2020-09-01 02:14:07 <Jiggy> Anyone running 3.12 mixed w/ edge packages on a baremetal HDD install? I would like to upgrade to Caddy 2.x, but not sure if I can wait another 4-6 months for 3.13 :P
2020-09-01 02:23:51 <Jiggy> I'm thinking it may be best to stay on 3.12 instead of edge and just run caddy 2 containerized on Host network
2020-09-01 02:52:26 <TBB> it's risky most of the time to mix edge packages with stable ones. containerization is a good idea, or alternatively you could produce the package for stable.
2020-09-01 03:27:47 <artok> building caddy as static binary and all is good...
2020-09-01 03:44:20 <Jiggy> Thank you for the comments TBB and artok - I'm making a snapshot now (on 3.11.6), going to 3.12, and then either using a container or manual binary install
2020-09-01 04:27:15 <jjuran> 
ACTION waves at the familiar names in the room

2020-09-01 04:45:51 <baconicsynergy> So I noticed the bluetooth wiki page was quite sparse. I'm assuming its not considered a priority at this stage of development? Which is totally fine, I'm just curious
2020-09-01 04:47:56 <baconicsynergy> I noticed bluez is available in the repos. I installed it and I'm trying to run bluetoothctl, which hangs on "Waiting to connect to bluetoothd..."
2020-09-01 04:48:32 <baconicsynergy> Is bluetoothd able to run under openrc? or is it tied to systemd? Forgive me, I'm not very knowledgable about this subject
2020-09-01 04:54:52 <baconicsynergy> I tried starting bluetoothd with openrc and failed miserably ;_;
2020-09-01 04:57:37 <baconicsynergy> omg I got it!!!
2020-09-01 04:58:16 <baconicsynergy> sudo rc-update add bluetooth, sudo rc-service bluetooth start
2020-09-01 04:58:34 <baconicsynergy> gnome bluetooth menu is now populated!! I'm so happy <3
2020-09-01 05:00:28 <baconicsynergy> I have to admit I got kinda lucky here
2020-09-01 05:00:52 <baconicsynergy> I am not a very smart individual... but damn I feel good right now
2020-09-01 05:42:08 <shhhum[m]> Can anyone help me test weston if it actually launches or I am doing something wrong?
2020-09-01 05:43:15 <shhhum[m]> Steps to reproduce: https://gitlab.alpinelinux.org/alpine/aports/-/issues/11907
2020-09-01 05:47:55 <baconicsynergy> I can test! sounds like fun
2020-09-01 05:48:13 <baconicsynergy> I'm not very good at understanding wayland but i'll see if i can reproduce
2020-09-01 05:49:11 <shhhum[m]> Ok, thank you.
2020-09-01 06:14:20 <baconicsynergy> shhhum[m]: greetings from weston!
2020-09-01 06:14:53 <shhhum[m]> Yeah? Working fine?
2020-09-01 06:15:04 <shhhum[m]> Strange.
2020-09-01 06:15:10 <baconicsynergy> seems to be fine, tmux in weechat is running in weston-terminal right now
2020-09-01 06:15:19 <baconicsynergy> weechat in tmux*
2020-09-01 06:15:28 <shhhum[m]> What am I doing wrong then....
2020-09-01 06:15:35 <shhhum[m]> If Sway runs just fine.
2020-09-01 06:15:43 <baconicsynergy> sway runs fine for me too. how odd
2020-09-01 06:17:09 <monado> Hey everyone
2020-09-01 06:17:24 <shhhum[m]> What hardware do you have? In particular – GPU?
2020-09-01 06:17:52 <shhhum[m]> monado: Hey.
2020-09-01 06:17:56 <baconicsynergy> hi monado
2020-09-01 06:18:11 <monado> albr is gone
2020-09-01 06:18:12 <monado> Woot woot
2020-09-01 06:18:17 <baconicsynergy> I'm running alpine edge on a thinkpad x250
2020-09-01 06:18:31 <baconicsynergy> so, I believe I am using intel integrated graphics
2020-09-01 06:19:07 <shhhum[m]> Oh, alright. Thank you for testing.
2020-09-01 06:19:43 <baconicsynergy> jane:~$ lspci | grep -i vga
2020-09-01 06:19:44 <baconicsynergy> 00:02.0 VGA compatible controller: Intel Corporation HD Graphics 5500 (rev 09)
2020-09-01 06:19:54 <baconicsynergy> no problem, i hope you find the answer!!
2020-09-01 06:20:41 <shhhum[m]> Hm. I have 2 GPUs – integrated (Intel 4th gen) and RX590.
2020-09-01 06:21:02 <shhhum[m]> Maybe Weston do not support multi-gpu and somehow confuses?
2020-09-01 06:21:14 <baconicsynergy> sounds like it could definitely be a possibility
2020-09-01 06:21:33 <shhhum[m]> I'll test when I get back home.
2020-09-01 06:22:02 <baconicsynergy> I'll test it on my gaming rig with an rx480
2020-09-01 06:22:18 <baconicsynergy> i'll ping you if I encounter the hang
2020-09-01 06:22:37 <shhhum[m]> Thank you!
2020-09-01 06:29:17 <shhhum[m]> Seems like it could be it: https://gitlab.freedesktop.org/wayland/weston/-/issues/76
2020-09-01 06:33:54 <tsujp> is there a package for alpine which is ruby but with jemalloc instead? or must I compile my own
2020-09-01 06:40:24 <mps> tsujp: afaik no one pkg in alpine is built with jemalloc
2020-09-01 08:04:20 <gillesmauve> Baconicsynergy: gaming with wine?
2020-09-01 11:34:27 <Hello71> tsujp: you can load jemalloc with LD_PRELOAD
2020-09-01 11:34:54 <Hello71> although jemalloc is still in unmaintained
2020-09-01 12:48:55 <Guest41> Hi all, I'm trying to turn my RPi 4 into a Wireguard router. On a fresh 3.12 I've tried installing wireguard-rpi4@edge-community but it doesn't let me install , citing "ERROR: unsatisfiable constraints: linux-rpi4-5.4.61-r0: masked in: @edge-main satisfies: wireguard-rpi4-5.4.61-r0[linux-rpi4=5.4.61-r0]
2020-09-01 12:50:51 <Guest41> I updated the kernel using update-kernel, pointing it towards the edge-main repository for the latest kernel, but the problem persisted. I tried sudo apk add linux-rpi4@edge-main but that fails after a ton of errors. How should I proceed?
2020-09-01 12:55:56 <Guest41> Switching to 'Huest66' since this connection is unstable and I'm new to IRC (and also a Linux newbie if you haven't already noticed)
2020-09-01 12:59:18 <Hello71> mixing packages is a bad idea unless you definitely know what you're doing
2020-09-01 12:59:27 <Hello71> isn't there wireguard-dkms
2020-09-01 13:00:02 <ikke> We don't have dkms
2020-09-01 13:29:53 <Huest66> I'm still trying to get the hang of IRC so please 'mention' me when you reply so I might see the response if I automatically disconnect
2020-09-01 13:31:43 <ikke> Huest66: is there a reason you are trying to install wireguerd from edge?
2020-09-01 13:31:57 <ikke> It is available for 3.12: https://pkgs.alpinelinux.org/package/v3.12/community/armv7/wireguard-rpi4
2020-09-01 13:32:26 <ikke> And because these modules are tied to a specific kernel version, you need to use it from the same repository as the kernel you are using
2020-09-01 13:33:43 <PureTryOut[m]> Huest66: you can also join this channel from Matrix if you want, it's a bit more like modern messengers and doesn't disconnect you when you close the chat application #freenode_#alpine-linux:matrix.org
2020-09-01 13:41:51 <Huest66> @ikke oh I see what you were getting at. Come to think of it I have no real reason to use the edge version of wireguard. Guess I just wanted the latest version. Will test right away
2020-09-01 13:42:42 <Huest66> PureTryOut[m] I'm using kiwiirc via browser because it's the simplest. I'll explore other IRC applications when I have time, but thanks
2020-09-01 13:43:31 <PureTryOut[m]> Matrix is not an IRC application though 😉
2020-09-01 13:43:46 <Huest66> Is it where Neo is god?
2020-09-01 13:43:49 <PureTryOut[m]> It's a different thing entirely that just has abilities to connect to IRC stuff
2020-09-01 13:46:29 <Huest66> @ikke so 'apk add wireguard-rpi4' leads to the following message: "ERROR: unsatisfiable constraints: linux-rpi4-5.4.61-r0: masked in: @edge-main satisfies: wireguard-rpi4-5.4.61-r0[linux-rpi4=5.4.61-r0]  wireguard-rpi4-5.4.61-r0: masked in: @edge-community satisfies: world[wireguard-rpi4]"
2020-09-01 13:48:13 <Huest66> Just for reference my /etc/apk/repositories points to /media/mmcblk0p1/apks, https://alpine.global.ssl.fastly.net/alpine/latest-stable/main, with @edge-* connecting to various edge repositories of that mirror (I like HTTPS--makes me feel safe)
2020-09-01 13:48:39 <Huest66> I'll check out Matrix right now
2020-09-01 13:48:46 <ikke> Huest66: note that https://dl-cdn.alpinelinux.org/alpine works now
2020-09-01 13:48:55 <ikke> no need to use global.ssl.fastly.net anymore
2020-09-01 13:49:45 <ikke> Huest66: It's best to avoid these pinned repositories like Huest66 recommended
2020-09-01 13:49:46 <Huest66> Thanks for this--much easier to remember. Time to update my notes
2020-09-01 13:52:09 <Huest66> PureTryOut[m] I'm on Chrome OS now and I don't want to mess with my Qubes so I'll just have to make do with kiwiirc
2020-09-01 13:53:05 <Huest66> @ikke so if I switch to the link you provided things should be better? Is it because that particular repository is 'pinned'? (un-updated?)
2020-09-01 13:53:15 <ikke> No
2020-09-01 13:53:28 <ikke> pinned repostories are those preprended with an @
2020-09-01 13:53:35 <ikke> @foo https://...
2020-09-01 13:54:53 <Huest66> Got it. Aside from packages not available elsewhere (e.g. dante-server and dnscrypt-proxy), I'll avoid @pinned as much as I can. Good to know
2020-09-01 13:55:10 <PureTryOut[m]> Huest66: you can use Matrix in a browser too... https://app.element.io
2020-09-01 13:55:28 <PureTryOut[m]> You don't have to do it now, or ever, just look into it
2020-09-01 13:56:19 <Huest66> Will do
2020-09-01 15:48:04 <me1> jkl
2020-09-01 15:51:16 <me1> help i am getting seg fault in waybar on alpine edge
2020-09-01 18:12:58 <mckay-trabajo> hey.. anyone.. -j4 consumes more ram rather than -j2 ????
2020-09-01 18:16:19 <c705> no idea what you're talking about
2020-09-01 18:16:31 <mckay-trabajo> compiler flags!. please!
2020-09-01 18:17:05 <mps> make flags
2020-09-01 18:17:25 <mps> number of paralel jobs
2020-09-01 18:17:31 <mckay-trabajo> mps:  core usage.. i have 64 pcu but not enouoght ram!
2020-09-01 18:17:42 <mps> yes, sure it will need more RAM
2020-09-01 18:17:45 <mckay-trabajo> so i get many memory exausted!
2020-09-01 18:18:17 <mckay-trabajo> but i not have more money XD mps so i ned to limit the number of paralels jobs!
2020-09-01 18:18:24 <mckay-trabajo> or not?
2020-09-01 18:18:51 <mps> you have to test this yourself and decide
2020-09-01 18:19:03 <mckay-trabajo> ¬_¬u
2020-09-01 18:19:13 <mps> you can add zran swap maybe
2020-09-01 18:19:24 <mps> zram*
2020-09-01 18:19:39 <mckay-trabajo> no.. is a test build.. must work without that.. cos i will make an APKBUILD
2020-09-01 18:22:48 <mps> APKBUILD uses JOBS from /etc/abuild.conf, you don't need to set -j in APKBUILD, except in special cases
2020-09-01 18:23:21 <mckay-trabajo> my special case is my special build environment.. that the case..
2020-09-01 18:23:44 <mckay-trabajo> i only have 16Gb of ram
2020-09-01 18:24:22 <mps> with 8GB I built big pkgs as firefox
2020-09-01 18:24:48 <mckay-trabajo> firefox got memory exhaust with 64 cores.. uummm well i must test.. puff
2020-09-01 18:25:08 <mps> -j8
2020-09-01 18:25:29 <dmb> not enough cores
2020-09-01 18:25:50 <mckay-trabajo> yeah.. not enought
2020-09-01 18:25:57 <mckay-trabajo> 64 is not enought XD
2020-09-01 18:26:31 <mps> some of our builders have just 64
2020-09-01 18:35:01 <opal> ya gonna switch my pi to a disk install lol
2020-09-01 18:35:42 <opal> i think replacing init with s6 / s6-rc, it's kind of the easiest way to go about it just to put it all on disk
2020-09-01 19:31:35 <nacelle> everyime someone does a pi with disk install on microsd a baby kitten weeps, or something like that.
2020-09-01 19:32:40 <opal> what's a separate /var and /home partition
2020-09-01 19:32:59 <opal> if stuff on your system commonly writes to /etc or /usr idk how to help you
2020-09-01 19:35:00 <ikke> opal: are we playing jeopordy?
2020-09-01 19:35:52 <opal> ya
2020-09-01 19:36:52 <opal> i need to get some new hard drives tbh
2020-09-01 19:51:13 <vpseg> Any ideas for traffic shaping for an interface or process? For various reasons, tc and GNU trickle won't work for me.
2020-09-01 19:52:26 <mps> vpseg: describe problem
2020-09-01 19:53:07 <c705> vpseg: no need to cross post between various channels
2020-09-01 19:56:23 <vpseg> mps: I want to limit bandwidth for a specific process. If that is impossible with my constraints, then I want to limit it for a user or an interface. For example, I want to limit a process to 300kbps down, 400kbps up
2020-09-01 19:56:49 <vpseg> c705: I was just trying to get people that wouldn't be in both. Sorry
2020-09-01 19:57:34 <mps> vpseg: what is your problem
2020-09-01 20:01:49 <vpseg> mps: I am trying to limit speed tests to a certain rate
2020-09-01 20:03:18 <opal> that isnt your problem
2020-09-01 20:03:29 <opal> mps is referring to where you said stuff "won't work for" you
2020-09-01 20:03:42 <opal> state your _problem_
2020-09-01 20:03:58 <mps> vpseg: that is what you want, but what is problem to make this
2020-09-01 20:05:31 <vpseg> The main issues include tc affecting the entire interface (which I believe can be solved with classful qdiscs) and that our testing has shown tc is "wildly inaccurate" (I didn't do that research)
2020-09-01 20:05:47 <opal> mm
2020-09-01 20:06:21 <vpseg> Ignoring that issue, I personally can't make it do anything but give me the classic "RNETLINK answers: no such file or directory" on a valid device
2020-09-01 20:06:36 <mps> vpseg: did you read lartc doc, if not search on internet
2020-09-01 20:07:02 <mps> do you use alpine kernel
2020-09-01 20:07:07 <vpseg> I read there, and on Arch Linux's advanced traffic shaping documentation, various other articles, and I stumped someone on here yesterday
2020-09-01 20:07:46 <vpseg> mps: https://hatebin.com/qfrciabwiq
2020-09-01 20:07:48 <ikke> vpseg: why is something like tc not woroking?
2020-09-01 20:07:57 <ikke> oh, you just gave the message
2020-09-01 20:07:58 <vpseg> See that snippet ^
2020-09-01 20:08:04 <vpseg> :)
2020-09-01 20:08:37 <vpseg> mps: uname -a: Linux template-pi 4.14.0 #82 SMP Fri Dec 8 14:33:14 CST 2017 aarch64 Linux
2020-09-01 20:08:46 <ikke> Not sure that I can help, but maybe strace might give information about what device
2020-09-01 20:09:19 <mps> I dislike html paste services
2020-09-01 20:09:36 <vpseg> mps: What do you prefer?
2020-09-01 20:09:52 <mps> vpseg: that doesn't look like alpine kernel
2020-09-01 20:11:01 <mps> your commands looks fine
2020-09-01 20:11:24 <mps> probably this kernel doesn't have all needed modules/options
2020-09-01 20:11:28 <vpseg> ikke: I'll link strace output in a moment
2020-09-01 20:11:56 <opal> "What do you prefer?" something with raw text output
2020-09-01 20:12:15 <mps> ix.io, tpaste.us and similar
2020-09-01 20:12:17 <opal> something we can curl https://whatever | less
2020-09-01 20:12:32 <opal> ix.io blocks tor doesnt it?
2020-09-01 20:12:36 <mps> opal: yes +++1
2020-09-01 20:12:53 <opal> ive had trouble accessing ix.io sadly
2020-09-01 20:12:54 <mps> tor, don't make me laugh
2020-09-01 20:13:08 <vpseg> Here's this for now. strace output: https://termbin.com/r71i
2020-09-01 20:14:06 <opal> im not trying to make anyone laugh
2020-09-01 20:14:11 <mps> that is not needed, 'RTNETLINK answers: No such file or directory' says all
2020-09-01 20:14:38 <vpseg> (in response to ikke)
2020-09-01 20:14:51 <vpseg> mps: What more can I do to troubleshoot?
2020-09-01 20:15:10 <mps> install alpine kernel
2020-09-01 20:18:41 <c705> sorry your kernel is from 2017? time to update
2020-09-01 20:26:35 <mps> vpseg: wait, you are trying to delete non attached qdisc from interface
2020-09-01 20:29:22 <vpseg> mps: Would that explain the failed add command following?
2020-09-01 20:30:00 <mps> yes
2020-09-01 20:30:38 <mps> you have to add/replace it first, than manipulate
2020-09-01 20:31:54 <mps> example 'tc qdisc replace dev br0 root fq_codel'
2020-09-01 20:32:08 <mps> then you can 'del' or 'replace' it
2020-09-01 20:32:36 <vpseg> I'm assuming I replace br0 with eth0
2020-09-01 20:32:59 <mps> yes
2020-09-01 20:33:13 <vpseg> So if I understand correctly, that would replace pfifo_fast with fq_codel (which I thought was supposed to be the default). Why do I have to do this?
2020-09-01 20:33:35 <mps> I think so
2020-09-01 20:33:38 <vpseg> I just updated the kernel, rebooting
2020-09-01 20:34:11 <mps> i'm going out
2020-09-01 20:34:42 <vpseg> Sounds good, thank you for your help today
2020-09-01 21:18:56 <vpseg> I'm trying to compile tcng and I'm getting an error about missing execinfo.h, which is apparently part of gnulib and implements a backtrace method. Many Cygwin users are told to just remove this because it doesn't work in Cygwin. Can I get it on Alpine?
2020-09-01 21:21:32 <c705> vpseg: provided by libexecinfo-dev
2020-09-01 21:22:25 <vpseg> c705: Oh wow. Thank you. Did you just know that or how did you find out?
2020-09-01 21:22:39 <c705> vpseg: https://pkgs.alpinelinux.org/contents
2020-09-01 21:23:06 <vpseg> :o
2020-09-01 21:25:38 <vpseg> Now I'm getting an error from aarch64-alpine-linux-musl/bin/ld: cannot find -lfl
2020-09-01 21:26:11 <vpseg> It's probably basic but I don't know how to fix this either. I'm assuming it's trying to link whatever lfl is?
2020-09-01 21:47:27 <ikke> libfl?
2020-09-01 21:47:47 <ikke> flex-dev
2020-09-01 21:48:04 <ikke> https://pkgs.alpinelinux.org/contents?file=libfl*.so*&path=&name=&branch=edge
2020-09-01 21:48:53 <vpseg> Thank you, flex-dev did it
2020-09-01 21:49:45 <ikke> it's basicaly -l fl
2020-09-01 21:49:51 <ikke> it it looks for libfl.so
2020-09-01 21:51:05 <vpseg> Makes sense. Thank you
2020-09-01 21:55:11 <Hello71> iirc fq_codel is default only on systemd because reasons
2020-09-01 22:12:04 <mps> Hello71: yes, I'm 'playing' cake and pie nowadays
2020-09-01 22:12:35 <mps> codel was from my old script :)
2020-09-01 22:23:38 <baconicsynergy> so, what was the primary motivation for using busybox over the gnu/bsd coreutils? size, license, security? I'm very curious :)
2020-09-01 22:26:25 <u0jQx9gPyrYg> all of them.
2020-09-01 22:30:06 <Cogitri> Mostly size I think
2020-09-01 22:31:11 <Hello71> coreutils and util-linux have many dubious features
2020-09-01 22:31:18 <Hello71> and I don't know of any "bsd coreutils"
2020-09-02 00:06:26 <baconicsynergy> what's something really cool about alpine that most people dont know about?
2020-09-02 00:07:54 <c705>  it's not just a container distribution
2020-09-02 00:14:05 <baconicsynergy> I'm using it as my main laptop driver right now! You are 100% correct
2020-09-02 00:16:56 <baconicsynergy> I'm sure it's a fantastic container distribution though
2020-09-02 00:17:19 <baconicsynergy> How many concurrent alpine containers can theoretically run on a server?
2020-09-02 00:17:32 <baconicsynergy> Let's say we're working with a supercomputer
2020-09-02 00:18:59 <c705> i don;t know. that's not a really important question really
2020-09-02 01:25:24 <Hello71> it depends on what "concurrent" and "run" and "server" mean
2020-09-02 01:38:21 <Hello71> and also "container"
2020-09-02 01:39:26 <Hello71> also, if you're talking supercomputers, also "on"
2020-09-02 01:41:20 <opal> alpine's decent for containers but i think i'd rather see myself build static programs "into" a container from an outside package manager and then run my chosen process tree within that
2020-09-02 01:41:25 <opal> i know there's docker, but... docker
2020-09-02 01:42:25 <opal> docker might actually not be as bad as im thinking it is, but ive heard several counts of "don't use it". mainly it's probably a case of it having reached critical mass and being used for the wrong reasons
2020-09-02 01:43:14 <opal> i used to use alpine for desktop but got tired of some of the hacks required to use it in my setup. plus i wanted a source-based distribution anyway
2020-09-02 01:43:32 <opal> using openbsd for my laptop since i dont need exotic hardware support of linux
2020-09-02 01:44:23 <opal> now im just using alpine for pi and a vps, and some vms/containers hanging out here and there that i'll probably replace eventually
2020-09-02 01:44:50 <opal> < baconicsy>…what's something really cool about alpine that most people dont know about?
2020-09-02 01:44:57 <opal> to answer your question: imo the community isnt shit
2020-09-02 01:45:06 <opal> might not matter to everyone but in my case it matters a lot
2020-09-02 01:45:52 <opal> i dont favour a distro that doesnt feel like it has an approachable set of developers and maintainers
2020-09-02 01:46:58 <opal> gentoo's an exception; i'm not really part of its community but it's a metadistribution anyhow, so i can very trivially maintain my own customisations and scratch my itches without having to worry about mainlining
2020-09-02 02:00:27 <Hello71> the main problem about docker is that it's advertised as and people think it's a useful security boundary
2020-09-02 02:00:49 <Hello71> combined with it encouraging downloading and executing random internet binaries
2020-09-02 02:01:23 <Hello71> it's also kind of heavy, but not *terribly* so
2020-09-02 02:01:47 <Hello71> most of the other complaints are against container architecture, not docker
2020-09-02 02:05:01 <opal> gotcha
2020-09-02 02:05:09 <opal> yeah security's a common point ive seen brought up
2020-09-02 02:09:12 <wsinatra> containers done right can result in vastly simplistic reproducible infrastructure
2020-09-02 02:09:39 <wsinatra> a lot of people abuse them to bodge together things, and manually compile software, which isn't a good use case
2020-09-02 02:10:09 <wsinatra> but if you have to deploy infrastructure at scale they're absolutely delightful
2020-09-02 02:12:24 <Hello71> sure, they're not bad if you control or reasonably trust the contents
2020-09-02 02:12:31 <Hello71> and aren't picky about moderate overhead
2020-09-02 02:13:06 <opal> where's the bulk of docker's overhead?
2020-09-02 02:13:35 <opal> say, comparatively to lxc
2020-09-02 02:14:25 <wsinatra> Trust is a whole other thing. Maintain your own registry for production, including base images, that at least helps. Nothing will be perfect though
2020-09-02 02:17:22 <opal> "It works on my machine!" -- https://www.docker.com/use-cases
2020-09-02 02:17:46 <opal> ohh that was the other thing. this mentality is interpreted negatively by a lot of people
2020-09-02 02:18:04 <opal> "i can't ensure a reproducible environment so i'll just package the environment ;^)"
2020-09-02 02:18:45 <opal> that's a super easy philosophy to take overboard
2020-09-02 02:20:21 <opal> it can be done right, there is merit to "shipping your environment along with your software", but it's no substitute for well-designed software in the first place
2020-09-02 02:22:43 <opal> something i hold against flatpak and the like as well
2020-09-02 02:22:45 <jjuran> What package contains the Clang profiling libraries?
2020-09-02 02:27:24 <Hello71> jjuran: https://pkgs.alpinelinux.org/contents
2020-09-02 02:29:52 <jjuran> Hello71: Sweet, thanks
2020-09-02 02:29:54 <jjuran> It was compiler-rt-static
2020-09-02 02:36:41 <Hello71> uh... static?
2020-09-02 02:46:00 <jjuran> Indeed, I was seeking libclang_rt.profile-x86_64.a
2020-09-02 02:46:41 <jjuran> https://pkgs.alpinelinux.org/contents?file=libclang_rt.profile-x86_64.a&path=&name=&branch=v3.12&arch=x86_64
2020-09-02 05:59:44 <baconicsynergy> has anyone been able to get the arc theme working? It's available in the repos but neither GNOME nor Xfce can find it
2020-09-02 06:00:00 <baconicsynergy> its such a lovely theme
2020-09-02 07:16:15 <abenz> hi, my system crashed while attempting to bring up a qemu vm with gpu passthrough. Not a big deal as those gpus are a bit moody.. however when trying to look into it I figured I would install the "linux-firmware-amd-ucode" package, but this fails
2020-09-02 07:16:25 <abenz> it gets downloaded but fails upon installation
2020-09-02 07:16:43 <abenz> can anyone with amd cpu check if they have this installed successfully (3.12)
2020-09-02 07:19:54 <gjabell> does firefox on edge segfault for anyone else?
2020-09-02 07:20:53 <ikke> Are you using X11 or wayland?
2020-09-02 07:21:02 <gjabell> wayland
2020-09-02 07:21:19 <gjabell> was working fine yesterday, not sure what I updated last night tho
2020-09-02 07:22:30 <gjabell> hmm does seem to be a wayland issue, unsetting MOZ_ENABLE_WAYLAND lets it start
2020-09-02 07:22:37 <ikke> dbus has been disabled, but on wayland, that causes segfaults
2020-09-02 07:23:01 <ikke> https://gitlab.alpinelinux.org/alpine/aports/-/issues/11901
2020-09-02 07:23:17 <gjabell> ah gotcha
2020-09-02 07:23:30 <ikke> starting it with --no-remote would also fix it
2020-09-02 07:25:28 <mps> I think kiss linux maintainer working on patch
2020-09-02 07:38:51 <mps> hmm, no, he is working on va-api fixes
2020-09-02 07:39:01 <mps> https://bugzilla.mozilla.org/show_bug.cgi?id=1662496
2020-09-02 12:19:15 <Thermi> Is there a package that provides gnu libc? I'd need to test if a bug I found in a software is due to being linked against musl or if it's a general programming error
2020-09-02 12:19:40 <ikke> Thermi: no, there is not
2020-09-02 12:19:47 <Thermi> Hmh. Inconvenient.
2020-09-02 12:19:56 <Thermi> ty for the answer though
2020-09-02 12:20:33 <ikke> I'm not sure if that would even work (linking a single package against glibc, while everything else is linked against musl))
2020-09-02 12:21:47 <Cogitri> Nop, that'd blow up
2020-09-02 12:22:09 <Cogitri> You'd need to build the package and all dependencies of it against glibc
2020-09-02 12:24:38 <Thermi> :|
2020-09-02 12:37:08 <Thermi> Is it smart or stupid to put the call to ./configure into prepare() instead of build() (assuming prepare is already in use)
2020-09-02 12:37:39 <PureTryOut[m]> It's wrong mainly, no other APKBUILD does that
2020-09-02 12:37:45 <Thermi> Okay, ty
2020-09-02 12:38:18 <Thermi> (thought it'd be the proper way to do it because strictly speaking, the call to ./configure prepares, but not builds the source, only make and its "clones" does that)
2020-09-02 12:46:28 <ikke> yeah, it's more by convention that configure is run in the build stage
2020-09-02 13:11:40 <shhhum[m]> Cogitri: Hey, I've just installed gnome to test some stuff and while Wayland session works without problems, x11 does not – it results in grey screen with only cursor displaying.
2020-09-02 13:12:14 <shhhum[m]> Is it works for you?
2020-09-02 13:14:49 <Cogitri> Worked on my laptop a few months ago, but I haven't used GNOME Xorg for quite some time now
2020-09-02 13:15:03 <Cogitri> I can test if it still works later today
2020-09-02 13:16:50 <shhhum[m]> I will appreciate it.
2020-09-02 13:16:54 <Cogitri> Probably best to checkout GDM's log in the meantime
2020-09-02 13:24:33 <shhhum[m]> It appears that log does not contain anything useful.
2020-09-02 13:24:39 <shhhum[m]> https://gitlab.alpinelinux.org/alpine/aports/uploads/f474366a425c97d0fb3c0583ef8ed300/greeter.log
2020-09-02 13:26:06 <shhhum[m]> Hmm. Or maybe CRTC gamma stuff is useful after all?
2020-09-02 13:46:53 <dmb> CRTC sounds a lot like crt monitor stuff
2020-09-02 13:50:51 <shhhum[m]> Well, I have 2 monitors – one (HDMI) connected to amdgpu, and another (VGA) connected to integrated card (Intel).
2020-09-02 13:51:57 <Cogitri> Probably unrelated
2020-09-02 13:52:11 <Cogitri> GNOME Shell is dying for some reason
2020-09-02 14:08:23 <fcolista> Was anyone able to use two video cards with Alpine? I've a radeon and an integrated Intel with DisplayPort
2020-09-02 14:08:44 <abenz> I have two amdgpu cards, one for passthrough and one for host
2020-09-02 14:08:52 <fcolista> lshw -C display shows the two display
2020-09-02 14:08:57 <abenz> but I guess you ask about using them together to drive more monitors?
2020-09-02 14:08:59 <fcolista> xrandr --listmonitors does not
2020-09-02 14:09:06 <fcolista> correct abenz
2020-09-02 14:09:15 <fcolista> the integrated videocard is enabled
2020-09-02 14:09:55 <fcolista> I can't see the DP-0 port
2020-09-02 14:11:05 <Cogitri> FWIW I can switch to between my intel iGPU and AMD dGPU with PRIME and my monitor works via the DP out via USB-c and the HDMI out
2020-09-02 14:14:29 <fcolista> Cogitri, so you can't use the DP directly to a monitor right?
2020-09-02 14:23:15 <fcolista> Wondering what to do to enable the DP...seems somehow disabled
2020-09-02 14:35:17 <fcolista> I think that I don't have a module loaded for displayport
2020-09-02 14:54:06 <Cogitri> I can use the DP directly to a monitor
2020-09-02 14:54:20 <Cogitri> Both to a USB-C monitor I have and to a usb-c dock with DisplayPort
2020-09-02 14:54:43 <Cogitri> I don't think there's a specific module for Displayport, that should be in your GPU module
2020-09-02 14:59:29 <ikke> ter
2020-09-02 15:00:27 <fcolista> I'm trying to figure what's the option for i915 to enable display port
2020-09-02 15:03:56 <Cogitri> I don't think you have to supply any option to it 🤔
2020-09-02 16:19:12 <crosbymichael> 2/c
2020-09-02 16:19:36 <crosbymichael> opps, my bad
2020-09-02 16:58:28 <dmb> 2/0*
2020-09-02 18:34:19 <opal> huh
2020-09-02 18:34:40 <opal> fcolista consider identifying on irc with sasl
2020-09-02 18:34:59 <fcolista> ?
2020-09-02 18:36:32 <c705> I thought you had to use sasl now
2020-09-02 18:36:33 <opal> https://freenode.net/kb/answer/sasl - "SASL is a method that allows identification to services (NickServ) during the connection process, before anything else happens - therefore eliminating the need to /msg nickserv identify." it prevents race conditions e.g. joining registered-only channels before you log in, stuff like that
2020-09-02 18:36:37 <opal> https://freenode.net/kb/answer/hexchat
2020-09-02 18:36:46 <opal> c705 for certain vpns like the one youre using, yes
2020-09-02 18:37:01 <c705> ah
2020-09-02 18:37:06 <opal> fcolista i can tell you arent using it because i see a bunch of "- fcolista has identified as fcolista" messages when you join
2020-09-02 18:37:43 <c705> not to mention plaintext passwords flying over the wire (unless you use IRCS I suppose)
2020-09-02 18:37:43 <opal> SASL also has the benefit of being standard, so any irc network that supports it, it can use the same "SASL username" / "SASL password" blanks
2020-09-02 18:37:53 <opal> you can use SASL EXTERNAL to use a client certificate if you wish
2020-09-02 18:38:02 <opal> fcolista uses tls
2020-09-02 18:39:09 <opal> "/msg nickserv" in general is unsafe though, since if you forget which irc network youre on, or a network is misconfigured and someone else uses that nick, well, they get your passwords now
2020-09-02 18:39:34 <opal> speaking from experience on that one
2020-09-02 18:39:39 <opal> got a lot of passwords that way :^)
2020-09-02 18:52:57 <Cogitri> Unfortunately Matrix can't do SASL yet I think
2020-09-02 18:53:51 <opal> i highly suggest not logging into nickserv over matrix at all
2020-09-02 20:09:40 <Cogitri> opal: Well, gotta authenticate somehow
2020-09-02 20:09:49 <opal> over irc :>
2020-09-02 20:10:33 <opal> i dont trust the competency of those who write the matrix-appservice-irc bridge as well as the people who run freenode's instance of it
2020-09-02 20:10:42 <opal> nor do a majority of freenode staff, really
2020-09-02 20:16:59 <c705> Cogitri: SASL is the authentication mechanism that is being suggested
2020-09-02 22:51:31 <raiju> Is there any advantage of running alpine on RAM versus disk, or is the main benefit that you can run it on ram without needing external disks (I.E. from a hardware perspective). Is there much of a performance difference?
2020-09-02 22:51:55 <opal> do benchmarks if youre really worried
2020-09-02 22:52:07 <opal> if you have to ask, likely you dont need a diskless setup
2020-09-02 22:52:32 <c705> well, there are advantages and disadvanages
2020-09-02 22:52:43 <opal> yeah but unrelated to general performance
2020-09-02 22:52:53 <opal> more related to what devices youre using
2020-09-02 22:52:55 <c705> running live makes disk i/o super fast
2020-09-02 22:53:19 <opal> os caches hot files in ram anyway so thats negligible unless your i/o is really slow
2020-09-02 22:53:19 <c705> i suppose modern ssds provide pretrty close equivilents without being volitile
2020-09-02 22:53:44 <c705> or your cache isn't built
2020-09-02 22:54:01 <c705> i run live, but i'd say for most cases, it's more pain than the value you get from it
2020-09-02 22:54:14 <raiju> Mostly just curious, not "worried" about anything. Just trying to learn. Most articles don't really give an in depth reason for why you would use it, just the way it is
2020-09-02 22:54:25 <opal> i see
2020-09-02 22:54:29 <opal> yeah mostly its for embedded
2020-09-02 22:54:53 <c705> I use it so  can destroy my operating system and not have to care
2020-09-02 22:55:43 <raiju> @c705 so what happens to configs/states that happen while running diskless, is it destroyed and reboots fresh or saved in a small ROM or similar?
2020-09-02 22:56:11 <c705> raiju: mostly the latter
2020-09-02 22:56:24 <c705> you can use lbu to setup mostly the same thing, but yeah
2020-09-02 22:56:36 <opal> yeah lbu saves packages and /etc if im not mistaken
2020-09-02 22:56:59 <opal> and a package cache saves your system from having to download packages from network every reboot
2020-09-02 22:57:06 <c705> \/etc by default, anything else by a conf change
2020-09-02 22:57:34 <opal> `/ /etc` should work to escape slash on irssi
2020-09-02 22:58:58 <raiju> Ahhh excellent, that's really cool!
2020-09-02 23:59:14 <baconicsynergy> anyone else running GNOME on alpine? :)
2020-09-03 00:00:36 <afontain_> Cogi tri without space
2020-09-03 00:00:46 <afontain_> (I'm avoiding the ping)
2020-09-03 01:17:31 <maldridge> clearly you should use unicode confusables
2020-09-03 01:17:58 <opal> clearly you should yolo and ping them anyway
2020-09-03 01:18:16 <opal> bet their ears are tingling
2020-09-03 03:48:48 <hbug> i can't find out where the standard gnu inetutils ftp client is packaged.
2020-09-03 03:49:34 <hbug> there is lftp but it's learning curve is unacceptable for such a simple protocoll
2020-09-03 04:27:38 <JakeSays> anyone aware of an alpine port to the jetson nano?
2020-09-03 09:01:52 <afontain_> I believe this issue is fixed https://gitlab.alpinelinux.org/alpine/aports/-/issues/11895
2020-09-03 09:07:39 <mps> afontain_: yes
2020-09-03 09:08:16 <afontain_> maybe you could close it
2020-09-03 09:13:06 <mps> ofc, will do later
2020-09-03 09:42:41 <telmich> In some (older, but fully upgraded?) alpine installations wpa_supplicant is started after networking, which makes address acquisition hanging
2020-09-03 09:42:48 <telmich> What's the right way to modify the boot order?
2020-09-03 09:44:06 <mps> before or after in conf.d or init.d scripts
2020-09-03 10:04:33 <abenz> whats the name of the package that contains spice-client (I'm running a kvm in my alpine box, and I want to access that kvm vm with spice client)
2020-09-03 10:04:45 <abenz> I do apk search spice-client, I get spice-gtk as a result, so I install that
2020-09-03 10:05:10 <abenz> still not getting the client though (spice-client not found)
2020-09-03 10:10:44 <abenz> oh, so I need to install virt-viewer, and the binary is called "remote-viewer"
2020-09-03 10:12:08 <mps> afontain_: closed bug about installkernel
2020-09-03 10:13:28 <ikke> Was that about the missing checksum?
2020-09-03 10:15:16 <mps> ikke: yes
2020-09-03 10:15:43 <ikke> ok
2020-09-03 11:06:01 <Hund> I cannot get the date and time to sync in Alpine.
2020-09-03 11:06:26 <Hund> `date` shows me it's "Fri May 29 12:06:11 CEST 2020". It's currently 13:06 here now.
2020-09-03 11:06:31 <Hund> It's not even the right date?
2020-09-03 11:08:21 <ikke> Hund: is there an ntp service running?
2020-09-03 11:08:27 <Hund> openntpd.
2020-09-03 11:08:50 <ikke> ok, not familiar with it
2020-09-03 11:09:13 <Hund> me neither
2020-09-03 11:09:18 <ikke> What does `ntpd -n` return?
2020-09-03 11:09:34 <Hund> configuration OK
2020-09-03 11:10:05 <ikke> and grep server /etc/ntpd.conf
2020-09-03 11:11:09 <Hund> servers pool.ntp.org
2020-09-03 11:11:48 <ikke> `ntpd -s -d` should sync the time now
2020-09-03 11:16:24 <Hund> https://dpaste.org/2srp/slim
2020-09-03 11:16:51 <Hund> I waited a few minutes before killing it. I don't know why it got stuck there.
2020-09-03 11:21:10 <ikke> grep constraints /etc/ntpd.conf
2020-09-03 11:23:47 <mps> install chrony instead of busybox ntpd
2020-09-03 11:24:08 <mps> it is better and more secure
2020-09-03 11:24:23 <ikke> this is openntpd?
2020-09-03 11:24:40 <mps> huh
2020-09-03 11:25:09 <ikke> Hund mentioned they use openntpd
2020-09-03 11:25:41 <Hund> mps: I used to run that but since it didn't work I wanted to try something else.
2020-09-03 11:25:52 <mps> rephrasing: install chrony instead of openntpd
2020-09-03 11:27:15 <mps> long time ago stopped to use openntpd because it was always problematic for me
2020-09-03 11:28:11 <MY-R> ^ agree
2020-09-03 11:28:59 <Hund> How did I switch back again? I forgot already..
2020-09-03 11:29:36 <MY-R> I'm using chrony as server and on everything what can and never got any problem and times are much more accurate, openntpd was problematic for me too
2020-09-03 11:30:18 <Hund> It was setup-ntp.
2020-09-03 11:30:20 <Hund> So.
2020-09-03 11:30:22 <Hund> Now what?
2020-09-03 11:30:41 <Hund> How do I sync the clock with chrony?
2020-09-03 11:30:48 <MY-R> uninstall it and install what want
2020-09-03 11:31:21 <MY-R> /etc/chrony/chrony.conf
2020-09-03 11:32:07 <Hund> It only installed the packages it removed.
2020-09-03 11:32:18 <Hund> What should I change in the file?
2020-09-03 11:33:13 <ikke> I think the default should work in most cases
2020-09-03 11:33:29 <Hund> What can I do to trigger a sync?
2020-09-03 11:33:48 <ikke> You could add initstepslew <value> to make it step in one go if it's of by more then the specified value
2020-09-03 11:33:57 <MY-R> default will point to close enough ntp servers so, you can change to use your country ones if want
2020-09-03 11:34:26 <ikke> You can also run chronyc makestep to make it skip time in one go
2020-09-03 11:35:53 <MY-R> Hund, rc-service chronyd start
2020-09-03 11:36:08 <ikke> rc-update add chronyd as well
2020-09-03 11:36:44 <MY-R> and to check which servers are in use: chronyc sources
2020-09-03 11:36:57 <Hund> https://dpaste.org/depF/slim
2020-09-03 11:37:26 <ikke> chronyc makestep
2020-09-03 11:38:12 <MY-R> or add to chrony.conf: makestep 1.0 3
2020-09-03 11:39:01 <mps> pool pool.ntp.org iburst, in /etc/chrony/chrony.conf
2020-09-03 11:39:25 <MY-R> ye mps that is default
2020-09-03 11:40:44 <MY-R> "initstepslew 10 pool.ntp.org" dont need if using "makestep 1 3"
2020-09-03 11:40:46 <Hund> I added "makestep 1.0 3" and restarted the service, it's still say may.
2020-09-03 11:41:21 <MY-R> Hund, correct your bios date and be sure that is UTC time
2020-09-03 11:41:22 <Hund> Okey.
2020-09-03 11:41:46 <ikke> How would that matter?
2020-09-03 11:42:20 <MY-R> it would be faster probably :D
2020-09-03 11:42:25 <Hund> A Raspberry Pi doesn't have a "BIOS" and I don't know why I would need to change anything there. I have never done so in the past.
2020-09-03 11:42:28 <MY-R> ahh
2020-09-03 11:42:30 <MY-R> no rtc too
2020-09-03 11:42:39 <MY-R> so ignore me :P
2020-09-03 11:42:41 <ikke> Hund: what's the output of chronyc sources?
2020-09-03 11:43:15 <Hund> https://dpaste.org/UiAp/slim
2020-09-03 11:43:39 <MY-R> so all fine
2020-09-03 11:44:52 <mps> restart service after change conf
2020-09-03 11:45:08 <Hund> I did.
2020-09-03 11:45:09 <ikke> Hund: and chronyc tracking?
2020-09-03 11:45:20 <MY-R> or check if previous ntp isnt running in background
2020-09-03 11:45:41 <Hund> https://dpaste.org/GUyJ/slim
2020-09-03 11:45:55 <Hund> It's not.
2020-09-03 11:46:12 <MY-R> looks good
2020-09-03 11:46:16 <MY-R> restart Pi :D
2020-09-03 11:46:18 <Hund> It works now. I might just needed some time.
2020-09-03 11:46:30 <MY-R> good, enjoy!
2020-09-03 11:46:41 <Hund> Thank you for your time! :)
2020-09-03 11:46:45 <ikke> :D
2020-09-03 11:46:50 <MY-R> restart Pi to be sure :D
2020-09-03 11:47:11 <ikke> don't forget to lbu commit, otherwise, you go back in time ;-)
2020-09-03 11:47:20 <Hund> I'm never going to restart it. ;P
2020-09-03 11:47:35 <MY-R> get rtc module!
2020-09-03 11:47:36 <Hund> I added an alias for that so I can save every change quick and easy. :)
2020-09-03 11:48:25 <MY-R> ahh 24/7 so dont need rtc clock
2020-09-03 14:49:41 <Trilby> Does anyone know the status of the alpine forums announced in the post linked below? The URLs in that post are dead links: https://www.alpinelinux.org/posts/Alpine-Linux-forums.html
2020-09-03 14:50:10 <ikke> Trilby: the forums have been discontinued
2020-09-03 14:50:54 <Trilby> That's a shame - thanks for the answer.
2020-09-03 18:26:27 <opal> bring back newsgroups
2020-09-03 18:27:00 <shhhum[m]> Cogitri: Sorry to bother you once again, but did you test GNOME with x11 session?
2020-09-03 20:00:19 <TBB> I've actually been thinking about that, opal
2020-09-03 20:01:06 <TBB> how is it that we had a fully functioning conversation system already decades ago, but we let it go and chose to mess everything up by using lame web forums instead?...
2020-09-03 20:01:07 <mps> I would salute that
2020-09-03 20:02:28 <mps> TBB: modernism supported by corporations marketing
2020-09-03 20:02:48 <TBB> and instead of even improving the web forums, more and more of them just move to Facebook, the shitties conversation system of them all
2020-09-03 20:03:24 <mps> one can't have its (false) photo on newsgroup
2020-09-03 20:04:03 <ikke> IRC is getting out of fashion as well
2020-09-03 20:04:40 <TBB> well, IRC does have a problem with requiring a full time connection so you won't miss anything
2020-09-03 20:04:48 <maldridge> mps: eternal september of the corporate open source project
2020-09-03 20:05:25 <ikke> TBB: but scrolling to backlog does not take an hour :P
2020-09-03 20:05:40 <opal> TBB / mps: nntp is actually something im planning to set up once i get all my other services squared away, so i'm gonna try to revive that medium somewhat
2020-09-03 20:06:03 <TBB> I remember when this shit started for me personally
2020-09-03 20:06:04 <opal> i wish i could say i cared about irc but imo it's going out of fashion rightfully so
2020-09-03 20:06:12 <opal> i'd rather xmpp be improved
2020-09-03 20:06:16 <opal> and mucs actually work well
2020-09-03 20:06:46 <TBB> I was working for a telecom who wanted to implement an internal discussion forum. We suggested an NNTP server; the incompetent fucks in charge wanted it built on top of some Oracle monstrosity
2020-09-03 20:06:50 <opal> nntp, the protocol is alright but the clients for it are probably cumbersome for new users
2020-09-03 20:07:36 <mps> opal: do popular mail clients still support nntp
2020-09-03 20:07:54 <TBB> opal, that could be addressed, fortunately. it could even be a web frontend, they do have their advantages such as easy to upgrade and such
2020-09-03 20:09:38 <opal> mps: does popular equal good? i.e. are you asking the right question?
2020-09-03 20:09:47 <TBB> (those some incompetent fucks funneled money to their own external partners by basically ignoring every recommendation by the in-house tech experts; they even got busted but with no consequences other than all techs fleeing to other companies)
2020-09-03 20:09:47 <opal> thunderbird is popular, it's steaming shit, but yes it supports nntp
2020-09-03 20:09:55 <opal> claws-mail is a decent gui client and supports it as well
2020-09-03 20:11:12 <opal> TBB businesses dont value technical and security consulting despite our world quickly moving in the direction where it matters, more at 11
2020-09-03 20:11:15 <opal> its a sad fact
2020-09-03 20:11:30 <opal> cut those corners
2020-09-03 20:11:51 <mps> opal: mutt ;)
2020-09-03 20:12:31 <opal> whats that in response to
2020-09-03 20:12:39 <opal> mutt is quite the antithesis to accessibility
2020-09-03 20:12:47 <TBB> yup, and at some point we'll pay a terrible price for it. one could argue that we already do in the form of systematic privacy erosion
2020-09-03 20:12:47 <opal> i thought that was our concern with this subject
2020-09-03 20:14:04 <opal> i think the first question to answer is "how do we make mail more appealing to people" then we can figure out nntp easier
2020-09-03 20:14:29 <opal> because at this stage people just think of mail as "that thing where i have 100k unread messages"
2020-09-03 20:15:16 <opal> and thats whats killing better adoption of it. people feel like theyre forced to use it rather than having a comfortable workflow with it
2020-09-03 20:15:27 <opal> i could say similar about snail mail honestly, large volume of it is junk
2020-09-03 20:15:34 <opal> makes it unpleasant even to have a mailbox
2020-09-03 20:34:15 <Gnjurac> hi i want to install alpine on rpi4 to host some stuff, does alpine not have nfs(want to share a dist in lan) or is it named diferent?
2020-09-03 20:35:20 <ikke> https://pkgs.alpinelinux.org/packages?name=nfs*&branch=edge
2020-09-03 20:35:53 <mps> Гњурац ?
2020-09-03 20:36:14 <mps> Gnjurac: yes, it have
2020-09-03 20:36:21 <Gnjurac> oh searhing with no * on end it shows nothing
2020-09-03 20:36:43 <ikke> yeah, it's an exact search by default
2020-09-03 20:36:51 <Gnjurac> dident know that
2020-09-03 20:37:29 <Gnjurac> Србија до Токија :)
2020-09-03 20:37:38 <mps> heh
2020-09-03 20:38:01 <mps> but we should use English here, I was just curious
2020-09-03 20:38:10 <ikke> :)
2020-09-03 20:38:30 <mps> ikke: you translated it? :D
2020-09-03 20:38:46 <ikke> no
2020-09-03 20:39:03 <mps> then better don't :)
2020-09-03 20:40:32 <Gnjurac> :)
2020-09-03 20:46:23 <Gnjurac> for rpi4 i should just extract this to sdcard right?
2020-09-03 20:47:35 <mps> you need separate boot partition iirc
2020-09-03 20:47:36 <Gnjurac> rpi one is not iso
2020-09-03 20:47:51 <mps> but I don't have RPi4 so not sure
2020-09-03 20:48:29 <Gnjurac> so fat32 /boot 1gb and hmm second / restgb
2020-09-03 20:48:42 <mps> I have script to unpack/prepare rpi zero w
2020-09-03 20:48:57 <mps> with serial console, not hdmi
2020-09-03 20:49:34 <Gnjurac> what you running form pi zero, i have both pi zero w and pi4 but tough pz is too weak for anyhting
2020-09-03 20:49:49 <Gnjurac> pzero will use as kodi hopfully
2020-09-03 20:50:02 <mps> I only test linux releases, nothing more
2020-09-03 20:50:10 <Gnjurac> oh
2020-09-03 20:50:41 <mps> also when have time trying to make mainline kernel to run on it for alpine next stable
2020-09-03 21:06:16 <mps> warning for all users: please do not install linux-edge to arm32 boards, it will not boot
2020-09-03 21:09:51 <yann-kaelig> Hi. I'm curious to know where this sort of application should be installed in my system. https://github.com/Miso-K/lxd-api-gateway The install script ( which is written for Ubuntu install it to /home/ubuntu ) but I wonder if it's really the better place and if there is no other dir for this purpose in particulary on Alpine Linux
2020-09-03 21:16:15 <CcxWrk> Hrm, my Xorg is segfaulting as of yesterday. Any known issue, perhaps with intel driver?
2020-09-03 21:16:46 <CcxWrk> I'm on edge.
2020-09-03 21:21:46 <matthewcroughan> Hey again!
2020-09-03 21:22:12 <matthewcroughan> Anyone around to help me understand alpine's repos a bit better? :D
2020-09-03 21:22:27 <matthewcroughan> If I want to add and maintain something, how would I do it?
2020-09-03 21:23:19 <matthewcroughan> I go to https://gitlab.alpinelinux.org/, and aports right?
2020-09-03 21:23:36 <mps> https://wiki.alpinelinux.org/wiki/Category_talk:Developer_Documentation
2020-09-03 21:29:36 <matthewcroughan> mps: holy crap the aports git is so big my webpage just can't deal with it :D
2020-09-03 21:29:40 <matthewcroughan> gonna have to clone this haha
2020-09-03 21:30:44 <mps> yes, clone is better
2020-09-03 21:30:53 <matthewcroughan> mps:  I cannot find ffmpeg in here though, can you help me find it?
2020-09-03 21:31:15 <matthewcroughan> I see fprobe, for example, which somehow uses ffmpeg
2020-09-03 21:31:17 <mps> apk search ffmpeg
2020-09-03 21:31:24 <matthewcroughan> ?
2020-09-03 21:31:29 <matthewcroughan> No I mean in the aports on gi
2020-09-03 21:31:30 <matthewcroughan> git*
2020-09-03 21:31:51 <mps> community/ffmpeg
2020-09-03 21:31:58 <matthewcroughan> Oh I see!
2020-09-03 21:32:48 <matthewcroughan> mps: Do you think a patch to compile ffmpeg with omx would be accepted into aports?
2020-09-03 21:32:55 <matthewcroughan> or do you think this is too obscure?
2020-09-03 21:33:17 <mps> omx?
2020-09-03 21:33:18 <matthewcroughan> The reason for this is because the raspberry pi needs this to encode, and it would greatly improve my life if I did not have to compile ffmpeg :D
2020-09-03 21:33:36 <matthewcroughan> yeah, -enable-omx --enable-omx-rpi
2020-09-03 21:33:59 <mps> what is omx?
2020-09-03 21:34:20 <matthewcroughan> It's very..
2020-09-03 21:34:23 <matthewcroughan> I can't even google it lmfao
2020-09-03 21:34:36 <matthewcroughan> It's the way the Raspberry Pi's GPU is leveraged to encode video with ffmpeg.
2020-09-03 21:35:13 <matthewcroughan> Ah! OpenMAX
2020-09-03 21:35:18 <mps> is it free and open?
2020-09-03 21:35:49 <matthewcroughan> > OpenMAX (Open Media Acceleration), often shortened as "OMX", is a non-proprietary and royalty-free cross-platform set of C-language programming interfaces.
2020-09-03 21:36:38 <mps> source is free to download and use?
2020-09-03 21:37:03 <mps> i.e. not some cumbersome 'fake-free' license?
2020-09-03 21:37:27 <Gnjurac> so hmm if i want to self host some stuff, 2-3 dockers, should i put nginx as normal install for reverse proxy to dockers or should i have ngnix in each docker container?
2020-09-03 21:37:36 <matthewcroughan> mps: source for what exactly?
2020-09-03 21:37:48 <matthewcroughan> The thing FFMPEG uses to implement OpenMAX ?
2020-09-03 21:38:12 <matthewcroughan> this is just a compile time option to use OMX inside of FFMPEG, what exactly is it that needs to be downloaded?
2020-09-03 21:38:29 <mps> matthewcroughan: this is just configure switch in ffmpeg?
2020-09-03 21:38:36 <matthewcroughan> yeah it's a compile time option
2020-09-03 21:38:42 <matthewcroughan> --enable-omx --enable-omx-rpi
2020-09-03 21:38:42 <mps> aha
2020-09-03 21:39:09 <mps> then you can create merge request and wait
2020-09-03 21:40:27 <matthewcroughan> mps: hmm, can I build ffmpeg using the APKBUILD?
2020-09-03 21:40:33 <mps> Gnjurac: that depends on your use case
2020-09-03 21:40:40 <mps> matthewcroughan: sure
2020-09-03 21:40:59 <mps> read url I posted above
2020-09-03 21:41:20 <Gnjurac> i think its better cuz this way i will have mostly updated nginx
2020-09-03 21:41:40 <mps> Gnjurac: I use haproxy for reverse proxies
2020-09-03 21:42:00 <Gnjurac> insted dockerized ones, ye i think i will use ngxni or haproxy infort of dockers
2020-09-03 21:42:14 <matthewcroughan> mps: the docs keep telling me to refer to "making a package"
2020-09-03 21:42:20 <matthewcroughan> but I do not want to make a package, I want to build an APKBUILD
2020-09-03 21:42:23 <matthewcroughan> https://wiki.alpinelinux.org/wiki/APKBUILD_Reference
2020-09-03 21:42:35 <mps> Gnjurac: yes I do same but with LXC
2020-09-03 21:42:43 <matthewcroughan> ah okay, sorry, I missed the first line, which links to abuild
2020-09-03 21:43:19 <mps> matthewcroughan: np
2020-09-03 21:43:53 <mps> if you want to work with building/tweaking/making pkgs this guide is all we have
2020-09-03 21:44:11 <matthewcroughan> mps: so, would the workflow be: Run an Alpine Docker Container. Make changes to APKBUILD. Get it working. Submit PR?
2020-09-03 21:44:37 <matthewcroughan> get it working via abuild, then submit PR
2020-09-03 21:44:54 <mps> well, mostly. and not PR but MR, we are using gitlab at gitlab.a.o
2020-09-03 21:52:39 <matthewcroughan> mps: the ffmpeg apkbuild gives me unsatisfiable constraints
2020-09-03 21:52:41 <matthewcroughan> any ideas?
2020-09-03 21:52:48 <matthewcroughan> vidstab-dev (missing):
2020-09-03 21:53:11 <matthewcroughan> it requires an edge package, OH I am not checked out in the right branch
2020-09-03 21:53:24 <mps> heh
2020-09-03 21:54:42 <mps> Experience is most brutal teacher my dear, but you will learn, you will learn dear - Alice in wonderland, Lewis Caroll :)
2020-09-03 21:55:17 <matthewcroughan> mps: so is master branch edge?
2020-09-03 21:55:28 <mps> yes
2020-09-03 21:56:04 <Gnjurac> so hpaproxy now supports sitatic files too?
2020-09-03 21:56:12 <Gnjurac> serwing
2020-09-03 21:56:50 <mps> Gnjurac: never tried, but it have these error pages so maybe
2020-09-03 21:57:16 <Gnjurac> duno if i should nginx it or hpa it
2020-09-03 21:57:27 <Gnjurac> i like newr shiny stuff
2020-09-03 21:58:06 <matthewcroughan> mps: what do you think is more likely? Do you think I will get this merged into ffmpeg, or do you think there will be a `ffmpeg-raspi` package instead?
2020-09-03 21:58:09 <mps> but you can use darkhttpd for simple static pages
2020-09-03 21:58:54 <Gnjurac> i need some proxy to dockers anywya
2020-09-03 21:59:07 <mps> matthewcroughan: I'm not sure, I have no experience with RPis (bad broadcom based things)
2020-09-03 21:59:25 <Gnjurac> (bad broadcom) ?
2020-09-03 21:59:44 <matthewcroughan> Yes, so do you think we are unlikely to get the bad broadcom support merged into the main `ffmpeg` package on alpine?
2020-09-03 21:59:57 <matthewcroughan> maybe ikke could comment? :D
2020-09-03 22:00:51 <mps> matthewcroughan: you can create bug/issue/whish report on gitlab.a.o and ask other people and developers what they think
2020-09-03 22:01:03 <matthewcroughan> mps: I cannot figure out how to get abuild to use all my cores
2020-09-03 22:01:08 <matthewcroughan> it seems to only be using 1 core
2020-09-03 22:01:30 <mps> Gnjurac: broadcom is too closed
2020-09-03 22:02:00 <mps> there are a lot more open ARM SBCs on market
2020-09-03 22:02:09 <Gnjurac> duno i dont know any
2020-09-03 22:02:22 <Gnjurac> RiscV?
2020-09-03 22:02:25 <matthewcroughan> this is going to take forever on one core on an old x5650 xeon 24 thread server hahaha
2020-09-03 22:02:28 <matthewcroughan> noooo
2020-09-03 22:02:47 <mps> Gnjurac: I'm talking about ARM boards
2020-09-03 22:03:19 <Gnjurac> and anyway whats point if its closed when i even if its open you dident build it so no way to know whats dumped in it
2020-09-03 22:03:23 <mps> and yes, I would like to have riscv and usable one
2020-09-03 22:03:31 <matthewcroughan> Please, someone help me, this can't be the case :D
2020-09-03 22:03:42 <matthewcroughan> I am reading at 10,000 MPH, how do I parallelise the build
2020-09-03 22:03:49 <Gnjurac> i would like to have silicon fabric :)
2020-09-03 22:03:58 <Gnjurac> *fabrick
2020-09-03 22:04:29 <Gnjurac> *factory
2020-09-03 22:04:32 <mps> Gnjurac: broadcom drivers are closed and not much is upstreamed in mainline kernel
2020-09-03 22:04:35 <Gnjurac> duno what i am talking anymore
2020-09-03 22:04:53 <Gnjurac> what ARMs are better
2020-09-03 22:04:55 <Gnjurac> RK999
2020-09-03 22:05:16 <u0jQx9gPyrYg> matthewcroughan: just edit the APKBUILD and add -j24 to the lines where make is called
2020-09-03 22:05:17 <mps> nowadays in latest kernel there is improvement for RPis but still not polished
2020-09-03 22:05:28 <matthewcroughan> u0jQx9gPyrYg: is this really it though?
2020-09-03 22:05:29 <u0jQx9gPyrYg> and then remove the -j24 before you commit the MR
2020-09-03 22:05:35 <Gnjurac> Rockchip RK3328 ?
2020-09-03 22:05:41 <matthewcroughan> surely abuild has a way of giving args to make?
2020-09-03 22:06:05 <mps> u0jQx9gPyrYg: /etc/abuild.conf, JOBS=24
2020-09-03 22:06:13 <matthewcroughan> mps: !!!!!!!!!11
2020-09-03 22:06:18 <matthewcroughan> should be in the docs :D
2020-09-03 22:07:15 <mps> Gnjurac: by coincidence I'm typing right now from rk3399 box :)
2020-09-03 22:08:19 <mps> Gnjurac: also Allwinner boards are fine
2020-09-03 22:08:26 <Gnjurac> pinetop?
2020-09-03 22:08:56 <mps> Gnjurac: gru-kevin samsung chromebook, rk3399
2020-09-03 22:09:40 <Gnjurac> i tough chrombooks is some google sypwhere hardware
2020-09-03 22:10:23 <mps> yes, they are till you hack them to make it linux native machines
2020-09-03 22:11:27 <mps> but they are less spywhere than intel CPUs with ME
2020-09-03 22:11:38 <mps> spyware*
2020-09-03 22:12:20 <mps> Gnjurac: btw, are we neighbors?
2020-09-03 22:12:32 <Gnjurac> duno
2020-09-03 22:12:42 <mps> I'm in Ruma
2020-09-03 22:12:54 <Gnjurac> i am from Bg, Cerak
2020-09-03 22:13:08 <mps> not far, nice
2020-09-03 22:15:06 <matthewcroughan> mps: where does abuild put the output? O.o
2020-09-03 22:15:31 <matthewcroughan> oh, by default ~/packages
2020-09-03 22:15:38 <mps> $HOME/packages/
2020-09-03 22:17:47 <matthewcroughan> Alpine's build system is pretty cool on the caching wow
2020-09-03 22:17:51 <matthewcroughan> how the hell
2020-09-03 22:19:35 <mps> ok, too late is here, going to sleep, good night all
2020-09-03 22:20:42 <matthewcroughan> Or it just didn't build t hem because it saw the files already existed haha
2020-09-03 22:20:48 <matthewcroughan> I thought that was way too fast O.o
2020-09-03 22:20:49 <Gnjurac> night
2020-09-03 22:20:54 <matthewcroughan> gnight
2020-09-03 22:22:33 <matthewcroughan> hmm I cannot figure out how to build a local APKBUILD
2020-09-03 22:22:37 <matthewcroughan> Docs are not helpful
2020-09-03 22:22:55 <matthewcroughan> I did not realise that by doing "abuild checksum && abuild -r" it was fetching remotely
2020-09-03 22:23:10 <zoobab> try file:/// ?
2020-09-03 22:23:16 <matthewcroughan>  ?
2020-09-03 22:23:28 <zoobab> let me check
2020-09-03 22:24:11 <u0jQx9gPyrYg> abuild -r does more or less this: abuild fetch deps checksum clean unpack prepare build rootpkg clean undeps
2020-09-03 22:24:26 <u0jQx9gPyrYg> but anyway if you wanna figure it out, abuild is just a shell script you can read and learn
2020-09-03 22:24:41 <matthewcroughan> I am not interested in learning that detail
2020-09-03 22:24:44 <matthewcroughan> I simply want to package
2020-09-03 22:24:50 <matthewcroughan> I am reading, a lot.
2020-09-03 22:24:52 <u0jQx9gPyrYg> abuild -r should do it
2020-09-03 22:24:58 <matthewcroughan> from the current directory?
2020-09-03 22:25:06 <u0jQx9gPyrYg> where the APK BUILD file is
2020-09-03 22:25:08 <matthewcroughan> I cannot find any statement in the docs that suggests that is true
2020-09-03 22:25:46 <matthewcroughan> ah no you are correct
2020-09-03 22:25:53 <matthewcroughan> it does get it from the CD
2020-09-03 22:26:36 <matthewcroughan> it seems somehow my build is not being done with the ./configure args I wanted :D
2020-09-03 22:27:23 <matthewcroughan> Some caching issue I am guessing
2020-09-03 22:28:36 <matthewcroughan> No. For some reason, even after I have reset the whole github repo and cleaned it, it will not build again
2020-09-03 22:29:01 <matthewcroughan> Ah, nevermind. I mistook the output. Since it was purging 180 packages it filled up my terminal and I thought it had completed, when there had in fact been an error.
2020-09-03 22:30:01 <matthewcroughan> Oh no :D ERROR: OpenMAX IL headers from raspberrypi/firmware not found
2020-09-03 22:36:26 <matthewcroughan> OKAY, now I have a snag. u0jQx9gPyrYg
2020-09-03 22:36:26 <matthewcroughan> I need raspberrypi-dev to build this. But that's only available for armv7, armhf and aarch64.
2020-09-03 22:36:41 <matthewcroughan> https://gitlab.alpinelinux.org/alpine/aports/-/tree/master/community/ffmpeg
2020-09-03 22:37:03 <matthewcroughan> I do not understand how this ends up building for *all* architectures?
2020-09-03 22:39:21 <matthewcroughan> Additionally, I cannot find raspberrypi-dev on here https://gitlab.alpinelinux.org/alpine/aports, any help?
2020-09-03 22:43:19 <matthewcroughan> https://gitlab.alpinelinux.org/alpine/aports/-/blob/master/community/omxplayer/APKBUILD
2020-09-03 22:43:36 <matthewcroughan> For example, omxplayer references it
2020-09-03 22:43:36 <matthewcroughan> Additionally, the docs do not detail how one would test for arm?
2020-09-03 22:43:36 <matthewcroughan> but I cannot find an APKBUILD definition for raspberrypi-dev
2020-09-03 22:44:11 <matthewcroughan> https://git.alpinelinux.org/aports/tree/main/raspberrypi/APKBUILD
2020-09-03 22:44:17 <matthewcroughan> OK, I see that this is defined in subpackages=
2020-09-03 22:44:42 <matthewcroughan> Now I just need to understand how I am supposed to test ARM packages on x86
2020-09-03 22:49:13 <matthewcroughan> 0     0    0     0    0     0      0      0 --:--:--  0:00:04 --:--:--     0curl: (6) Could not resolve host: ftp.gnu.org
2020-09-03 22:49:14 <matthewcroughan> ahahhaha
2020-09-03 22:56:42 <matthewcroughan> when trying to run bootstrap.sh armhf I get binutils-armhf (missing):
2020-09-03 23:03:40 <u0jQx9gPyrYg> i don't think this will work on x86, there is no crosscompiling
2020-09-03 23:03:46 <matthewcroughan> that is terrible :d
2020-09-03 23:03:58 <matthewcroughan> so I have to do this on an actual raspberry pi?
