesd3jifsdn> I wasn't asked for a password change
2020-01-25 09:27:43 <chgangesd3jifsdn> only for a username, now I can't install new packages :(
2020-01-25 09:28:46 <chgangesd3jifsdn> find /mnt/c -name "home" 2>/dev/null  ; I will just change root password manually
2020-01-25 09:28:52 <chgangesd3jifsdn> if I can find the rootfs on windows c:
2020-01-25 09:29:02 <chgangesd3jifsdn> ...
2020-01-25 09:31:23 <chgangesd3jifsdn> root:!::0:::::
2020-01-25 09:31:29 <_ikke_> no password set
2020-01-25 09:31:34 <_ikke_> ie, disabled
2020-01-25 09:31:41 <_ikke_> would sudo work?
2020-01-25 09:33:05 <chgangesd3jifsdn> "DESKTOP-OLH495L:~$ f3c2e807fc928347fb9aca7fff43fcb0a02a7096752b1f42dfb9a4d0f66a4eb92cd7c7a0913176e07859c247eecaac23f5fb66b7ebb5fe5237f3a66e0adc0ee4 ^C                                                                                          DESKTOP-OLH495L:~$ su -c "apk add curl"                                                                                 Password:
2020-01-25 09:33:05 <chgangesd3jifsdn>                                                       su: Authentication service cannot retrieve authentication info   " shit
2020-01-25 09:33:07 <chgangesd3jifsdn> nope
2020-01-25 09:33:10 <chgangesd3jifsdn> sudo not installed in alpine
2020-01-25 09:33:11 <chgangesd3jifsdn> by default
2020-01-25 09:34:45 <_ikke_>  https://github.com/agowa338/WSL-DistroLauncher-Alpine this is the source
2020-01-25 09:35:47 <_ikke_> sorry, no
2020-01-25 09:38:05 <_ikke_> https://github.com/agowa338/WSL-DistroLauncher-Alpine/blob/master/DistroLauncher/DistroLauncher.cpp
2020-01-25 09:39:45 <_ikke_> I have no clue how you are supposed to get root access. As you noticed, it's a 3rd party that maintains this
2020-01-25 09:40:40 <chgangesd3jifsdn> I am in the rootfs on ntfs
2020-01-25 09:40:44 <chgangesd3jifsdn> and I openned the shadow file
2020-01-25 09:41:14 <chgangesd3jifsdn> root:$1$yes$618d0e0df183b627759508d80009c9a7:18286:0:99999:7::: ; I try to set password "yes" the current one is the md5sum of "yesyes" hash
2020-01-25 09:41:33 <chgangesd3jifsdn> I also tried with salt
2020-01-25 09:42:29 <chgangesd3jifsdn> ah
2020-01-25 09:42:31 <chgangesd3jifsdn> I see now
2020-01-25 09:42:40 <chgangesd3jifsdn> $1$nA76FjpS$giKUq.0GOoQEE/F1flCpF/ ; mkpasswd --method=md5
2020-01-25 09:42:46 <chgangesd3jifsdn> it is not in base64
2020-01-25 09:43:10 <chgangesd3jifsdn> I am in
2020-01-25 09:43:37 <_ikke_> nice
2020-01-25 12:23:33 <Newbie69> Hi all, quick question, what option  do I need to enable to allow root login via putty (ssh)
2020-01-25 12:28:39 <russkel> Newbie69, that will be in your sshd configuration file found in /etc/ssh/sshd_config
2020-01-25 12:29:09 <russkel> PermitRootLogin is the option
2020-01-25 12:29:20 <Newbie69> Thanks :)
2020-01-25 12:30:01 <russkel> also, run 'man sshd_config' and then search for PermitRoot
2020-01-25 12:33:36 <Newbie69> Got it, thanks for you help
2020-01-25 13:02:14 <MY-R> sweet! I wanted to ask if can do anything to speed up mkinitfs to generate images and boom somebody did (almost!) it: <algitbot> mkinitfs:master |Sören Tempel| mkinitfs: add optional support for pigz | http://dup.pw/mkinitfs/59204d36
2020-01-25 23:23:56 <blueeyes> so I'm not really sure what's causing this, but every time I turn on my laptop now something is triggering a shutdown
2020-01-25 23:24:18 <blueeyes> not really sure what's doing it. how do I find the entry point of the last shutdown?
2020-01-25 23:27:24 <blueeyes> is there a way for me to inhibit shutdown to figure this out?
2020-01-25 23:27:33 <blueeyes> it keeps shutting down within like 30 seconds
2020-01-25 23:27:48 <_ikke_> 30 seconds after what exactly?
2020-01-25 23:28:05 <_ikke_> Cannot think of anything that would trigger a shutdown in Alpine Linux by default
2020-01-25 23:28:38 <blueeyes> booting
2020-01-25 23:29:02 <_ikke_> do you have a DM installed?
2020-01-25 23:29:19 <blueeyes> I use sway, but this happens from at the TTY after logging in
2020-01-25 23:29:26 <blueeyes> or even if I don't log in
2020-01-25 23:29:58 <_ikke_> did you try tailing /var/log/messages?
2020-01-25 23:30:03 <blueeyes> let me try that
2020-01-25 23:30:41 <blueeyes> oh
2020-01-25 23:30:43 <blueeyes> sorry
2020-01-25 23:30:54 <_ikke_> or dmesg -w
2020-01-25 23:30:55 <blueeyes> the battery died
2020-01-25 23:31:01 <blueeyes> I thought it was plugged in
2020-01-25 23:31:04 <_ikke_> doesn't help :-)
2020-01-25 23:31:08 <blueeyes> but apparently I unplugged the charger
2020-01-25 23:31:30 <_ikke_> I had something similar happen the other day (when helping someone else)
2020-01-25 23:31:46 <MY-R> maybe it wanted put laptop in sleep mode all the time?
2020-01-25 23:32:11 <_ikke_> yeah, perhaps
2020-01-25 23:39:06 <blueeyes> I created a qemu vm just so I could install windows 10 and try wsl
2020-01-25 23:39:13 <blueeyes> lol
2020-01-26 07:14:20 <Tazy> 
ACTION quietly wonders why firefox been build for all but armhf

2020-01-26 07:41:47 <_ikke_> Tazy: 0b6cd5c4ee48b8d8ce1f8c039afedddb834627da
2020-01-26 07:55:20 <Tazy> _ikke_, thanks! ..and oh well, maybe someday!
2020-01-26 14:21:23 <asimpson> How do folks here run Dropbox in Alpine? Flatpak? Chroot? Docker/postman?
2020-01-26 14:23:12 <asimpson> Podman* not postman
2020-01-26 14:35:30 <Cogitri> Not at all is the most probable answer to that
2020-01-26 14:35:50 <Cogitri> I'd recommend you to use something that's not Dropbox
2020-01-26 14:36:58 <alxsdf> asimpson: I'd suggest using a chroot
2020-01-26 14:37:50 <asimpson> Cogitri: I realize Dropbox is not FOSS and something like owncloud or syncthing would be better, but I still have occasion to use Dropbox from time to time.
2020-01-26 14:38:29 <alxsdf> I've never done it with dropbox but I've used a chroot for a bunch of other things now, it seems to be viable
2020-01-26 14:38:43 <asimpson> alxsdf: Thanks! yeah, I was leaning that direction.
2020-01-26 14:39:10 <alxsdf> I have an arch linux chroot, as I didn't get on with the debian chroot installer or any of the others
2020-01-26 14:39:34 <alxsdf> I can paste up some scripts I use for ease of use and some details of my setup if you want?
2020-01-26 14:39:57 <asimpson> alxsdf: if you’re willing. That would be great.
2020-01-26 14:39:58 <Cogitri> I'd use the Flatpak since that probably easier to set up, but a Chroot is fine too
2020-01-26 14:41:48 <alxsdf> hmm, in my experience flatpack is bad if you're on a musl system. you still have to use a chroot because almost everyone and their mother relies on non-standard properties of gnulib, that *might* not be the case with dropbox so it might be worth trying that first, as it'd certainly be easier
2020-01-26 14:46:16 <alxsdf> asimpson: http://tpaste.us/6Vvz
2020-01-26 14:46:35 <alxsdf> if you want explanation for any of the magic feel free to prod me :)
2020-01-26 14:47:03 <asimpson> alxsdf: thank you kindly! I'll muddle through and reach out if I get really stuck :)
2020-01-26 14:47:15 <Cogitri> Flatpak works just fine for me on Alpine, I use it everyday
2020-01-26 14:47:50 <alxsdf> asimpson: oh, worth noting I have my chroot in an external partition mounted to ~/u/
2020-01-26 14:48:08 <alxsdf> hence why everything references that
2020-01-26 14:48:17 <asimpson> Cogitri: I've pulled in Signal from Flatpak and it seems to work fine. I should compare Dropbox there to Dropbox in a chroot and see which works better.
2020-01-26 14:48:43 <asimpson> alxsdf: makes sense
2020-01-26 14:48:46 <alxsdf> Cogitri: huh! I should check those out then, that'd be much easier than running in a chroot
2020-01-26 15:43:50 <nico[m]6> Wow... it seems to actually work
2020-01-26 15:44:35 <Cogitri> Yup, Flatpak is a lot easier than using a chroot :)
2020-01-26 15:44:44 <Cogitri> And usually integrates better into your desktop
2020-01-26 17:16:06 <giaco> hello guys, I'm here because you have more update packages than debian stable
2020-01-26 17:16:53 <quadrathoch2> giaco, wb :)
2020-01-26 17:18:39 <giaco> quadrathoch2: if I plug an usb eth card, do you think I will get ssh access without plugging hdmi?
2020-01-26 17:21:09 <quadrathoch2> giaco, oh, honestly on alpine i got no idea, because back then I still had a tv for access
2020-01-26 17:21:47 <_ikke_> You'll probably have to run setup-alpine first
2020-01-26 17:22:19 <quadrathoch2> yeah that's what I also thought, as you are booting into a 'live system'
2020-01-26 17:28:12 <giaco> quadrathoch2: I think I've asked too quickly: https://wiki.alpinelinux.org/wiki/Raspberry_Pi_-_Headless_Installation
2020-01-26 17:28:52 <quadrathoch2> giaco, nice. will bookmark that myself :)
2020-01-26 21:02:14 <giaco> lol, first execution of alpine -> setup-alpine -> select keymap. Wrong mapping, can't type / to select right timezone. Repeat
2020-01-26 21:02:59 <giaco> I can surely stay on us-us
2020-01-26 21:19:51 <giaco> solved, it was a qemu error
2020-01-26 21:23:19 <quadrathoch2> giaco, nice :)
2020-01-26 21:33:06 <phanter> hi. I have alpine on ZFS root. Now I want to look at a file inside a snapshot. So I do this: sudo ls /.zfs/snapshot/zfs-auto-snap_hourly-2020-01-26-0100/
2020-01-26 21:33:21 <phanter> But then I get the following error: ls: /.zfs/snapshot/zfs-auto-snap_hourly-2020-01-26-0100/: Symbolic link loop
2020-01-26 21:33:43 <phanter> How can I use these snapshots and/or access its files?
2020-01-26 22:00:46 <phanter> I followed this guide:https://wiki.alpinelinux.org/wiki/Alpine_Linux_with_root_on_ZFS_with_native_encryption#Creating_the_...(i switched from freebsd to alpine linux yesterday...)
2020-01-26 23:28:10 <atkka> hello
2020-01-26 23:29:29 <atkka> so I installed alpine, lvm on luks with a swap volume
2020-01-26 23:30:26 <atkka> I added the swap to fstab but I have no swap at boot, I can start swap fine with swapon /dev/vg0/swap
2020-01-26 23:30:36 <atkka> any ideas?
2020-01-27 01:36:49 <User743537> Is there a wiki page describing how built packages are secured in the repos?
2020-01-27 02:02:28 <danieli> User743537: elaborate? what do you mean by "secured"?
2020-01-27 02:32:17 <User743537> danieli: how the build servers and package signing work
2020-01-27 02:42:21 <User743537> Also will setup-alpine work with UEFI? It appears to.. but doesn't add anything to my BIOS' menu
2020-01-27 02:45:18 <User743537> It says "installing for x86_64-efi platform"
2020-01-27 05:04:36 <_c705> curious - why is my iptables version marked: iptables v1.8.4 (legacy)? is there something else I should be using?
2020-01-27 07:29:04 <daftaupe> nftables ?
2020-01-27 14:25:47 <gjabell> dante's socksify doesn't function because of no bindresvport, which should be defined in netinet/in.h - is this a musl issue and if so are there workarounds?
2020-01-27 14:31:06 <Cogitri> What? You're trying to build the thing?
2020-01-27 14:31:15 <Cogitri> If so, please provide a log
2020-01-27 14:39:18 <gjabell> it's an issue for me with the current package as well, although during building it does have the following warnings
2020-01-27 14:39:19 <gjabell> https://paste.sr.ht/~gjabell/df32f73c49c85c064552488309d963cba242f3f2
2020-01-27 14:43:05 <gjabell> I can also replicate the same warning with an example program: https://paste.sr.ht/~gjabell/17262f67faf71b255230ed50be0a0b95e623795c
2020-01-27 15:06:28 <Cogitri> I guess libtirpc could provide that
2020-01-27 15:20:23 <gjabell> looks like that at least lets the example compile; let me see if I can get dante working with it too
2020-01-27 15:45:18 <mps> it is not in POSIX afaik
2020-01-27 16:08:01 <gjabell> hmm
2020-01-27 16:10:09 <gjabell> I'm able to suppress the warning by including krb5-dev (dep of libtirpc that includes bindresvport) but the configure script doesn't seem to recognize that it is supported (checking for bindresvport... no)
2020-01-27 16:11:02 <gjabell> passing in LIBS=-lgssrpc as a ./configure arg which works to build the example code from before, not sure if there is something else I need to specify?
2020-01-27 17:33:48 <Cogitri> Maybe you need to do something ala `LDFLAGS="$LDFLAGS -lgssrpc"` (or properly integrate support for it in the configure script)
2020-01-27 17:59:03 <Piraty> is upstreaming patches generally a goal of alpine? for example open-vm-tools carries a lot of patches that made their way to OE as well, but only very few seem to be addressed upstream
2020-01-27 17:59:17 <Piraty> ncopa: (they are your patches, afterall)
2020-01-27 17:59:34 <Piraty> https://layers.openembedded.org/layerindex/recipe/67416/
2020-01-27 18:06:49 <Cogitri> I suppose we should upstream wherever possible, but sadly we have loads of downstream patches, yes :c
2020-01-27 18:07:32 <Cogitri> Upstreaming things should generally be a goal since it lessens the workload on basically everyone, maybe we should make a sprint to upstream everything we can
2020-01-27 18:11:25 <Piraty> good to hear. but good luck with the vmware stuff, i see some musl or otherrelated PR sit for a while...
2020-01-27 18:14:18 <Cogitri> Well, if upstream doesn't cooperate we can't really do something about that, I suppose
2020-01-27 18:20:40 <mps> what Cogitri says, plus some upstream are quite responsive and willing to work with us but some are even don't answer to our question or sent patches
2020-01-27 18:42:38 <phanter> Hi. I did a root on zfs installation (without encryption) and roughly followed this guide: https://wiki.alpinelinux.org/wiki/Alpine_Linux_with_root_on_ZFS_with_native_encryption
2020-01-27 18:43:05 <phanter> It works, but I have issues accessing the snapshots (I installed auto-snapshot)
2020-01-27 18:43:42 <phanter> when trying to view a file from a snapshot, i get an error: "cd: can't cd to zfs-auto-snap_daily-2020-01-26-0100/: Symbolic link loop"
2020-01-27 18:46:31 <Cogitri> Glad the guide has helped you :)
2020-01-27 18:46:47 <Cogitri> Hm, what command did you run to make the snapshots available as files?
2020-01-27 18:46:52 <daftaupe> phanter may I ask why you did leave FreeBSD ?
2020-01-27 18:48:53 <phanter> Performance was not good. It (seemed to) stall often, for example... it sometimes took half a minute to ssh into the machine. I know that my server is not super powerful, but it was becoming rather painful. Also, I am _much_ more comfortable with Linux...
2020-01-27 18:49:27 <phanter> Cogitri: I did not run any command. I understood that I could simply cd into the snapshot folder and access those files...
2020-01-27 18:50:52 <daftaupe> Ok, thanks for explaining phanter
2020-01-27 18:52:12 <Cogitri> You need to do `zfs set snapdir=visible <volume>` in order to be able to access that
2020-01-27 18:52:45 <phanter> <volume> is the pool name? (like rpool or something)?
2020-01-27 18:53:44 <Cogitri> It's the name of the volume you want to make available, not the pool
2020-01-27 18:54:19 <Cogitri> E.g. if you want to make rpool/ROOT/alpine's snapshots visible use rpool/ROOT/alpine as <volume>
2020-01-27 18:54:42 <phanter> oke... you answered my next question right there :-)
2020-01-27 18:55:55 <phanter> I did this: sudo zfs set snapdir=visible rpool/ROOT/alpine
2020-01-27 18:56:14 <phanter> and got no output. The I try to cd into that snapshot folder again, with the same result I posted before...
2020-01-27 18:57:32 <Cogitri> Hm, works for me, I just did `sudo zfs set snapdir=visible expool/ROOT/alpine` and could cd to `/.zfs/snapshot` afterwards
2020-01-27 19:00:00 <phanter> yes, I can cd into that folder, but not into an actual snapshot folder. Could you try that?
2020-01-27 19:05:13 <Cogitri> Ah sure, just a moment
2020-01-27 19:19:42 <phanter> Cogitri: any update on cd'ing into a snapshot?
2020-01-27 19:39:15 <lastebil> I've looked through the various documents and tried a number of things, but I can't seem to get the boot service "consolefont" or "loadkmap" to actually run at boot.
2020-01-27 19:39:51 <lastebil> they do run manually, using rc-service... and any/all errors in the configs have been repaired; I am beginning to think something needs to update something in the initrd?
2020-01-27 19:45:59 <mps> lastebil: rc-update | grep consolefont
2020-01-27 19:49:31 <lastebil> that returns default, though I thought I had removed it from default and put it in boot (which is what is indicated should be used by the configuration file)
2020-01-27 19:50:41 <lastebil> loadkmap does return 'boot' as excpected.  (It does not, however, set the keymap.)
2020-01-27 19:52:28 <mps> consolefont should be in boot
2020-01-27 19:52:55 <lastebil> yes. it is now; I seem to have confused the -u option as something I would need to run every time I did an add
2020-01-27 19:53:04 <mps> I'm not sure for loadkmap because I don't use it
2020-01-27 19:55:35 <lastebil> it seems it should be per documentation I found (rebooting again, so can't say exactly where) - but it may be that it was being overridden by loadkeys (which was in default.)
2020-01-27 19:56:52 <lastebil> consolefont did not load the font, however.  (running it via rc-service, both with and without debug and verbose, DOES change it... so I am very confused as to what could be stopping it from operating)
2020-01-27 19:57:56 <mps> sorry, but did you check if you have set correct fonts and they are installed
2020-01-27 19:58:38 <Cogitri> phanter: So I just tagged a snapshot manually via `zfs snapshot expool/ROOT/alpine@test` and I can cd just fine into /.zfs/snapshot/test
2020-01-27 19:58:56 <lastebil> yes; as per "running it via rc-service"
2020-01-27 19:59:25 <phanter> Cogitri: I will try that now as well, creating a manual snapshot
2020-01-27 20:00:18 <mps> lastebil: hmm, strange, never had issue with consolefont on boot, except when set invalid fonts
2020-01-27 20:00:37 <phanter> sudo zfs snapshot rpool/ROOT/alpine@test
2020-01-27 20:00:45 <phanter> cd /.zfs/snapshot/test/
2020-01-27 20:00:50 <phanter> -ash: cd: can't cd to /.zfs/snapshot/test/: Symbolic link loop
2020-01-27 20:01:00 <phanter> so no success on my side.
2020-01-27 20:01:19 <phanter> zfs --version
2020-01-27 20:01:21 <phanter> zfs-0.8.2-1
2020-01-27 20:01:23 <phanter> zfs-kmod-0.8.2-1
2020-01-27 20:01:32 <phanter> just to be sure we are running the same stuff...
2020-01-27 20:01:43 <lastebil> "rc-service -vD consolefont start" results in "Executing: /lib/rc/sh/openrc-run.sh /lib/rc/sh/openrc-run.sh /etc/init.d/consolefont start" and "Setting console font [spleen/spleen-16x32.psfu.gz]"
2020-01-27 20:02:54 <lastebil> it just seems as if it is not executing this; as if it does not see it as part of the initrd on boot, so ... it won't _run_ on boot.  But updating the initrd does not help, either. So I'm stumped ):
2020-01-27 20:05:48 <lastebil> I've rebuilt the initrd and have now set verbose mode for openrc - that may lead me to what is wrong.
2020-01-27 20:06:15 <lastebil> knowing that yours works from boot is helpful, though, mps (:
2020-01-27 20:06:24 <mps> what is full path of spleen-16x32.psfu.gz
2020-01-27 20:06:43 <mps> you don't need it in initramfs
2020-01-27 20:06:56 <Cogitri> phanter: Huh, well, that's super odd
2020-01-27 20:07:01 <mps> at least before it works
2020-01-27 20:08:18 <phanter> Any hints on what to check?
2020-01-27 20:08:32 <phanter> I am a bit lost...
2020-01-27 20:09:16 <mps> Cogitri: ah, did you read my mind? :) just thought to tell that the new zfs is released
2020-01-27 20:09:16 <lastebil>  /usr/share/consolefonts/spleen   (which is why the spleen/ is in front of the path.)  loadfont also works this way.
2020-01-27 20:10:21 <north1> @mps i told him 2 days ago
2020-01-27 20:10:31 <mps> you have /usr/share/consolefonts/spleen/spleen-16x32.psfu.gz? sorry if sounds stupid question
2020-01-27 20:10:37 <Cogitri> Heh, finally got to that :P
2020-01-27 20:10:46 <lastebil> mps: yes.
2020-01-27 20:10:48 <Cogitri> Finally accelerated acceleration, hooray
2020-01-27 20:11:28 <lastebil> mps: and again, if run manually, i.e. after the boot process, with rc-service, it changes the font.  The service script, when run, actually does what is expected.
2020-01-27 20:11:53 <lastebil> I am now turning on more and more ooptions to log what happens with rc at boot to see if it _is_ running.
2020-01-27 20:12:00 <mps> lastebil: could you disable it from boot (or any runlevel) and try by hand only
2020-01-27 20:12:39 <mps> maybe '/etc/init.d/consolefont start'
2020-01-27 20:12:41 <lastebil> mps: sure. (I also should reboot after removal to test fully, yes?)
2020-01-27 20:13:00 <mps> and add 'set -x' at top of init script
2020-01-27 20:14:17 <mps> btw, do you run it on 'normal' box or some container
2020-01-27 20:14:35 <lastebil> does openrc-run have an -x flag?  (I know shells do, but - )
2020-01-27 20:14:45 <lastebil> (the shebang is #!/sbin/openrc-run )
2020-01-27 20:15:17 <lastebil> this is a normal box, however it _is_ running in dom0.  (installed via the alpine-xen iso.)
2020-01-27 20:15:56 <lastebil> that is a good question, though - it may be that the dom0 startup is slightly different than vanilla
2020-01-27 20:16:34 <mps> hm, not sure if it support 'set -x', forgot that it is openrc and not /bin/sh :)
2020-01-27 20:17:12 <mps> only left option in my head is that 'setfont' read fonts only from hardcoded dirs
2020-01-27 20:17:48 <mps> try some fonts from /usr/share/consolefonts to see does it works at all
2020-01-27 20:18:08 <lastebil> ok. I'll try sun-12x22
2020-01-27 20:18:32 <lastebil> (the issue with other fonts is they are too small to read (:  )
2020-01-27 20:19:09 <mps> I use ter-u22b.psf.gz on hidpi display
2020-01-27 20:20:32 <lastebil> spleen (which is the default openbsd console font) is slightly larger; it actually has one font which is too large for the Linux kernel (:
2020-01-27 20:21:01 <lastebil> (it won't load fonts that are 32 pixels or 64 pixels (:  )
2020-01-27 20:21:16 <mps> ehm, it is not easy to run BSD on linux :)
2020-01-27 20:21:58 <lastebil> no, and I'm not trying to - well, ok, eventually I'll put a openbsd domu on this machine, but that's an entirely DIFFERENT situation (:
2020-01-27 20:22:35 <lastebil> so an interesting thing:  when using sun12x22 - it does work, but throws an error that zcat cannot find the file (:
2020-01-27 20:22:44 <mps> but, iiuc consolefont settings work but not with your fonts?
2020-01-27 20:22:54 <lastebil> no, you misunderstand
2020-01-27 20:23:05 <lastebil> they work with my fonts, and the standard fonts.
2020-01-27 20:23:20 <lastebil> the script that SHOULD run at boot - does not.
2020-01-27 20:23:30 <mps> aha
2020-01-27 20:24:06 <lastebil> now - just to be 100% sure - I am going to put consolefont back in boot runlevel (with the sun12x22 font) and see if it works.
2020-01-27 20:25:19 <mps> is the /usr/share available in boot level
2020-01-27 20:26:25 <lastebil> I don't know, but you know, I can see why it would NOT be
2020-01-27 20:26:36 <lastebil> (which is why I would imagine it would need the font in the initrd)
2020-01-27 20:26:40 <freem_> Hello. How could I find content of a specific package through web interfaces?
2020-01-27 20:26:58 <lastebil> 
ACTION will allow someone to answer freem's question before continuing

2020-01-27 20:27:49 <mps> https://pkgs.alpinelinux.org/contents
2020-01-27 20:29:10 <freem_> thanks, it will give me the leads I need
2020-01-27 20:31:17 <lastebil> mps: so it did not work with sun12x22.  When you set yours up with Terminus, did you have to do anything to the initrd or anything similar?
2020-01-27 20:32:33 <mps> nothing, works 'out of box' on a lot of machines
2020-01-27 20:33:18 <lastebil> hmm. you know, I'm going to try something - I'm going to try it WITHOUT xen (option 0) and ignore all the parts where it can't read /dev/hvc0  (:
2020-01-27 20:34:03 <mps> heh, I think it works only /dev/tty*
2020-01-27 20:34:41 <mps> and it have flag to not run on xenu
2020-01-27 20:35:14 <lastebil> I read xenu as "virutalized domu" not "hypervisor dom0"  - am I incorrect here?
2020-01-27 20:35:17 <lastebil> ...
2020-01-27 20:35:42 <lastebil> you know what is working outside of dom0 ?  Yes, the consolefont script.
2020-01-27 20:35:54 <mps> I mean, it runs on virtual consoles
2020-01-27 20:36:22 <mps> don't have experience with xen
2020-01-27 20:37:08 <lastebil> I have quite a bit - which is why I am thrilled alpine still has an installer that installs dom0 quite nicely, and minimally
2020-01-27 20:38:54 <lastebil> but we have narrowed it down to: this does NOT run at boot when running as dom0, and it works "out of the box" otherwise.
2020-01-27 20:39:33 <lastebil> which means - as normally this will be headless - I can just run a script _if_ I need to use the console, and fire these up.
2020-01-27 20:39:39 <lastebil> I'm fine with that (:
2020-01-27 20:40:46 <mps> maybe you can hack start script to set it
2020-01-27 20:41:45 <mps> it is not complicated
2020-01-27 20:42:05 <lastebil> it may indeed be that xenu flag. I'm going to try removign that and see if it does it.
2020-01-27 20:45:45 <lastebil> yes. that flag is the one.  Apparently, it does not care if it's a dom0 or domu... it skips if either (:
2020-01-27 20:46:06 <lastebil> which makes sense for domu - as that would have no framebuffer, normally.  (at least for a paravirtualized machine.)
2020-01-27 20:46:26 <lastebil> or a hardware device for the console that you could change the font.
2020-01-27 20:46:58 <lastebil> ... and the same would be the case for the keyboard mapping... yes, this all makes 100% sense as to why it would not run in domu.
2020-01-27 20:47:38 <lastebil> and since most of the time these dom0's are headless, I'll bet I'm basically the only person who is wishing to do this (:
2020-01-27 20:48:51 <freem_> you were trying to change console font to terminus for your dom0 system? Can I ask why, or is it a stupid question?
2020-01-27 20:49:22 <lastebil> not a stupid question; (and I'm changing to Spleen, though terminus would also fit the bill.)
2020-01-27 20:50:03 <lastebil> Because I am currently doing some work on the console, without proper bridged networking and so forth, for testing - and my eyes are not very good anymore.
2020-01-27 20:50:39 <lastebil> the machine I am testing with has only hdmi output - and the screens I have that do hdmi are all hi-dpi; so the fonts are tiny, tiny little things.
2020-01-27 20:50:57 <freem_> oh. And there is no way to use some sort of old terminal stuff for that, so that you could just use your favorite x-terminal-emulator with your font?
2020-01-27 20:51:13 <lastebil> well, I _could_ but I don't want to install X; it's not needed (:
2020-01-27 20:51:22 <freem_> I mean, with baudrate and the like
2020-01-27 20:51:39 <freem_> wow, I wasn't even thinking about X11, would be way too overkill
2020-01-27 20:51:41 <lastebil> oh, yes, if I had a serial terminal for this.  Which I _will_ in the data center.  This is at home.
2020-01-27 20:52:11 <lastebil> and this machine does not have proper serial output.  (or ipmi)
2020-01-27 20:52:30 <lastebil> I _will_ want proper serial output for the dom0, though; specifically _for_ ipmi (:
2020-01-27 20:52:31 <freem_> hum... I have never played with xen (not a sysadmin, unfortunately I think sometimes) but would not a unix socket do the same job?
2020-01-27 20:53:06 <lastebil> good question: for kvm you could do that (for the virtual machine, though, not the host.  dom0 = host)
2020-01-27 20:53:41 <freem_> for what I know of Xen, you have the physical machines, that host several virtual ones, dom0 being one of them
2020-01-27 20:53:46 <freem_> don't know much more
2020-01-27 20:53:49 <lastebil> the dom0 is actually a hypervisor layer that the kernel runs within.  It actually can be set up to restrict the dom0 to only using certain parts of the hardware.
2020-01-27 20:54:11 <lastebil> (such as restricting the dom0 to only use so much memory; leaving all the rest for the virtuals.)
2020-01-27 20:54:28 <freem_> interesting, so dom0 is not at all virtualized?
2020-01-27 20:54:51 <mps> latest kernels have options to enable bigger fonts in kernel
2020-01-27 20:55:07 <lastebil> mps: cool, looking forward to that (:
2020-01-27 20:55:30 <lastebil> freem_: it is a full linux kernel, yes.  Not virtual.  Just - well - you can restrict it (:
2020-01-27 20:55:30 <freem_> thinking about font's sizes... those old font formats do support multiple sizes per glyphs
2020-01-27 20:55:37 <mps> for some time I'm using that options on my arm boxes
2020-01-27 20:55:47 <nico[m8> mps: doesn't that exist since Linux 2.4 is the vga= command line or does that not exist anymore?