2020-09-03 23:04:17 <u0jQx9gPyrYg> you have to build it on an arm. or you need to remove the dep on the raspi-dev pkg
2020-09-03 23:04:43 <matthewcroughan> I do not think this will get into the official ffmpeg because of this dep
2020-09-03 23:04:56 <matthewcroughan> I will have to make a new package like ffmpeg-rpi
2020-09-03 23:06:17 <matthewcroughan> oh god I cannot even clone the git repo ahhaha
2020-09-04 07:41:30 <gjabell> does linux-edge support zfs?
2020-09-04 07:45:31 <ikke> gjabell: You'd have to compile the kernel modules yourself
2020-09-04 07:45:44 <gjabell> gotcha, thanks
2020-09-04 07:45:59 <ikke> https://git.alpinelinux.org/aports/tree/main/zfs-lts/APKBUILD
2020-09-04 07:46:06 <ikke> You can set FLAVOR to edge
2020-09-04 07:46:36 <gjabell> how often does linux-edge update?
2020-09-04 07:46:46 <ikke> I think mps updates it regularly
2020-09-04 07:46:57 <ikke> He's quite on top of these things
2020-09-04 07:47:07 <gjabell> ah yeah looks to be pretty often
2020-09-04 07:47:22 <gjabell> cool, thanks
2020-09-04 08:04:11 <mps> yes, updates are usually after new version is released, day or two
2020-09-04 08:05:11 <mps> but current 5.8.x series is problematic on some arches
2020-09-04 08:05:22 <mps> s/is/are/
2020-09-04 08:05:47 <mps> on some arches it doesn't boot
2020-09-04 08:06:27 <mps> problem is in combo of binutils, gcc and kernel
2020-09-04 08:06:43 <mps> patch to fix this is in linux-next
2020-09-04 08:07:19 <mps> I hope it will soon land in stable branch
2020-09-04 10:21:37 <Gnjurac> hi
2020-09-04 10:24:12 <Gnjurac> is thia post install old
2020-09-04 10:25:12 <Gnjurac> cuz from wha i see i allredy have setup  root passwors hostname networking while doing setup alpine
2020-09-04 10:26:44 <Gnjurac> and 1 thing tahat confuses me is this lba commit -d
2020-09-04 10:27:07 <Gnjurac> and after taht it says i can make presistan storage
2020-09-04 10:56:43 <Hkone> Hi
2020-09-04 10:57:29 <Hkone> I've got a problem setting up xorg on Alpine
2020-09-04 10:57:52 <Hkone> Am I in the right place to ask for help?
2020-09-04 10:58:36 <shhhum[m]> Yes.
2020-09-04 10:58:48 <shhhum[m]> Did you run: sudo setup-xorg-base?
2020-09-04 10:59:03 <Hkone> Yes, thats what I did
2020-09-04 10:59:29 <Hkone> but startx says that there's no /dev/dri/card0
2020-09-04 10:59:53 <Cogitri> Probably forgot to load the module for your card/enable udev
2020-09-04 11:00:58 <Hkone> OpenRC says udev is started during boot, but I guess I should check it again
2020-09-04 11:02:58 <Hkone> Also modprobe amdgpu says Invalid argument
2020-09-04 11:03:08 <shhhum[m]> Ah.
2020-09-04 11:03:19 <Cogitri> Make sure to also enable the udev-settle and udev-trigger service
2020-09-04 11:03:35 <shhhum[m]> I think you need to get rid of nomodestting in grub configuration.
2020-09-04 11:03:38 <Cogitri> And udev-postmount
2020-09-04 11:03:49 <ikke> make sure nomodeset is not enabled
2020-09-04 11:04:10 <ikke> it's a kernel argument
2020-09-04 11:04:15 <ikke> it used to be enabled by default
2020-09-04 11:05:03 <Hkone> ok, gonna try to disable nomodeset
2020-09-04 11:09:16 <Hkone> thanks alot, X now starts
2020-09-04 11:14:55 <sebarosanul> I can't install man
2020-09-04 11:15:05 <sebarosanul> Is it not in the repositories?
2020-09-04 11:15:20 <davew> man-pages and mandoc
2020-09-04 11:16:17 <sebarosanul> So mandoc instead of man?
2020-09-04 11:16:29 <PureTryOut[m]> mdocml would also work
2020-09-04 11:16:38 <ikke> instead of mandoc, you can also install `docs`, which will pull in the man pages of all installed packages
2020-09-04 11:16:44 <ikke> mdocml is old
2020-09-04 11:16:55 <yunfan> mps:  hi again
2020-09-04 11:17:05 <ikke> mdocml was renamed to mandoc
2020-09-04 11:17:11 <ikke> " mandoc (formerly mdocml) UNIX manpage compiler toolset"
2020-09-04 11:17:20 <sebarosanul> https://wiki.alpinelinux.org/wiki/Alpine_newbie_apk_packages said to install man and man-pages
2020-09-04 11:17:23 <yunfan> do you know how to use ipv6 route discovery in alpine/
2020-09-04 11:17:42 <yunfan> i am configuring my vps's ipv6 feature
2020-09-04 11:19:31 <ikke> You mean router advertisements / slaac?
2020-09-04 11:21:56 <mps> yunfan: what ikke says, and I think even dnsmasq supports that
2020-09-04 11:22:06 <davew> I'm having some trouble setting up networking on alpine.  I get an ip address via dhcp, but the hostname isn't sent to the router.
2020-09-04 11:22:11 <mps> and there is radvd in alpine
2020-09-04 11:22:55 <ikke> davew: you can add hostname $(hostname) in /etc/network/interfaces
2020-09-04 11:23:16 <davew> If I run udhcpc -x hostname:example then it works
2020-09-04 11:23:19 <mps> yunfan: long time didn't seen you. I hope you are well and your family especially little new programmer :)
2020-09-04 11:23:20 <ikke> mps: not sure if he wants to use it as a client or as a router
2020-09-04 11:23:28 <davew> yes, you'd think so ikke :-/
2020-09-04 11:24:14 <mps> client? loading ipv6 module does this automatically. or I misunderstood
2020-09-04 11:25:22 <davew> I have even tried hostname example rather than $(hostname) and neither work
2020-09-04 11:25:51 <yunfan> ikke: i use static
2020-09-04 11:25:55 <davew> I've also added udhcpc_opts -x hostname:$(hostname) (and also replacing $hostname with the host name)
2020-09-04 11:26:06 <yunfan> ikke: my vps is provided by vultr, they gave a /64 prefix
2020-09-04 11:26:24 <yunfan> but they said use routr discovery on gateway configure
2020-09-04 11:26:56 <yunfan> mps: i am very well, and i am busy on my project and take care my baby girl
2020-09-04 11:27:41 <mps> yunfan: nice to hear
2020-09-04 11:28:15 <mps> yunfan: if you have to run RA on some 'gateway' radvd is quite simple to setup
2020-09-04 11:28:16 <yunfan> the current problem is outside ipv6 request cant be handle by my vps
2020-09-04 11:28:36 <ikke> yunfan: what does ip -6 route return?
2020-09-04 11:28:53 <yunfan> while on the vps, ping -6 google.com says network unreachable, i guess its my lack of route record
2020-09-04 11:30:15 <yunfan> ikke: https://paste.ubuntu.com/p/38vGtg9Xs8/  check this
2020-09-04 11:33:29 <sebarosanul> Should I ise alpine as a desktop or is it best used as a server?
2020-09-04 11:33:38 <sebarosanul> use*
2020-09-04 11:34:19 <PureTryOut[m]> I use it as a desktop perfectly fine, but also as a server. You can use it for whatever you want
2020-09-04 11:34:50 <yunfan> mps: btw, i need to report a problem while using alpine 1.12 of armv7 , many packegs report "Error relocating /usr/bin/nim: __clock_gettime64: symbol not found"
2020-09-04 11:35:02 <yunfan> too many report this
2020-09-04 11:35:15 <sebarosanul> How do you get encryption set up if the install is automatic?
2020-09-04 11:35:17 <ikke> yunfan: means you need to upgrade
2020-09-04 11:35:29 <yunfan> ikke: to which version?
2020-09-04 11:35:39 <yunfan> isnt 1.12 the latest one?
2020-09-04 11:35:56 <ikke> yunfan: did you install packages from edge?
2020-09-04 11:36:02 <ikke> 3.12 is the latest, but I assume you mean that
2020-09-04 11:36:13 <yunfan> ikke: yes, i need neovim and nim
2020-09-04 11:36:23 <yunfan> ikke: yes, is 3.12, sorry
2020-09-04 11:36:29 <ikke> THat's the issue then
2020-09-04 11:36:46 <ikke> we upgraded musl to 1.2
2020-09-04 11:37:09 <yunfan> ikke: aha, after using apk upgrade, it works :D
2020-09-04 11:37:11 <ikke> so packages built against musl-1.2 are not going to work with musl 1.1
2020-09-04 11:37:27 <yunfan> thanks, so how can i solve my ipv6 issue?
2020-09-04 11:39:08 <mps> yunfan: last night I wrote here: if you use linux-edge don't upgrade it to 5.8 series, it will not boot on armv7 probably
2020-09-04 11:39:42 <mps> yunfan: look at radvd
2020-09-04 11:39:48 <mps> for RA
2020-09-04 11:40:02 <ikke> idn't that for sending router advertisements?
2020-09-04 11:40:29 <yunfan> mps: i had configured pre-up  echo 2 to accept_ra
2020-09-04 11:40:29 <mps> ikke: yes
2020-09-04 11:40:41 <yunfan> should i still need to install radvd?
2020-09-04 11:40:55 <ikke> yunfan: did you enable forwarding?
2020-09-04 11:41:03 <yunfan> ikke: yes
2020-09-04 11:41:08 <mps> yunfan: no if provider sends them to your VPS
2020-09-04 11:41:08 <ikke> ok, that that's necessary indeed
2020-09-04 11:41:21 <ikke> mps: shouldn't the kernel handle RA's?
2020-09-04 11:41:47 <mps> ikke: yes, it accept them but don't forward
2020-09-04 11:42:08 <ikke> right, but this is about the vps itself not getting the routes, right?
2020-09-04 11:42:13 <ikke> not about other clients
2020-09-04 11:42:30 <mps> for clients to work nothing is needed to do, except maybe load ipv6 module if not loaded by default
2020-09-04 11:42:31 <yunfan> mps: but my problem is the reverse forward, it could accept ipv6 request, cant send ipv6 ping
2020-09-04 11:42:36 <davew> so, e.g. this, as suggested by ikke, https://paste.ubuntu.com/p/KpgZZH2Hng/, doesn't provide hostname to the router so I can't access it from other machines by name
2020-09-04 11:42:57 <davew> but then I run udhcpc -q -x hostname:$(hostname) and it's there.
2020-09-04 11:43:14 <yunfan> you can check this by viewing https://test.ipv6.tuple.info its a ipv6 only domain host on that vps
2020-09-04 11:43:51 <mps> yunfan: ikke: I have few VMs on different providers, all worked out-of-the-box, I didn't do anything
2020-09-04 11:43:52 <yunfan> what the hell, it just works now ...
2020-09-04 11:43:59 <ikke> This is the default route: 'default via fe80::fc00:2ff:fee6:1c07 dev eth0 '
2020-09-04 11:44:13 <yunfan> after i run rdisc6 eth0
2020-09-04 11:44:17 <yunfan> a while
2020-09-04 11:44:48 <ikke> "RDisc6 is an Unix program which implements the ICMPv6 Router Discovery in userland (it is normally done by the kernel)"
2020-09-04 11:44:57 <mps> fe80 are link-local
2020-09-04 11:45:00 <dne> if you have installed a package from a "tagged" repo, how can you make apk forget/unpin it (i.e I'd like to reinstall from the default repo)?
2020-09-04 11:45:23 <ikke> yunfan: what does `cat /proc/sys/net/ipv6/conf/all/accept_ra_defrtr ` return?
2020-09-04 11:45:52 <yunfan> ikke: its 1
2020-09-04 11:46:21 <mps> yunfan: are you sure your problem is not at the provider setup
2020-09-04 11:47:25 <yunfan> mps: maybe its caused by my previouse configuration which disable accept_ra
2020-09-04 11:47:43 <yunfan> i am recently a little zealot on ipv6
2020-09-04 11:47:44 <mps> ime, alpine as 'client' works 'out of the box' in ipv6
2020-09-04 11:48:05 <yunfan> but the route stuff just got me headache
2020-09-04 11:48:05 <ikke> mps: if you enable forwarding, it disables accept_ra, unless you set that to 2
2020-09-04 11:48:15 <ikke> so there might be other things disabled as well
2020-09-04 11:48:46 <yunfan> mps: now  my vps could got response via ping -6 google.com
2020-09-04 11:49:02 <yunfan> but in my container, it can not
2020-09-04 11:49:09 <mps> yes, check files /proc/sys/net/ipv6/conf/ and if something is changed in /etc/sysctl files
2020-09-04 11:49:24 <ikke> yunfan: I think you need to do something special to enable ipv6 in docker
2020-09-04 11:49:48 <mps> nowadays it is moved in /usr/lib or /usr/share (I forgot that nonsenses)
2020-09-04 11:50:05 <yunfan> ikke: i had enable it, that's why in my container i had saw real ipv6 addr
2020-09-04 11:50:29 <mps> ah, this is dockers, not bare metal emulator
2020-09-04 11:51:16 <yunfan> yep
2020-09-04 11:51:45 <mps> then I can't help, don't have any experience with such setup
2020-09-04 11:52:09 <yunfan> i had delploy a shellinabox container, so i could use my vps's ipv4 add to service v4 world
2020-09-04 11:52:15 <yunfan> :[
2020-09-04 11:52:30 <yunfan> ok will try again
2020-09-04 11:52:41 <mps> good luck :)
2020-09-04 11:52:48 <yunfan> at lease my cubieboard had ipv6 and ddns :D
2020-09-04 13:29:35 <Gnjurac47> am
2020-09-04 13:47:14 <yunfan> ikke: do we have ndppd in alpine?
2020-09-04 13:49:20 <wsinatra> yunfan: a quick apk search doesn't show anything
2020-09-04 13:49:39 <wsinatra> https://pkgs.alpinelinux.org/packages?name=ndppd&branch=edge
2020-09-04 13:49:57 <wsinatra> looks like a no, but you could package if you need/want to :)
2020-09-04 13:50:09 <ikke> Not sure if it's the same, but we do have: https://pkgs.alpinelinux.org/package/edge/testing/armv7/npd6
2020-09-04 13:51:07 <yunfan> i found someone's article about enable ipv6 with vultr and docker
2020-09-04 13:51:20 <yunfan> but he use debian as the host system while i am using alpine
2020-09-04 13:51:31 <yunfan> the only difference is he use ndppd
2020-09-04 13:52:08 <ikke> I think npd6 does the same
2020-09-04 13:52:15 <ikke> Which is what I have been using
2020-09-04 13:52:28 <Gnjurac47> does anyine think puting coin on rpi4 cpu have any effect on colling?
2020-09-04 13:52:43 <ikke> ?
2020-09-04 13:53:15 <yunfan> yes, i think so by reading its introduction
2020-09-04 13:53:17 <wsinatra> https://gitlab.com/Durrendal/WS-Alpine-Packages/-/blob/master/testing/ndppd/APKBUILD no clue if this works, but I ran the AUR package through toAPK
2020-09-04 13:53:33 <wsinatra> should be a decent start if you wanted to take a crack at packaging it
2020-09-04 13:53:53 <wsinatra> it's by no means perfect though
2020-09-04 13:54:46 <Gnjurac47> lol somone allredy did this
2020-09-04 13:54:47 <Hello71> better to use ip n add proxy
2020-09-04 13:54:47 <Gnjurac47> https://www.youtube.com/watch?v=6HIiGMJxu_A
2020-09-04 13:54:54 <Hello71> unless you really need dynamic
2020-09-04 14:10:33 <yunfan> mps: ikke  looks like i had met a bug https://forums.docker.com/t/ipv6-routing-problem/53821/4
2020-09-04 14:11:32 <ikke> aha, interesting
2020-09-04 14:27:09 <Gnjurac47> hmm why i cant install ufw
2020-09-04 14:28:43 <Gnjurac47> i get  ERROR:unsatisfiable constrains ufw (missing) requred bt: world(ufw)
2020-09-04 14:29:20 <ikke> Gnjurac47: Do you have the community repo enabled?
2020-09-04 14:29:30 <Gnjurac47> dont think so
2020-09-04 14:29:41 <Gnjurac47> oh i see
2020-09-04 14:29:46 <Gnjurac47> :)
2020-09-04 14:51:07 <Hkone> How to prevent udhcpc from pausing init while "sending discover" (5 attempts, 3sec each)?
2020-09-04 15:07:41 <lorux> Quid de Alpinephone?
2020-09-04 15:08:21 <lorux> Bien le bonjour Linuxiens!
2020-09-04 15:57:08 <Gnjurac47> bonjour :)
2020-09-04 15:59:04 <Gnjurac47> hmm i end up using nftables but for some reason my ssh or nftabels are not cooperating
2020-09-04 15:59:16 <Gnjurac47> cna i somehow see on what port sshd is running
2020-09-04 15:59:30 <Gnjurac47> i put Port 8022
2020-09-04 15:59:47 <Gnjurac47> but when i try now to connect from other pc it just timeouts
2020-09-04 16:00:05 <Gnjurac47> guess will first tunr off nftables to confirm its not killng me
2020-09-04 16:00:40 <dne> netstat -anp|grep sshd
2020-09-04 16:00:44 <zcrayfish> ^
2020-09-04 16:11:15 <Gnjurac47> wtf
2020-09-04 16:12:01 <Gnjurac47> is says  222 3696
2020-09-04 16:12:18 <Gnjurac47> is says  22 3696
2020-09-04 16:12:44 <Gnjurac47> i put in conf Port 8022 why is it ignoring config
2020-09-04 16:18:45 <zcrayfish> /etc/ssh/sshd_config ?
2020-09-04 16:18:48 <zcrayfish> Port 8022
2020-09-04 16:20:24 <Gnjurac47> yes under all i just added Port 8022
2020-09-04 16:24:05 <dne> have you restarted? ;)
2020-09-04 16:25:03 <Gnjurac47> rc-services sshd restard
2020-09-04 16:25:05 <Gnjurac47> ye 10 times
2020-09-04 16:25:23 <Gnjurac47> do i have to restart whole rpi?
2020-09-04 16:25:32 <dne> nah
2020-09-04 16:26:47 <Gnjurac47> i even turned of nftables rc-services nftables stop
2020-09-04 16:27:11 <Gnjurac47> like wierd why is it engoing my confihg
2020-09-04 16:27:19 <zcrayfish> make sure the sshd was not started with a -p option. perhaps `ps aux | grep sshd'
2020-09-04 16:27:44 <Gnjurac47> maybe i know
2020-09-04 16:27:45 <Gnjurac47> sec
2020-09-04 16:28:51 <Gnjurac47> lol
2020-09-04 16:29:00 <Gnjurac47> was in ssh insted sshd_config
2020-09-04 16:29:30 <zcrayfish> xD
2020-09-04 16:30:09 <mps> `apk add iproute2-ss` and then `ss -ltnp`
2020-09-04 16:59:09 <Gnjurac> yo
2020-09-04 16:59:16 <Gnjurac> i created a ntfable file
2020-09-04 16:59:29 <Gnjurac> but it only works when i run  nft -f path
2020-09-04 17:00:08 <zcrayfish> Is it in /etc/nftables.d/ ??
2020-09-04 17:00:38 <Gnjurac> i am retard again
2020-09-04 17:00:39 <Gnjurac> ahahha
2020-09-04 17:00:44 <Gnjurac> i upt it in /etc/
2020-09-04 17:01:41 <zcrayfish> as long as the default file (/etc/nftables.nft) is there, placing a *.nft file in /etc/nftables.d/ should do what you want =)
2020-09-04 17:02:25 <Gnjurac> .nft
2020-09-04 17:02:31 <Gnjurac> i named my .conf
2020-09-04 17:02:34 <Gnjurac> will rename
2020-09-04 17:08:20 <Gnjurac> so it will load any.nif file?
2020-09-04 17:11:29 <zcrayfish> yeah, any *.nft file in the /etc/nftables.d/ directory will be loaded by default
2020-09-04 18:23:16 <Gnjurac> so nftables to start on boot i should "rc-update add nftables boot"
2020-09-04 18:23:19 <Gnjurac> right?
2020-09-04 18:42:31 <Gnjurac> right right?
2020-09-04 18:44:26 <Gnjurac> or sysinit ?
2020-09-04 18:45:20 <opal> when do you want nftables to start
2020-09-04 18:45:21 <Gnjurac> and why i have acpid crashed
2020-09-04 18:45:34 <Gnjurac> i want to start as soon as as i can
2020-09-04 18:45:47 <Gnjurac> so filrewall is blocking as soon it power ups
2020-09-04 18:45:58 <Gnjurac> guessing before netwokr starts working :)
2020-09-04 18:48:33 <Gnjurac> save me from my own stupididty
2020-09-04 18:50:03 <opal> re acpid check logs
2020-09-04 18:50:25 <opal> re nftables, check what the service depends on
2020-09-04 18:50:45 <Gnjurac> re not found, is re a commnad
2020-09-04 18:51:15 <Gnjurac> and its community :(
2020-09-04 18:54:02 <Gnjurac> acpid: /dev/input/event0: No such file or directory
2020-09-04 18:54:29 <Gnjurac> this i get for log, guessing cuz its headless why should there be one?
2020-09-04 18:54:35 <Gnjurac> or am i reading this wronk
2020-09-04 18:58:22 <Gnjurac> :(
2020-09-04 19:00:30 <opal> imo you should take time seeing how components of your system work together, and how they pertain to your specific use case
2020-09-04 19:00:46 <opal> and not count on people giving you exact commands and instructions for everything
2020-09-04 19:00:57 <opal> especially with firewalls which are very system-specific
2020-09-04 19:01:06 <Gnjurac> where
2020-09-04 19:01:08 <Gnjurac> to go see
2020-09-04 19:01:24 <Gnjurac> my logic was to turn firewall as soon i can
2020-09-04 19:03:14 <Gnjurac> duno whats wrong with that i just asked if i should put it to what run-lvl
2020-09-04 19:04:19 <Gnjurac> boot or default
2020-09-04 19:06:11 <mckayshirou> it is possible to kill PID 1 on alpine?
2020-09-04 19:10:13 <u0_a305> Is there an easy way to mmake a live usb file system not read only? Trying to use lbu. Tried "mount -o remount, rw"
2020-09-04 19:14:51 <mckayshirou> i am not so experienced ( dont believe my answer) but the PC interpretes the USB drive as a CD (that is because thay name is Live CD) and CD´s are read only
2020-09-04 19:15:26 <mckayshirou> remember "dont believe my answer" i am not expert
2020-09-04 19:16:14 <Gnjurac> mount  -o umask=0000
2020-09-04 19:17:26 <mps> mckayshirou: as root you can everything
2020-09-04 19:17:45 <mps> I do `kill -HUP 1` to reload /etc/inittab
2020-09-04 19:19:10 <opal> sending a fatal uncatchable kill signal to pid 1 is unadvised unless you want an unsafe way to halt your system
2020-09-04 19:19:16 <mckayshirou> thanks i may try the secondary effects tonigth
2020-09-04 19:19:38 <mckayshirou> i dont want t halt the system just see what can happen
2020-09-04 19:19:43 <opal> u0_a305 depends on filesystem type
2020-09-04 19:21:50 <mckayshirou> another question: in this group is allowed hating Windows?
2020-09-04 19:23:10 <u0_a305> @opal I just ran the alpine iso through etcher to a usb
2020-09-04 19:23:14 <mps> mckayshirou: you can hate whatever you want but please don't spam this channel. it is for helping our users
2020-09-04 19:24:06 <Gnjurac> mps: can you spoonfeed me for nftables should i start it at boot or default lvl?
2020-09-04 19:25:27 <mckayshirou> ok i understand
2020-09-04 19:26:04 <mps> Gnjurac: I would not start it on boot
2020-09-04 19:27:16 <mps> Gnjurac: even sysinit not good imo
2020-09-04 19:28:30 <Gnjurac> so by system of eleminaiton nonetwork?
2020-09-04 19:28:52 <Gnjurac> nonetwokr or default cuz shotdown its not
2020-09-04 19:29:30 <Gnjurac> where can i see how this runlvls go
2020-09-04 19:29:52 <Gnjurac> i tryed googleing open-rc runlvls but got nothing usfull
2020-09-04 19:29:56 <mps> /etc/rc.conf
2020-09-04 19:32:51 <mps> Gnjurac: boot and sysinit runlevels are to setting system in working state, i.e. load modules, renaming interfaces etc..
2020-09-04 19:33:19 <mps> so, I doubt that setting nftables is ok before its kernel modules are loaded
2020-09-04 19:34:54 <Gnjurac> i am reading /etc/rc.conf atm, if i dont figure out i have a basiment i can kidnap you
2020-09-04 19:35:28 <Hello71> based on your performance so far I doubt you will be able to organize it
2020-09-04 19:35:56 <Gnjurac> dude i did armed roberis as tean, police newer got me for that
2020-09-04 19:36:06 <mps> Gnjurac: /etc/inittab
2020-09-04 19:37:25 <Gnjurac> still dide not found 1 text explaing to me, "look this runlvl is used for this and this for that"
2020-09-04 19:37:33 <Gnjurac> guess i should put to default
2020-09-04 19:37:36 <Gnjurac> and hope for best
2020-09-04 19:38:36 <mps> Gnjurac: there is no explanation, just gives you idea how things works in openrc
2020-09-04 19:39:09 <mps> actually I never tried to learn where order of runlevels are set
2020-09-04 19:39:21 <mps> I'm fine with defaults
2020-09-04 22:20:56 <Gnjurac> hmm if i install docker will my rpi memory go to doom cuz evry img will go to ram?
2020-09-04 22:21:53 <Gnjurac> or can i store docker images on usb/hdd and run them from there
2020-09-05 07:33:46 <ikke> GNUmoon: sure, you can create a separate mountpoint for /var/lib/docker
2020-09-05 07:47:46 <mps> ikke: I think you wrote to wrong nick
2020-09-05 07:50:25 <ikke> yes
2020-09-05 07:50:39 <ikke> Gnjurac already left
2020-09-05 07:54:37 <GNUmoon> ikke: yeah, it wasn't me :)
2020-09-05 07:58:36 <ikke> heh
2020-09-05 08:00:33 <detha> is there an apk incantation to get the list of installed packages plus the repo they came from?
2020-09-05 08:02:40 <ikke> I believe apk does not store that
2020-09-05 08:03:36 <ikke> apk info -a does not return it
2020-09-05 08:04:08 <detha> I was afraid so. Reasoning behind it: if I enable testing, install something, then disable testing the next upgrade leaves the installed package intact, but removes the -doc
2020-09-05 08:04:50 <detha> So, next question was: what is installed here that is currently in testing
2020-09-05 08:14:52 <ikke> detha: not an exact answer to your question, but you can use apk version -l '?' to find packages that are not found in a repo anymore
2020-09-05 08:16:46 <mps> detha: best I can think of is 'apk policy `apk info`'
2020-09-05 08:17:03 <mps> maybe add some grep/sed
2020-09-05 08:17:57 <detha> ah. with some use of -X that for each repository that could give me what I need yes, thnks
2020-09-05 08:39:38 <mous[m]> 
ACTION sent a long message:  < https://matrix.org/_matrix/media/r0/download/matrix.org/lBMlfwMmXSUAcsyABMIpjVgT/message.txt >

2020-09-05 08:39:53 <ikke> " * │ mous[m] sent a long message:"
2020-09-05 08:41:28 <zcrayfish> what happens if you try to start the iscsi daemon directly instead of using the init script?
2020-09-05 08:41:34 <zcrayfish> something like: iscsid --pidfile /var/run/iscsid.pid -w 100 -- -i /etc/iscsi/initiatorname.iscsi
2020-09-05 08:42:30 <ikke> 1) it's harder to manage. 2) You cannot easily restart it. 3) It won't start at boot
2020-09-05 08:42:44 <zcrayfish> I was replying to mous[m]'s long message
2020-09-05 08:42:56 <zcrayfish> basically his iscsi daemon fails to start via the init script
2020-09-05 08:43:12 <zcrayfish> was hoping a better error message would be dumped if he started it directly
2020-09-05 08:43:42 <ikke> zcrayfish: ah, it sounded like a separate question
2020-09-05 08:44:05 <zcrayfish> :)
2020-09-05 08:51:10 <mous[m]> <zcrayfish "something like: iscsid --pidfile"> this synthax didn't work, I modded like this :
2020-09-05 08:52:31 <mous[m]> Another log output :
2020-09-05 08:52:31 <mous[m]> https://paste.ubuntu.com/p/cGTSbdvQrz/
2020-09-05 08:53:01 <ikke> So you are missing kernel modules?
2020-09-05 08:53:39 <mous[m]> Thank you, I am pleasantly surprised by your reactivity.
2020-09-05 08:54:57 <mous[m]> it seems so, but I don't know how to force it.
2020-09-05 08:55:22 <zcrayfish> try "modprobe iscsi_tcp"
2020-09-05 08:57:48 <mous[m]> modprobe: FATAL: Module iscsi_tcp not found in directory /lib/modules/3.10.108
2020-09-05 08:57:48 <mous[m]> lg-bullhead:/home/mous# ls /lib/modules
2020-09-05 08:57:48 <mous[m]> ls: /lib/modules: No such file or directory
2020-09-05 09:05:38 <zcrayfish> custom kernel by chance?
2020-09-05 09:06:29 <mous[m]> yes
2020-09-05 09:06:59 <mous[m]> postmarketOS >linux5X>usb Networking
2020-09-05 09:07:34 <PureTryOut[m]> Well if it's just missing kernel modules, that's easy to resolve
2020-09-05 09:07:57 <PureTryOut[m]> mous: like I mentioned, don't use multiline messages here. This room is a majority IRC room and it won't show up properly on that side
2020-09-05 09:08:28 <ikke> gunzip -c /proc/config.gz | grep CONFIG_ISCSI_TCP
2020-09-05 09:09:17 <mous[m]> <ikke "gunzip -c /proc/config.gz | grep"> no output
2020-09-05 09:09:50 <zcrayfish> it seems to me that your kernel was compiled without that module
2020-09-05 09:09:54 <ikke> yes
2020-09-05 09:09:57 <mps> modprobe configs
2020-09-05 09:10:17 <ikke> mps: I think config.gz exists
2020-09-05 09:10:26 <ikke> otherwise they would have gotten an error message
2020-09-05 09:10:40 <mps> yes, but just in case
2020-09-05 09:10:46 <mous[m]> modprobe configs return > fatal error too
2020-09-05 09:10:54 <mous[m]>  FATAL: Module configs not found in directory /lib/modules/3.10.108
2020-09-05 09:11:06 <ikke> mous[m]: what does uname -r return?
2020-09-05 09:11:29 <mous[m]> 3.10.108
2020-09-05 09:11:49 <ikke> ok, that's good
2020-09-05 09:12:01 <ikke> hmm
2020-09-05 09:12:08 <ikke> does /lib/modules/3.10.108 exist?
2020-09-05 09:12:14 <mous[m]> no
2020-09-05 09:12:17 <ikke> aha
2020-09-05 09:12:36 <ikke> what directories exist in /lib/modules?
2020-09-05 09:12:51 <zcrayfish> 04:57:48       mous[m] ls: /lib/modules: No such file or directory
2020-09-05 09:12:53 <PureTryOut[m]> Not sure why we're still talking about this. The module is clearly not enabled and it needs to be for this to work 🤷‍♂️
2020-09-05 09:13:07 <ikke> PureTryOut[m]: ^^
2020-09-05 09:13:13 <ikke> /lib/modules does not exist?
2020-09-05 09:13:32 <mous[m]> dosent not exist yes
2020-09-05 09:13:56 <mous[m]> PureTryOut: how to load modules in my case?
2020-09-05 09:14:16 <PureTryOut[m]> Once the module is available, just `modprobe <module name>`
2020-09-05 09:14:20 <ikke> mous[m]: are you using some kind of virtual image?
2020-09-05 09:15:00 <ikke> are you using a container or something like that?
2020-09-05 09:15:22 <PureTryOut[m]> Nah he is using postmarketOS which uses custom kernels for most devices
2020-09-05 09:15:28 <mous[m]> ikke: PostmarketOs on smartphone
2020-09-05 09:15:37 <zcrayfish> Looked like an LG Bullhead (Google Nexus 5x)
2020-09-05 09:15:39 <ikke> PureTryOut[m]: I wonder why /lib/modules does not exist?
2020-09-05 09:15:55 <ikke> shouldn't that come with the kernel?
2020-09-05 09:17:01 <PureTryOut[m]> Normally yeah, idk why that is the case for him
2020-09-05 09:17:50 <ikke> apk info -l <kernel-package>
2020-09-05 09:18:09 <ikke> -L, sorry
2020-09-05 09:18:13 <ikke> what is the name of the kernel package?
2020-09-05 09:19:42 <mous[m]> https://paste.ubuntu.com/p/GxcRGf9r9F/
2020-09-05 09:20:51 <mous[m]> <ikke "what is the name of the kernel p"> open-iscsi
2020-09-05 09:26:22 <mous[m]> <PureTryOut[m] "Normally yeah, idk why that is t"> I chose during the installation, "no user interface".Can it have a link?
2020-09-05 09:26:34 <PureTryOut[m]> No
2020-09-05 09:30:35 <mous[m]> from : https://wiki.alpinelinux.org/wiki/Architecture#Module_Loading
2020-09-05 09:30:55 <mous[m]> lg-bullhead:/home/mous# cat /etc/modules
2020-09-05 09:30:55 <mous[m]> af_packet
2020-09-05 09:30:55 <mous[m]> ipv6
2020-09-05 09:32:12 <zcrayfish> the entries in /etc/modules are processed by modprobe; thus if running modprobe manually won't work, adding a module in /etc/modules won't help either.
2020-09-05 09:32:41 <mous[m]> 
ACTION sent a long message:  < https://matrix.org/_matrix/media/r0/download/matrix.org/sVXfAdcNQvehNQasGPJvtyUF/message.txt >

2020-09-05 09:34:21 <mous[m]> all directory are empty
2020-09-05 09:36:39 <PureTryOut[m]> At this point it's a pmOS bug (missing kernel modules), so you could move this conversation back to #main:postmarketos.org mous
2020-09-05 09:38:11 <mous[m]> ok PureTryOut
2020-09-05 09:38:21 <mous[m]> merci for you support
2020-09-05 09:41:01 <Gnjurac> hmm installed nfs-utils
2020-09-05 09:41:34 <Gnjurac> but hmm i have not /etc/nfs.conf or /etc/nfsmount.conf
2020-09-05 09:41:38 <Gnjurac> is that normal?
2020-09-05 09:43:46 <zcrayfish> The only file put in /etc by the nfs-utils apk is /etc/exports
2020-09-05 09:50:56 <Gnjurac> how do i set port for nfs?
2020-09-05 09:52:11 <Gnjurac> nah just turned off firewall but still get mount.nfs: failed to apply fstab options
2020-09-05 09:52:34 <Gnjurac> tough hmm what fstab otpion i am just mounting not edinting fstab
2020-09-05 09:52:49 <Gnjurac> mount 192.168.1.40:/media/hdd1 /media/SHAREDSTORAGE/
2020-09-05 09:56:35 <Naib> hi, I am re-doing my VE's (under CENTOS) but I can't create a new alpine container
2020-09-05 09:56:46 <Naib> ERROR: http://ftp.tsukuba.wide.ad.jp/Linux/alpinev3.12/main: No such file or directory
2020-09-05 09:57:12 <ikke> Naib: sounds like you missed a / between alpine and v3.12
2020-09-05 09:57:13 <Naib> the mirrorlist appears to be wrong, it is missing a / between alpine and the version. any workaround
2020-09-05 09:57:38 <ikke> let me check
2020-09-05 09:57:52 <Naib> oh now it works
2020-09-05 09:58:00 <Naib> so there must have been an error because it now inserts it
2020-09-05 09:58:14 <Naib> lxc-create -n VE_nginx -t alpine
2020-09-05 09:58:19 <Naib> that was thje command executed
2020-09-05 09:58:35 <ikke> Naib: https://mirrors.alpinelinux.org/
2020-09-05 09:58:40 <ikke> We don't offer that mirror
2020-09-05 09:58:52 <Naib> fetch http://dl-3.alpinelinux.org/alpine/v3.12/main/x86_64/APKINDEX.tar.gz
2020-09-05 09:59:14 <Naib> ok ,so it must be centos and its mirror list. for some reason one of them had an issue, missing /
2020-09-05 09:59:21 <Naib> its now pulling from a valid one.
2020-09-05 09:59:27 <Naib> thanks :) back to creating my 3 VE's
2020-09-05 09:59:51 <ikke> What is a VE?
2020-09-05 10:00:48 <Naib> virtual environment
2020-09-05 10:00:51 <ikke> ah
2020-09-05 10:00:52 <Naib> containers basically
2020-09-05 10:01:06 <ikke> 
ACTION mumbles something about containers not being virtual :P

2020-09-05 10:01:30 <Naib> I know... but at work I try to describe containers and literally everyone goes "oh virtual machines" ...
2020-09-05 10:01:36 <ikke> heh
2020-09-05 10:01:48 <Naib> no ... they use the base system and are isolated
2020-09-05 10:01:53 <ikke> just show them the PS output on the host
2020-09-05 10:01:59 <ikke> and see all their processes listed
2020-09-05 10:02:03 <Naib> true
2020-09-05 10:04:56 <Gnjurac> mounting 192.168.1.40:/media/hdd1 failed, reason given by server: No such file or directory
2020-09-05 10:05:32 <Gnjurac> from etc/exports /media/hdd1 192.168.1.50(rw,sync,no_all_squash,root_squash,subtree_check)
2020-09-05 10:12:18 <Gnjurac> got it wokring by adding this fis=0
2020-09-05 10:12:25 <Gnjurac> now need to figure out that that means
2020-09-05 10:19:53 <mps> Naib: teach them and if they don't want to learn beat them :)
2020-09-05 10:20:49 <Gnjurac> pls dont hurt me, i am just a pesant...
2020-09-05 10:20:56 <mps> people should learn to use right terms
2020-09-05 10:34:17 <Gnjurac> mps that rk3399 is too weak for blender right?
2020-09-05 10:50:18 <Gnjurac> dont mean for rendering somingthing crazy am asking for light modeling
2020-09-05 10:50:27 <Gnjurac> tough duno if blender even runs on arm
2020-09-05 10:55:31 <mps> Gnjurac: if you run wayland with panfrost GPU driver it is not bad
2020-09-05 10:56:39 <Gnjurac> tryed it?
2020-09-05 10:56:49 <mps> no, I don't need blender
2020-09-05 10:57:18 <mps> and panfrost is not working fine with Xorg, and I don't like wayland
2020-09-05 10:57:22 <Gnjurac> duno i am personaly w8 for better ARMs or AMD emmbeded they saying they V2000 is going be octa core xD
2020-09-05 10:57:50 <mps> there is honeycomb arm64 board, 16 cores
2020-09-05 10:57:52 <Gnjurac> tough i hope they make some rpi sized one thats 5v so i can make my custom laptop
2020-09-05 10:58:13 <mps> and you can put different GPU boards on it
2020-09-05 10:59:10 <Gnjurac> yep but amd is allinone
2020-09-05 11:01:58 <mps> I think some problems with ARM will change soon. apple switch to use ARM on their new macbooks
2020-09-05 11:02:13 <mps> macbooks*
2020-09-05 11:27:21 <Gnjurac> meybe will see
2020-09-05 11:27:47 <Gnjurac> hmm this lbu commit -d dosent save dir i created for mounting in /mnt/
2020-09-05 11:35:38 <mps> no, you have to configure /etc/lbu/lbu.conf
2020-09-05 11:35:52 <Gnjurac> ok will check
2020-09-05 11:36:29 <mps> don't ask me, I do not use run-from-ram
2020-09-05 11:37:59 <Gnjurac> ye i think this run from ram brings a lot of doom
2020-09-05 11:39:03 <Gnjurac> https://wiki.alpinelinux.org/wiki/Alpine_local_backup
2020-09-05 11:39:09 <Gnjurac> nice found explanation
2020-09-05 11:41:53 <Gnjurac> just duno if running from RAM is waste of time if i cant save dockerimages to some other space cuz guess ram will get eaten fast that way
2020-09-05 11:48:36 <Gnjurac> hmm anyone know if this lba include can i just include 1 dir but nothing inside of it
2020-09-05 11:49:41 <Gnjurac> want to create /mnt/mountpoint << but if i include this and i mount hd to it i guess when i do lbu commit it will try to backup evrything form hd thats mounted at that point
2020-09-05 11:59:43 <ikke> I'm not sure
2020-09-05 12:00:51 <Gnjurac> can i switch to persistan somehow
2020-09-05 12:03:09 <Gnjurac> meybe i can lbu include /mount/hdd
2020-09-05 12:03:19 <Gnjurac> lbu exclude /mont/hdd/
2020-09-05 12:06:42 <Gnjurac> but hmm i think i should switch from diskless
2020-09-05 12:43:47 <Gnjurac> https://www.maxocull.com/2019/12/25/Alpine-Desktop-on-the-Raspberry-Pi-4/
2020-09-05 12:43:51 <Gnjurac> am fallowing this guide
2020-09-05 12:43:54 <Gnjurac> have 1 question
2020-09-05 12:44:09 <Gnjurac> <Your kernel parameters> root=/dev/mmcblk0p2
2020-09-05 12:44:52 <Gnjurac> so when i open cmdline.txt i get modules=loop,squashfs,sd-mod,usb-storage quiet console=tty1  so i should append root=/dev/mmcblk0p2
2020-09-05 12:45:05 <ikke> yes
2020-09-05 12:45:21 <Gnjurac> but my question is should i add    ,    after console=tty1
2020-09-05 12:45:35 <ikke> no
2020-09-05 12:45:43 <Gnjurac> ok ty was not shure
2020-09-05 12:45:58 <Gnjurac> and i would like to not have to attach kayboard agian if it dosent boot up
2020-09-05 12:46:38 <Gnjurac> now lets hope it works
2020-09-05 12:46:42 <Gnjurac> time for rebbot :)
2020-09-05 12:47:48 <Gnjurac> yey
2020-09-05 12:47:51 <Gnjurac> i think it worked
2020-09-05 12:59:35 <Gnjurac> x-systemd.device-timeout=10s
2020-09-05 13:00:09 <Gnjurac> is this for systemd only, for fstab want to make rpi4 boot even if external drive is missing or died
2020-09-05 13:58:40 <charl> Hello, I installed alpine on a laptop that has only a wifi interface but after reboot udhcpc fails to get a DHCP lease. If I start wpa_supplicant manually I can connect to internet. What's the correct way to start it at boot?
2020-09-05 13:59:55 <raiju> auto wlan0
2020-09-05 13:59:59 <raiju> ifacewlan inet dhcp
2020-09-05 14:00:09 <raiju> to /etc/network/interfaces
2020-09-05 14:00:35 <charl> that's what I have
2020-09-05 14:01:06 <raiju> rc-update add wpa_supplicant boot
2020-09-05 14:01:15 <raiju> I just set it up last night and it worked
2020-09-05 14:01:28 <charl> I found this 2 year old open issue https://gitlab.alpinelinux.org/alpine/aports/-/issues/8025
2020-09-05 14:02:15 <charl> @raiju I did that and it worked but I'm not sure if it's the correct way
2020-09-05 14:02:32 <raiju> That's the way the alpine wiki tells you to do it
2020-09-05 14:03:15 <raiju> https://wiki.alpinelinux.org/wiki/Connecting_to_a_wireless_access_point
2020-09-05 14:04:51 <charl> yeah I did that and it works it's just that I get a few lines "sending discover' before I get a lease
2020-09-05 14:05:39 <raiju> Probably not unusual, just like you usually lose because before ARP is performed and tables are updated. Could be wrong though no clue
2020-09-05 14:05:45 <raiju> lose pings*
2020-09-05 14:06:12 <charl> is there a way to put it on background so it can speed up the boot process?
2020-09-05 14:09:19 <raiju> There's a background opt you can try adding udhcpc_opts --background to interface config
2020-09-05 14:09:32 <charl> let me try that
2020-09-05 14:11:20 <raiju> https://unix.stackexchange.com/questions/267733/how-to-get-busybox-udhcpc-to-run-in-background-forever
2020-09-05 14:11:52 <raiju> It's from 16' but he replied to say it had to do with a busybox config. Maybe a good place to start searching
2020-09-05 14:12:26 <charl> it didn't work
2020-09-05 14:13:32 <charl> https://busybox.net/downloads/BusyBox.html#udhcpc
2020-09-05 14:21:49 <raiju> charl: check out the answer here and see if it works for you https://unix.stackexchange.com/questions/202839/overriding-udchcp-flags-used-by-ifup-in-busybox
2020-09-05 14:23:00 <charl> yeah I found that earlier didn't even try.. lets see
2020-09-05 14:25:23 <charl> it seems to work
2020-09-05 14:26:17 <charl> i just need to redirect output to /dev/null because it overwrites my "login: "
2020-09-05 14:29:30 <raiju> Cool that's good to know!
2020-09-05 14:30:24 <raiju> You can do that run in the dchp script likely
2020-09-05 14:30:53 <raiju> /usr/share/udchcpc/default.script
2020-09-05 14:31:28 <charl> "udhcpc_opts -b >/dev/null 2>&1 &"
2020-09-05 14:32:10 <charl> now lets see if upgrading to edge breaks it :p
2020-09-05 14:37:25 <charl> aaand it does
2020-09-05 14:37:38 <charl> I mean it doesn't go on background any more
2020-09-05 14:37:44 <charl> I guess I have to live with that
2020-09-05 15:30:21 <Gnjurac> hmm can i somehow disable rpcbind cuz i am using nfs v4
2020-09-05 15:30:35 <Gnjurac> when i start nfs service it auto runs rpcbind
2020-09-05 15:31:06 <Gnjurac> and if i try to shutdown rpcibind it shutdown nfs too
2020-09-05 15:31:37 <Gnjurac> beh will just close port 111
2020-09-05 16:51:38 <Gnjurac> does my rsync ignore exlcude cuz i am using ash or what?
2020-09-05 16:59:54 <Cogitri> Maybe you didn't quote it correctly or smth
2020-09-05 17:15:24 <raiju> Is it possible to run windows in virtualbox on alpine? Planning to try when I get home unless someone can save me the time
2020-09-05 17:16:30 <Cogitri> Should be possible but if I were you I'd use KVM via virt-manager because it tend to be way faster and less problematic than VirtualBox
2020-09-05 17:17:28 <raiju> ok great I'll look into it. I haven't used any vm software since like 2010 so I didn't know what's popular ha
2020-09-05 17:17:55 <raiju> School needs me to install some windows PLC sim software and I'd rather not dual boot
2020-09-05 17:48:31 <Gnjurac> so this /etc/periodicl/weakly that made for easyer chronejob running
2020-09-05 17:48:49 <Gnjurac> so if i put script inthere it will run or what
2020-09-05 17:53:01 <Gnjurac> guess i will create one named  1min and test myself
2020-09-05 17:53:42 <Gnjurac> https://superuser.com/questions/402781/what-is-run-parts-in-etc-crontab-and-how-do-i-use-it
2020-09-05 17:53:51 <Gnjurac> nice it was like i tought
2020-09-05 20:07:16 <charl> what is the package for apk man page?
2020-09-05 20:07:20 <charl> I'm on edge
2020-09-05 20:08:27 <ikke> mandoc
2020-09-05 20:08:58 <charl> I already have mandoc it's not it
2020-09-05 20:09:31 <charl> man: No entry for apk in the manual.
2020-09-05 20:12:53 <mps> apk add apk-tools-doc ?
2020-09-05 20:13:14 <charl> isn't that for apk-tools which is a different package?
2020-09-05 20:13:50 <charl> yeah that was it thanks mps
2020-09-05 20:16:57 <ikke> charl: you can also do `apk add docs` to automatically get all *-doc packages for installed packages
2020-09-05 20:17:13 <charl> nice one thanks!
2020-09-05 21:30:56 <Gnjurac> hmm
2020-09-05 21:31:20 <Gnjurac> so if i swiched for disklless i should not use anymore apk update
2020-09-05 21:31:26 <Gnjurac> or am i getting this wronk
2020-09-05 21:32:05 <ikke> That's unrelated
2020-09-06 06:56:07 <markand> there are no packages that modifies the /etc/os-release file?
2020-09-06 06:56:14 <markand> after switching to 3.12 it still lists 3.11
2020-09-06 06:56:29 <markand> apk info -W says there is no package that contains this file
2020-09-06 06:58:01 <ikke> https://pkgs.alpinelinux.org/contents?file=os-release&path=&name=&branch=edge&arch=x86_64
2020-09-06 06:58:04 <ikke> markand: ^
2020-09-06 06:58:50 <markand> hmmm, apk info -L alpine-base on mine now lists 0 files
2020-09-06 06:58:56 <markand> 🤔
2020-09-06 06:59:11 <ikke> apk fix alpine-base?
2020-09-06 06:59:43 <markand> I think it's not the first time where I upgrade and suddenly the package database says for some package that there aren't files associated with
2020-09-06 07:00:04 <ikke> yeah, haven't seen that either
2020-09-06 09:33:19 <afontain_> Cogitri: apparently, Games 3.37.0 depends on retro-gtk-1
2020-09-06 09:33:36 <afontain_> the thing is, it doesn't seem to exist upstream either https://gitlab.gnome.org/GNOME/retro-gtk
2020-09-06 09:35:16 <Cogitri> Ah right, I think I hit that too, maybe we need a git snapshot of retro-gtk?
2020-09-06 09:35:50 <Cogitri> But it's not too much of a problem if we don't get to bumping GNOME Games just now, only Nautilus pulls in tracker 3 right now
2020-09-06 09:36:03 <afontain_> ah, from the meson.build, dependency ('retro-gtk-1', version: '>=' + 0.18.0)
2020-09-06 09:36:06 <afontain_> so we're good
2020-09-06 09:36:19 <afontain_> but I have no idea how it finds it
2020-09-06 09:36:36 <afontain_> there's no package config or anything
2020-09-06 09:37:25 <afontain_> ah, yes, there is
2020-09-06 09:37:27 <afontain_> nevermind
2020-09-06 09:38:05 <afontain_> in our case, the pkg-config is called retro-gtk-0.14, not retro-gtk-1
2020-09-06 09:42:37 <Cogitri> Yes, I guess a new version which broke API will be released soon
2020-09-06 09:43:45 <ikke> 
ACTION points subtly to #alpine-devel ;-)

2020-09-06 09:46:22 <Cogitri> Oh right, didn't pay attention to the channel name :D
2020-09-06 09:48:20 <danpker> Hey, I was wondering if anyone could help me setting the log level for crond. I'm trying to debug a job that isn't running
2020-09-06 10:00:04 <zcrayfish> danpker: edit /etc/conf.d/crond
2020-09-06 10:00:44 <zcrayfish> add an -l option to the CRON_OPTS variable there
2020-09-06 10:00:48 <zcrayfish> and restart the service
2020-09-06 10:00:51 <zcrayfish> that should do what you want
2020-09-06 10:00:58 <Cogitri> https://wiki.alpinelinux.org/wiki/Alpine_Linux:FAQ#My_cron_jobs_don.27t_run.3F
2020-09-06 10:15:31 <Gnjurac> clear
2020-09-06 10:15:41 <Gnjurac> sorry
2020-09-06 10:35:12 <Gnjurac> hmm acme.sh certbot getssl they all on edge :(
2020-09-06 10:35:51 <Gnjurac> hmm guess i can just dl acme.sh script manualy
2020-09-06 10:37:32 <Gnjurac> ssl_client: raw.githubusercontent.com: ocsp verify failed: ocsp response not current
2020-09-06 10:37:39 <Gnjurac> why cant wget fatch it
2020-09-06 10:37:53 <Gnjurac> need to chedk whats this ocsp thingy
2020-09-06 10:38:02 <zcrayfish> OSCP is a check for certificate revocation status
2020-09-06 10:39:10 <zcrayfish> acme.sh is in the community repo
2020-09-06 10:40:37 <Gnjurac> oh i think i know why cert faild
2020-09-06 10:40:46 <Gnjurac> my time is off Tue Aug 18 23:57:34 CEST 2020
2020-09-06 10:41:03 <zcrayfish> Yes, OSCP needs reliable timekeeping
2020-09-06 10:41:11 <Gnjurac> hmm dident i got istalled that timer thingy on setup
2020-09-06 10:42:25 <Gnjurac> rc-service chronyd status
2020-09-06 10:42:25 <Gnjurac>  * status: starte
2020-09-06 10:43:00 <chieta> what should be installed https://bpa.st/4WMQ ?
2020-09-06 10:43:01 <Gnjurac> hmm then why my time off
2020-09-06 10:43:18 <Gnjurac> meybe nftables blocking it
2020-09-06 10:44:00 <zcrayfish> chieta: clang or GCC
2020-09-06 10:44:13 <ikke> chieta: build-base
2020-09-06 10:44:18 <zcrayfish> ^ even better
2020-09-06 10:45:01 <chieta> yes i have tried it apk --update add gcc zcrayfish
2020-09-06 10:45:19 <chieta> ahh got it ikke
2020-09-06 10:45:54 <ikke> build-base adds several basic build tools
2020-09-06 10:47:20 <Gnjurac> hmm i restarded chrony and my time is now normal
2020-09-06 10:47:23 <Gnjurac> wierd
2020-09-06 10:47:51 <zcrayfish> you might need to pick a more local NTP server.
2020-09-06 10:53:25 <Gnjurac> i will if it fails again
2020-09-06 11:02:58 <danpker> zcrayfish: Thanks, that’s what I was looking for
2020-09-06 11:03:49 <zcrayfish> np
2020-09-06 12:12:33 <raiju> Question about graphics, so running KDE on alpine I have seamless web browsing. When running i3wm it's extremely glitchy/laggy. Is there a service I need to manually start or similar?
2020-09-06 12:36:49 <MY-R> raiju, kde by default using compositing so probably firefox/chromium detecting it and enable HW (gpu) acceleration
2020-09-06 12:37:24 <MY-R> but you can enforce browsers to use it too
2020-09-06 12:41:28 <raiju> I think that's probably the key, I need to read more about. I found conflicting answers on quora saying don't use/use compositing in I3, guess I need to read up on how it works
2020-09-06 12:54:43 <MY-R> raiju, just enable in chromium "chrome://flags" "ignore-gpu-blacklist" and in firefox "about:config" change "layers.acceleration.force-enabled" to "true"
2020-09-06 12:56:28 <Hello71> I doubt that is related to compositing
2020-09-06 13:11:17 <raiju> :MY-R Thanks, I saved that to run when I get home
2020-09-06 16:17:05 <charl> I just installed doas. Is there a reason why /etc/doas.conf is read only?
2020-09-06 16:17:19 <charl> https://git.alpinelinux.org/aports/commit/?id=5d3999ccaee7d956bb75f5800e118e2e0e81dfa9
2020-09-06 16:22:01 <Hello71> same as sudo
2020-09-06 16:22:08 <Hello71> sudoers
2020-09-06 16:22:14 <Hello71> I think?
2020-09-06 16:22:37 <Hello71> hm, I think I'm missing something
2020-09-06 16:24:23 <charl> am I not supposed to edit it directly with vi or what?
2020-09-06 16:24:26 <charl> sudo has visudo
2020-09-06 16:25:07 <Hello71> ncopa: https://git.alpinelinux.org/aports/commit/main/doas/APKBUILD?id=dde225b4e88039a27e1ee33216a79c56775c446b?
2020-09-06 16:25:22 <Hello71> also apparently there's more than one portable doas
2020-09-06 16:28:52 <charl> Hello71 like which one?
2020-09-06 16:29:17 <charl> I see only main/doas
2020-09-06 16:35:53 <Hello71> portable
2020-09-06 16:48:03 <ikke> Yeah, there are at least 2 doas implementations
2020-09-06 16:48:33 <ikke> one is a ported version from BSD, the other is a different project
2020-09-06 17:01:22 <Hello71> I found https://github.com/Duncaen/OpenDoas and https://github.com/slicer69/doas
2020-09-06 17:02:40 <duncaen> the slicer69 one is bad
2020-09-06 17:04:23 <duncaen> https://git.alpinelinux.org/aports/tree/main/doas/APKBUILD?id=5d3999ccaee7d956bb75f5800e118e2e0e81dfa9#n37
2020-09-06 17:04:30 <duncaen> the configuration file comes from alpine
2020-09-06 17:07:15 <duncaen> https://github.com/slicer69/doas/blob/master/vidoas
2020-09-06 17:07:17 <duncaen> lol
2020-09-06 17:07:45 <duncaen> I would report this as security issue, but the author already banned me from last time
2020-09-06 17:09:32 <charl> got an answer from Drew himself: "The idea was to give you pause while editing it so that you were
2020-09-06 17:18:15 <Hello71> drew?
2020-09-06 17:23:54 <charl> Drew DeVault, doas package maintainer
2020-09-06 17:25:27 <Hello71> but ncopa committed that change
2020-09-06 17:25:33 <Hello71> meh
2020-09-06 17:34:35 <charl> the install line comes from the maintainer when he first added the port
2020-09-06 17:34:47 <charl> install -Dm440 "$srcdir"/doas.conf "$pkgdir"/etc/doas.conf
2020-09-06 17:35:19 <charl> or was it ncopa anyway doesn't matter :P
2020-09-06 17:35:34 <charl> sorry for pinging
2020-09-06 17:41:47 <charl> fcolista: man page of picom not in picom-doc
2020-09-07 10:18:32 <ncopa> im not really sure why we made doas.conf readonly. its a bit sensitive though. if you make a syntax error you may lock yourself out, and there are no visudo equivalent
2020-09-07 11:20:34 <ws2k3> i just installed alpine linux in a vm. when i try to ssh into it i get Too many authentication failures
2020-09-07 11:45:48 <ws2k3> anyone has an idea on how top install openssl 1.0.1f into alpine 3.10?
2020-09-07 11:49:15 <ikke> You'd have to manually build it most likely
2020-09-07 11:56:17 <ws2k3> ikke: i was trying to use alpine linux 3.8 since that has openssl 1.0 but then i could not install curl-dev
2020-09-07 11:56:45 <ikke> Why do you need such an old version?
2020-09-07 11:57:17 <ws2k3> cause i would like top compile php 5.5.9 with openssl
2020-09-07 11:58:22 <ikke> for the record, Alpine 3.8 should have curl-dev
2020-09-07 11:58:28 <ikke> But 3.8 is no longer maintained
2020-09-07 11:59:03 <ws2k3> ikke: https://pastebin.com/6bwEQg1X
2020-09-07 12:00:41 <ikke> Right, so 3.8 is still built with libressl, so it would conflict with openssl
2020-09-07 12:01:25 <ws2k3> ikke: / # openssl version
2020-09-07 12:01:25 <ws2k3> OpenSSL 1.0.2u  20 Dec 2019 or am i not understanding you?
2020-09-07 12:01:55 <ikke> ws2k3: All packages in Alpine 3.8 are built against libressl
2020-09-07 12:02:32 <ikke> so when you try to install curl-dev, it also tries to install libressl-dev, which conflicts with openssl-dev
2020-09-07 13:37:13 <Hello71> ncopa: well there is https://github.com/slicer69/doas/blob/master/vidoas, but it's, uh... pretty bad
2020-09-07 14:21:19 <Gnjurac> hmm is ngnix runnig under root or what hwen i do rc-services start ngnix
2020-09-07 14:23:19 <zcrayfish> Gnjurac: ps aux | grep nginx
2020-09-07 14:23:24 <Gnjurac> ty
2020-09-07 14:23:24 <zcrayfish> 2nd column is user it is runnning as
2020-09-07 14:24:25 <zcrayfish> but; I think that server might fork as a limited user
2020-09-07 14:24:27 <Gnjurac> ok nice it runs under normal user
2020-09-07 14:24:29 <Gnjurac> thats good
2020-09-07 14:24:40 <zcrayfish> check nginx.conf
2020-09-07 14:24:48 <zcrayfish> oh cool
2020-09-07 14:26:29 <Gnjurac> nah i mistyped
2020-09-07 14:26:58 <Gnjurac> i did ps aux | grep ngnix
2020-09-07 14:27:02 <Gnjurac> anyway this is what i get
2020-09-07 14:27:03 <Gnjurac>  3626 root      0:00 nginx: master process /usr/sbin/nginx -c /etc/nginx/nginx.conf
2020-09-07 14:27:04 <Gnjurac>  3635 root      0:00 grep nginx
2020-09-07 14:27:04 <Gnjurac>  3627 nginx     0:00 nginx: worker process
2020-09-07 14:27:28 <Gnjurac> so this shows 3 processes
2020-09-07 14:29:18 <Gnjurac> so will change to my user in ngnix conf right
2020-09-07 14:31:02 <Gnjurac>  3689 root      0:00 nginx: master process /usr/sbin/nginx -c /etc/nginx/nginx.conf
2020-09-07 14:31:29 <Gnjurac> changed to my user but this one still stays as root like while it was runing nngixn under ngnix user
2020-09-07 14:31:38 <Gnjurac> guess i go ask on ngxni to not harras here
2020-09-07 14:32:21 <zcrayfish> It probably goes as root to begin with in order to listen on port 80, that is a privileged port
2020-09-07 14:32:45 <zcrayfish> but, as you can see by PID 3627, it forks to a limited user
2020-09-07 14:32:56 <zcrayfish> when actually serving the requests
2020-09-07 14:33:03 <Gnjurac> i dident know process can for to other user
2020-09-07 14:35:40 <ikke> root can :)
2020-09-07 15:32:51 <Gnjurac> *fork, ok :)
2020-09-07 15:48:46 <mikn> i have a strange issue with zfs snapshots.
2020-09-07 15:49:40 <mikn> ls /some/where/.zfs/snapshots/mysnap results in "ls: cannot open directory 'mysnap': Symbolic link loop"
2020-09-07 15:51:12 <mikn> trying to rollback also fails with a listing of later snapshots and the message "can not destroy 'latersnapshot': dataset is busy"
2020-09-07 15:52:34 <mikn> the only reference to anything similar is in the alpine linux mailing list archives from three months ago with no responses
2020-09-07 15:54:04 <mikn> "I can access with the usual syntax on Alpine 3.10 (kernel 4.19), but with
2020-09-07 15:54:04 <mikn> Alpine 3.11 (kernel 5.4) I get" (symbolic link loop)
2020-09-07 15:55:17 <mikn> Soo, should I maybe downgrade kernel or what? Sorry if this is stupid, I'm still learning.
2020-09-07 16:00:13 <mikn> I'm on Alpine 3.12, kernel 5.4.43-1-lts btw
2020-09-07 20:14:25 <Guntoo> hei
2020-09-07 20:14:32 <Guntoo> latest upgrade amazing
2020-09-07 20:16:04 <Guntoo> when will it possible to run without NO_HARWARE_CURSOR ?
2020-09-07 20:16:24 <Guntoo> wihout that my curses becomes invisible
2020-09-07 20:16:47 <ikke> In what context?
2020-09-07 20:16:49 <Guntoo> I can click and all but I have to guess the coordinates each time
2020-09-07 20:17:52 <Guntoo> sorry export WLR_NO_HARDWARE_CURSORS=1
2020-09-07 20:19:32 <ikke> What environment?
2020-09-07 20:20:23 <Guntoo> sway
2020-09-07 20:21:44 <Guntoo> oops I meant asking in #sway
2020-09-07 20:23:34 <Guntoo> was musl 1.2.1 an important update?
2020-09-07 20:23:49 <ikke> Mostly for 32-bits systems
2020-09-07 20:24:03 <ikke> Because it changed the size of time_t
2020-09-07 20:24:42 <Guntoo> oh right, the thing still lingering around
2020-09-07 20:29:13 <maerwald> how do I get libffi.so.6 on edge?
2020-09-07 20:29:33 <maerwald> libffi is important for binary compatibility
2020-09-07 20:30:22 <Cogitri> You'd need to introduce a libffi6 package and build that locally, I suppose
2020-09-07 20:30:51 <Cogitri> We've upgraded to the new libffi and rebuilt everything against it so the old version is no more in the Alpine repos
2020-09-07 20:31:19 <maerwald> bad news for haskell then
2020-09-07 20:38:11 <ikke> Because it only works with libffi6?
2020-09-07 20:39:03 <maerwald> the binary packages yeah
2020-09-07 20:39:23 <maerwald> though some have a bundled libffi I think
2020-09-08 03:24:18 <u0_a305> Could switching repositories cause system files to be overwritten? I somehow managed to empty my firmware folder this weekend, and only thing I can think of was switching from versioned repositories to latest
2020-09-08 03:33:50 <ikke> u0_a305: No, that should not be the cause
2020-09-08 03:34:46 <ikke> u0_a305: what *can* happen is if you modify apk managed files not in /etc, then those changes could be overwritten (but that can happen with any upgrade)
2020-09-08 03:36:43 <u0_a305> Got ya, thanks. I should have copied my log files before I reinstalled because now I'm curious. I only remember changing repos and installing xfce and then poof.
2020-09-08 04:15:50 <opal> are any firmware packages still showing up as installed
2020-09-08 04:15:56 <opal> might have not been in world set
2020-09-08 04:21:16 <u0_a305> The install was only a day old so I just reinstalled and started fresh and copied all the firmware files to another location just in case ha. The entire folder was completely empty.
2020-09-08 09:23:03 <telmich> I think the last firefox upgrade made it sluggish again
2020-09-08 09:23:22 <telmich> It worked somewhat fine for about 1-2 weeks and now I have the hanging behaviour back
2020-09-08 09:26:19 <mps> result of bloated small simple ....
2020-09-08 09:26:48 <mps> yes, alpine starts to look and behave as other bloated distros
2020-09-08 09:28:39 <u0_a305> I had trouble with firefox on I3, but then I installed KDE and everything works great
2020-09-08 09:28:42 <mps> re firefox, I read somewhere that palemoon could be built on musl (sabotage linux iirc)
2020-09-08 09:58:35 <EF```> It is a shame Palemoon's devs won't consider a formal musl version
2020-09-08 10:01:30 <MY-R> for a week firefox was without dbus support so maybe need start/enable dbus daemon in system (without it firefox print many warnings) + get rid of XDG_RUNTIME_DIR (which dbus using too) warnings by setting it up in profile.d like here: https://wiki.alpinelinux.org/wiki/Wayland
2020-09-08 10:11:19 <mps> EF```: if I have time I would try to build palemoon on alpine
2020-09-08 10:12:45 <detha> I wish there were something like a libdbus0 that makes applications believe they can connect to dbus, but that does absolutely nothing, just to get rid of the 'can't connect to dbus' stderr spam
2020-09-08 10:13:42 <mps> detha: yes, I'm also thinking about it. something like libudev-zero
2020-09-08 10:15:49 <mps> https://github.com/illiliti/libudev-zero
2020-09-08 10:17:40 <EF```> mps: like you stated earlier, others have had a go but Palemoon is glibc and blinkers! Shame as it is a decent browser
2020-09-08 10:20:18 <mps> EF```: sabotage developer told me that they build and use it on sabotage linux which is musl based also
2020-09-08 10:20:47 <mps> but I don't have free time to make it on alpine
2020-09-08 10:21:15 <detha> whenever I hear Palemoon, I think of https://github.com/jasperla/openbsd-wip/issues/86
2020-09-08 10:24:07 <mps> aha, but build for my own usage is fine
2020-09-08 10:24:58 <EF```> Don't believe politics and personal issues should darken software. Not like they're arguing over some 0-day exploit!
2020-09-08 10:27:33 <u0jQx9gPyrYg> are the palemoon guys finally backporting all the security bugfixes from upstream? only checked a few years ago, and then it was this cesspool of long forgotten cves
2020-09-08 10:31:08 <mps> iiuc it is forked long time ago, so not sure do they need (or can) backport anything
2020-09-08 10:33:54 <Cogitri> They probably have to backport a ton of things but its next to impossible to do that with an ever diverging codebase
2020-09-08 10:36:24 <u0jQx9gPyrYg> why would anyone use a browser with tons of 100days?
2020-09-08 10:36:25 <mps> they don't use servo and rust
2020-09-08 10:37:03 <mps> u0jQx9gPyrYg: all browser are risky
2020-09-08 10:37:24 <mps> except maybe edbrowse ;)
2020-09-08 10:37:27 <u0jQx9gPyrYg> yes, but 0days are less than 100days
2020-09-08 10:38:11 <mps> I don't see big diff in 1day or 100days
2020-09-08 10:39:39 <ikke> the older the vulnerability, the more commonly it's tried to be exploited
2020-09-08 10:40:53 <mps> nowadays this is irrelevant, those who want to exploit someones something are ready even for 0day exploits
2020-09-08 10:42:51 <mps> security and easy to use (with a featurism attitidue) doesn't go in tandem
2020-09-08 10:44:34 <ikke> THe most secure system is one that does not do anything
2020-09-08 10:46:32 <detha> <ikke> the older the vulnerability, the more commonly it's tried to be exploited - disagree. Nobody is going to bother trying 10-year-old CVEs, because the percentage of machines that still applies to is insignificant
2020-09-08 10:46:53 <mps> huh, so called 'Godwin law'
2020-09-08 10:49:52 <mps> (when god created cosmos it made it as insecure 'place')
2020-09-08 10:51:45 <u0jQx9gPyrYg> the big diff is in the price of those exploits
2020-09-08 11:08:33 <charl> Looking at the wiki there are 2 methods to suspend: 1. pm-suspend, 2. echo mem > /sys/power/state. What is the difference between the 2?
2020-09-08 11:08:43 <charl> Which one do you guys use?
2020-09-08 11:11:52 <caskd> i usually use the /sys/power/state one at that relies on acpid afaik and doesn't need any extra stuff
2020-09-08 11:12:39 <caskd> you could also use pm-suspend but i think using acpi is simpler
2020-09-08 11:12:59 <MY-R> charl, pm-suspend is part of package "pm-utils" which include many scripts: https://pkgs.alpinelinux.org/contents?branch=edge&name=pm-utils&arch=x86_64&repo=main
2020-09-08 11:13:06 <charl> yeah the /sys/power/state method seems to work fine for me
2020-09-08 11:13:19 <charl> pm-utils is broken and seems upstream is abandoned
2020-09-08 11:13:42 <charl> # /usr/sbin/pm-suspend: line 199: /usr/lib/pm-utils/sleep.d/98video-quirk-db-handler: not found
2020-09-08 11:13:50 <charl> head -n1 /usr/lib/pm-utils/sleep.d/98video-quirk-db-handler
2020-09-08 11:14:42 <charl> also requires bash for one script ^ and bash isn't listed as dependency when you apk add pm-utils
2020-09-08 12:27:02 <Hello71> detha: tell that to all the shit kiddies spamming worldwide server logs
2020-09-08 18:20:08 <Guntoo> greets
2020-09-08 18:20:22 <Guntoo> a bizzarre thing is hapening
2020-09-08 18:20:47 <Guntoo> browser works fine,  I can type fast, other alacritty terms work fine as well.
2020-09-08 18:22:00 <Guntoo> yet with a terminal in root  sway goes up 60% cpu usage and typing becomes highly unresponsive. Just deleting lines it hangs in some letters then jumps to a random location
2020-09-08 18:22:28 <Guntoo> I thougth it would be the same in all the open terms. it isn't
2020-09-08 18:23:03 <Guntoo> even here in irssi is quick and responsive, as soon as I focus the terminal in root slows down to a crawl
2020-09-08 18:23:17 <Guntoo> how's this even possible
2020-09-08 18:24:29 <Guntoo> lawl, up to 107% usage
2020-09-08 18:26:48 <santosh> Hi
2020-09-08 18:27:04 <santosh> Which Alpine variant is suitable for my discover 4 plus tablet
2020-09-08 18:27:08 <santosh> Thanks
2020-09-08 18:27:28 <Guntoo> Guest56931: arm perhaps
2020-09-08 18:28:04 <Guest56931> Thanks
2020-09-08 18:28:13 <Guest56931> Aarm7 ?
2020-09-08 18:28:18 <Guest56931> arm7 ?
2020-09-08 18:29:09 <Guest56931> Will it work on tablet?
2020-09-08 18:29:13 <Guest56931> Touchscreen ?
2020-09-08 18:30:01 <Guest56931> How to install it on Tab ?
2020-09-08 18:30:12 <Guest56931> Please guide, as I'm in trouble
2020-09-08 18:31:04 <mps> Guest56931: look at postmarketOS, alpine for phones/tablets
2020-09-08 18:31:28 <Guest56931> postmarketos not available for download
2020-09-08 18:31:30 <Guest56931> Thanks
2020-09-08 18:32:23 <mps> postmarketos.org
2020-09-08 19:02:41 <user_mike> Hello! I'm digging into Alpine Linux on the PinePhone---- question... does anyone have information about getting HackRF running?
2020-09-08 19:06:45 <user_mike> I have found alternative APK equivalent packages for LibUSB, FFTW, and default-jdk, now
2020-09-08 19:07:08 <user_mike> https://github.com/pavsa/hackrf-spectrum-analyzer
2020-09-08 20:01:19 <telmich> Re sluggish firefox: I think the problem was that I booted the LTS kernel. FF is so far fine on the edge kernel
2020-09-08 20:01:55 <Cogitri> 👌
2020-09-08 20:02:47 <mps> yes, linux-edge is built with forced preempt while -lts without any preemption
2020-09-08 20:06:05 <mps> though I think ff responsiveness depends on site visited
2020-09-08 20:09:28 <telmich> Might be, however for me it often gets into a really unusable state after some time (some days) and then even scrolling any website takes seconds before it reacts
2020-09-08 20:12:44 <mps> for me some sites are slow, sites with heavy javascript
2020-09-08 20:14:04 <user_mike> Does Privacy badger help at all to cut out javasript?
2020-09-08 20:15:33 <Cogitri> Use uMatrix for that
2020-09-08 20:19:25 <mps> uMatrix is good but for some sites it have to allow JS to see content
2020-09-08 20:20:17 <Cogitri> Yes, but that's the point of uMatrix?
2020-09-08 20:20:35 <mps> yes
2020-09-08 20:21:05 <user_mike> Intrusive javascript content exchanges
2020-09-08 20:21:43 <mps> used for bitcoin mining ;)
2020-09-08 20:29:46 <user_mike> And reading the larynx of sub vocal thought
2020-09-08 20:33:38 <user_mike> The body reradiates FM/Microwaves with the relative EEG waveform... twhich is building the talking statue of AI especially in China these days
2020-09-08 20:49:03 <u0jQx9gPyrYg> user_mike: i maintain a bunch of packages related to sdrs (including hackrf) in the aports-ugly repo: https://github.com/aports-ugly/aports/ - you could take those and build them for your phone.
2020-09-08 20:49:44 <u0jQx9gPyrYg> i have a hackrf and drive it with those packages both on x86_64 and also on arm64
2020-09-08 21:02:34 <user_mike> Thanks You u0jQx9gPyrYg
2020-09-08 21:03:18 <user_mike> Do you think that JDK and UDev permissions are still required?
2020-09-08 21:05:28 <user_mike> I'm embarassed to say I can't find a 'sudo gedit' alternative in PostMarketOS & Phosh
2020-09-08 21:07:45 <Cogitri> Please don't launch GUI apps with root
2020-09-08 21:08:04 <Cogitri> Doesn't work on any Wayland environment
2020-09-08 21:09:31 <Cogitri> You could make a simple bash script that copies the file to $(mktemp -d)/file, opens that with gedit and sudo cp the temp file
2020-09-08 21:16:09 <Hello71> EDITOR=gedit sudoedit?
2020-09-08 21:16:41 <Cogitri> Right, that does basically that, right?
2020-09-08 21:19:52 <user_mike> (The following things have been tested on Ubuntu and Gentoo and may need to be adapted to other Linux distributions. In particular, your distro may have a group named something other than plugdev for this purpose.)
2020-09-08 21:19:52 <user_mike> To do that, you need to create a new rules file in the /etc/udev/rules.d folder. I called mine 52-hackrf.rules. Here is the content:
2020-09-08 21:19:52 <user_mike> ATTR{idVendor}=="1d50", ATTR{idProduct}=="6089", SYMLINK+="hackrf-one-%k", MODE="660", GROUP="plugdev"
2020-09-08 21:19:52 <user_mike> ATTR{idVendor}=="1d50", ATTR{idProduct}=="604b", SYMLINK+="hackrf-jawbreaker-%k", MODE="660", GROUP="plugdev"
2020-09-08 21:19:54 <user_mike> ATTR{idVendor}=="1fc9", ATTR{idProduct}=="000c", SYMLINK+="hackrf-dfu-%k", MODE="660", GROUP="plugdev"
2020-09-08 21:20:28 <user_mike> Just 3 lines for UDev :( ty
2020-09-08 21:32:34 <mps> few hours ago replaced udev with libudev-zero and tweaked mdev.conf
2020-09-08 21:34:03 <mps> so, I'll have to write some new mdev rules
2020-09-09 06:18:53 <JD_the_fish> Im having some problems with changing drive lettering in diskless mode. And my fstab entry for mounting my save location is seemingly ignored
2020-09-09 06:52:30 <JD_the_fish> where would one look in order to understand the alpine linux "automount" system? Because it certainly does some automatic mounting in /media/
2020-09-09 07:01:25 <Cogitri> In a base system? Usually you'd need something like gvfs or kio installed for automatic mounting
2020-09-09 07:03:11 <JD_the_fish> Well, for example there is an fstab entry to mount /dev/usbdisk at /media/usb
2020-09-09 07:03:33 <JD_the_fish> Tho i think whats assigned to /dev/usbdisk is decided by the kernel
2020-09-09 07:09:37 <JD_the_fish> The problem that i have is that alpine insists on mounting the second parition on the SD card under /media/sdb2
2020-09-09 07:09:48 <JD_the_fish> or sda2 depending on how the name changes
2020-09-09 07:10:22 <JD_the_fish> even if i specify an fstab entry to mount it by UUID under /media/pers
2020-09-09 07:13:53 <JD_the_fish> Im starting to suspect this stuff gets handled by the initram fs or something before the actual os init
2020-09-09 07:14:37 <JD_the_fish> It would be nice to have some documentation on the boot sequence in diskless mode
2020-09-09 07:21:27 <JD_the_fish> im about ready to give up on diskless mode
2020-09-09 12:37:59 <Hello71> Cogitri: I don't think that's right? at most that lets you access drives in gnome/kde gui, but I don't think it does
2020-09-09 12:38:10 <Hello71> for disk automount you need udisks
2020-09-09 12:38:24 <Hello71> or traditional fstab user flag
2020-09-09 12:38:47 <Cogitri> Right, gvfs uses udisks for that
2020-09-09 12:39:48 <Hello71> could be that gvfs explicitly invokes udisks, but you can use udisks without either, and use either without udisks (e.g. for network drives)
2020-09-09 12:40:45 <mps> I think udev/mdev can do that
2020-09-09 12:41:19 <mps> though I use udevil from cli
2020-09-09 12:56:04 <Hello71> yeah, you can make some custom hacks, but udisks is the Standard Solution
2020-09-09 13:00:39 <mps> standard? I wouldn't point to xkcd about standards :)
2020-09-09 13:01:35 <ikke> If you don't, I will :P https://xkcd.com/927/
2020-09-09 13:04:12 <mps> ;)
2020-09-09 14:00:38 <lxeiqr> Hey, is it possible to run Tor Browser on Alpine Linux (at least via gcompat)?
2020-09-09 14:55:42 <u0jQx9gPyrYg> lxeiqr: if not, you can always run it in a chroot
2020-09-09 17:14:49 <DusXMT> Greetings, I've just built the community/ffmpeg package using 'abuild -r', but I've noticed that the libraries seem to miss their version names, as well as the typical symlinks usually seen with libraries to handle ABI compatibility: https://dpaste.org/takX
2020-09-09 17:16:02 <DusXMT> I'm rather new to Alpine, so it's likely I'm missing something obvious
2020-09-09 17:22:37 <u0jQx9gPyrYg> that is normal
2020-09-09 17:22:46 <u0jQx9gPyrYg> for binary normal use pkgs you don't need that
2020-09-09 17:22:55 <u0jQx9gPyrYg> only in the -dev packages you have those symlinks...
2020-09-09 17:23:58 <DusXMT> Hmmm, then why does the package from Alpine's binary repos have those symlinks?
2020-09-09 18:14:30 <Hello71> DusXMT: tar -tzvf
2020-09-09 18:14:53 <Hello71> although modern tars autodetect z
2020-09-09 18:15:35 <DusXMT> Hello71: https://dpaste.org/JDZx
2020-09-09 20:44:37 <ismaell> hi
2020-09-09 20:44:55 <ismaell> any page where I can look how to setup the update of the initramfs?
2020-09-09 20:45:23 <ismaell> I want to script the customization
2020-09-09 20:48:08 <mps> ismaell: /sbin/mkinitfs is shell script
2020-09-09 20:48:36 <mps> so you can read it and change if you want
2020-09-09 20:48:42 <mps> want/need
2020-09-09 20:52:40 <ismaell> mps: thanks
2020-09-09 20:52:54 <c705> ismaell: you may also need to hack update-kernel depending on what you're trying to get done
2020-09-09 20:53:25 <ismaell> c705: I just want to trim it down
2020-09-09 20:53:30 <ismaell> I'm new to the distro
2020-09-09 20:53:36 <ismaell> so trying to familiarize
2020-09-09 20:53:55 <ismaell> my VPS died and I decided to switch from Debian to Alpine
2020-09-09 20:53:58 <ismaell> a good decision so far
2020-09-09 20:54:35 <ismaell> so I started scripting everything I need for alpine
2020-09-09 20:54:36 <ismaell> :)
2020-09-10 03:31:32 <opal> < mps> so you can read it and change if you want
2020-09-10 03:31:38 <opal> on upgrade wont changes be overwritten
2020-09-10 10:35:55 <petyy> Hi there!
2020-09-10 10:36:13 <petyy> can I have a question about alpine edge?
2020-09-10 10:38:18 <zcrayfish> no need for meta, simply ask your question, if someone knows they'll pipe up eventually
2020-09-10 10:40:16 <petyy> I would like to install libreoffice* packages, but I'm getting "ERROR: unsatisfiable constraints:"...packages are missing. Searches with Google gave me links to libreoffices version 6.4.6.2-r0, builded on 2020-08-27, but only as a cached results. When I try to access the link, I'm getting 404 error message. Also when I try to find libreoffice's packages on pkgs.alpinelinux.org, I'm getting zero results for edge version of Alpine.
2020-09-10 10:40:32 <petyy> What happened to libreoffice packages for edge, please?
2020-09-10 10:40:46 <Cogitri> Yes, libreoffice fails to build with GCC10
2020-09-10 10:41:00 <Cogitri> So we can't rebuild it right now
2020-09-10 10:42:05 <petyy> aha, ok. then I have to wait until libreoffice fixes their build compatibility with GCC10, so you'll be able to build the packages...
2020-09-10 10:42:51 <petyy> version which was removed had some problems so it was removed?
2020-09-10 10:43:39 <petyy> I mean...it was build already and present in repo...
2020-09-10 10:56:18 <petyy> anyway, thanks for answer :)
2020-09-10 10:56:26 <Cogitri> It was removed because poppler was bumped and made a soname bump, so we had to rebuild libreoffice for it
2020-09-10 11:11:47 <petyy> @Cogitri: thx! :)
2020-09-10 11:12:52 <Cogitri> The bug might actually be in GCC and not LO, so not sure when we'll get around to fixing that :/
2020-09-10 11:16:37 <petyy> Have a nice day :) Bye
2020-09-10 13:58:34 <ws2k3> is libsasl2-modules-sql available on alpine linux?
2020-09-10 13:59:12 <Cogitri> pkgs.alpinelinux.org might be able to answer that question
2020-09-10 14:00:02 <ws2k3> Cogitri: i already checked but i could not find it. only libsasl but i allmost cant imagen that its not available
2020-09-10 14:04:14 <Hello71> https://pkgs.alpinelinux.org/contents?branch=edge&name=*sasl*
2020-09-10 14:08:53 <repo> hi
2020-09-10 14:09:16 <repo> i'm trying to install wine (32 bit version) as a dependency in a 64 bit container
2020-09-10 14:09:25 <repo> or rather an image
2020-09-10 14:09:38 <repo> so i tried: RUN echo x86 > /etc/apk/arch && apk --update add wine
2020-09-10 14:09:53 <repo> but this seems to still install the 64 bit version
2020-09-10 14:10:08 <ikke> yeah, apk is hardcoded for just a single arch
2020-09-10 14:10:29 <repo> i thought i could circumvent that by overwriting the arch file
2020-09-10 14:10:30 <ikke> so whatever apk --print-arch returns, that's the arch it will fetch
2020-09-10 14:10:39 <repo> hm
2020-09-10 14:10:46 <repo> yeah it returns x86_64
2020-09-10 14:11:27 <ikke> and alpine is not multi-arch capable, so I wonder if that would even work
2020-09-10 14:12:32 <repo> hm ok ic
2020-09-10 16:12:36 <bala> hello
2020-09-10 16:13:25 <bala> someone can help me?
2020-09-10 16:14:13 <CcxWrk> bala: Not unless you ask actual question we can help with. ;-)
2020-09-10 16:16:14 <teraflops> 🤷
2020-09-10 20:46:24 <Bradley2> To be honest I've never heard of Alpine as a distro. Who is this for?
2020-09-10 20:48:27 <nshp> Bradley2: some people run it on routers and things like that, a lot of people use it as a base for generic docker containers, some people use it as a generic server distro... it's just another linux distro configurable for whatever you want
2020-09-10 21:35:25 <fluvf> Hey!I'm trying to setup a router using awall, that'll only allow access to WAN through openvpn,but I've hit a wall and can't get my box to actually route anything.It connects fine through the vpn, but my machines on LAN can't connect to the box, nor to the internet.I followed the Zero-to-Awall guide, and I've read the rest of the documentation that
2020-09-10 21:35:25 <fluvf> awall has to offer,but obviously I'm doing something wrong. Any and all help would be appreciated!
2020-09-10 23:14:58 <alxsdf> I don't actually use alpine linux as a server distribution
2020-09-10 23:15:01 <alxsdf> it's my main desktop
2020-09-10 23:15:10 <alxsdf> at least on my x200
2020-09-10 23:16:20 <alxsdf> Anyway, that's not why I'm here
2020-09-10 23:16:51 <alxsdf> I just upgraded my system (800 packages lol), and I get a message at boot, from pulseaudio about not using it as the root user
2020-09-10 23:20:21 <tehcloud> get rid of it and live a happier life
2020-09-10 23:22:07 <alxsdf> hm?
2020-09-10 23:25:05 <alxsdf> oh, no unfortunately pulseaudio is required for some things, frustratingly
2020-09-10 23:31:21 <alxsdf> is there anything wrong with replacing '--system' in the start-stop-daemon line to '--name pulse'
2020-09-10 23:35:30 <alxsdf> no, nvm, brain wasn't engaged
2020-09-11 00:04:09 <Hello71> pulse should be run under your user if you choose to use it
2020-09-11 00:04:17 <Hello71> it's still practical (barely) to use alsa directly
2020-09-11 01:00:29 <davew> Is there something that stops passwordless sudo on alpine?  I can't seem to set it up correctly. sudo -l shows NOPASSWD: but then when I use sudo -u I get asked for a password
2020-09-11 03:19:13 <jjuran> `apk search qemu` doesn't turn up any QEMU packages, which is surprising
2020-09-11 04:09:25 <jjuran> Ah, it's in the community repository
2020-09-11 04:13:21 <sysx1000> Hello! Is there any way to reduce amount of tmpfs for / ? It appears to take a half of a RAM, I need to set it to something particular
2020-09-11 04:45:04 <ikke> add 'rootflags=size=<size>' to the kernel parameters
2020-09-11 04:45:16 <ikke> sysx1000: ^
2020-09-11 04:46:17 <sysx1000> ikke, thanks
2020-09-11 04:48:46 <TBB> but then, I have doubts about whether that's how tmpfs works at all
2020-09-11 04:49:39 <ikke> TBB: What do you mean?
2020-09-11 04:50:29 <TBB> well the question doesn't make sense. the only "sizes" for tmpfs are the max size, which is just a limit but doesn't mean anything else, and how much exactly you have data on tmpfs
2020-09-11 04:51:20 <ikke> Maybe they need to increase this limit
2020-09-11 05:08:40 <ikke> Or maybe they want to limit the amount of ram that can be used
2020-09-11 05:24:26 <sysx1000> by default, tmpfs takes 50% of RAM
2020-09-11 05:24:34 <sysx1000> if no size specified
2020-09-11 05:24:40 <sysx1000> that's I have
2020-09-11 05:25:40 <sysx1000> but my system is memory strict (rpi3), so I have to get back unused tmpfs 'sectors' back to RAM. BTW it also could be done by mount -o remount,size=<desired> /
2020-09-11 05:27:16 <ikke> right, but that's not persistent
2020-09-11 05:27:21 <sysx1000> it is
2020-09-11 05:27:31 <ikke> not afte reboot..
2020-09-11 05:27:32 <sysx1000> rootflags works fine, btw
2020-09-11 05:27:37 <ikke> right
2020-09-11 06:14:48 <TBB> tmpfs doesn't reserve memory for the whole filesystem though, only what's used
2020-09-11 06:15:41 <TBB> but in any case, good to hear ikke's solution works for you
2020-09-11 10:06:34 <ws2k3> im trying to use auxprop with sasl postfix on alpine linux 3.10 i receive the following error: SASL authentication problem: unknown password verifier(s) auxprop am i missing some libs maby?
2020-09-11 11:34:12 <ws2k3> im trying to use auxprop with sasl postfix on alpine linux 3.10 i receive the following error: SASL authentication problem: unknown password verifier(s) auxprop am i missing some libs maby?
2020-09-11 11:34:16 <ws2k3> maby the sql plugin is just not available?
2020-09-11 11:39:01 <Hello71> whole point of tmpfs is it does not reserve ram
2020-09-11 11:39:09 <Hello71> as opposed to ramfs
2020-09-11 11:39:49 <Hello71> not sure what happens if you turn off overcommit
2020-09-11 14:14:09 <ws2k3> im trying to use auxprop with sasl postfix on alpine linux 3.10 i receive the following error: SASL authentication problem: unknown password verifier(s) auxprop am i missing some libs maby?
2020-09-11 14:14:15 <ws2k3> maby the sql plugin is just not available?
2020-09-11 15:05:49 <junialter> hi, I'm trying to build a firewall on alpine
2020-09-11 15:06:09 <junialter> the weird thing is the moment I enable IPv6 forwarding in kernel all my traffic is being block from that very moment
2020-09-11 15:06:33 <junialter> it's alpine 3.12
2020-09-11 15:07:15 <junialter> sysctl -w net.ipv6.conf.all.forwarding=1 is what I use
2020-09-11 15:33:32 <Hello71> probably you also want to set accept_ra=2
2020-09-11 15:49:08 <junialter> thank you but problem persists
2020-09-11 15:58:47 <junialter> I will try via radvd, the documentation says that radvd automatically will enable forwarding in kernel
2020-09-11 15:59:16 <junialter> the moment radvd starts, traffic goes to zero
2020-09-11 15:59:26 <junialter> I mean IPv6 traffic of course
2020-09-11 16:05:20 <junialter> dunno what's happening here. I just tried with a different alpine 3.12 host who doesn't suffer from this problem. Maybe I need to reinstall and look if I can somehow reproduce this. Thank you so far though
2020-09-11 16:32:32 <Guntoo> junialter: happening what where?
2020-09-11 16:38:07 <telmich> I get a strange error on some boxes with a proper looking interfaces file: `Error: inet prefix is expected rather than "10.3.0.44/64"`
2020-09-11 16:38:26 <telmich> It's one interface (eth1.10) with a section for ipv4 and one for ipv6
2020-09-11 16:38:36 <telmich> and it seems that ifup mixes them up somehow
2020-09-11 16:40:05 <telmich> I uploaded the config & error on https://www.nico.schottelius.org/temp/ifupdownerror-inetprefix - has anyone seen this before?
2020-09-11 16:46:28 <junialter> Guntoo: I try to make alpine a firewall and the moment I activate IPv6 Forwarding in kernel all IPv6 traffic is being blocked
2020-09-11 16:52:07 <junialter> I think Hello71 was right, I need to set accept_ra=2 but my first attempt didn't work...
2020-09-11 17:32:44 <junialter> the problem was, that I need to set accept_ra explicitly for the interfaces instead of just setting accept_ra.all=2
2020-09-11 17:37:10 <Hello71> there is no accept_ra.all, and you should read the documentation instead of blindly setting things
2020-09-11 17:42:16 <detha> telmich: I believe ifupdown can only set _one_ ipv4 and _one_ ipv6 on an interface
2020-09-11 17:43:21 <junialter> sure pal, I mean .all.accept_ra
2020-09-11 17:43:33 <junialter> this exists for sure, I'm no knewb
2020-09-11 17:44:41 <junialter> telmich: 10.3.0.44/64 is invalid, IPv4 only know metmasks up until /32
2020-09-11 17:55:54 <detha> the 64 is on an ipv6 address. it should not use that on the ipv4
2020-09-11 18:01:13 <detha> what might un-confuse it is to first set the v4 address, then the v6
2020-09-11 18:05:49 <mckay-trabajo> https://wiki.alpinelinux.org/wiki/FTP
2020-09-11 18:05:53 <mckay-trabajo> updated
2020-09-11 18:30:57 <junialter> well that language skills though
2020-09-11 20:52:42 <Hello71> you can't do multiple addresses in one stanza (properly)
2020-09-11 20:52:56 <Hello71> you need to start a new iface
2020-09-11 21:17:42 <mps> well, post-up ip address add dev name address/mask
2020-09-11 21:18:05 <mps> and many addresses appears on one interface
2020-09-11 21:25:37 <Hello71> but that's not an address directive
2020-09-11 21:26:11 <Hello71> at that rate all shell scripts are one liners
2020-09-11 21:52:01 <telmich> In Debian's interfaces, adding `iface ... ` with different protocols is a "typical" thing, which is why I assumed it should work in Alpine, too
2020-09-11 21:52:31 <telmich> junialter: yep, I'm aware of the lengths, the problem is I never specified that mask for the IPv4 range
2020-09-11 21:53:03 <telmich> detha: So you are saying the correct solution is to define either inet6 or inet and then add the rest via post-up?
2020-09-11 22:20:10 <Hello71> did you set another netmask
2020-09-12 05:24:10 <DeVol> I get what looks like ANSI'ish codes like  [30;17R  spamming when connecting through ssh... I never encountered this before on any other ssh-connection. I'm using the ssh-client from the standard git-installation on a Windows 10-box. Any idea what to check? Something with colors in the prompt?
2020-09-12 05:24:51 <DeVol> (I tried both sshd and dropbear)
2020-09-12 05:25:39 <teraflops> DeVol: terminfo issue?
2020-09-12 05:26:37 <teraflops> to see if that's the issue try e.g `export TERM=vt100`
2020-09-12 05:34:17 <DeVol> @teraflops doesn't like the issue
2020-09-12 05:34:20 <DeVol> also not prompt
2020-09-12 05:34:21 <DeVol> https://pasteboard.co/JqH1EJt.png
2020-09-12 06:05:04 <DeVol> apparently it's a client problem, i dont have it on another Win10-box
2020-09-12 06:39:45 <detha> telmich: you should be able to have both. But possibly order matters
2020-09-12 08:55:07 <Guntoo> greets
2020-09-12 08:55:55 <Guntoo> what's a lightweight drawing app? I need to draw a rectangle onto a jpg
2020-09-12 08:57:49 <afontain_> well, gimp and gnome paint are GTK2, so no wayland for you
2020-09-12 08:58:41 <arahael> I thought GTK works on wayland.
2020-09-12 08:58:52 <afontain_> inkscape is now GTK3, but well, it's vector graphics so it's not quite appropriate
2020-09-12 08:59:22 <afontain_> krita is a professional software and a KDE one, so I'm not sure it ticks the lightweight box ^^
2020-09-12 08:59:37 <afontain_> arahael: well, GTK3 works very well on Wayland
2020-09-12 08:59:49 <arahael> Ah, so it's specifically GTK2.
2020-09-12 08:59:58 <afontain_> but GTK2 has no support at all
2020-09-12 09:00:13 <arahael> It's surprising that gimp is using gtk2, still.
2020-09-12 09:00:21 <arahael> I mean, it's the *gimp toolkit*
2020-09-12 09:00:32 <afontain_> lol
2020-09-12 09:00:34 <Guntoo> dayn!!
2020-09-12 09:01:03 <Guntoo> none lightweight, inkscape could work as it has rudimentary jpg/png import.
2020-09-12 09:01:35 <afontain_> then there's stuff that operates on pdf rather than regular images, but that's not great either
2020-09-12 09:02:02 <afontain_> tbh, I'd use gimp with xwayland
2020-09-12 09:03:54 <Guntoo> there's this method but it requires a bit of a learning curve and finding out the dimension and pixel location etc.. --> convert logo: -fill none -stroke black -strokewidth 3 -draw "rectangle 10,10 630,470" logo_rect.png
2020-09-12 09:03:58 <arahael> I'm still using X. :(  Infact, I use xrdp.
2020-09-12 09:04:02 <Guntoo> I wanted to avoid that
2020-09-12 09:04:36 <afontain_> ah yes, that works too
2020-09-12 09:05:00 <Guntoo> another massochist example --> convert rose: -fill red -stroke black -strokewidth 2 \ -draw 'rectangle 0 0 %[fx:w-1] %[fx:h-1]' \  out_stroke.png
2020-09-12 09:05:34 <Guntoo> for this I'll have to rate alpine-linux only 4 stars.
2020-09-12 09:06:17 <Guntoo> I'll hold onto the 5th star until a decent media editing tools are available.
2020-09-12 09:06:52 <mps> who cares of ratings :)
2020-09-12 09:07:24 <afontain_> works with X.org or with XWayland ¯_(ツ)_/¯
2020-09-12 09:14:46 <mps> I have to prepare for wayland my aarc64 moving chromebook
2020-09-12 09:15:18 <mps> only don't know how will I live without awesomeWM
2020-09-12 09:17:09 <afontain_> there's way-cooler, IIRC
2020-09-12 09:17:09 <afontain_> dunno if it's any good
2020-09-12 09:17:56 <mps> do we have it in repo
2020-09-12 09:18:58 <mps> rust!? :P
2020-09-12 09:19:29 <mps> but yes, looks similar to awesomeWM
2020-09-12 09:19:37 <Guntoo> mps: eventually all WM/DE will have to migrate to wayland
2020-09-12 09:20:23 <Guntoo> I've converted sway into pseudo-*box WM
2020-09-12 09:20:39 <mps> Guntoo: yes, I feel that. what a bad future
2020-09-12 09:21:21 <Guntoo> mps: yeah, wayland config is a bit insane at first
2020-09-12 09:21:26 <mps> maybe for the start I could go with xwayland
2020-09-12 09:22:02 <mps> though I don't know even how to start it
2020-09-12 09:23:24 <afontain_> that reminds me of this http://way-cooler.org/blog/2019/04/29/rewriting-way-cooler-in-c.html
2020-09-12 09:23:32 <Guntoo> mps: alpine wiki got some one-liners to install it.
2020-09-12 09:24:24 <afontain_> Rewriting Way Cooler in C
2020-09-12 09:24:28 <afontain_> Starting with this PR[1], the compositor for Way Cooler will be written in C.
2020-09-12 09:24:28 <Guntoo> mps: basically, follow the guide, it pulls some deps, others are missing such as gallium and that's all.
2020-09-12 09:24:48 <afontain_> https://github.com/way-cooler/way-cooler/pull/609
2020-09-12 09:24:54 <Guntoo> mps: which image will you be insalling?
2020-09-12 09:25:06 <mps> Guntoo: I know how to install (at the end I'm alpine developer :) ) but don't know how to start use it
2020-09-12 09:25:33 <Guntoo> mps: oops :/ my bad
2020-09-12 09:25:40 <mps> np ;)
2020-09-12 09:26:02 <mps> afontain_: heh, good info
2020-09-12 09:26:19 <Guntoo> mps: somehow it seemed familiar your nick, but didn't know from which channel
2020-09-12 09:26:35 <mps> so maybe it is promising, will look how to package it for alpine
2020-09-12 09:26:46 <afontain_> hmm
2020-09-12 09:27:06 <afontain_> the repository is archived
2020-09-12 09:27:33 <afontain_> http://way-cooler.org/blog/2020/01/09/way-cooler-post-mortem.html
2020-09-12 09:27:50 <mps> Guntoo: I'm not on many channels, just alpine ones, musl and some linux-ARM channels
2020-09-12 09:27:54 <Guntoo> "Note that the client part of Way Cooler (the part that will implement the AwesomeWM compatibility) is still written in Rust. This is the most pragmatic solution because it can utilize wayland-rs and rlua."
2020-09-12 09:28:57 <Guntoo> rewritten in C, with a wrapper of Rust ir order to use Lua oO
2020-09-12 09:29:00 <Guntoo> what a mess
2020-09-12 09:30:57 <afontain_> « The library was very unsound – meaning it was possible to cause memory unsafety from what we erroneously reported was “safe” Rust code. If we spent more time learning the language and sought help from the wider Rust community in wrapping the library we would have avoided a lot of headache. Instead, we plunged ahead despite not knowing the language well. »
2020-09-12 09:31:13 <mps> but after little thinking, I doubt I will have time for all this
2020-09-12 09:31:44 <mps> working on next -lts kernel and tools take a lot of time
2020-09-12 09:34:15 <Guntoo> mps: you need an assistant then
2020-09-12 09:35:27 <mps> Guntoo: well, most important things are actually done/ready
2020-09-12 09:36:07 <mps> last and most complicated is to test mainline kernels on RPis
2020-09-12 09:36:59 <mps> though I have mainline 5.8 running on RPi zero W, without any patch or special tweak
2020-09-12 09:38:06 <mps> in next few weeks more time will be used on testing and tweaking some things
2020-09-12 09:39:07 <mps> and I hope that will make armhf kernel in linux-edge soon
2020-09-12 09:41:10 <mps> if anyone wants to help with tweaking and testing stable kernels on arm, will be welcomed warmly ;)
2020-09-12 17:27:34 <c705> mps: 5.9 is out I think
2020-09-12 17:27:43 <c705> sorry, nevermind
2020-09-12 17:27:53 <c705> 5.8.9
2020-09-12 17:27:56 <mps> 5.8.9
2020-09-12 17:28:37 <mps> nothing serious, but anyway already prepared, just waiting for test build before push
2020-09-12 17:29:13 <c705> thanks!
2020-09-12 18:03:08 <mps> c705: pushed. hope it will be on mirrors in about one hour
2020-09-12 18:04:03 <mps> c705: and, no, you don't have to sorry: `uname -a` => Linux zarya 5.9.0-rc4-3-gru #5 SMP PREEMPT Thu Sep 10 16:26:28 UTC 2020 aarch64 GNU/Linux
2020-09-12 18:04:29 <mps> so, 5.9 is out, -rc though :)
2020-09-12 18:56:59 <c705> mps: yeah, but mainline only
2020-09-12 18:57:03 <c705> not stable yet
2020-09-12 18:58:04 <mps> yes, but that is how these things works
2020-09-12 18:58:32 <c705> of course
2020-09-12 19:31:05 <mps> there are no menus in RPis bootloader?
2020-09-13 03:13:55 <abenz> is it just me or is firefox getting stuck even more now (after bump to 80) ?
2020-09-13 05:37:56 <bluezinc> abenz: I think that's a fair observation across the board, not just for alpine.
2020-09-13 05:42:50 <abenz> bluezinc: I see, I assumed it was due to the glibc/dbus/etc bugs that were mostly triggered by video playback
2020-09-13 05:44:42 <bluezinc> unfortunately not.  It's been happening on my ubuntu/debian boxes as well.
2020-09-13 05:46:16 <abenz> I see
2020-09-13 08:29:00 <nacelle> i've not seen this, but i'm running 80.0.1
2020-09-13 08:29:29 <nacelle> and I use it frequently
2020-09-13 09:23:23 <MY-R> ye 80 and 80.0.1 working great again on x86_64
2020-09-13 09:40:19 <abenz> I'm running 3.12, 80 is the latest available
2020-09-13 10:32:29 <Gnjurac> hi anyone running an arm64 alpine linux in qemu, am trying to setup so i dont have to bloat my rpi4 with build tools and stuff
2020-09-13 10:33:53 <mps> I posted guides few times here
2020-09-13 10:34:09 <mps> look at irclogs.alpinelinux.org
2020-09-13 10:49:32 <Gnjurac> ok will try
2020-09-13 10:50:37 <Gnjurac> mps: any fast wahy to go ower logs whout writing a script?
2020-09-13 10:54:06 <Gnjurac>  <mps> qemu-system-arm  -machine virt  -cpu cortex-a15 -m 512 -bios boot/u-boot.bin  -drive if=none,file=armv7-work.img,id=disk1,fo...........
2020-09-13 10:54:11 <Gnjurac> found this
2020-09-13 10:55:34 <Gnjurac> tough thing that confuses me is  -bios/u-boot.bin but i have alpine-virt-3.12.0-aarch64.iso file so from where does it get this bios file
2020-09-13 10:55:42 <Gnjurac> do i have to extract iso first?
2020-09-13 10:57:24 <Gnjurac> mps: you alive, help your neighboor so your cow lives xD
2020-09-13 10:58:28 <jjuran> Please don't gore anyone's ox :-)
2020-09-13 11:02:33 <Gnjurac> and question number 2 why is nim so old in repo?
2020-09-13 11:04:04 <Gnjurac> like i am 1.2.5 on void-linux and thats stable version, in alpine packages nim is like ver 0.17
2020-09-13 11:08:16 <ikke> Gnjurac: Depends on the maintainer of that package
2020-09-13 11:08:21 <ikke> if there is any
2020-09-13 11:10:02 <Gnjurac> so i have to find that Jakub Jirutka and bring him to my basment for forced labour
2020-09-13 11:11:14 <ikke> Alternatively, you provide a merge request yourself
2020-09-13 11:15:21 <mps> Gnjurac: https://tpaste.us/Kywn
2020-09-13 11:15:37 <mps> sorry, I don't have time to explain details now
2020-09-13 11:16:41 <mps> look at comments and try to make it working, it is not complicated though not for linux newbies
2020-09-13 11:47:14 <Gnjurac>  /usr/share/ovmf ? what is this dir i dont even have this dir on my pc
2020-09-13 11:47:26 <Gnjurac> and xbps-query -RS ovmh dosent give me anything
2020-09-13 11:48:33 <Gnjurac> *ovmf
2020-09-13 12:13:53 <mps> apk info ovmf => Open Virtual Machine Firmware (OVMF) BIOS
2020-09-13 16:42:51 <Gnjurac> kmeeeeee
2020-09-13 18:34:42 <Gnjurac> kmeeeeeee
2020-09-13 18:35:24 <Gnjurac> do i need to isntall ubuntu to get that /usr/share/ovmf
2020-09-13 18:56:16 <MY-R> Gnjurac, enable "community" repository and install package "ovmf"
2020-09-13 18:57:56 <Gnjurac> there is no in void repo :(
2020-09-13 18:58:11 <Gnjurac> maybe i can build it from source
2020-09-13 19:06:05 <Hello71> ...
2020-09-13 19:54:20 <Gnjurac> why ... ?
2020-09-13 19:59:22 <charl> I think you should ask in #voidlinux
2020-09-13 19:59:58 <Gnjurac> i am trying to run alpine-linux in quemu so i started here
2020-09-13 20:00:32 <Gnjurac> need it to cross copile to aarm64
2020-09-13 20:01:10 <charl> I was talking about the ovmf question
2020-09-13 20:02:59 <c705> but your ovmf issue is an issue with your host system
2020-09-13 20:05:21 <Gnjurac> ye i eat poop ty
2020-09-13 20:06:37 <MY-R> https://github.com/void-linux/void-packages/pull/17225
2020-09-13 20:09:24 <c705> ...
2020-09-13 20:40:05 <charl> how to make busybox less display colors?
2020-09-13 20:44:54 <c705> what?
2020-09-13 20:46:20 <charl> How do i get colored output from busybox less?
2020-09-13 20:47:04 <zcrayfish> charl: hint: ls --help
2020-09-13 20:47:05 <charl> To view colored man pages when I use it as a PAGER for example.
2020-09-13 20:47:13 <zcrayfish> oops I misread sorry
2020-09-13 20:47:40 <charl> PAGER for man*
2020-09-13 20:50:01 <charl> From the source I see busybox is configured with LESS_RAW and LESS_ENV but when I 'export LESS=-R' I see no colors
2020-09-13 20:55:33 <charl> the less from apk works out of the box but it still bugs me that the busybox one is not working
2020-09-13 20:56:46 <zcrayfish> I don't see how to make the less work like you want either offhand :/
2020-09-13 20:57:06 <zcrayfish> when I do, for instance: ls -la --color=always | less
2020-09-13 20:57:10 <zcrayfish> the color is stripped :(
2020-09-13 20:57:36 <afontain_> `| less -R` ?
2020-09-13 20:58:21 <charl> no difference
2020-09-13 21:02:16 <sm_urf> I'm having trouble PXE booting Alpine.  My host uses a USB adapter for networking.  Would I need to rebuild initramfs?
2020-09-13 21:15:56 <u0jQx9gPyrYg> does busybox less even support ansi escape sequences?
2020-09-13 21:26:41 <zcrayfish> One would think so judging from 541-557 of the source. https://git.busybox.net/busybox/tree/miscutils/less.c
2020-09-13 21:36:11 <gjabell> did anbox recently break for anyone else?
2020-09-13 21:37:01 <afontain_> how did it break?
2020-09-13 21:37:42 <gjabell> getting this when trying to run an app: Application manager service is not running yet
2020-09-13 21:38:22 <afontain_> I pushed something yesterday to make it installable again: https://gitlab.alpinelinux.org/alpine/aports/-/commits/master
2020-09-13 21:38:23 <afontain_> ah
2020-09-13 21:38:25 <afontain_> so it's not that
2020-09-13 21:38:41 <afontain_> can you start the app manager manually?
2020-09-13 21:39:05 <afontain_> *session manager
2020-09-13 21:39:06 <afontain_> `anbox session-manager` in a graphical terminal, as your user
2020-09-13 21:40:17 <gjabell> that returns: Failed to start as either binder or ashmem kernel drivers are not loaded
2020-09-13 21:40:27 <gjabell> wonder if this is an issue with running the edge kernel
2020-09-13 21:40:45 <afontain_> yup, edge kernel has neither binder or ashmem
2020-09-13 21:41:19 <gjabell> ahh
2020-09-13 21:41:30 <gjabell> alright, I'll reboot into the other kernel and see if that fixes it, thanks
2020-09-13 21:52:46 <charl> u0jQx9gPyrYg zcrayfish apparently it doesn't actually interpret ansi sequences. It strips them to produce readable output
2020-09-13 21:54:52 <charl> less in main repo works fine though if anyone wants colors
2020-09-13 21:57:15 <u0jQx9gPyrYg> that's what i expected
2020-09-13 21:57:20 <zcrayfish> And it's not GNU? cool
2020-09-13 21:58:58 <charl> it is GNU
2020-09-13 22:00:43 <zcrayfish> oh, yeah it is actually... I see the GNU site just redirects to  www.greenwoodsoftware.com/less/
2020-09-14 03:40:32 <codebam> has anyone here used zoom on alpine? I don't want to click the join because the meeting is tomorrow morning. just trying to figure out if it's web based and will work in firefox?
2020-09-14 04:43:06 <abenz> codebam: there is a web client with limited functionality.. so it should work, I would check audio/video if you will be expected to use those though..
2020-09-14 04:44:15 <abenz> alternatively if you need full fuctionality, a qemu vm will work nice, I used zoom on a windows VM running on alpine (use qxl for display driver)
2020-09-14 05:13:57 <abenz> PureTryOut[m]: just wanted to say thanks for trying to sort the kdenlive issue with upstream, I'm following closely
2020-09-14 07:13:12 <PureTryOut[m]> abenz: np!
2020-09-14 07:41:59 <charl> after regenerating grub config consolefont refuses to change my font on boot
2020-09-14 07:42:50 <charl> the only thing I changed was change GRUB_TIMEOUT from 2 to 0 and then grub-mkconfig'd
2020-09-14 07:44:49 <charl> any idea how to debug this?
2020-09-14 08:26:45 <MY-R> charl, I would start from checking if font which you set in "/etc/conf.d/consolefonts" exist in "/usr/share/consolefonts"
2020-09-14 08:27:49 <charl> it does exist and the font changes if I setfont or if I restart the consolefont service
2020-09-14 08:31:56 <MY-R> weird
2020-09-14 14:00:05 <Gnjurac> i booted alpine in qemu just ignored that ovmf booted non efi one i think
2020-09-14 14:01:17 <arew> hello, I installed alpine linux. I have a problem with fonts rendering in the terminal and firefox
2020-09-14 14:01:32 <arew> some utf8 characters are not displayed.
2020-09-14 14:01:54 <arew> What font do you recommend for large coverage of utf8 (including emoji)
2020-09-14 14:02:31 <arew> I assume it is a font problem
2020-09-14 14:03:07 <arew> At the moment firefox is configure with 'Nimbus Sans L'
2020-09-14 14:03:48 <arew> And allacrity use an Ubuntu font
2020-09-14 14:04:07 <zcrayfish> arew: unless someone else has a better suggestion, I'd probably just download  Noto Emoji (or Noto Color Emoji)
2020-09-14 14:04:10 <zcrayfish> https://www.google.com/get/noto/#emoji-zsye
2020-09-14 14:04:14 <zcrayfish> YMMV. But that is the one I use.
2020-09-14 14:21:25 <Gnjurac> i am getting  * ERROR: modloop failed to start
2020-09-14 14:21:35 <Gnjurac> but it stills boots, is this a big issue?
2020-09-14 14:59:02 <Gnjurac> i think this moodloop is for drivers cuz from what i see i cant mount drive now
2020-09-14 14:59:19 <arew> zcrayfish: what about the other chars that are not displayed in firefox, in particular, the open file dialog only display a square with hex digits
2020-09-14 15:00:59 <zcrayfish> unifont would be my first choice to cover all the non-emoji stuff
2020-09-14 15:01:06 <zcrayfish> unifont is in alpines repos too
2020-09-14 15:17:11 <arew> I think it used to work. It stopped working when I installed some gnome stuff :/
2020-09-14 16:17:23 <xmastree> Where does one pass boot options to the standard alpine image? I've got a real old machine here that is getting stuck on the installing packages stage. The error used the word "devant" I think it was which is a ballerina term google says.
2020-09-14 16:19:35 <xmastree> I thought this might be something for the experts, because at first I was all like wtf. But it seems the old HDD is having acpi issues among other issues.
2020-09-14 16:22:36 <xmastree> compaq armada 110, and alpine ran smooth as butter on this relic. Anyone else tried Alpine on a similarly old machine?
2020-09-14 16:25:31 <xmastree> (anyone know when the next alpine release is scheduled?)
2020-09-14 16:26:37 <c705> boot options can go on the kernel cmdline
2020-09-14 16:27:14 <xmastree> c705 yes, but where is that on booting the standard image?
2020-09-14 16:28:11 <xmastree> if I hit buttons soon enough can I use the ISOLINUX boot: prompt and if so are there any differences?
2020-09-14 17:36:43 <Gnjurac> hmm can i make qemu alpine imge have bigger root size
2020-09-14 17:36:54 <Gnjurac> i was compiling nim and just now came
2020-09-14 17:36:56 <Gnjurac> rror: unhandled exception: errno: 28 `No space left on device` [IOError]
2020-09-14 17:36:59 <Gnjurac> kmeeeeeeeeeee
2020-09-14 17:37:16 <Gnjurac> worst thing i think it was wery little short of space
2020-09-14 17:38:25 <mps> Gnjurac: create new bigger image, partition it and format it, and copy old one to new one
2020-09-14 17:39:19 <Gnjurac> how
2020-09-14 17:39:38 <Gnjurac> and do you maybe know why i get that i am getting  * ERROR: modloop failed to start
2020-09-14 17:39:58 <Gnjurac> maybe cuz i not running uefi?
2020-09-14 17:40:25 <mps> or, qemu-img resize
2020-09-14 17:41:17 <mps> man qemu-img, is your friend (or foe if you feel it as such)
2020-09-14 17:41:52 <mps> you can download ovmf from alpine, unpack it and use it on other systems
2020-09-14 17:42:16 <mps> but you have to download ovmf for destination arch you use
2020-09-14 17:43:04 <Gnjurac> http://ix.io/2xz7
2020-09-14 17:44:03 <Gnjurac> for void there is no ovmf, they are atm hacking it using arch ovmf but again from what i saw there is no bios.bin in arch one files
2020-09-14 17:45:10 <Gnjurac> hmm i made 4gb img but its not mounted alpine boots to ram
2020-09-14 17:45:57 <mps> use ovmf from alpine, I tested it and it works
2020-09-14 17:46:41 <c705> i concur
2020-09-14 17:46:49 <c705> ovmf works
2020-09-14 17:46:53 <mps> actual file is /usr/share/OVMF/QEMU_EFI.fd
2020-09-14 17:52:00 <Gnjurac> from where can i get that one
2020-09-14 17:52:55 <Gnjurac> https://github.com/tianocore/tianocore.github.io/wiki/EDK-II ?
2020-09-14 17:53:05 <Gnjurac> https://pkgs.alpinelinux.org/package/edge/community/x86_64/ovmf
2020-09-14 17:53:10 <Gnjurac> there no dl button
2020-09-14 17:53:35 <Gnjurac> ok found it in mirrors i think
2020-09-14 17:54:48 <Gnjurac> ovmf-0.0.201908-r0.apk how to extract this
2020-09-14 17:55:07 <mps> not x86_64, but aarch64
2020-09-14 17:55:23 <mps> tar xf filename.apk
2020-09-14 17:55:41 <Gnjurac> ok it gived me soem errors but gess i can ingore this
2020-09-14 17:55:45 <Gnjurac> cheksucm just
2020-09-14 17:55:52 <Gnjurac> tar: Ignoring unknown extended header keyword 'APK-TOOLS.checksum.SHA1'
2020-09-14 17:57:26 <Gnjurac> ok this one has bios
2020-09-14 18:02:55 <Gnjurac> hmm  * ERROR: modloop failed to start
2020-09-14 18:02:59 <Gnjurac> got this again
2020-09-14 18:03:24 <Gnjurac> even i added -bios SMRT/usr/share/ovmf/bios.bin
2020-09-14 18:03:46 <Gnjurac> SMRT is ofc dir here
2020-09-14 18:04:18 <Gnjurac> what to do with athat usr/share/OVMF/QEMU_EFI.fd
2020-09-14 18:04:33 <Gnjurac> -append "modloop=modloop-lts alpine_repo=http://dl-cdn.alpinelinux.org/alpine/latest-stable/main" \
2020-09-14 18:05:08 <Gnjurac> and about this i just hacked this moodloop but why do i have to point alpine_repo=...
2020-09-14 18:05:43 <mps> alpine repo is needed on install to fetch packages
2020-09-14 18:07:03 <mps> -bios QEMU_EFI.fd (wherever it is)
2020-09-14 18:07:49 <Gnjurac> ok sec
2020-09-14 18:12:44 <Gnjurac> still am getting that modloop error
2020-09-14 18:25:59 <mps> Gnjurac: I can't help much running qemu alpine arm on other distros
2020-09-14 18:26:36 <mps> install alpine on your workstation and then maybe I and other people here will/can help you
2020-09-14 18:26:40 <Gnjurac> kmeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee
2020-09-14 18:27:03 <Gnjurac> but if i install alpine on my workstation then i cant play games :)
2020-09-14 18:27:11 <Gnjurac> no steam
2020-09-14 18:27:18 <mps> who cares for games
2020-09-14 18:27:18 <ikke> flatpak
2020-09-14 18:27:21 <artok> flatp
2020-09-14 18:27:37 <mps> artok: \o
2020-09-14 18:27:37 <Gnjurac> flatpack works with steam?
2020-09-14 18:27:48 <artok> hello! \o
2020-09-14 18:27:53 <Gnjurac> but arent games saved localy?
2020-09-14 18:28:09 <mps> artok: can you try linux-edge on your RPi2?
2020-09-14 18:28:09 <Gnjurac> do games then go to flatpack too?
2020-09-14 18:28:23 <artok> i have rpi3b+ and rpi4
2020-09-14 18:28:30 <mps> ah
2020-09-14 18:28:36 <artok> on those, yes
2020-09-14 18:28:39 <Gnjurac> i have zero w and 4 i win
2020-09-14 18:28:44 <Gnjurac> now give me power
2020-09-14 18:29:52 <mps> linux-edge armv7 have basic things for RPis, but I think only RPi2
2020-09-14 18:29:54 <Gnjurac> 1 moree issue dont thing i will be able to run Unity on alpine :(, godot maybe
2020-09-14 18:30:15 <Gnjurac> does even vscode work on alpine?
2020-09-14 18:30:31 <Gnjurac> vscodium maybe
2020-09-14 18:30:45 <Gnjurac> musl is hard :(
2020-09-14 18:30:52 <mps> Gnjurac: If you are ready to hack and think I can post you script which I used to install aarch64 under qemu on x86_64 host
2020-09-14 18:31:29 <mps> it was with u-boot for qemu
2020-09-14 18:31:41 <Gnjurac> what
2020-09-14 18:32:02 <mps> Gnjurac: read here first http://arvanta.net/mps/
2020-09-14 18:33:12 <mps> and don't expect much from me this night, I'm on fourth glass of red wine ;) (cheers)
2020-09-14 18:34:19 <Gnjurac> why dirnk wine, worst hangower are from wine, tekila only good alocohol
2020-09-14 18:34:43 <Gnjurac> anywya alcoholl is bad, cheers with water :)
2020-09-14 18:35:57 <artok> gonna get me beer
2020-09-14 18:36:18 <artok> then might also have some red
2020-09-14 18:36:32 <artok> then start emacs =)
2020-09-14 18:36:43 <mps> artok: :)
2020-09-14 18:37:29 <mps> make sense, only if I'm drunk will think of emacs
2020-09-14 18:43:27 <artok> tried working with VS Code, atom, sublime.. somehow emacs does better for me
2020-09-14 18:46:23 <artok> but nothing works without alcohol =)
2020-09-14 18:49:20 <mps> yes, alcohol is the main fuel for the civilization
2020-09-14 19:24:36 <yann-kaelig> Hi, looking for mariadb-connector-j on Alpine. Is it available, perhaps with another naming ?
2020-09-14 19:27:12 <c705> yann-kaelig: just download the driver off their site
2020-09-14 20:03:16 <gjabell> is it possible to extend service definitions and configurations in /etc/{init.d,conf.d}/ without modifying the upstream copies?
2020-09-14 20:03:25 <gjabell> ie to prevent having *.apk-new copies
2020-09-14 20:03:41 <ikke> No, as far as I know, there is not
2020-09-14 20:04:06 <gjabell> ok, thanks
2020-09-15 02:31:37 <twb> Remind me - while running alpine live, what's the thing I put into /etc/apk/repositories?
2020-09-15 02:32:07 <twb> After installing to a disk, it puts comments in there I can uncomment, but there's no comment there when running live :-(
2020-09-15 02:34:33 <twb> Looks like I can just run "setup-apkrepos" and it'll do it for me
2020-09-15 02:43:52 <twb> (FYI the context was some dumb docker user and I was like "here's a oneliner that doesn't suck")
2020-09-15 02:44:37 <twb> "ip link set eth0 up; udhcpc; setup-apkrepos; apk update; apk add speedtest-cli; speedtest-cli"  instead of whatever javascript would show here: https://hub.docker.com/r/tianon/speedtest/dockerfile
2020-09-15 07:01:16 <Guest95> Can somebody explain what I'm doing wrong? I'm trying to synchronize the "git aports" repository locally with the remote one, but I only get "Already up to date." But from https://git.alpinelinux.org/aports/ I see that there are changes...
2020-09-15 07:02:17 <ikke> Guest95: hard to say without more details
2020-09-15 07:02:18 <Guest95> git fetch && git pull
2020-09-15 07:03:04 <ikke> git pull also does a git fetch. So when you fetch before that, git pull will not actually fetch something
2020-09-15 07:03:12 <ikke> git merge @{upstream}
2020-09-15 07:03:14 <Guest95> Oh...
2020-09-15 07:03:19 <ikke> git merge --ff-only @{upstream}
2020-09-15 07:03:50 <Guest95> ikke "Already up to date."
2020-09-15 07:04:01 <ikke> What does git status return
2020-09-15 07:04:24 <Guest95> ikke On branch master
2020-09-15 07:04:24 <Guest95> Your branch is up to date with 'origin/master'.
2020-09-15 07:04:25 <Guest95> nothing to commit, working tree clean
2020-09-15 07:04:42 <ikke> git rev-parse HEAD
2020-09-15 07:04:56 <Guest95> ikke 426d80579bcb4b8645f1b89472385030ad8b9ba5
2020-09-15 07:05:19 <ikke> git remote -v
2020-09-15 07:05:36 <Guest95> origin  git://dev.alpinelinux.org/aports (fetch)
2020-09-15 07:05:37 <Guest95> origin  git://dev.alpinelinux.org/aports (push)
2020-09-15 07:07:31 <Guest95> Also, i didnt change anything in a repo, just clone it long time ago.
2020-09-15 07:07:47 <ikke> This is an old url, apparently the repo there is not updated anymore
2020-09-15 07:08:01 <ikke> git remote set-url origin https://gitlab.alpinelinux.org/alpine/aports.git
2020-09-15 07:08:18 <Guest95> But ... https://wiki.alpinelinux.org/wiki/Aports_tree
2020-09-15 07:08:42 <ikke> it points to git.alpinelinux.org, not dev.alpinelinux.org
2020-09-15 07:08:48 <Guest95> Ah, well, now i see.
2020-09-15 07:09:20 <ikke> git.alpinelinux.org is still up-to-date
2020-09-15 07:11:01 <Guest95> ikke now its works! thanks alot, have a good day :3
2020-09-15 07:11:07 <ikke> you as well
2020-09-15 07:53:30 <sxiii> Greetings everybody :) I am trying to build a hobby project based on alpine and I stuck a little bit with graphics driver autodetection problem... Is there anybody who could lead me the correct way? I could try to describe my goal and steps
2020-09-15 07:55:56 <Cogitri> Autoloading modules would usually be done via udev
2020-09-15 07:57:22 <sxiii> Cogitri: is there a possibility to make he live image for alpine that will detect amdgpu, nvidia or intel cards and load the according driver or is it real only with some very-generic one?
2020-09-15 08:00:59 <twb> sxiii: surely that already happens?
2020-09-15 08:01:26 <Cogitri> The live image certainly contains the required modules for that
2020-09-15 08:01:45 <Cogitri> But I'm not sure if udev is enabled on the live images (or if mdev is setup properly do to the trick as well)
2020-09-15 08:01:50 <twb> maybe you want the sucky non-free drivers insead of the reliable ones?
2020-09-15 08:02:31 <Cogitri> twb: For Nvidia that statement isn't true unfortunately, the OSS drivers aren't great especially for new cards
2020-09-15 08:03:21 <twb> nouveau is unreliable on new cards?
2020-09-15 08:05:49 <Cogitri> Yup
2020-09-15 08:07:09 <Cogitri> And since Nvidia doesn't release signed firmware for cards >=GTX 1xxx (actually not sure if they even did it for GTX 9xx cards) reclocking doesn't work :/
2020-09-15 08:09:10 <twb> Sounds like a good reason not to run nvidia except in VGA mode
2020-09-15 08:10:04 <Cogitri> Sounds like a good reason to use AMD or Intel, yes :D
2020-09-15 08:10:32 <Cogitri> sxii: Anyway, sorry for going a bit offtopic. What exactly is your problem? The modules don't load automatically?
2020-09-15 08:13:49 <sxiii> twb: I don't care for the type of the drivers; what I need is to make sure I can use full resolution on both 1 and 2 screens.
2020-09-15 08:15:04 <qman__> it depends
2020-09-15 08:15:17 <sxiii> My current issue is when I boot alpine ISO on my machine with Radeon HD5850, it just cannot initialize X. What I really need is fullscreen "sent" program from suckless tools open (I probably don't even need a window manager) as well as possibility to open xpdf.
2020-09-15 08:15:32 <sxiii> This is intended to make presentations. So I could boot right into the presenation.
2020-09-15 08:15:46 <qman__> yeah, the older radeon HDs don't "just work" anymore
2020-09-15 08:16:01 <Cogitri> They should, the radeon module is still a thing
2020-09-15 08:16:06 <qman__> I gave up and swapped in a GTX970 instead of my 7950
2020-09-15 08:16:45 <qman__> I don't know what changed, but they don't, experienced it myself
2020-09-15 08:17:19 <TBB> 
ACTION moves his "year of the Linux desktop" prediction marker 5 years forward, again

2020-09-15 08:17:27 <qman__> I spent about three days trying to get dual monitors to work
2020-09-15 08:17:30 <sxiii> Hmmm... So two ways for me then? To try to debug the old radeon or just swap the card to something other :D ?
2020-09-15 08:17:32 <Cogitri> I have a HD6670 around for when my server doesn't boot (so I can debug the thing) and that just works
2020-09-15 08:17:45 <sxiii> I would be actually okay with some generic VESA driver if it can handle resolution
2020-09-15 08:18:21 <qman__> it works on default console resolutions, but when you try to set up X with dual monitors, hard nope
2020-09-15 08:18:46 <qman__> I'm not kidding when I say I spent about 3 days screwing with it
2020-09-15 08:18:52 <sxiii> I actually have a problem even with a single screen. It doesn't work :) With two monitors, it would be a "next quest"...
2020-09-15 08:19:03 <abenz> is there any reason why firefox on 3.12 isn't bumped to 80.0.1 ? Given its unusable as is, and 80.0.1 is already on edge (work already done?)
2020-09-15 08:19:16 <qman__> put in the 970 and it worked perfect, first try
2020-09-15 08:19:27 <qman__> I'm also guessing amdgpu compatible cards will just work
2020-09-15 08:19:38 <qman__> but something broke with the older radeon HDs
2020-09-15 08:19:45 <twb> This is why last time I did presentations, I just used beamer or s5 and dumped presentation.pdf on an HTTP server somewhere
2020-09-15 08:19:48 <Cogitri> abenz: Probably that no one did the work for backporting it/I forgot about it when bumping it on edge
2020-09-15 08:19:59 <abenz> qman__: are you aware amd ported amdgpu to older gcn cards recently ? (I think 5.9)
2020-09-15 08:20:27 <abenz> I think your 7950 would work on amdgpu (with full acceleration and vulkan support)
2020-09-15 08:20:35 <qman__> possibly if that's a recent change
2020-09-15 08:20:38 <qman__> I did this setup on 3.11
2020-09-15 08:20:51 <abenz> kernel 5.9 iIRC
2020-09-15 08:21:10 <sxiii> abenz: do you know what cards the backport covers?
2020-09-15 08:21:52 <qman__> alpine 3.11, that is, not the kernel
2020-09-15 08:23:57 <abenz> not my ideal choice to read tech news, but this website posted a quick summary:
2020-09-15 08:24:03 <sxiii> Oh I found an artice about this on linuxreviews org - but it seems that only "more recent" cards are supported
2020-09-15 08:24:12 <abenz> https://linuxreviews.org/Linux_5.9_Brings_Hardware_Accelerated_Video_Playback_To_10+_Year_Old_AMD_Graphics_Cards_Using_The_%22amdgpu%22_Linux_Kernel_Driver
2020-09-15 08:24:21 <sxiii> Yeah that exact article ;)
2020-09-15 08:24:47 <abenz> Cogitri: ah so it would require some work to port to 3.12? I thought it would be a simply bump
2020-09-15 08:25:38 <abenz> s,simply,simple,g
2020-09-15 08:26:27 <Cogitri> It would require a cherry-pick onto the 3.12-stable branch
2020-09-15 08:26:36 <Cogitri> So it's not a lot of work, but someone has to do it :)
2020-09-15 08:32:41 <Cogitri> abenz: https://gitlab.alpinelinux.org/alpine/aports/-/merge_requests/12473
2020-09-15 08:33:23 <mps> use edge, Luke ;)
2020-09-15 08:34:30 <mps> linux-edge 5.9 will be in repo in about 2-3 weeks
2020-09-15 08:46:39 <abenz> I see
2020-09-15 08:47:40 <Cogitri> FF 80.0.1 should be in 3.12 once the builders are done building it now, so probably 1~2 hours
2020-09-15 08:48:33 <abenz> many thanks, the changelogs talk about some fixes, but we'll see about crashes
2020-09-15 08:48:39 <abenz> is it usable on your end?
2020-09-15 08:49:33 <Cogitri> I didn't have any crashes with FF for months on edge :)
2020-09-15 08:49:45 <Cogitri> But that means I didn't have problems with FF 80 either
2020-09-15 08:50:11 <mps> I confirm
2020-09-15 08:50:27 <Cogitri> The crashes or it being stable? :D
2020-09-15 08:51:47 <abenz> Cogitri: oh, I thought it was crashing for eveyone (including non-alpine users)
2020-09-15 08:52:03 <abenz> with mozilla downsizing and all, some were even jumping ship
2020-09-15 08:52:36 <abenz> it crashes over 20 times a day for me :(
2020-09-15 08:53:32 <mps> Cogitri: stable, ofc. Only have oom sometimes on -esr in 3.12 aarch64
2020-09-15 13:06:58 <oliv3r[m]> is there a 'apk download' file to only download an apk file (for any arch)?
2020-09-15 13:07:56 <sxiii> oliv3r[m]: I think you'd just "wget repo_url/packagename" :)
2020-09-15 13:08:36 <oliv3r[m]> lol, well yes, but i'd want it to take into account /etc/apk/repo and do signature checks and all that :p but yes, that is a way
2020-09-15 13:08:49 <mps> apk fetch pkgname
2020-09-15 13:09:16 <sxiii> yeah; I just also checked "apk fetch" routine :)
2020-09-15 13:09:59 <mps> apk --arch architecture fetch pkgname
2020-09-15 13:11:22 <oliv3r[m]> of course! thanks mps
2020-09-15 13:11:53 <mps> oliv3r[m]: you are welcome :)
2020-09-15 13:12:47 <oliv3r[m]> next question will probably to see if i can force an install of this package :p but that part I should be able to figure out :D
2020-09-15 13:13:51 <mps> apk --root / --arch add .....
2020-09-15 13:14:01 <mps> not sure will this work
2020-09-15 13:14:18 <oliv3r[m]> i know installing an --arch package (without first fetching it) won't work
2020-09-15 13:14:23 <mps> apk --root / --arch other-arch add .....