2020-01-27 20:55:50 <lastebil> mps: I do also want to try arm xen soon.
2020-01-27 20:56:07 <lastebil> nico[m8: it does, due to the framebuffer.
2020-01-27 20:56:16 <mps> well, it exist but don't work good on hidpi
2020-01-27 20:56:20 <freem_> or, you might be able to change /dev/fb0's resolution, somehow
2020-01-27 20:56:36 <mps> but with new addition of terminus it is quite acceptable
2020-01-27 20:57:01 <lastebil> mps: btw - https://www.cambus.net/spleen-monospaced-bitmap-fonts  (also on github)
2020-01-27 20:57:43 <mps> lastebil: will look, but I'm on terminus for years. thanks
2020-01-27 20:58:18 <freem_> I recently had to hack a bit around framebuffer... was not able to find really decent tools to manage it, so I'm not sure there is binaries able to change resolution, and even less if highdpi
2020-01-27 20:58:18 <lastebil> yes, not saying you should change at all (:
2020-01-27 20:58:48 <freem_> but in theory, /usr/include/linux/fb.h permits it
2020-01-27 20:59:28 <freem_> directfb may work, not sure. Didn't much for the uses my job needed
2020-01-27 20:59:30 <lastebil> freem_: in most cases you want to set that prior to the initialization of the screen - for example, on my 4 display machine, with 4 different monitor sizes, and one rotated - only one displays "fully" - the rest are just "partial coverage of the entire screen area"
2020-01-27 20:59:36 <mps> I even started to hack kernel to add terminus by default but someone did it before I finished
2020-01-27 21:00:16 <freem_> lastebil, tbh, I'll be grateful to any link to setup a multi-monitor framebuffer setup
2020-01-27 21:00:40 <lastebil> freem_: will let you know as soon as I find one.  (I wonder if fbterm would actually do that at all...)
2020-01-27 21:00:57 <lastebil> my GUESS is it would not, and it would take some serious work to do it in kernel.
2020-01-27 21:00:59 <freem_> that'd be nice
2020-01-27 21:01:04 <mps> lastebil: this spleen fonts looks similar to terminus, in my eyes
2020-01-27 21:02:06 <lastebil> mps: they do (well, not the thin variety.)
2020-01-27 21:02:09 <freem_> a way to use one TTY per physical screen, would allow me to reduce even more the mem usage of some very old computer of mine
2020-01-27 21:02:59 <lastebil> freem_: not sure that would reduce mem use, except for character mapped displays... and I don't think we have those since the 80's (:
2020-01-27 21:03:42 <lastebil> mps: I did not say thank you; your suggestions got me to where I wanted to be. Kiitos!  (:
2020-01-27 21:05:06 <mps> lastebil: you are welcome :)
2020-01-27 21:06:10 <mps> also. I learned something new about xen
2020-01-27 21:08:26 <lastebil> the wiki may need a few updates about it (as some things have changed) but I'll just do some more fiddling before I suggest any changes
2020-01-27 21:08:58 <lastebil> (the process with the boot cd is easier than the wiki indicates (:  )
2020-01-27 21:30:15 <freem_> lastebil, about TTYs on multiple screen, just found that: https://dafyddcrosby.com/dual-framebuffers/ can't try it now, for several reasons, but from there I can always read source code, so I'll be fine to find any info I might need about multiple screens without graphical server
2020-01-27 21:48:34 <ncopa> Piraty: upstreaming is good and should be done whenever possible
2020-01-27 21:48:48 <ncopa> i dont remember if i have submitted the vmware patches upstream
2020-01-27 21:49:05 <ncopa> i suspect i was not able to create them in an upstreamable format
2020-01-27 23:06:56 <atkka> hello
2020-01-27 23:07:33 <atkka> clear
2020-01-28 00:36:48 <pltrz> can I see the contents of a package that I _haven't_ installed, using apk? I know I could use pkgs.alpinelinux.org/contents
2020-01-28 03:52:36 <codebam> what package do I need for icons to work in nautilus and anything else that uses icons? I have an icon pack installed and selected but none of the icons work
2020-01-28 03:52:53 <codebam> the icon pack works for rofi though
2020-01-28 03:53:01 <codebam> just doesn't work with gtk apps
2020-01-28 05:33:54 <Cogitri> librsvg/gdk-pixbuf possibly?
2020-01-28 05:56:20 <anoduck> Getting an error on boot that alpine fails to mount sysroot on virt kernel, but boots into vanilla kernel just fine.
2020-01-28 05:56:26 <anoduck> Any ideas?
2020-01-28 05:59:39 <Cogitri> Well, what FS is your / ?
2020-01-28 06:02:07 <anoduck> ext4
2020-01-28 06:03:14 <anoduck> I am probing around looking at the configuration files. Because in extlinux.conf both options 1) virt and 2) vanilla have the same value for sysroot.
2020-01-28 06:09:44 <anoduck> I have been running virt for all this time, but I have not figured out why. I am starting to think I installed the wrong flavor to begin with, and vanilla is what I should be on regardless.
2020-01-28 06:10:19 <Cogitri> Probably lts not vanilla
2020-01-28 06:12:11 <anoduck> I have been running edge, and primarily use the box to serve docker containers. Not seeing lts in the downloads page, isn't that ubuntu?
2020-01-28 06:14:05 <anoduck> It appears I just need to move the option to boot virt down on the list, and place vanilla as primary boot. Virt is for running from within a virtualized environment and not for running as a host for virtualized environments. Correct?
2020-01-28 06:23:32 <Cogitri> I mean you should use linux-lts, not linux-vanilla
2020-01-28 06:23:57 <Cogitri> linux-lts is kernel 5.4 now, vanilla stays on 4.19 (and is dead)
2020-01-28 06:24:19 <Cogitri> And yes, linux-virt is for being a guest, not a host
2020-01-28 06:24:26 <maldridge> will there eventually be a wireguard kernel, or will that just be enabled in all?
2020-01-28 06:35:16 <anoduck> @Cogitri Wow, that is nice to know...I am on 4.19.87-0-vanilla. Guess I need to update with apk...
2020-01-28 06:38:37 <anoduck> Alpine is such a smooth and cool customer to work with. Loved it since I first tried it.
2020-01-28 06:40:31 <anoduck> @Cogitri is there anyway of getting rid of the virt flavor from my installation?
2020-01-28 06:44:01 <anoduck> Nevermind, I should just uninstall linux-virt and tada...done. Thanks again.
2020-01-28 07:05:54 <Cogitri> maldridge: I don't see why Wireguard shouldn't be enabled in all kernels and new flavours are a lot of work, so I guess it'll be enabled in all flavours
2020-01-28 07:06:18 <Cogitri> anoduck: Glad it works now
2020-01-28 07:11:49 <maldridge> Cogitri: cool
2020-01-28 07:12:01 <maldridge> I know the void plan is to just switch it on, but I was curious alpine's intentions
2020-01-28 07:18:07 <Cogitri> Well, ncopa has the final say on kernel stuff soooo :P
2020-01-28 07:21:52 <anoduck> Cogitri: Like a dream
2020-01-28 08:03:07 <mps> if we follow current kernel practice to use only LTS kernel then we will wait 'some time' for wireguard in kernel
2020-01-28 08:08:55 <nepugia> On kernel stuff... any change the wacom modules could be packaged? (i.e as a package to install later)
2020-01-28 08:09:11 <nepugia> I think there is already a wacom driver for Xorg in testing
2020-01-28 08:09:18 <nepugia> (as in the userspace component)
2020-01-28 08:15:48 <Cogitri> Is there an issue on that on Gitlab or smth? I don't use Wacom stuff so I don't really know what's missing
2020-01-28 08:15:59 <mps> nepugia: you mean enabled
2020-01-28 08:17:48 <nepugia> mps, not quite, i don't expect the default kernel install to install them, i would like them to be available as a package to install later
2020-01-28 08:18:00 <nepugia> that wouldn't bother smaller installs with those modules :3
2020-01-28 08:19:12 <Cogitri> It'd be a lot of effort to maintain that though
2020-01-28 08:19:48 <nepugia> I suppose that depends on how the build system looks like, and i don't really know, so i am just offering a suggestion
2020-01-28 08:23:08 <Cogitri> Ah yup, but keeping those modules in sync with the kernel version isn't fun
2020-01-28 08:24:47 <nepugia> Suppose so, have some problems with that already if the kernel gets updated since my system needs kernel modules to boot and i have the (then older) kernel in the ESP
2020-01-28 08:25:28 <mps> it would be a lot of work to have every kernel module in separate package
2020-01-28 09:04:10 <Hund> I'm having issues with my Raspberry Pi figuring out how to mount my secondary partition on my sdcard as /home and as persistent storage. Have anyone figured this out?
2020-01-28 11:05:11 <misterunknown> clandmeter: Routinely question: Did you have any issues upgrading your lxc hostsystems from 3.10 to 3.11?
2020-01-28 11:06:11 <clandmeter> misterunknown: hi, im not sure we did that upgrade yet.
2020-01-28 11:10:26 <misterunknown> clandmeter: Ah, ok. The change from lxc 3.1 to 3.2.1 shouldn't be a problem, at least following the release notes. So I guess I'll simply try it.
2020-01-28 11:12:20 <mps> misterunknown: I upgraded few without issues
2020-01-28 11:12:45 <misterunknown> mps: Nice, thanks.
2020-01-28 11:25:57 <misterunknown> clandmeter, mps: Just upgraded, hadn't any issues either.
2020-01-28 11:28:14 <mps> good to hear :)
2020-01-28 11:35:01 <clandmeter> misterunknown: thanks for the feedback, appreciated :)
2020-01-28 14:19:39 <telmich> In some days my replacement notebook (xps 13 again) should arrive - I'll give alpine another try, last time the kernel was too old. Really looking forward to it
2020-01-28 14:20:32 <daftaupe> too old for what telmich ?
2020-01-28 14:23:14 <mps> telmich: 5.4.14 is in testing
2020-01-28 14:28:31 <misterunknown> mps: Is there a simple way to only install the kernel from testing while keep all other packages from (main|community)?
2020-01-28 14:29:20 <_ikke_> misterunknown: pinned repos
2020-01-28 14:30:11 <mps> misterunknown: sorry, i didn't mean testing repo but testing release which is edge, ofc
2020-01-28 14:30:24 <mps> it is in main repo on edge
2020-01-28 14:31:21 <_ikke_> answer would still be the same
2020-01-28 14:33:22 <mps> _ikke_: yes, your explanation is good about installing pkgs from testing repo
2020-01-28 14:34:25 <_ikke_> misterunknown: https://wiki.alpinelinux.org/wiki/Alpine_Linux_package_management#Repository_pinning
2020-01-28 14:40:46 <misterunknown> _ikke_, mps: Thanks, didn't know this is possible.
2020-01-28 14:41:49 <_ikke_> misterunknown: note that usually it's not recommended/supported to install packages from a different release, as there might be dependency conflicts, but this does not matter for the kernel
2020-01-28 14:51:32 <telmich_> daftaupe: for the wifi and for the intel driver; this has been fixed though with 5.4.x already
2020-01-28 14:51:47 <telmich_> mps: ack, that should work
2020-01-28 14:52:14 <telmich_> What I still wonder is why my other notebook (an older dell with touch screen) does not emit any events on the touchscreen, even though libinput sees it
2020-01-28 14:52:52 <telmich_> Has anyone seen that a touch screen appears on /dev/input/xx and that there are no events produced?
2020-01-28 14:52:55 <mps> do you know what driver it needs enabled in kernel
2020-01-28 14:53:20 <telmich_> No, but let me just boot it
2020-01-28 14:53:41 <mps> yes, I had it with uclogic tablet/drawing board
2020-01-28 14:53:51 <telmich_> Ah, nice
2020-01-28 14:53:59 <telmich_> (nice as in: not alone here)
2020-01-28 14:54:45 <mps> it is appeared in log but didn't worked without kernel driver because kernel driver understand it's protocol
2020-01-28 14:59:50 <telmich> Ok, so how do I best trace back the xinput list to a kernel driver..hmm
2020-01-28 15:00:08 <telmich> It's an `WCOM484D:00 056A:484D Touchscreen pad`
2020-01-28 15:04:59 <telmich> I do have a hid_multitouch loaded and the device from xinput is `Device Node (265):	"/dev/input/event3"`and /sys/class/input/event3/device/phys contains i2c-WCOM484D:00. So it is probably one of the i2c drivers
2020-01-28 15:10:28 <mps> can't see any identifier in kernel source, but that doesn't mean it is not there, maybe other identifier
2020-01-28 15:47:17 <newber2020> How do I switch from the default networking options (wpa_supplicant + udhcpc) to Connman? Looks like I can't just install dbus+connman and `rc-update add connman boot` (and dbus of course) since it appears to clash? But when I remove the `networking` service I'm no longer able to bring up my wifi.
2020-01-28 15:52:10 <mps> connman is buggy currently
2020-01-28 15:53:18 <newber2020> Ah I see
2020-01-28 15:54:57 <mps> some works about fixing issue started and probably in next release it will be better
2020-01-28 15:55:35 <newber2020> Thanks for pointing it out :) Looking forward to it
2020-01-28 15:56:51 <mps> also I'm looking for it and watching progress
2020-01-28 16:02:28 <newber2020> Is there a corresponding issue?
2020-01-28 16:16:49 <mps> not that I know for any, in alpine I mean
2020-01-28 17:29:43 <dansolo42> @hund: try doing this:  create the file system and add its entry to /etc/fstab.  lbu add /etc/fstab && lbu commit -d  and then reboot and see what happens
2020-01-28 17:30:29 <simon_> hello! newbie here, looking for help! :-)
2020-01-28 17:31:35 <simon_> have been running into some issues running alpine via docker...
2020-01-28 17:31:56 <simon_> issues are described here: https://forums.docker.com/t/easiest-way-for-container-to-use-host-user-and-how-to-fix-gdb-crashing-container-issue/88017
2020-01-28 17:32:55 <simon_> anybody know why gdb is bringing down docker and alpine ?
2020-01-28 17:33:32 <simon_> and is there a better way to run off-the-shelf alpine with a non-root user that can sudo?
2020-01-28 17:37:04 <Cogitri> Why would you mount /etc/passwd and so on into the docker container?
2020-01-28 17:37:36 <jan78> hey guys
2020-01-28 17:38:03 <jan78> i have a wired problem with alpine on a nuc and a sata connected ssd
2020-01-28 17:38:51 <jan78> after i reboot and the ssd partition mounts (ext4) all the folders are owed by the user chrony
2020-01-28 17:40:30 <jan78> i have to manually chown -R user.user on the mount folder...
2020-01-28 17:40:33 <simon_> @Cogitri just wanted to have a quick way of using my host user inside the alpine docker container, otherwise root user inside container creates files as root user which is annoying for host... is there a better way with alpine and/or docker to get the alpine container to create files with the same user on host?
2020-01-28 18:06:06 <S1m0nHF> Cogitri: Is there a better / preferred way to do it?
2020-01-28 18:07:49 <_ikke_> S1m0nHF: a Dockerfile that creates a user and sets sudo config
2020-01-28 18:09:29 <S1m0nHF> _ikke_: any example of how to do that on the docker command line invoking the container?
2020-01-28 18:10:47 <_ikke_> S1m0nHF: why not build an image with a Dockerfile?