2020-09-15 13:15:02 <mps> didn't tried this
2020-09-15 13:15:36 <oliv3r[m]> yeah it will just download your current arch; but even fetch, will download your current arch
2020-09-15 13:16:30 <oliv3r[m]> so i have to setup a fake root, download it, and then see what's possible :S
2020-09-15 13:16:55 <oliv3r[m]> it's kinda annoying that the --arch flag sounds like it would 'just work' but is restricted to your current arch only :(
2020-09-15 13:18:18 <mps> it works for me when I make other arch lxcs and putting other arch on disk/mmc/usb
2020-09-15 13:21:26 <oliv3r[m]> hmm, on edge it doesn't seem to be working :(
2020-09-15 13:22:44 <oliv3r[m]> / # apk --root / --arch arm32v7 add klibc-dash
2020-09-15 13:22:44 <oliv3r[m]>  /usr/lib/klibc-x86_64/
2020-09-15 13:22:45 <oliv3r[m]> (1/2) Installing libklibc (2.0.8-r0)
2020-09-15 13:23:25 <mps> few days ago I bootstrapped armhf 3.12 on aarch64 edge
2020-09-15 13:23:43 <oliv3r[m]> if however, i create a /tmp/arm32v7 and add it that, that works of course
2020-09-15 13:23:51 <mps> --arch armv7 ?
2020-09-15 13:23:56 <oliv3r[m]> yes, bootstrapping! that works :)
2020-09-15 13:24:44 <oliv3r[m]> mips: / # apk --root / --arch potato add klibc-dash
2020-09-15 13:24:44 <oliv3r[m]> (1/2) Installing libklibc (2.0.8-r0)
2020-09-15 13:24:51 <mps> we don't have arch arm32v7
2020-09-15 13:25:17 <oliv3r[m]> i know, docker vs alpine always gets confusing :) I actually like the docker naming better :p
2020-09-15 13:26:13 <oliv3r[m]> in any case, I don't want to bootstrap; so i guess i have to fake that part. I think i can do /tmp/armv7 for the root, just copy the certs is the minimal needed thing?
2020-09-15 13:28:37 <mps> https://tpaste.us/b4rX is how I bootstrapped armhf on aarch64
2020-09-15 13:30:16 <oliv3r[m]> yeah bootstrapping is easy; i just want to 'cross-install' packages :p which isn't supported (yet)
2020-09-15 13:30:23 <oliv3r[m]> mps but this works so far: apk --root /tmp/arm32v7 --initdb --arch armv7 add klibc-dash busybox --repository=htt
2020-09-15 13:30:23 <oliv3r[m]> g/alpine/edge/main --allow-untrusted
2020-09-15 13:30:23 <oliv3r[m]> p://dl-cdn.alpinelinux.org/alpine/edge/testing/ --repository=http://dl-cdn.alpinelinux.or
2020-09-15 13:30:48 <oliv3r[m]> though I guess the busybox thing cannot be avoided due to the postinstall /bin/sh need
2020-09-15 13:31:26 <mps> yes
2020-09-15 13:36:50 <oliv3r[m]> sad, but that's fine; though klibc-dash-static could provide /bin/sh :D
2020-09-15 13:43:22 <oliv3r[m]> aww, doesn't work; but then it wasn't intended to be used that way
2020-09-15 14:06:48 <oliv3r[m]> mps: do you know what the minimal thing is we can do to initialize the db? i only know add --initdb busybox :( but i want something less even; as i can then do apk fetch --arch --root /tmp :p
2020-09-15 14:09:21 <oliv3r[m]> guess alpine-keys is the smallest one :)
2020-09-15 14:56:53 <mps> oliv3r[m]: did you tried without any pkg? not sure it will work but doesn't cost to try
2020-09-15 16:08:12 <tux19> hi everyone! I have tried to install alpine on a UEFI laptop alongside with other two OS, I already use rEFInd as bootloader. I have mounted the partition where I wanted to install Alpine and run the setup. However, when rebooting, I had a kernel panic and a message like " /sysroot no such file or directory". My arch is x86_64
2020-09-15 16:22:17 <Hello71> I don't think it's possible to get those messages together
2020-09-15 16:31:56 <tux19> i don't remember exactly, but i couldn't boot alpine..is there anything i have to care about when i install it for a multiboot?
2020-09-15 19:26:32 <maerwald> it seem bash is broken in i386/alpine:edge
2020-09-15 19:27:11 <ikke> maerwald: musl needs to be updated
2020-09-15 19:27:28 <ikke> ncopa will do an edge snapshot soon that will fix it
2020-09-15 19:27:41 <ikke> but for now, do an apk upgrade -U (or apk add -u musl)
2020-09-15 19:28:17 <maerwald> that worked
2020-09-15 22:32:11 <bbuccianti> firefox doesn't detect my mic
2020-09-15 22:32:23 <bbuccianti> can't find any resource on setting it properly
2020-09-15 22:32:41 <bbuccianti> I can record something with arecord -Ddefault test.wav
2020-09-15 22:32:57 <bbuccianti> and it works properly when I aplay -Ddefault test.wav
2020-09-15 22:33:03 <bbuccianti> I can hear myself talking
2020-09-15 22:33:15 <bbuccianti> why it doesn't show in firefox?
2020-09-15 22:59:57 <Hello71> did you install pulse
2020-09-16 03:39:29 <u0_a305> Anyone here using Krita on Alpine
2020-09-16 08:04:40 <JD_the_fish> Im trying to use an answer file for setup-alpine. How do i make it accept that dns is configured via dhcp?
2020-09-16 13:19:30 <sxiii> Greetings guys :) I am building a specialized ISO build of Alpine for presentation purposes... And I'm a bit stuck
2020-09-16 13:20:17 <sxiii> I think I basically have two options: either to use the sources to build ISO as told on Alpine Wiki page about mkimage.sh
2020-09-16 13:20:54 <sxiii> Or, to build my ISO image from current live system. This would be easier for me, but I don't know how to do that, sadly.
2020-09-16 13:22:18 <sxiii> So when I am using documentation from the Wiki, I am adding the package that I want (and it's dependencies) into three files: genapkovl-dhcp.sh (I took this one as I also need DHCP), mkimg.pres.sh (this is my "building profile") and, lastly, to live system (I add packages to the Dockerfile as I build in Docker).
2020-09-16 13:22:53 <sxiii> Finally, when I run the building, I got this: `#15 129.4 $apks: unable to select package (or its dependencies)
2020-09-16 13:22:53 <sxiii> failed to solve with frontend dockerfile.v0: failed to build LLB: executor failed running [/bin/sh -c sh mkimage.sh --tag edge --outdir ~/iso --arch x86_64 --repository http://dl-cdn.alpinelinux.org/alpine/edge/main --profile pres --repository http://dl-cdn.alpinelinux.org/alpine/edge/community]: runc did not terminate sucessfully
2020-09-16 13:22:53 <sxiii> ------
2020-09-16 13:22:54 <sxiii> `
2020-09-16 13:23:19 <sxiii> I'm using alpine:edge channel
2020-09-16 14:48:16 <cfengineuser> Hi, I'm new to awall and wondering if there is a way to use a dump of exiting firewall rules via iptables -L to make a set of equivalent awall configs
2020-09-16 15:15:38 <ikke> No, I don't think that's possible
2020-09-16 15:16:40 <ikke> awall config is on a higher level than iptables
2020-09-16 17:54:26 <maldridge> can awall target other backends?
2020-09-17 00:22:38 <enyc> Sean[m]1: meow ;p
2020-09-17 00:40:31 <bbuccianti> seems like firefox and firefox-esr are compiled with --disable-pulseaudio
2020-09-17 00:40:49 <bbuccianti> maybe that's why I can't use my mic?
2020-09-17 00:41:08 <bbuccianti> I'm trying to compile firefox-esr with --enable-pulseaudio to see if that fix it
2020-09-17 00:46:34 <twb> bbuccianti: last year firefox *only* supported audio via pulseaudio on linux
2020-09-17 00:46:44 <twb> bbuccianti: but I believe that has since been fixed and you can use plain alsa again
2020-09-17 00:48:57 <twb> Here on Debian 10, alsa, no pulse, firefox-esr=68.2.0 can do audio *out* at least.  /me double-checks
2020-09-17 00:51:38 <twb> Yep, confirmed using random test video: https://www.youtube-nocookie.com/embed/rZuwO2njgBY?wmode=transparent&iv_load_policy=3&autoplay=1&html5=1&showinfo=0&rel=0&modestbranding=1&playsinline=1&theme=light
2020-09-17 01:17:49 <Hello71> bbuccianti: I asked you and you never answered
2020-09-17 01:50:23 <bbuccianti> Hello71: sorry, you asked if I had pulseaudio installed
2020-09-17 01:51:04 <bbuccianti> twb: I'm trying to use a microphone to talk through some meeting browser apps
2020-09-17 01:51:24 <bbuccianti> output works fine
2020-09-17 01:52:08 <twb> bbuccianti: OK interesting
2020-09-17 01:52:23 <twb> bbuccianti: I did briefly try microphone (w/ firefox, no pulse) and it didn't Just Work, but I assumed it was something else I had done
2020-09-17 01:52:52 <bbuccianti> it didn't just work?
2020-09-17 01:53:09 <twb> I mean that I see the same symptoms as you
2020-09-17 01:53:24 <twb> audio output to speakers works, audio input from microphone doesn't work
2020-09-17 01:53:47 <bbuccianti> I'm trying to compile with --enable-pulseaudio to see if that makes a difference
2020-09-17 01:54:01 <twb> I did not test for very long, because complicated web pages tend to trigger a bug in my SSD firmware that hangs my entire laptop
2020-09-17 01:54:21 <bbuccianti> wow
2020-09-17 01:54:22 <twb> I was testing using https://meet.jit.si/$(xkcdpass)
2020-09-17 01:54:48 <bbuccianti> yeah, jitsi doesn't recognize a mic
2020-09-17 01:57:05 <bbuccianti> close to two hours compiling
2020-09-17 01:57:25 <bbuccianti> browser compilation suck
2020-09-17 02:05:24 <enyc> 
ACTION meows

2020-09-17 02:05:44 <yjftsjthsd> It really does. On gentoo, I always just grabbed the official binaries from mozilla because compiling would take ages, but that's not an option on Alpine unless possibly the glibc-compat stuff is up for it (and that seems a little dicey)
2020-09-17 02:17:50 <twb> I don't even have enough disk space to download the firefox sources, let alone compile them
2020-09-17 02:21:05 <yjftsjthsd> It also needs gobs of RAM to build:)
2020-09-17 02:21:29 <yjftsjthsd> (although officially I think 8GB works and that's not... too... terrible anymore)
2020-09-17 03:06:58 <baconicsynergy[m> Hello friends! My search engine kung fu has failed me and I was hoping for some advice. I followed all of the instructions from https://wiki.alpinelinux.org/wiki/KVM and started my vm with "qemu-system-x86_64 -enable-kvm -m 2048 -nic user,model=virtio -driv
2020-09-17 03:06:58 <baconicsynergy[m> e file=ubuntu20_04.qcow2,media=disk,if=virtio" However, running "virsh list" shows nothing, even when started as root!
2020-09-17 03:08:29 <bbuccianti> I can confirm that compiling firefox-esr (almost 3h in my pc) with --enable-pulseaudio allows mic to record audio
2020-09-17 03:08:31 <bbuccianti> :D
2020-09-17 03:15:01 <vivi> Choosing Alpine for our business was the best decision we've made - wanted to say thanks :D <3
2020-09-17 03:22:12 <twb> what business?
2020-09-17 03:22:57 <twb> baconicsynergy[m: "virsh list --all" can only see VMs managed by libvirtd.  Running qemu-system-x86_64 by hand is not managed by libvirtd.
2020-09-17 03:23:22 <vivi> twb: we sniff 802.11 for proximity to report metrics of nearby mobile phones
2020-09-17 03:24:20 <baconicsynergy[m> twb, oh!! I will research how to manage a VM with libvirtd :) Thank you very much!
2020-09-17 03:24:21 <twb> baconicsynergy[m: if you want to use virsh/virtd/virt-manager/gnome-boxes you need to create a "domain" for each VM you want to run, then "virsh start my-cool-vm", which will run qemu-system-x86_64 under the hood
2020-09-17 03:24:25 <vivi> this + gps data is sent over the 3g network (gRPC) where it's indexed in Elasticsearch and our customers can visualize
2020-09-17 03:24:35 <twb> baconicsynergy[m: do you have a linux GUI anywhere
2020-09-17 03:24:42 <twb> baconicsynergy[m: if so, by far easiest is to use virt-manager
2020-09-17 03:25:19 <baconicsynergy[m> I'm running GNOME successfully on Alpine Edge :) I'll check out virt-manager! So much to learn :o
2020-09-17 03:25:22 <twb> baconicsynergy[m: doesn't have to be the same host as virtd runs on, virt-manager can connect over SSH
2020-09-17 03:25:28 <baconicsynergy[m> oh okay!
2020-09-17 03:25:38 <vivi> so basically wifi+gps+modem golang -> grpc -> wireguard -> golang -> elasticsearch
2020-09-17 03:25:58 <twb> baconicsynergy[m: OK cool.  For local, you might need to add your GUI user to a system group like "kvm" or "libvirtd", I'm not sure.
2020-09-17 03:26:25 <twb> baconicsynergy[m: you can talk to #virt on irc.oftc.net for help with virt* stack.
2020-09-17 03:27:09 <twb> vivi: so the high-level user story is something like "tell me which store displays my shoppers linger in front of" ?
2020-09-17 03:27:54 <twb> vivi: I thought all phones now used randomized MACs prior associating, to avoid that
2020-09-17 03:28:57 <vivi> twb: as to point #2 you're right! except the OUI (vendor) is still relevant and we don't count dupes within 5min window
2020-09-17 03:29:02 <twb> (I also thought they were smart enough to power down the radio until the GPS reported they were within ~1km of a "home base" AP)
2020-09-17 03:29:21 <vivi> Additionally, these aren't stationary devices which require additional care in design - power environments are rather flaky in our field.
2020-09-17 03:29:41 <twb> vivi: ah they don't randomize the vendor part of the MAC?  Interesting
2020-09-17 03:29:59 <twb> I've only actually looked at the code on systemd-networkd
2020-09-17 03:30:30 <vivi> I'm not entirely sure the mobile vendors open source their mac-randomization
2020-09-17 03:30:43 <vivi> Additionally, we don't really care - accuracy isn't really a thing in 'advertisement'
2020-09-17 03:30:47 <twb> my only phone runs alpine so I have NFI :-)
2020-09-17 03:31:06 <vivi> (Coming from a deecade of bs director titles advertisement)
2020-09-17 03:31:44 <vivi> anyway the buildroot option didn't work very well for us, earlier this year I was in this channel asking for help and happened to find some of the best offerings.
2020-09-17 03:32:07 <vivi> We're now at a point where our v4 is absolutely rock-solid and I can push OTA udates - We *LOVE* Alpine Linux.
2020-09-17 03:32:31 <vivi> twb: I want an alpine phone!
2020-09-17 03:32:54 <twb> https://store.pine64.org/product/pinephone-community-edition-postmarketos-with-convergence-package-limited-edition-linux-smartphone/
2020-09-17 03:33:13 <twb> It doesn't actually work too well yet :-)
2020-09-17 03:33:44 <vivi> Oh that's this? Right so I'm awaiting some AR for a different project and will pick one up
2020-09-17 03:33:45 <twb> None of my chargers are charging it quickly, and HDMI isn't working with my monitors
2020-09-17 03:34:06 <twb> Strictly speaking it ships pmOS, which is *based on* alpine
2020-09-17 03:35:08 <vivi> here's all that matters - openrc :heavy-green-check: , musl :heavy-green-check: arm: :heavy-green-check:, systemd: HELL NO
2020-09-17 03:37:15 <vivi> jokes aside, I have a pretty cool writeup on netbooting raspi's that I'd blog but I don't have the time to maintain a blog so if anyone here is interested in a professional and (hopefully) well-written bit of documentation on netbooting the raspberry pi's and loading an ovl over the net - I got one nearly ready.
2020-09-17 03:37:15 <twb> Benno Rice's keynote summarizes my opinions about that pretty perfectly
2020-09-17 03:37:53 <twb> (TL;DR: launchd is a good idea; systemd is a bad implementation of that idea)
2020-09-17 03:38:07 <vivi> matter of fact this could suit well in the Wiki - lmk any mods here if interested
2020-09-17 03:38:17 <vivi> twb: Agree
2020-09-17 03:38:49 <vivi> For me it's about simplicity and when building the types of systems that I built - simplicity is tantamount to performance
2020-09-17 03:39:41 <vivi> As a small company, shipping devices worldwide is prohibitively costly, if I can assure that "it'll be slow but it'll work" that's good enough for me - we're capped at 5KB/s anyway :P
2020-09-17 03:39:57 <twb> I "leaned into" it on Debian, figuring if I had to have the suck, I might as well get the benefits
2020-09-17 03:40:22 <twb> some parts I quite like, e.g. MemoryHigh= instead of LD_PRELOAD'ing nocache
2020-09-17 03:40:45 <twb> And the seccomp BPF is simple to use
2020-09-17 03:41:44 <twb> Like you can say "this is the NTP server, it needs to to normal daemon things and also time things; block all other syscalls"
2020-09-17 03:41:44 <vivi> All valid arguments - for me it's workflow and memory.
2020-09-17 03:42:02 <vivi> as in - "This works, I know it, it's not broken, please stop"
2020-09-17 03:42:09 <twb> yeah I 100% get that
2020-09-17 03:42:21 <twb> Large parts of it are still just broken for some use cases
2020-09-17 03:42:30 <vivi>  /etc/init.d is my thing - systemctl <what> <do when?> <wait did I get this in proper order?>
2020-09-17 03:43:14 <vivi> then there's all the .service  vs. whatever else and it's like "I thought I told dnsmasq to stop but the .service is still running, for some reason systemd thinks there more than one thing I don't want it to do?"
2020-09-17 03:43:24 <twb> e.g. "try to do DHCPv4 on all ethernet interfaces; if ANY of them come up, then consider the network to be 'up'" currently requires patching .service scripts
2020-09-17 03:43:58 <vivi> Believe me, this hurts IRL on the daily
2020-09-17 03:44:30 <twb> MountOptions= is a scalar, but ExecStart= is an array :-)
2020-09-17 03:45:07 <vivi> Oh good, yeah once again a scary 'new' thing I'm just not interested in dealing with (yet)
2020-09-17 03:49:20 <baconicsynergy[m> twb, it worked :) thank you for the assistance, it is much appreciated! 'sudo virsh list' shows my virtual machine now :)
2020-09-17 03:49:37 <twb> baconicsynergy[m: you probably won't need sudo anymore for taht either
2020-09-17 03:49:56 <twb> Also if you're in GUI land, "pkexec" not "sudo"
2020-09-17 03:50:15 <baconicsynergy[m> i thought so too since my user was added to the libvirt group, but virsh list shows nothing...hmm
2020-09-17 03:50:31 <twb> baconicsynergy[m: you need to log out & back in for group membership to take effect
2020-09-17 03:50:45 <baconicsynergy[m> I did that :p still nothing!
2020-09-17 03:50:56 <twb> baconicsynergy[m: dunno then sorry
2020-09-17 03:51:05 <baconicsynergy[m> thats okay!
2020-09-17 06:51:09 <rnkn> hello
2020-09-17 06:52:30 <rnkn> I'm looking to get a Pinebook Pro and I'd like to run Alpine; what are the chances the Pinebook-specific stuff from postmarketOS will get ported back to OG Alpine?
2020-09-17 07:04:40 <mps> rnkn: what needs to be 'ported'?
2020-09-17 07:05:30 <rnkn> mps: so apparently postmarketOS has implemented a bunch of compatibility with power management, wifi, etc.
2020-09-17 07:05:39 <rnkn> sorry I'm not more familiar than what I've read in the forums
2020-09-17 07:07:26 <mps> is it rockchip or allwinner SOC (I always forget)
2020-09-17 07:09:54 <rnkn> Rockchip
2020-09-17 07:10:08 <mps> yes, found their site
2020-09-17 07:10:52 <mps> it is rk3399, and I type right now on rk3399 box with native alpine
2020-09-17 07:11:18 <mps> wifi works very nice
2020-09-17 07:11:48 <mps> what you mean by 'power management'?
2020-09-17 07:12:23 <rnkn> I don't really know sorry, I'm just going off a forum thread I found
2020-09-17 07:12:28 <rnkn> one sec...
2020-09-17 07:12:52 <rnkn> https://forum.pine64.org/showthread.php?tid=8793
2020-09-17 07:15:10 <mps> huh, too long to read
2020-09-17 07:16:50 <rnkn> sorry I'm getting kicked outta this cafe
2020-09-17 07:18:14 <mps> np
2020-09-17 07:19:30 <mps> if anything you know should be 'ported back' from pmOS you can tell here (or #alpine-devel), or open/create issue/bug on gitlab.a.o
2020-09-17 07:20:09 <mps> we can then look and see what can be done
2020-09-17 07:31:28 <artok> well then
2020-09-17 07:31:41 <artok> kicked out of cafe. that's a merit
2020-09-17 07:41:34 <rnkn> the cafe was closing
2020-09-17 07:43:08 <rnkn> mps: thanks, I'm still a prisoner of Apple/macOS atm
2020-09-17 07:45:55 <mps> rnkn: ok. can tell you that alpine works fine on my rk3399 gru-kevin chromebook for more that two years
2020-09-17 12:17:24 <PureTryOut[m]> rnkn: if you want the Pinebook Pro running on Alpine (so do I but I don't have the device 😢) you should coordinate with MartijnBraam
2020-09-17 12:21:08 <mps> hmm, also I would like to use it, but don't have idea how to get it
2020-09-17 12:21:39 <mps> maybe ask them to send me one for making alpine working on it 'out-of-the-box'
2020-09-17 12:26:04 <PureTryOut[m]> Lol, then you should do that
2020-09-17 12:26:21 <PureTryOut[m]> I was working on supporting it in Alpine, but it needs for one https://gitlab.alpinelinux.org/alpine/aports/-/merge_requests/6677
2020-09-17 12:28:51 <mps> yes, I remember, and I was move for it to main
2020-09-17 12:30:36 <PureTryOut[m]> Well I'm still waiting for somebody to either merge it or shoot the MR down indefinitely 🤷‍♂️
2020-09-17 12:32:28 <mps> hmm, maybe I should ask again there and if no one object move gcc-cross-embeded to main
2020-09-17 12:32:58 <mps> move u-boot and because that ATF to community is not option
2020-09-17 12:37:02 <PureTryOut[m]> Yeah I'd think the same
2020-09-17 12:37:03 <PureTryOut[m]> iirc ncopa wasn't a fan but he never actively agreed or disagreed with the option
2020-09-17 12:38:25 <mps> if we want to support more arm64 boxes then we have to do that
2020-09-17 12:52:24 <rnkn> pine64 and alpine seem like an attractive match
2020-09-17 12:53:33 <mps> there is 'alpine' SBC/SOC
2020-09-17 12:53:44 <mps> :)
2020-09-17 12:57:52 <rnkn> mps: ??
2020-09-17 12:57:58 <rnkn> an alpine SoC?
2020-09-17 12:59:44 <mps> arm64 platform name
2020-09-17 13:00:42 <mps> CONFIG_ARCH_ALPINE in kernel config
2020-09-17 13:01:00 <detha> is there a more official way of adding a reoo plus key to /etc/apk than just 'edit /a/a/r and copy the key file in place' ?
2020-09-17 13:02:21 <mps> Annapurna Labs Alpine SOC
2020-09-17 13:04:21 <rnkn> I see... nothing to do with Megan Ellison
2020-09-17 13:07:32 <wsinatra_> pine64 and alpine is an absolute must if you ask me
2020-09-17 13:10:05 <wsinatra_> PureTryOut[m]: what does the arm-trusted-firmware offer to the rockchip based boards? just easier to work with them if we've got it?
2020-09-17 13:11:17 <rnkn> I guess there's also the problem of no arm64 widevine too
2020-09-17 13:12:45 <PureTryOut[m]> wsinatra_: it makes it bootable lol
2020-09-17 13:13:04 <PureTryOut[m]> I'm not an expert on this, MartijnBraam knows more about it and he ported pmOS to it
2020-09-17 13:13:13 <PureTryOut[m]> So you'll have to ask him for any details
2020-09-17 13:13:50 <mps> it is needed to make u-boot for rk3399
2020-09-17 13:13:57 <mps> wsinatra_: ^
2020-09-17 13:14:18 <wsinatra_> ah gotcha, that makes more sense
2020-09-17 13:15:05 <wsinatra_> I just saw the pinebook chatter and my weekend project was looking at getting alpine on mine. guess there are some stumbling blocks
2020-09-17 13:15:41 <mps> though there are hacks on net to make u-boot for rk3399 without gcc-cross-embeded but no one tried that for alpine afaik
2020-09-17 13:15:56 <rnkn> there is the postmarketOS, which is based on alpine
2020-09-17 13:16:10 <rnkn> I only raised the topic because I'd prefer Pure Alpine tm
2020-09-17 13:16:41 <mps> simple solution is to get u-boot from pmOS, flash it and install alpine
2020-09-17 13:16:46 <PureTryOut[m]> postmarketOS just uses arm-trusted-firmware build with gcc-cross-embedded
2020-09-17 13:17:11 <wsinatra_> yeah it sounds like the solution is the arm firmware package
2020-09-17 13:18:03 <mps> flood pine by mail to send one to me and we will see ;)
2020-09-17 13:21:09 <rnkn> did google send you a free chromebook though?
2020-09-17 13:21:32 <mps> if it can be found in some shop near me I would go right now and buy one
2020-09-17 13:22:01 <mps> rnkn: ha, nice joke :)
2020-09-17 13:22:30 <wsinatra_> mps: if I had a spare rockpro I'd definitely send it your way
2020-09-17 13:22:39 <wsinatra_> the month long wait sucks
2020-09-17 13:23:36 <mps> wsinatra_: thanks, but I think if you send it right now it will arrive in a month or two to me :)
2020-09-17 13:24:38 <wsinatra_> hahah yeah I'm in New England, if I remember we're at least an ocean apart
2020-09-17 13:27:06 <mps> yes
2020-09-17 13:54:32 <MartijnBraam> rnkn: the correct way to add alpine support to the pinebook pro is having the WIP u-boot in the SPI flash on it, then it can just boot an uefi aarch64 OS image from SD or eMMC
2020-09-17 13:54:50 <MartijnBraam> except that the kernel in alpine doesn't have pinebook pro patches and it isn't fully functional in mainline
2020-09-17 13:55:17 <MartijnBraam> and with 5.5+ kernels the boot hangs when using that WIP uboot
2020-09-17 13:55:32 <MartijnBraam> but <5.5 kernels don't support most pbp things
2020-09-17 13:55:55 <PureTryOut[m]> linux-edge would help which is 5.8 currently
2020-09-17 13:55:58 <Jasper[m]> Didn't realize it supports UEFI. Would that be supported on the rockpro64 too?
2020-09-17 13:56:44 <MartijnBraam> 5.8 on linux edge doesn't have the display driver for the pinebook
2020-09-17 13:57:17 <MartijnBraam> which makes for a pretty boring linux experience
2020-09-17 13:57:20 <PureTryOut[m]> But yeah still not full functionality
2020-09-17 13:57:24 <MartijnBraam> Jasper: I think that would work fine, if somebody worked on making it work in uboot
2020-09-17 13:57:25 <Jasper[m]> God matrix home server is slow again
2020-09-17 13:57:26 <MartijnBraam> takes some config
2020-09-17 13:57:27 <MartijnBraam> this is the WIP u-boot: https://github.com/samueldr/wip-pinebook-pro/pull/13
2020-09-17 14:00:17 <PureTryOut[m]> Oh graphical u-boot to SPI that supports UEFI boot, hell yes.
2020-09-17 14:03:25 <MartijnBraam> yeah it's awesome, if it works
2020-09-17 14:03:47 <MartijnBraam> I did try to run the regular alpine image from the download page using that uboot, and it does boot neatly and shows grub, gets to linux and then fails because the kernel doesn't know that hardware
2020-09-17 14:03:53 <PureTryOut[m]> Yeah ok that's a next problem lol
2020-09-17 14:03:57 <MartijnBraam> I even patched the nixos logo out of uboot :P
2020-09-17 14:04:04 <PureTryOut[m]> Why does it even have a NixOS logo anyway? It should be distro independent
2020-09-17 14:04:08 <MartijnBraam> because it's build by nixos
2020-09-17 14:04:10 <Jasper[m]> <MartijnBraam "Jasper: I think that would work "> Interesting that it may not yet. They're practically the same board
2020-09-17 14:04:11 <MartijnBraam> it was quite annoying to  get it even building in alpine
2020-09-17 14:04:12 <PureTryOut[m]> Oh you just mean the pre-built u-boot
2020-09-17 14:04:14 <MartijnBraam> well it's in their patched source, there's no pre-built image
2020-09-17 14:04:16 <MartijnBraam> this is their build: https://user-images.githubusercontent.com/132835/87493303-84f01f80-c61a-11ea-9bf7-c52ed629a435.jpg
2020-09-17 14:04:19 <MartijnBraam> this is mine: https://fosstodon.b-cdn.net/media_attachments/files/104/756/622/320/785/105/original/322e15f0af32c410.jpg
2020-09-17 14:04:25 <MartijnBraam> (logo is higher up because the console scrolled)
2020-09-17 14:05:07 <MartijnBraam> yep, you can probably take most of the patches that's used for the uefi uboot for the pbp, but fix up display so it outputs to hdmi instead
2020-09-17 14:05:45 <mps> MartijnBraam: would you tell what should be enabled in kernel
2020-09-17 14:05:53 <mps> or create MR
2020-09-17 14:06:12 <MartijnBraam> don't know yet, haven't made that part work
2020-09-17 14:06:44 <MartijnBraam> in theory this tree should work: https://gitlab.manjaro.org/tsys/linux-pinebook-pro/-/tree/v5.8-rc1
2020-09-17 14:07:09 <mps> ok, will look on Arch alarm, maybe there is something which can be give hint
2020-09-17 14:08:57 <MartijnBraam> I did run that manjaro kernel for a while and it works great, except not in combination with the nice uboot
2020-09-17 14:13:33 <mps> the problem is no one alpine developer have this box so we can't test and fix what is missing
2020-09-17 14:15:33 <MartijnBraam> well I had it working in postmarketOS, the difficult part is how to make it work in alpine since the way that worked was with the u-boot that needs compilers from testing
2020-09-17 14:16:50 <MartijnBraam> having the graphical uboot stuff work would just fix that since u-boot doesn't need to be in alpine
2020-09-17 14:16:51 <PureTryOut[m]> mps: MartijnBraam has it and is an Alpine developer by proxy 😜
2020-09-17 14:18:56 <MartijnBraam> I also dropped the container for all the screws and I lost them so I can't put my pbp back together
2020-09-17 14:19:12 <mps> PureTryOut[m]: now I understand why I thought you own this box :)
2020-09-17 14:21:02 <mps> MartijnBraam: if you want to test alpine kernels I'm ready to enable needed config parts in linux-edge
2020-09-17 14:23:51 <mps> 5.8.10 is just released
2020-09-17 14:25:00 <mps> MartijnBraam: do you have kernel .config file to send me to look in diffs in linux-edge, maybe we can find what should be enabled
2020-09-17 14:25:21 <MartijnBraam> it's not just config, mainline linux is missing drivers
2020-09-17 14:25:22 <PureTryOut[m]> It's not just stuff not being enabled, but also stuff not being in mainline
2020-09-17 14:25:26 <mps> I'll make and push 5.8.10 in a few hours
2020-09-17 14:25:28 <PureTryOut[m]> What he said 😜
2020-09-17 14:25:58 <mps> ah, but we are against patches in kernels
2020-09-17 14:26:07 <MartijnBraam> I'm going through my local branches, trying to find my wip pbp stuff
2020-09-17 14:26:15 <MartijnBraam> mps: yes that's the main issue
2020-09-17 14:26:20 <mps> that is some kind of unwritten policy
2020-09-17 14:26:39 <MartijnBraam> maybe it can be something like what's done for the raspberry pi kernel?
2020-09-17 14:27:25 <mps> if you are ready to maintain it long term then ask ncopa
2020-09-17 14:27:39 <PureTryOut[m]> Well eventually it can be moved over to mainline, just not right now
2020-09-17 14:27:53 <mps> btw, I hope with next -lts we will not need -rpi kernels
2020-09-17 14:27:59 <MartijnBraam> yep, 5.9 has a lot of it
2020-09-17 14:28:18 <mps> and 5.9 has a lot of bugs ;)
2020-09-17 14:28:29 <rnkn> well this escalated quickly
2020-09-17 14:28:37 <mps> uname -a => Linux zarya 5.9.0-rc5-1-gru #2 SMP PREEMPT Mon Sep 14 07:05:10 UTC 2020 aarch64 GNU/Linux
2020-09-17 14:28:48 <PureTryOut[m]> Oh 5.9 has -rpi stuff needed? That's awesome
2020-09-17 14:28:51 <mps> this is on rk3399
2020-09-17 14:29:42 <mps> PureTryOut[m]: I built mainline 5.8.6 and run it quite fine on RPi zero, no external pathes
2020-09-17 14:29:57 <PureTryOut[m]> Nice!
2020-09-17 14:30:02 <mps> patches*
2020-09-17 14:30:31 <mps> but that is only RPi I have
2020-09-17 14:30:41 <PureTryOut[m]> My RPi3 still runs on Raspbian currently with Linux 4.19.66, but I'd love to switch it to Alpine stable at some point. Won't do that till it's possible with linux-edge or linux-lts probably though
2020-09-17 14:31:00 <MartijnBraam> the rpi4 is the one that has a lot of new stuff, new SoC, new GPU
2020-09-17 14:34:13 <mps> huh, vgacon: remove software scrollback support in 5.8.10
2020-09-17 14:35:43 <Hello71> fbcon too
2020-09-17 14:36:41 <mps> yes
2020-09-17 14:44:11 <rnkn> well this all seems promising... if I give in to my overzealous side and order a Pinebook Pro, what are the chances Alpine will be boot-ready in the month or two it'd take to get to Australia?
2020-09-17 14:46:24 <Cogitri> Well, if Alpine doesn't boot using pmOS should work
2020-09-17 14:47:39 <rnkn> there's always that
2020-09-17 14:48:18 <mps> sometimes I feel that we are pmOS lab :)
2020-09-17 14:48:55 <mps> or vice versa
2020-09-17 14:59:27 <MartijnBraam> there's a shortage on the panels used in the pbp
2020-09-17 14:59:42 <MartijnBraam> hmm looks like the uboot stuff was updated 3 days ago
2020-09-17 15:01:12 <Hello71> mps: it's called open source
2020-09-17 15:06:21 <mps> MartijnBraam: what is 'pbp'
2020-09-17 15:06:36 <MartijnBraam> pinebook pro
2020-09-17 15:06:49 <mps> ah, thanks :)
2020-09-17 15:07:02 <mps> have to remember
2020-09-17 15:14:37 <rnkn> but the ANSI US keyboard models are still in stock
2020-09-17 15:17:06 <MartijnBraam> ah
2020-09-17 15:17:22 <yjftsjthsd> "Sorry, we do not have enough "14″ PINEBOOK Pro LINUX LAPTOP (ANSI, US Keyboard)" in stock to fulfill your order right now. Please try again in 4320 minutes or edit your cart and try again. We apologise for any inconvenience caused." ?
2020-09-17 15:17:31 <yjftsjthsd> (I just tried to pull it up on the pine store)
2020-09-17 15:23:26 <rnkn> oh darn, I hadn't tried putting it in the cart
2020-09-17 15:23:46 <rnkn> their newsletter said they still had plenty of ANSI models in the store
2020-09-17 15:24:14 <MY-R> wth those signs mean "🤷‍ÿ♂️" ??
2020-09-17 15:59:03 <bbuccianti> I tried to install linux-edge but uninstalled linux-lts withouth trying it first so I'm stuck in Loading initial ramdisk
2020-09-17 15:59:19 <bbuccianti> Chrooted to it from a pendrive install
2020-09-17 15:59:33 <bbuccianti> How can I reinstall kernel properly?
2020-09-17 15:59:40 <ikke> apk add linux-lts
2020-09-17 15:59:48 <bbuccianti> I've done that
2020-09-17 16:00:01 <bbuccianti> But still the same problem, Loading initial ramdisk
2020-09-17 16:00:51 <bbuccianti> Also when I run mkinitfs it complains for /lib/modules/5.4.43 doesn't exist
2020-09-17 16:00:56 <yjftsjthsd> Is there a kernel in /boot?
2020-09-17 16:01:12 <bbuccianti> But installed kernel is 5.4.65
2020-09-17 16:01:51 <bbuccianti> I have /boot/initramfs-lts and /boot/vmlinuz-lts
2020-09-17 16:02:26 <yjftsjthsd> Can you verify that the bootloader is pointed at them?
2020-09-17 16:02:41 <bbuccianti> Grub.cfg?
2020-09-17 16:02:44 <bbuccianti> Let me check
2020-09-17 16:03:02 <yjftsjthsd> I was going to guess /boot/syslinux/syslinux.cfg, but I'm not in front of my Alpine box to verify
2020-09-17 16:04:24 <bbuccianti> Well... grub.cfg have linux /vmlinuz-lts blabla
2020-09-17 16:04:39 <bbuccianti> And initrd /initramfs-lts bla bla
2020-09-17 16:05:01 <bbuccianti> I need /boot/initramfs-lts there?
2020-09-17 16:05:19 <bbuccianti> Same on linux /boot/vmlinuz-lts ?
2020-09-17 16:05:30 <yjftsjthsd> I'd expect that to work, actually; I *think* the bootloader treats / as the root of *its* filesystem
2020-09-17 16:05:41 <bbuccianti> Oh I see
2020-09-17 16:06:00 <yjftsjthsd> But of course that begs the question of why yours isn't working:)
2020-09-17 16:06:35 <yjftsjthsd> To be clear: You're booting with GRUB? I thought alpine defaulted to alpine
2020-09-17 16:06:38 <bbuccianti> I can boot to recovery mode properly
2020-09-17 16:06:41 <yjftsjthsd> sorry, to syslinux
2020-09-17 16:06:47 <bbuccianti> I'm using grub yes
2020-09-17 16:07:23 <yjftsjthsd> Wait, you've got a regular boot and a recovery boot, and one works and the other doesn't? Same root filesystem?
2020-09-17 16:07:59 <bbuccianti> Seems like recovery mode doesn't work too
2020-09-17 16:08:19 <yjftsjthsd> darn, that would've been an easy way to hunt down the problem:\
2020-09-17 16:08:45 <bbuccianti> I can see a prompt but then it says sh: can't access tty; job control turned off
2020-09-17 16:08:52 <yjftsjthsd> Oh!
2020-09-17 16:08:55 <yjftsjthsd> That's good news
2020-09-17 16:09:11 <yjftsjthsd> That means you're loading the kernel and getting to *some* sort of userspace
2020-09-17 16:09:29 <yjftsjthsd> does it complain about not being able to mount root?
2020-09-17 16:10:08 <bbuccianti> It complains about tsc
2020-09-17 16:10:19 <bbuccianti> Switched to clocksource tsc
2020-09-17 16:10:24 <bbuccianti> And it hangs there
2020-09-17 16:11:13 <yjftsjthsd> Does your grub config pass the root= kernel option? Is it pointing to the right partition?
2020-09-17 16:11:22 <bbuccianti> Let me check
2020-09-17 16:12:01 <bbuccianti> Do you think would be better to use syslinux?
2020-09-17 16:12:50 <yjftsjthsd> I don't have a strong opinion.... it's possible that you're missing out on some tooling that automates setting up something, but I'd *expect* grub to work
2020-09-17 16:20:04 <bbuccianti> I can't found the root=kernel option
2020-09-17 16:22:54 <yjftsjthsd> That's.... odd
2020-09-17 16:23:05 <bbuccianti> There is another tool to configure booting on grub?
2020-09-17 16:24:07 <yjftsjthsd> I meant that I would not be surprised if Alpine ships scripts that automatically ex. stick a kernel where the alpine syslinux package expects it by default
2020-09-17 16:24:10 <yjftsjthsd> I don't know that
2020-09-17 16:24:15 <yjftsjthsd> but some distros do that kind of thing
2020-09-17 16:24:40 <yjftsjthsd> hang on, let me boot my alpine box
2020-09-17 16:27:07 <yjftsjthsd> Okay, so I've got an extlinux.conf that has APPEND root=UUID=blah modules=sd-mod,usb-storage,ext4 nomodeset quiet rootfstype=ext4
2020-09-17 16:27:19 <yjftsjthsd> If you don't have *any* root= then that could be the problem
2020-09-17 16:27:39 <yjftsjthsd> maybe see what partition your root is on (lsblk, df, whatever)
2020-09-17 16:28:22 <yjftsjthsd> then reboot, hit tab or e or whatever to edit the boot option
2020-09-17 16:28:33 <yjftsjthsd> and manually add root=/dev/sdX or whatever
2020-09-17 16:28:45 <bbuccianti> I think I have that extlinux.conf
2020-09-17 16:28:50 <bbuccianti> Let me check
2020-09-17 16:29:49 <yjftsjthsd> If you have config files for grub AND extlinux (syslinux variant for ext filesystems; not sure why they had to name it differently...), then you should make sure that the one actually booting is the one you expect
2020-09-17 16:34:34 <bbuccianti> I have that append root with the correct uuid and m
2020-09-17 16:34:35 <bbuccianti> odules
2020-09-17 16:39:50 <bbuccianti> Well I deleted the quiet kernel option
2020-09-17 16:40:17 <bbuccianti> And now I can see that it's hanging on fb0: switching to amdgpudrmfb from simple
2020-09-17 16:42:54 <vivi> RUN --mount=type=secret,id=abuild-key,required=true CARCH=armhf CTARGET=armhf PACKAGER_PRIVKEY="/run/secrets/abuild-key" abuild -F && find /root/packages -type f -name '*.apk' -exec cp '{}' ./ \;
2020-09-17 16:43:18 <vivi> CARCH=armv7 CTARGET=armv7 works fine, CARCH=armhf does not - is it armv6?
2020-09-17 16:46:35 <vivi> Found it, sorry https://git.unixpeople.org/hanez/abuild/src/commit/26ec31c6c2b10a9afe93335a948797a3ceeebfd5/functions.sh.in
2020-09-17 16:46:54 <bbuccianti> Well, installing linux-firmware-amdgpu doesn't help
2020-09-17 16:48:04 <bbuccianti> Still hanging on that switch to amdgpudrmfb
2020-09-17 16:48:40 <yjftsjthsd> Still odd to me that that would drop you into a shell:|
2020-09-17 16:52:22 <MartijnBraam> \o/ pinebookpro booted
2020-09-17 16:52:48 <MartijnBraam> with graphical u-boot, the manjaro kernel and postmarketos
2020-09-17 16:53:14 <MartijnBraam> so now the only requirement for alpine on the pbp is getting a kernel that's working
2020-09-17 16:56:03 <Jasper[m]> Wait does u-boot plainly not support hdmi output or just not for the rk3399?
2020-09-17 16:57:00 <MartijnBraam> all u-boot stuff is mostly dependent on the hardware, not sure if hdmi support exists for the rk3399
2020-09-17 16:58:31 <Jasper[m]> <MartijnBraam "all u-boot stuff is mostly depen"> I'm guessing the screen on the pbp is connected over eDP or LVDS then
2020-09-17 16:59:01 <MartijnBraam> DSI
2020-09-17 16:59:09 <Jasper[m]> DSI? Oh
2020-09-17 16:59:21 <MartijnBraam> I think at least
2020-09-17 16:59:25 <MartijnBraam> might be eDP
2020-09-17 17:00:11 <Jasper[m]> http://u-boot.10912.n7.nabble.com/PATCH-0-4-rockchip-rk3399-Fix-HDMI-out-td397637.html hmm
2020-09-17 17:01:13 <mps> eDP-1 connected primary 2400x1600+0+0 (normal left inverted right x axis y axis) 0mm x 0mm
2020-09-17 17:01:24 <mps> rockchip-drm driver
2020-09-17 17:02:19 <MartijnBraam> rk3399 support in general on mainline uboot is pretty fresh
2020-09-17 17:03:21 <mps> no, I run mainline from 5.0 (maybe even 4.20 can't remember)
2020-09-17 17:03:27 <MartijnBraam> hmmm linux-edge has a pinebook pro dtb
2020-09-17 17:15:41 <Jasper[m]> <mps "eDP-1 connected primary 2400x160"> I wonder if doing usbc to hdmi/dp would work
2020-09-17 17:15:50 <Jasper[m]> Since that is doing usbc dp alt mode
2020-09-17 17:16:08 <mps> Jasper[m]: yes, it works
2020-09-17 17:17:01 <mps> DP-1 disconnected (normal left inverted right x axis y axis)
2020-09-17 17:19:33 <MartijnBraam> the usb-c DP needs propriatary firmware though
2020-09-17 17:25:36 <mps> hmm, it worked for me without anything not in alpine
2020-09-17 17:26:02 <bbuccianti> I booted successfully with modprobe.blacklist=amdgpu
2020-09-17 17:26:29 <bbuccianti> I just need to make that permanent?
2020-09-17 17:26:36 <bbuccianti> or maybe there is a workaround?
2020-09-17 17:29:58 <MartijnBraam> postmarketOS packages firmware for the wifi stuff and displayport firmware for the pinebook pro currently
2020-09-17 17:33:04 <mps> which wifi card is in pbp
2020-09-17 17:33:54 <MartijnBraam> ap6256
2020-09-17 17:34:58 <mps> aha, then pbp is not open as they say
2020-09-17 17:37:39 <MartijnBraam> they?
2020-09-17 17:39:36 <mps> company behind it
2020-09-17 17:39:59 <MartijnBraam> heh pine64 doesn't claim at all that it's fully open
2020-09-17 17:40:44 <mps> good to know (to forget it :) )
2020-09-17 17:41:31 <MartijnBraam> the stuff is at least made so it can be as open as possible, but there weren't wifi chipsets available without firmware issues
2020-09-17 17:41:42 <MartijnBraam> and nothing closed running in the host OS :)
2020-09-17 17:42:33 <MartijnBraam> no display on linux-edge
2020-09-17 17:43:14 <MartijnBraam> at least not the way I'm booting it, you might be able to get an uefi console slightly working but I'm not uefi booting right now
2020-09-17 17:44:00 <mps> do you know which panel should be enabled
2020-09-17 17:46:06 <MartijnBraam> I think it's CONFIG_DRM_PANEL_FEIYANG_FY07024DI26A30D
2020-09-17 17:46:30 <mps> ok, will look
2020-09-17 17:47:05 <MartijnBraam> this is the defconfig I'm currently using: https://gitlab.manjaro.org/tsys/linux-pinebook-pro/-/blob/v5.8-rc1/arch/arm64/configs/pinebook_pro_defconfig
2020-09-17 17:47:30 <mps> I pushed 5.8.10 to builders few minutes ago
2020-09-17 17:48:03 <MartijnBraam> hmm no dts says boe,nv140fhmn49
2020-09-17 17:48:26 <Jasper[m]> <Jasper[m] "http://u-boot.10912.n7.nabble.co"> Hmm this seems to have been merged in april. In theory hdmi output in u-boot should work. USB keyboard on the other hand not though
2020-09-17 17:49:20 <mps> MartijnBraam: this is patched kernel for manjaro?
2020-09-17 17:49:21 <MartijnBraam> hmm I'll just check the schematic, this is weird
2020-09-17 17:49:25 <MartijnBraam> yes it is
2020-09-17 17:51:19 <mps> in mainline there is rk3399-pinebook-pro.dts
2020-09-17 17:53:09 <MartijnBraam> oh that panel is handled by panel-simple
2020-09-17 17:54:09 <mps> but there is config DRM_PANEL_FEIYANG_FY07024DI26A30D in 5.9-rc
2020-09-17 17:55:35 <MartijnBraam> yeah, but sure why that panel is in that defconfig, it's not needed
2020-09-17 17:55:50 <MartijnBraam> it needs panel_simple, pwm_bl and analogix_dp
2020-09-17 17:56:15 <mps> yes, same as samsung rk3399 chromebook
2020-09-17 17:57:49 <MartijnBraam> I don't see pwm_bl enabled
2020-09-17 17:57:57 <MartijnBraam> also it needs rockchipdrm ofcourse
2020-09-17 17:58:11 <MartijnBraam> 
ACTION has trouble locating the uart cable

2020-09-17 17:59:04 <mps> no, it is not enabled in linux-edge
2020-09-17 18:01:10 <mps> CONFIG_BACKLIGHT_PWM=m
2020-09-17 18:01:37 <mps> CONFIG_PWM_ROCKCHIP=m
2020-09-17 18:02:22 <mps> # CONFIG_DRM_PANEL_SIMPLE is not set
2020-09-17 18:02:37 <MartijnBraam> ah
2020-09-17 18:02:49 <mps> that is current in linux-edge 5.8.10, and previous
2020-09-17 18:08:21 <bbuccianti> yjftsjthsd: thanks for your help!
2020-09-17 18:08:42 <bbuccianti> yjftsjthsd: seems like mem_encrypt=off (stole from gentoo wiki) let me boot normally
2020-09-17 18:08:57 <bbuccianti> yjftsjthsd: on linux-edge
2020-09-17 18:10:23 <yjftsjthsd> Well I never would have thought of that, but I'm glad that it's working for you now:)
2020-09-17 18:12:20 <bbuccianti> I don't know anything about amdgpu SME
2020-09-17 18:12:25 <bbuccianti> but seems like a security thing?
2020-09-17 18:12:40 <bbuccianti> I don't understand why it's failing
2020-09-17 18:18:21 <MartijnBraam> mps: CONFIG_ROCKCHIP_CDN_DP needs to be disabled
2020-09-17 18:20:30 <mps> not sure, I think then my chromebook will not activate DP
2020-09-17 18:21:09 <mps> I played with it some time ago and forgot details
2020-09-17 18:21:24 <mps> hmm, could be that sound will not work
2020-09-17 18:21:25 <MartijnBraam> CDN_DP is broken on 5.8+
2020-09-17 18:22:50 <mps> ok, will test tomorrow
2020-09-17 18:28:47 <charl> after running setup-xorg-base which replaces mdev with eudev, consolefont service breaks and I get around 10 blank lines on my boot screen after "* Waiting for uevents to be processed"
2020-09-17 18:32:46 <charl> that is if I have consolefont service on boot. If I add it to default runlevel font changes successfully but I still get those 10 blank lines on boot screen
2020-09-17 18:35:32 <charl> I found the offending line:89 of `/etc/init.d/udev-trigger`: udevadm trigger --type=devices --action=add
2020-09-17 18:41:32 <charl> more specifically this rule here line:5 of `/lib/udev/rules.d/80-drivers.rules`: `ENV{MODALIAS}=="?*", RUN{builtin}+="kmod load $env{MODALIAS}"`
2020-09-17 18:41:38 <charl> which I have no idea what it does
2020-09-17 19:17:39 <charl> https://i.imgur.com/PzXhYEa.png
2020-09-17 19:34:47 <anonzzz> I have a Lenovo Chromebook N42. Keyboard is not working so I cannot complete an install without a USB keyboard. Wonder if there are images with a kernel that supports Chromebooks?
2020-09-17 19:35:26 <anonzzz> ...or maybe I would need to install using USB then compile a kernel with the Chromebook bits?
2020-09-17 19:52:44 <anonzzz> *USB=USB keyboard
2020-09-18 00:40:05 <gillesmauve> Weird. Getting just a black screen with bspwm--no hotkeys are working. Maybe I should see if sxhkd is on, but does anyone else know why? The rc files are all +x.
2020-09-18 00:56:03 <charl> how did you start it?
2020-09-18 01:07:26 <charl> gillesmauve: https://github.com/baskerville/bspwm/wiki
2020-09-18 01:08:07 <charl> on alpine the example rc's are in /usr/share/doc...
2020-09-18 01:09:17 <charl> also make sure you install sxhkd as it is not bundled with bspwm as in other distros
2020-09-18 01:20:38 <baconicsynergy[m> Hello friends! I'm on a quest to get pianobar and spotifyd working on Alpine Linux and postmarketOS and document the process so others can get them up and running too! I'm encountering the following error when I run pianobar: https://bpa.st/5VPA - this libao error looks like pianobar is trying to use alsa when it should use pulse. Normally, I would edit /etc/libao.conf to "default_driver=pulse", but this
2020-09-18 01:20:38 <baconicsynergy[m> strangely doesn't work... any ideas?
2020-09-18 01:43:52 <baconicsynergy[m> Oof looks like I overlooked editing /usr/share/alsa/alsa.conf to set the default soundcard and I added my user to the audio group. Sound is playing through my speakers! But it's not playing through my bluetooth headphones for some reason
2020-09-18 02:13:16 <baconicsynergy[m> Okay, I tinkered around with /etc/libao.conf, and I can confirm that it /definitely/ needs to be set to "default_driver=pulse"
2020-09-18 02:13:40 <baconicsynergy[m> as well as having libao installed, of course
2020-09-18 02:18:37 <Hello71> and pulse?
2020-09-18 02:51:10 <gillesmauve> baconicsynergy: I have a solution that you're not going to like.
2020-09-18 03:40:01 <baconicsynergy[m> gillesmauve shoot
2020-09-18 04:17:16 <gillesmauve> baconicsynergy: get rid of everything that you own that remotely mentions bluetooth
2020-09-18 04:17:29 <gillesmauve> it's a gateway drug to systemd
2020-09-18 04:40:31 <anonzzz> gillesmauve: lol
2020-09-18 04:43:52 <anonzzz> *maybe pulse is a gateway drug to systemd
2020-09-18 04:44:19 <baconicsynergy[m> :(
2020-09-18 04:50:20 <gillesmauve> bluetooth fuckin sucks
2020-09-18 04:54:45 <yjftsjthsd> Right, it's a gateway drug because it acclimates you to complex systems that theoretically look nice but have a thousand edge cases in practice;)
2020-09-18 04:54:54 <gillesmauve> precisely
2020-09-18 04:59:46 <winny> how's the state of libvirt nowadays with the couple backported patches to fix the nasty upstream deadlock behaviors?
2020-09-18 05:04:17 <winny> i'm asking primarily if anybody has ran into issues since, because it sounds like the problems might be a plethora of various malloc usages in libvirt
2020-09-18 05:24:09 <gillesmauve> why is malloc so bad when it comes to gcc abuse
2020-09-18 06:05:33 <paper_> is there a package which provides qmake? I can't find it with pkgs.alpinelinux.org
2020-09-18 06:06:28 <mps> paper_: apk search cmd:qmake
2020-09-18 06:06:45 <mps> qt5-qtbase-dev
2020-09-18 06:07:02 <paper_> how did I miss it, thanks a lot
2020-09-18 06:12:41 <yjftsjthsd> TIL `apk search` accepts actual expressions
2020-09-18 06:13:38 <mps> regex are not cheap
2020-09-18 06:13:47 <yjftsjthsd> ... does apk have a manpage? There's no apk-doc package
2020-09-18 06:14:14 <mps> apk-tools-doc
2020-09-18 06:14:35 <mps> apk search apk-tools
2020-09-18 06:14:38 <mps> :)
2020-09-18 06:16:49 <yjftsjthsd> not in 3.12 main or community...
2020-09-18 06:17:08 <mps> no, on edge
2020-09-18 06:17:12 <yjftsjthsd> Aha.
2020-09-18 06:17:16 <mps> it is new thing
2020-09-18 06:17:21 <yjftsjthsd> Ah, okay
2020-09-18 06:17:31 <yjftsjthsd> Was just going to ask why on earth that wouldn't be in stable
2020-09-18 06:18:12 <mps> it didn't existed when the 3.12 released, iirc
2020-09-18 06:20:28 <yjftsjthsd> https://gitlab.alpinelinux.org/alpine/apk-tools/-/commits/master/doc/apk-add.8.scd shows it being added early January 2020, so it would've been months old by the time 3.12 was cut per https://wiki.alpinelinux.org/wiki/Alpine_Linux:Releases but I can accept that it might've missed the ... does alpine do a freeze window? something like that
2020-09-18 06:22:25 <mps> some kind of 'freeze' exists but it is not strict. so some thing 'pass freeze' but some don't
2020-09-18 06:23:15 <mps> it depends of importance or developer/user persistence :)
2020-09-18 06:23:27 <mps> s/of/on/
2020-09-18 06:26:39 <yjftsjthsd> I'm honestly surprised that it took until 2020 for apk to get manpages at all
2020-09-18 06:26:52 <yjftsjthsd> But no matter; 3.13 will come soon enough:)
2020-09-18 06:27:05 <yjftsjthsd> Or I could just pull in apk-tools-doc@edge; that should be safe enough...
2020-09-18 06:28:43 <mps> apk -help is quite enough, imo
2020-09-18 06:36:42 <yjftsjthsd> So I mentioned the lack of manpages because I was wishing I could've learned about `apk search cmd:foo`. Even `apk search --help` doesn't tell me that that exists
2020-09-18 06:38:23 <yjftsjthsd> Although https://gitlab.alpinelinux.org/alpine/apk-tools/-/tree/master/doc still shows no page for apk-search :(
2020-09-18 09:58:38 <mps> what terminal emulator we have in alpine to use with sway? alacritty crashes
2020-09-18 09:59:20 <Cogitri> fwiw alacritty should work again once they make a new release
2020-09-18 09:59:37 <Cogitri> You could use Tilix or GNOME Terminal in the meantime
2020-09-18 10:01:25 <mps> thanks, will try
2020-09-18 10:34:25 <charl> is there some chroot helper script in extended iso?
2020-09-18 10:38:09 <sxiii> Hmmm... Did someone had any success with building a customized ISO from alpine? I'm following a documentation but I can't get my packages into ISO no matter how I try
2020-09-18 10:39:11 <mariuss_> Hi guys, I am just trying to install alpine-standard-3.12.0-x86_64.iso from USB to an old 2011 "Lenovo Ideacentre Q180".
2020-09-18 10:40:03 <mariuss_> It has no UEFI support, so I'd need to prepare a MBR install. I can't find some tutorial how to achieve that
2020-09-18 10:40:06 <mps> Cogitri: tilix also crashes
2020-09-18 10:40:13 <Cogitri> Huh
2020-09-18 10:40:34 <mariuss_> First install with EFI partition lead to a "GRUB_" notice after reboot
2020-09-18 10:40:34 <Cogitri> How does it crash?
2020-09-18 10:40:59 <mariuss_> any ideas how I use custom partitioning during installation?
2020-09-18 10:41:00 <mps> but I gave up. this all is quite unstable on my arm64
2020-09-18 10:41:55 <mps> only what worked was sway exit shortcut key combo
2020-09-18 11:58:26 <Hello71> mps: can't you use xterm or something?
2020-09-18 11:58:46 <Hello71> charl: what do you mean?
2020-09-18 11:59:00 <Hello71> sxiii: you have to apk add
2020-09-18 11:59:35 <mps> Hello71: xterm on wayland?
2020-09-18 11:59:49 <Hello71> mariuss_: are you having problems booting the image or a sys install?
2020-09-18 11:59:58 <Hello71> mps: xwayland is a thing?
2020-09-18 12:00:35 <mps> Hello71: but I deleted this hipsters packages
2020-09-18 12:00:48 <mps> xorg is still best
2020-09-18 12:02:19 <Hello71> well if you only have half an install then that might be your problem
2020-09-18 12:02:47 <Hello71> there are many other pure wayland TEs: kitty, konsole, probably anything gtk/vte
2020-09-18 12:03:50 <mps> systemd-logindWelogind ;)
2020-09-18 12:55:17 <mariuss_> Hello71 Well I am a step further, I manually partitioned hdd to dos style
2020-09-18 12:55:38 <mariuss_> primary 500MB first, then 2G swap, then extenden
2020-09-18 12:55:48 <mariuss_> with 32g ext4 for /
2020-09-18 12:55:54 <mariuss_> and the rest for /home
2020-09-18 12:56:07 <mariuss_> I think, i just have to figure out
2020-09-18 12:56:15 <mariuss_> how to setup the bootloade
2020-09-18 12:56:56 <mariuss_> I mounted sda1 to /mnt/boot, sda5 to /mnt/ and sda6 to /mnt/home
2020-09-18 12:57:04 <mariuss_> and fired "setup-disk -m sys /mnt"
2020-09-18 12:57:34 <mariuss_> But when I reboot without stick, I only get a blinking "_"
2020-09-18 12:57:36 <mariuss_> _
2020-09-18 12:57:49 <mariuss_> so I guess i need to install some bootloader?
2020-09-18 12:58:35 <mariuss_> also /dev/sda1 has proper boot flag
2020-09-18 13:12:03 <mariuss_> i might have did it, I needed to copy a mbr bin to /dev/sda, no alpine starts
2020-09-18 13:12:19 <mariuss_> dd if=/usr/share/syslinux/mbr.bin of=/dev/sda bs=440 count=1
2020-09-18 22:21:12 <MartijnBraam> Is there a list of machines the alpine-standard-3.12.0-aarch64.iso is known to work on?
2020-09-18 22:29:05 <c705> i doubt it
2020-09-18 22:29:22 <c705> if it runs linux, no reason to think alpine would have any issues, besides specific driver support
2020-09-18 22:35:26 <MartijnBraam> well that's the whole point
2020-09-18 22:35:33 <MartijnBraam> aarch64...
2020-09-18 22:35:51 <c705> what's the whole point?
2020-09-18 22:37:01 <MartijnBraam> with ARM you don't know if it just runs that particular linux kernel
2020-09-18 22:40:56 <c705> are you asking if we support a specific _architecture_
2020-09-18 22:41:43 <MartijnBraam> no I'm asking if there's any actual machine that you can plop alpine-standard-3.12.0-aarch64.iso into and it actually boots
2020-09-18 22:43:28 <c705> any machine that has a chip that runs aarch64
2020-09-18 22:44:49 <MartijnBraam> no that's not how ARM works
2020-09-18 22:45:55 <c705> really/ that seems rather dumb
2020-09-18 22:45:58 <c705> https://wiki.alpinelinux.org/wiki/Alpine_on_ARM
2020-09-18 22:47:41 <MartijnBraam> that's just a dump of the DTBs included, not what actually works.
2020-09-18 22:50:47 <vivi> MartijnBraam: perhaps providing your chipset could help?
2020-09-18 22:51:03 <MartijnBraam> I don't have one, I'm wondering what works at all to get one
2020-09-18 22:53:52 <yjftsjthsd> c705: yes, it is dumb, but yes, ARM is that ... diverse... in booting
2020-09-18 22:55:23 <c705> and people still say arm is the future/ ok then
2020-09-18 22:56:10 <MartijnBraam> arm is more standardized if you get the expensive ones, but that's not the ones you just get to try to boot alpine on
2020-09-19 04:16:18 <arahael> c705: Technically, iPhones, iPad, the Apple Watch, and even the "touch bar" and soon, the newer apple computers are all going to be "ARM", too.
2020-09-19 04:17:17 <arahael> c705: And you can rent ARM systems at Amazon AWS, too, and many Anddroid phones, and Raspberry Pi...
2020-09-19 10:01:04 <hazz> hi there
2020-09-19 10:01:50 <hazz> having probs on reading man
2020-09-19 10:02:19 <ikke> By default, there are no man pages installed
2020-09-19 10:02:23 <ikke> apk add mandoc docs
2020-09-19 10:02:31 <ikke> if you are on edge
2020-09-19 10:02:56 <ikke> or 3.12
2020-09-19 10:03:08 <ikke> on 3.11 or earlier, it's called mdocml
2020-09-19 10:03:28 <hazz> error relocating /usr/lib/libmandoc.so: __lstat_time64 no symbol found
2020-09-19 10:03:34 <ikke> ah
2020-09-19 10:03:35 <hazz> ikke: i know
2020-09-19 10:03:45 <ikke> apk version musl
2020-09-19 10:03:49 <hazz> 32 bits distro
2020-09-19 10:04:01 <ikke> You need to make sure you have musl-1.21
2020-09-19 10:04:23 <hazz> distro downloaded 3 days ago
2020-09-19 10:04:32 <hazz> has to be the last one
2020-09-19 10:04:40 <ikke> did you do an apk upgrade since?
2020-09-19 10:04:43 <ikke> I assume you did not
2020-09-19 10:04:54 <hazz> for sure
2020-09-19 10:05:49 <ikke> What version?
2020-09-19 10:06:18 <hazz> 1.2
2020-09-19 10:06:19 <mps> when reporting problem say which arch, which release and maybe content of /etc/apk/repositories
2020-09-19 10:06:55 <mps> ikke: maybe you can add this to topic on channel
2020-09-19 10:08:04 <hazz> i'm doing anothere upgrade right now
2020-09-19 10:08:12 <mps> (and delete ' the secure lightweight distro' :) )
2020-09-19 10:08:25 <hazz> i got even probs with less cmd
2020-09-19 10:11:32 <hazz> ok man man working right now
2020-09-19 10:11:39 <hazz> after upgrade 1.24
2020-09-19 10:11:45 <mps> https://docs.alpinelinux.org
2020-09-19 10:14:52 <hazz> on the way tips for installing gpm on terminal
2020-09-19 10:15:41 <mps> there is consolation pkg which is more modern than gpm
2020-09-19 10:15:50 <mps> it is libinput based
2020-09-19 10:16:38 <hazz> thank you
2020-09-19 10:16:49 <mps> np
2020-09-19 10:16:59 <hazz> I have to say that Alpinelinux impressed my
2020-09-19 10:17:07 <hazz> I have to say that Alpinelinux impressed me so much
2020-09-19 10:17:37 <hazz> fast no freezing on old laptop almost 10 years
2020-09-19 10:18:10 <mps> did you firefox or chromium ;)
2020-09-19 10:18:21 <hazz> i tried to install ARCHlinux 32 BIT but was quite a mess installing binaries
2020-09-19 10:18:24 <hazz> links
2020-09-19 10:18:34 <hazz> no X for now
2020-09-19 10:18:48 <mps> ah, ok
2020-09-19 10:19:13 <mps> 'big' software is slow wherever run
2020-09-19 10:20:37 <hazz> i used mulinux many years ago on a very old laptop a compaq with 16bit ram
2020-09-19 10:21:07 <hazz> i do not remember well
2020-09-19 10:23:20 <mps> my first linux was on 386 without FPU, 1MB (yes) RAM
2020-09-19 10:24:29 <mps> but not for serious work, just to test. first for work was i486 with 4MB RAM
2020-09-19 10:26:41 <hazz> mps: something similar to me,my first pc was a vic20 and played with a zx Spectrum
2020-09-19 10:27:20 <hazz> uses to play with basic, then i quit everything/
2020-09-19 10:27:42 <hazz> used to play with basic
2020-09-19 10:27:51 <mps> hazz: heh, yes, also I started with ZX, and comodore 64 (iirc right model)
2020-09-19 10:29:08 <hazz> sometimes seems like yesterday
2020-09-19 10:29:11 <mps> basic was OS on zx
2020-09-19 10:29:47 <mps> hazz: it was yesterday, isn't it :)
2020-09-19 10:30:11 <hazz> yes it is
2020-09-19 10:30:59 <hazz> those are mind flash memory more fast then nowadays memories 16 64 GB RAM
2020-09-19 10:31:00 <mps> 48KB RAM, hey
2020-09-19 10:32:36 <MY-R> on older CPUs or just low power ones windows 10 with 8 GB of ram and SSD working nice too until run firefox/chrome...
2020-09-19 10:32:48 <hazz> Jesus Christ there is Radare2 apk
2020-09-19 10:33:03 <hazz> thumb up
2020-09-19 10:33:05 <mps> Clive Sinclair is genius with his idea of use damaged memory chips
2020-09-19 10:33:30 <hazz> Yep Sinclair was a good one
2020-09-19 10:34:12 <hazz> those consumer pc were so powerfull
2020-09-19 10:34:20 <Stacy[m]> Hi, how do I fix an issue with starting gdm on alpine linux 3.12 in virt-manager?
2020-09-19 10:35:12 <Stacy[m]> Whenever i type rc-service gdm start, it tries to do something but then screen turns black with a blinking cursor and I can type in it
2020-09-19 10:35:29 <Stacy[m]> Or is GNOME only working in edge?
2020-09-19 10:36:45 <hazz> Sorry Stacy[m] can't answer back
2020-09-19 10:37:03 <mps> gnome works on 3.12 I think
2020-09-19 10:37:17 <mps> never tried though, Cogitri ^
2020-09-19 10:38:06 <mps> Stacy[m]: does X starts
2020-09-19 10:38:24 <Stacy[m]> I followed the GNOME install guide from wiki
2020-09-19 10:38:52 <Stacy[m]> <mps "Stacy: does X starts"> I mean, it doesn't throw any errors, so I assume it does
2020-09-19 10:39:35 <mps> try startx from console on VM, it is qemu?
2020-09-19 10:40:26 <Stacy[m]> yes
2020-09-19 10:43:15 <mps> i'm not sure but maybe xf86-video-modesetting is need to be installed
2020-09-19 10:44:06 <mps> last time I tested alpine X in qemu is about 3 years ago
2020-09-19 10:44:13 <charl> check xorg logs
2020-09-19 10:44:33 <mps> yes, /var/log/Xorg.0.log
2020-09-19 10:45:47 <charl> also, not sure but eudev might be a requirement for gnome. Did you setup-xorg-base?
2020-09-19 10:46:21 <Stacy[m]> Yes
2020-09-19 10:47:05 <charl> check xorg logs then. Warnings or Errors should give you a hint
2020-09-19 10:49:07 <Stacy[m]> No screens found error
2020-09-19 10:52:17 <mps> so, X driver probably missing
2020-09-19 10:52:44 <hazz> maybe you need to put the .conf file somewhere else
2020-09-19 10:52:52 <hazz> check it
2020-09-19 10:52:59 <Cogitri> GNOME should certainly work on 3.12. I guess either the right xf86-video driver is missing or use isn't set up
2020-09-19 10:53:12 <Cogitri> I doubt you need a .conf file, Xorg is good at autodetecting this stuff
2020-09-19 10:54:11 <mps> yes, Xorg works without .conf files, though tweaking libinput is not bad idea
2020-09-19 10:57:18 <Stacy[m]> Well the driver seems to be qxl, it is installed, but still xorg throws the same error
2020-09-19 10:57:57 <Stacy[m]> i can use software if qxl is broken or something, but how do i set to use software driver?
2020-09-19 10:58:06 <charl> "QEMU can emulate several types of VGA card. The card type is passed in the -vga type command line option and can be std, qxl, vmware, virtio, cirrus or none."
2020-09-19 10:58:16 <charl> https://wiki.archlinux.org/index.php/QEMU
2020-09-19 10:58:34 <hazz> db main
2020-09-19 10:58:40 <hazz> dc
2020-09-19 10:58:46 <Stacy[m]> Yes, it is qxl
2020-09-19 10:58:47 <hazz> sorry guys
2020-09-19 10:59:10 <charl> did you pass the -vga argument on qemu?
2020-09-19 11:00:21 <Stacy[m]> No, but I set qxl in virt-manager (it's like a gui for qemu)
2020-09-19 11:00:58 <mps> qxl only works on x86
2020-09-19 11:02:23 <charl> dmesg | grep drm
2020-09-19 11:04:19 <Stacy[m]> <charl "dmesg | grep drm"> No output
2020-09-19 11:05:09 <charl> so maybe you need to load qxl modules on your kernel
2020-09-19 11:05:14 <MY-R> I would uninstall qxl and try with VirtIO and modesetting
2020-09-19 11:05:39 <charl> or yeah what MY-R said
2020-09-19 11:06:37 <charl> "On Linux guests, the qxl and bochs_drm kernel modules must be loaded in order to gain a decent performance."
2020-09-19 11:06:41 <charl> from arch wiki
2020-09-19 11:09:15 <Stacy[m]> <charl ""On Linux guests, the qxl and bo"> Oh thanks! I had to add qxl to the modules argument during boot, now it works
2020-09-19 11:09:20 <Stacy[m]> Thanks for help
2020-09-19 11:16:07 <hazz> can some
2020-09-19 11:16:54 <hazz> post a config file to boot wlan0 with wpa_supplicant?
2020-09-19 11:17:15 <hazz> i got probs about rfkill at boot time
2020-09-19 11:17:36 <hazz> it's a b43 firmware
2020-09-19 11:19:29 <abenz> Stacy[m]: I'd use virtio gpu like the others said
2020-09-19 11:19:36 <abenz> it will give you very good performance
2020-09-19 11:19:55 <charl> hazz https://wiki.alpinelinux.org/wiki/Connecting_to_a_wireless_access_point
2020-09-19 11:19:59 <charl> bottom of page
2020-09-19 11:20:07 <abenz> of course this depends on what you plan on doing on the VM, but if it involves anything with graphics, virtio is superior
2020-09-19 11:24:39 <hazz> can someone tell me if rfkill can be run at boot time?
2020-09-19 11:24:44 <hazz> if yes how
2020-09-19 11:29:08 <charl> https://wiki.alpinelinux.org/wiki/Writing_Init_Scripts
2020-09-19 11:31:25 <Stacy[m]> <abenz "of course this depends on what y"> Do i have to load virtio module then?
2020-09-19 11:32:39 <charl> virtio is loaded by default
2020-09-19 11:33:53 <charl> you just need -vga virtio with qemu
2020-09-19 11:35:56 <ikke> abenz: did you. See the reply to the kdenlive issue?
2020-09-19 11:39:04 <abenz> ikke: wonderful! you said that just in time, I was about to fire up my editing VM
2020-09-19 11:39:24 <abenz> will try it now
2020-09-19 11:40:39 <hazz> anyone having prob with tmux server exited unexpectedly?
2020-09-19 11:40:49 <abenz> Stacy[m]: most kernels have it included, if you are using virt-manager, you will need to do two things iirc, select virtio, and also tick "Enable 3D acceleration"
2020-09-19 11:41:26 <abenz> or its "OpenGL acceleration", something like that
2020-09-19 11:41:54 <mps> hazz: you can try iwd instead of wpa_s
2020-09-19 11:42:49 <abenz> with this working properly, you get near host/native speeds (in my case, accelerated browsing, and kdenlive detects HW acceleration)
2020-09-19 11:43:08 <abenz> test in guest by dmesg | grep drm, you should see messages about "virgl loaded successfully"
2020-09-19 11:43:44 <Stacy[m]> I see, will try
2020-09-19 11:44:09 <charl> +1 for iwd. Reduced my boot time because I could't figure out how to run wpa_sup before udhcpc. Thanks for porting it mps!
2020-09-19 11:44:14 <hazz> mps: thanks
2020-09-19 11:45:32 <mps> charl: hazz: you are welcome
2020-09-19 11:47:58 <hazz> the last alpha release 3.13.0_alpha got probs with tmux
2020-09-19 11:48:10 <hazz> i need to install with git
2020-09-19 11:48:48 <mps> charl: there is also iwgtk (gui for iwd) if the networkmanager/connman are overkill
2020-09-19 11:49:23 <charl> mps I'm good with just iwd + openresolv
2020-09-19 11:49:27 <mps> in testing still, have to move it to community
2020-09-19 11:49:59 <mps> yes, iwctl is quite fine, but some people like gui
2020-09-19 11:52:03 <abenz> ikke: ah that patch requires targets mlt 6.22 and requires a compile (I'd need alpine-sdk?)
2020-09-19 11:52:23 <abenz> https://wiki.alpinelinux.org/wiki/Abuild_and_Helpers
2020-09-19 11:52:51 <abenz> is this the wiki page I should be studying for creating a custom version of an existing pkg?
2020-09-19 11:53:02 <abenz> s,targets,,g
2020-09-19 11:53:45 <ikke> I can make a merge request for that patch
2020-09-19 11:53:51 <ikke> then it will build a package
2020-09-19 11:54:00 <abenz> please do that would be great
2020-09-19 11:54:29 <abenz> on 3.12 as well if possible!
2020-09-19 11:54:37 <ikke> ah
2020-09-19 11:54:47 <ikke> should be possible
2020-09-19 11:55:02 <ikke> Can you test it against edge first, or is that harder?
2020-09-19 11:55:24 <abenz> no prob at all, I have another edge diskless thumb I can load
2020-09-19 11:56:07 <abenz> let me know once pushed
2020-09-19 11:57:24 <hazz> Sorry guys but doing a build -r or tmux where can i find the bin?
2020-09-19 11:57:43 <ikke> hazz: ~/packages/<repo>/<arch>
2020-09-19 11:57:45 <hazz> or - of
2020-09-19 12:04:42 <hazz> ikke: ok compiled,then do i have to reboot,sorry for the silly question
2020-09-19 12:06:30 <ikke> no
2020-09-19 12:07:06 <mps> reboot on alpine is not needed, except for kernel upgrade
2020-09-19 12:07:43 <mps> on musl system, to be correct
2020-09-19 12:07:52 <hazz> ./tmux or what
2020-09-19 12:08:00 <ikke> hazz: You need to install the package
2020-09-19 12:08:12 <hazz> installed via git
2020-09-19 12:08:21 <ikke> git does not install packages
2020-09-19 12:08:24 <hazz> and i got tmux.apk
2020-09-19 12:10:03 <mps> hazz: apk add --allow-untrusted pkg-filename.apk
2020-09-19 12:13:07 <ikke> abenz: https://gitlab.alpinelinux.org/kdaudt/aports/-/jobs/209864
2020-09-19 12:13:28 <ikke> The package will be in the artificats once it finished
2020-09-19 12:13:43 <hazz> sorry i didn't understand so what's the diff about apk add file, and cd aports and abuild -r
2020-09-19 12:14:37 <mps> apk is user tool to install/remove pkgs
2020-09-19 12:14:37 <ikke> abuild -r just builds a package
2020-09-19 12:15:04 <mps> aports is repository where you build packages with abuild
2020-09-19 12:15:08 <hazz> how can i use that package build
2020-09-19 12:16:05 <ikke> hazz: mps told you how
2020-09-19 12:16:32 <ikke> Or you add that path (/home/username/packages/<repo>) to /etc/apk/repositories
2020-09-19 12:17:14 <hazz> ok done it i presum to have understood
2020-09-19 12:17:27 <hazz> not for sure,sorry about that
2020-09-19 12:19:15 <hazz> all the aports are updated automaticly or what?
2020-09-19 12:21:33 <mps> hazz: aports is git repository
2020-09-19 12:22:04 <hazz> ok seen that,interesting
2020-09-19 12:22:15 <mps> that is where alpine development is done and do
2020-09-19 12:22:16 <hazz> well done
2020-09-19 12:22:48 <mps> you can clone it to local machines and make changes
2020-09-19 12:23:10 <hazz> not that good one,
2020-09-19 12:23:18 <hazz> i will try
2020-09-19 12:24:31 <mps> you can at alpine wiki section about development and how all this is used and works
2020-09-19 12:24:32 <hazz> one day or another like blondie
2020-09-19 12:24:42 <ikke> abenz: let me know if that one works for you
2020-09-19 12:25:56 <hazz> thanks mps for the info,putting back to iwd is there any tut where i can view that?
2020-09-19 12:26:25 <hazz> just on arch i saw something
2020-09-19 12:27:09 <mps> sorry no, but arch is quite fine, except part of systemd
2020-09-19 12:27:41 <hazz> yep
2020-09-19 12:28:20 <mps> I'm thinking to write but don't have time and I'm not good in english
2020-09-19 12:29:34 <hazz> or maybe in Spanish France Deutch?
2020-09-19 12:30:17 <hazz> that will be ok do you have one of yours to pass me?
2020-09-19 12:31:26 <mps> :)
2020-09-19 12:35:35 <mps> my Deutch is faded away mostly, Spanish never tried and France learning was my fiasco :)
2020-09-19 12:42:07 <hazz> uhm are you from east europe?
2020-09-19 12:42:54 <abenz> ikke: booting diskless now to test, thanks a lot
2020-09-19 12:44:16 <mps> hazz: me? I'm from central europe
2020-09-19 12:44:29 <hazz> mps: yes
2020-09-19 13:29:07 <sxiii> Guys can somebody teach me or gimme some docs for adding APKVOL into a ISO to make Alpine Live bootable ISO with some *pkgname* from repos?
2020-09-19 13:29:59 <sxiii> (/s/APKVOL/APKOVL/g) I've checked the official wiki on ISO creation as well as on APK Overlays and I just can't finalize it >,,<
2020-09-19 13:38:50 <abenz> ikke: kdenlive still gets stuck on splash
2020-09-19 13:38:56 <abenz> I run it with:
2020-09-19 13:39:02 <abenz> dbus-run-session kdenlive
2020-09-19 13:39:13 <abenz> dbus service is running (status=started)
2020-09-19 13:39:28 <abenz> is there anything else I overlooked? this is with a freshly upgraded edge
2020-09-19 13:43:50 <abenz> PureTryOut[m]: Hi, can you give ikke's build on your end too?
2020-09-19 13:44:04 <abenz> s,buid,build a shot,g
2020-09-19 13:48:11 <ikke> abenz: how did you install it?
2020-09-19 13:51:18 <ikke> abenz: note that I pushed a new commit that bumped pkgrel, which I forgot in the previous version
2020-09-19 13:54:16 <abenz> ikke: booted thumb, ensured repos are pointing to edge, update && upgrade -U, install kdenlive (was not installed)
2020-09-19 13:54:46 <abenz> installed dbus (this should be a dependency for kdenlive doesnt it?), and finally tried to launch: dbus-run-session kdenlive
2020-09-19 13:55:24 <ikke> aha, then you don't have the version from that MR
2020-09-19 13:55:37 <ikke> I did not push it to alpine/aports yet
2020-09-19 13:56:04 <ikke> https://gitlab.alpinelinux.org/kdaudt/aports/-/jobs/209871/artifacts/browse/packages/community/x86_64/
2020-09-19 13:56:04 <abenz> oh, I thought thats what the job paste implied
2020-09-19 13:56:13 <ikke> That's just CI
2020-09-19 13:59:50 <abenz> am I right in assuming I will need to remove mlt and then apk add --allow-untrusted mlt_u_sent.apk ?
2020-09-19 14:01:56 <ikke> because of the pgkrel bump, you don't need to remove ot
2020-09-19 14:01:59 <ikke> it
2020-09-19 14:02:05 <ikke> so just the latter
2020-09-19 14:02:19 <ikke> (you cannot even remove it without uninstalling kdenlive)
2020-09-19 14:08:27 <abenz> now I see a lot of messages when attempting to run kdenlive from terminal
2020-09-19 14:08:44 <abenz> and it auto terminates after showing splash (Doesnt get stuck like earlier)
2020-09-19 14:09:00 <abenz> last message complains about "QtQuick.Controls version 1.4 is not installed
2020-09-19 14:09:38 <abenz> but qt5-qtquickcontrols is installed
2020-09-19 14:10:34 <ikke> apk version qt5-qtquickcontrols
2020-09-19 14:11:35 <abenz> apk version qt5-qtquickcontrols
2020-09-19 14:11:35 <abenz> Installed:                                Available:
2020-09-19 14:11:42 <abenz> apk version qt5-qtquickcontrols2
2020-09-19 14:11:42 <abenz> Installed:                                Available:
2020-09-19 14:11:42 <abenz> qt5-qtquickcontrols2-5.15.1-r0          = 5.15.1-r0
2020-09-19 14:12:43 <abenz> it worked!
2020-09-19 14:12:46 <ikke> ?
2020-09-19 14:12:48 <abenz> at last
2020-09-19 14:12:49 <ikke> what worked
2020-09-19 14:13:44 <abenz> I installed qtquickcontrols (there's qtquickcontrols2), now I get a kdenlive popup box complaining about missing dependencies
2020-09-19 14:13:52 <abenz> shall I compile a list of the missing ones?
2020-09-19 14:14:12 <ikke> Yes
2020-09-19 14:14:18 <abenz> eg dbus, qtquickcontrols, breeze-icons.. among others
2020-09-19 14:14:32 <abenz> can you also please push the mlt fix?
2020-09-19 14:14:41 <ikke> yes, I will
2020-09-19 14:14:47 <abenz> cheers :)
2020-09-19 14:14:53 <ikke> And maybe open up a new issue for the missing dependencies
2020-09-19 14:15:10 <abenz> will do
2020-09-19 14:25:59 <sxiii> Maybe someone knows how to enable autologin on Alpine Live ISO? :)
2020-09-19 14:26:53 <sxiii> The problem is I probably have to install agetty into the image that I'm building. But I wasn't able to add any program into this image.
2020-09-19 15:04:19 <Stacy[m]> I would wait for expert to answer this, but I also remember there is a utility for installing alpine to usb
2020-09-19 15:07:44 <Hello71> dd
2020-09-19 15:35:46 <Guntoo> how to decrpyt gpg ?
2020-09-19 15:35:53 <Guntoo> gpgme or gnupg?
2020-09-19 15:37:03 <mps> gpgme is just 'easy' frontend to gnupg
2020-09-19 15:37:17 <mps> so, whatever you like
2020-09-19 15:37:32 <mps> Guntoo: ^
2020-09-19 15:37:37 <Guntoo> thangs
2020-09-19 15:37:41 <Guntoo> ks
2020-09-19 16:17:48 <whoismarvin42> Hi everyone, is anybody here being able to netboot RPi4 with latest Alpine (or Edge) ?
2020-09-19 16:22:21 <whoismarvin42> UEFI is OK, iPXE is OK too, kernel + initrd loaded (from CDN, *-rpi flavor, since lts crashes) + params inspired by boot.alpinelinux.org ... booting:EFI stub: Booting Linux Kernel...EFI stub: Generating empty DTBEFI stub: Exiting boot services and installing virtual address map...then nothing!
2020-09-19 16:23:35 <minimal> whoismarvin42: for RPI4 shouldn't you be using *-rpi4 rather than *-rpi?
2020-09-19 16:23:43 <crosbymichael> 2/c
2020-09-19 16:24:57 <whoismarvin42> no vmlinuz-rpi4 or initrd-rpi4 (only *-rpi .... or *-lts, aka kernel flavors)
2020-09-19 16:25:29 <whoismarvin42> https://dl-cdn.alpinelinux.org/alpine/v3.12/releases/aarch64/netboot-3.12.0/
2020-09-19 16:28:18 <minimal> haven't tried to netboot Alpine on a RPI but in terms of alpine packages there are linux-rpi, linux-rpi2, and linux-rpi4 packages and linux-rpi4 installs /boot/vmlinuz-rpi4 file
2020-09-19 16:28:19 <whoismarvin42> It was already a challenge to make previous steps happen ... but since files *-rpi are made available on the website/CDN, I suppose somebody had more success ... even if I'm not able to find a lots of docs around the topic
2020-09-19 16:28:43 <minimal> so I'd assume there should also be separate kernels for netbooting rpi, rpi2, and rpi4
2020-09-19 16:29:32 <whoismarvin42> thanks for your input
2020-09-19 16:32:36 <minimal> there's a kernel config file there, lemme have a quick look
2020-09-19 16:34:21 <whoismarvin42> Before going further, I'm using the files from the CDN trying to minimize sources of errors = mimic boot.alpinelinux.org ... which is outdated
2020-09-19 16:35:12 <whoismarvin42> So params / config / files / etc. are based on https://boot.alpinelinux.org/boot.ipxe
2020-09-19 16:36:24 <whoismarvin42> Everything working great (with latest or edge) ... loading files online (from CDN) ... but only for x86 ... not aarch64
2020-09-19 16:37:25 <minimal> I've only ever netbooted a RPI using Raspian where it pulls not only the kernel & initramfs from the network but also the RPI config.txt, usercfg.txt, etc from the network
2020-09-19 16:37:55 <whoismarvin42> Yep ... done that ... root using NFS
2020-09-19 16:38:38 <minimal> yeah, for RPI specifically I'm assuming either some hardware isn't initialised right or a kernel driver is missing, or overlays for RPI4 are missing from initramfs
2020-09-19 16:39:38 <whoismarvin42> working great ... but now seeking to find the ultimate answer = full remote boot without the need of local infrastructure (like NFS) = everything from the Web (Cloud ;)
2020-09-19 16:41:17 <minimal> the "Generating empty DTBEFI stub" is what makes me thing about the RPI DTB overlays
2020-09-19 16:42:22 <minimal> the DTB files are used to make various parts of the RPI hardware (network interface, video, etc) visible to the kernel so that it can then load drivers for them
2020-09-19 16:42:31 <whoismarvin42> Probably ... about the hardware or something missing ... based on what I read, RPi4 SoC has a bug regarding RAM (requiring to use only 3GB) ... but based on kernel available, it should be patched
2020-09-19 16:43:07 <whoismarvin42> Yep ...I agree with you about dtb stuff
2020-09-19 16:43:22 <minimal> the other thing is these files are from May - from memory not much more than 1 year since the RPI4 originally came out, they might be using an older version of the RPI firmware files that still have RPI4 issues
2020-09-19 16:43:34 <minimal> have you tried Edge?
2020-09-19 16:44:06 <whoismarvin42> but didn't find any doc on how to improve things in this context
2020-09-19 16:44:12 <whoismarvin42> Yep ... tried Edge
2020-09-19 16:44:40 <minimal> and did Edge work?
2020-09-19 16:44:40 <whoismarvin42> and all "compatible" ARM architectures
2020-09-19 16:44:43 <whoismarvin42> Nope
2020-09-19 16:44:48 <minimal> same problem?
2020-09-19 16:44:49 <whoismarvin42> Same message
2020-09-19 16:45:57 <minimal> looking at the ipxe config you referenced, have you tried changing line 7 to remove the "quiet" option, which should make the kernel give some output?
2020-09-19 16:46:02 <whoismarvin42> With other ARM arch (or lts flavor instead of rpi), it's not the same message: # Could not select: Exec format error (http://ipxe.org/2e008081)# Could not boot: Exec format error (http://ipxe.org/2e008081)
2020-09-19 16:46:13 <whoismarvin42> yep ... done that
2020-09-19 16:46:21 <minimal> and it made no diff?
2020-09-19 16:46:46 <whoismarvin42> I'm actually listining via GPIO (UART) = I have access to early boot messages
2020-09-19 16:47:04 <whoismarvin42> (aka serial console over USB)
2020-09-19 16:47:28 <whoismarvin42> nope no diff ... but I can provide all the boot messages
2020-09-19 16:47:50 <whoismarvin42> (not helping ... saying everything is fine ... until the 3 lines I posted)
2020-09-19 16:48:06 <minimal> also you're using EFI, I remember seeing something about a WIP uboot mod to do EFI on RPI, is that what this is using?
2020-09-19 16:48:52 <minimal> from the boot messages what exactly does the kernel do before it hangs? does it load any drivers, etc
2020-09-19 16:49:01 <whoismarvin42> EFI working super greate = can do whatever you want it ... and iPXE too (chainloaded) = I can use shell / tools / etc.
2020-09-19 16:51:02 <whoismarvin42> ESC (setup), F1 (shell), ENTER (boot)......>>Start PXE over IPv4.  Station IP address is 10.93.250.221  Server IP address is 10.93.250.11  NBP filename is efi/arm/netboot-arm_64.efi  NBP filesize is 221280 Bytes Downloading NBP file...  NBP file downloaded successfully.iPXE initialising devices...okiPXE 1.20.1+ (g4bd0) -- Open Source Network Boot
2020-09-19 16:51:02 <whoismarvin42> Firmware -- http://ipxe.orgFeatures: DNS HTTP HTTPS iSCSI TFTP AoE EFI MenuWaiting for link-up on net0...... okConfiguring (net0 dc:a6:32:58:71:c5)... okhttp://10.93.250.11:1080/boot.ipxe... okConfiguring (net0 dc:a6:32:58:71:c5)... okhttp://dl-cdn.alpinelinux.org/alpine/edge/releases/aarch64/netboot-20200917/vmlinuz-rpi... ok
2020-09-19 16:51:03 <whoismarvin42> http://dl-cdn.alpinelinux.org/alpine/edge/releases/aarch64/netboot-20200917/initramfs-rpi... okEFI stub: Booting Linux Kernel...EFI stub: Generating empty DTBEFI stub: Exiting boot services and installing virtual address map...
2020-09-19 16:52:08 <whoismarvin42> So you see: UEFI paylod -> iPXE payload -> Loading kernel + initrd -> booting ... failing with EFI messages
2020-09-19 16:53:08 <whoismarvin42> Driving me crazy  to be so close ;)
2020-09-19 16:53:32 <minimal> looking at that ipxe config, the aarch64 looks generic to me, nothing RPI specific. Normally when booting a RPI from SC card if I "cat /proc/cmdline" there are some entries there are appear to be generated by the RPI's bootcode.bin regarding DTB
2020-09-19 16:54:15 <minimal> I suspect that's you problem, its not a aarch64 issue, its a RPI issue where you need to pass some additional stuff to the kernel's command line
2020-09-19 16:57:09 <mps> yes, if kernel + initramfs load, userspace usually works. it is same for all aarch64
2020-09-19 16:57:18 <whoismarvin42> Yep ... I definitely agree with you ... it probably requires some extra stuff / params ... was hoping to find someone here with that specific knowledge since I was not able to find anywhere else ... it's too much of a guessing process otherwise
2020-09-19 16:57:27 <whoismarvin42> Thanks for your help / input
2020-09-19 16:58:25 <mps> I only tested on RPi zero, and not netboot but mmc
2020-09-19 16:58:50 <mps> with some tweaks to user cfg it boots fine
2020-09-19 16:59:04 <minimal> looking at a RPI2 next to me, /proc/cmdline shows entries for coherent_pool, 8250.nr_uarts, snd_bcm2835.enable.compat, snd_bcm2835.enable_hdmi, snd_bcm2835.enable_headphones, bcm2835_fb.fbwidth, smsc95xx.macaddr etc
2020-09-19 17:00:16 <minimal> note in particular the MAC address one and the snd_bcm3835.enable_hdmi one - one to tell the kernel the MAC address to use for the network port and the other to indicate if HDMI output should be enabled or not
2020-09-19 17:00:26 <mps> yes, these extras are from RPi bootloader, I think
2020-09-19 17:01:09 <minimal> they're set up by bootcode.bin normally (including read from config.txt and usercfg.txt). In your netboot config I don't know where those sort of settings are supposed to come from
2020-09-19 17:02:26 <minimal> whoismarvin42: so any idea where these sort of settings are supposed to come from in a netboot setup?
2020-09-19 17:03:20 <whoismarvin42> I'm probably going to create a local server instead of relying on Alpine CDN ... being able to tweak my conf + "play" with all these ideas
2020-09-19 17:04:09 <minimal> I haven't looked into netbooting Alpine so far and so haven't looked to see how the netboot images are being created
2020-09-19 17:04:41 <minimal> I've assuming there's a script for it like there is for chroot and VM image creation
2020-09-19 17:05:13 <whoismarvin42> @minimal all params are passed as arguments in the "kernel line" (look at the pxe script)
2020-09-19 17:06:43 <minimal> whoismarvin42: my point is in addition to what is specified in the "kernel line" part for a RPI normally bootcode.bin also adds additional entries and they're what appear to be missing
2020-09-19 17:07:38 <whoismarvin42> I didn't want to mess with the process of creating the netboot images ... was expecting the one provided to work ... but since I cannot crack this last step ... I'm probably going to dockerize all the process ... to be able to build / package all the components
2020-09-19 17:07:58 <minimal> looking at https://gitlab.alpinelinux.org/alpine/aports/-/blob/master/scripts/mkimg.netboot.sh there's mention of rpi and rpi2 but no mention of rpi4
2020-09-19 17:09:03 <whoismarvin42> I thing aarch64+ rpi = rpi4
2020-09-19 17:09:43 <whoismarvin42> (at least, it's how I understood it ... since rpi is not aarch64)
2020-09-19 17:10:11 <artok> aarch64 rpi = rpi3 ?
2020-09-19 17:10:26 <whoismarvin42> also
2020-09-19 17:10:44 <artok> and linux-rpi4 = rpi4
2020-09-19 17:10:53 <minimal> for Alpine on aarch64 normally the linux-rpi package is for RPI2 and RPI3 and the linux-rpi4 package is (obviously) for RPI4
2020-09-19 17:11:18 <minimal> I'm running Alpine on a RPI3+ using aarch linux-rpi package
2020-09-19 17:11:36 <artok> linux-rpi aarch64 is not rpi2 runnable
2020-09-19 17:11:41 <minimal> the older RPIs didn't support 64bit, think it was added to one of the RPI2 revisions
2020-09-19 17:12:00 <artok> surely rpi3 is the first 64bit?
2020-09-19 17:12:02 <minimal> so some RPI2's run 64bit, some don't
2020-09-19 17:12:10 <artok> which ones run?
2020-09-19 17:12:53 <minimal> AFAIK in a later revision of the RPI2 boards they changed the processor to a newer version which does do 64bit
2020-09-19 17:13:02 <whoismarvin42> yep agree with that (talking about the package) ... but in the scrpit you've mentionned (https://gitlab.alpinelinux.org/alpine/aports/-/blob/master/scripts/mkimg.netboot.sh) ... I think the shell switch case "aarch64 + rpi" is refering to RPi4
2020-09-19 17:14:00 <minimal> artok: a guess search gives "RPi 2B version 1.2 BCM2837 SoC"
2020-09-19 17:16:27 <whoismarvin42> minimal: regarding your remark about additional arguments on the kernel line ... I agree ... extra params are probably required ... on top of that ... you can load additional files if needed before loading the kernel ... my problem is, in fact, the lack of documentation
2020-09-19 17:16:41 <minimal> whoismarvin42: line 29? going by Alpine packaging terminology, "rpi" refers to "linux-rpi" package for aarch64 which is for RPI2&3, the following line's reference to "rpi" and "rpi2" for armv7 appears to fine in with this as linux-rpi and linux-rpi2 for armv7 are packages for everything but the RPI4.
2020-09-19 17:17:50 <minimal> I suspect that lines 29 & 30 need to have "rpi4" added to them before they're used to regenerate netboot images to then (hopefully) create images suitable for RPI4
2020-09-19 17:22:22 <whoismarvin42> I think (event if it's unforunate) that 'rpi' does not have the same meaning in line 29 and 30 (rpi v1 on line 30, rpi v4 on line 29 since rpi v1 is not aarch53) ... but no matter what ... I'm going to take some time to build a "local" solution ... being able to fully automated my next tests
2020-09-19 17:22:38 <whoismarvin42> One more time, thanks for your input
2020-09-19 17:26:02 <minimal> whoismarvin42: on lines 29 & 30 "rpi" doesn't mean a RPI1 machine, its shorthand for Alpine package "linux-rpi"
2020-09-19 17:26:03 <whoismarvin42> btw, do you know if there is any logs / history that I can search (looking for infos / people having the same problem) ... or any other pointer ?
2020-09-19 17:26:21 <minimal> I'm interested to find out how you get on as I intend to make use of netboot myself in the near future
2020-09-19 17:27:20 <minimal> I'm having a look at the ipxe source currently to see how they handle aarch64 and if there is anything RPI-specific in there. The last time I looked at ipxe was before they added aarch64 support to it.
2020-09-19 17:28:48 <whoismarvin42> iPXE is fully functional ... just have to build it for aarch64
2020-09-19 17:29:21 <abenz> ikke: https://gitlab.alpinelinux.org/alpine/aports/-/issues/11959
2020-09-19 17:29:23 <whoismarvin42> But before being able to use it you'll have to play with EFI firmware
2020-09-19 17:29:52 <minimal> whoismarvin42: I don't really thing so - I found the same issue recently trying to diagnose a different RPI issue and went on the raspberrypi IRC channel but didn't get much help fronm there
2020-09-19 17:30:21 <abenz> ikke: even when dbus is not started (but installed), I can run kdenlive fine but it must be started with "dbus-run-session kdenlive", is this expected?
2020-09-19 17:30:42 <minimal> whoismarvin42: where is the EFI firmware though? Isn't it coming from alpine package alpine-ipxe-snp_efi (which is from ipxe)?
2020-09-19 17:30:49 <whoismarvin42> I'm actually working on packaging my previous steps to use the SPI in RPi4 ... and remove the need of SDCard = build + netboot cluster
2020-09-19 17:32:44 <whoismarvin42> Nope ... iPXE is "just" one of the payload you can boot from you EFI booloader
2020-09-19 17:34:32 <minimal> whoismarvin42: for multi RPI2 node (whether clustered or standalone) I add an RTC and a POE adaptor and write a prepared Alpine image to SD cards for each machine, I prefer them able to boot without needing a netboot server present. I've got the RPIs each mounted onto a piece of plastic so they are be vertically slide into a frame for mounting side-by-side under the POE switch
2020-09-19 17:36:17 <minimal> whoismarvin42: so where does the EFI come from? you've got the Pi doing basic netboot to pull EFI file which then loads and pulls ipxe file which then pulls its config file?
2020-09-19 17:36:35 <whoismarvin42> I'have a similar setup (19" rack with 3d printed frame) ... but my goal for the end of this year is to a single image as base for all my needs
2020-09-19 17:38:13 <mps> some other ARM SOCs are more open than Broadcom (RPi)
2020-09-19 17:38:45 <minimal> whoismarvin42: done that - cloud-init enabled image written to SD card with a partition containing the cloud-init YAML file. Want 3 RPIs up and running? write the image to 3 SD cards and edit the YAML on each card to taste - specify whether you want static or dhcp networking, which Alpine repo to use, whuch users and groups to create, which SSH keys, etc. 1st boot of the RPIs cloud-init then does its magic
2020-09-19 17:39:02 <whoismarvin42> for now: SDCard a UEFI bootlaoder => TFTP for iPXE => netboot (load kernel + initrd + whatever is needed)
2020-09-19 17:39:56 <minimal> whoismarvin42: so where did you get the EFI bootloader from?
2020-09-19 17:39:58 <whoismarvin42> I want to get rid of the SDCard (and failures coming with it) ... at least, it's my goal
2020-09-19 17:40:31 <whoismarvin42> https://www.virtuallyghetto.com/2020/07/two-methods-to-network-boot-raspberry-pi-4.html
2020-09-19 17:40:49 <whoismarvin42> look at "Alternative Network Boot on rPI"
2020-09-19 17:43:24 <whoismarvin42> But ubuntu doesn't boot (probably why he's not answering the question in comments) ... provided binaries are for the wrong arch ... back to similar problem as Alpine
2020-09-19 17:44:21 <whoismarvin42> for now of course ... probably more and more people are going to ask these questions ... and we'll find answers eventually ;)
2020-09-19 17:45:22 <whoismarvin42> You got the link to the UEFI firmware ?
2020-09-19 17:45:37 <minimal> whoismarvin42: right, that's experimental 3rd party UEFI for RPI4. So won't work on the 3/3+ machine I have here
2020-09-19 17:46:18 <minimal> whoismarvin42: BTW did you notice this commit to the upstream project: https://github.com/tianocore/edk2-platforms/commit/e6d70ee53e209de2fc65c2c76c3182913db710b2
2020-09-19 17:46:19 <whoismarvin42> I'm only targetting RPi4 because of the SPI flash
2020-09-19 17:48:26 <minimal> the "Limit RAM to 3G" setting in particular might be relevant, as someone else mentioned at the start of this discussion when the 4GB RPI4 first came out there was some problem that was kludged around by only using 3Gb........its probably been fixed in more recent firmware
2020-09-19 17:48:32 <whoismarvin42> Yep ... about the commit (that's why I was talking about the 3GB RAM problem)
2020-09-19 17:51:00 <minimal> with the 8GB RPI4 out now I expect more people to look into this sort of stuff. I intend to get a 8GB RPI4 to test how well it works as a container cluster node
2020-09-19 17:53:25 <whoismarvin42> All versions are working quite well (depending on the load / and you heat dissipation system of course) ... but the 8GB is probably the way to go for kubernetes
2020-09-19 17:54:10 <minimal> whoismarvin42: as the UEFI stuff is not packaged for Alpine I suspect that netboot has never been tested against it. The aarch64 linux kernel in general may well have UEFI support but the issue is whether the RPI kernel built for Alpine have any/all EFI stuff enabled in them.
2020-09-19 17:56:19 <whoismarvin42> I'm suspecting that too
2020-09-19 17:56:22 <minimal> basically the Alpine RPI netboot is expecting to interact with RPI's bootcode.bin to hand over some info in a particular way. the UEFI bootloader you're using want to handover any info in a UEFI type of way which this kernel may not be ready for/may not understand.
2020-09-19 17:58:15 <whoismarvin42> I'm suspecting that too ... and since there is not rpi4 in releases/aarch64/netboot-3.12.0/dtbs-lts/broadcom/ ... you're probably more than right ... it's probably still too soon
2020-09-19 17:58:18 <mps> minimal: next -lts will probably work for RPi
2020-09-19 17:58:22 <minimal> unfortunately you may need to build your own RPI4 kernels to experiment. I'm settings up a machine here currently for Alpine armv7 & aarch64 package builds as my typical method for build Alpine packages locallly in general (QEMU emulating armv7/aarch64) is not realistic for kernels - they'd take forever to build
2020-09-19 17:59:30 <mps> I'm tested mainline kernel 5.8 on RPi zero, it boots and works fine, without any patches
2020-09-19 18:00:19 <mps> so probably it will work on other RPis
2020-09-19 18:00:21 <whoismarvin42> I intend to do so (build my own kernel) ... but was hoping to have a "working point" to begin
2020-09-19 18:00:57 <minimal> whoismarvin42: have you tried looking at the contents of the initramfs you're using to verify it contains the DTB overlay files for RPI4? Its basically a cpio archive.
2020-09-19 18:01:20 <mps> I'm ready to 'tweak' linux-edge aarch64 for RPi if someone are ready to test it, I don't have board
2020-09-19 18:02:47 <whoismarvin42> I didn't ... but based on what we said previously  ... guessing it's lacking the required content
2020-09-19 18:09:35 <minimal> whoismarvin42:  zcat initramfs | cpio -t
2020-09-19 18:16:00 <whoismarvin42> ..modloopbinbin/busyboxbin/shdevetcetc/apketc/apk/keysetc/apk/keys/alpine-devel@lists.alpinelinux.org-524d27bb.rsa.pubetc/apk/keys/alpine-devel@lists.alpinelinux.org-58199dcc.rsa.pubetc/fstabetc/groupetc/mdev.confetc/modprobe.detc/modprobe.d/aliases.confetc/modprobe.d/blacklist.confetc/modprobe.d/i386.confetc/modprobe.d/kms.confetc/passwdinitliblib
2020-09-19 18:16:01 <whoismarvin42> /firmwarelib/firmware/ene-ub6250lib/firmware/ene-ub6250/ms_init.binlib/firmware/ene-ub6250/ms_rdwr.binlib/firmware/ene-ub6250/msp_rdwr.binlib/firmware/ene-ub6250/sd_init1.binlib/firmware/ene-ub6250/sd_init2.binlib/firmware/ene-ub6250/sd_rdwr.binlib/ld-musl-aarch64.so.1lib/libblkid.so.1lib/libblkid.so.1.1.0lib/libc.musl-aarch64.so.1lib/libcrypto.so.
2020-09-19 18:16:01 <whoismarvin42> 1.1lib/libcryptsetup.so.12lib/libcryptsetup.so.12.6.0lib/libdevmapper.so.1.02lib/libkmod.so.2lib/libkmod.so.2.3.5lib/libssl.so.1.1lib/libuuid.so.1lib/libuuid.so.1.3.0lib/libz.so.1lib/libz.so.1.2.11lib/mdevlib/mdev/dvbdevlib/mdev/ide_linkslib/mdev/usbdevlib/mdev/usbdisk_linklib/mdev/xvd_linkslib/moduleslib/modules/5.4.43-0-rpilib/modules/5.4.43-0-rp
2020-09-19 18:16:02 <whoismarvin42> i/kernellib/modules/5.4.43-0-rpi/kernel/driverslib/modules/5.4.43-0-rpi/kernel/drivers/hidlib/modules/5.4.43-0-rpi/kernel/drivers/hid/hid-apple.kolib/modules/5.4.43-0-rpi/kernel/drivers/hid/hid-cherry.kolib/modules/5.4.43-0-rpi/kernel/drivers/netlib/modules/5.4.43-0-rpi/kernel/drivers/net/ethernetlib/modules/5.4.43-0-rpi/kernel/drivers/net/ethernet
2020-09-19 18:16:02 <whoismarvin42> /microchiplib/modules/5.4.43-0-rpi/kernel/drivers/net/ethernet/microchip/enc28j60.kolib/modules/5.4.43-0-rpi/kernel/drivers/net/ethernet/qualcommlib/modules/5.4.43-0-rpi/kernel/drivers/net/ethernet/qualcomm/qca_7k_common.kolib/modules/5.4.43-0-rpi/kernel/drivers/net/ethernet/qualcomm/qcaspi.kolib/modules/5.4.43-0-rpi/kernel/drivers/net/ethernet/wiz
2020-09-19 18:16:03 <whoismarvin42> netlib/modules/5.4.43-0-rpi/kernel/drivers/net/ethernet/wiznet/w5100-spi.kolib/modules/5.4.43-0-rpi/kernel/drivers/net/ethernet/wiznet/w5100.kolib/modules/5.4.43-0-rpi/kernel/drivers/usblib/modules/5.4.43-0-rpi/kernel/drivers/usb/storagelib/modules/5.4.43-0-rpi/kernel/drivers/usb/storage/uas.kolib/modules/5.4.43-0-rpi/kernel/drivers/usb/storage/ums
2020-09-19 18:16:03 <whoismarvin42> -alauda.kolib/modules/5.4.43-0-rpi/kernel/drivers/usb/storage/ums-cypress.kolib/modules/5.4.43-0-rpi/kernel/drivers/usb/storage/ums-datafab.kolib/modules/5.4.43-0-rpi/kernel/drivers/usb/storage/ums-eneub6250.kolib/modules/5.4.43-0-rpi/kernel/drivers/usb/storage/ums-freecom.kolib/modules/5.4.43-0-rpi/kernel/drivers/usb/storage/ums-isd200.kolib/modul
2020-09-19 18:16:04 <whoismarvin42> es/5.4.43-0-rpi/kernel/drivers/usb/storage/ums-jumpshot.kolib/modules/5.4.43-0-rpi/kernel/drivers/usb/storage/ums-karma.kolib/modules/5.4.43-0-rpi/kernel/drivers/usb/storage/ums-onetouch.kolib/modules/5.4.43-0-rpi/kernel/drivers/usb/storage/ums-realtek.kolib/modules/5.4.43-0-rpi/kernel/drivers/usb/storage/ums-sddr09.kolib/modules/5.4.43-0-rpi/kerne
2020-09-19 18:16:05 <whoismarvin42> l/drivers/usb/storage/ums-sddr55.kolib/modules/5.4.43-0-rpi/kernel/drivers/usb/storage/ums-usbat.kolib/modules/5.4.43-0-rpi/kernel/fslib/modules/5.4.43-0-rpi/kernel/fs/nlslib/modules/5.4.43-0-rpi/kernel/fs/nls/nls_cp1250.kolib/modules/5.4.43-0-rpi/kernel/fs/nls/nls_cp1251.kolib/modules/5.4.43-0-rpi/kernel/fs/nls/nls_cp1255.kolib/modules/5.4.43-0-rp
2020-09-19 18:16:05 <whoismarvin42> i/kernel/fs/nls/nls_cp737.kolib/modules/5.4.43-0-rpi/kernel/fs/nls/nls_cp775.kolib/modules/5.4.43-0-rpi/kernel/fs/nls/nls_cp850.kolib/modules/5.4.43-0-rpi/kernel/fs/nls/nls_cp852.kolib/modules/5.4.43-0-rpi/kernel/fs/nls/nls_cp855.kolib/modules/5.4.43-0-rpi/kernel/fs/nls/nls_cp857.kolib/modules/5.4.43-0-rpi/kernel/fs/nls/nls_cp860.kolib/modules/5.4.
2020-09-19 18:16:05 <whoismarvin42> 43-0-rpi/kernel/fs/nls/nls_cp861.kolib/modules/5.4.43-0-rpi/kernel/fs/nls/nls_cp862.kolib/modules/5.4.43-0-rpi/kernel/fs/nls/nls_cp863.kolib/modules/5.4.43-0-rpi/kernel/fs/nls/nls_cp864.kolib/modules/5.4.43-0-rpi/kernel/fs/nls/nls_cp865.kolib/modules/5.4.43-0-rpi/kernel/fs/nls/nls_cp866.kolib/modules/5.4.43-0-rpi/kernel/fs/nls/nls_cp869.kolib/modul
2020-09-19 18:16:06 <whoismarvin42> es/5.4.43-0-rpi/kernel/fs/nls/nls_cp874.kolib/modules/5.4.43-0-rpi/kernel/fs/nls/nls_cp932.kolib/modules/5.4.43-0-rpi/kernel/fs/nls/nls_cp936.kolib/modules/5.4.43-0-rpi/kernel/fs/nls/nls_cp949.kolib/modules/5.4.43-0-rpi/kernel/fs/nls/nls_cp950.kolib/modules/5.4.43-0-rpi/kernel/fs/nls/nls_euc-jp.kolib/modules/5.4.43-0-rpi/kernel/fs/nls/nls_iso8859-1
2020-09-19 18:16:06 <whoismarvin42> .kolib/modules/5.4.43-0-rpi/kernel/fs/nls/nls_iso8859-13.kolib/modules/5.4.43-0-rpi/kernel/fs/nls/nls_iso8859-14.kolib/modules/5.4.43-0-rpi/kernel/fs/nls/nls_iso8859-15.kolib/modules/5.4.43-0-rpi/kernel/fs/nls/nls_iso8859-2.kolib/modules/5.4.43-0-rpi/kernel/fs/nls/nls_iso8859-3.kolib/modules/5.4.43-0-rpi/kernel/fs/nls/nls_iso8859-4.kolib/modules/5.
2020-09-19 18:16:07 <whoismarvin42> 4.43-0-rpi/kernel/fs/nls/nls_iso8859-5.kolib/modules/5.4.43-0-rpi/kernel/fs/nls/nls_iso8859-6.kolib/modules/5.4.43-0-rpi/kernel/fs/nls/nls_iso8859-7.kolib/modules/5.4.43-0-rpi/kernel/fs/nls/nls_iso8859-9.kolib/modules/5.4.43-0-rpi/kernel/fs/nls/nls_koi8-r.kolib/modules/5.4.43-0-rpi/kernel/fs/nls/nls_koi8-ru.kolib/modules/5.4.43-0-rpi/kernel/fs/nls/
2020-09-19 18:16:07 <whoismarvin42> nls_koi8-u.kolib/modules/5.4.43-0-rpi/kernel/fs/nls/nls_utf8.kolib/modules/5.4.43-0-rpi/kernel/fs/overlayfslib/modules/5.4.43-0-rpi/kernel/fs/overlayfs/overlay.kolib/modules/5.4.43-0-rpi/kernel/fs/squashfslib/modules/5.4.43-0-rpi/kernel/fs/squashfs/squashfs.kolib/modules/5.4.43-0-rpi/kernel/netlib/modules/5.4.43-0-rpi/kernel/net/packetlib/modules/5
2020-09-19 18:16:08 <whoismarvin42> .4.43-0-rpi/kernel/net/packet/af_packet.kolib/modules/5.4.43-0-rpi/modules.aliaslib/modules/5.4.43-0-rpi/modules.alias.binlib/modules/5.4.43-0-rpi/modules.builtinlib/modules/5.4.43-0-rpi/modules.builtin.alias.binlib/modules/5.4.43-0-rpi/modules.builtin.binlib/modules/5.4.43-0-rpi/modules.deplib/modules/5.4.43-0-rpi/modules.dep.binlib/modules/5.4.43
2020-09-19 18:16:08 <whoismarvin42> -0-rpi/modules.devnamelib/modules/5.4.43-0-rpi/modules.orderlib/modules/5.4.43-0-rpi/modules.softdeplib/modules/5.4.43-0-rpi/modules.symbolslib/modules/5.4.43-0-rpi/modules.symbols.binmediamedia/cdrommedia/floppymedia/usbnewrootprocrunsbinsbin/apksbin/nlplug-findfssysusrusr/binusr/bin/ssl_clientusr/libusr/lib/libargon2.so.1usr/lib/libjson-c.so.5usr
2020-09-19 18:16:09 <whoismarvin42> /lib/libjson-c.so.5.0.0usr/lib/liblzma.so.5usr/lib/liblzma.so.5.2.5usr/lib/libtls-standalone.so.1usr/lib/libtls-standalone.so.1.0.0usr/shareusr/share/udhcpcusr/share/udhcpc/default.scriptvarvar/cachevar/cache/miscvar/cache/misc/modloop-rpi.SIGN.RSA.alpine-devel@lists.alpinelinux.org-58199dcc.rsa.pub15622 blocks
2020-09-19 18:16:18 <ikke> whoismarvin42: Please use a paste service
2020-09-19 18:16:23 <ikke> This is just spam
2020-09-19 18:16:26 <whoismarvin42> sorry
2020-09-19 18:16:31 <minimal> you're looking for things like bcm2711-rpi-4-b.dtb and also bcm2835_smi_dev.ko which is a kernel module specific to the RPI4
2020-09-19 18:17:44 <minimal> (or at least specific to the Alpine kernel packaging for linux-rpi4)
2020-09-19 18:19:08 <whoismarvin42> Nope, nothing there
2020-09-19 18:19:13 <whoismarvin42> As suspected
2020-09-19 18:22:25 <minimal> just looking at the aarch linux-rpi kernel config on an RPI2 - I see that CONFIG_FB_EFI is not set which is the Framebuffer driver for UEFI use. the linux-rpi2 and linux-rpi4 packages use the same kernel config and so as that is not set it may explain why you see nothing on screen...
2020-09-19 18:23:54 <minimal> its not set because Alpine RPI doesn't use UEFI. I don't know if any other RPI distributions like Ubuntu or Raspbian set that or any other UEFI-related kernel settings.
2020-09-19 18:25:03 <minimal> basically if you want to use UEFI you'll have to configure and build a kernel package yourself
2020-09-19 18:25:36 <minimal> whoismarvin42: why not stick with the normal netboot for RPI?
2020-09-19 18:26:49 <whoismarvin42> "normal" netboot requires SDCard + TFTP + NFS ... targetted solution ... UEFI Firmware in SPI flash + HTTP boot = zero local requirements
2020-09-19 18:27:50 <minimal> whoismarvin42: why does it need a SDcard (apart from an initial one to boot off to change the eeprom settings)?
2020-09-19 18:27:53 <whoismarvin42> (and less problems with SD card failures ... with Rpi on remote locations)
2020-09-19 18:29:37 <minimal> I helped with the original netboot testing for the RPI3 several years ago and it (sort of) worked without an SDcard. [In my case they had some switch-specific misbehaviours that so I needed only bootcode.bin on an SDcard but this was fixed in the built-in firmware as of 3+ onwards AFAIK]
2020-09-19 18:30:23 <whoismarvin42> even if you remove the SDCard ... you still need TFTP + NFS ... u-boot cannot achieve HTTP boot ... so no other solution than praying for UEFI ... which I have waited for a long time ;)
2020-09-19 18:30:37 <minimal> perhaps you're referring to iPXE boot ("booting over the Internet") which is not exactly the same thing
2020-09-19 18:30:53 <minimal> for normal netboot you shouldn't need an SD card
2020-09-19 18:33:49 <whoismarvin42> I agree about the SDCard ... and you'right about RPi3  (btw, having only to do read operations, based on the scenario you're referring, the SDcard may not die as soon as expected) ... but the ultimate goal is to achieve HTTPS netboot
2020-09-19 18:34:22 <whoismarvin42> (at least be capable of)
2020-09-19 18:35:42 <minimal> whoismarvin42: the EFI bootloader is clearly flagged as experimental - so unfortunately you're on your own really..........I'll be glad in a few months time when you've figured it all out (including kernel config changes) and documented it and perhaps submitted Alpine MRs for things like kernel config and they I can get it to work in 5 mins ;-)
2020-09-19 18:37:55 <whoismarvin42> Thank you so much for your input ... as you are pointing ... it's probably a matter of time ... I'll probably give it another try around Christmas ... who knows ... good things tend to happen in that period
2020-09-19 18:44:18 <minimal> whoismarvin42: I've always liked the idea of iPXE's https boot (EFI HTTP is a more recent thing & from memory its optional, not mandatory, in a UEFI stack)
2020-09-19 18:45:36 <whoismarvin42> I'm in love <3 witg this killing feature (and the capability to check / verifiy payloads)
2020-09-19 18:46:41 <minimal> are you using your own custom certs/CA?
2020-09-19 18:47:16 <whoismarvin42> and praying for them to implement DoH one day
2020-09-19 18:47:37 <minimal> DoH for UEFI?
2020-09-19 18:47:57 <whoismarvin42> for iPXE payloads
2020-09-19 18:48:51 <whoismarvin42> Had trouble with custom certs/CA
2020-09-19 18:48:54 <minimal> both DoE in iPXE and HTTPS for iPXE and UEFI would have the issue of built-in CAs and how to handle expired or blacklisted CAs
2020-09-19 18:53:50 <whoismarvin42> at least it would be an improvment compared to what is is actually ... and people could start thinking about it ... and how to do it in a better way ... probably the opportunities around RPi / IoT and all this "new devices / sbc" will provoke a discussion ...
2020-09-19 18:54:11 <whoismarvin42> in the "near future"
2020-09-19 18:54:58 <whoismarvin42> if you're looking for iPXE with custom certs / CA ... look at netboo.xyz
2020-09-19 18:55:05 <minimal> whoismarvin42: yes I'd looked at iPXE with HTTPS and a private CA a while ago as a way to provide simple plug-it-in-and-go servers
2020-09-19 18:55:07 <whoismarvin42> netboot.xyz
2020-09-19 18:55:07 <Guntoo> whoismarvin42: sync your system time
2020-09-19 18:55:24 <whoismarvin42> He provides an Ansible playbook
2020-09-19 18:55:43 <Guntoo> minimal: doesn't exist yet ipxe+https ?
2020-09-19 18:55:59 <minimal> whoismarvin42: do you have an RTC on your RPI(s)?
2020-09-19 18:56:44 <minimal> Guntoo: https://ipxe.org/crypto
2020-09-19 18:57:17 <whoismarvin42> NTP for the RPI(s) ... RTC on another SBC (ntp server)
2020-09-19 18:59:45 <minimal> whoismarvin42: used iPXE on x86_64 in a previous place for automating the bootstrapping large clusters
2020-09-19 19:00:43 <whoismarvin42> iPXE is a great project indeed
2020-09-19 19:00:54 <Guntoo> minimal: clusters of IoTs?
2020-09-19 19:01:22 <Guntoo> minimal: do those work with self-certs?
2020-09-19 19:01:41 <minimal> Guntoo: nope, large (and I really mean large) clusters for enterprise in-memory databases
2020-09-19 19:03:00 <Guntoo> minimal: so, what does that solve, controlling everything from one termiinal, such power on/off/reboot from a single terminal without having to walk into each machine?
2020-09-19 19:05:29 <minimal> Guntoo: private cloud bare metal servers dedicated to large customers - automation built and destroyed the clusters via iPXE (which was built into the network adaptors)
2020-09-19 19:08:52 <Guntoo> destroy themL
2020-09-19 19:08:54 <Guntoo> ?
2020-09-19 19:09:17 <minimal> Guntoo: yupe, automation used IPMI to poweron machines to built the clusters and powered them off when clusters decommissoned. You wouldn't walk to machines when they're in multiple Data Centers around the world ;-)
2020-09-19 19:10:25 <minimal> Guntoo: destroy == decommissioned. When customer is finished with the cluster the disks are wiped and the machines powered off.
2020-09-19 19:16:28 <Guntoo> wow
2020-09-19 19:16:30 <Guntoo> insane
2020-09-19 19:18:32 <whoismarvin42> mps: sorry, I missed your message. If you want to "play" with Alpine on RPi ... I could give it a try ... and tests your kernels / whatever you need
2020-09-19 19:19:09 <whoismarvin42> minimal: nice ... you probably have learned a lot in that context
2020-09-19 19:20:41 <minimal> whoismarvin42: never got to "play" with machines as large (memory, cores, SSD space) anywhere else
2020-09-19 19:27:48 <mps> whoismarvin42: np. just tell or write what drivers and options should be enabled in linux-edge, https://gitlab.alpinelinux.org/alpine/aports/-/blob/master/testing/linux-edge/config-edge.aarch64
2020-09-19 19:28:20 <mps> and test it after it is built, ofc :)
2020-09-19 19:46:53 <whoismarvin42> mps: sorry for the delay ... I don't have a clue of what driver / options should be enabled / what parameters should be passed / flag enabled (or not) ... for now ... but, I'll probably come back soon (or publish the solution if I find it). I have to go right now, lost track of the time ... as always with with "playful" topics ;)
2020-09-19 19:48:02 <mps> whoismarvin42: np, we have time till the next -lts kernel announce
2020-09-19 19:48:23 <minimal> whoismarvin: you'll need to go through the set of CONFIG_EFI* settings to see what you need
2020-09-19 19:49:30 <whoismarvin42> Yep .. I was expecting it to work "out of the box" ... it didn't = I'll have to dig ... thank you all
2020-09-19 19:50:16 <whoismarvin42> Have a nice WE
2020-09-19 21:35:45 <q66> Ariadne: yo, just btw, i'm gonna pick up the mesa aarch64 patch you added into alpine's mesa for void
2020-09-19 21:36:14 <q66> just tested it on my honeycomb box with rx5500 and finally no more x11 triangle junk (on either glibc or musl)
2020-09-19 21:36:38 <Ariadne> there's an updated one
2020-09-19 21:36:44 <Ariadne> I need to add it
2020-09-19 21:36:47 <Ariadne> will do after nap
2020-09-19 21:37:00 <q66> if you can link me that'd be great
2020-09-19 21:37:05 <q66> i want to get it in with 20.1.8 update
2020-09-19 22:13:09 <yjftsjthsd> Does anyone run Steam on Alpine? I tried it a while back with flatpak but got nowhere so I ended up just using Arch for games, but now I'm looking at rebuilding that box and it would be super nice to use Alpine, even if I need to shove steam in a container or something (actually given that it's proprietary software with some iffy practices that
2020-09-19 22:13:09 <yjftsjthsd> would be ideal)
2020-09-19 22:15:47 <Cogitri> I use Steam in Flatpak and that just works
2020-09-19 22:16:48 <yjftsjthsd>  Odd, I'll have to retry and see if it works now. Thanks
2020-09-19 22:42:11 <shodan45> is it safe to upgrade a 3.7 install directly to 3.12?
2020-09-19 23:07:41 <shodan45> answering my own question: yeah, pretty much :)
2020-09-19 23:09:50 <zcrayfish> nice
2020-09-19 23:19:40 <shodan45> seems there's no package or plan to add dehydrated to alpine :(
2020-09-19 23:20:35 <shodan45> which is a shame, I much prefer it over the standard lets encrypt client
2020-09-19 23:22:39 <shodan45> it even fits really well with alpine's "small simple secure" creed
2020-09-19 23:22:52 <shodan45> slogan? tagline?
2020-09-19 23:23:13 <zcrayfish> I use acme.sh for that purpose, personally
2020-09-19 23:23:47 <shodan45> zcrayfish: I'll take a look
2020-09-20 03:04:28 <abenz> PureTryOut[m]: hi, ikke pushed the fix onto 3.12, when fixing dependencies, please bump kdenlive to 20.08.1 (on 3.12) while at it
2020-09-20 03:04:43 <abenz> if its not too much trouble
2020-09-20 03:04:47 <abenz> thanks a lot :)
2020-09-20 03:09:01 <atkka> hello everyone
2020-09-20 03:11:53 <atkka> i'm having issues getting audio to work with flatpaks, no sound in firefox or wire messenger, dbus or pulseaudio issue probably
2020-09-20 03:12:56 <atkka> anyone know what needs to be done to allow flatpaks to access pulse/alsa?
2020-09-20 06:27:46 <ikke> abenz: nice, I've reported it to mlt upstream (on an already existing issue), and it will now be fixed in 6.24
2020-09-20 07:06:24 <Ariadne> q66: pushing mesa 20.1.5-r2 with the updated patch
2020-09-20 07:16:32 <Ariadne> q66: are you in solidrun discord?
2020-09-20 07:16:45 <Ariadne> q66: if you are not and you want to be i can generate an invite link
2020-09-20 08:01:49 <abenz> ikke: many thanks! But you also pushed the mlt fix locally on alpine 3.12/edge right? I should be able to apk upgrade -U into the fixed version in a while, no?
2020-09-20 08:02:48 <abenz> will mlt's version be incremented (or is the system aware that a different version exist and refetch/reinstall ?
2020-09-20 08:08:54 <ikke> Yes, I've pushed it to 3.12 with a pkgrel bump, so it should automatically be installed on upgrade
2020-09-20 09:26:31 <abenz> ikke: cheers!
2020-09-20 10:05:18 <paper_> can I build alpine packages in an alpine chroot? I am getting "ERROR: Unable to lock database: Bad file descriptor" and "ERROR: Failed to open apk database: Bad file descriptor"
2020-09-20 10:05:54 <paper_> that's running "abuild -r" as an unpriviledged user in abuild group
2020-09-20 10:06:08 <Cogitri> Maybe you forgot to mount /proc /sys and /dev?
2020-09-20 10:08:12 <paper_> not sure, I think the script is mounting it: https://ttm.sh/hw0.sh
2020-09-20 10:08:45 <paper_> full output of abuild -r: https://ttm.sh/hwq.txt
2020-09-20 10:10:50 <paper_> mount says "none on /proc type proc (rw,relatime)", is that okay?
2020-09-20 10:11:54 <Cogitri> Yes, that should be fine
2020-09-20 10:12:19 <Cogitri> Ah actually, I guess there's no apk database in your chroot for some reason?
2020-09-20 10:12:36 <ikke> paper_: does /lib/apk/db/lock exist?
2020-09-20 10:14:01 <paper_> ikke: yes
2020-09-20 10:14:11 <ikke> Then remove it
2020-09-20 10:14:30 <ikke> Assuming no other process is using apk
2020-09-20 10:15:01 <paper_> I removed it and abuild -r still says the same
2020-09-20 10:16:06 <paper_> it's an old chroot from, I think, 3.10 updated to edge. Do you think creating a new 3.12 chroot would help?
2020-09-20 10:55:44 <hazz> hi there trying to configure iwd from command line
2020-09-20 10:56:03 <hazz> i got probs passing the passphrase
2020-09-20 10:56:17 <hazz> i got operation failed
2020-09-20 10:56:52 <hazz> any tips
2020-09-20 10:57:01 <mps> hazz: you don't have to type passphrase in cli, wait iwctl to ask for it and then type
2020-09-20 10:57:50 <hazz> i know but is a hexcode of ndigit
2020-09-20 10:58:22 <hazz> i tried severals time always the same problems
2020-09-20 10:59:06 <hazz> i can't install or work the gpm or libinput mouse on the terminal console
2020-09-20 10:59:44 <hazz> to do a paste and copy if it does work a P&C
2020-09-20 11:00:23 <mps> these pkgs are in testing repo, so you have to enable testing in /etc/apk/repositories
2020-09-20 11:01:38 <hazz> ok thanks i will try
2020-09-20 11:06:48 <hazz> mps: Is there a way to compile not in automatic form but giving a consens form?
2020-09-20 11:09:40 <mps> what you mean by consens or automatic form? I don't understand
2020-09-20 11:13:19 <hazz> i mean every passages that makes it will says confirm that you want to compile it or something similar
2020-09-20 11:15:08 <mps> ah, one by one: abuild checksum deps unpack prepare build package rootpkg index
2020-09-20 11:16:41 <hazz> wit a moment that i have to write it down
2020-09-20 11:17:04 <mps> abuild -h, will tell more
2020-09-20 11:26:35 <fridim> Hi o/   I still have issues with amdgpu here on the alpine. Same hardware and no problem with archlinux (different kernels).  Here is the dmesg log https://gist.github.com/fridim/690ccd0e52a7ada4f4f9f620f330e658     Not sure if it's because it's fixed in the other kernel, or if it's a configuration issue
2020-09-20 11:33:23 <hazz> sorry guys is it possible to find the records of this channel
2020-09-20 11:38:09 <mps> log?
2020-09-20 11:38:15 <hazz> yes
2020-09-20 11:38:16 <mps> irclogs.a.o
2020-09-20 11:38:27 <mps> irclogs.alpinelinux.org
2020-09-20 11:41:19 <hazz> thanks so much mps
2020-09-20 11:42:16 <mps> hazz: np :)
2020-09-20 11:53:44 <hazz> mps do you know if there is a soft to download those logs and use it out of the box
2020-09-20 11:58:10 <mps> hmm, curl or wget
2020-09-20 12:00:35 <hazz> mps tryng with wget
2020-09-20 12:00:49 <mps> w3m with some options to strip html
2020-09-20 12:08:24 <mps> `curl 'https://irclogs.alpinelinux.org/%23alpine-linux-2020-09.log' | less`
2020-09-20 12:24:33 <hazz> thank you i tried to mirror it no chance
2020-09-20 12:57:27 <sxiii> Heyhey guys :) Some update on the presentation project. I've managed to build the ISO with the files I need inside of the ISO
2020-09-20 12:58:00 <sxiii> However, they are in the ISO, but not installed/available by default on the Live ISO system. So after I boot up I can add them with "apk add" but I need them to be available on boot-time.
2020-09-20 12:58:06 <sxiii> What is my best way of doing it?
2020-09-20 12:59:14 <sxiii> Do I have to deal with APKOVL in a specific way or do the "lbu commit" commands before my image is build?
2020-09-20 13:03:24 <abenz> sxiii: not sure about the context of an ISO (on a CDROM?), but on my USB, I simply place an apkovl and it gets extracted onto the running system
2020-09-20 13:03:54 <abenz> you can place any file you want, including eg content in /home, obviously all the important stuff in /etc as well
2020-09-20 13:04:35 <sxiii> I'm building a live-ISO in Docker. I need several apps to auto-start when booting this ISO.
2020-09-20 13:05:01 <sxiii> So the apps are added to the package management base, but they ain't added to the actual "running" system when I start the ISO.
2020-09-20 13:05:14 <sxiii> Until I do "apk add sent" and then they all are added from the ISO...
2020-09-20 13:05:48 <abenz> yes they need to be listed in etc/apk/world for them to be auto installed
2020-09-20 13:06:30 <abenz> maybe someone else will be better equipped to answer, I only know about the usb context, good luck
2020-09-20 13:08:29 <sxiii> I'm not sure that ISO context differs in any way from the USB context if we're speaking about freshly boot system.
2020-09-20 13:08:49 <sxiii> I am rebuilding now the image with stuff inside etc/apk/world, let's see if it works right away :)
2020-09-20 13:10:01 <abenz> :)
2020-09-20 13:10:11 <abenz> you will need the cache of all those applications as well
2020-09-20 13:10:30 <abenz> eg /media/usb/cache
2020-09-20 13:10:51 <abenz> all the .apks you have listed in world should reside there
2020-09-20 13:11:32 <abenz> it would be far simpler if you booted into a live system, say with a usb, then ran setup-lbu, setup-apkcache ...etc
2020-09-20 13:12:04 <abenz> once you have you system configured just as you like, then generate a fresh apkovl and adapt to CD use
2020-09-20 13:13:54 <abenz> I imagine you will only need to update /etc/lbu/lbu.conf, /etc/apk/repositories , /etc/fstab
2020-09-20 13:14:20 <abenz> ah, and the thing that caused me headache, you need the symlink in /etc/apk/cache to point to your real cache location
2020-09-20 13:15:01 <abenz> mine is /media/usb/cache, your case would be /media/cdrom/cache I guess
2020-09-20 13:15:58 <abenz> the above I didn't realize initially, and would have the unpack process zip through quickly as the system was unable to find cache (was pointed to /media/sdxx/cache, which changes every other reboot)
2020-09-20 13:24:56 <sxiii> abenz: I've added sent to /etc/apk/world and it works!