2020-01-28 18:11:02 <_ikke_> That's the way you do these things in Docker
2020-01-28 18:12:07 <_ikke_> The only alternative is to create a complex shell oneliner
2020-01-28 18:12:12 <clandmeter> if you want to match file owner with hosts user, you need to match the same uid/guid
2020-01-28 18:12:30 <S1m0nHF> _ikke_: I'd prefer to just use an off-the-shelf docker image. Why? If I post a bug somewhere with details for reproduction then there's no special image involved and it's just a one-line copy and paste to reproduce... make sense?
2020-01-28 18:12:40 <_ikke_> no
2020-01-28 18:13:00 <_ikke_> You make things more complex this way
2020-01-28 18:13:36 <S1m0nHF> _ikke_: any ideas why gdb is bringing down docker and alpine in my example?
2020-01-28 18:13:37 <_ikke_> A Dockerfile is more readable, and docker is built around the idea of building on top of images
2020-01-28 18:14:40 <S1m0nHF> _ikke_: any examples of such a docker file which dynamically creates the same user as in the host?
2020-01-28 18:16:05 <S1m0nHF> clandmeter: yes, that's right... which is why I 'borrowed' the passwd file etc from the host in my example... but is there a better / preferred why to do this?
2020-01-28 23:17:14 <reidrankin> I'm having to do some dev work on an ubuntu box and I keep getting pissed that "apk-get install" doesn't work
2020-01-28 23:19:41 <mps> reidrankin: apk-get or apt-get
2020-01-28 23:20:53 <reidrankin> i've tried "apk-get install", "apt-get add", "opkg update", and "apt-get fetch coffee" nothing works
2020-01-28 23:22:00 <mps> hmm, understand, sometimes I do similar mistakes
2020-01-28 23:22:43 <mps> also 'apk add' instead of 'apt install'
2020-01-28 23:23:04 <mps> sometimes even 'apk install'
2020-01-29 01:29:50 <bl4ckb0ne> is there a package for `progress` ?
2020-01-29 03:53:39 <atkka> when using the console/tty in alpine the bottom 50px or so of my display isn't updated or drawn to
2020-01-29 03:54:10 <atkka> if I change the font using setfont the problem is fixed
2020-01-29 03:57:59 <atkka> I'm also on a hypbrid laptop with nvidia and intel, I haven't installed any drivers is there a way to tell what gpu im using or intel vs nouveau
2020-01-29 04:08:04 <atkka> fixed it, changing the bitmap font size worked
2020-01-29 04:09:01 <atkka> I'd still like to know how to tell which gpu is being utilized by default if anyone knows
2020-01-29 05:10:19 <S1m0nHF> what's the process for getting an extra perl module built so that apk add can just add it?
2020-01-29 05:10:42 <S1m0nHF> at the moment I use cpanm to build it and it's sooooo much slower than apk ad :-)
2020-01-29 05:43:44 <_ikke_> S1m0nHF: there are plenty of examples in aports, maybe look at one of them and use that as a basis
2020-01-29 05:44:16 <_ikke_> S1m0nHF: if your not comfortable with that, you could make a package request in gitlab
2020-01-29 05:44:57 <S1m0nHF> _ikke_: thanks! do you have a gitlab link for that?
2020-01-29 05:45:18 <_ikke_> gitlab.a.o
2020-01-29 05:45:35 <_ikke_> https://gitlab.alpinelinux.org/alpine/aports/issues
2020-01-29 05:46:17 <S1m0nHF> _ikke_: thanks!
2020-01-29 05:47:35 <S1m0nHF> looks like it doesn't show up in aports... so I'll try to make the package request...
2020-01-29 06:00:35 <S1m0nHF> request is here: https://gitlab.alpinelinux.org/alpine/aports/issues/11168
2020-01-29 16:38:19 <mtnt> Hi there! Im noob, dont hit too hard. Installed Alpine, X11, dwm. Mice wont move. It is Ligitech G502
2020-01-29 16:53:51 <mtnt> okay, gtg. thanks
2020-01-29 18:34:57 <telmich> Is there any notion in Alpine to allow non-root users to modify/create/delete network interfaces as non-root users?
2020-01-29 18:35:36 <telmich> I'm asking for the development of uncloud, which most of the things can manage as a regular user, but creating bridges and vxlan interfaces so far looks like we've to reach out via sudo
2020-01-29 19:47:03 <detha> telmich: give the users CAP_NET_ADMIN?
2020-01-29 21:30:20 <maldridge> has anyone tried to build osquery on alpine?
2020-01-29 21:34:59 <mps> maldridge: isn't it big for distro which tend to be small
2020-01-29 21:48:12 <maldridge> its security software for people who believe that sort of thing helps
2020-01-29 21:53:48 <mps> 'big' software and security doesn't fit together usually
2020-01-29 21:54:15 <mps> except for managers
2020-01-29 21:55:18 <bl4ckb0ne> yet wpa_supplican is using wlan0 correctly
2020-01-29 21:56:34 <mps> yes, it works fine on my RPi zero
2020-01-29 22:02:24 <S1m0nHF> hello!
2020-01-29 22:02:38 <S1m0nHF> vfprintf (f=0x18, f@entry=0x7ffff66ea688, fmt=0x7ffff6ad2ac0 "...", ap=<optimized out>) at src/stdio/vfprintf.c:686
2020-01-29 22:03:12 <S1m0nHF> I'm getting a seg fault in vfprintf() ... is there an apk add to grab the source code for debugging?
2020-01-29 22:03:41 <_ikke_> musl-dbg I think
2020-01-29 22:05:19 <dalias> if f=0x18 then an invalid FILE* was passed
2020-01-29 22:05:32 <dalias> hmm, f@entry differs tho
2020-01-29 22:05:47 <dalias> so perhaps you have a stack-based buffer overflow
2020-01-29 22:06:54 <dalias> yeah the crash happens right after the buffer flush
2020-01-29 22:06:58 <S1m0nHF> #0  vfprintf (f=0x18, f@entry=0x7ffff66ea688, fmt=0x7ffff6ad2ac0 "... coroutine voodoo ...", ap=<optimized out>) at src/stdio/vfprintf.c:686
2020-01-29 22:07:04 <dalias> so it's pretty likely that the buffer flush clobbered f
2020-01-29 22:07:09 <S1m0nHF> #1  0x00007ffff6ac7039 in vsnprintf (s=<optimized out>, n=<optimized out>, fmt=<optimized out>, ap=<optimized out>) at src/stdio/vsnprintf.c:54
2020-01-29 22:07:10 <dalias> what's the rest of the backtrace?
2020-01-29 22:07:38 <S1m0nHF> interestingly I'm calling vsnprintf() so I don't expect any FILE* to be used...
2020-01-29 22:08:01 <dalias> the FILE* exists on vsnprintf's stack and is a string-backed one
2020-01-29 22:08:18 <S1m0nHF> kk
2020-01-29 22:08:29 <dalias> where does the string it's writing to live?
2020-01-29 22:08:35 <dalias> and did you pass the correct buffer size?
2020-01-29 22:09:11 <S1m0nHF> hehe I thought I did and it all 'works' on glibc... I'll scrutinize all that stuff again...
2020-01-29 22:09:52 <dalias> 0x18 is an odd value for it to have been clobbered with
2020-01-29 22:10:02 <dalias> not text-like
2020-01-29 22:10:40 <S1m0nHF> the string it's writing to is a global char array... so not on the stack and not dynamically allocated etc
2020-01-29 22:12:15 <S1m0nHF> it's the type of bug where if you comment out some tiny bit of code then it doesn't seg fault... so I'm suspecting there is some type of corruption going on...
2020-01-29 22:12:38 <dalias> is the code public?
2020-01-29 22:12:49 <S1m0nHF> unfortunately not... :-(
2020-01-29 22:14:18 <S1m0nHF> is there some kind of extra debug mode or extra assert layer or something for *printf() functions in musl?
2020-01-29 22:14:59 <S1m0nHF> my debug strategy is to try and slowly simplify the code until the trouble-maker turns up...
2020-01-29 22:16:19 <dalias> run under valgrind?
2020-01-29 22:16:55 <fyr> does alpine not have pandoc packaged?
2020-01-29 22:17:07 <_ikke_> fyr: I think we had in the past but it had issues
2020-01-29 22:17:20 <fyr> yikes,
2020-01-29 22:17:42 <_ikke_> hmm, no, apparently not even
2020-01-29 22:17:43 <fyr> other markdown converters are always so bad
2020-01-29 22:18:06 <fyr> and rely on npm
2020-01-29 22:20:14 <S1m0nHF> _ikke_: I installed musl-dbg but (gdb) list src/stdio/vfprintf.c:686 does not show any source code... any ideas?
2020-01-29 22:21:08 <dalias> you could just extract the musl source and run the debugger from the dir containing the source
2020-01-29 22:21:34 <dalias> upstream 1.1.24 unpatched should be close enough; this file is definitely not patched by alpine
2020-01-29 22:21:35 <_ikke_> S1m0nHF: that package only has the symbols, not the source
2020-01-29 22:23:07 <S1m0nHF> dalias: what would be the command to get that source code?
2020-01-29 22:27:31 <mps> hmm, after 3 days with linux-lts 5.4.15 again got OOM
2020-01-29 22:28:31 <mps> lets see linux-lts 5.4.16, will it survive more than 3 days
2020-01-29 22:34:00 <bl4ckb0ne> S1m0nHF: download the source from the same version and add them to the gdb path
2020-01-29 22:36:59 <codebam> where is the ftp or telnet commands in our package repos?
2020-01-29 22:37:05 <codebam> s/is/are
2020-01-29 22:37:17 <S1m0nHF> bl4ckb0ne: thanks! how do I match the versions and where do I download the source code from?
2020-01-29 22:37:22 <codebam> I can't find it searching for `cmd:ftp` or `cmd:telnet`
2020-01-29 22:37:55 <mps> codebam: busybox-extra
2020-01-29 22:38:04 <codebam> thank you!
2020-01-29 22:38:15 <mps> for ftp I would use lftp
2020-01-29 22:38:36 <bl4ckb0ne> read the apkbuild https://gitlab.alpinelinux.org/alpine/aports/blob/master/main/musl/APKBUILD
2020-01-29 22:38:36 <codebam> lftp isn't the same. doesn't show verbose output properly
2020-01-29 22:39:19 <mps> I'm not sure busybox has ftp applet
2020-01-29 22:40:57 <codebam> yeah it's not in extras
2020-01-29 22:40:57 <gu0tvuNHFV8o> telnet can be had via netcat
2020-01-29 22:41:05 <codebam> ohh true I forgot about that gu0tvuNHFV8o
2020-01-29 22:41:08 <codebam> thanks
2020-01-29 22:41:17 <gu0tvuNHFV8o> ftp can be done via netcat too :)
2020-01-29 22:41:49 <gu0tvuNHFV8o> you could also try socat
2020-01-29 22:42:23 <gu0tvuNHFV8o> curl and wget also do ftp btw
2020-01-29 22:43:18 <mps> and firefox also can do ftp ;)
2020-01-29 22:43:24 <gu0tvuNHFV8o> what do you mean with > lftp isn't the same. doesn't show verbose output properly
2020-01-29 22:43:26 <codebam> yeah that's why I was using telnet. I wanted to see the raw output
2020-01-29 22:43:29 <codebam> but I can just use netcat
2020-01-29 22:56:28 <dalias> netcat != telnet
2020-01-29 22:56:46 <dalias> lftp has very different command interface than ftp
2020-01-29 23:03:34 <telmich> <detha "nico: give the users CAP_NET_ADM"> I'll try that, thanks for the hint
2020-01-29 23:04:50 <gu0tvuNHFV8o> sure firefox also has very different command interface than ftp.
2020-01-29 23:45:13 <vivi> Good day, any Alpine devs available for hire for a brief consultancy - if so do PM me rate / availability.
2020-01-30 00:40:41 <daemon> hey all is it possible to download the raw packages and install them locally via docker, for instance gcc-8.3.0-r0
2020-01-30 00:41:38 <kaniini> wow
2020-01-30 00:42:29 <kaniini> vivi: what do you need precisely
2020-01-30 00:43:05 <kaniini> daemon: yes it is possible.  apk add filename.apk
2020-01-30 00:43:49 <daemon> kaniini, cool any quick commands to grab a package and all its deps and pull them in as the raw apk's rather than manually tracing there locations
2020-01-30 00:43:59 <kaniini> apk fetch -R
2020-01-30 00:44:07 <daemon> perfect, thank you
2020-01-30 00:49:11 <vivi> kaniini, I have a propreitary binary (go) that needs to ship with a prebuilt-alpine (I'm using apk fetch -r + abuild etc)
2020-01-30 00:49:51 <vivi> Only problem is that the /etc/apk/world has all the deps, /media/mmcblk0p1 has all the necessary files / directories (cache, apks/armhf ...)
2020-01-30 00:50:25 <vivi> Issue is that the following packages (despite following proper documentation) do not ship with the pre-built OS:
2020-01-30 00:51:03 <vivi> udev, udev-init-scripts, dbus, bluez, udev-init-scripts-openrc, hwids-net, hwids-pci, hwids-udev, hwids-usb
2020-01-30 00:51:47 <vivi> Even though the apkvol is accurate and the apkindex is generated and signed against keys in /etc/apk/keys
2020-01-30 00:52:18 <vivi> libusb, libpcap -- all the things I need work just fine in /etc/apk/world -- except for that list I provided above.
2020-01-30 00:55:25 <vivi> Once again - happy to pay - kaniini I believe I may have already shared this with you but now there's money on the table :P    https://git.riboflav.in/snippets/7
2020-01-30 01:07:10 <bl4ckb0ne> i have a wpa_supplicant launched by supervise-daemon, where can i disable it?
2020-01-30 01:07:20 <bl4ckb0ne> i think it is stepping on my networkmanager ocnfig
2020-01-30 01:07:57 <north1> with openrc
2020-01-30 01:08:55 <bl4ckb0ne> i also have wpa_supplicant -u launched by it
2020-01-30 01:09:00 <bl4ckb0ne> will it disable both?
2020-01-30 01:12:26 <reidrankin> vivi: to clarify here, the gobuild and apkbuild stages are working?
2020-01-30 01:12:59 <kaniini> ironically I am working on a software tool that builds Alpine appliances
2020-01-30 01:13:01 <kaniini> lol
2020-01-30 01:13:37 <reidrankin> ...you don't say (guess what I'm doing)
2020-01-30 01:14:27 <reidrankin> (though I'm exclusively targeting the run-from-ram style)
2020-01-30 01:14:32 <vivi> reidrankin, indeed -- the apk stuff is A-OK -- it's the apk fetch -R and the index signing against world that turds itself
2020-01-30 01:14:50 <S1m0nHF> bl4ckb0ne: cat /etc/os-release 2>&1 | egrep PRETTY shows PRETTY_NAME="Alpine Linux v3.11"
2020-01-30 01:15:09 <kaniini> reidrankin: that is an objective of mine to support for boards that can do it
2020-01-30 01:15:12 <vivi> Basically go / bin / packaging has zero to do with this issue.  It's when the pi boots it says "package unavailable ... even though in repo"
2020-01-30 01:16:03 <vivi> dbus used to me so much fun, now it's just an annoyance ( I have a dependency on bluez and there's no way I can find in code to do what I need so I default to bluez )
2020-01-30 01:16:37 <kaniini> I can look into your issue this weekend I guess
2020-01-30 01:16:43 <reidrankin> vivi: am I missing something? I don't see a fetch command in that pastebin?
2020-01-30 01:16:56 <vivi> Allow me to update with latest -
2020-01-30 01:19:37 <vivi> https://e3db7f4a.ngrok.io/
2020-01-30 01:19:57 <vivi> just ran docker nginx $(pwd) + ngrok so I didn't have to make this internet fame
2020-01-30 01:20:15 <vivi> (This runs on a raspberry pi3B+ with gitlab-runner (shell))
2020-01-30 01:24:01 <vivi> reidrankin, I must be missing something very peculliar here - I followed the docs as best they made sense several times over the past two weeks - still can't ship an image that boots with dbus and udev
2020-01-30 01:24:07 <reidrankin> is the fetch step actually getting the apks? or is it a runtime issue?
2020-01-30 01:24:15 <vivi> It's runtime.
2020-01-30 01:25:06 <vivi> Basically when the pi boots it loads about 50% of the apacks from /media/mmcblk0 (on the pi) and takes a dump at the packages listed above - despite all others in that list working perfectly.
2020-01-30 01:25:42 <vivi> Could it perhaps be that the build machine fetches packages from community and the apkindex isn't signed against all the different sources? dunno this is getting ... tedious
2020-01-30 01:27:51 <reidrankin> well, you're in luck; I've been doing some work on apk over the last few days and have had to figure out a lot of where it diverges from the documentation
2020-01-30 01:29:17 <vivi> reidrankin, you're my saviour man... my own personal Jesus Christ.
2020-01-30 01:29:35 <reidrankin> is there anything in /etc/apk/repositories?
2020-01-30 01:32:17 <vivi> https://e3db7f4a.ngrok.io/repositories.html
2020-01-30 01:32:56 <vivi> https://e3db7f4a.ngrok.io/world.html
2020-01-30 01:33:56 <vivi> Don't ask why html - it could be the sandwich in front of me begging to be eaten.
2020-01-30 01:36:18 <reidrankin> unless I missed it, it doesn't look like there's a symlink at /etc/apk/cache to /media/mmcblk0/apks/custom?
2020-01-30 01:42:01 <vivi> It's a step outside of the build.sh - but is /media/mmcblk0/cache absolutely necessary?
2020-01-30 01:42:45 <vivi> I mean if *every* other package does the what it's supposed to do - that leads me to believe that it has something to do with PKI
2020-01-30 01:43:47 <vivi> I'm just really grasping at straws now though :(
2020-01-30 01:44:16 <reidrankin> apk can either load from cache or a repo, and how it looks for the APKINDEX.tar.gz is different for those
2020-01-30 01:44:43 <reidrankin> can I take a look at the index file?
2020-01-30 01:50:51 <reidrankin> vivi: try apk verify on your index to make sure it got signed properly, and try it on one of the behaving and one of the misbehaving APKs to see if it's PKI-related
2020-01-30 01:51:53 <vivi> That's a perfect step to add to the ci/cd pipeline - thanks for the suggestion reidrankin!
2020-01-30 01:52:41 <vivi> I'm in a time crunch thus the poor quality code - I have OS+API+ETL+DB to deal with before the end of February (AWS credits!!)
2020-01-30 01:53:32 <reidrankin> yeah... I know the feeling
2020-01-30 01:55:23 <vivi> We scan nearby radio signals to detect potential advertisement impressions on moving properties
2020-01-30 01:55:36 <vivi> Previous version was buildroot and it did - until the sd cards popped
2020-01-30 01:57:09 <reidrankin> I hate that. Had it happen to me in a server with an HPe OEM microSD whose $200 pricetag was supposed to be justified by it being reliable.
2020-01-30 01:57:51 <vivi> We have 100's of devices in our fleet across North America - when an SD card goes, we must ship IoT devices across borders which is :money_with_wings:
2020-01-30 01:58:17 <reidrankin> It started failing silently. I only found out after comparing hashes of the image I'd blown on and finding out it was different every time.
2020-01-30 01:58:33 <vivi> It's rotten, Jim.
2020-01-30 02:00:12 <reidrankin> vivi: what is the precise error message you're getting on boot? And are you getting an "initramfs emergency shell" prompt?
2020-01-30 02:00:45 <vivi> reidrankin, one moment I took a photograph.
2020-01-30 02:02:11 <vivi> https://imgur.com/a/CgvDvRW
2020-01-30 02:02:54 <vivi> This is somewhat old, - but that's what appears now with the aformentioned apacks (udev, dbus...)
2020-01-30 02:03:39 <vivi> ^^ Also note that's before the 'raspberries' appear - meaning likely in the initrd
2020-01-30 02:03:55 <vivi> s/likely/absolutely
2020-01-30 02:04:20 <daemon> /
2020-01-30 02:04:25 <daemon> close your regexs
2020-01-30 02:05:06 <reidrankin> s%s/likely/absolutely%s/likely/absolutely/%
2020-01-30 02:05:07 <vivi> s/likely/absolutely/g
2020-01-30 02:05:35 <daemon> valid both of you well done :)
2020-01-30 02:05:48 <reidrankin> huh I guess the bot doesn't do global replaces
2020-01-30 02:06:01 <reidrankin> foo bar foo baz
2020-01-30 02:06:06 <daemon> reidrankin, very little can understand or handle the power of perlre
2020-01-30 02:06:08 <reidrankin> s/foo/bash/
2020-01-30 02:06:33 <reidrankin> ... where's my autoreplacment bot?
2020-01-30 02:06:43 <vivi> Alright, time for rest. I'll try your suggestion in a few hrs reidrankin - appreciate your help.
2020-01-30 02:09:56 <reidrankin> vivi: It won't help, it's not a PKI issue; that error only comes from one place in apk, and it's caused by an inability to open a package
2020-01-30 02:39:27 <reidrankin> vivi: I'm pretty sure all the failing packages have the "noarch" architecture and apk is looking for them in /media/mmcblk0/apks/custom/noarch; you can use "--rewrite-arch=armhf" with the "apk index" command to force the index to list them all in the armhf architecture, which will make apk search "/media/mmcblk0/apks/custom/armhf" where they actually are
2020-01-30 02:41:37 <reidrankin> vivi: that said, I was actually somewhat surprised by the "make your own repo" approach: when I've wanted to do similar things before, I've just made sure the apkovl contains a symlink /etc/apk/cache to the directory of choice (/media/sda2/apk-cache in my install) and stuck all the packages and index there
2020-01-30 02:42:26 <reidrankin> vivi: there are different codepaths in apk for cache and repo lookup, so you don't have to worry about the architecture folders like that
2020-01-30 02:43:13 <reidrankin> vivi: I gotta sleep too, if you need more help I'll be on tomorrow - or drop me a line (reidrankin@gmail.com) and I'll see what I can do
2020-01-30 02:51:57 <reidrankin> vivi: oh yeah, it might be a good idea to make sure apk fetch is fetching armhf packages (I'm betting your AWS instance is x86_64)
2020-01-30 02:56:36 <reidrankin> if you don't use the --rewrite-arch flag when creating the index, `tar xzOf APKINDEX.tar.gz APKINDEX | grep '^A:' | cut -d : -f 2 | sort -u` will list the architectures of the packages you're working with
2020-01-30 03:01:25 <reidrankin> also I'm *pretty* sure you don't need the index at all if you use the cache directory
2020-01-30 03:20:25 <vivi> reidrankin, Incredible I think you may have solved it. I just pushed a build that looks solid - I'll follow up after testing tomorrow.
2020-01-30 03:28:18 <vivi> As a side note reidrankin I'm actually building on raspberry pi's connected to the physical hardware we ship so I can execute end-to-end tests.
2020-01-30 03:29:41 <vivi> libpcap is a bugger to build against musl & armhf for go
2020-01-30 03:30:04 <vivi> I should have just chose rust but .. again timecrunch.
2020-01-30 03:44:51 <vivi> Holy GUAGAMOLE reidrankin you nailed it! Perfect solution BEST!
2020-01-30 03:45:23 <russkel> well this is a bit dumb https://pythonspeed.com/articles/alpine-docker-python/
2020-01-30 03:45:33 <russkel> who wants to tell him py3-matplotlib exists :D
2020-01-30 03:45:49 <russkel> "Alpine makes Python Docker builds 50× slower, and images 2× larger"
2020-01-30 03:49:25 <vivi> On occasion, you find misinformed editorials online :(
2020-01-30 03:49:35 <vivi> #4
2020-01-30 03:50:41 <north1> @russkel to be fair, it is in testing, which means it isn't in any stable release, which is what people use.
2020-01-30 03:50:53 <north1> same for py3-pandas
2020-01-30 03:51:31 <north1> The solution is the maintainer or someone (who  might take maintainership) moving it to community/ or even main/
2020-01-30 03:51:42 <russkel> testing doesn't necessarily mean it is a testing version does it
2020-01-30 03:51:54 <russkel> there's tons of stable release versions in testing
2020-01-30 03:52:03 <north1> no
2020-01-30 03:52:41 <russkel> stable as per the developer releases, not this "LTS" idea of stable
2020-01-30 03:52:52 <north1> it means the maintainer thinks it has some quirks that need to be ironed or they are not ready to maintain a stable version for 6 months (community/) or 2  years (main/)
2020-01-30 03:53:17 <north1> I have some stuff in testing like telegram-desktop because it is not ready for being maintained in the long term
2020-01-30 03:54:05 <russkel> it's also the recommended/expected starting place for aports, from what I have been told
2020-01-30 03:54:23 <north1> yes, almost all aports start there
2020-01-30 03:54:29 <north1> some are immediatelly moved to community/ though
2020-01-30 03:55:33 <russkel> I know I have a few aports in testing which are stable versions hehe, but i haven't bothered to get them moved to community yet
2020-01-30 03:56:00 <codebam> how hard would it be to modify the APKBUILD for linux-lts to instead build from master?
2020-01-30 03:56:25 <russkel> codebam, as in linux master branch?
2020-01-30 03:57:10 <codebam> yeah
2020-01-30 03:57:34 <russkel> probably quite straight forward, but wether or not it compiles is another thing
2020-01-30 03:57:53 <codebam> oh okay
2020-01-30 03:58:14 <russkel> check out the APKBUILD for linux-lts and see for yourself
2020-01-30 03:58:35 <russkel> why would you use lts rather than the most recent package?
2020-01-30 03:58:51 <codebam> is lts not the stable package?
2020-01-30 03:59:13 <russkel> lts means long term support
2020-01-30 03:59:28 <codebam> yes but there's no regular package for linux
2020-01-30 03:59:31 <codebam> only linux-lts
2020-01-30 03:59:49 <russkel> uhh
2020-01-30 04:00:04 <codebam> if you don't believe me, check for yourself
2020-01-30 04:00:42 <russkel> ah I didn't know that, oops
2020-01-30 04:01:38 <codebam> how would I download master as a tarball?
2020-01-30 04:01:44 <codebam> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/
2020-01-30 04:02:00 <codebam> or does APKBUILD allow me to specify git repos in `source=`?
2020-01-30 04:02:38 <russkel> github mirrors those repos I am pretty sure, so you could download from there
2020-01-30 04:03:03 <codebam> oh right, forgot about that. thanks!
2020-01-30 04:03:09 <russkel> sorry mate need to run, good luck
2020-01-30 04:04:07 <codebam> thanks for the help! :)
2020-01-30 05:57:37 <bleb> im reformatting an external hard drive
2020-01-30 05:58:59 <bleb> busybox fdisk lets me create a dos partition table but it says "the number of cylinders is larger than 1024 and could cause some problems with ... booting and partitioning software from other os's"
2020-01-30 05:59:19 <bleb> is it a bad idea to use a dos partition table for such a large disk?
2020-01-30 05:59:28 <bleb> should i use some other os which can make an efi table?
2020-01-30 05:59:42 <bleb> gpt i mean
2020-01-30 06:05:09 <codebam> bleb probably best to use gpt if it's warning you
2020-01-30 06:05:21 <codebam> although I'm not sure if booting would be an issue or not
2020-01-30 06:26:08 <hfingler> probably not the correct channel, but I don't know what to do anymore. I'm trying to compile aws-cpp-sdk against musl statically, but the end .a still has the __dso_handle thing from glibc, which messes up my final linking, does anyone know what could this be?
2020-01-30 06:43:02 <Cogitri> You're compiling this on an Alpine Linux machine, hfingler?
2020-01-30 06:43:30 <hfingler> alpine docker container
2020-01-30 06:45:20 <maldridge> mps: I agree with your stance, unfortunately some of the folks I work with have less self-consistent views
2020-01-30 06:51:58 <hfingler> forgot to tag you @Cogitiri, its an alpine docker container
2020-01-30 06:52:23 <hfingler> whoops, @Cogitri
2020-01-30 06:56:15 <Cogitri> Huh
2020-01-30 06:57:24 <Cogitri> I don't see how it'd end up with that since glibc shouldn't be available at all in that Docker containwr
2020-01-30 07:00:30 <Cogitri> Can you show us your Dockerfile ?
2020-01-30 07:13:12 <Cogitri> (Can't help much due to no WiFi in train tho)
2020-01-30 07:20:11 <hfingler> @Cogitri I didn't do a dockerfile, just spun up an alpine container and opened an it bash into it. Then i installed the required packages to compile it
2020-01-30 07:24:15 <Cogitri> Ah, that makes it pretty hard to replicate this for us then
2020-01-30 07:24:48 <Cogitri> Would be nice if you could make a Dockerfile
2020-01-30 07:24:52 <hfingler> yeah, I'll try writing a docker file tomorrow
2020-01-30 07:34:48 <Cogitri> Okie
2020-01-30 07:34:51 <Cogitri> Feel free to get back at us then :)
2020-01-30 07:40:51 <hfingler> will do, thanks!
2020-01-30 08:33:09 <misterunknown> Hi. I have a problem with mariadb after upgrading from alpine 3.10 to 3.11. Mysqldump says "mysqldump: Couldn't execute 'SHOW FUNCTION STATUS WHERE Db = 'cacti'': Cannot load from mysql.proc. The table is probably corrupted (1728)"
2020-01-30 08:34:17 <misterunknown> So I tried to run mysql_upgrade, but this fails with "Error: Server version (10.4.10-MariaDB-log) does not match with the version of the server (10.4.12-MariaDB) with which this program was built/distributed. You can use --skip-version-check to skip this check."
2020-01-30 08:35:32 <misterunknown> But I don't find any package with version "10.4.10". All related packages have version "10.4.12-r0".
2020-01-30 08:37:27 <clandmeter> misterunknown: hi
2020-01-30 08:38:19 <clandmeter> i pushed an update yesterday from 10 -> 12
2020-01-30 08:38:56 <clandmeter> is your mariadb client and server the same version?
2020-01-30 08:44:18 <misterunknown> clandmeter: Yes, at least the package versions are correct: https://termbin.com/gvyr
2020-01-30 08:45:50 <misterunknown> clandmeter: Ok, nevermind, obviously the upgrade didn't restart the mariadb server.
2020-01-30 08:47:43 <misterunknown> Now everything works. I upgraded to 3.11 two days ago, and restarted the whole container, so I was "sure" that it couldn't be the running instance. But I configured unattended updates via ansible. I guess the new versions got installed without restarting the mariadb server.
2020-01-30 11:10:07 <mps> maldridge: eh, understand. we all have to earn money for living and make because of that to make 'compromises' with managers and marketers
2020-01-30 11:12:34 <mps> and, I'm not against packaging osquery (or any 'big' software) because I know some people need these for different reasons and I'm not one to judge
2020-01-30 11:18:59 <bjorn`> mirror sync to ftp.acc.umu.se seems currently broken, who do I notify?
2020-01-30 11:23:04 <mps> bjorn`: it is already known by infra team, they works to solve it
2020-01-30 11:23:36 <bjorn`> cool
2020-01-30 12:43:45 <misterunknown> Does anyone use the webdav cli client "cadaver"?