2020-09-20 13:25:26 <sxiii> My apps does not require cache as they are intended to be like fresh ones/presentation/kiosk mode (it's sent, xpdf and maybe firefox)
2020-09-20 13:26:15 <sxiii> Now. The two last things I have to solve in my iso: auto-login (it asks for root username strangely but logs-in without password) and, lastly, the xserver
2020-09-20 13:27:15 <sxiii> abenz: I have to way of running anything after I am booting the ISO. It has to boot and autorun the presentaton without any user interference (that's the idea of the project)
2020-09-20 13:27:19 <mps> sxiii: util-linux have aggety which have autologin option
2020-09-20 13:27:52 <sxiii> ok so I add util-linux package now, do I have to configure agetty additionally so it works right after the iso bootup?
2020-09-20 13:28:06 <abenz> what do you plan to use for WM?
2020-09-20 13:28:31 <abenz> if you eg use openbox (simple and light), you can place whatever you want to autostart in ~/.config/openbox/autostart
2020-09-20 13:28:34 <mps> but you can also run X display manager with autologin enabled
2020-09-20 13:28:34 <sxiii> abenz: I'm not going to use CD-ROM, nei. I'm going to use ISO to burn to flash drive in the future.
2020-09-20 13:29:10 <sxiii> abenz: I actually would go without WM, if possible. I need only 3 fullscreen apps to work (each time only one of them)
2020-09-20 13:29:28 <abenz> so plain X? no borders or anything?
2020-09-20 13:29:33 <sxiii> The only problem is I will need a high resolutions and multi-monitor support (as I said this ISO will be for presentations)
2020-09-20 13:29:55 <sxiii> no; no borders; anything. I don't need them. Just the 100% of the viewport should take the app itself
2020-09-20 13:30:38 <sxiii> I was using Openbox in my recent NixOS build (I'm doing my tests with both Alpine and NixOS); but I was rather stick NOT to use any WM at all, if possible.
2020-09-20 13:30:51 <sxiii> The only quest is setting resolution for two screens at max...
2020-09-20 13:30:56 <abenz> with firefox that would be F11, but other apps?
2020-09-20 13:31:19 <sxiii> sent is already a fullscreen app. xpdf has the fullscreen flag
2020-09-20 13:31:48 <abenz> firefox too actually
2020-09-20 13:31:51 <abenz> firefox -kiosk
2020-09-20 13:31:59 <abenz> will lanuch it into full screen
2020-09-20 13:32:03 <sxiii> yes. I used this flag in NixOS build and it works well. but I use openbox there.
2020-09-20 13:32:35 <sxiii> I'm not sure if having no WM will work okay as at some point in the past, when I tried, I got "not-so-fullscreen" windows running when I was without any WM
2020-09-20 13:32:59 <sxiii> I think the windows was like 800x600 when the screen was 1280x1024 or something like that
2020-09-20 13:33:55 <sxiii> mps: what's the easier way to do the autologin: util-linux + agetty + adding my own .xinitrc or "run X display manager with autologin enabled" can you elaborate a little?
2020-09-20 13:34:55 <mps> if you want to go to X, then simply rc-update add XDM (whichever) and you don't need console autologin
2020-09-20 13:35:19 <mps> I have slim configured to autologin me on start
2020-09-20 13:36:06 <mps> and I put all commands I need in $HOME/.xinitrc
2020-09-20 13:44:41 <sxiii> yeah so I should try adding slim and configure it, I think it's the best bet as I need X anyways
2020-09-20 13:50:43 <mps> you can also enable autologin on console in case you need it
2020-09-20 13:51:44 <mps> that is, both, though slim will start X automatically, console could be 'emergency' backup
2020-09-20 13:59:53 <sxiii> yeah but the thing is if I got access to emergency console I can type in the user without any problems :)
2020-09-20 14:00:13 <sxiii> But the main idea is of course have only X fullscreen app launch after ISO boots up
2020-09-20 14:49:24 <q66> Ariadne: oh, this change, alright, thanks
2020-09-20 14:49:54 <Ariadne> q66: are you on the discord ?  do you want to be on the discord if not ?
2020-09-20 14:50:43 <q66> i am not, i usually prefer not to do discord but i have it anyway because of some friends so i guess i might as well
2020-09-20 14:53:03 <q66> feel free to give a link, might pop in in a bit
2020-09-20 17:38:42 <nahum> Hey all, on the alpine linux wiki it seems to suggest that contributions need to be in the form of full articles, but a lot of the documentation might be able to be a bit more clearly written. Would it be acceptable to add points of clarity and expand on things within the wiki's pages as a contribution?
2020-09-20 17:39:13 <ikke> Should be fine
2020-09-20 17:39:16 <ikke> that's just about creating new pages
2020-09-20 18:07:02 <abenz> ikke: was wondering how long it takes the buildbots to generate the new binaries..
2020-09-20 18:07:10 <abenz> I'm using this mirror btw dl-cdn.alpinelinux.org/alpine/v3.12/community...
2020-09-20 18:07:19 <abenz> still no update for mlt
2020-09-20 18:07:45 <abenz> not rushing, just wondering when to expect it :)
2020-09-20 18:09:21 <ikke> It should have already been there
2020-09-20 18:10:14 <ikke> maybe I forgot to merge
2020-09-20 18:21:07 <ikke> abenz: oh, on 3.12 stable, the pkgrel bump was accidentaly ommitted
2020-09-20 18:22:22 <ikke> abenz: it should be there in about 30 minutes
2020-09-20 18:33:05 <abenz> ikke: cheers!
2020-09-20 18:48:01 <abenz> at last, kdenlive runs natively!
2020-09-20 18:51:04 <abenz> only thing left is for PureTryOut[m] to bump to 20.08.1 and I'd be all set, because now I get..
2020-09-20 18:51:07 <abenz> This project type is unsupported (version 1) and cannot be loaded.
2020-09-20 18:51:08 <abenz> Please consider upgrading your Kdenlive version.
2020-09-20 18:53:20 <abenz> or I could bump my system to edge, does that sound like a reasonable thing on a production system?
2020-09-20 18:53:45 <abenz> desktop user that is, not a server
2020-09-20 18:55:05 <c705> if it's truely production, I wouldn't be running edge
2020-09-20 18:55:38 <c705> then again, I run edge on my 'production" desktop machine and it works fine for me. Submit a MR on the apk package to bump the version and the Pm will assist
2020-09-20 19:09:27 <abenz> c705: I'm too new to alpine for a MR, but from the logs I see its actually just a simple version bump for kdenlive
2020-09-20 19:10:24 <abenz> but yes, you have a point its about time I get familiar with abuild and bump packages locally if needed
2020-09-20 19:25:54 <c705> clone the repo, fork it, make the change, submit it back as a mr
2020-09-20 19:26:05 <c705> or ping the package maintainer here and get them to do it
2020-09-20 19:33:24 <PureTryOut[m]> abenz: I didn't realize a 20.08.1 was out yet lol, will do tomorrow, thanks for the notify!
2020-09-20 20:33:46 <bbuccianti> to which package belongs the ascii man page?
2020-09-20 20:34:56 <c705> bbuccianti: possibly `asciidoc-doc`
2020-09-20 20:36:17 <bbuccianti> nope that's asciidoc package
2020-09-20 20:36:22 <bbuccianti> I found man-pages
2020-09-20 20:36:25 <bbuccianti> thanks!
2020-09-20 20:50:23 <Hello71> bbuccianti: https://pkgs.alpinelinux.org/contents?file=ascii.*&path=%2Fusr%2Fshare%2Fman%2F*&branch=edge
2020-09-20 21:40:08 <atkka> hello again, can anyone point me in the right direction to get audio working with flatpaks?
2020-09-20 21:41:12 <Cogitri> Just works for me with pa
2020-09-20 21:42:13 <atkka> hmm, thank you. did you do anything special, are you running a DE?
2020-09-20 21:42:36 <atkka> I have dbus and pulse audio, but no sound from firefox or other flatpaks
2020-09-20 21:42:37 <Cogitri> I use GNOME, I didn't do anything
2020-09-20 21:43:29 <atkka> ok, I wonder if some env stuff needs sorted, I'm not running a DE, just xorg and dwm, thanks for the info
2020-09-20 21:47:33 <Cogitri> Do you use pulseaudio?
2020-09-20 21:48:16 <atkka> not usually but I have it insalled now and it works for native apps just not flatpaks
2020-09-20 22:02:40 <charl> I need help understanding how loadkeys/keymaps are supposed to work. Lets say I `loadkeys gr` in console how am I supposed to switch back to default/us keymap?
2020-09-20 22:23:44 <atkka> Cogitri: can you see if pulseaudio is exporting its socket to /tmp
2020-09-20 22:24:30 <atkka> I suspect this is the issue, I don't think the firefox flatpak can mount /tmp so it can't access pulse
2020-09-20 22:25:58 <c705> your user also has to be in the `audio` group i think
2020-09-20 22:26:20 <atkka> c705: I am, native application audio works just not flatpaks
2020-09-20 22:26:54 <c705> doesn't flatpak apps exec with its own user
2020-09-20 22:27:34 <c705> or is it just seccomp?
2020-09-20 22:27:56 <atkka> I'm not sure, I'm new to flatpaks myself
2020-09-20 22:28:36 <charl> it does
2020-09-20 22:28:39 <charl> --with-system-helper-user=flatpak
2020-09-20 22:29:19 <charl> try adding flatpak user to audio group
2020-09-20 22:29:29 <c705> ^
2020-09-20 22:30:09 <atkka> ok one sec
2020-09-20 22:36:34 <atkka> hmm, no luck, user flatpak is a member of the audio group
2020-09-20 22:37:07 <c705> I guess you'd have to pass the pulse socket into the container as well, but no idea how
2020-09-20 22:37:31 <atkka> flatpak messages indicate [OnMediaSinkAudioError]
2020-09-20 22:37:44 <atkka> but not flatpak firefox works fine
2020-09-20 22:38:05 <atkka> yeah, I'm wondering why if you have a DE installed it seems to work
2020-09-20 22:39:27 <c705> dbus maybe
2020-09-20 22:39:31 <charl> you don't have a DE?
2020-09-20 22:39:33 <c705> pretty sure pulse uses dbus
2020-09-20 22:40:15 <atkka> I have dbus installed and service running, flatpak was bitching about it until I started the service so I know that's working now
2020-09-20 22:40:24 <charl> do you have eudev?
2020-09-20 22:41:10 <atkka> I have udev and whatever setup-xorg-minimal installed
2020-09-20 22:41:10 <u0jQx9gPyrYg> have you tried stracing it?
2020-09-20 22:42:28 <atkka> setup-xorg-base I mean
2020-09-20 22:42:38 <atkka> u0jQx9gPyrYg: I haven't straced it
2020-09-20 22:45:38 <atkka> anyway, I'm running xorg-base, pulseaudio, dbus, dbus-x11, and dwm. Native firefox and applications both pulse/alsa work fine. Flatpaks run fine but have no sound, I have tried firefox and wire messenger flatpaks
2020-09-20 22:51:28 <atkka> anyway, thanks for the help so far, I am going to take a break for an hour or two but will be back if you find anything! thansk again!
2020-09-20 23:06:49 <MY-R> atkka, did you setup XDG_RUNTIME_DIR?
2020-09-20 23:08:03 <charl> what happens to alpine if ncopa leaves or God Forbid something happens to him?
2020-09-20 23:08:21 <charl> 3k+ maintained packages.. I see his name everywhere
2020-09-20 23:08:43 <MY-R> create "/etc/profile.d/xdg_runtime_dir.sh" and paste there script from here: https://wiki.alpinelinux.org/wiki/Wayland (doesnt matter that not using wayland)
2020-09-20 23:08:50 <MY-R> atkka ^
2020-09-20 23:18:23 <Hello71> it should be in /run?
2020-09-20 23:19:10 <Hello71> er
2020-09-21 00:51:20 <atkka> MY-R: thanks for the tip, unfortunately it didn't work
2020-09-21 00:53:03 <atkka> may it have something to do with installing flatpaks via sudo? every example I've seen doesn't install as sudo/root however flatpak install *whatever* fails without sudo: no permissions
2020-09-21 01:04:19 <charl> are you running with `flatpak --user`?
2020-09-21 01:04:53 <charl> you can install or run as user with `--user`
2020-09-21 01:11:29 <charl> atkka ^
2020-09-21 03:35:57 <atkka> figured out flatpak audio, had to export XDG_RUNTIME_DIR=/tmp/user/$(id -u) with a script
2020-09-21 03:36:33 <atkka> to start a flatpak with pulseaudio access I have to: pulseaudio --start && flatpak run *app*
2020-09-21 03:38:12 <atkka> thanks for the help everyone
2020-09-21 04:56:32 <yjftsjthsd> Speaking as someone who just hit that exact problem today (Steam works, but games without sound are... worse off), it would be really helpful if that got documented (probably at https://wiki.alpinelinux.org/wiki/Flatpak), or if we could somehow rig up the relevant packages to make it work out of the box.
2020-09-21 04:57:55 <yjftsjthsd> In the interest of not wandering in and saying "someone should do this work" and then leaving, I'm going to add it to my personal task list in the event that nobody else gets there first, although it would be nice if atkka could do it since they actually did it and have the exact steps already fresh in their mind.
2020-09-21 04:58:18 <abenz_> c705: thanks for the tip, will get up to speed on weekend
2020-09-21 04:59:01 <abenz_> PureTryOut[m]: great, thanks :)
2020-09-21 05:07:07 <winny> a fun gotcha... turns out if you follow the wiki instructions for zfs, and change your disk enumeration, it will likely break and degrade pool.  the fix is likely to use eudev and refer to disks by uuids.
2020-09-21 05:08:10 <winny> btw is it possible to run 'zpool status' as non-root?
2020-09-21 05:15:49 <yjftsjthsd> winny I haven't actually tried it, but I'm pretty sure it's just down to permissions on /dev/zfs?
2020-09-21 05:16:25 <yjftsjthsd> Failing that, there's always doas/sudo with nopasswd rules
2020-09-21 05:20:59 <atkka> yjftsjthsd: I put it on my list, I think I have a wiki account already
2020-09-21 05:22:18 <atkka> yjftsjthsd: I would like to do some more testing, a fresh install with the minimal number of packages/steps needed and with more flatpaks like steam. I have tested with firefox and wire and they work flawlessly, mic/speakers/headphones
2020-09-21 05:25:04 <atkka> I can do most of that tomorrow and make a rough draft for the wiki
2020-09-21 05:25:09 <yjftsjthsd> Excellent, thank you:) Since I literally hit that today and haven't actually fixed it on my box yet, I would be happy to be a tester once you get it up:)
2020-09-21 05:25:50 <atkka> yjftsjthsd: do you have bluetooth set up to test?
2020-09-21 05:25:57 <yjftsjthsd> I do not
2020-09-21 05:26:01 <atkka> ok
2020-09-21 05:26:16 <atkka> that's one thing I've never gotten working on alpine
2020-09-21 05:26:32 <atkka> pulse plus bluez or whatever you need these days
2020-09-21 05:27:05 <yjftsjthsd> TBH, while I have used BT on "prepackaged" distros (ubuntu, fedora, etc.), I generally view it as "never going to be worth the pain".
2020-09-21 05:27:27 <yjftsjthsd> (And really, BT kinda sucks in general, IMO.)
2020-09-21 05:27:36 <atkka> yeah, I know what you mean, I think pulse is a hard depends now anyway
2020-09-21 05:29:03 <atkka> I'm glad that I got flatpak working though, can have firefox-esr native and the flatpakff for the latest and greatest testing
2020-09-21 05:32:04 <yjftsjthsd> Yeah, I have mixed feelings about FP overall, but it definitely solves some real problems quite well:)
2020-09-21 05:33:46 <atkka> on musl systems for sure
2020-09-21 05:36:24 <yjftsjthsd> Quite:) It was either this, or using full-on containers (docker/podman), which isn't that different but more DIY work
2020-09-21 05:37:30 <atkka> yeah I was going to go the chroot route but found that flatpaks could solve that problem pretty easily
2020-09-21 05:39:02 <atkka> do you know if export is a bash only builtin?
2020-09-21 05:39:13 <yjftsjthsd> No, pretty sure it's POSIX
2020-09-21 05:41:22 <yjftsjthsd> Works in dash and ash
2020-09-21 05:41:45 <atkka> hmm I'm talking about the shell itself, it looks like it is a bash shell builtin, my ash shell doesn't seem to have it
2020-09-21 05:42:25 <atkka> ok nvm, it just doesn't tab complete but you can type it
2020-09-21 05:42:38 <yjftsjthsd> Ah, yeah, ash has different controls
2020-09-21 05:42:44 <yjftsjthsd> (huge pain sometimes)
2020-09-21 05:44:03 <atkka> well if you want a quick rundown on the flatpak audio issue its pretty simple
2020-09-21 05:46:39 <atkka> just export XDG_RUNTIME_DIR=/tmp/usr/$(id -u)
2020-09-21 05:47:12 <atkka> then launch your application with pulseaudio --start
2020-09-21 05:47:48 <atkka> pulseaudio --start && flatpak run org.mozilla.firefox
2020-09-21 05:48:19 <atkka> for example
2020-09-21 05:51:48 <atkka> you may need to create that directory (/tmp/usr/) but I think its auto generated, you can automate everything with a script and alias
2020-09-21 07:31:10 <Milka> Greetings. I'm trying to get Alpine working on btrfs root. So far, the boot process stops in emergency shell with "/sbin/btrfs not found" and "Failed to scan for btrfs devices" errors. Ofc after 'exit' kernel panics. Is there any way to fix this?
2020-09-21 07:32:18 <Cogitri> Install btrfs on your host system, make sure the btrfs feature is in your mkinitfs.conf and regenerate the initramfs
2020-09-21 07:32:21 <Cogitri> then it should work
2020-09-21 07:34:37 <Milka> Thanks. How do I get initramfs regenerated manually? Initially I got it automatically through kernel installation hook.
2020-09-21 07:41:09 <gjabell> mkinitfs should do it
2020-09-21 07:42:08 <Milka> Thanks
2020-09-21 08:25:26 <maerwald> I'm recently getting linker errors on edge: https://gitlab.haskell.org/haskell/ghcup-hs/-/jobs/447862#L1570
2020-09-21 08:25:39 <maerwald> /usr/lib/gcc/x86_64-alpine-linux-musl/10.2.0/../../../../x86_64-alpine-linux-musl/bin/ld: /usr/lib/gcc/x86_64-alpine-linux-musl/10.2.0/crtend.o: relocation R_X86_64_32 against `.ctors' can not be used when making a shared object; recompile with -fPIC
2020-09-21 08:34:58 <Cogitri> I guess the upgrade to GCC10.x may cause that
2020-09-21 08:40:44 <maerwald> how do I fix this
2020-09-21 08:43:31 <Milka> Cogitri: I've managed to get btrfs into initramfs, but now there is no device nodes for any disks in there, so root mounting fails. What did I miss?
2020-09-21 08:44:49 <Milka> blkid and fdisk -l also show nothing
2020-09-21 08:46:00 <gillesmauve> Install gentoo
2020-09-21 08:47:59 <gillesmauve> maerwald: unrelated to solving your issue, will you tell me why you're on edge?
2020-09-21 08:49:36 <maerwald> gillesmauve: no particular reason
2020-09-21 08:49:49 <maerwald> also, gentoo?
2020-09-21 08:49:59 <gillesmauve> It's a meme.
2020-09-21 08:50:14 <maerwald> Well, I don't know that meme then.
2020-09-21 08:50:22 <gillesmauve> No particular reason? What events led to you changing the repo over to edge?
2020-09-21 08:50:26 <gillesmauve> You do now.
2020-09-21 08:50:41 <maerwald> I can't really follow why it's a meme, though
2020-09-21 08:50:58 <gillesmauve> There are ways to pin packages to specific versions, but that's more likely to blow up on you when on edge.
2020-09-21 08:51:00 <maerwald> gillesmauve: bad GHC support
2020-09-21 08:51:16 <gillesmauve> Gnu Hadron Collider, I assume
2020-09-21 08:51:25 <maerwald> glasgow haskell compiler
2020-09-21 08:51:35 <gillesmauve> Figures they'd get around to building one of their own
2020-09-21 08:51:45 <maerwald> which I fixed myself, so I might be able to swithch to a different branch
2020-09-21 08:51:58 <gillesmauve> Ah, nice work.
2020-09-21 08:52:15 <gillesmauve> How'd you get anything to come out of a haskell compiler? Sounds like a side effect.
2020-09-21 08:52:48 <gillesmauve> In the future, it should be apparent that anyone making jokes is too stupid to help you solve the problem outright, by the way. That's the oldest meme in the book.
2020-09-21 08:53:02 <maerwald> glad you're helping me then
2020-09-21 09:15:31 <Cogitri> Milka: You set root and rootfstype in your bootflags?
2020-09-21 09:20:55 <Milka> Cogitri: yes, I specified root=UUID=xxxx... and rootfstype=btrfs
2020-09-21 09:21:42 <mps> reading kernel commit log messages, I wouldn't use btrfs for anything serious. maybe only to play with it
2020-09-21 09:23:45 <gillesmauve> finally on a normal machine, maerwald: have you already fixed your problem?
2020-09-21 09:24:16 <maerwald> no
2020-09-21 09:24:26 <maerwald> I was installing gentoo since 2 hours
2020-09-21 09:24:36 <gillesmauve> i'd just set up a virtual machine with the devtools you need on it temporarily until the package is fixed; i don't want to recommend the package pinning method
2020-09-21 09:24:41 <u0jQx9gPyrYg> gillesmauve:zomg you are funny, but not helpful at all
2020-09-21 09:24:48 <gillesmauve> u0jQx9gPyrYg: that's what i was saying.
2020-09-21 09:25:32 <Milka> mps: but the thing is, there are no disk device nodes in /dev and blkid shows nothing, and I've got xfs for /home, /tmp and /var, so I don't think   it 's btrfs specific issue, I guess I just missed smth
2020-09-21 09:25:33 <u0jQx9gPyrYg> yeah, but poor maerwald was hopeful that you would be
2020-09-21 09:26:22 <mps> Milka: I'm not speaking about your issue, but in general about btrfs
2020-09-21 09:27:07 <gillesmauve> u0jQx9gPyrYg: i've given at least two alternatives
2020-09-21 09:27:21 <mps> Milka: after you make new initramfs you can unpack it somewhere and check if btrfs module is actully there
2020-09-21 09:28:31 <mps> and also is the /sbin/btrfs in in initramfs
2020-09-21 09:29:35 <gillesmauve> what's milka doing?
2020-09-21 09:30:15 <mps> also add 'loglevel=7' (or above) and remove quiet from kernel cmdline to see boot messages on console
2020-09-21 09:36:04 <Milka> mps: I've checked for /sbin/btrfs from the emergency shell, and also there are no btrfs related errors displayed any more. It just can't find the root.
2020-09-21 09:36:53 <mps> stupid question, but did you made btrfs rootfs?
2020-09-21 09:37:43 <Cogitri> Milka: Do you happen to use nvme as your root disk?
2020-09-21 09:38:32 <Milka> yes, nvme0n1
2020-09-21 09:39:14 <Cogitri> And you've added nvme to your mkinitfs features?
2020-09-21 09:40:16 <Milka> Oops I guess no, gonna check it
2020-09-21 09:40:58 <Cogitri> Yeah, was surprising to me at first too that one would need to explicitly enable support for it
2020-09-21 09:41:39 <MY-R> I think many people gave up because of mising "nvme" in mkinitfs
2020-09-21 09:42:26 <mps> or mmc :)
2020-09-21 09:42:30 <MY-R> ye
2020-09-21 09:43:01 <MY-R> mkinitfs could be little bit more smarter :\
2020-09-21 09:47:18 <Cogitri> Yup
2020-09-21 09:47:31 <Cogitri> What's also annoying is that mkinitfs doesn't abort on missing files
2020-09-21 09:47:48 <Cogitri> So if you have the btrfs feature enabled but don't have btrfs-progs installed mkinitfs won't warn you or error out
2020-09-21 09:48:15 <Cogitri> So you'll only notice that things are broken when you try to start your system and the initramfs can't find the btrfs executable
2020-09-21 09:51:41 <MY-R> the price of simplicity
2020-09-21 09:52:05 <mps> MY-R: no, bonus
2020-09-21 09:52:17 <gillesmauve> the boners of simplicity
2020-09-21 09:52:36 <charl> spitting warnings is bloat
2020-09-21 09:54:00 <charl> does anyone here use a keymap other than us? which one? I want to test something
2020-09-21 09:54:27 <mps> charl: X or console
2020-09-21 09:54:32 <charl> console
2020-09-21 09:54:43 <mps> I don't, sorry
2020-09-21 09:55:44 <MY-R> charl, use: pl
2020-09-21 09:57:25 <charl> MY-R does pl keymap have a way of switching to and from pl/us language?
2020-09-21 09:57:56 <MY-R> to check if working push right alt + zxcase which should give you with correct fonts that: żźćąśę
2020-09-21 09:58:10 <MY-R> I have no idea
2020-09-21 09:59:07 <charl> my problem is if I set gr keymap I start typing in greek and I have no way of switching to us
2020-09-21 09:59:20 <charl> so I can't even login
2020-09-21 09:59:54 <Milka> Cogitri, mps, thanks, it booted! I've got a noob question tho: I used my /etc/fstab as fstab for mkinitfs. Should I do this usually?
2020-09-21 10:00:12 <Cogitri> Nice that it boots now :)
2020-09-21 10:00:50 <MY-R> charl, isnt better to just set up 'us' as default and after login switch to 'gr' ?
2020-09-21 10:01:17 <Cogitri> I don't think I use a fstab for my mkinitfs? I only mount / via root= in the initramfs, the rest happens later on via openrc
2020-09-21 10:03:42 <MY-R> charl, or dunno how you login in and out then could probably set it up somewhere to bring back us
2020-09-21 10:04:41 <charl> MY-R the point is after changing to greek with loadkeys or loadkmap there's nothing I can do on the shell
2020-09-21 10:05:00 <charl> because everything i type is in greek
2020-09-21 10:05:23 <charl> how it is supposed to work is: I alt shift and switch from and to us/gr
2020-09-21 10:06:00 <charl> https://git.kernel.org/pub/scm/linux/kernel/git/legion/kbd.git/tree/data/keymaps/i386/qwerty/gr.map
2020-09-21 10:06:52 <abenz> I installed abobe-fonts-source-code-pro as I read online its good for terminal, but I cant find it, do I need to do a step/regenerate a cache/index?
2020-09-21 10:07:14 <abenz> also if anyone has suggestions for open/free fonts suitable for terminal
2020-09-21 10:08:11 <abenz> maximum readability is my goal, not aesthetics
2020-09-21 10:08:34 <mps> Milka: as Cogitri wrote, initramfs set / from root=whatever kernel cmdline, so no need to put in in /etc/fstab
2020-09-21 10:09:14 <gillesmauve> abenz: what do you think of Input? https://djr.com/input/
2020-09-21 10:09:27 <gillesmauve> (paired with output: https://djr.com/output/ )
2020-09-21 10:09:39 <charl> abenz how did you install it? apk or manual?
2020-09-21 10:09:58 <charl> if apk I think it does fc-cache by itself
2020-09-21 10:10:09 <charl> fc-cache -f
2020-09-21 10:11:12 <charl> as for fonts I prefer terminus-font or spleen from openbsd
2020-09-21 10:11:59 <abenz> gillesmauve: I see "Buy it now" in the links you provided, with all the free stuff available (and my intent to use in terminal) I'm leaning towards open/free stuff
2020-09-21 10:12:12 <abenz> charl: yes with apk, I cant find it in lxterminal's list
2020-09-21 10:12:20 <gillesmauve> to my understanding, it's completely free for normal use
2020-09-21 10:12:30 <abenz> I just ran fc-cache -f again and I still cant find it, let me close all terminals instances and see
2020-09-21 10:13:35 <gillesmauve> https://input.fontbureau.com/download/ private use, etc etc
2020-09-21 10:14:15 <abenz> I see them now after I closed all terminal instances and reopened
2020-09-21 10:15:31 <abenz> gillesmauve: thanks
2020-09-21 10:15:40 <gillesmauve> aye, comrade
2020-09-21 10:15:42 <gillesmauve> go and code
2020-09-21 10:15:45 <abenz> I will test a few and see
2020-09-21 10:15:48 <gillesmauve> try not to make anyone rich
2020-09-21 10:27:46 <charl> I think I found the source of my problem: https://git.alpinelinux.org/aports/tree/main/kbd/APKBUILD
2020-09-21 10:28:38 <charl> for some reason ncopa uses ckbcomp to convert keymaps and then uses those to generate binary keymaps
2020-09-21 10:28:57 <charl> and those binary keymaps are then used by setup-keymap/loadkmap
2020-09-21 10:30:36 <charl> if I loadkeys the "legacy" maps directly, greek works fine and I can switch to en/gr with alt+shift
2020-09-21 11:34:17 <Stacy[m]> how to save boot options in alpine permanently?
2020-09-21 11:34:29 <Stacy[m]> the system is installed to a hdd in sys mode
2020-09-21 11:46:07 <mps> charl: your question about ncopa (BDFL), there are a lot of pkgs which have maintainer field but practically not maintained by official maintainer but with other people
2020-09-21 11:47:11 <mps> actually, maintainer field is not important as in some other linux/unix distros
2020-09-21 11:47:51 <mps> Stacy[m]: you mean boot loaders options/configs?
2020-09-21 11:48:02 <ncopa> charl: im open to ideas how to fix it. please file an issue and explain the details
2020-09-21 11:48:21 <Stacy[m]> <mps "Stacy: you mean boot loaders opt"> Yes
2020-09-21 11:49:40 <mps> Stacy[m]: you update grub or syslinux config files, that's all
2020-09-21 11:50:23 <mps> ncopa: give maintainership of all of them to charl :)
2020-09-21 11:51:18 <charl> ncopa: will do, thanks
2020-09-21 11:51:23 <charl> mps: lmao
2020-09-21 11:51:28 <Stacy[m]> How do I do it on alpine
2020-09-21 11:51:57 <mps> Stacy[m]: what are you using, grub or syslinux
2020-09-21 11:52:06 <Stacy[m]> I know what file to edit and what command to use in debian, but not in alpine
2020-09-21 11:52:23 <Stacy[m]> Probably syslinux
2020-09-21 11:52:27 <charl> mps: so basically committers do all the work but still.. doesn't the maintainer have to approve?
2020-09-21 11:53:45 <mps> Stacy[m]: /etc/update-extlinux.conf
2020-09-21 11:54:58 <mps> charl: well, still maintainers do a lot of work but if maintainer is not active or responds than anyone can make changes and create merge request
2020-09-21 11:55:18 <Stacy[m]> And after editing?
2020-09-21 11:55:27 <charl> I see
2020-09-21 11:55:34 <mps> Stacy[m]: and run update-extlinux
2020-09-21 11:56:18 <mps> for grub, files are in /etc/grub.d/
2020-09-21 11:56:50 <Stacy[m]> Okay then I guess it is it
2020-09-21 11:58:36 <mps> charl: alpine is not so rigid in development process
2020-09-21 12:00:12 <mps> but yes, maintainers who are active are important but still they usually are not final authorities. some kind of consensus and good 'will' usually prevail
2020-09-21 12:00:48 <mps> and ofc, sometimes conflicts happens but not often and not serious
2020-09-21 12:02:11 <shhhum[m]> mps: Can libudev-zero with mdev completely replace eudev?
2020-09-21 12:03:29 <mps> shhhum[m]: well, not completely I think
2020-09-21 12:03:35 <Stacy[m]> yes, thanks it worked
2020-09-21 12:04:01 <mps> some things need to re/write rules from udev to mdev
2020-09-21 12:05:00 <mps> for example I had to write hackish script for backlight driver which worked out-of-the-box with udev
2020-09-21 12:05:15 <mps> but most things works
2020-09-21 12:05:40 <mps> plug/unplug works, that is most important :)
2020-09-21 12:07:07 <shhhum[m]> Usev rules is just shell scripts or..?
2020-09-21 12:07:13 <shhhum[m]> Udev*
2020-09-21 12:07:33 <mps> shhhum[m]: I'm using it for about two weeks on two my machines, one is my main workstation and second is small SBC which do some hardware/device controls
2020-09-21 12:08:13 <mps> shhhum[m]: no, udev rules are well, rules for udev daemon how to handle devices
2020-09-21 12:08:26 <shhhum[m]> Gotcha.
2020-09-21 12:08:43 <mps> they are more declarative configs
2020-09-21 12:08:52 <mps> more like*
2020-09-21 12:09:22 <mps> and there are udev helpers written in shell scripts
2020-09-21 12:09:41 <mps> also mdev helpers are usually shell scripts
2020-09-21 12:13:30 <shhhum[m]> Okay, thank you! So I guess we can not completely replace eudev in Alpine and I should just close my issue in bugtracker, right?
2020-09-21 12:13:34 <shhhum[m]> https://gitlab.alpinelinux.org/alpine/aports/-/issues/11965
2020-09-21 12:19:15 <mps> shhhum[m]: you can close it if you want, but also you can keep it open for some time, maybe someone have to add comments about that
2020-09-21 12:20:47 <mps> some people expressed already interest in it, to test it and some even to improve it, and some to add configs and/or scripts
2020-09-21 12:21:02 <mps> also I plan to improve it over time
2020-09-21 12:25:12 <shhhum[m]> Okay, I leave it open then. When I get back home, I'm also planing to install libudev-zero and test it, hope everything goes well.
2020-09-21 12:26:36 <mps> just don't forget to add hwdrivers and mdev to sysinit runlevel, then it should work
2020-09-21 12:26:40 <Milka> shhhum[m]: What happens now if you 'apk upgrade -U -a' with mdev and libudev-zero instead of eudev?
2020-09-21 12:27:40 <Milka> you've mentioned it in your issue
2020-09-21 12:30:12 <Hello71> mps: no, for grub you should edit /etc/default/grub
2020-09-21 12:30:16 <mps> good idea is to remove udev-init-scripts-openrc and eudev-openrc when installing libudev-zero
2020-09-21 12:30:37 <mps> Hello71: right, I forgot where it is
2020-09-21 12:30:45 <mps> I don't use grub much
2020-09-21 12:31:03 <shhhum[m]> Thing is libudev-zero do not replace eudev, which is installed on most systems. So one should `apk del eudev` and then `apk add libudev-zero` manually.
2020-09-21 12:31:04 <Hello71> /etc/grub.d scripts source /etc/default/grub
2020-09-21 12:31:25 <mps> shhhum[m]: yes
2020-09-21 12:31:26 <shhhum[m]> mps: Okay, thank you.
2020-09-21 12:31:50 <mps> that is now intentional, to not make mess with users system
2020-09-21 12:32:43 <mps> one have forcefully and knowingly replave eudev with libudev-zero, not blindly
2020-09-21 12:32:56 <mps> have to*
2020-09-21 12:33:12 <shhhum[m]> Oh, makes sense.
2020-09-21 12:35:05 <mps> Hello71: I think /etc/default/grub should go in /etc/conf.d
2020-09-21 12:35:19 <mps> but now it is as it is copied from debian
2020-09-21 12:35:33 <Hello71> not exactly
2020-09-21 12:36:07 <mps> well, its place in /etc
2020-09-21 12:36:15 <Hello71> ok
2020-09-21 12:36:32 <Hello71> but /etc/conf.d is for openrc init scripts
2020-09-21 12:36:55 <Hello71> really it should be /etc/grub-mkconfig.conf
2020-09-21 12:37:19 <mps> well, I agree
2020-09-21 12:37:38 <mps> something like that
2020-09-21 12:37:52 <Hello71> /etc/conf.d shouldn't be generic place for config files, that's already what /etc means
2020-09-21 12:38:14 <mps> yes, yes ...
2020-09-21 12:38:42 <Hello71> actually /etc/conf.d is misnamed too, it should be /etc/init.d/conf or /etc/initconf or something :p
2020-09-21 12:38:45 <Hello71> but probably too late now
2020-09-21 12:39:58 <mps> hmm, /etc/openrc.conf.d ?
2020-09-21 12:41:12 <mps> yes, I feel it will stay for long time as it is now
2020-09-21 12:49:50 <charl> ncopa: done https://gitlab.alpinelinux.org/alpine/aports/-/issues/11966
2020-09-21 12:50:29 <charl> I wonder if I should file a bug report in busybox too because I'm unable to create a valid keymap using dumpkmap/loadkmap
2020-09-21 12:51:24 <charl> for now loadkeys + kbd-legacy works
2020-09-21 13:38:33 <charl> can't confirm email in alpine wiki
2020-09-21 13:38:36 <charl> Alpine Linux could not send your confirmation mail. Please check your email address for invalid characters.
2020-09-21 18:43:14 <c705> are security fixes backported to the linux-lts package in v3.12? how does that work?
2020-09-21 18:47:02 <mps> c705: by updating linux-lts (though from time to time, and not whenever it is released)
2020-09-21 20:57:47 <shhhum[m]> mps: Hmm, I am getting a lot of `sh: can't create /tmp/.libudev-zero/uevent.#####: nonexistent directory `. Should I also add helper.sh from libudev-zero hotplugging section?
2020-09-21 20:58:12 <shhhum[m]> Besides `mdev.conf`.
2020-09-21 20:58:48 <mps> shhhum[m]: no, it should work out of the box
2020-09-21 20:59:28 <mps> but maybe your kernel doesn't have /proc/sys/kernel/hotplug ?
2020-09-21 21:00:59 <mps> cat /proc/sys/kernel/hotplug should show /sbin/mdev after /etc/init.d/mdev is run
2020-09-21 21:01:49 <mps> but you add helper.sh as /sbin/hotplug executable
2020-09-21 21:02:05 <mps> it doesn't make any mess
2020-09-21 21:03:09 <mps> here is my /etc/mdev.conf https://tpaste.us/Xg5J
2020-09-21 21:03:58 <mps> last line (commented) can be used for debug by removing comment
2020-09-21 21:08:11 <shhhum[m]> Hmm. So I should not add `-.* root:root 660 *env > /tmp/.libudev-zero/uevent.$$` as libudev-zero suggests?
2020-09-21 21:08:17 <shhhum[m]> https://github.com/illiliti/libudev-zero/blob/master/contrib/mdev.conf
2020-09-21 21:10:36 <mps> well, I don't have it as you can see
2020-09-21 21:12:36 <mps> and I don't have /sbin/hotplug file, shell script or binary from libudev-zero contrib
2020-09-21 21:12:57 <shhhum[m]> Okay, I'll delete it then. Thanks for your help.
2020-09-21 21:15:22 <mps> shhhum[m]: np. thanks for trying and testing it
2020-09-21 21:22:07 <atkka> For those who requested: https://wiki.alpinelinux.org/wiki/Flatpak#Fixing_Audio_Issues
2020-09-21 21:23:15 <atkka> I will get some more testing done and clean it up as I figure things out, but that is all I have time for today. Thanks everyone!
2020-09-22 03:55:30 <atkka> hello, I created an alpine package and tested it, lints and installs fine. what is the next step to get it accepted?
2020-09-22 03:56:11 <atkka> github/gitlab?
2020-09-22 04:00:43 <yjftsjthsd> Thank you, atkka! I tried the audio fix just in a shell and it worked; I appreciate you updating the wiki. FWIW, it does work with flatpaks not installed with `--user` - I didn't realize that flag existed when I was setting up my box, so all my flatpaks were installed as root. Wouldn't do that again, but it does provide a datapoint...
2020-09-22 04:01:06 <atkka> yjftsjthsd: you're welcome!
2020-09-22 04:01:49 <atkka> good to know about having it work with --system installs
2020-09-22 04:02:27 <atkka> heads up if you want to install flatpaks with --user you have to also configure the remote repos with --user as well
2020-09-22 06:51:45 <Milka> mps: Greetings. I have trouble to make libinput work with libudev-zero (sees no input after boot). I guess the problem is in absence of /tmp/.libudev-zero directory during boot. mdev tries to follow the rules for input and create uevent file, but ends up reporting that directory is nonexistent. BUT if I later insert *any* input device - libinput detects them all. Have you encountered that kind of issue?
2020-09-22 06:53:52 <mps> Milka: no, it worked (and works) for me fine
2020-09-22 06:54:21 <Milka> I guess I missed something again
2020-09-22 06:54:26 <mps> (looks like I must write short guide for using libudev-zero on alpine)
2020-09-22 06:55:05 <mps> you enabled hwdrivers and mdev in sysinit runlevel?
2020-09-22 06:55:27 <mps> rc-update add hwdrivers sysinit
2020-09-22 06:55:41 <mps> rc-update add mdev sysinit
2020-09-22 06:56:39 <mps> and your kernel have /proc/sys/kernel/hotplug file?
2020-09-22 06:57:38 <mps> and, apk del udev-init-scripts-openrc eudev-openrc
2020-09-22 06:59:24 <Milka> I guess I forgot hwdrivers
2020-09-22 07:00:01 <Milka> I use default kernel and did not install eudev on this machine
2020-09-22 07:02:18 <mps> on which arch?
2020-09-22 07:02:21 <Milka> mps: ok, thanks, now it works, mdev still reports the error during boot though
2020-09-22 07:02:31 <Milka> 64
2020-09-22 07:02:35 <Milka> i
2020-09-22 07:03:01 <Milka> x86_64
2020-09-22 07:03:37 <mps> nice to hear
2020-09-22 07:03:43 <mps> which error?
2020-09-22 07:04:27 <mps> and what shows `cat /proc/sys/kernel/hotplug`
2020-09-22 07:05:43 <Milka> mdev's path
2020-09-22 07:06:26 <Milka> Error about absence of /tmp/.libudev-zero directory
2020-09-22 07:06:43 <Milka> while trying to add event file there
2020-09-22 07:07:15 <mps> ok, let me reboot and look again
2020-09-22 07:08:49 <Milka> I have it just after 'starting mdev'
2020-09-22 07:10:08 <mps> Milka: here is my boot log on serial console on armv7 board https://tpaste.us/MQO5
2020-09-22 07:14:53 <Milka> strange... I guess this must be rules I added to mdev.conf to support hotplug on libudev-zero
2020-09-22 07:15:30 <Milka> https://raw.githubusercontent.com/illiliti/libudev-zero/master/contrib/mdev.conf
2020-09-22 07:16:18 <mps> Milka: actually you don't need this
2020-09-22 07:17:54 <mps> Milka: sorry for curiosity but is your nick your real name? I know some with real name Milka
2020-09-22 07:18:42 <mps> it was quite popular female name when I was young
2020-09-22 07:22:27 <Milka> That's a lovely name, but unfortunately, not mine
2020-09-22 07:22:53 <mps> :)
2020-09-22 07:23:49 <mps> you probably have seen my real name, and it is 'similar' to your nick
2020-09-22 07:25:53 <Milka> Not sure
2020-09-22 07:27:25 <mps> look at maintainer for libudev-zero ;)
2020-09-22 07:33:12 <Milka> Oh, lovely name too. And thanks for libudev-zero. It's not widely spread among other distribs. I was looking for udev replacement since... forever, and never would've found it without your work.
2020-09-22 07:35:11 <PureTryOut[m]> atkka: submit it to https://gitlab.alpinelinux.org/alpine/aports
2020-09-22 07:36:33 <Milka> mps:  just found out that hotplug doesnt work without those rules
2020-09-22 07:37:57 <mps> Milka: hmm, strange. I will investigate later. now I preparing for business meeting and don't have free time
2020-09-22 07:38:21 <Milka> Ok, good luck
2020-09-22 07:56:49 <shhhum[m]> Milka: Yeah, I've reported the same thing yesterday to mps. It shows exactly the same errors during boot.
2020-09-22 08:52:03 <Milka> shhhum[m]: actually it's not critical. Still hard to believe that we can get wayland desktop without elogind and eudev
2020-09-22 08:53:11 <shhhum[m]> Milka: I am currently running Sway without elogind, but I am not using things like swayidle.
2020-09-22 08:54:43 <Milka> swayidle also works
2020-09-22 08:54:53 <Milka> and swaylock
2020-09-22 08:55:04 <Milka> why wouldn't they?
2020-09-22 10:05:55 <shhhum[m]> Hmm. I thought sway{idle,lock} required elogind, oh well.
2020-09-22 10:12:10 <abenz> I can use an app natively now on alpine, but I noticed that some parts of the interface are blurry..
2020-09-22 10:12:25 <abenz> perhaps not as crisp as I'd like is a better term
2020-09-22 10:12:43 <abenz> on my VM the same app looks crisp, this is likely due to missing fonts isnt it?
2020-09-22 10:13:03 <abenz> how do I go about identifying the ideal/missing fonts?
2020-09-22 11:48:22 <Hello71> shhhum[m]: swayidle didn't interact with logind for a long time. swaylock still doesn't afaik
2020-09-22 12:09:30 <alhassanaraouf> Why Firefox and Chromium Don't Play Netflix or any DRM Content? is it known problem or i did something wrong
2020-09-22 12:18:22 <abenz> its a known problem with netflix and drm ;)
2020-09-22 12:19:55 <abenz> in firefox settings, there should be a checkbox (play drm content), do you have that enabled?
2020-09-22 12:21:13 <abenz> I think you need some special plugins
2020-09-22 12:29:18 <u0jQx9gPyrYg> widevine is the keyword i believe
2020-09-22 12:29:43 <alhassanaraouf> yes i have enabled and the plugin installed but every time i play something it says plugin crashed
2020-09-22 12:30:02 <u0jQx9gPyrYg> aah. is that a binary plugin depending on glibc?
2020-09-22 12:30:05 <alhassanaraouf> u0jQx9gPyrYg yes that plugin
2020-09-22 12:30:58 <alhassanaraouf> i thought that too but not sure i will keep searching
2020-09-22 12:32:02 <alhassanaraouf> but also i installed chromium with chromium-chromedriver and also didn't work
2020-09-22 14:27:41 <luja> Good day for all. Where I can see screen shots of alpine linux. It is similar to puppy?
2020-09-22 14:30:03 <wsinatra> luja: alpine doesn't really look any specific way. It doesn't ship a default desktop or anything like that
2020-09-22 14:32:04 <luja> wsinatra: Oh!
2020-09-22 14:34:22 <luja> wsinatra: It's similar to Arch-linux but lightweight?
2020-09-22 14:37:02 <wsinatra> That's a decent way to look at it, the installer is CLI just like arch's, but there's a nifty set of setup scripts that automate the nitty gritty stuff for you
2020-09-22 14:37:57 <wsinatra> Though nothing's stopping you from manually installing by any means, heck, you can even create a custom ISO to handle builds for specific systems if you wanted to. Or just run everything from RAM
2020-09-22 14:39:08 <yjftsjthsd> Ah, but we *have* an installer, and helper scripts to help set things up. There's tradeoffs vs Arch's "no installers" approach.
2020-09-22 14:39:32 <yjftsjthsd> (That said, you're still expected to set up DEs yourself)
2020-09-22 14:40:18 <wsinatra> It's a wonderful balance honestly
2020-09-22 14:40:50 <wsinatra> it's enough in the weeds to satiate that desire to be in control of our systems, but friendly enough for people to pick up on a whim
2020-09-22 14:41:34 <wsinatra> plus the packaging is dead simple, so not having the AUR isn't really an issue. There are even tools to convert AUR packages to APKBUILDs if you find you're missing something
2020-09-22 14:41:54 <luja> Wow!, it's interesting. I'm a user of Linux Mint, albeit it's a good distribution I don't give you flexibility on desktop for example: to change from mate to xfce
2020-09-22 14:42:09 <luja> excuse me
2020-09-22 14:42:22 <luja> It don't gives you flexibility ...
2020-09-22 14:42:57 <luja> Alpine sounds like similar to LFS
2020-09-22 14:43:31 <luja> Linux From Scratch but a plus on scripts for configurations and others
2020-09-22 14:44:09 <wsinatra> Hahaha it isn't quiet compiling your own kernel and modules. But we won't stop you either, I've got Alpine edge running on a Droid4 using that approach
2020-09-22 14:44:23 <wsinatra> 's/quiet/quite/'
2020-09-22 14:45:13 <luja> wow!, ok for RE
2020-09-22 14:46:32 <luja> Interesting, I have a older machine now with puppy linux, but I will try alpine then
2020-09-22 14:46:59 <charl> It requires you to be somewhat comfortable installing stuff from the command line
2020-09-22 14:47:01 <luja> 's/a/an/' ;-)
2020-09-22 14:47:23 <charl> don't expect fancy graphic installers etc
2020-09-22 14:47:53 <charl> documentation is lacking, at least compared to archlinux amazing wiki
2020-09-22 14:49:31 <luja> yeap, I see. But if you are supporting here ...
2020-09-22 14:49:38 <charl> but if you need help setting up stuff we're here
2020-09-22 14:49:47 <luja> ;-)
2020-09-22 14:50:38 <wsinatra> Yep plenty of people here who are happy to help if you get stuck
2020-09-22 14:51:12 <wsinatra> documentation from the arch wiki can be helpful still, if you take all of that systemd junk and toss it out
2020-09-22 14:51:15 <luja> Thanks a lot.
2020-09-22 14:52:28 <luja> Ups 'if you take all of that systemd junk and toss it out'. I'm a tiny ignorant about then coulbe easy :D
2020-09-22 14:55:19 <luja> You have a document about differences between systemd and musl libc (Is it the core of afirmation?)
2020-09-22 14:55:58 <yjftsjthsd> Those are separate issues. Alpine uses openrc rather than systemd, and musl rather than glibc.
2020-09-22 14:56:03 <wsinatra> the difference would actually be between glibc and musl libc, and systemd vs openrc
2020-09-22 14:56:17 <yjftsjthsd> But systemd supports musl, and openrc supports glibc.
2020-09-22 14:56:29 <wsinatra> https://www.musl-libc.org/
2020-09-22 14:56:50 <wsinatra> plenty of info on musl here though, there's also #musl (I think that's the channel)
2020-09-22 14:57:00 <yjftsjthsd> (Gentoo uses openrc with glibc. I don't actually know of anyone using systemd with musl but I'm pretty sure it's supposed to work)
2020-09-22 14:57:13 <wsinatra> https://wiki.gentoo.org/wiki/Comparison_of_init_systems <- and they document things decently too ;)
2020-09-22 14:58:19 <luja> I go to read now the last link, thanks
2020-09-22 15:08:00 <luja> Well I see some diferences but somes are very technical for me :(, In the bottom of topic I can't see the advantages from one over the other :(
2020-09-22 15:09:22 <mps> systemd doesn't work on musl, thanks good providence
2020-09-22 15:09:32 <luja> I am a newbie for linux techical details yet.
2020-09-22 15:09:57 <luja> je,je
2020-09-22 15:10:41 <luja> mps, could you explain to me grosso modo the advantage from musl over systemd
2020-09-22 15:11:30 <mps> huh, they are different things
2020-09-22 15:11:58 <mps> can't compare grandma's with frogs
2020-09-22 15:12:21 <luja> ups!!
2020-09-22 15:12:48 <mps> systemd can be compared to openrc (and other init systems)
2020-09-22 15:12:53 <luja> openrc vs systemd
2020-09-22 15:13:06 <luja> grosso modo
2020-09-22 15:13:16 <mps> musl can be compared with glibc, and other libc implementations
2020-09-22 15:13:52 <nshp> luja: mostly what you'll care about as a user is just: when the archwiki tells you to do e.g. "systemctl restart <service>", you'll do "rc-service <service> restart"
2020-09-22 15:14:00 <programmerq> openrc is script based and is a lot more user-servicable. systemd is a completely different approach to how to launch system services and favors a compiled monolithic binary instead.
2020-09-22 15:14:06 <mps> well, simple: openrc is simple and imperative while systemd is complex and declarative
2020-09-22 15:15:07 <luja> nshp: Clear
2020-09-22 15:15:30 <mps> musl: https://www.etalabs.net/compare_libcs.html
2020-09-22 15:15:56 <mps> little outdated, though still good
2020-09-22 15:16:07 <opal> openrc has some shortcomings too, namely not supervising
2020-09-22 15:16:38 <mps> opal: it has supervisor
2020-09-22 15:16:43 <opal> anyone else had issues with transmission-3.00-r0 crashing on verifies? i'll prob get a coredump and see whats going on, but wondering if im the only one or not
2020-09-22 15:17:00 <opal> lol supervise-daemon doesnt count mps
2020-09-22 15:17:35 <opal> it's some weak shim
2020-09-22 15:17:51 <mps> opal: yes, and I dislike supervisors
2020-09-22 15:18:12 <mps> though I use some
2020-09-22 15:19:08 <mps> s6 and/or runit could be used beside openrc
2020-09-22 15:19:21 <opal> yeah thats what i do for now
2020-09-22 15:19:28 <opal> i wanna familiarise myself with s6-rc and try it at some point
2020-09-22 15:20:09 <nshp> I rolled my own
2020-09-22 15:20:16 <nshp> turns out init really doesn't have to be complicated
2020-09-22 15:20:46 <Milka> ncopa: is jack really required as hard dependency for mpv?
2020-09-22 15:20:47 <luja> Argg!!, I'm overheating my brain. It's very technical for me. Definetively I am a informed user only :(, I hope at least .(
2020-09-22 15:20:50 <opal> ya in the end the bare minimum is to start and reap ;)
2020-09-22 15:22:03 <wsinatra> luja: it's not too bad, just in head first. Throw up a virtual machine and poke the OS a bit and see if you like it
2020-09-22 15:22:52 <opal> luja alpine linux is hands-down the best binary-package linux distro ive used
2020-09-22 15:22:55 <mps> Milka: no, it is not needed, that is why I build mpv locally without unneeded 'things' (some will call it bloat)
2020-09-22 15:23:03 <opal> the biggest part about any distro is gonna be the package management
2020-09-22 15:23:21 <luja> wsinatra: ok. I will try.
2020-09-22 15:23:40 <opal> and to a lesser but still-important extent: the packages that are available, default init, libc, stuff like that
2020-09-22 15:23:47 <luja> opal: thanks
2020-09-22 15:23:57 <wsinatra> apk is a SOLID package manager
2020-09-22 15:24:15 <opal> all distros are "linux" and in my experience alpine gets out of my way and lets me manage it as "just linux" head-on
2020-09-22 15:24:17 <wsinatra> lighting fast in comparison to things like apt or dnf
2020-09-22 15:24:42 <opal> it's fast, and more importantly it's correct
2020-09-22 15:25:01 <luja> I initially think in apk for android
2020-09-22 15:25:13 <ikke> Alpine used apk before androi
2020-09-22 15:25:15 <opal> haha yeah i think a lot of people including myself thought that
2020-09-22 15:25:15 <ikke> android
2020-09-22 15:26:28 <luja> I feeling now that I am entering in a new world ;)
2020-09-22 15:26:53 <luja> 's/I feeling/I am feeling/'
2020-09-22 15:27:16 <opal> there are some shortcomings with alpine as with anything, but mostly comes down to my use case, and mostly when using it as a desktop distro
2020-09-22 15:27:29 <opal> i have no serious complaints, just some machines i find it better to use something else
2020-09-22 15:27:47 <opal> much of the reason i switched to gentoo on desktop
2020-09-22 15:27:47 <luja> opal: ok
2020-09-22 15:28:44 <luja> Is gentoo recomended for Stallman? I believe that
2020-09-22 15:28:50 <Milka> mps:  Hard dependency for that of stuff (full sound server for player that can do it's job on bare alsa) *sounds* pretty insane to do in upstream. Unless there is no way to  build it with Jack support, but exclude it from runtime dependencies.
2020-09-22 15:30:05 <opal> stallman does not recommend gentoo, no
2020-09-22 15:30:11 <opal> nor does fsf
2020-09-22 15:30:23 <opal> i'd take their criteria with a grain of salt
2020-09-22 15:31:01 <nshp> does it not pass the test because installing proprietary software is "too easy"?
2020-09-22 15:31:06 <opal> yes
2020-09-22 15:31:09 <nshp> that's a good one
2020-09-22 15:31:18 <opal> yeah i dont take the fsf seriously
2020-09-22 15:31:30 <opal> thats my choice those, more power to those who do
2020-09-22 15:31:35 <opal> s/those/though/
2020-09-22 15:31:54 <wsinatra> opal: what edge cases have you hit?
2020-09-22 15:31:54 <cfengineuser> hey guys i am trying to install pidgin
2020-09-22 15:31:59 <cfengineuser> have some trouble
2020-09-22 15:32:06 <opal> wsinatra for alpine on desktop?
2020-09-22 15:32:25 <wsinatra> yeah, just curious what other people are seeing. I use it as a daily driver myself
2020-09-22 15:32:45 <cfengineuser> when i look in the aports 3.12 i see that pidgin should be version 2.14.1
2020-09-22 15:32:58 <cfengineuser> but when i apk add pidgin i get version 2.13
2020-09-22 15:33:07 <opal> bought this desktop prebuilt when i was younger, was getting it for college and just wanted something quick and cheap, didnt know how bad nvidia was on linux, so thats one thing
2020-09-22 15:33:17 <opal> with alpine im stuck with nouveau or igpu which is less than desirable for my use
2020-09-22 15:33:23 <mps> Milka: yes, that is how it is made so build it with 'options' or not is solution
2020-09-22 15:33:37 <opal> also i play some stuff on occasion such as minecraft
2020-09-22 15:34:02 <opal> and some stuff like wine and pcsx2 pretty much require multilib support on 64-bit
2020-09-22 15:34:54 <mps> cfengineuser: pidgin in aports is upgraded after 3.12 is released
2020-09-22 15:36:09 <Milka> mps: so, upstream package is not gonna change?
2020-09-22 15:36:38 <cfengineuser> so what am i to do?
2020-09-22 15:36:41 <cfengineuser> mps
2020-09-22 15:36:59 <mps> use edge
2020-09-22 15:37:26 <mps> or rebuild pidgin locally on 3.12
2020-09-22 15:37:28 <cfengineuser> you mean just for that one package?
2020-09-22 15:37:44 <cfengineuser> i tried rebuilding but missing libgnt-dev
2020-09-22 15:38:36 <wsinatra> opal: ah this makes sense. I worked around some gaming issues with Steam, but it doesn't help if your GPU isn't well supported.
2020-09-22 15:38:42 <mps> first build libgnt and install it
2020-09-22 15:38:46 <mps> then pidgin
2020-09-22 15:38:58 <cfengineuser> no its libgnt-dev
2020-09-22 15:39:25 <wsinatra> I think my biggest oddity has been the lack of mingw packages and other such cross compilation tools is lacking, and I don't have the time to fix it immediately, end up resorting to massive Debian docker containers to work around things
2020-09-22 15:39:45 <mps> libgnt contains libgnt-dev
2020-09-22 15:39:45 <cfengineuser> I'm confused
2020-09-22 15:39:49 <cfengineuser> oh
2020-09-22 15:40:10 <cfengineuser> ok lets try that
2020-09-22 15:40:51 <nshp> the whole -dev package thing is mainly a debianism (and rhel?), other distros don't tend to split packages like that
2020-09-22 15:42:35 <mps> nshp: splitting pkgs is good idea
2020-09-22 15:43:33 <mps> Milka: I did understood what you mean
2020-09-22 15:43:40 <nshp> depends on what you want. for a workstation, as a developer, I don't want to bother installing every -dev package under the sun. if I run it, I'm probably going to be compiling it at some point.
2020-09-22 15:43:58 <nshp> if you aren't compiling all the time and/or care about space, then sure
2020-09-22 15:44:40 <mps> as a developer I want small workstation and install -dev when I need them and remove when I don't need them
2020-09-22 15:45:32 <mps> on alpine this is dead simple using metapackages
2020-09-22 15:46:48 <mps> and keep in mind that alpine is used a lot for servers and embeded systems, where -dev is not needed
2020-09-22 15:47:18 <nshp> then that's where we differ. I'd rather be a lazy data hoarder. I'll just plug another drive into my RAID rather than deleting things :)
2020-09-22 15:47:30 <nshp> absolutely, not arguing that it isn't a good idea there.
2020-09-22 15:49:35 <opal> whenever that argument about resource use is brought up, i always retort with "i want to use my space for useful data"
2020-09-22 15:49:41 <nshp> I didn't actually realize alpine had the -dev split as well when I said that. I don't use it interactively much, I'm just a lurker
2020-09-22 15:49:41 <opal> but yeah when laziness is a factor i cant argue there
2020-09-22 15:50:35 <opal> having "options" like globally installing *-man or *-dev or *-openrc would be nice
2020-09-22 15:51:33 <ikke> For *-man we have docs
2020-09-22 15:51:35 <charl> for man there is `apk add docs`
2020-09-22 15:51:49 <opal> ohh
2020-09-22 15:51:54 <charl> anything you have installed or install after that, installs man pages
2020-09-22 15:52:02 <opal> shit thats cool
2020-09-22 15:55:19 <luja> Hey guys thanks for the support I will try alpine on a virtual machine. I have already marked on hexchat this chanel ;-)
2020-09-22 15:56:28 <charl> good luck luja
2020-09-22 15:57:26 <wsinatra> good luck!
2020-09-22 15:57:27 <luja> charl: ;-)
2020-09-22 15:57:34 <luja> thank a lot
2020-09-22 15:57:53 <luja> s/thank/thanks/
2020-09-22 16:37:17 <Batoeh> Good afternoon. Would it be reasonable to attempt to make a alpine system a dedicated ZFS NAS/SANS machine?
2020-09-22 16:38:26 <wsinatra> I believe that's been done before
2020-09-22 16:38:45 <Batoeh> Sweet!
2020-09-22 16:40:04 <wsinatra> https://wiki.alpinelinux.org/wiki/Alpine_Linux_with_root_on_ZFS_with_native_encryption
2020-09-22 16:40:07 <wsinatra> might come in handy
2020-09-22 16:41:03 <wsinatra> http://lambdacreate.com/posts/14 <- I actually built an Alpine based NAS, though I just went with a simple RAID1 on 2 ext4 drives.
2020-09-22 16:41:21 <wsinatra> regardless, this definitely should be feasible
2020-09-22 16:41:26 <Batoeh> Thank you wsinatra. I really appreciate the assistance so far.
2020-09-22 16:41:56 <mps> wsinatra: what protocol you use for NAS file access
2020-09-22 16:42:28 <mps> samba, I see
2020-09-22 16:42:31 <wsinatra> mps: Samba4, the wife is still using Windows so it was the easiest method
2020-09-22 16:43:12 <mps> yes, I see
2020-09-22 16:43:16 <wsinatra> In all honesty, I probably would have gone with NFS if I was just worried about my systems. Right now I just rsync things onto and treat it like a big data warehouse (hence the name)
2020-09-22 18:20:49 <telmich> It seems there are some bugs in the current ifupdown scripts - I get a weird ipv6 route that looks like an ipv4 mapped address - given the previous problems (ipv4 has to come before ipv6  in the interfaces file) I guess that there is some variable leakage in the scripts
2020-09-22 18:21:35 <ikke> telmich: What version of alpine?
2020-09-22 18:23:28 <telmich> It also seems to be that this triggers that eth1.8 is not getting a link/set into up state on boot
2020-09-22 18:24:34 <ikke> is this ifupdown, or ifupdown-ng?
2020-09-22 18:26:23 <telmich> https://www.nico.schottelius.org/temp/failing_interfaces_alpine_linux-20200922
2020-09-22 18:26:38 <ikke> application/octet-stream :(
2020-09-22 18:26:52 <telmich> It has quite some funny/screwed up netmasks in it as well
2020-09-22 18:27:41 <ikke> telmich: ?
2020-09-22 18:28:47 <ikke> telmich: What version of alpine is this?
2020-09-22 18:30:50 <cfengineuser> thanks mps all good
2020-09-22 18:31:22 <mps> cfengineuser: you are welcome :)
2020-09-22 18:31:54 <mps> actually, I built all this on 3.12-stable before pushed upgrades to edge
2020-09-22 19:10:21 <zoobab> F/win 8
2020-09-22 21:50:24 <bovis> After updating wpa_supplicant.conf, what's needed to reload it and switch networks? rc-service wpa_supplicant restart && rc-service networking restart don't seem to do it.
2020-09-22 21:59:12 <j`ey> anyone know how to run the virtual iso from https://alpinelinux.org/downloads/ ?
2020-09-22 21:59:36 <j`ey> (specifically the aarch64 ones)
2020-09-22 22:04:10 <charl> bovis I don't use wpa but I think you need wpa_cli https://wiki.alpinelinux.org/wiki/Connecting_to_a_wireless_access_point#Launching_udhcpc_through_wpa_cli_actions
2020-09-22 22:37:13 <wsinatra> j'ey: I've used the aarch64 iso with qemu with great success
2020-09-22 22:37:44 <wsinatra> I wrote about it briefly here: lambdacreate.com/posts/11
2020-09-22 22:38:50 <bovis> Thanks charl , I'm looking into it
2020-09-22 22:42:00 <j`ey> wsinatra: ah, maybe I needed ttyAMA instead of ttySO
2020-09-22 23:08:42 <wsinatra> spotty network, but j'ey that's potential. I wrote a script to automate it for me. But I can share the config/script if you get stuck
2020-09-22 23:14:55 <j`ey> wsinatra: with AMA instaed of S0 it worked, thanks
2020-09-22 23:15:43 <wsinatra> awesome! glad I could help
2020-09-22 23:18:40 <atkka> anyone know how to start a service along with startx? .xinitrc won't work because you'd need permissions
2020-09-22 23:26:21 <Hello71> j`ey: I think it depends on -M type. virtconsole should be faster but I'm not sure if it works on arm
2020-09-22 23:27:00 <Hello71> (not that anything is fast with emulation)
2020-09-23 00:21:28 <atkka> I'm new to git, I have cloned aports from gitlab, made my changes and commited them, pushed to my own gitlab fork, getting ready to merge them (this is for a new package)
2020-09-23 00:22:15 <atkka> how do I make sure I'm synced with Alpine's master?
2020-09-23 00:22:47 <atkka> looks like I'm one commit behind
2020-09-23 00:24:09 <atkka> git rebase says branch master is up to date, how can I check that?
2020-09-23 00:26:10 <yjftsjthsd> Doesn't gitlab show your state compared to master?
2020-09-23 00:27:20 <atkka> yes on the website and I'm one commit behind since I cloned it even after git fetch --all
2020-09-23 00:27:36 <atkka> how can I pull from upstream to stay in sync iguess
2020-09-23 00:30:12 <atkka> or through the commandline is there a way to see the latest commit by someone else?
2020-09-23 00:31:13 <atkka> git remote -v gives me my fork as origin and upstream as https://gitlab.alpinelinux.org/alpine/aports
2020-09-23 00:34:59 <yjftsjthsd> Ah... so I kinda thought gitlab had a nice web-gui way to do a rebase. I know one of the web-guis I've used supported it, and I kinda always use that since I never remember the commands
2020-09-23 00:36:00 <atkka> ah ok, well I think I'm good there won't be any conflicts or anything. I'm going to submit a merge request and see what happens.
2020-09-23 00:36:09 <nshp> gitlab will do the simplest kind of non-conflicting rebase for you
2020-09-23 00:36:32 <atkka> I just want to see the changes since I cloned the repo
2020-09-23 00:37:15 <atkka> the web shows alpine/aports had a commit a few hours ago but git log doesn't have that commit
2020-09-23 00:37:49 <atkka> I rebased and fetched but still can't see the commit locally? am I supposed to? not sure how it all works yet
2020-09-23 00:39:50 <nshp> as in `git fetch upstream && git rebase -i upstream/master`?
2020-09-23 00:42:33 <atkka> as it https://wiki.alpinelinux.org/wiki/Creating_patches#Rebasing_against_Alpine_Linux.27s_master
2020-09-23 00:43:06 <nshp> try my rebase command
2020-09-23 00:43:24 <atkka> I did, what are you supposed to do when it brings up the editor?