2020-01-30 14:04:36 <daemon> hey all how do I install a .apk package from a local file without reaching out to the internet what so ever
2020-01-30 14:05:08 <daemon> I have all my apk files mounted
2020-01-30 14:05:08 <daemon> binutils-2.33.1-r0.apk  gmp-6.1.2-r1.apk
2020-01-30 14:05:09 <daemon> etc ...
2020-01-30 14:05:19 <vivi> add path to /etc/apk/repositories
2020-01-30 14:05:21 <daemon> but when I do apk add ./ ... it tries to fetch some index
2020-01-30 14:05:23 <daemon> and then fails
2020-01-30 14:05:32 <daemon> cool cheers vivi
2020-01-30 14:06:07 <vivi> If there's no APKINDEX you'll have to create one, lmk if you run into trouble.
2020-01-30 14:06:27 <daemon> yeah ran straight into it
2020-01-30 14:06:29 <daemon> no index
2020-01-30 14:06:55 <daemon> https://dpaste.org/2kFN
2020-01-30 14:07:01 <daemon> are the only apk's I have and need
2020-01-30 14:07:13 <daemon> notable gcc and its deps
2020-01-30 14:07:23 <vivi> You doing apk fetch?
2020-01-30 14:07:31 <daemon> to get those originally I did apk fetch -R
2020-01-30 14:07:40 <daemon> but where this container is going cannot make outbound connections
2020-01-30 14:07:51 <daemon> so I can ship the files but not the index
2020-01-30 14:07:56 <daemon> unless I downloaded the index file I suppose
2020-01-30 14:07:56 <vivi> abuild-keygen -a -i
2020-01-30 14:08:08 <vivi> You can sign the index with your own key if you include it in /etc/apk/keys
2020-01-30 14:08:20 <daemon> sh: abuild-keygen: not found
2020-01-30 14:08:27 <vivi> You'll need apk add abuild
2020-01-30 14:08:42 <daemon> maybe copying the index file would be wiser
2020-01-30 14:09:39 <vivi> That's a solution as well - if you're not shipping any custom apacks use the index from the mirror you pulled the packages from.
2020-01-30 14:10:11 <vivi> I just went through all this (but with custom apacks) so I was forced to include all the PKI stuff.
2020-01-30 14:10:15 <daemon> yep all this is, is kind of a base build system for another image, so no custom apk's everything else is source built
2020-01-30 14:11:50 <vivi> GRRR I reverted code because of this... http://mirror.csclub.uwaterloo.ca/alpine/v3.11/community/armhf/APKINDEX.tar.gz
2020-01-30 14:13:20 <daemon> it complained
2020-01-30 14:13:22 <daemon> but did it
2020-01-30 14:13:22 <daemon> https://dpaste.org/dz08
2020-01-30 14:14:35 <daemon> ah interesting
2020-01-30 14:14:40 <daemon> even without an index I can do it
2020-01-30 14:14:44 <daemon> as long as I install them in the correct order
2020-01-30 14:14:49 <daemon> like you would build them
2020-01-30 14:18:22 <tboerger[m]> https://mirrors.alpinelinux.org/#mirror28 mirror status doesn't look good for the choosen one
2020-01-30 14:18:43 <vivi> That's my default - too bad.
2020-01-30 14:19:51 <daemon> I doubt this will be helpful for anyone else, but if you want a mini alpine as a build engine, https://dpaste.org/tkBH
2020-01-30 14:19:58 <daemon> is the order you have to install the apk's (without index)
2020-01-30 14:20:59 <daemon> now just to remove libssl apk and all the other things in base
2020-01-30 14:21:34 <daemon> if it will let me
2020-01-30 14:22:00 <daemon> Ok I take it I am not allowed: https://dpaste.org/uFLD
2020-01-30 14:22:05 <daemon> or is there some trick to this bit
2020-01-30 14:22:58 <daemon> I don't suppose it really matters anyhow
2020-01-30 14:25:45 <vivi> It's likely those are OS level dependencies as apk supports TLS endpoints thus libssl/libressl whatever ssl is required.
2020-01-30 14:25:49 <vivi> ^ speculation.
2020-01-30 14:26:24 <daemon> more than likely
2020-01-30 14:26:52 <vivi> What's the size constraint? Or you just want as thin an image as possible?
2020-01-30 14:27:11 <daemon> no size constraint, its more a removing potential problems thing
2020-01-30 14:27:52 <reidrankin_> daemon, vivi: just put the apks in the cache
2020-01-30 14:28:01 <reidrankin_> no need to index or sign the index
2020-01-30 14:28:21 <daemon> oh liek a package cache? where abouts is that locale
2020-01-30 14:28:21 <daemon> located*
2020-01-30 14:28:25 <daemon>  /var/db/pkg (pure guess)
2020-01-30 14:28:40 <reidrankin_> if you're using an apkovl, just make sure it contains a symlink at /etc/apk/cache to /media/whatever/somecachedir
2020-01-30 14:28:46 <vivi> reidrankin_, is correct - he saved my life last night :P
2020-01-30 14:29:06 <daemon> perfect
2020-01-30 14:29:16 <daemon> reidrankin, you would not know how to remove apk its self as well as libssl etc
2020-01-30 14:29:37 <mps> daemon: apk add /path/dir/pkgname-ver-apk
2020-01-30 14:29:38 <daemon> effectively everything but sh and gcc
2020-01-30 14:29:55 <reidrankin> I think you can do that if you're operating on a detached root
2020-01-30 14:30:03 <mps> no need to install alpine-sdk
2020-01-30 14:30:04 <reidrankin> with the --root option
2020-01-30 14:30:10 <daemon> ah cool
2020-01-30 14:30:14 <daemon> thank you kindly
2020-01-30 14:30:45 <vivi> reidrankin, that fix unblocked a two week clogged pipe, I can finally work on the daemon lol
2020-01-30 14:30:55 <reidrankin> FYI: the *only* cache dir apk ever looks for is /etc/apk/cache, even if it's /var/cache/aok (the default) that's because alpine-baselayout (i think) has the a symlink there
2020-01-30 14:31:13 <daemon> I see
2020-01-30 14:31:14 <mps> for example 'apk add linux-lts-5.4.16-r0.apk'
2020-01-30 14:31:44 <reidrankin> vivi: happy to help, and I know go as well... just sayin'
2020-01-30 14:32:10 <reidrankin> s%/var/cache/aok%/var/cache/apk%
2020-01-30 14:33:10 <vivi> A fellow gopher - excellent!  Quick question (specific to raspi armhf) -- https://github.com/goodtft/LCD-show  this works fine on raspbian but that's for the birds (j/k)
2020-01-30 14:33:30 <vivi> I copied the dbd files and updated the cmdline and config -- no love (this bash is ... yucky)
2020-01-30 14:33:36 <reidrankin> daemon: I haven't tried 'apk del apk' before, but I bet you could do it on the live system by using the binary from apk-static
2020-01-30 14:34:16 <reidrankin> obviously you'd want to remove that package too, but there's nothing that says you can't copy the static-linked binary out of the way, use that, and then delete it
2020-01-30 14:34:27 <daemon> makes sense
2020-01-30 14:34:36 <vivi> This is all it's doing : https://github.com/goodtft/LCD-show/blob/master/LCD35-show  --- I don't need the X11 /xorg stuff, just need cmdline + config to load that dbd file or whatever and hopefully console output does what it's supposed to do.
2020-01-30 14:34:43 <daemon> because its all a static binary via busybox on the default images
2020-01-30 14:36:16 <vivi> raspbian and alpine obvs do not share kernel args making this a spaghetti mystery
2020-01-30 14:36:29 <reidrankin> daemon: actually, the default image uses dynamic-linked apk
2020-01-30 14:37:00 <reidrankin> so don't just `cp /sbin/apk /somewhere`
2020-01-30 14:43:17 <reidrankin> vivi: that's because dtoverlay isn't a linux kernel option
2020-01-30 14:43:49 <reidrankin> it's interpreted by the raspian init system, probably in the initramfs stage
2020-01-30 14:44:39 <reidrankin> needless to say alpine's init system doesn't handle devicetree overlays
2020-01-30 14:45:06 <reidrankin> (this is not to say you're out of luck; you just have to load it yourself)
2020-01-30 14:50:10 <mps> why we then have dtbo files for RPi
2020-01-30 14:50:11 <daemon> ok so the next part of the plan is to source build gcc/mfr/isl/..... (same version as the apk) and install them onto a scratch image (as well as bourne shell etc...)
2020-01-30 14:50:20 <daemon> this will be fun :)
2020-01-30 14:50:38 <daemon> 
ACTION opens the LFS documentation

2020-01-30 14:52:37 <reidrankin> mps: that's actually a good question. There might be something different in the initramfs shipped with the pi image; I'm checking now.
2020-01-30 14:52:58 <vivi> reidrankin, makes sense - not a showstopper, would just be neat for demo purposes.
2020-01-30 14:54:37 <mps> maybe RPis bootloader (which is not u-boot) know how to load/use dtbo files
2020-01-30 14:54:58 <mps> but I don't have much experience with RPis
2020-01-30 14:55:44 <reidrankin> dtbo's are applied by the kernel, so I can imagine it might stick them somewhere special where the kernel's looking for them
2020-01-30 14:56:01 <reidrankin> which would explain the lack of userspace involvement
2020-01-30 14:58:23 <clandmeter> mps, what is the solution to that intel  gpu lockups
2020-01-30 14:58:47 <clandmeter> mps: is there something i can do from ssh to reset it?
2020-01-30 14:58:54 <reidrankin> looks like that's what's happening; the raspi 'start.elf' bootloader reads config.txt and loads the dtb and dtbo's before handling off to the kernel
2020-01-30 14:59:09 <mps> clandmeter: to me helps adding some swap, I have 2GB
2020-01-30 14:59:19 <clandmeter> i mean it hangs now
2020-01-30 14:59:24 <clandmeter> anything i can do to unhang it?
2020-01-30 14:59:50 <mps> wait some or try to switch to VC
2020-01-30 14:59:51 <reidrankin> seems I misunderstood; dtoverlay was never a *kernel* parameter, it was for the bootloader
2020-01-30 15:00:02 <mps> wait some time
2020-01-30 15:00:27 <clandmeter> VC?
2020-01-30 15:00:44 <mps> virtual console
2020-01-30 15:00:53 <clandmeter> http://tpaste.us/pKox <- this is similar issue right?
2020-01-30 15:01:01 <mps> ctrl-alt-F1 :)
2020-01-30 15:01:12 <clandmeter> nah nothing works
2020-01-30 15:01:14 <clandmeter> just ssh
2020-01-30 15:01:24 <mps> yes, that is
2020-01-30 15:01:53 <mps> clandmeter: what is your kernel version
2020-01-30 15:02:33 <clandmeter> 5.4.12-144.current
2020-01-30 15:02:54 <mps> ah, ncopa added patch in 5.4.14
2020-01-30 15:03:22 <mps> your are outdated
2020-01-30 15:04:05 <reidrankin> vivi: have you tried copying tft35a-overlay.dtb to /boot/overlays/tft35a-overlay.dtb and /boot/tft35a.dtbo, and adding 'dtparam=i2c_arm=on', 'dtparam=spi=on', 'enable_uart=1', and 'dtoverlay=tft35a:rotate=90' to /boot/config.txt?
2020-01-30 15:04:08 <clandmeter> i am not on alpine ;-)
2020-01-30 15:04:12 <clandmeter> dont tell anyone
2020-01-30 15:04:12 <mps> reidrankin: yes, bootloader fetches in load dtb's and dtbo's, not kernel
2020-01-30 15:04:55 <mps> clandmeter: be assured, no one will hear from me that you don't use alpine ;)
2020-01-30 15:05:31 <clandmeter> thats not what i said
2020-01-30 15:05:40 <clandmeter> dont twist my words! :p
2020-01-30 15:05:58 <mps> :)
2020-01-30 15:06:35 <mps> don't worry, also I have few lxc boxes with something else and not alpine :)
2020-01-30 15:09:46 <mps> reidrankin: here is alpine  example of u-boot loading kernel and dtb, http://tpaste.us/WRZ0
2020-01-30 15:09:55 <mps> on RPi zero W
2020-01-30 15:19:07 <ncopa> clandmeter: you coudl try the alpine kernel
2020-01-30 15:19:22 <ncopa> i backported the patch
2020-01-30 15:19:31 <ncopa> i guess it should be reported upstream
2020-01-30 15:19:47 <ncopa> to kernel.org
2020-01-30 15:19:59 <clandmeter> i guess its a known issue?
2020-01-30 15:20:29 <ncopa> yes and they fixed it in drm-tip (or what they call it) and it looks like it was  included in 5.5 kernel
2020-01-30 15:35:18 <mps> ncopa: when we talking about it, this should be 'backported' to 3.11-stable
2020-01-30 15:35:51 <ncopa> what should be backported?
2020-01-30 15:36:01 <mps> kernel
2020-01-30 15:36:10 <ncopa> the gpu fix?
2020-01-30 15:36:20 <misterunknown> mps: Does that affect 3.11? 3.11 uses kernel 4.19
2020-01-30 15:36:27 <mps> because of gpu fix
2020-01-30 15:36:33 <ncopa> 3.10 uses 4.19 kernel
2020-01-30 15:36:40 <ncopa> 3.11 uses 5.4
2020-01-30 15:37:24 <telmich_> good evening
2020-01-30 15:37:42 <telmich_> I've an alpine specific keepalived problem and was wondering if anyone has seen this before: https://github.com/acassen/keepalived/issues/1490#issuecomment-580290972
2020-01-30 15:38:15 <misterunknown> ncopa: Hm, I'm still on 4.19 somehow...
2020-01-30 15:38:17 <mps> ncopa: I know that you are busy but some people on 3.11-stable told that they have this gpu issue
2020-01-30 15:39:09 <mps> misterunknown: you upgraded from 3.10 to 3.11 and not installed 'from scratch'?
2020-01-30 15:39:16 <misterunknown> mps: Yes.
2020-01-30 15:39:36 <ncopa> misterunknown: you may need to apk add linux-lts && apk del linux-vanilla
2020-01-30 15:39:37 <mps> that's our luck :)
2020-01-30 15:39:44 <daemon> is there anyway to find out what compilation flags was used when building something before it got turned into an apk
2020-01-30 15:40:08 <ncopa> daemon: look at the APKBUILD or the build log
2020-01-30 15:40:12 <telmich_> argh... it might be the rp_filter setting
2020-01-30 15:40:17 <mps> I would expect more bug reports if all those who upgraded also installed linux-lts
2020-01-30 15:40:37 <daemon> ncopa, cool where abouts would I locate those
2020-01-30 15:40:46 <misterunknown> ncopa, mps Ok, so I'm lucky I didn't upgrade? Should I upgrade now? Or is it risky?
2020-01-30 15:41:49 <mps> misterunknown: if you upgrade kernel to linux-lts and don't remove linux-vanilla you will have both
2020-01-30 15:42:06 <mps> s/upgrade/install/
2020-01-30 15:42:35 <mps> and you will have option which one to boot at boot menu
2020-01-30 15:42:35 <ncopa> daemon: apk search --origin --exact thepackage
2020-01-30 15:42:41 <ncopa> will tell you the aports name
2020-01-30 15:42:49 <ncopa> it is often same as the package name
2020-01-30 15:43:08 <daemon> cool cheers ncopa
2020-01-30 15:43:38 <ncopa> apk policy thepackage will tell what repo it came from
2020-01-30 15:43:42 <ncopa> eg main or community
2020-01-30 15:43:59 <ncopa> then you find it in https://git.alpinelinux.org/aports/tree
2020-01-30 15:44:30 <misterunknown> mps: Hm, I have a root server, which I don't can access other than ssh, so I would have to remove the old kernel. But I meant should I wait until the bugfix you just talked about is backported?
2020-01-30 15:45:08 <mps> on the remote server you don't run X?
2020-01-30 15:45:17 <ncopa> only needed to wait if you use X with intel GPU
2020-01-30 15:45:57 <telmich_> hmm... resetting rp_filter to 0 on all interfaces doesn't help either
2020-01-30 15:46:12 <misterunknown> Ok. I don't run X, so I'll try it.
2020-01-30 15:47:00 <ncopa> telmich_: i have not seen that problem before
2020-01-30 15:47:13 <mps> I run it on my x86_64 build box with intel gpu but without X, never had issue
2020-01-30 15:47:32 <telmich_> ncopa: thanks for the quick follow up. Me neither and it's very strange because tcpdump sees the packets, but keepalived not
2020-01-30 16:00:04 <daemon> after I have found what a pkgs real name is for instance gcc-9.2.0-r3 how do I use that to find out its 'APKBUILD' or build log
2020-01-30 16:00:09 <daemon> I assume its listed somewhere online?
2020-01-30 16:00:55 <_ikke_> https://pkgs.alpinelinux.org/packages?name=gcc&branch=edge
2020-01-30 16:01:20 <daemon> perfect cheers _ikke_
2020-01-30 16:42:36 <ncopa> https://git.alpinelinux.org/aports/tree/main/gcc/APKBUILD
2020-01-30 17:35:38 <maldridge> mps: the problem is I don't think its packageable.  they're not shipping the cleanest cmake
2020-01-30 17:42:26 <mps> maldridge: maybe this can be starting point https://git.archlinux.org/svntogit/community.git/tree/trunk/PKGBUILD?h=packages/osquery
2020-01-30 17:44:05 <mps> sometime Arch's PKGBUILDs help me with 'hard' upstream packages
2020-01-30 18:11:45 <maldridge> there were two problems as best as I could tell: 1) they can't find openssl; 2) they aggressively whitelist platforms they're willing to build for
2020-01-30 18:28:15 <mps> well, I wrote 'starting point', but I have no doubt you can make it
2020-01-30 19:11:08 <hfingler> I've been trying to compile an aws sdk to a musl static library, but somehow after compiling it in an Alpine container, the library still contains some glibc symbols, such as __dso_handle, does anyone know what could be happening? I have a gist to building it: https://gist.github.com/hfingler/f5597511af0ddf9d0b8ba4ca82bdf781
2020-01-30 20:33:11 <mps> clandmeter: https://git.launchpad.net/~ubuntu-kernel/ubuntu/+source/linux/+git/focal/commit/?h=master-next&id=8c6f3bd6fbb9d3a135baf3ad642e48781afc3c48
2020-01-30 20:33:36 <mps> they backported same patch for i915 as ncopa did
2020-01-30 20:40:45 <clandmeter> I will hope it will land in my os sooner than later
2020-01-30 20:43:53 <mps> btw, they backported also wireguard to 5.4
2020-01-30 20:44:40 <mps> and fix I need for my chromebook, rockchip-drm annoying clock bug
2020-01-30 20:44:58 <mps> going to build new kernel for it :)
2020-01-30 20:45:03 <bsdlain> Hi
2020-01-30 20:49:53 <Cogitri> mps: imho there's not much point to backport Wireguard when there's the dkms module
2020-01-30 20:51:07 <mps> well, maybe, but they did it for their users, to not have hassle of using dkms
2020-01-30 20:53:28 <Cogitri> Well, the hassle on Alpine is adding one module :)
2020-01-30 20:53:35 <Cogitri> s/module/package/
2020-01-30 20:54:06 <Cogitri> If you want to spare users from that declare a dep on wireguard for linux-lts instead of backporting the patch
2020-01-30 20:55:42 <mps> if I'm kernel maintainer I would add wireguard patch in main/linux-vanilla/lts long time ago :)
2020-01-30 21:18:36 <vivi_> https://termbin.com/9nr3v Thoughts?
2020-01-30 21:21:09 <Cogitri> Probably forgot to link against the rpi libs or smth
2020-01-30 21:23:05 <vivi_> Yeah, my thoughts as well. I'll keep hacking away at it.
2020-01-30 21:24:05 <Cogitri> Good luck, I suppose
2020-01-30 21:42:33 <reidrankin> vivi_: try putting -lm -lpthread -lbcm_host at the end, after the .o's
2020-01-30 21:43:12 <vivi_> Currently looks like :          g++ -L/opt/vc/lib -lm -lpthread -lbcm_host -o $(EXECUTABLE) main.o mailbox.o sample.o wave_reader.o transmitter.o
2020-01-30 21:43:38 <reidrankin> instead do: g++ -L/opt/vc/lib -o $(EXECUTABLE) main.o mailbox.o sample.o wave_reader.o transmitter.o -lm -lpthread -lbcm_host
2020-01-30 21:44:12 <reidrankin> and upgrade to c++17, we miss you over here in the future
2020-01-30 21:44:48 <reidrankin> 
ACTION has been waiting for c++20 to come out for several years

2020-01-30 21:45:26 <reidrankin> seriously, "for (auto&& thing : container_of_things) {}" is worth its weight in gold
2020-01-30 21:46:08 <Cogitri> I'm waiting for module support so I can finally stop guessing if the cryptic error I'm getting is because I forgot to include some header or due to something else :^)
2020-01-30 21:46:18 <vivi_> I'm 10x more familiar with Rust over C++ - by that I mean I have exactly one rust project in production and zero C++ - this is a strange universe to me
2020-01-30 21:48:15 <bl4ckb0ne> 10*0 is 0
2020-01-30 21:48:21 <vivi_> ^^ precisely.
2020-01-30 21:48:31 <reidrankin> oh, yeah, c++17 also has inline static class members, so you don't have to make a .cpp file with like three lines just to add a static instance of your class
2020-01-30 21:48:33 <Cogitri> I mostly write D/Rust for my OSS projects too (luckily), but I have to use C++ at work
2020-01-30 21:48:51 <Cogitri> (But we wrap about all of std::*, so it's luckily not that bad)
2020-01-30 21:49:05 <vivi_> Go / Python and unfortunately some web bs
2020-01-30 21:49:35 <Cogitri> Ah, I write in a C++ webframework :^)
2020-01-30 21:49:54 <vivi_> reidrankin, good guess but she's not gonna take.  I don't know if I want to bother you with this silly problem - it's just for debugging in the field.
2020-01-30 21:49:57 <Cogitri> But honestly I prefer that over writing the JS for it I think
2020-01-30 21:51:24 <vivi_> Yeah, keeping up with the latest react library is a full-time gig; you must be < 30 years of age and have a fridge stocked of Mtn Dew (Code Red) and a drawer full of aderall
2020-01-30 21:52:56 <vivi_> And then in three weeks when the lib is 'depracated' or a slightly more hip version surfaces five mins later you have to explain to your lead why you're re-opening an old ticket.
2020-01-30 21:54:50 <vivi_> PS reidrankin found the missing thing in `/opt/vc/lib` - need to figure how to add that to LDD path or w/e
2020-01-30 21:54:54 <Cogitri> Yeaah, JS sure is fun
2020-01-30 21:55:19 <Cogitri> `-L` is for adding library paths during linking
2020-01-30 21:55:59 <vivi_> Sick, you guys are the absolute most helpful IRC friends ever :')
2020-01-30 21:57:37 <vivi_> <3   build-armhf-1:~/fm_transmitter# ./fm_transmitter
2020-01-30 21:57:37 <vivi_> Usage: fm_transmitter [-f <frequency>] [-b <bandwidth>] [-d <dma_channel>] [-r] <file>
2020-01-30 21:58:05 <daemon> just for the cutes: (my dog emulating the appropriate way to sleep after a day of debugging perl and lots of alcohol) https://itsosticky.com/pr3jfm-1h25p71-c6more
2020-01-30 21:58:16 <Cogitri> Nice :)
2020-01-30 21:58:32 <daemon> ;)
2020-01-30 23:25:26 <daemon> hmm
2020-01-30 23:25:26 <daemon> # apk fetch glibc
2020-01-30 23:25:26 <daemon> glibc: unable to select package (or its dependencies)
2020-01-30 23:25:29 <daemon> no glibc?
2020-01-30 23:25:48 <north1> no glibc, this is Alpine Linux
2020-01-30 23:26:06 <daemon> thisi s where musl steps in right
2020-01-30 23:26:25 <north1> yes, musl is the libc used in Alpine, instead of glibc (GNU libc)
2020-01-30 23:26:38 <daemon> ok so I need to find out howto build binutils with musl
2020-01-30 23:26:40 <daemon> got it
2020-01-30 23:26:41 <daemon> cheers
2020-01-30 23:30:09 <north1> Should be the same as glibc
2020-01-30 23:30:17 <daemon>  /mnt/opusos/src/binutils-2.33.1/build # ../configure --prefix=$lfs_tools --with-sysroot=$lfs --with-lib-path=$lfs_tools/lib --target=$lfs_tgt --disable-nls --disable-werror
2020-01-30 23:30:21 <daemon> apparently not
2020-01-30 23:30:33 <north1> Well, there is an Alpine LInux APKBUILD for it
2020-01-30 23:30:38 <daemon> boots out with gcc cannot create executables
2020-01-30 23:30:41 <daemon> and it can
2020-01-30 23:31:13 <daemon> https://dpaste.org/aNRr
2020-01-30 23:31:17 <daemon> seems to be the issue
2020-01-30 23:31:28 <daemon> hmm wonder wat Scrt* are
2020-01-30 23:32:29 <daemon> ah multi arch support
2020-01-30 23:32:34 <daemon> wonder if I can disable multilib for binutils
2020-01-30 23:39:30 <daemon> got it :)
2020-01-30 23:39:34 <daemon> musl-dev + libtool
2020-01-31 00:04:18 <vktec> Hey, I'm trying to build linux off an alpine install but getting the error 'asm/types.h: no such file or directory'. What package would asm/types.h be in?
2020-01-31 00:05:32 <north1> linux-headers holds /usr/include/asm/types.h
2020-01-31 00:07:39 <vktec> Ah cool, thanks
2020-01-31 01:35:30 <daemon> hey all having a bit of a problem compiling an application: https://dpaste.org/JNrh
2020-01-31 01:35:34 <daemon> gcc infact
2020-01-31 01:35:38 <daemon> anything special I have to pass it?
2020-01-31 01:52:40 <vivi> daemon, not sure there's enough info there to determine
2020-01-31 01:53:02 <daemon> vivi, whatdo you need I have it all infront of me but honestly this log is pretty darn long :P
2020-01-31 01:53:57 <vivi> config.status: executing default commands -- what are the 'default commands' ?
2020-01-31 01:54:13 <daemon> vivi, https://dpaste.org/XHQb may shed some light
2020-01-31 01:54:16 <daemon> let me see
2020-01-31 01:55:23 <reidrankin> run your compile command with ' 2>&1 | nc termbin.com 9999' on the end
2020-01-31 01:55:34 <reidrankin> (you'll need netcat)
2020-01-31 01:55:40 <daemon> adjusting!
2020-01-31 01:56:03 <reidrankin> that'll give us a log with all the things
2020-01-31 01:56:18 <vivi> alias tb="nc termbin.com 9999"
2020-01-31 01:56:27 <vivi> ^^ love the hell out of this service
2020-01-31 01:57:19 <daemon> hmm I wonder if its configure or make that is failing ....
2020-01-31 01:57:36 <daemon> guessing make
2020-01-31 01:58:15 <daemon> this will take 5 minutes or so ... gcc is not the only thing being compiled and I need to --no-cache it as I added netcat
2020-01-31 01:58:32 <reidrankin> termbin.com, dweet.io, and ipify.org are some of my favorite public services
2020-01-31 02:00:43 <daemon> https://termbin.com/abyq
2020-01-31 02:00:53 <daemon> config.status: creating Makefile
2020-01-31 02:00:53 <daemon> https://termbin.com/abyq
2020-01-31 02:00:55 <daemon> hmm no crash
2020-01-31 02:01:23 <daemon> iis it make install that is failing
2020-01-31 02:01:30 <daemon> that does not look complete
2020-01-31 02:01:31 <daemon> for make
2020-01-31 02:01:41 <reidrankin> i guess termbin has a timeout
2020-01-31 02:02:11 <daemon> thats ok ill run netcat on one of my hosts and pipe it to a file
2020-01-31 02:02:25 <reidrankin> try adding ' &>/tmp/foo; </tmp/foo nc termbin.com 9999' instead
2020-01-31 02:02:38 <daemon> ah letit finish first
2020-01-31 02:02:42 <reidrankin> yeah, that would do
2020-01-31 02:03:27 <daemon> watching stuff compile is hypnotic
2020-01-31 02:04:46 <daemon> well system is at 100% on one core so presume its in the build
2020-01-31 02:05:12 <daemon> I dropped the -j4 just incase it created some whackyness in the order of reporting
2020-01-31 02:05:32 <daemon> https://termbin.com/j83i
2020-01-31 02:05:58 <daemon> so its actually mpc that is bailing
2020-01-31 02:08:05 <vivi> Aw heck I spilt flower dust all over my lenovo
2020-01-31 02:08:28 <daemon> doing some baking?
2020-01-31 02:08:36 <vivi> Indubidably.
2020-01-31 02:08:40 <daemon> :)
2020-01-31 02:09:42 <reidrankin> are you cross-compling musl gcc from a libc system, or libc gcc from a musl system?
2020-01-31 02:10:06 <vivi> 'Indubitably' was misspelt - should certainly signal some clandestine behaviour.
2020-01-31 02:10:06 <daemon> reidrankin, trying not to, the host is standard alpine, so musl and the gcc I am trying to build should also be musl
2020-01-31 02:10:30 <daemon> it should really not be trying to do anything glibc'ish
2020-01-31 02:12:58 <reidrankin> the log says "checking build system type... x86_64-pc-linux-musl" "
2020-01-31 02:12:58 <reidrankin> checking host system type... x86_64-pc-linux-musl
2020-01-31 02:12:58 <reidrankin> " "checking target system type... x86_64-lfs-linux-gnu"
2020-01-31 02:13:18 <reidrankin> yousure you're not using any odd target flags?
2020-01-31 02:13:20 <daemon> I cannot see where the hell its pulling thatfro
2020-01-31 02:13:21 <daemon> m
2020-01-31 02:13:22 <reidrankin> to configure
2020-01-31 02:13:53 <daemon> https://dpaste.org/XHQb
2020-01-31 02:13:59 <daemon> is all the args
2020-01-31 02:14:10 <daemon> ah
2020-01-31 02:14:11 <reidrankin> oh right
2020-01-31 02:14:15 <daemon> remove the target?
2020-01-31 02:14:31 <reidrankin> no, change it to x86_64-pc-linux-musl
2020-01-31 02:15:27 <daemon> I really should have spotted that
2020-01-31 02:15:34 <daemon> though in theory it should be x86_64-lfs-linux-musl
2020-01-31 02:15:39 <daemon> though I imagine that would make little difference
2020-01-31 02:16:57 <reidrankin> well, it'll control where things get looked for, it's not like someone's parsing the string and seeing "oh i see you're gnu not musl"
2020-01-31 02:17:17 <daemon> hmm I should write my self a note
2020-01-31 02:17:29 <daemon> because I THINK that becomes important a bit later on
2020-01-31 02:17:49 <reidrankin> so unless all your stuff gets stuck in places like /usr/lib/x86_64-lfs-linux-musl instead of /usr/lib/x86_64-pc-linux-musl you're gonna have a bad time
2020-01-31 02:18:29 <daemon> I believe it actually does
2020-01-31 02:18:36 <daemon> I am actually following the linux from scratch guide
2020-01-31 02:18:41 <daemon> though it is targetted towards glibc
2020-01-31 02:18:45 <daemon> I am removing references to it
2020-01-31 02:18:47 <reidrankin> ...oh. that would change things somewhat...