2020-09-23 00:43:56 <nshp> assuming you don't want to do anything fancy, observe, write and quit
2020-09-23 00:44:16 <nshp> if you don't -i it won't bring up the editor, I just do that out of habit
2020-09-23 00:44:32 <atkka> ahh ok well now everything is there when I use git log
2020-09-23 00:45:19 <atkka> so that is local right, should I push the rebase changes to my repo?
2020-09-23 00:45:27 <atkka> fork i mean
2020-09-23 00:45:42 <nshp> correct. you'll have to force-push as well if you had already pushed your commit
2020-09-23 00:46:01 <nshp> because you're rewriting history in your fork by pulling in history from upstream
2020-09-23 00:47:11 <atkka> hmm ok, let me see
2020-09-23 00:51:27 <atkka> so I merged branch master of gitlab.alpinelinux.org:atka/aports, I can see my commit twice now with one with a new id/hash and I can see all the recent community commits
2020-09-23 00:51:33 <atkka> is that right?
2020-09-23 00:52:43 <atkka> I have yet to make my merge request for alpine/aports
2020-09-23 00:55:05 <atkka> ehh thats not right, I need to undo 2 commits that aren't needed
2020-09-23 00:59:45 <nshp> oh, you did not want to merge
2020-09-23 00:59:55 <atkka> yeah I see that now
2020-09-23 01:00:06 <atkka> when trying to force git told me to do that
2020-09-23 01:00:16 <nshp> heh, yeah, git likes to think it's smart
2020-09-23 01:00:30 <atkka> but I see now I need to undo that, its all a leaning experience
2020-09-23 01:00:46 <atkka> though git workflows seem like a pain in the ass at first
2020-09-23 01:01:02 <atkka> I nice general order of operations would be nice
2020-09-23 01:01:11 <nshp> so you never want to force-push to a shared branch, because if other people then pull that branch later, git gets confused
2020-09-23 01:01:21 <nshp> but for your own forks or feature branches, force-push is normal
2020-09-23 01:01:46 <atkka> all I want is to contribute packages to alpine and stay in sync with both them and my fork
2020-09-23 01:02:12 <nshp> you're very much on the right track, you just have to ignore git when it's trying to be helpful :)
2020-09-23 01:03:15 <atkka> thanks for the info, so I went back to a previous state with git checkout, now how do I make that stick?
2020-09-23 01:03:22 <atkka> force push from here?
2020-09-23 01:04:10 <nshp> `git checkout` is just like thumbing back a few pages in the book, it doesn't actually remove the pages ahead of you
2020-09-23 01:04:36 <nshp> checkout your master again, then use `git reset --hard <revision>` to actually chop off the pages back to <revision>
2020-09-23 01:04:56 <nshp> and *then* you should be able to force-push
2020-09-23 01:04:56 <atkka> where do I find the revision?
2020-09-23 01:05:14 <nshp> same thing you just checked out, I assume? the hash found in git log?
2020-09-23 01:05:20 <atkka> ah ok
2020-09-23 01:06:30 <nshp> and btw, if `git push --force` gave you an error rather than just a warning, that's the server side telling you no. gitlab has some optional settings to protect your master branch by rejecting force-pushes, but you can disable that in your repo settings on gitlab
2020-09-23 01:07:25 <atkka> says I'm not allowed to push code to a protected branch
2020-09-23 01:07:44 <nshp> yep, that's the thing I said. you can change that in the web UI.
2020-09-23 01:08:04 <nshp> that's more for cases where a bunch of people are doing work on feature branches in the same repo, rather than for private forks like yours
2020-09-23 01:08:05 <atkka> webhooks?
2020-09-23 01:08:15 <nshp> nope
2020-09-23 01:08:24 <atkka> just looking for the setting
2020-09-23 01:08:36 <nshp> I'm not sure which tab, I don't have a gitlab repo up at the moment
2020-09-23 01:09:00 <nshp> https://docs.gitlab.com/ee/user/project/protected_branches.html
2020-09-23 01:09:14 <atkka> thank you
2020-09-23 01:10:53 <atkka> ok, I am synced with alpine/aports
2020-09-23 01:11:02 <atkka> we are at the same head
2020-09-23 01:11:19 <atkka> thanks for the help btw
2020-09-23 01:11:51 <atkka> can you tell me what command I need to run to stay in sync before I commit in the future?
2020-09-23 01:12:03 <atkka> can I just pull -all now?
2020-09-23 01:13:14 <atkka> then push changes to my fork, then do my commit, then merge request?
2020-09-23 01:17:07 <nshp> sounds just about correct, but I honestly don't know if there is a `pull --all`
2020-09-23 01:17:27 <nshp> I tend to do what I said above: `git fetch upstream && git rebase -i upstream/master`
2020-09-23 01:17:37 <nshp> probably overly explicit, but that's what's burned into my fingers at this point
2020-09-23 01:18:33 <atkka> do you do that everytime then make your changes commit and push?
2020-09-23 01:18:39 <nshp> you can drop the -i to avoid pulling up an editor, I just use that as an extra opportunity to make sure I'm doing what I think I'm doing
2020-09-23 01:18:41 <nshp> yup
2020-09-23 01:18:56 <atkka> ok ok I think I'm getting it
2020-09-23 01:19:04 <atkka> need to spend a day with the docs
2020-09-23 01:19:49 <atkka> I've been reading the git docs, gitlab git docs, and alpine's wiki got a little confusing
2020-09-23 01:21:09 <atkka> do you know if abuild sanitycheck is the same as apkbuild-lint?
2020-09-23 01:21:56 <nshp> here's where I drop the sad news that I have no clue how to use alpine or apk, I just randomly saw a git question :)
2020-09-23 01:22:48 <atkka> haha its all good, everything lints and passes checks, I'm ready for my first merge request
2020-09-23 03:44:40 <winny> hi, I installed eudev via setup-udev, and i see I have a /dev/disk/by-uuid/ but only half of my disks have UUIDs.  I'm not sure what is going on here.... i thought all disks will have UUIDs?
2020-09-23 03:47:29 <winny> interestingly, looking at lsblk -f, multiple disks have the same uuid o.o
2020-09-23 07:32:36 <abenz> anyone had firefox 81 crash yet (with hw acceleration enabled) ?
2020-09-23 09:18:02 <ncopa> Milka: apparently it is
2020-09-23 09:18:53 <ncopa> jack support was enabled in commit e7e262263edfd61caa4f32d82e1313f5571a169b
2020-09-23 11:25:52 <MY-R> winny, run "blkid" as root
2020-09-23 11:46:48 <winny> MY-R; thanks, i asked around and it turns out one can't use UUIDs to refer to individual ZFS disks as each disk in the zpool gets the same UUID.  https://gist.github.com/winny-/4830f65419faed2ef81e368dcf1b9b99#file-gistfile3-txt .  I resolved the issue by exporting the zpool, and properly importing it using /dev/disk/by-id (i was misled by the alpinewiki zfs tutorials using /dev/sdX which is
2020-09-23 11:46:50 <winny> apparently a smell in zfs-land)
2020-09-23 12:04:21 <Milka> Anyone using sway without elogind?
2020-09-23 12:10:24 <shhhum[m]> Milka: I am using, but I build sway locally and then do `chmod a+s /usr/local/bin/sway`.
2020-09-23 12:11:42 <shhhum[m]> Milka: I am using, but I build sway locally and then do chmod a+s /usr/local/bin/sway.
2020-09-23 12:12:26 <Milka> shhhum[m]: why? actually sway comes suid out of the box
2020-09-23 12:12:36 <Milka> in alpine
2020-09-23 12:13:31 <shhhum[m]> I like to be on bleeding edge :^)
2020-09-23 12:13:57 <Milka> so you build master branch?
2020-09-23 12:13:59 <Hello71> winny: it's generally not a great idea on linux because linux makes minimal attempt to keep /dev/sdx consistent across boots
2020-09-23 12:14:03 <shhhum[m]> Yeah.
2020-09-23 12:14:28 <Milka> I've got several problems with sway
2020-09-23 12:14:29 <Hello71> although in practice as long as your disks are all connected at boot and via pci I think they should almost always stay the same
2020-09-23 12:15:20 <Milka> first one - I'm unable to switch ttys while sway is running
2020-09-23 12:15:38 <Milka> shhhum[m]: you also have this?
2020-09-23 12:15:53 <shhhum[m]> No, I can switch.
2020-09-23 12:16:22 <shhhum[m]> I think your problem is with mkinitfs.conf
2020-09-23 12:17:24 <shhhum[m]> Give me a sec.
2020-09-23 12:17:34 <Milka> sway just falls to tty and freezes
2020-09-23 12:17:55 <Milka> and only power button helps
2020-09-23 12:19:57 <shhhum[m]> So, /etc/mkinitfs/mkinitfs.conf should contain something like this: features="ata base cdrom ext4 keymap kms mmc raid scsi usb virtio". After that, you have to manually run sudo mkinitfs.
2020-09-23 12:21:56 <Milka> it's the same for me
2020-09-23 12:22:42 <shhhum[m]> Got it. And what's in your /etc/modules?
2020-09-23 12:24:02 <Milka> just af_packet and ipv6
2020-09-23 12:24:37 <Milka> all my modules are loaded with hwdrivers service I think
2020-09-23 12:25:23 <shhhum[m]> Okay, so you need to read this two articles based on which hardware you're using: https://wiki.alpinelinux.org/wiki/Intel_Video and https://wiki.alpinelinux.org/wiki/Radeon_Video
2020-09-23 12:27:41 <winny> Hello71; indeed, i had to install udev to get /dev/disk/... and follow the debian zfs wiki article for best practices not present on the alpine wiki
2020-09-23 12:27:45 <Hello71> pretty sure that only enables early boot kms
2020-09-23 12:28:01 <shhhum[m]> Or for a test, like wiki says, you can run modprobe fbcon and try to switch TTYs.
2020-09-23 12:28:11 <Hello71> which is nice if you use encryption or something but as long as you have udev that doesn't change post-boot behavior
2020-09-23 12:28:42 <shhhum[m]> To see if it actually solves your issue.
2020-09-23 12:33:49 <Hello71> fbcon.ko was removed in 4.14
2020-09-23 12:35:11 <shhhum[m]> Oh.
2020-09-23 12:35:17 <shhhum[m]> Really?
2020-09-23 12:36:19 <Hello71> also setup-xorg-base already sets up udev
2020-09-23 12:36:42 <Hello71> it also doesn't make sense to randomly install mkinitfs if you aren't already using it
2020-09-23 13:10:12 <abenz> playing 4k nature videos, firefox yet to crash. Not celebrating yet, just wanted to let you know it might finally be fixed :)
2020-09-23 15:45:49 <cfengineuser> whenever i start bitcoin on alpine, it kills itself with signal 6
2020-09-23 16:40:07 <ikke> that's a feature :P
2020-09-23 16:51:50 <jjuran> :-D
2020-09-23 17:17:05 <kciN> I just heard about alpine and that it does not utilize any gnu code, what core utils does it use then, plan9's?
2020-09-23 17:17:14 <c705> busybox
2020-09-23 17:17:29 <c705> by default. you can install coreutils yourself, however
2020-09-23 17:19:50 <kciN> I see, thanks.
2020-09-23 17:20:02 <c705> cheers
2020-09-23 17:28:40 <Stacy[m]> It has a lot of gnu code considering it is built with gcc
2020-09-23 17:29:06 <dmb> would be interesting if it used coreutils from the bsd's or solaris :)
2020-09-23 17:29:23 <dmb> for them its just a part of the os utilies, not seperate packages
2020-09-23 17:30:48 <kciN> what compiler does it come with?
2020-09-23 17:30:58 <Stacy[m]> it isn't imposible, but it would require someone to port bsd's libc
2020-09-23 17:31:05 <c705> musl
2020-09-23 17:31:21 <ikke> gcc
2020-09-23 17:31:40 <kciN> wat
2020-09-23 17:31:44 <c705> right, musl is the libc
2020-09-23 17:31:48 <c705> gcc is the compiler
2020-09-23 17:32:52 <kciN> kinda disappointed. would've liked to see 8c or tcc :)
2020-09-23 17:33:23 <j`ey> tcc cant build Linux kernel
2020-09-23 17:33:33 <kciN> true
2020-09-23 17:33:49 <kciN> I hope it gets agnostic soon
2020-09-23 17:34:08 <dmb> why not llvm/clang?
2020-09-23 17:34:09 <j`ey> doubt it, clang just added support for another extension (asm goto)
2020-09-23 17:34:11 <afontain_> tcc can't build busybox
2020-09-23 17:34:40 <afontain_> tcc can't build a gtk app with some gresources.
2020-09-23 17:34:43 <dmb> i'm not really sure purpose of alpine though is to be gpl-less
2020-09-23 17:35:08 <afontain_> tcc isn't too practical, I think
2020-09-23 17:35:13 <j`ey> Im assuming busybox is for the size, not the gpl-less
2020-09-23 17:35:15 <dmb> more of to be lightweight, musl is lightweight but also happens to be licensed with a freeer license
2020-09-23 17:36:36 <kciN> where does the busyboxs code come from though
2020-09-23 17:36:51 <j`ey> https://www.busybox.net/
2020-09-23 17:37:14 <j`ey> (also busybox is gpl anyway, forgot that)
2020-09-23 17:37:46 <kciN> I mean did they write it from scratch or ported from gnu or plan9
2020-09-23 17:38:42 <Stacy[m]> I believe kernel got more agnostic
2020-09-23 17:39:10 <kciN> gcc is a virus
2020-09-23 17:39:13 <Stacy[m]> They have triggers for whether compiler supports goto and some other feature in config
2020-09-23 17:40:25 <kciN> oh and the shell, is it dash?
2020-09-23 17:42:24 <j`ey> Stacy[m]: x86 relies on asm goto
2020-09-23 17:44:11 <wsinatra> kciN: busybox ash by default
2020-09-23 17:45:17 <c705> ash is roughly the same as dash
2020-09-23 17:47:08 <dmb> wheres mah zsh
2020-09-23 17:47:59 <c705> apk add zsh
2020-09-23 17:50:39 <ikke> Apk add zsh-doc
2020-09-23 17:56:40 <kciN> Got stuck on "boot:"
2020-09-23 18:06:59 <dmb> got stuck on LiLO
2020-09-23 19:36:32 <abenz> .. and firefox finally got stuck
2020-09-23 19:36:57 <c705> haven't had any ffox issues myself lately
2020-09-23 19:37:32 <abenz> c705: yes I think its solid once you turn off HW acceleration
2020-09-23 19:37:39 <abenz> doing so though makes for a very poor user experience
2020-09-23 19:37:45 <c705> i never had to do that
2020-09-23 19:38:03 <c705> firefox was having issues a couple weeks back, but I believe those were patched
2020-09-23 19:38:29 <abenz> its off by default on linux systems afaik, I had to enable it to be able to play high res videos or scroll through pages without cpu ramping up fans
2020-09-23 19:38:59 <abenz> to check, type about:support
2020-09-23 19:39:14 <abenz> Ctrl+F: compositing
2020-09-23 19:39:50 <c705> doesn't it just use OpenGL instead at that point?
2020-09-23 19:40:01 <c705> or webgl, even
2020-09-23 19:40:33 <abenz> I'm using webrender, I thought that was to be the new default? in any case thats what leaves my cpu almost idle while browsing
2020-09-23 19:41:34 <c705> I dunno. I use firefox and it works
2020-09-23 19:46:42 <abenz> c705: can you play uhd content on youtube? eg: https://www.youtube.com/watch?v=LXb3EKWsInQ
2020-09-23 19:47:08 <abenz> also are you using completely stock FF or did you change anything in about:config?
2020-09-23 19:47:35 <c705> totally stock. I can check
2020-09-23 19:47:59 <c705> well..ehh. The package is stock. I'm running it in firejail, but it's not changing any fcnality
2020-09-23 19:48:32 <c705> yup, watching it. green lizard is green
2020-09-23 19:48:42 <c705> Looks like the GPU is being used too
2020-09-23 19:49:00 <c705> I would imagine it's just using the DRI API
2020-09-23 19:49:06 <abenz> at 4k quality?
2020-09-23 19:49:09 <c705> yessir
2020-09-23 19:49:32 <abenz> thanks for the tip on firejail, first time I hear about it
2020-09-23 19:49:41 <c705> firejail is nice
2020-09-23 19:49:46 <abenz> last thing please
2020-09-23 19:50:10 <abenz> can you paste the decision log (from about:support) ?
2020-09-23 19:50:13 <abenz> I'd appreciate it
2020-09-23 19:52:26 <c705> abenz: https://termbin.com/l7ys7
2020-09-23 19:53:26 <abenz> c705: cheers
2020-09-23 19:53:51 <c705> :)
2020-09-23 20:37:28 <Milka> shhhum[m]: what packages did you install for xwayland to work correctly? I have sway crashing on every xwayland app starting.
2020-09-23 20:43:38 <c705> did you set XDG_RUNTIME_DIR
2020-09-23 20:47:19 <Hello71> shouldn't be related, and also they left already
2020-09-23 21:02:50 <shhhum[m]> Just the xorg-server-xwayland package.
2020-09-23 21:04:22 <shhhum[m]> Milka: Also, have you installed mesa-dri-swrast package?
2020-09-23 21:04:57 <shhhum[m]> I've had similar issues as you: https://gitlab.alpinelinux.org/alpine/aports/-/issues/10778
2020-09-23 21:27:38 <Milka> shhhum[m]: this package doesn't seem to exist
2020-09-23 21:28:14 <shhhum[m]> Oh, you're right.
2020-09-23 21:29:07 <shhhum[m]> Now it's called mesa-dri-gallium.
2020-09-23 21:29:39 <Milka> yea, I installed it, sway wouldn't start without it
2020-09-23 21:30:22 <shhhum[m]> Hmm.
2020-09-23 21:30:43 <Milka> it crashes without any debug messages on tty switch and on any xwayland-using app
2020-09-23 21:30:53 <shhhum[m]> I am actually have all the mesa-* packages.
2020-09-23 22:15:53 <atkka> i'm installing alpine on laptop with both nvidia and intel about to mkinitfs but I'd like to blacklist nvidia as I won't be using it
2020-09-23 22:17:17 <atkka> do I just make a blacklist.conf with nvidia and nouveau in it?
2020-09-23 22:19:39 <atkka> or do I need to add nv nvidia-drm nvidia-modeset et al?
2020-09-23 22:25:03 <atkka> also on modernish intel (skylake), is xf86-intel-video still recommended or should I use the kernel modesetting driver? I just want to avoid screen tearing and retain vaapi acceleration.
2020-09-23 22:33:32 <atkka> one more question, I see now that blacklist.conf is prepopulated with entries, with all the intel vulnerabilities is blacklisting microcode still recommended? what does enabling ucode updates on alpine entail?
2020-09-23 22:52:38 <Hello71> modesetting is better
2020-09-23 22:52:58 <Hello71> lots of the defaults are bad, but microcode isn't that useful because you should use initramfs loading anyways
2020-09-23 22:53:47 <Hello71> apk add /boot/intel-ucode.img; grub-mkconfig etc
2020-09-23 22:57:50 <atkka> Hello71: thanks for the info
2020-09-23 22:58:01 <Hello71> uh
2020-09-23 22:58:15 <Hello71> apk add intel-ucode I mean
2020-09-23 22:58:29 <Hello71> not sure if update-extlinux adds it by default
2020-09-23 22:58:48 <atkka> I got what you meant, I'm using grub-efi
2020-09-23 23:09:50 <Hello71> https://wiki.archlinux.org/index.php/Microcode
2020-09-23 23:49:47 <atkka> Hello71: perfect, thanks
2020-09-24 03:25:45 <enyc> 
ACTION meows

2020-09-24 04:30:12 <atkka> can anyone tell me what kind of stuff I would need dbus for? a few wiki guides and whatnot say you need dbus or dbus-x11, I have omitted them from a fresh install and everything seems to work, X, firefox etc
2020-09-24 04:31:09 <atkka> when I shutdown there is a warning about polkit needing the dbus service which doesn't exist but that's all i've noticed
2020-09-24 05:38:38 <atkka> I guess a better question would be, if I have a minimal desktop, no DE, and a few gui apps will much be broken other than notifications and such?
2020-09-24 05:40:37 <zcrayfish> Nothing is wrong with just using a window manager.
2020-09-24 05:49:21 <k0takbasser> Hello
2020-09-24 05:50:17 <k0takbasser> There is a problem that sometimes our local mirror is not avaliable. So I want apk to fallback to some other mirror. Is it possible to configure fallback mirrors?
2020-09-24 05:51:37 <ikke> I don't think apk supports that
2020-09-24 05:55:06 <k0takbasser> Oh thats sad :(
2020-09-24 06:57:28 <k0takbasser> Do you think that the mirror fallbacking feature would be helpful or it neseserally comlicates the package manager?
2020-09-24 06:58:08 <k0takbasser> *unnecessarily
2020-09-24 06:58:10 <ikke> You can open a feature request in https://gitlab.alpinelinux.org/alpine/apk-tools
2020-09-24 08:39:47 <jagungal> Hello, I'm very new to Linux and I managed to connect to my wifi, but I rebooted and now I can no longer connect. Running `wpa_supplicant -i wlan0 -c /etc/wpa_supplicant/wpa_suppicant.conf` returns a series of `Authentication request to the driver failed` errors. Is there anything I can do to fix this?
2020-09-24 09:05:31 <shmileee> Hi everyone. I'm using Alpine Linux 3.12 standard distro for hosting Drone-CI server on-pre and running it as a Docker container. Has anyone faced any container redirect issues with Docker or could point me out towards network troubleshooting path? Thanks in advance.
2020-09-24 09:06:22 <ikke> What kind of container redirect issues?
2020-09-24 09:10:10 <shmileee> ikke, it's more droneci-related, - oauth2 integration with bitbucket provider. I've tested same naked Docker setup on CentOS 8 in the same network and everything is working, hence the assumption Docker on Alpine (or perhaps Alpine itself) is somehow altering my network which I don't know how to troubleshoot.
2020-09-24 09:11:35 <ikke> shmileee: without more details, it's hard to say, but I would not expect that to be the case
2020-09-24 09:13:00 <shmileee> I'm using Packer to preprovision pure Alpine 3.12, here's gist of what the system looks like: https://gist.github.com/shmileee/ae511be2b0d820c918dccd32f2c3e3f3
2020-09-24 09:15:49 <shmileee> ikke, I'm assuming pure alpine does not have any firewall activated by default?
2020-09-24 09:17:23 <ikke> no
2020-09-24 10:35:12 <shmileee> ikke, i've narrowed down the issue to drone running on alpine natively, still the same network error :'(  if anyone has managed to run drone-ci sever on alpine 3.12 - please let me know
2020-09-24 13:00:51 <Gnjurac> hi 1 dumb question, i added alies for acme.sh but when i su, it reads bashrc and says
2020-09-24 13:01:08 <Gnjurac> bash: acme.sh=/root/.acme.sh/acme.sh: No such file or directory ?
2020-09-24 13:01:47 <Gnjurac> but i can run acme.sh no problem and it exists
2020-09-24 13:01:52 <Gnjurac> so hmm wtf?
2020-09-24 13:12:47 <Gnjurac> lol i frogot to add alias
2020-09-24 13:18:48 <Gnjurac> for cron /etc/periodic/foo i can use symbolic links right i dont have to put scrip itslef
2020-09-24 13:18:51 <Gnjurac> there
2020-09-24 13:32:40 <Gnjurac> tough hmm i think maybe i should not run this acme.sh as root
2020-09-24 13:33:38 <Cogitri> Nop, best to make a separate user for it
2020-09-24 13:34:26 <Gnjurac> i was just thinking of that
2020-09-24 13:37:49 <Cogitri> And since acme.sh preserves the ownership of files it's pretty easy to setup
2020-09-24 13:38:36 <Cogitri> Just touch the file where the key and cert are saved, chown them to acme:root and done
2020-09-24 13:39:16 <Cogitri> (And make sure to also make the files 440/660)
2020-09-24 13:40:44 <Gnjurac> you using acme.sh from apk add or scrip directly from git?
2020-09-24 13:49:14 <Cogitri> From apk
2020-09-24 13:49:18 <Cogitri> But it shouldn't really matter
2020-09-24 13:53:29 <Gnjurac> ok so, creat user with no special group and home login, ok so to make nginx key and that other file owner acme:root and chmod 440 or 660
2020-09-24 13:54:16 <Gnjurac> and add chronjob for that user for acme.sh to renew i guess
2020-09-24 13:56:00 <Cogitri> Yup
2020-09-24 13:56:31 <Cogitri> And then create your key and cert as that user on first run
2020-09-24 13:56:45 <Cogitri> Afterwards acme.sh --cron will automatically refresh everything for you
2020-09-24 13:57:13 <Gnjurac> ok will try its my first time setting up nignix and acme but i think i will get it
2020-09-24 14:42:44 <Gnjurac> thos owner acme:root i use acme only if my new user name is acme right? cuz i named my new user foracme so guess it should  be foracme:root
2020-09-24 14:53:33 <Cogitri> Yup
2020-09-24 17:27:15 <Gnjurac> hmm 1 question there is no /etc/chrontab/foracme
2020-09-24 17:27:25 <Gnjurac> how will it autoupdate then
2020-09-24 17:32:11 <Gnjurac> guess i can just add acme.sh --renew -d example.com --force and run monthly
2020-09-24 17:36:19 <Gnjurac> oh nice so for renewal i dont need to supply token again
2020-09-24 18:02:18 <sam3> Hello.  Is there a way to tell the date and time that a package was installed w/ apk?
2020-09-24 18:10:51 <ikke> sam3: No, that's not recorded afaik
2020-09-24 18:15:03 <Milka> shhhum[m]: what mesa version do you use?
2020-09-24 18:15:21 <shhhum[m]> Milka: I'm on Edge.
2020-09-24 18:17:13 <shhhum[m]> So it is 20.1.5-r2.
2020-09-24 18:17:21 <Milka> the bug that crashes sway for me on xwayland apps also crashes gdb when I try to attach gdb to sway process to catch it
2020-09-24 18:19:08 <Milka> asked sway guys for support and they made great effort to help me, but after gdb crashed they were like :O
2020-09-24 18:21:12 <Milka> shhhum[m]: do you have several gpus on your setup?
2020-09-24 18:21:47 <c705> for what it's worth, I'm using sway on edge with little to no issues
2020-09-24 18:22:02 <c705> I also have 2 GPus, one of them is owned by vfio however
2020-09-24 18:22:51 <shhhum[m]> Milka: Yes I do.
2020-09-24 18:25:19 <Milka> c705, shhhum[m] would you guys please provide some instructions about how did you install it? What packages were involved? And do you use xwayland?
2020-09-24 18:29:24 <shhhum[m]> Yeah, I am using xwayland with sway. I am not sure what else I can provide besides what I've already told you a little while ago.
2020-09-24 18:30:07 <shhhum[m]> Have you read this: https://wiki.alpinelinux.org/wiki/Sway ?
2020-09-24 18:35:26 <c705> I ditto shhhum[m]. All i did was install some packages
2020-09-24 18:35:50 <c705> I also had to export XDG_RUNTIME_DIR per the Alpine Wiki docs
2020-09-24 18:35:58 <Gnjurac> hmm have 1 question about cert renew like if i am runing it under foracme user this --reloadcmd     "service nginx force-reload" it wont run cuz like foracme dosent have rights to rc-service ngix restart it needs root/sudo
2020-09-24 18:36:17 <shhhum[m]> You can also try to run sudo apk upgrade -U -a
2020-09-24 19:30:33 <c705> is alsa on edge broken for anyone else? Can't run speaker-test: Unable to set hw params for playback: Out of memory
2020-09-24 19:30:39 <c705> i can assure you i am not OOM
2020-09-24 19:44:53 <c705> changing my asound.conf resolved this issue. It was working 3 hours ago and magically stopped for whatever reason
2020-09-24 19:46:17 <c705> no hardware changes, no aports updates I can see that would affect that. Super annoying and weird
2020-09-24 19:51:41 <opal> interesting
2020-09-24 19:51:56 <opal> i actually need to move my sound stuff to work with jack
2020-09-24 19:52:03 <opal> im feeling alsa's limitations in my setups
2020-09-24 19:52:41 <c705> me too
2020-09-24 19:53:28 <c705> I even rebooted into my live linux image that I know for a fact has been working for the last 2 weeks (the past 1 year even) and today after 2Pm, just magically strats throwing the OOm error
2020-09-24 19:53:34 <c705> same kernel and everything. I don;t get it
2020-09-24 19:53:49 <c705> maybe a lib got updated or something, I don't care enough to look into it
2020-09-24 21:12:53 <nsz> c705: you can try to run the command under strace -f and see what syscall fails with ENOMEM
2020-09-24 21:13:07 <nsz> it can be some weird alsa related ioctl
2020-09-24 21:13:41 <nsz> (but it also can be something in libc and then you wont see it on the syscall level)
2020-09-24 21:14:48 <c705> i already resolved it, but thanks
2020-09-24 21:15:12 <c705> (ltrace would catch the libc error, if there was one)
2020-09-25 07:22:30 <abenz_> c705: sway on edge? FF on wayland uses a differrent compositor, and is accelerated!
2020-09-25 07:23:16 <abenz> this is re our convo yesterday on how FF runs smoothly for you and is not accelerated for me by default (I'm on Xorg)
2020-09-25 07:23:55 <abenz> if there was an openbox alternative on wayland, I'd jump ship instantly
2020-09-25 09:51:48 <gillesmauve> Has that video-freezing bug gone away yet?
2020-09-25 09:53:41 <Cogitri> Can't reproduce it so I couldn't tell
2020-09-25 10:38:49 <abenz> gillesmauve: if you're on Xorg and using webrender, its still there
2020-09-25 10:39:02 <abenz> wayland seems stable though
2020-09-25 10:40:54 <MY-R> imho webrender from ff 80 working great on xorg
2020-09-25 10:41:30 <abenz> could it be related to gpu used? I'm using an amd card, amdgpu driver
2020-09-25 10:42:07 <abenz> MY-R: what gpu?
2020-09-25 10:42:15 <MY-R> radeon driver
2020-09-25 10:42:58 <MY-R> on versions before 80 I saw in dmesg lot of weird errors with webrender turned on
2020-09-25 10:43:07 <MY-R> now all is clean and clear
2020-09-25 10:44:23 <abenz> MY-R: can I trouble you with pasting your Decision log from about:support ?
2020-09-25 10:46:17 <MY-R> abenz, dont have alpine anymore runnig so it wont tell you much
2020-09-25 10:47:16 <abenz> I see
2020-09-25 11:49:56 <cfengineuser> How do I add a disk to alpine
2020-09-25 11:49:57 <cfengineuser> ?
2020-09-25 11:51:19 <cfengineuser> oh got it
2020-09-25 16:37:02 <dirkson> Hey all. Download page - What's the difference between standard->aarch64 and generic arm->aarch64 ?
2020-09-25 16:55:59 <Hello71> iso vs tar
2020-09-25 16:56:22 <Hello71> also maybe slightly different packages
2020-09-25 18:43:15 <gjabell> is there a way to refer to disks by id/uuid rather than sdX mapping?
2020-09-25 18:44:59 <artok> gjabell: on fstab and kernel cmdline, sure
2020-09-25 18:45:35 <gjabell> I'm looking to pass a disk into a VM but it'd be nice to use a path that won't change
2020-09-25 18:46:17 <c705> gjabell: check /dev/disk/bu-uuid/
2020-09-25 18:46:25 <c705> /dev/disk/by-uuid/
2020-09-25 18:46:44 <gjabell> was looking for that but it doesn't seem to be a directory on my machine
2020-09-25 18:46:58 <gjabell> is there something I need to do to set it up?
2020-09-25 18:47:10 <c705> err..I guess so, but I don't know what provides that
2020-09-25 18:47:18 <c705> Perhaps udev?
2020-09-25 18:47:21 <gjabell> hmm
2020-09-25 18:48:37 <gjabell> yeah looks like it's a udev thing
2020-09-25 18:48:49 <c705> interesting
2020-09-25 18:50:37 <mps> blkid
2020-09-25 18:50:48 <mps> no need for udev
2020-09-25 18:51:04 <gjabell> does blkid create those directories though?
2020-09-25 18:51:32 <mps> no
2020-09-25 18:52:00 <mps> it just show block devices with their 'attributes'
2020-09-25 18:52:01 <c705> blkid isn't going to privide path names
2020-09-25 18:52:07 <gjabell> ah, looking for a way to have them created, not find the id
2020-09-25 18:52:19 <gjabell> doesn't matter if it provides pathnames anyways if they aren't static
2020-09-25 18:52:29 <artok> can't mount take UUID as source?
2020-09-25 18:52:30 <gjabell> I can always refer with /dev/sdX but that might change
2020-09-25 18:52:41 <gjabell> mount can I think but libvirt can't :/
2020-09-25 18:52:43 <mps> c705: try ti run it and look
2020-09-25 18:53:05 <mps> artok: yes mount UUID=........
2020-09-25 18:53:31 <c705> I know what blkid does, it's still not going to help with what this person wants
2020-09-25 18:53:57 <gjabell> I think udev is what I needed, unf have to reboot to check it haha
2020-09-25 18:54:06 <c705> gjabell: correct
2020-09-25 18:54:37 <mps> c705: ah, sorry, reread it and now see, yes you are right
2020-09-25 18:55:02 <mps> I thought mount option UUID is needed
2020-09-25 19:06:40 <Hello71> mdev can probably be configured to do it
2020-09-25 19:08:26 <gjabell> looks like udev works, thanks for the tip c705 :)
2020-09-25 19:08:46 <gjabell> might be nice to just use mdev tho
2020-09-25 19:09:39 <gjabell> hmm now my zfs drives arent found
2020-09-25 19:09:46 <gjabell> thats a bit awkward
2020-09-25 19:48:52 <mps> Hello71: right, I have done that with my tests of libudev-zero. with help from some scripts from https://github.com/slashbeast/mdev-like-a-boss
2020-09-26 01:00:44 <evilcat> yo. I'm using this script and am having some trouble getting networking working in qemu, would anyone be able to give me a hand setting up a tun/tap interface so I can use the arm version of alpine?
2020-09-26 01:00:46 <evilcat> https://gist.github.com/tuxmartin/f348d4247beda04e6029366da494f164
2020-09-26 01:01:04 <evilcat> at the moment I'm building a tun tap interface like this:
2020-09-26 01:01:08 <evilcat> sudo ip tuntap add dev tap3 mode tapsudo ip link set up dev tap3sudo ip link set tap3 master br99sudo sysctl -w net.ipv4.ip_forward=1
2020-09-26 01:01:14 <evilcat> and have made a bridge in networkmanager
2020-09-26 01:02:11 <evilcat> I'm using these args to start the 'qemu-system-arm' '-net nic,vlan=0 -net tap,vlan=0,ifname=tap3'
2020-09-26 01:03:06 <evilcat> I'm not 100% if I'm doing this correctly as when I do ifconfig, I don't see any interfaces and in /sys/devices/virtual/net besides lo
2020-09-26 07:29:24 <Milka> I'm getting unsatisfiable constraints while trying to install some stuff, about libsasl-2.1.27-r9 requiring libgdm.so.4 that is (missing). What's wrong?
2020-09-26 07:29:53 <Milka> libgdbm.so.4 I mean
2020-09-26 07:34:16 <ikke> Milka: it's being rebuilt atm
2020-09-26 07:34:51 <Milka> ikke: ok, thanks
2020-09-26 09:16:40 <Guest85195> help /m
2020-09-26 09:17:10 <Guest85195> test123
2020-09-26 09:17:46 <Guest85195> hi everybody, first time using irssi, don't know well how to use it but learning
2020-09-26 09:17:59 <Guest85195> question about native zfs with alpine
2020-09-26 09:18:12 <Guest85195> would like to add a swapfile
2020-09-26 09:18:15 <mps> there is #test channel on freenode
2020-09-26 09:18:21 <Guest85195> thanks
2020-09-26 09:18:41 <Guest85195> don't have any /dev/zvol devices listed
2020-09-26 09:18:48 <Guest85195> so it seems that udev rules missing
2020-09-26 09:19:16 <Guest85195> I have got this kind of file such as rules.d/60-zvol.rules
2020-09-26 09:20:04 <Guest85195> but no /dev/zvol, even after reboot, any chance I forgot anything?
2020-09-26 09:21:01 <Guest85195> actually on alpine on encrypted zfs on my lenovo x1 carbon
2020-09-26 09:53:57 <Guest85195> nevermind, surely copied the rule at a bad location, should surely be under /etc/udev/rules.d instead  of /lib/udev/rules.d/, have a nice day
2020-09-26 10:40:14 <CactiChameleon9> Hello... I have a question. Why are lots of doc pages missing in the wiki?
2020-09-26 10:40:50 <mps> CactiChameleon9: probably because not written ;)
2020-09-26 10:40:59 <CactiChameleon9> Ah... OK
2020-09-26 10:41:22 <CactiChameleon9> How feasible would it be to use alpine as my main linux distro?
2020-09-26 10:41:34 <mps> maybe some are removed because they are outdated
2020-09-26 10:41:38 <CactiChameleon9> (with bedrock linux to support other package managers)
2020-09-26 10:42:01 <mps> depends on your use cases
2020-09-26 10:42:50 <CactiChameleon9> A lightweight system that avoids loads of unnecessary reads and writes, with pleanty of flexability
2020-09-26 10:42:57 <CactiChameleon9> I think alpine fits that description
2020-09-26 10:43:53 <ikke> As long as you have at least a bit of a background in Linux, alpine should work fine
2020-09-26 10:43:59 <ikke> But there is not a lot of hand-holding
2020-09-26 10:44:10 <CactiChameleon9> OK, thanks
2020-09-26 10:45:10 <CactiChameleon9> I'll get started then. Thanks for the info
2020-09-26 15:47:26 <alxsdf> Hey, I'm getting problems with unsatisfiable constraints on an upgrade
2020-09-26 15:50:11 <ikke> alxsdf: What errors exactly?
2020-09-26 15:54:35 <alxsdf> sorry I was helping someone in the doom emacs discord
2020-09-26 15:54:46 <alxsdf> https://tpaste.us/5Vj0
2020-09-26 15:54:51 <alxsdf> ikke: the errors ^
2020-09-26 15:55:54 <ikke> thanks
2020-09-26 15:57:34 <alxsdf> for reference I'm on alpine edge
2020-09-26 15:58:31 <ikke> Right, that was my next question (and I'm also interrupted with diferent things)
2020-09-26 15:59:06 <alxsdf> It's ok, I understand
2020-09-26 15:59:46 <alxsdf> I'm an old hat to posting for help on IRC and waiting half a day for responses, you're fairly quick responding :)
2020-09-26 16:00:06 <ikke> heh :)
2020-09-26 16:00:41 <ikke> ok first, what is the output of `apk version -l '?'`
2020-09-26 16:00:56 <ikke> This usually means there is a package installed which is no longer present, and it's holding things back
2020-09-26 16:01:38 <alxsdf> uhhhhhhhhhhhhhhhhhhh quite a lot
2020-09-26 16:01:39 <alxsdf> https://tpaste.us/vZrL
2020-09-26 16:01:43 <alxsdf> o-o
2020-09-26 16:02:22 <ikke> right, I see
2020-09-26 16:02:53 <ikke> the py-* stuff should be explicitly py2 (if still exists) or py3
2020-09-26 16:04:21 <ikke> hmm, libreoffice is temporarily disabled, because it has build issues with gcc-10
2020-09-26 16:04:25 <alxsdf> The general strategy would be to uninstall things and then reinstall the properly versioned ones?
2020-09-26 16:04:32 <alxsdf> I don't really depend on libreoffice at the moment
2020-09-26 16:04:58 <alxsdf> or properly present ones, rather
2020-09-26 16:05:05 <ikke> yes, indeed
2020-09-26 16:05:15 <alxsdf> I can forsee maybe qt-x11 causing problems, I'm hesitating to guess that that has a lot of dependencies
2020-09-26 16:05:20 <ikke> You could just edit /etc/apk/world, and then run apk fix afterwards
2020-09-26 16:05:45 <alxsdf> ah, you mean edit the packages so they're the present packages, and then ask apk to fix it? Or just remove them
2020-09-26 16:05:55 <ikke> either
2020-09-26 16:06:12 <alxsdf> Interesting. Thanks for your prompt and useful help :)
2020-09-26 16:33:03 <winny> not sure if this is related to alpine, but I've noticed my .ssh has a weird listing output: drwx--S---    2 ansible  ansible       4096 Aug 27 01:31 .ssh
2020-09-26 16:33:26 <winny> right, it's actuall 2700
2020-09-26 16:33:48 <winny> not quite sure how i've managed that on all my ansible deploys
2020-09-26 16:38:02 <winny> umask 0077 && mkdir thing && ls -ld thing -> 2700 perms.  Is that the wrong umask?
2020-09-26 16:39:23 <winny> tested in an ubuntu environment as well, it created the dir with 0700 perms.
2020-09-26 16:44:02 <ikke> umask 0077; touch foo
2020-09-26 16:44:09 <ikke> -rw------- 1 build build    0 Sep 26 16:43 foo
2020-09-26 16:45:09 <winny> interesting, let me look further.
2020-09-26 16:48:06 <winny> i had just tested it in docker, and it did the weird behavior, but after doing a docker pull it works as intended.  On a couple servers I manage it still happens though, perhaps a recent update fixed it?
2020-09-26 17:37:42 <flox> Hi all. I have a question: is there an alpine iso that comes with a build chain (llvm, autotools, etc.) pre-installed and offline capable?
2020-09-26 17:38:03 <ikke> No
2020-09-26 17:38:16 <ikke> What do you mean with offline-capable?
2020-09-26 17:39:13 <flox> Ideally something like bootstrappable, like includes sources for all installed packages
2020-09-26 17:39:17 <winny> hosting the entire repo locally is do-able, in case you cannot get back online
2020-09-26 17:39:58 <ikke> No, we have nothing like that
2020-09-26 17:40:08 <ikke> even if you have aports, you still do not have all the sources
2020-09-26 17:42:24 <flox> I'm not looking for all the sources, more something like an iso that you can use to rebuild a system from scratch, eg. using something like a linux from scratch handbook
2020-09-26 17:56:21 <winny> https://wiki.alpinelinux.org/wiki/How_to_setup_a_Alpine_Linux_mirror could use this as inspiration for your project, as it sounds like you aren't exactly sure what exactly you want installed, so you can just create a local mirror (i think amirror can also be a directory too, in case you don't want to set up a web server)
2020-09-26 17:57:21 <winny> the mirror is small enough it can live on a SDXC card
2020-09-26 18:00:23 <winny> guess they didn't want to put in the work
2020-09-26 18:41:33 <winny> looks like that umask/mkdir issue was fixed within the last couple months.  i wrote the list with a reproducer in case anybody is interested.  just a bit nutters that something like umask/mkdir simply does the wrong thing, in some configuration
2020-09-26 18:48:18 <szgy> hi y'all
2020-09-26 18:48:40 <szgy> so, I've been trying to install alpine to a partition in my computer.
2020-09-26 18:48:48 <szgy> On the wiki there is this https://wiki.alpinelinux.org/wiki/Installing_Alpine_on_HDD_dualbooting
2020-09-26 18:49:07 <szgy> but it said that it doesn't work with recent alpine linux versions
2020-09-26 18:51:05 <szgy> but I followed anyways to see what happened. I could install it on a partition but I can't boot to it. It boots but gives me a single user prompt (not a grub prompt) with the error "cant mount /dev/sda2 to /sysroot : no such file or directory"
2020-09-26 18:52:14 <szgy> Up to now I tried:
2020-09-26 18:52:18 <szgy> 1. regenerating initramfs vía a chroot
2020-09-26 18:53:11 <szgy> 2. Installing grub2 on alpine linux instead of keeping it on my main os (artix). Also didn't work and the same error happenened, which lead me to believe that the problem is not on the bootloader...
2020-09-26 18:53:58 <szgy> 3. after getting to the single user prompt y "mount -f -t ext3 /dev/sda2 /sysroot" and Ctrl-D (this gave a kernel panic :/   )
2020-09-26 18:55:38 <szgy> anyways, I think that the problem is mainly with the installation process. But since I followed an outdated wiki, I should count on it working... If someone has any pointers, I'll kindly appreciate it!
2020-09-26 18:56:29 <szgy> * I shouldn't count on it working...
2020-09-26 20:13:29 <velix> Hey. Isn't edge available under Docker's alpine:edge anymore? I'm getting an error here.
2020-09-26 20:13:49 <velix> "Error: getting image docker.io/library/alpine:edge from image store failed: image "docker.io/library/alpine:edge": not found"
2020-09-26 20:14:59 <ikke> velix: should be stil available
2020-09-26 20:15:08 <velix> ikke: hmm... :latest works
2020-09-26 20:15:22 <ikke> velix: docker pull alpine:edge works for me
2020-09-26 20:15:29 <ikke> latest is the latest stable version
2020-09-26 20:15:33 <ikke> ie, 3.12 now
2020-09-26 20:15:51 <velix> Yeah... hmm. Maybe I've got a problem with caches. Let me try to clean them.
2020-09-26 20:16:45 <velix> Since I need inkscape 1.0.1, I either need to backport them or use edge.
2020-09-26 20:16:51 <velix> But I'm too tired to backport today.
2020-09-26 20:17:55 <velix> Heck, now it works
2020-09-26 20:18:00 <velix> \o/
2020-09-26 20:19:05 <velix> Note to myself: don't name a container "imagemagick" when it's for "inkscape" ;)
2020-09-26 20:19:40 <velix> Anyone with an idea, why LibreOffice 7 still didn't make it? Last time I've read about several musl incompatibilies.
2020-09-26 20:22:16 <ikke> The current version does not even build with gcc 10
2020-09-26 20:22:23 <velix> :D
2020-09-26 20:23:11 <velix> ikke: Some days ago, I've tried to find a regression and tried some very old versions of Open- und Libre Office.
2020-09-26 20:23:18 <velix> They all were based on Java...
2020-09-26 20:23:29 <velix> It's incredible, how they got this out of the core :D
2020-09-27 02:54:02 <opal> im glad libreoffice is out of my setup and i just use latex
2020-09-27 02:54:14 <opal> not suitable for everyone ofc but works for me
2020-09-27 03:22:49 <atkka> opal: all i need is to be able to make pdfs and invoices, I would love a commandline option, would latex work well for that?
2020-09-27 03:23:25 <atkka> simple pdfs at that, a picture now and then and a couple columns of text
2020-09-27 03:27:15 <opal> latex is more geared toward typesetting. https://wiki.archlinux.org/index.php/PDF might list some useful utilities to do more like what you want
2020-09-27 03:27:34 <opal> i think the last thing i needed other than latex was something to edit / draw on pdfs
2020-09-27 03:35:08 <atkka> i just want to type some text in vi and run text through a program that formats it and makes a pdf
2020-09-27 03:35:15 <atkka> i'll check out the link
2020-09-27 03:36:33 <opal> i think ghostscript can do quite a bit
2020-09-27 05:20:33 <mps> atkka: I use perl (PDF::API2 module) to create invoices programatically
2020-09-27 05:23:28 <mps> but I still need libreoffice, when someone send me documents in 'office' formats to convert them
2020-09-27 05:33:26 <atkka> mps: thanks! that looks pretty awesome
2020-09-27 05:42:25 <mps> atkka: np
2020-09-27 05:43:20 <mps> atkka: what was reason you packaging sct you didn't use setcolortemperature from github
2020-09-27 05:44:24 <mps> btw, perl-pdf-api2 is in testing
2020-09-27 05:44:37 <atkka> I wanted practice packaging something and going through the submission process
2020-09-27 05:45:19 <mps> I think setcolortemperature is more complete, I have it in my local aports
2020-09-27 05:45:35 <atkka> is it a fork from sct?
2020-09-27 05:45:59 <mps> some kind of fork
2020-09-27 05:47:12 <mps> https://github.com/Tookmund/setcolortemperature
2020-09-27 05:47:18 <atkka> oh, I just used sct because I have been using it for years and it does what I need as simple as can be, sct 3000, didn't know there were more features to have
2020-09-27 05:48:58 <mps> eh, I also used it when I was on intels, and even prepared aports to push. but it doesn't work on my arm boxes so I forgot it till I saw your MR
2020-09-27 05:49:31 <atkka> does the sct I packaged not work on arm?
2020-09-27 05:49:42 <mps> no
2020-09-27 05:50:06 <atkka> huh, it builds for them though
2020-09-27 05:50:11 <atkka> does the github one?
2020-09-27 05:50:45 <mps> no
2020-09-27 05:51:00 <atkka> do you know why?
2020-09-27 05:51:06 <mps> this is not sct issue but drivers on ARMs
2020-09-27 05:51:31 <atkka> oh thats a bummer, I want to get a pinebook or something someday
2020-09-27 05:51:47 <mps> even xgamma doesn't work on ARM display drivers
2020-09-27 05:52:08 <atkka> do you know of an sct for the tty? right now I have to startx, sct 3000, kill dwm to get tty sct
2020-09-27 05:52:19 <mps> maybe it can work with panfrost or lima drivers, I didn't tested that
2020-09-27 05:53:35 <mps> no, I don't know. I use console on for rescue and to startx
2020-09-27 05:54:06 <atkka> what ARMs do you run? I have a couple pis that run alpine but would like an arm laptop
2020-09-27 05:54:18 <mps> I use X as better terminal multiplexer :)
2020-09-27 05:54:28 <atkka> i'm dvtm all the way
2020-09-27 05:54:40 <dysfun> i have a pinebook pro and i just use x
2020-09-27 05:55:01 <atkka> dysfun: can you run mainline alpine on it?
2020-09-27 05:55:12 <atkka> or is it postmarketOS
2020-09-27 05:55:31 <dysfun> it's currently some manjaro thing
2020-09-27 05:55:49 <mps> I have 3 chromebooks, one samsung Peach pi (exynos-5800 arm32), one Acer R13 (mediatek arm64) and one samsung (rk3399 arm64)
2020-09-27 05:55:50 <dysfun> i was intending to install alpine last night
2020-09-27 05:56:13 <atkka> i would love to see if it can run the stock alpine image
2020-09-27 05:56:34 <atkka> mps: good experiences with them?
2020-09-27 05:56:44 <atkka> battery life?
2020-09-27 05:57:12 <mps> atkka: yes, all 3 are fine, all have 4GB RAM
2020-09-27 05:57:16 <dysfun> the stock alpine arm image seems to come as a tarball and i'm not quite sure what i'm supposed to do with it - install it on FAT?
2020-09-27 05:57:42 <atkka> I think you extract it to a vfat sd card
2020-09-27 05:57:55 <dysfun> ah right
2020-09-27 05:58:02 <dysfun> well i'll probably give that a go later
2020-09-27 05:58:10 <atkka> dunno how you would "install" it on a pinebook though with the emmc
2020-09-27 05:58:39 <dysfun> well usually i just insert a usb pen
2020-09-27 05:59:18 <atkka> running it from ram on a usb drive? or installing it from a usb drives
2020-09-27 05:59:24 <dysfun> installing it
2020-09-27 05:59:48 <dysfun> i don't see how it would be any different from installing this machine?
2020-09-27 06:00:02 <atkka> oh cool, I didn't know the pinebook could do that, I though they needed special images and the usb-emmc flasher
2020-09-27 06:00:43 <dysfun> i don't think so. but one of my friends has been making SEL4 run on his, so i'll ask
2020-09-27 06:01:09 <atkka> any info you dig up or try would be good for the wiki
2020-09-27 06:01:21 <dysfun> yes, i think i am going to have to start some wiki pages
2020-09-27 06:01:58 <atkka> even just a works/doesn't work would be nice at this stage
2020-09-27 06:02:03 <dysfun> heh
2020-09-27 06:02:11 <dysfun> well i'm definitely going to try it
2020-09-27 06:02:16 <dysfun> manjaro is a bit shit
2020-09-27 06:02:37 <atkka> I find most distros lacking after using alpine
2020-09-27 06:02:48 <dysfun> yes
2020-09-27 06:03:27 <atkka> void is/was another good one but I haven't ran it for a long while
2020-09-27 06:03:34 <mps> one pmOS developer tested alpine on pinebook and told that alpine kernel doesn't work
2020-09-27 06:03:43 <atkka> ahh crap
2020-09-27 06:03:54 <mps> MartijnBraam: iirc
2020-09-27 06:04:50 <dysfun> i'm not afraid to build my own kernels, but it would be nice to have like a working alpine to do that from that's the same architecture, which i don't have
2020-09-27 06:05:14 <atkka> it would be nice to not need patched kernels at this stage
2020-09-27 06:05:30 <mps> dysfun: also u-boot from alpine is not made to work on pinebook
2020-09-27 06:05:31 <atkka> pmOS is alpine I suppose you could snag that kernel to start?
2020-09-27 06:05:47 <dysfun> mps: oh. what's needed?
2020-09-27 06:05:57 <mps> atkka: you take words from my mouth :)
2020-09-27 06:06:53 <mps> dysfun: we need to move gcc-cross to main to build u-boot for arm64 boards
2020-09-27 06:07:03 <dysfun> ah
2020-09-27 06:07:05 <mps> gcc-cross-embeded
2020-09-27 06:07:31 <atkka> can you do that locally?
2020-09-27 06:07:31 <dysfun> 
ACTION hates toolchains

2020-09-27 06:07:45 <mps> no one want to take resposibility for such move
2020-09-27 06:08:40 <mps> I plan to test this move locally and if it works maybe I will advocate move
2020-09-27 06:08:40 <atkka> well arm is becoming more popular, won't it have to happen at some point?
2020-09-27 06:08:42 <dysfun> oh no, we'll have to trick them into it
2020-09-27 06:08:56 <mps> though I already voted for move
2020-09-27 06:09:21 <dysfun> atkka: arm is also about to become part of nvidia, so the future might not be as bright as previously thought
2020-09-27 06:09:35 <atkka> oh yeah....thanks for the reminder
2020-09-27 06:09:51 <atkka> to riscv everyone!
2020-09-27 06:09:54 <dysfun> still, huge opportunities for riscv
2020-09-27 06:09:55 <dysfun> lol
2020-09-27 06:10:00 <mps> dysfun: yes, this is sad
2020-09-27 06:10:30 <mps> and yes, would be nice to have riscv soon
2020-09-27 06:10:55 <atkka> I need a low power laptop, efficient, fanless
2020-09-27 06:10:59 <atkka> open as possible
2020-09-27 06:11:00 <dysfun> still waiting for riscv hardware i'd consider worth using though
2020-09-27 06:11:07 <atkka> must run mainline, that is all I ask
2020-09-27 06:11:20 <dysfun> would be a lot of pain getting everything going on riscv, would want a nice machine
2020-09-27 06:11:55 <mps> atkka: your best bet is samsung exynos Peach Pi then
2020-09-27 06:12:15 <mps> all works out of the box, though it is arm32
2020-09-27 06:12:32 <mps> but this is most responsive machine I had
2020-09-27 06:12:38 <atkka> laptop?
2020-09-27 06:12:48 <mps> chromebook
2020-09-27 06:13:47 <atkka> XE500C12
2020-09-27 06:14:24 <mps> I forgot 'code' of it, but could be
2020-09-27 06:14:52 <atkka> oh no that was the intel one
2020-09-27 06:14:59 <mps> hm, no. this one is intel
2020-09-27 06:15:07 <atkka> samsung chromebook 2 though
2020-09-27 06:15:21 <atkka> XE503C32
2020-09-27 06:15:36 <atkka> that is the 13in 1080p, exynos
2020-09-27 06:15:39 <atkka> 4gb
2020-09-27 06:16:19 <mps> http://linux-exynos.org/wiki/Samsung_Chromebook_2_XE503C32
2020-09-27 06:18:17 <atkka> looks pretty nice, accelerated video playback with mpv?
2020-09-27 06:18:28 <atkka> encoding too?
2020-09-27 06:19:02 <mps> no, no accelerated but quite fine with mpv
2020-09-27 06:19:43 <atkka> I'm mainly concerned with powerconsumption
2020-09-27 06:19:48 <mps> didn't tried lima driver because I gave it to my son before 5 months
2020-09-27 06:20:29 <atkka> do you know if flatpak runs on arm stuff?
2020-09-27 06:21:00 <mps> well, power consumption depends on load, and ime there is not 'good' battery if the load is high
2020-09-27 06:21:20 <mps> uh, why anyone use flatpack
2020-09-27 06:22:12 <atkka> sandboxing of webapps, browsers (I don't want some apps to have access to my home folder for instance), libc issues
2020-09-27 06:22:32 <atkka> but mainly I'm tied to one damn application, wire messenger
2020-09-27 06:23:26 <mps> I use native apps only, so don't have experience with things like flatpack
2020-09-27 06:23:49 <mps> for sandboxing I use firejail or bubblewrap
2020-09-27 06:26:02 <atkka> yeah, I prefer native over anything really and just started trying flatpak last week. there are some nice use cases.
2020-09-27 06:26:29 <atkka> honestly, if I can get rid of wire I can get rid of X completely
2020-09-27 06:26:47 <dysfun> but x is a great way to have lots of terminals
2020-09-27 06:26:59 <atkka> i use DVTM
2020-09-27 06:27:17 <atkka> dwm but for the framebuffer, works very well
2020-09-27 06:27:40 <atkka> mouse support if you want it, copy and paste, works great
2020-09-27 06:28:08 <dysfun> i never got along with dwm particularly
2020-09-27 06:28:53 <dysfun> quite tempted to try wayland, but it looks rather like i'll end up writing my own compositor if i go that way, as if i didn't have enough projects to ignore
2020-09-27 06:30:03 <atkka> I have too much time right now
2020-09-27 06:30:11 <afontain_> I'd have a hard time surviving without Firefox.
2020-09-27 06:30:19 <dysfun> 
ACTION would also need firefox

2020-09-27 06:30:23 <afontain_> or maybe without qutebrowser
2020-09-27 06:30:34 <helby> qute makes no sense
2020-09-27 06:30:37 <atkka> you still have firefox, browsh
2020-09-27 06:31:05 <atkka> headless firefox renders pages for your terminal
2020-09-27 06:31:16 <dysfun> and how about video?
2020-09-27 06:31:24 <afontain_> huh
2020-09-27 06:31:27 <atkka> works fine, accelerated even
2020-09-27 06:31:35 <dysfun> hmm
2020-09-27 06:32:23 <afontain_> do you have a screenshot?
2020-09-27 06:32:50 <dysfun> https://www.brow.sh/
2020-09-27 06:32:50 <atkka> not currently, I can get one tomorrow though
2020-09-27 06:33:02 <atkka> I'm on dwm right now
2020-09-27 06:33:14 <dysfun> apparently you can ssh to their demo
2020-09-27 06:33:21 <atkka> is it back up?
2020-09-27 06:33:29 <dysfun> no idea, i'm just looking at their site the first time
2020-09-27 06:34:45 <afontain_> well, the ssh demo seems broken
2020-09-27 06:34:45 <atkka> pictures, gifs, pdfs, videos all work fine without x or wayland
2020-09-27 06:35:23 <atkka> there are frambuffer window managers as well
2020-09-27 06:36:55 <afontain_> the https demo seems down as well
2020-09-27 06:38:11 <atkka> yeah they were down a month ago when I checked
2020-09-27 06:38:27 <atkka> netsurf is another graphical browser you can use
2020-09-27 06:38:48 <atkka> w3m can display images and open videos/yt with mpv
2020-09-27 06:39:45 <atkka> I plan on packaging some framebuffer apps for alpine like FIM
2020-09-27 06:40:07 <atkka> want to get a couple packaged here tomorrow or early in the week
2020-09-27 06:40:42 <atkka> fbpdf is another that is missing
2020-09-27 06:41:26 <mps> atkka: sct pulls some -dev packages on install
2020-09-27 06:42:44 <atkka> https://github.com/caramelli/higfxback/wiki/Linux-Framebuffer
2020-09-27 06:43:14 <atkka> mps: libX11-dev libxrandr-dev?
2020-09-27 06:43:50 <atkka> is there something that needs to be changed?
2020-09-27 06:46:52 <mps> yes, I will look later when finish with some things I work right now
2020-09-27 06:52:56 <atkka> ok, sounds good. please let me know.
2020-09-27 07:05:27 <atkka> mps: should I have them under depends-dev or something other than depends?
2020-09-27 07:05:37 <atkka> will they get removed that way?
2020-09-27 07:07:55 <mps> makedepends
2020-09-27 07:09:56 <mps>  makedepends="libx11-dev libxrandr-dev"
2020-09-27 07:10:29 <mps> and you don't need -I'files'
2020-09-27 07:11:19 <mps> atkka: https://tpaste.us/K5Ob :)
2020-09-27 07:23:15 <atkka> mps: thank you, I'll get it fixed tomorrow!
2020-09-27 07:24:19 <mps> atkka: can I propose to take my APKBUILD and put yourself as maintainer, so we can work on it together?
2020-09-27 07:24:30 <mps> you take*
2020-09-27 07:25:56 <atkka> sure, I would like that. It would be a good learning experience
2020-09-27 07:26:45 <mps> nice
2020-09-27 07:44:33 <atkka> well I'm off to bed, thanks for everything
2020-09-27 07:45:11 <atkka> mps: I'll be around tomorrow, night
2020-09-27 09:06:20 <nono> Hi all
2020-09-27 09:08:44 <nono> Do you know if bcache should work as it in Alpine Linux? I need to manually register my devices each reboot to trigger udev rules. Maybe the package lacks some features/hook for initrd ?
2020-09-27 11:15:29 <CactiChameleon9> Hello, where can I report issues
2020-09-27 11:15:52 <CactiChameleon9> I am running the rolling relese alpine
2020-09-27 11:16:26 <CactiChameleon9> I have pickedup a couple of issues with dependencies and an issue with gnome-web
2020-09-27 11:17:42 <mps> CactiChameleon9: gitlab.alpinelinux.org, choose project and open new issue report
2020-09-27 11:17:51 <CactiChameleon9> OK, thanks
2020-09-27 11:18:05 <mps> it is edge, not rolling
2020-09-27 11:18:50 <CactiChameleon9> Yeah... I forgot the name
2020-09-27 11:18:53 <CactiChameleon9> Sorry
2020-09-27 11:19:07 <CactiChameleon9> But it is rolling release, so I though that was better than beta
2020-09-27 11:20:01 <Cogitri> Is your issue with GNOME Web that it doesn't load some websites?
2020-09-27 11:20:06 <CactiChameleon9> Yeah
2020-09-27 11:20:10 <CactiChameleon9> Not much seems to load
2020-09-27 11:20:15 <Cogitri> Because that's a known bug - but unfortunately one with an unclear cause
2020-09-27 11:20:22 <CactiChameleon9> hmm...
2020-09-27 11:20:25 <CactiChameleon9> OK
2020-09-27 11:20:33 <CactiChameleon9> and where should I report a dpependencies error
2020-09-27 11:20:35 <Cogitri> Yeah, all sites which use more than a minimal amount of JS are problematic
2020-09-27 11:20:48 <Cogitri> For some reason webkit2gtk has some issues on musl
2020-09-27 11:21:19 <Cogitri> Feel free to report the dependencies error on https://gitlab.alpinelinux.org/alpine/aports/issues/new
2020-09-27 11:21:32 <Cogitri> Thanks for telling us about it :)
2020-09-27 11:21:46 <CactiChameleon9> Yeah the dependency error is for polari
2020-09-27 11:22:15 <CactiChameleon9> It requires the telepathy packages listed here to work properly https://aur.archlinux.org/packages/polari-git/
2020-09-27 11:24:10 <CactiChameleon9> What is aport?
2020-09-27 11:24:55 <mps> CactiChameleon9: no, alpine don't have rolling release. edge is development branch
2020-09-27 11:24:56 <CactiChameleon9> oh.. Alpine packages build scripts. Is that the right place to state a install dependcy issue then?
2020-09-27 11:25:32 <CactiChameleon9> ok, thanks mps. I am still very new to alpine, but so far it has been amazing
2020-09-27 11:25:33 <mps> and as development branch there can be (and there are) some problems
2020-09-27 11:26:12 <mps> CactiChameleon9: np :) and it is good to report issues/bugs
2020-09-27 11:26:48 <CactiChameleon9> is this a good title? "Polari more runtime dependancies needed"
2020-09-27 11:27:25 <CactiChameleon9> and should I mention edge, or not
2020-09-27 11:27:48 <mps> keep in mind that we 'want' minimum dependencies
2020-09-27 11:27:55 <CactiChameleon9> yeah...
2020-09-27 11:28:02 <mps> only those which are essential
2020-09-27 11:28:09 <CactiChameleon9> should I try ininstalling each one to find the main cause of the issue?
2020-09-27 11:28:19 <mps> i.e. package doesn't work without them
2020-09-27 11:28:43 <CactiChameleon9> It refused to add an irc network - which is the point of polari
2020-09-27 11:30:56 <mps> sure, you can open bug/issue and someone will look at it
2020-09-27 11:33:18 <CactiChameleon9> Done
2020-09-27 11:33:20 <CactiChameleon9> https://gitlab.alpinelinux.org/alpine/aports/-/issues/11981
2020-09-27 11:33:25 <CactiChameleon9> Thanks for the help
2020-09-27 11:33:30 <Cogitri> Thanks 👍
2020-09-27 11:37:00 <CactiChameleon9> Cogitri: Where is the gnome web issue?
2020-09-27 11:38:24 <Cogitri> Hm, seems like I haven't create an issue for it yet (only talked about it on IRC)
2020-09-27 11:38:38 <Cogitri> Would be good to have a tracker issue for that too
2020-09-27 12:33:05 <ikke> It certainly would
2020-09-27 12:40:34 <CactiChameleon9> I have a question, what the best method to get proper graphics driver in alpine (intel)
2020-09-27 12:42:15 <Cogitri> Installing xf86-video-intel should be sufficient
2020-09-27 12:43:15 <CactiChameleon9> OK
2020-09-27 12:43:35 <CactiChameleon9> My main reason for asking is android studio's ui rendering is refusing to work
2020-09-27 12:43:47 <CactiChameleon9> with the error "com.android.tools.idea.layoutlib.RenderingException: The rendering library could not be initialized"
2020-09-27 12:44:06 <helby> I don't use xf86, just mesa dri
2020-09-27 12:44:19 <CactiChameleon9> how do I do that?
2020-09-27 12:44:47 <CactiChameleon9> I think some distro's state not to uuse xf86 drivers for intel so it may help
2020-09-27 12:44:50 <helby> I am using sway/wayland, not sure if xorg requires xf86, long time did not use that shit
2020-09-27 12:44:56 <helby> but probably does not need
2020-09-27 12:45:09 <CactiChameleon9> I am using gnome, so it may be wayland?
2020-09-27 12:45:17 <helby> probably yes
2020-09-27 12:45:52 <helby> for me, sway will not start without mesa-dri
2020-09-27 12:45:58 <helby> it's even in wiki somehwere
2020-09-27 12:46:07 <helby> but wiki is mess anyway
2020-09-27 12:47:12 <CactiChameleon9> I have removed xf86-video-intel
2020-09-27 12:47:32 <CactiChameleon9> Should I reinstall libva-intel-driver along with the mesa that was uninstalled?
2020-09-27 12:49:18 <CactiChameleon9> OK, I will try that and see
2020-09-27 12:49:36 <CactiChameleon9> I probably need to rebot
2020-09-27 12:49:40 <CactiChameleon9> *reboot
2020-09-27 12:56:23 <helby> so how it looks now with mesa?
2020-09-27 12:57:14 <MY-R> CactiChameleon9, if your CPU is Broadwell or newer then can install package "intel-media-driver" too
2020-09-27 12:57:19 <helby> for fuck sake I want to install some other distro on another disk for case I will fuck up something, but I can't find something useful, everyehre just a lot of crap
2020-09-27 12:58:55 <gillesmauve> helby: xorg is better than wayland
2020-09-27 12:59:10 <helby> gillesmauve: congratz then
2020-09-27 12:59:19 <helby> gillesmauve: I will not that touch more
2020-09-27 12:59:26 <gillesmauve> doesn't have anything to do with me, lmao
2020-09-27 12:59:36 <helby> how not? u probably using it
2020-09-27 13:00:03 <gillesmauve> that wouldn't have something to do with someone making the claim that it's better
2020-09-27 13:00:13 <helby> how?