2020-01-31 02:18:48 <daemon> and totally missed the target -_-
2020-01-31 02:19:12 <reidrankin> sorry about that.
2020-01-31 02:19:32 <daemon> hmm for what? I never said that information till now :P
2020-01-31 02:19:36 <reidrankin> in that case you should change it to whatever you're using as a platform string
2020-01-31 02:19:53 <daemon> yep I went for LFS_TARGET=x86_64-lfs-linux-musl
2020-01-31 02:20:09 <vivi> These awful Hauweii modems require usb_modeswitch - they default to /dev/sr0 (cdrom) and you gotta kick them into serial mode.  Thus far I wrote it as syscalls into the go bin we ship but now that reidrankin solved the packages on boot - I think I'll go back to old-school ppp + chatscripts
2020-01-31 02:20:25 <daemon> ug usb modeswitch
2020-01-31 02:20:27 <vivi> With that now I must read a whole load on udev + openrc and cross my fingers.
2020-01-31 02:20:29 <daemon> remind me of cheaper printer/scanner
2020-01-31 02:20:38 <reidrankin> what the heck are they doing as a cdrom drive?
2020-01-31 02:21:22 <vivi> reidrankin, for the Windows chuds - they need an autoexec.bat to use their computerbox.
2020-01-31 02:21:55 <daemon> checking build system type... Invalid configuration `x86_64-pc-linux-musl': machine `x86_64-pc-linux' not recognized
2020-01-31 02:21:56 <daemon> hmm
2020-01-31 02:22:02 <reidrankin> i have a printer that does that, but at least it's composite device with both the printer and the cdrom
2020-01-31 02:22:11 <daemon> let me try it with pc instead..
2020-01-31 02:22:11 <vivi> insert usb into Windows ME -> UPnP (topkek) fires a thing that execs autoexec - now that the driver's installed it'll go serial and actually be useful.
2020-01-31 02:23:48 <vivi> I also have this wicked task of determining product+vendor hwids to determine which ppp provider script to run - but I gotta do the old-school hayes AT commands to determine if Canada || USA (CA : $30/mo per device -- USA $30/YR per device){
2020-01-31 02:24:31 <vivi> Also ppp cannot be doing a thing with certain modems due to /dev/ttyUSB[1,5] accepting different types of commands... it's tedious.
2020-01-31 02:27:09 <vivi> Ahem - different interfaces accepting different types of functionality.  /dev/ttyUSB0 is useless on most, but on the Quectels (CAT-M1 LTE) could be /dev/ttyUSB[1,5] whilst the ZTE or Hauweii on Rogers mobile may only data over /dev/ttyUSB2 - if I need to monitor CSQ,COPS (signal strength, network operators etc) I have to serial to /dev/ttyUSB1
2020-01-31 02:41:38 <daemon> hmm alternative plan
2020-01-31 02:41:53 <daemon> how would I install alpine ontop of scratch as a 'standard os' without busybox/ static linked binaries
2020-01-31 02:47:02 <reidrankin> well, you could just throw apk-static on there
2020-01-31 02:47:43 <reidrankin> add virtual packages for busybox and other stuff you've got already so it won't try to install those
2020-01-31 02:47:55 <reidrankin> then everything should work fine, probably
2020-01-31 03:26:26 <simon_> hello again!
2020-01-31 03:26:29 <codebam> does openrc have user scripts?
2020-01-31 03:27:03 <simon_> I did: apk add xxhash-dev
2020-01-31 03:27:18 <simon_> but get: undefined reference to `XXH64'
2020-01-31 03:27:40 <simon_> how do I link in xxhash to my C executable?
2020-01-31 03:54:35 <ToBeLinux> Hello everyone
2020-01-31 03:55:58 <ToBeLinux> Does anyone have any idea how to install LXDE on Alpine? I have been searching left and right and got nowhere, unfortunately.
2020-01-31 04:15:40 <simon_> If I type nm /usr/lib/libxxhash.so.0.7.2 then it says: nm: /usr/lib/libxxhash.so.0.7.2: no symbols
2020-01-31 04:28:19 <simon_> hmmmm... maybe everything is self contained in xxhash.h
2020-01-31 04:28:37 <simon_> but compiler always says it is missing XXH64() function... :-(
2020-01-31 04:32:54 <simon_> hmmmm... the xxhash.h in the apk is NOT the xxhash.h from github... it's much shorter...
2020-01-31 04:38:36 <simon_> got it to work if I take the actual xxhash.h and xxh3.h files from https://github.com/Cyan4973/xxHash and include xxhash.h in my C source file directly...
2020-01-31 04:39:06 <simon_> could not figure out how to use or link against the apk xxhash packages... they appear to be useless?
2020-01-31 04:39:24 <simon_> or very specifically designed for python?
2020-01-31 05:37:51 <yashi> Guys, how to compile QT5 apps on alpine? I've installed qt5-qtbase-dev etc
2020-01-31 05:38:08 <yashi> but when I run ./configure it complains that QTDIR is not set (which it is, to /usr/lib/qt5)
2020-01-31 05:48:22 <markand> which program are you trying to build?
2020-01-31 05:59:33 <yashi> lyx
2020-01-31 07:27:48 <vktec> Hey, I'm having some issues with xorg. I think I've installed all the relevant packages but it doesn't seem to be finding my GPU in /dev - maybe a udev/mdev issue? Here's my /var/log/Xorg.0.log https://termbin.com/gdvx
2020-01-31 07:46:59 <AinNero> vktec: how would a gpu in /dev look like?
2020-01-31 07:48:14 <vktec> /dev/dri/card0 is what it's looking for, I just guessed at it being the GPU given that I only know DRI as Direct Rendering Infrastructure
2020-01-31 07:49:31 <AinNero> have you uses setup-xorg-base ?
2020-01-31 07:49:36 <vktec> Yes
2020-01-31 08:45:01 <misterunknown> j
2020-01-31 12:24:25 <daemon> is there anyway to do it without busybox what so ever
2020-01-31 12:24:35 <daemon> basically tell apk to install sh etc.. to /mnt/target
2020-01-31 12:24:41 <daemon> then copy from that directory to scratch
2020-01-31 12:28:46 <AinNero> daemon: apk has the --root option
2020-01-31 12:29:00 <AinNero> similar workflow to bootstrapping a chroot
2020-01-31 12:29:09 <daemon> ah cool
2020-01-31 12:30:09 <daemon> so all I would need to do is install the non-static version of the required base tools and send them to /target with --root
2020-01-31 12:32:39 <AinNero> https://wiki.alpinelinux.org/wiki/Alpine_Linux_in_a_chroot#Install_the_alpine_base_installation_onto_the_chroot
2020-01-31 12:33:42 <daemon> hmmm do I still need to populate /dev etc
2020-01-31 12:33:47 <daemon> bearing in mind the targetis a docker image
2020-01-31 12:34:07 <daemon> I really just want a docker image that has no static binaries on it
2020-01-31 14:33:47 <phanter> hi. since a week I am using alpine linux on my NAS on a ZFS root. I am creating snapshots. When I try to view those snapshots I get the following error: -ash: cd: can't cd to zfs-auto-snap_hourly-2020-01-16-2100/: Symbolic link loop
2020-01-31 14:34:03 <phanter> any ideas for how to resolve this?
2020-01-31 14:34:49 <phanter> hi. since a week I am using alpine linux on my NAS on a ZFS root. I am creating snapshots. When I try to view those snapshots I get the following error: -ash: cd: can't cd to zfs-auto-snap_hourly-2020-01-16-2100/: Symbolic link loop
2020-01-31 14:35:39 <AinNero> phanter: call `stat` on above path
2020-01-31 14:36:04 <phanter> hi. since a week I am using alpine linux on my NAS on a ZFS root. I am creating snapshots. When I try to view those snapshots I get the following error: -ash: cd: can't cd to zfs-auto-snap_hourly-2020-01-16-2100/: Symbolic link loop
2020-01-31 14:36:22 <phanter> sorry about reposting... I am messing up my copy-pastes :-(
2020-01-31 14:36:31 <phanter> AinNero: will do and report back
2020-01-31 14:36:50 <AinNero> i'd like to see the output from stat
2020-01-31 14:37:19 <phanter> stat: can't stat 'zfs-auto-snap_hourly-2020-01-16-2100/': Symbolic link loop
2020-01-31 14:38:24 <phanter> ow, and I get this error on _all_ snapshots, even in different zpools
2020-01-31 14:40:03 <AinNero> stat does not fail with link loop when called on a self-pointing symlink
2020-01-31 14:40:12 <AinNero> so i guess its a zfs fuckup
2020-01-31 14:40:26 <phanter> so, I should go for a reinstall?
2020-01-31 14:40:51 <phanter> however, I have this issue both in my virtual machine install as well as on the bare metal install...
2020-01-31 14:40:55 <AinNero> i dont know about zfs so i cant really help you
2020-01-31 14:54:08 <Cogitri> Did I head ZFS? :P
2020-01-31 15:12:40 <pltrz> how can I get my encrypted home logical volume to decrypt on bootup? I added an entry to /etc/crypttab with "home /dev/mapper/machina-vault" where machina is my volume group, and added "cryptsetup" to mkinitfs.conf
2020-01-31 15:13:18 <gu0tvuNHFV8o> did you redo "mkinitfs"?
2020-01-31 15:13:25 <pltrz> but I have to manually "cryptsetup open /dev/machina/vault home"
2020-01-31 15:13:31 <pltrz> then mount it
2020-01-31 15:15:57 <pltrz> yes, and just did it again and rebooted to be sure
2020-01-31 15:17:34 <phanter> hi Cogitri . I have been asking the question I wrote before a few times and we talked about it before as well (maybe you do remember).
2020-01-31 15:18:56 <phanter> The issues was that, at least in my case, the auto mounting of the snapshot does not work (or something like that).
2020-01-31 15:20:07 <phanter> when I mount the snapshot explicitely (mount -t zfs pool/fs@snapshot /mount/point), I can indeed see the snapshot. This does (for me) not explain the error message as such, but it does mean that my snapshot functionality is working.
2020-01-31 15:37:46 <kaey> phanter: i can reproduce your problem on alpine edge, but not on archlinux
2020-01-31 15:38:16 <kaey> both ship 0.8.3, guess something in kernel is configured differently
2020-01-31 16:50:52 <phanter> kaey: you are right... it was an alpine issue and I was in the wrong channel... sorry
2020-01-31 17:18:57 <simon_> If I type nm /usr/lib/libxxhash.so.0.7.2 then it says: nm: /usr/lib/libxxhash.so.0.7.2: no symbols
2020-01-31 17:19:58 <simon_> how can I use xxhash on Alpine in a C program after apk add xxhash-dev ?
2020-01-31 17:53:42 <gu0tvuNHFV8o> simon_: nm -D is what your are lookign for i guess
2020-01-31 17:54:34 <simon_> thanks!
2020-01-31 17:54:43 <simon_> that makes the functions show up :-)
2020-01-31 17:54:51 <gu0tvuNHFV8o> yeah, it should
2020-01-31 17:55:57 <simon_> now for the next mystery: if I run the 2 line C program on this page https://github.com/Cyan4973/xxHash then it always complains that the XXH64() function is undefined...
2020-01-31 17:56:47 <simon_> how do I link in that libxxhash.so file to my C program to avoid the undefined reference?
2020-01-31 18:02:21 <gu0tvuNHFV8o> what is your gcc line?
2020-01-31 18:02:32 <phanter> when booting up my nas, it shows errors like: cannot import 'zpool': pool already exists.
2020-01-31 18:03:04 <phanter> The pool works, but it there a way to not have these errors (and fix the issue causing the error)?
2020-01-31 18:04:18 <Cogitri> I guess it might try to import your pool when the initramfs has already imported it?
2020-01-31 18:14:07 <simon_> gu0tvuNHFV8o, this is my program: void main(void) { char buffer[] = "hello"; XXH64_hash_t hash = XXH64(buffer, sizeof(buffer), 123); printf("hash=0x%lx\n", hash); }
2020-01-31 18:14:25 <simon_> gu0tvuNHFV8o, and this is the gcc line: gcc -o myxxh myxxh.c
2020-01-31 18:14:44 <simon_> gu0tvuNHFV8o, and this is the error: myxxh.c:(.text+0x36): undefined reference to `XXH64'
2020-01-31 18:14:52 <mps> simon_: missing -lxxhash
2020-01-31 18:15:23 <mps> gcc -o myxxh myxxh.c -lxxhash
2020-01-31 18:16:08 <simon_> mps, thanks, that works... now to figure out why it doesn't work in my much larger code I'm trying to compile :-)
2020-01-31 18:17:38 <mps> simon_: welcome to C programming world ;)
2020-01-31 18:18:35 <simon_> hehe :-)
2020-01-31 18:19:27 <mps> but, best programming world, imo
2020-01-31 18:38:40 <Cogitri> Meh :P
2020-01-31 18:46:16 <mps> :)
2020-01-31 19:25:13 <simon_> mps, problem is that bigger project compiles with -static
2020-01-31 19:25:25 <simon_> so results in: /usr/lib/gcc/x86_64-alpine-linux-musl/9.2.0/../../../../x86_64-alpine-linux-musl/bin/ld: cannot find -lxxhash
2020-01-31 19:25:42 <simon_> same result for: gcc -static -o myxxh myxxh.c -lxxhash
2020-01-31 19:26:41 <simon_> mps, does alpine have a way to link xxhash statically somehow?
2020-01-31 19:27:05 <AinNero> simon_: do you have lxxhash.a ?
2020-01-31 19:27:15 <AinNero> .so for dynamic, .a for static linking, or something like that
2020-01-31 19:27:32 <AinNero> or maybe it was .la
2020-01-31 19:27:54 <simon_> AinNero, AFAICT apline apks do not contain a .a or .la for xxhash...
2020-01-31 19:28:29 <simon_> is that 'normal' not to provide a static version or something like xxhash?
2020-01-31 19:28:44 <AinNero> i think yes
2020-01-31 19:28:52 <mps> simon_: I think xxhash doesn't have static libs built
2020-01-31 19:31:47 <simon_> kk
2020-01-31 19:40:29 <alxsdf> you can probably alter the APKBUILD and add a -static
2020-01-31 19:41:14 <simon_> alxsdf, how do you mean?
2020-01-31 19:47:35 <alxsdf> simon_: All of the sources for the alpine packages are in a repo called 'aports' you can navigate to the right folder, alter the APKBUILD to add CFLAGS="-static" and maybe a few extra linking steps, and run abuild to make it into a package
2020-01-31 19:47:55 <alxsdf> then lxxhash can be managed by your package manager or whatever
2020-01-31 22:23:59 <vivi> Wow, after writing this go code in a way I thought would be awesome it's actually the opposite.
2020-01-31 22:24:42 <vivi> I did a struct that encapsulates the state of all other structs - I set the values directly without any mutex or anything and now I'm eating my hat.
2020-01-31 22:24:59 <vivi> Time to redesign with channels
2020-01-31 23:41:39 <casaca> p
2020-01-31 23:46:04 <north1> i