2020-09-27 13:00:31 <helby> if u use it, it's better than wayland, then I just congratulate to you for using it as a better system :)
2020-09-27 13:00:58 <helby> I got fed up with fuckering with all x files long time ago
2020-09-27 13:01:15 <helby> sway works well for me, without any issue for my usage
2020-09-27 13:01:39 <helby> and don't be bad, just tired several distros and openbsd and all deleted withing a minute
2020-09-27 13:01:42 <CactiChameleon9> to be honest I dont get the difference between the two, I believe that wayland may have better sandboxing or something like that making it a little harder to work with
2020-09-27 13:01:59 <helby> for me it's huge difference
2020-09-27 13:02:05 <helby> I install sway and that crap works
2020-09-27 13:02:08 <Cogitri> The nice thing about Wayland is better HiDPI support
2020-09-27 13:02:16 <CactiChameleon9> I forgot about that
2020-09-27 13:02:22 <helby> without spenind a weekend in xinit xresources and tons of other xfiles
2020-09-27 13:02:28 <CactiChameleon9> Why can they not just put that in xorg
2020-09-27 13:03:56 <helby> forget xorg
2020-09-27 13:04:05 <helby> it's long time legacy already ;)
2020-09-27 13:06:51 <CactiChameleon9> OK, I have installed intel-media-driver, and rebooted and I will see if it works now
2020-09-27 13:07:03 <CactiChameleon9> nope....
2020-09-27 13:07:47 <CactiChameleon9> Maybe its the openjdk instead of oracle java?
2020-09-27 13:08:04 <CactiChameleon9> I cannot get the install to work from the wiki however
2020-09-27 13:08:51 <helby> check archlinux wiki
2020-09-27 13:09:01 <CactiChameleon9> what on?
2020-09-27 13:09:07 <helby> whatever you need
2020-09-27 13:09:12 <helby> alpine wiki is junk
2020-09-27 13:09:23 <CactiChameleon9> android studio, a doubt they have a section on it
2020-09-27 13:09:56 <gillesmauve> Cogitri: what's the core of hidpi support?
2020-09-27 13:10:03 <helby> there is little bit about it
2020-09-27 13:10:16 <helby> but still first get right your intel shit
2020-09-27 13:10:23 <gillesmauve> helby: you seem like a chore that thinks the world revolves around your specific use case
2020-09-27 13:10:40 <gillesmauve> i'm glad that you're going to be whining into ddevault's ear instead of anyone worth keeping around
2020-09-27 13:10:50 <helby> gillesmauve: yes, I seem
2020-09-27 13:10:57 <gillesmauve> everyone knows he needs a couple pegs knocked off of his board, so at least there's someone out there to do it
2020-09-27 13:11:17 <helby> lol
2020-09-27 13:11:21 <helby> why to him?
2020-09-27 13:11:24 <gillesmauve> either that or you'll be on os x and put in your place in the next year
2020-09-27 13:11:41 <helby> wtf is your story about?
2020-09-27 13:11:56 <gillesmauve> the book's purpose isn't to tell you about the story
2020-09-27 13:11:59 <gillesmauve> it's to tell you the story
2020-09-27 13:12:15 <gillesmauve> it's your job to learn what you want, not the book's to teach you
2020-09-27 13:13:36 <helby> not sure what you want to tell me, but okay
2020-09-27 13:13:58 <Cogitri> gillesmauve: E.g. mixed resolutions just work on Wayland
2020-09-27 13:14:19 <gillesmauve> Cogitri: xrandr never gave me issues on x
2020-09-27 13:14:42 <gillesmauve> perhaps i live in 2020 and you guys live in 1984
2020-09-27 13:15:19 <gillesmauve> this reminds me of the conversation most commonly had about x versus sprayandprayland: omg the screen tear
2020-09-27 13:15:28 <gillesmauve> which was an nvidia driver problem, not an x problem
2020-09-27 13:15:57 <helby> it seems you are talking with yourself dude, u only one fighting xorg-wayland :D
2020-09-27 13:16:21 <helby> rest seems to be okay with whatever they are using
2020-09-27 13:17:22 <Cogitri> I had screen tearing on amd and intel too by default
2020-09-27 13:17:28 <gillesmauve> "you're the only one," says the guy who was shitting on x earlier over the course of a dozen lines
2020-09-27 13:17:41 <gillesmauve> cogitri: still not an x problem
2020-09-27 13:17:50 <gillesmauve> nvidia set the standard and everyone followed suit
2020-09-27 13:18:12 <gillesmauve> graphics drivers? problems with graphics? nah that's x's problem, definitely doesn't happen on every single computer ever made
2020-09-27 13:18:34 <gillesmauve> blame upstream, must be their problem, since they obviously have access to any of our blobbed garbage, amirite
2020-09-27 13:18:56 <gillesmauve> how do i remotely set off 100 autoplaying videos of linus flipping off nvidia
2020-09-27 13:21:30 <CactiChameleon9> Question... what does ubuntu install intel graphics and java related by default?
2020-09-27 13:21:42 <CactiChameleon9> (that alpine does not)
2020-09-27 13:22:04 <helby> probably just 500 extra packages :0
2020-09-27 13:22:12 <CactiChameleon9> Thats a valid point
2020-09-27 13:22:27 <CactiChameleon9> I havent had this issue with arch either though
2020-09-27 13:22:29 <helby> not sure,but they install probably all intel crap
2020-09-27 13:22:45 <helby> for older, newer processors and so on, so something always works
2020-09-27 13:22:54 <helby> what's your issue btw?
2020-09-27 13:23:14 <CactiChameleon9> Android studio's ui rendering function is erroring
2020-09-27 13:23:35 <CactiChameleon9> com.android.tools.idea.layoutlib.RenderingException: The rendering library could not be initialized
2020-09-27 13:23:50 <helby> good
2020-09-27 13:23:54 <helby> dont use android studio then
2020-09-27 13:23:59 <helby> sell smartphone too ;0
2020-09-27 13:24:04 <helby> it's a signal from above
2020-09-27 13:24:05 <CactiChameleon9> I am very used to it thanks...
2020-09-27 13:24:17 <helby> just joking
2020-09-27 13:24:19 <CactiChameleon9> And its the only google product I use
2020-09-27 13:24:24 <CactiChameleon9> (degoogle lineage)
2020-09-27 13:24:31 <CactiChameleon9> ik
2020-09-27 13:25:28 <CactiChameleon9> alpine linux really isnt a massive desktop distro is it
2020-09-27 13:25:52 <helby> probably not, I use it, but I m oaky with minimal shit, that's the reason I use it
2020-09-27 13:25:57 <CactiChameleon9> yeah
2020-09-27 13:26:01 <CactiChameleon9> I love it so far
2020-09-27 13:26:11 <helby> of course when I tried some crap I always get some kind of error like you
2020-09-27 13:26:20 <helby> bc. I miss who knowswhat
2020-09-27 13:26:40 <CactiChameleon9> I could try adding a ubuntu strata (bedrock linux) and seing if it somehow helps
2020-09-27 13:26:40 <helby> no dbus, no greater session and many other crap. so I don't expect that soe bloat will work well
2020-09-27 13:27:14 <helby> if I need some bloat, I would use arch probably
2020-09-27 13:27:29 <CactiChameleon9> also, I try much more than ides... my computer is my entertainment :D
2020-09-27 13:27:41 <CactiChameleon9> Yeah... but apk is amazing
2020-09-27 13:27:48 <helby> yes it is
2020-09-27 13:27:50 <CactiChameleon9> and it runs so lightweight
2020-09-27 13:27:52 <helby> not much bullshit
2020-09-27 13:28:05 <helby> everyting is done fast without nonsense, like it too
2020-09-27 13:28:12 <CactiChameleon9> I will try adding a ubuntu strata and seeing what happens
2020-09-27 13:28:56 <CactiChameleon9> I should be doing work now....
2020-09-27 13:28:59 <CactiChameleon9> :D]
2020-09-27 13:29:13 <Cogitri> Oh, you're using Bedrock Linux? How does that work for you?
2020-09-27 13:30:03 <CactiChameleon9> Its is good
2020-09-27 13:30:54 <CactiChameleon9> Arch linux does not intergrate too well, and you have to write some systemd services if you want to run a service from your non init strata, but apart from that great
2020-09-27 13:31:37 <Cogitri> Neat
2020-09-27 13:31:43 <CactiChameleon9> Bedrock linux is the reason I am trying alpine, because it is so easy to install and switch too and then disable my ubuntu
2020-09-27 13:31:56 <tehcloud> never heard of this until now
2020-09-27 13:32:00 <tehcloud> very interesting...
2020-09-27 13:32:14 <CactiChameleon9> It basicually intergrates lots of distros,
2020-09-27 13:32:46 <CactiChameleon9> I dont use it like the example, I just have it for one off programs or for easily setting up a chroot for compiling some program which i can delete later for neatness
2020-09-27 13:33:15 <tehcloud> ahh... so it throws distros into chroots?
2020-09-27 13:33:21 <CactiChameleon9> fancy chroots
2020-09-27 13:33:35 <CactiChameleon9> it then links files to different places and other things to make it all play nice
2020-09-27 13:34:07 <CactiChameleon9> It makes it like, on the surface, running one distro
2020-09-27 13:34:14 <helby> it can be cool mess after two days of usage ;)
2020-09-27 13:34:22 <CactiChameleon9> Mine is file
2020-09-27 13:34:26 <tehcloud> I am typing from Arch right now
2020-09-27 13:34:26 <CactiChameleon9> *fine
2020-09-27 13:34:33 <tehcloud> I'm only on Arch on this machine because of nvidia
2020-09-27 13:34:37 <CactiChameleon9> Its irriversable btw
2020-09-27 13:34:48 <tehcloud> but I have an AMD GPU in the mail and will move this machine to Alpine soon
2020-09-27 13:35:05 <helby> tehcloud: $5 for arch users, sorry
2020-09-27 13:35:08 <tehcloud> sort of want to test bedrock now
2020-09-27 13:35:21 <CactiChameleon9> so you cannot easily undo bedrock installation, apart from a full reinstall of something
2020-09-27 13:35:38 <CactiChameleon9> I did enjoy arch linux, but I got bored
2020-09-27 13:35:49 <tehcloud> I can't say I'm a huge fan of Arch
2020-09-27 13:35:53 <helby> yep,that's the reason for swith :)
2020-09-27 13:35:55 <tehcloud> but on this machine I needed a glibc distro
2020-09-27 13:36:00 <CactiChameleon9> I had fixed everything and it worked great, so I then moved to bedrock linux
2020-09-27 13:36:02 <tehcloud> and Debian is fucked nowadays
2020-09-27 13:36:17 <helby> still arch is one of the best from my view
2020-09-27 13:36:31 <helby> everything there, quality wiki, works well,new soft
2020-09-27 13:36:31 <CactiChameleon9> Has anyone here tried gentoo properly
2020-09-27 13:36:32 <tehcloud> it gets the job done and hasn't broken too badly
2020-09-27 13:36:41 <tehcloud> CactiChameleon9, I ran Gentoo for 3 years
2020-09-27 13:36:46 <tehcloud> but that was over a decade ago
2020-09-27 13:36:48 <helby> and importantnotmuch fucked custom tweaks
2020-09-27 13:36:52 <tehcloud> got tired of compiling everything
2020-09-27 13:36:59 <CactiChameleon9> I tried but struggled. It is harder than arch
2020-09-27 13:37:14 <CactiChameleon9> I have no idea why arch gets any credit for being dificult
2020-09-27 13:37:22 <tehcloud> I ran FreeBSD as my main OS for a long time back in the really old days
2020-09-27 13:37:23 <helby> bc.installation
2020-09-27 13:37:27 <tehcloud> it was awesome back then
2020-09-27 13:37:28 <CactiChameleon9> I installed it first try when I was still a noob
2020-09-27 13:37:30 <helby> nobody remember all steps
2020-09-27 13:37:42 <CactiChameleon9> freeBSD?
2020-09-27 13:37:43 <helby> yes, but u have to follow copy/paste
2020-09-27 13:38:00 <CactiChameleon9> Yeah, but compared to gentoo it is nothing
2020-09-27 13:38:07 <tehcloud> Gentoo basically started out as an attempt to bring FreeBSD's ports system to Linux
2020-09-27 13:38:08 <helby> oh yes freebsd, the most messy system
2020-09-27 13:38:25 <CactiChameleon9> Then again I did try to compile a custom kernal, maybe without it would have been easier
2020-09-27 13:38:42 <helby> I spend one night with compiling too
2020-09-27 13:38:51 <helby> after then, never touched it again :)
2020-09-27 13:39:10 <CactiChameleon9> I have done that... with a pi zero compiling a minecraft server software
2020-09-27 13:39:19 <CactiChameleon9> because it was not compatable with armv6
2020-09-27 13:39:40 <CactiChameleon9> That is painful, the server was the most lightweight one also
2020-09-27 13:39:43 <helby> it's relatively okay to compile something special if needed, but compile whole system permanently ...
2020-09-27 13:39:51 <CactiChameleon9> valid
2020-09-27 13:40:03 <helby> like lovely browser ... one need a vacation for that
2020-09-27 13:40:21 <CactiChameleon9> ?
2020-09-27 13:48:26 <CactiChameleon9> It works now
2020-09-27 13:48:41 <CactiChameleon9> I installed ubuntu, and the java 8 jdk from ubuntu and it seems to work
2020-09-27 13:50:13 <CactiChameleon9> And android studio does not complain about openjdk
2020-09-27 13:50:22 <CactiChameleon9> maybe ubuntu uses oracle's version?
2020-09-27 13:50:28 <CactiChameleon9> idk tbh
2020-09-27 13:50:34 <CactiChameleon9> Thanks for the help
2020-09-27 13:51:39 <PureTryOut[m]> I'm reinstalling Alpine on my desktop which means NetworkManager again refuses to take control over my network devices. eth0 does not appear in `/etc/network/interfaces` and the logs tell me NetworkManager knows about it, but it doesn't change the status from unmanaged to managed.
2020-09-27 13:51:40 <PureTryOut[m]> Cogitri: any chance you can help me out?
2020-09-27 13:59:32 <Cogitri> Hm, maybe some other service already manages it?
2020-09-27 14:52:07 <CactiChameleon9> no libreoffice?
2020-09-27 14:52:41 <ikke> CactiChameleon9: fails to build atm with gcc10
2020-09-27 14:52:56 <CactiChameleon9> whatever that means... OK
2020-09-27 14:53:34 <ikke> We upgraded the compiler to gcc 10, libreoffice now won't build until it's fixed
2020-09-27 14:54:24 <CactiChameleon9> Oh.. OK
2020-09-27 14:54:45 <helby> is anything other than intel, i915 firmware somehow usefull for old thinkpad? anybody know?
2020-09-27 18:43:01 <abenz_> PureTryOut[m]: I thought everyone on alpine used setup-interfaces :)
2020-09-27 20:17:49 <gillesmauve> PureTryOut: how about not using networkmanager
2020-09-27 20:19:09 <Cogitri> That doesn't integrate into the UI then though
2020-09-27 20:19:22 <Cogitri> I think disabling the network service should probably fix it
2020-09-27 20:42:42 <PureTryOut[m]> gillesmauve: that is a completely useless suggestion, please don't do that
2020-09-27 20:42:45 <PureTryOut[m]> Cogitri: no it doesn't, that's an obvious solution for which I already made sure it wasn't disabled 😉
2020-09-27 20:51:10 <Cogitri> Hm, and there really aren't any other network related services which could grab the device?
2020-09-27 22:41:32 <opal> gillesmauve you're in a distro support channel that happens to package networkmanager among other packages perhaps you or i dont like. get used to it
2020-09-27 22:45:54 <tehcloud> what's funny about this is that with a lot of other distros you get users requesting more packages
2020-09-27 22:46:02 <tehcloud> with Alpine you get users trying to get rid of packages
2020-09-27 22:56:08 <opal> i was gonna contribute gnucash as a package once i figured out how to compile the latest version
2020-09-27 22:56:14 <opal> then i stopped using gnucash so i have no incentive lol
2020-09-27 22:57:07 <opal> i might revisit that just because it's nice to have a library of software that plays nicely with musl-libc
2020-09-27 23:08:43 <tehcloud> I write all my software against musl to begin with nowadays
2020-09-27 23:09:02 <tehcloud> it has exposed a lot of my shitty programming techniques
2020-09-27 23:10:33 <gillesmauve> opal: this isn't a support channel, and you don't pay for alpine
2020-09-27 23:10:55 <gillesmauve> package maintainers are not representatives of a brand name
2020-09-27 23:11:04 <gillesmauve> don't let your brain rot like the rest of your cohort, opal
2020-09-27 23:11:23 <gillesmauve> puretryout: quit using networkmanager
2020-09-27 23:11:38 <opal> community support is still support.
2020-09-27 23:11:48 <gillesmauve> even your meaninglessly exalted "production environments" only use it when they haven't learned how to properly administer their network
2020-09-27 23:11:54 <gillesmauve> opal: you know what you meant.
2020-09-27 23:12:13 <opal> what i meant is YOU are antagonising another user's choice to use the software they wish
2020-09-27 23:12:28 <opal> you are not being helpful here, you are not adding anything of value here
2020-09-27 23:12:37 <tehcloud> is this something worth getting worked up over?
2020-09-27 23:12:43 <gillesmauve> https://i.imgur.com/Jjbomam.png look at this dipshit
2020-09-27 23:13:17 <gillesmauve> wouldn't the mere existence of someone this capable of being that far up their own ass just make you seethe with rage every time they talked in a channel, knowing how many of your fellow users are going to have to deal with his shit-eating obsessions?
2020-09-27 23:13:39 <tehcloud> gillesmauve, who is that?
2020-09-27 23:13:45 <gillesmauve> yeah, who is that?
2020-09-27 23:14:05 <gillesmauve> a question that likely rocks them to their core every time they're even so much as grazed by it
2020-09-27 23:14:59 <gillesmauve> how dare you, tehcloud from freenode, not bow and quiver in your loins for the SIMPLE OPPORTUNITY to aid them on their quest to flash off all forms of personality and replace them with cold, dead "support"
2020-09-27 23:15:22 <tehcloud> sorry, I should have known better!
2020-09-27 23:15:33 <gillesmauve> at least opal will feel like they fit the fuck in somewhere for once when they get to unduly serve His One True Opulence
2020-09-27 23:16:30 <gillesmauve> Oh Utter PureTryOut, thank you for your divine presence in my meager s-support channel! forgive me my sins, for i am born unwell! forged in error!
2020-09-27 23:32:41 <hxx> Hi anyone?
2020-09-27 23:50:41 <hxx> Noob here, oes anyone know the 'good' repos for a clean install of kde plasma? (AL 3.12)
2020-09-28 00:33:12 <gildarts> Hello! I was hoping for some help getting alpine installed on my desktop. I’m booting to a usb stick with alpine loaded, but after the “Loading hardware drivers” message the screen goes wonky and it appears to stop loading.
2020-09-28 00:34:18 <Hello71> try adding nomodeset to kernel parameters
2020-09-28 00:35:48 <gildarts> It appears that is already there...
2020-09-28 00:41:13 <gildarts> Grub setting is Linux ... nomodeset
2020-09-28 00:52:10 <hxx> Hi, is kde plasma an official package for AL 3.12? i am geting a lot of dependencies error when running apk
2020-09-28 00:52:29 <Hello71> hxx: don't randomly mix repos
2020-09-28 00:52:39 <Hello71> gildarts: what?
2020-09-28 00:55:06 <gildarts> The boot config in grub already included nomodeset at the end of the Linux line.
2020-09-28 00:59:26 <Hello71> so you mean linux, not Linux?
2020-09-28 01:00:40 <hxx> Hello71:beg your pardon?
2020-09-28 01:01:23 <hxx> What made you think my request for help is random?
2020-09-28 01:05:05 <hxx> Oops sorry hello71, I misread your sufggestion, a bit tired
2020-09-28 01:05:29 <gildarts> Yes sorry, I’m on a phone that keeps capitalizing it.
2020-09-28 01:06:23 <Hello71> unless you know what you're doing and are ready to pick up the pieces, use only repos from a single alpine version
2020-09-28 01:06:44 <Hello71> gildarts: can you be more specific about "wonky"
2020-09-28 01:07:55 <hxx> gotcha. So I should have only one repo ref. in /etc/apk/repositories? pls confirm
2020-09-28 01:09:59 <hxx> This repo, http://dl-cdn.alpinelinux.org/alpine/v3.12/main/x86_64/, is labelled 'official', so I should just that one then?
2020-09-28 01:19:06 <hxx> *use
2020-09-28 01:20:59 <gildarts> Not sure how else to describe it. https://usercontent.irccloud-cdn.com/file/3s8SUuSr/1601256039.JPG
2020-09-28 01:21:06 <gildarts> Hello71: ^
2020-09-28 01:22:53 <Hello71> gildarts: what graphics card do you have
2020-09-28 01:23:16 <Hello71> hxx: you also need community for plasma
2020-09-28 01:23:30 <Hello71> but you need to pick if you want 3.12 or edge, you cannot select both
2020-09-28 01:23:33 <hxx> hi
2020-09-28 01:23:49 <hxx> k
2020-09-28 01:24:34 <gildarts> Hello71: have an Nvidia RTX 2080. This is originally a windows/gaming box, but was trying to install Alpine to learn and try to get away from Windows as much as I can.
2020-09-28 01:24:57 <hxx> thx Hello71, I will try that tomorrow
2020-09-28 01:25:03 <hxx> bye now
2020-09-28 01:26:20 <Hello71> gildarts: nvidia and linux don't go well together. especially nvidia and any sort of configuration that isn't completely conventional
2020-09-28 01:26:58 <tehcloud> 
ACTION has a Radeon in the mail right now

2020-09-28 01:26:58 <Hello71> you may have better luck without nomodeset
2020-09-28 01:27:33 <Hello71> well, it would be more precise to say that nvidia and open source don't go well together
2020-09-28 01:27:54 <tehcloud> a decade ago you basically needed nvidia for their closed source driver because everything else sucked
2020-09-28 01:28:00 <tehcloud> now the tables have turned completely
2020-09-28 01:28:37 <gildarts> Makes sense. Will give it a try and if changing that doesn’t work, will try Ubuntu for the metal and try Alpine in a VM.
2020-09-28 02:44:11 <Hello71> radeon was already pretty decent for r600+ in 2010, even if not as fast as fglrx
2020-09-28 06:57:03 <Milka> I'm wondering, why resume= kernel parameter only accepts /dev/disk nodes, but not UUID ('Stat: UUID=xxxx no such file or directory')?
2020-09-28 06:58:24 <afontain_> I've seen that in the nixOS config, you must also use /dev/disk/by-…
2020-09-28 07:06:46 <Milka> afontain_: Is it possible in Alpine?
2020-09-28 07:08:30 <ikke> If I read https://gitlab.alpinelinux.org/alpine/mkinitfs/-/blob/master/initramfs-init.in#L489 correctly, it seems resume does not even accept an argument?
2020-09-28 07:09:01 <ikke> Or maybe it's just the kernel that reads it
2020-09-28 07:42:52 <afontain_> I don't know, I tried a year ago, and didn't succeed. Now, in-between, kernel support has been added for it
2020-09-28 07:47:24 <Milka> ikke: I think it's just for kernel
2020-09-28 07:53:47 <Milka> Is it possible to catch suspend/resume events with busybox acpid?
2020-09-28 09:16:37 <hxx> Good morning - I have installed KDE plasma on AL 3.18, via http://dl-cdn.alpinelinux.org/alpine/v3.12/, no errors to report. I notice that one can start plasma either through wayland or x11, how do you set it up?
2020-09-28 10:12:39 <hxx> and how do i do that?
2020-09-28 10:42:08 <Cogitri> You should be able to choose what session to start in SDDM
2020-09-28 11:20:56 <hxx> accordinfg to the docs, installing kde plasma metapackage via apk also installed and configured sddm but when booting, sddm prompt is not displayed
2020-09-28 11:22:04 <Cogitri> You need to enable the sddm service with `rc-update`
2020-09-28 11:22:45 <Cogitri> rc-update add sddm default
2020-09-28 11:23:55 <hxx> k
2020-09-28 11:24:03 <hxx> trying now
2020-09-28 11:29:40 <hxx> k - now i m getting sddm prompt but only user profiles loaded are 'Linux user'...
2020-09-28 11:32:08 <hxx> managed to enter user login manually but getting a dark screen after successful login
2020-09-28 11:33:36 <hxx> but I remember seeing a powerdevil message about 'cannot connect to ck' when i ran xinit and ended up with a similar dark screen
2020-09-28 11:36:13 <hxx> it seems to be related to my screen brightness
2020-09-28 14:24:46 <hxx> Hello - i have re-installed AL3.12 + xfce, as described here: https://wiki.alpinelinux.org/wiki/Xfce_Setup, all services are starting up w\o any noticeable errors but lightdm is not launching
2020-09-28 14:25:58 <Hello71> did you run rc-update add lightdm
2020-09-28 14:30:52 <hxx> Hi Hello71 - yup
2020-09-28 14:31:26 <wsinatra> did you check the /var/log/lightdm directory? there's a Xorg log, and a couple of other log files that should tell you what's going on
2020-09-28 14:34:39 <jmondi> Hello, maybe a stupid question, but: I'm booting an Alpine 3.12-aarch64 minirootfs on NFS from an embedded device, I get to the login but user 'root' gets refused with "Login incorrect"
2020-09-28 14:35:12 <jmondi> I haven't found any mention of other users being enabled in the wiki pages
2020-09-28 14:35:52 <Hello71> also /var/log/messages
2020-09-28 14:36:03 <Hello71> jmondi: securetty
2020-09-28 14:38:56 <hxx> the only errors i have are failures to load module 'nv'and 'vesa'
2020-09-28 14:39:47 <hxx> in /var/log/Xorg.0.log
2020-09-28 14:39:48 <jmondi> Hello71: which indeed lists tty1 already :/
2020-09-28 14:40:39 <Hello71> you said "embedded device" so maybe ttyS1
2020-09-28 14:40:42 <Hello71> er, 0
2020-09-28 14:40:48 <Hello71> !paste
2020-09-28 14:40:59 <jmondi> added ttyS0,S1 and S2 :) I have an USB keryboard connected
2020-09-28 14:41:25 <jmondi> minirootfs does not seem to have ssh enabled, that's why I connected a keyboard
2020-09-28 14:41:56 <Hello71> you can cross-apk add
2020-09-28 14:42:26 <Hello71> but I'm pretty sure it also doesn't have networking enabled by default
2020-09-28 14:43:29 <jmondi> I can ping the device, maybe just because I had forced an ip address through a kernel arg to configure the nfsroot
2020-09-28 14:45:56 <Hello71> sure, it doesn't turn *off* the networking
2020-09-28 14:46:09 <hxx> k got it dbus failed to start
2020-09-28 14:46:42 <hxx> sorry lightdm service failed
2020-09-28 14:49:17 <jmondi> Hello71: yeah, but without the kernel arg that forces an ip address (or a dhcp query) I would not have had eth0 initialized :)
2020-09-28 14:50:35 <steinbach> hey, what do i ahve to put on boot command line to receive terminal output when installing alpine on a non-graphics computer
2020-09-28 14:50:40 <steinbach> i only have 1 serial port
2020-09-28 14:51:15 <Hello71> did you try console=ttyS0
2020-09-28 14:51:21 <Hello71> isn't that the default
2020-09-28 14:53:33 <steinbach> mmmmmm, or i dont get a feedback... ergo something is wrong or i am typing something wrong
2020-09-28 14:55:33 <steinbach> on older versions you had to type vanilla
2020-09-28 14:55:36 <steinbach> what is it now?
2020-09-28 14:55:43 <ikke> lts
2020-09-28 14:58:41 <steinbach> ok, that worked! thanks
2020-09-28 14:58:54 <steinbach> lts console=ttyS0,115200
2020-09-28 14:58:57 <steinbach> booted
2020-09-28 14:59:01 <steinbach> and getting feedback
2020-09-28 15:03:17 <jmondi> Hello71: meh, I removed securetty and now I can type in root and get asked for a password
2020-09-28 15:03:36 <jmondi> but the empy password doesn't work, nor does 'root' 'root'
2020-09-28 15:04:02 <jmondi> so I did `echo "root" | openssl passwd -1 -stdin` and placed the output in /etc/shadow
2020-09-28 15:04:10 <jmondi> and now I can login with 'root' 'root'
2020-09-28 15:04:14 <jmondi> weird
2020-09-28 15:04:17 <Hello71> could be
2020-09-28 15:08:33 <toogley> how big is the chance you will use sr.ht patch style of contributing after sourcehut has completed 1.0 ?
2020-09-28 15:08:57 <toogley> and i mean also to use its build system, its bug tracker, etc.
2020-09-28 15:09:41 <ikke> Not very high
2020-09-28 15:22:45 <Hello71> my understanding is that people (I think most loudly but not exclusively Ariadne) are grumpy about over-promising and under-delivering on the mailing list product
2020-09-28 15:23:37 <Ariadne> Hello71: I don't really care.  what I care about is that others complain and it causes problems with using the lists as intended.
2020-09-28 15:24:42 <Ariadne> personally I agree with a lot of Drew's opinions
2020-09-28 15:25:01 <Ariadne> most of sourcehut is very no nonsense and I appreciate that
2020-09-28 15:25:25 <Ariadne> but the lists are indeed problematic
2020-09-28 15:25:48 <Ariadne> and we will get mailman3 going eventually.
2020-09-28 15:26:00 <toogley> what do you mean by "but the lists are indeed problematic" ?
2020-09-28 15:26:22 <mps> jmondi: miniroots are intended to be used in docker (and similar) and not quite suited for native or full VM
2020-09-28 15:26:27 <Ariadne> some people refuse to use the lists for discussions.
2020-09-28 15:26:33 <Hello71> it may be more precise to say that Ariadne was conveying complaints, which may have appeared like Ariadne complaining. personally, I don't care. mailman, srht, as long as it has basic mailing functionality. just don't switch to sourceforge
2020-09-28 15:26:40 <Ariadne> that is problematic
2020-09-28 15:26:57 <mps> Hello71: I'm against sr.ht
2020-09-28 15:27:15 <Ariadne> hello71: yes, that is basically my position.  I rather use the software which has behavior that people like.
2020-09-28 15:27:39 <Ariadne> I would prefer the lists bounce mails back to me, but I can live with it
2020-09-28 15:27:52 <jmondi> mps: I'm noticing minirootfs is really, well, mini :)
2020-09-28 15:27:56 <Ariadne> I think that's mps problem too
2020-09-28 15:28:00 <Hello71> doesn't that cause backscatter
2020-09-28 15:28:07 <Hello71> jmondi: that's the point
2020-09-28 15:28:27 <mps> Hello71: not because of technical points but something 'else', which I will not discus here
2020-09-28 15:28:42 <Hello71> you can add packages before boot with cross-apk, but it can't run hooks
2020-09-28 15:28:53 <Hello71> which may cause certain packages to not work properly
2020-09-28 15:28:58 <Ariadne> Hello71: sure but most clients can deal with it, and when you have multiple clients accessing the same mailing list, it makes things easier
2020-09-28 15:29:08 <toogley> i like the idea of not having to create an account to contribute and i think its interesting and productive to use patches for pull requests. i think it is more straightforward compared to the typical github/gitlab/etc. style of forking/creating branches and creating pull requests
2020-09-28 15:29:18 <Ariadne> great
2020-09-28 15:29:29 <mps> jmondi: yes, if you install in 'full virtualization' box or native miniroots need some (a lot) of preparation.
2020-09-28 15:29:30 <toogley> but of course that is not possible at the moment, since sr.ht has not finished 1.0
2020-09-28 15:29:33 <Ariadne> you don't have to create an account even with gitlab first approach
2020-09-28 15:29:45 <jmondi> Hello71: indeed! I've gone through the whole setup and I got sshd to work, but, in example, udhcpd doesn't work well with my network dhcp server and such..
2020-09-28 15:29:47 <toogley> Ariadne: no?
2020-09-28 15:29:55 <Ariadne> you can always find a developer on irc and work with them directly
2020-09-28 15:30:09 <mps> Ariadne: not only me, there are (were) more developers who don't like 'sr.ht'
2020-09-28 15:30:11 <jmondi> mps: I'm not plannig to do any virtualization, I'm using minirootfs as my embedded system NFS root :)
2020-09-28 15:30:21 <Ariadne> mps: yes I know.
2020-09-28 15:30:26 <Hello71> jmondi: perhaps, but that's not udhcpd related, that's running multiple dhcp servers related
2020-09-28 15:30:42 <Ariadne> mps: and we plan to drop it, but I've been very busy as has been ikke
2020-09-28 15:30:44 <toogley> Ariadne: does that mean, i can also send some person a diff via IRC/mail and this person pushes that directly, without pull request on gitlab?
2020-09-28 15:30:51 <Ariadne> yes
2020-09-28 15:30:59 <toogley> Ariadne: interesting. thanks
2020-09-28 15:31:36 <jmondi> Hello71: I really hope I only have one running.. anyway, I like busybox, but I might need a slighlty more complex userspace environment..
2020-09-28 15:31:42 <Ariadne> gitlab flow is just better for most people because you don't have to build that relationship
2020-09-28 15:31:48 <jmondi> better switch to the full release probably
2020-09-28 15:31:55 <mps> jmondi: what arch?
2020-09-28 15:32:22 <Ariadne> anyway I am going to shortly be committing some crimes to bring gcc9 back on mips
2020-09-28 15:32:32 <Ariadne> just a heads up
2020-09-28 15:32:34 <Ariadne> lol
2020-09-28 15:32:40 <jmondi> mps: aarch64
2020-09-28 15:32:52 <jmondi> arm64, sorry.. well, you got it :)
2020-09-28 15:32:59 <Ariadne> I also agree with mps on the non-technical issue
2020-09-28 15:33:05 <Hello71> I don't think anybody is looking to rehash those sr.ht arguments now. I just wanted to tell toogley that there are some people specifically against sr.ht, mainly because of lacking mailing lists implementation. I also don't think we can rule out sr.ht entirely, but we almost certainly won't use it in the short-medium term, as we just switched to gitlab
2020-09-28 15:33:18 <mps> jmondi: then you can use iso or some of tarballs
2020-09-28 15:33:18 <Ariadne> we can't depend on somebody like drew for infrastructure
2020-09-28 15:33:38 <Ariadne> if it breaks at 3 am he is likely to just say go fuck yourself
2020-09-28 15:33:40 <Ariadne> lol
2020-09-28 15:33:48 <toogley> Hello71: yes, i agree
2020-09-28 15:34:09 <jmondi> mps: yes I should, not the iso as I want a tarball, but I'm sure I can find one
2020-09-28 15:34:13 <toogley> Ariadne: well that is the same for gitlab, i think? or do you use managed gitlab?
2020-09-28 15:34:16 <Hello71> Ariadne: sourcehut is it then? dunno which is which
2020-09-28 15:34:17 <jmondi> or loop  mount the iso and do that myself
2020-09-28 15:34:41 <Ariadne> toogley: we do have an SLA for gitlab
2020-09-28 15:34:47 <toogley> ah
2020-09-28 15:35:03 <Hello71> jmondi: or maybe you want udhcpc, not udhcpd
2020-09-28 15:35:11 <Hello71> Ariadne: uh?
2020-09-28 15:35:41 <Ariadne> hello71: ? :)
2020-09-28 15:35:47 <mps> jmondi: https://alpinelinux.org/downloads/ and Generic Arm, aarch64 is not big
2020-09-28 15:35:57 <Ariadne> if gitlab breaks we can escalate to gitlab and they will fix it if need be
2020-09-28 15:36:05 <Cogitri> toogley: See https://lists.alpinelinux.org/~alpine/devel/%3C24b91bd507e8151d41ac1d9866a4fd7a07febfe0.camel%40cogitri.dev%3E for the discussion on the ML about mail patches
2020-09-28 15:36:16 <Ariadne> it is one of the  main reasons we adopted gitlab as forge
2020-09-28 15:36:17 <jmondi> Hello71: yeah, just typed in wrong :)
2020-09-28 15:36:25 <jmondi> mps: I'll try that one
2020-09-28 15:36:28 <jmondi> thanks everyone
2020-09-28 15:36:53 <Ariadne> anyway there's plenty of alternatives to using the aports mailing list
2020-09-28 15:37:11 <Cogitri> TL;DR mail patches aren't great because they're a pain to review, doesn't have easy CI, managing vX patches sucks and most core devs don't like working with it (me included)
2020-09-28 15:37:18 <toogley> Cogitri thanks
2020-09-28 15:37:19 <Hello71> Ariadne: I see. so not really "SLA", is more "support contract"
2020-09-28 15:37:46 <Hello71> Cogitri: well, that's the point (theory) of sourcehut/sr.ht
2020-09-28 15:37:48 <Ariadne> yes that is what is meant by SLA
2020-09-28 15:39:21 <Ariadne> [09:37] <b39775Hello71> Cogitri: well, that's the point (theory) of sourcehut/sr.ht
2020-09-28 15:39:47 <Ariadne> unfortunately that requires a huge commitment from infra team to support and a lot of that code doesn't yet exist anyway
2020-09-28 15:39:47 <Hello71> Ariadne: imo "SLA" means that they operate it and guarantee a certain level of service. "support contract" means that they will come fix it if *you* break it. but as long as we understand each other I guess it's fine
2020-09-28 15:40:02 <Hello71> Ariadne: point (theory (hope)) :p
2020-09-28 15:40:42 <Ariadne> not that I'm involved with infra but :)
2020-09-28 15:41:15 <Hello71> maybe one could say it is "smokeware" :p
2020-09-28 15:41:55 <gildarts> st.ht's uptime stats have been fantastic compared to github at least, not familiar with gitlab's numbers.
2020-09-28 15:44:27 <Ariadne> who cares
2020-09-28 15:44:46 <Ariadne> we're not using a hosted forge, so it doesnt really matter
2020-09-28 15:45:11 <Ariadne> and we are only using sr.ht lists, which many people have complained about, so we will eventually stop using that too
2020-09-28 15:46:13 <ikke> Our gitlab instance has a great uptime
2020-09-28 15:46:34 <Ariadne> it's just every time
2020-09-28 15:46:46 <toogley> why did people have complained about sr.ht mls ? so why will you switch to mailman?
2020-09-28 15:46:51 <Ariadne> i have time to work on mailman3 packaging + dockerization
2020-09-28 15:46:57 <Ariadne> some shit hits the fan
2020-09-28 15:46:58 <Ariadne> lol
2020-09-28 15:47:06 <gildarts> Just a comment around the "if gitlab goes down we can call gitlab" comment earlier.
2020-09-28 15:47:11 <ikke> toogley: it's mostly the backlash of switching things
2020-09-28 15:47:17 <toogley> ah
2020-09-28 15:47:21 <gildarts> Don't have an opinion on what should be used.
2020-09-28 15:47:33 <Ariadne> mailman3 we can at least configure to behave as it did originally
2020-09-28 15:47:57 <ikke> toogley: without any warning in advance, it broke people's workflow
2020-09-28 15:48:39 <Ariadne> yes i agree that was the largest problem
2020-09-28 15:49:50 <mps> huh (I tested completed mailman3 but in lxc based on debian. waiting for Ariadne to make it on alpine)
2020-09-28 15:50:09 <mps> complete*
2020-09-28 15:50:25 <Ariadne> ikke: i think drew's response to people who complained made it a lot worse too
2020-09-28 15:50:35 <ikke> nod
2020-09-28 15:50:46 <mps> I don't know python to the degree to make mailman3 working on alpine
2020-09-28 15:50:58 <Ariadne> mps: i have some of the packaging already done
2020-09-28 15:51:11 <Ariadne> and this week i think i can finally make time to finish it
2020-09-28 15:51:27 <mps> Ariadne: yes, and I'm patient, though I don't forget :)
2020-09-28 15:51:32 <ikke> OK, so we need to start working on exporting the existing DB
2020-09-28 15:51:55 <mps> ikke: where is the DB?
2020-09-28 15:51:58 <Ariadne> yeah i know enough about how srht works (from trying to find solutions that made both drew and the annoyed users happy)
2020-09-28 15:52:53 <ikke> mps: on one of our infra servers
2020-09-28 15:53:45 <mps> ah, do I have read access to DB?
2020-09-28 15:53:52 <ikke> I don't know
2020-09-28 15:54:12 <ikke> I think you just have access to a couple of lxc containers, right?
2020-09-28 15:54:26 <mps> maybe we can discuss this on infra
2020-09-28 16:04:14 <toogley> btw, i'm thinking about running alpine linux on a thinkpad as desktop OS. with wayland and sway. i know that alpine is not used much on the desktop and rather on embedded systems, routers, and servers. i'm wondering if there is something i will be missing? i know already that i will have to run some packages in a glibc chroot. So i wonder if there's something that is really difficult to use
2020-09-28 16:04:16 <toogley> on alpine? something where a chroot is not possible?
2020-09-28 16:04:57 <ikke> nvidia drivers
2020-09-28 16:05:16 <toogley> won't need them, don't have a dedicated graphic card on this machine
2020-09-28 16:05:59 <toogley> the thinkpad is very old, T520
2020-09-28 16:06:50 <mps> toogley: binary only which depends on glibc will not work
2020-09-28 16:07:02 <mps> widevine for example
2020-09-28 16:07:06 <ikke> mps: should work in a chroot / flatpak, right?
2020-09-28 16:07:14 <mps> or it is videwine
2020-09-28 16:07:26 <ikke> widevine
2020-09-28 16:07:39 <toogley> ah, yeah. i don't need that. (in general, i don't need copyright DRM things in my browser)
2020-09-28 16:08:34 <mps> ikke: I don't have experience with flatpack, though was tempted to try dart lang in it
2020-09-28 16:09:18 <mps> toogley: I'm using alpine as my workstation about 2 and a half years
2020-09-28 16:09:49 <Cogitri> It's best to just try it, I suppose
2020-09-28 16:09:58 <mps> though I don't use some 'special' software
2020-09-28 16:10:56 <toogley> Cogitri: yeah, makes sense
2020-09-28 16:43:20 <toogley> (thanks everyone)
2020-09-28 17:20:53 <toogley> is there a reason why you don't have an account in the fediverse yet? on mastodon or something?
2020-09-28 17:23:59 <toogley> https://fosstodon.org/explore would be a suitable instance
2020-09-28 17:26:32 <detha> https://xkcd.com/1810 comes to mind
2020-09-28 17:28:51 <helby> toogley: just yesterday bought T420
2020-09-28 17:29:01 <helby> everything works well
2020-09-28 17:29:24 <helby> toogley: but I have too minimal system, so media keys don't work, did not even tried them
2020-09-28 17:29:39 <rodrigue7973> hi all
2020-09-28 17:29:53 <rodrigue7973> i am the problem for locale langugage french x)
2020-09-28 17:30:04 <rodrigue7973> i am begennier
2020-09-28 17:30:22 <toogley> detha: yes, but the fediverse is a) based on free software und b) its rising in popularity
2020-09-28 17:30:22 <helby> toogley: not sure about touchpad double click, if it requires some libinput or whatever setting, bc. it does not work in defualt alpine for me
2020-09-28 17:30:31 <toogley> rodrigue7973: bonjour!
2020-09-28 17:30:41 <rodrigue7973> toogley: ah toi france ?
2020-09-28 17:31:05 <toogley> rodrigue7973: no, not really. :D
2020-09-28 17:31:07 <rodrigue7973> toogley: you form france?
2020-09-28 17:31:13 <rodrigue7973> zute!
2020-09-28 17:32:21 <toogley> rodrigue7973: https://www.deepl.com/fr/translator ?
2020-09-28 17:33:13 <rodrigue7973> sorry good evening,
2020-09-28 17:33:14 <rodrigue7973> I had a local problem that was still an English keyboard in the Alps.
2020-09-28 17:33:16 <rodrigue7973> please help me
2020-09-28 17:38:38 <rodrigue7973> u!p !
2020-09-28 17:39:40 <helby> rodrigue7973: did u tried 'setup-keymap' command as root /doas/sudo ?
2020-09-28 17:42:43 <rodrigue7973> ok
2020-09-28 17:43:43 <rodrigue7973> thank !
2020-09-28 18:01:06 <rodrigue7973> I have a question for seafile but it says "disappeared" but pkoi it no longer has an alpine linux package?
2020-09-28 18:01:07 <rodrigue7973> thank you
2020-09-28 18:29:55 <CactiChameleon9> does alpine have tor browser in its repos?
2020-09-28 18:30:38 <c705> seems like no. maybe in testing?
2020-09-28 18:31:02 <CactiChameleon9> I cannot find it, however apk search does not seem very safisticated
2020-09-28 18:31:37 <c705> then no
2020-09-28 18:37:36 <helby> but tor is there
2020-09-28 18:41:22 <u0jQx9gPyrYg> someone(TM) needs to package tor browser
2020-09-28 18:41:32 <ikke> u0jQx9gPyrYg: feel free ;-)
2020-09-28 18:41:37 <c705> u0jQx9gPyrYg: feel free to do it yourself
2020-09-28 18:42:07 <c705> if you seriously *need* to use tor browser for anything, you should be using tails anyways
2020-09-28 18:42:22 <u0jQx9gPyrYg> i was reacting to CactiChameleon9
2020-09-28 18:42:35 <ikke> oh, right, I see
2020-09-28 18:43:09 <helby> or just encrypt your nick on irc like u0jQx9gPyrYg and u don't need any tor
2020-09-28 18:44:42 <CactiChameleon9> I dont need it, its just I block google using pi hole and refuse to use anything that needs it without tor
2020-09-28 18:44:56 <c705> alrighty then
2020-09-28 18:48:24 <helby> smart boy
2020-09-28 20:27:25 <toogley> hey. i'm currently installing alpine with the "layers" /dev/sda3 (partition containing LVM volume things) -> luks1 crypt -> finally the lvm volume. so the LVM is inside crypto.
2020-09-28 20:27:32 <toogley> and i'm reading https://wiki.alpinelinux.org/wiki/LVM_on_LUKS#Installing_Alpine_Linux
2020-09-28 20:27:40 <toogley> there, it says
2020-09-28 20:27:43 <toogley> To enable the operating system to decrypt the PV at boot time, create the /mnt/etc/crypttab file. Enter the following line into the file to decrypt the /dev/sda2 partition using the luks module and map it to the lvmcrypt name:
2020-09-28 20:28:13 <toogley> and i'm now confused, because /mnt/etc/crypttab is inside the root lvm and that is inside the luks crypto
2020-09-28 20:31:05 <c705> toogley: you'd need to unlock the lvm within grub if the root fs is on the encrypted device
2020-09-28 20:31:16 <c705> the cryptydisk, rather
2020-09-28 20:32:17 <toogley> yes, is this difficult?
2020-09-28 20:32:53 <c705> that's subjective. see the grub documentation
2020-09-28 20:37:06 <toogley> c705: hm, thanks
2020-09-28 21:00:56 <bbuccianti> how can I make abuild to install local package as a deps for another one?
2020-09-28 21:05:40 <Cogitri> Sorry?
2020-09-28 21:10:49 <bbuccianti> suppose that you're building a package A that depends on package B, but you want to provide your APKBUILD for package B before building package A
2020-09-28 21:13:45 <Cogitri> Just build package B with a higher pkgver/pkgrel than what's in our repos and then build package A
2020-09-28 21:21:26 <bbuccianti> thanks!
2020-09-28 21:48:19 <toogley>  hey. so i try to configure my grub2 bootloader to unlock my cryptodevice which contains lvm which contains root, swap, home. the alpine docs seem to be outdated so i
2020-09-28 21:48:21 <toogley> try to read voidlinux and archlinux docs.
2020-09-28 21:48:55 <toogley>  on archlinux there is this: https://wiki.archlinux.org/index.php/Dm-crypt/Encrypting_an_entire_system#Configuring_the_boot_loader_2 and the voidlinux docs use at this point the rd.luks.name=device-UUID=cryptlvm root=/dev/MyVolGroup/root syntax
2020-09-28 21:50:12 <toogley>  dracut seems to be a package which alpinelinux doesn't seem to have
2020-09-28 21:50:54 <toogley> see https://docs.voidlinux.org/installation/guides/fde.html strg+ f GRUB_CMDLINE_LINUX_DEFAULT regarding voidlinux docs
2020-09-28 21:52:13 <toogley> so in essence, i try to understand what the correct content of my GRUB_CMDLINE_LINUX_DEFAULT parameter in /etc/default/grub is
2020-09-28 21:52:37 <Cogitri> We use mkinitfs and not dracut so you need different parameters
2020-09-28 21:53:22 <Cogitri> `cryptdm=pool cryptroot=/dev/nvme0n1p4 modules=btrfs,nvme root=/dev/mapper/pool rootfstype=btrfs`
2020-09-28 21:53:47 <Cogitri> Change btrfs for whatever FS you use and cryptroot to point to the right partition
2020-09-28 21:54:02 <toogley> what is nvme?
2020-09-28 21:54:18 <toogley> ah the physical volume?
2020-09-28 21:54:26 <Cogitri> pcie based storage
2020-09-28 21:54:35 <toogley> ah
2020-09-28 21:54:44 <Cogitri> If you don't have an NVM.e SSD you don't need that
2020-09-28 21:58:29 <toogley> https://wiki.alpinelinux.org/wiki/LVM_on_LUKS#Grub_with_UEFI says "cryptdm=lvmcrypt" - i guess yours is the most current way to specify that or did that not change?
2020-09-28 21:59:46 <Cogitri> I think the name doesn't matter
2020-09-28 22:00:06 <Cogitri> Just make sure it's consistent in both cryptdm= and root=
2020-09-28 22:00:21 <Cogitri> Since root= needs to be /dev/mapper/$cryptdm
2020-09-28 22:02:52 <toogley> my fs is ext4, do i need modules=ext4 ?
2020-09-28 22:03:19 <toogley> or where is that documented?
2020-09-28 22:12:10 <toogley> hm, i do that tomorrow. good night! :) and thanks for your help, Cogitri
2020-09-29 14:45:47 <fling> Hello.
2020-09-29 14:46:10 <fling> How to configure apk to use a proxy?
2020-09-29 14:46:11 <fling> https://wiki.alpinelinux.org/wiki/Apk
2020-09-29 14:46:15 <fling> I'm reading this ^
2020-09-29 14:49:36 <rodrigue7973> good evening,
2020-09-29 14:49:37 <rodrigue7973> why is seafile-server missing?
2020-09-29 14:49:39 <rodrigue7973> thank you
2020-09-29 14:56:12 <u0jQx9gPyrYg> fling: i would expect the standard way to work. set the proxy via env vars
2020-09-29 14:58:32 <fling> u0jQx9gPyrYg: do I need http or https?
2020-09-29 14:59:04 <detha> whichever one you use in /etc/apk/repositories
2020-09-29 14:59:05 <u0jQx9gPyrYg> depends on what you set as the url in /etc/apk/repositories
2020-09-29 14:59:29 <fling> ok got it
2020-09-29 15:00:40 <fling> will not lxd be available in 3.12?
2020-09-29 15:01:12 <mps> fling: no, but it is in testing
2020-09-29 15:02:17 <fling> testing is edge? https://pkgs.alpinelinux.org/packages?name=lxd&branch=edge
2020-09-29 15:02:59 <mps> yes and no, testing is where pkgs are tested
2020-09-29 15:03:27 <mps> testing is always edge
2020-09-29 15:03:34 <fling> so testing is just an extra repository for 3.12?
2020-09-29 15:03:53 <mps> no
2020-09-29 15:04:05 <fling> ok I don't see it here https://pkgs.alpinelinux.org/packages?name=lxd&branch=v3.12&repo=testing
2020-09-29 15:04:13 <mps> it is not officially released, wver
2020-09-29 15:04:18 <mps> ever*
2020-09-29 15:23:08 <fling> how to debug 'Error: Permission denied' in apk?
2020-09-29 15:23:24 <fling> update works
2020-09-29 15:23:28 <fling> but add and upgrade errors out
2020-09-29 15:23:33 <afontain_> Are you running apk as root?
2020-09-29 15:23:38 <fling> yes
2020-09-29 15:24:11 <afontain_> can you give the full error message?
2020-09-29 15:27:19 <fling> afontain_: https://bpa.st/YIHQ
2020-09-29 15:35:44 <Hello71> read-only root
2020-09-29 15:36:00 <fling> but 'update' works?
2020-09-29 15:36:16 <Hello71> files are cached
2020-09-29 15:37:14 <fling> / is writeable
2020-09-29 15:48:37 <afontain_> I'm not sure why you get these, to be honest
2020-09-29 15:48:50 <afontain_> strace may give you the failing syscall
2020-09-29 16:09:31 <fling> afontain_: sh: strace: not found
2020-09-29 16:10:55 <afontain_> (yeah, you need to install it)
2020-09-29 16:12:06 <Hello71> lol
2020-09-29 16:15:02 <afontain_> now that I think of it, that's kinda annoying if the package manager can't install anything
2020-09-29 16:15:13 <afontain_> are you running this inside lxd?
2020-09-29 16:21:04 <fling> afontain_: yes
2020-09-29 16:21:08 <fling> Hello71: haha :D
2020-09-29 16:21:49 <Hello71> how did you check "/ is writeable"
2020-09-29 16:22:06 <fling> touch /test
2020-09-29 16:22:28 <fling> Hello71: also apk writes something so after the error upgrade gives the same error
2020-09-29 16:22:47 <Hello71> what?
2020-09-29 16:23:35 <afontain_> I kinda expect there is some capability or syscall forbidden
2020-09-29 16:23:55 <afontain_> but I've never used lxd
2020-09-29 16:24:38 <mps> yes, lxd is not well tested on alpine and because that it is still in testing
2020-09-29 16:25:54 <fling> afontain_: apk add lxd ; lxd init ; lxc launch images:alpine/3.12 my-alpine ; lxc exec my-alpine sh ; apk add nano
2020-09-29 16:26:22 <fling> mps: I've used it from the beginning on gentoo
2020-09-29 16:26:53 <mps> maintainer on alpine told that it needs more testing
2020-09-29 16:27:20 <mps> fling: did you used it built with musl or glibc
2020-09-29 16:27:23 <fling> anyway lxd works fine for me, only getting issues with these new alpine images both 3.12 and edge
2020-09-29 16:28:53 <fling> mps: how to check?
2020-09-29 16:29:31 <fling> I will try a vm
2020-09-29 16:44:50 <mps> look at /lib/libc
2020-09-29 16:50:26 <fling> mps: ls: /lib/libc: No such file or directory
2020-09-29 16:50:53 <fling> I found ld-musl-x86_64.so.1
2020-09-29 16:52:28 <mps> fling: on gento?
2020-09-29 16:52:40 <fling> on alpine
2020-09-29 16:53:25 <mps> we know what is on alpine, I asked to look at gento
2020-09-29 16:53:41 <fling> I don't have musl on gentoo, I have glibc
2020-09-29 16:54:00 <mps> heh
2020-09-29 16:54:02 <fling> I can install musl
2020-09-29 16:54:35 <mps> I asked do you run lxd on glibc or musl on gento, and you told you don't know
2020-09-29 16:54:50 <fling> oh
2020-09-29 16:55:00 <mps> now it is more clear
2020-09-29 16:55:14 <fling> you said 'it' I thought you asked about the image not lxd package
2020-09-29 16:55:18 <mps> lxd is not well tested on musl, as I wrote above
2020-09-29 16:55:29 <mps> ok, no problem
2020-09-29 16:55:55 <mps> on IRC it is easy to fall in these 'traps' of misunderstanding
2020-09-29 16:56:17 <fling> should I install musl and lxd on gentoo?
2020-09-29 16:56:35 <mps> I don't know really, never used gento
2020-09-29 17:04:38 <fling> what is the default root password?
2020-09-29 17:07:09 <Cogitri> I think it's unset by default
2020-09-29 17:07:19 <Cogitri> So you can't login without a passwd
2020-09-29 17:08:05 <fling> it asks me to login on the console hmm hmmmm
2020-09-29 17:11:04 <mps> Enter key
2020-09-29 17:11:19 <fling> Login incorrect
2020-09-29 17:13:20 <fling> I should probably use lxd-agent instead but when I try to lxc exec I get `Error: Failed to connect to lxd-agent`
2020-09-29 17:13:47 <mps> hmm, type root at login prompt, and on password prompt press 'Enter'
2020-09-29 17:14:32 <mps> ah, maybe /etc/securetty missing console
2020-09-29 17:14:39 <fling> mps: this is what I tried, the password is probably unset
2020-09-29 17:15:18 <mps> yes
2020-09-29 17:15:45 <mps> so /etc/securetty maybe need something
2020-09-29 17:21:25 <c705> last time I ran into something like this, it turns out the problem was with one of the packages that was in my apkovl. The package failed to install, no error messages, and it lead to the apkovl.tar.gz not being unpacked and /etc/securetty not being present
2020-09-29 17:57:32 <toogley> test?
2020-09-29 18:17:45 <fling> toogley: Evaluate conditional expression?
2020-09-29 18:18:33 <toogley> fling: no, was just trying if i was still alive in this channel
2020-09-29 18:27:29 <fling> Where is the source script for producing lxd images?
2020-09-29 18:39:51 <miskatonic> is it possible to install packages from testing under a different directory than main packages, such as under /usr/local or /opt?
2020-09-29 18:49:35 <northbrae> Hello. I am using Alpine as a bastion host in a ProxyJump setup. I want the user to successfully authenticate to the bastion via "ssh -J bastion target" but I don't want the user to be able to gain a shell on the bastion (so "ssh bastion" should fail). I have tried various things, such as setting the user's shell to /bin/true or /bin/false, or
2020-09-29 18:49:36 <northbrae> setting "Match User / ForceCommand" to /bin/false. These all fail, because it seems that Alpine insists that the user be able to gain a real shell. Is there way in Alpine to do what I want?
2020-09-29 18:50:52 <fling> miskatonic: in a chroot?
2020-09-29 21:14:44 <atkka> is this still correct? https://wiki.alpinelinux.org/wiki/LXC
2020-09-29 21:16:19 <mps> atkka: skip part about grsec
2020-09-29 21:16:24 <atkka> I'm getting errors: get_template_path: 887 No such file or directory - bad template: alpine
2020-09-29 21:16:27 <atkka> mps I did
2020-09-29 21:17:32 <mps> and don't forget to start cgroups
2020-09-29 21:17:52 <mps> /etc/init.d/cgroups
2020-09-29 21:19:58 <atkka> cgroups is started
2020-09-29 21:20:07 <atkka> error persists when I run: lxc-create -n guest1 -f /etc/lxc/lxc.conf -t alpine
2020-09-29 21:20:47 <atkka> I installed lxc-templates as well (needed?) logged out and in
2020-09-29 21:21:12 <mps> lxc-templates-legacy-alpine is also needed
2020-09-29 21:23:05 <atkka> huh, ok. why is it considered a legacy template? and why does it indicate you don't need any templates is you are making alpine containers?
2020-09-29 21:24:15 <mps> hm, never looked why but without it lxc doesn't work for me
2020-09-29 21:24:48 <atkka> the first paragraph of that wiki says you need no templates for alpine containers, very frustrating
2020-09-29 21:25:41 <mps> that's nice world of wikis :)
2020-09-29 21:27:26 <atkka> well, I don't see the point of having any information at all if the information is incorrect.
2020-09-29 21:27:44 <atkka> incorrect pages should be pruned
2020-09-29 21:28:29 <mps> well, `rm -rf /whole_internet`
2020-09-29 21:29:59 <atkka> honestly the would would be better off
2020-09-29 21:30:08 <atkka> world
2020-09-29 21:30:53 <mps> actually I create lxcs 'by hand', editing config files and using 'apk --arch arch --root /path ...' to install base and rest of pkgs
2020-09-29 21:31:35 <mps> I'm not yet decided when I will 'rm -rf / ' :)
2020-09-29 21:33:47 <atkka> yeah I'm pretty close to rm -rf /myself
2020-09-29 21:33:47 <mps> atkka: sorry, I'm now tired, going to smoke one pipe and go to bed. good night
2020-09-29 21:34:07 <atkka> no problem, have a good night
2020-09-29 22:59:40 <motte_> hi, could anyone tell me what package i need to install to solve ld: cannot find -lz? i thought it was zlib-dev, but it seems not
2020-09-29 23:01:56 <atkka> lz4-libs?
2020-09-29 23:16:36 <Hello71> i bet trying to compile in 32 bit or something
2020-09-29 23:16:44 <Hello71> motte_: full command and output?
2020-09-29 23:31:08 <motte_> my god is haskell compiling slow
2020-09-29 23:31:31 <motte_> it just finished.. and lz4-libs didn't do the trick
2020-09-29 23:31:53 <atkka> thanks for the info
2020-09-29 23:33:10 <motte_> Hello71: i'm trying to compile a fully static binary with docker
2020-09-29 23:34:01 <motte_> Hello71: the full output is insanely long, but the complete error message is this:
2020-09-29 23:34:10 <motte_> /usr/lib/gcc/x86_64-alpine-linux-musl/10.2.0/../../../../x86_64-alpine-linux-musl/bin/ld: cannot find -lz
2020-09-29 23:35:13 <motte_> and the base Dockerfile looks like this: http://ix.io/2zdt
2020-09-29 23:39:56 <Hello71> ugh
2020-09-30 06:14:06 <CactiChameleon9> I have an issue with nextcloud-desktop
2020-09-30 06:14:24 <CactiChameleon9> It doesnt seem to open
2020-09-30 06:14:41 <CactiChameleon9> And I cannot find any of its arch dependencies in the repos
2020-09-30 06:56:25 <Cogitri> CactiChaneleon9: fwiw it works just fine for me
2020-09-30 06:56:39 <Cogitri> I guess arch might name qt5 packages differently
2020-09-30 06:59:20 <afontain_> motte_: https://pkgs.alpinelinux.org/contents?file=libz.so*&path=&name=&branch=edge
2020-09-30 07:01:24 <afontain_> if you want the static one, it's in zlib-static
2020-09-30 07:01:33 <afontain_> (https://pkgs.alpinelinux.org/contents?file=libz.a*&path=&name=&branch=edge)
2020-09-30 11:09:40 <motte> afontain_: thanks, zlib-static did the trick
2020-09-30 11:48:13 <kn_> hello. Is there any reason why I can't get an output from who/w/users?
2020-09-30 11:49:18 <Hello71> alpine doesn't enable utmp
2020-09-30 13:21:06 <detha> my apk-search foo is failing, what would be a viable postgres connector for python3?
2020-09-30 13:28:17 <t0mato> detha: I've always used py3-psycopg2, not sure if there's something better available
2020-09-30 13:31:07 <detha> thnks, that is what I used to use, but for some reason apk didn't find it
2020-09-30 13:34:52 <detha> hmm. apk search psyco finds it, apk search postgres doesn't
2020-09-30 13:38:31 <t0mato> very weird, even when explicitly searching with "apk search -d postgres" I can't seem to find the package
2020-09-30 14:11:49 <fling> Is strace the proper thing for debugging apk?
2020-09-30 14:20:05 <ikke> depends on what you want to debug
2020-09-30 14:20:12 <ikke> strace only shows systemcalls
2020-09-30 14:20:39 <fling> ikke: error permission denied
2020-09-30 14:22:12 <ikke> fling: are you running it in docker?
2020-09-30 14:22:22 <fling> ikke: in lxd
2020-09-30 14:25:22 <u0jQx9gPyrYg> strace is a proper tool to figure out permission issues
2020-09-30 15:16:49 <ikke> fling: You need SYS_CAP_PTRACE before you are able to use strace
2020-09-30 15:17:07 <ikke> fling: By default, at least with LXC, containers have limited capabilities
2020-09-30 16:51:22 <abenz> can I unplug an alpine usb (diskless mode) and continue using the system normally? (given I don't use lbu after unplugging the USB) ?
2020-09-30 16:56:08 <artok> abenz: if your usb isn't mounted anywhere, sure
2020-09-30 16:56:41 <abenz> artok: its mounted at /usb/media (I have apkcache and lbu setup)
2020-09-30 16:56:50 <tboerger> is it possible that edge is currently broken on arm32? i'm receiving errors like this on cloud.drone.io: ncurses-terminfo-base-6.2_p20200926-r1: temporary error (try again later)
2020-09-30 16:56:55 <tboerger> this happens only for arm32, currently i'm testing it only with dl-cdn mirrors, you can see the whole output at https://cloud.drone.io/toolhippie/kubectl/3/2/2
2020-09-30 16:57:29 <abenz> artok: when I try to unmount /media/usb it says target is busy
2020-09-30 16:57:51 <tboerger> and i can reproduce that reliable within drone ci for arm32 only, amd64 and arm64 are building totally fine with the same mirror config :(
2020-09-30 17:09:59 <ikke> You need to stop the modloop service
2020-09-30 17:11:33 <abenz> ikke: is it sane? thats what I'm asking, eg say I want to boot up a few machines with the same usb
2020-09-30 17:11:50 <ikke> tboerger: might be related to the 64-bits time_t change
2020-09-30 17:12:09 <abenz> after stopping modloop what do I lose in terms of functionality ?
2020-09-30 17:12:23 <ikke> abenz: loading new kernel modules and firmware
2020-09-30 17:15:12 <abenz> ikke: https://paste.debian.net/1165315/
2020-09-30 17:15:30 <ikke> tboerger: if it's the same problem we ran into, where libseccomp gives an incorrect response to non-existing system calls, which prevents musl from properly falling back
2020-09-30 17:16:01 <kn_> Is there a way to see users logged in with Alpine?
2020-09-30 17:16:21 <ikke> No
2020-09-30 17:18:23 <monado> Hi everyone
2020-09-30 19:04:37 <mps> tboerger: arm32? which one: armhf or armv7
2020-09-30 19:05:15 <mps> and please if you can use paste service without JS, or better plain text paste service
2020-09-30 19:18:08 <Hello71> ikke: is not libseccomp, is docker misuse of seccomp
2020-09-30 19:28:41 <usblion> Hi! I'd like to use Intel GVT-G for a VM, and according to the Arch Wiki I need the vfio-mdev and kvmgt modules, but I can't find them. Are they included in some package or do I need to rebuild the kernel myself?
2020-09-30 19:29:27 <c705> usblion: vfio is part of the kernel
2020-09-30 19:31:44 <ikke> Hello71: oh, right
2020-09-30 19:32:14 <Hello71> also technically seccomp is not libseccomp
2020-09-30 19:39:44 <tboerger> mps: it's simple a public ci system where all the logs are available. it's armhf which gets also printed. the only error i'm getting you can see above, "temporary error (try again later)" for multiple packages.
2020-09-30 19:40:33 <tboerger> ikke: and what does that mean? arm64 is even building fine, it's really just the 32bit version on edge while all other versions are working.
2020-09-30 19:41:17 <tboerger> and previously it was working totally fine, idk when this started to fail
2020-09-30 19:42:31 <ikke> tboerger: It means that musl tries the 64-bit syscall, expects ENOSYS, and falls back to 32-bits versions
2020-09-30 19:43:04 <ikke> seccomp (or dockers way of using seccomp) makes it return EPERM
2020-09-30 19:43:20 <ikke> so musl never falls back to the 32-bits variants
2020-09-30 19:43:35 <tboerger> and how can this be fixed for docker images? :D
2020-09-30 19:44:16 <mps> ah, CI log. ok
2020-09-30 20:21:53 <ikke> tboerger: you can't
2020-09-30 20:21:57 <ikke> it would need to be fixed on the host
2020-09-30 21:38:15 <tboerger> ikke: do you also know how?
2020-09-30 21:39:11 <ikke> Latest docker version should fix it, or use the seccomp profile from github that includes these syscalls