2023-11-01 00:07:23 <zcrayfish> It works fine for many people, but it isn't the default choice; probably because chronyd works much better on machines with intermittent connectivity and/or power management features.
2023-11-01 02:01:56 <Saijin_Naib[m]> Oh wow, I always use built-in when using setup-alpine... I might need to rethink that
2023-11-01 02:06:48 <zcrayfish> I'm glad I'm off of Hughes FSS internet these days: none of the setup-ntp options support multicast.
2023-11-01 02:07:17 <omni> and chrony can do NTS
2023-11-01 05:21:03 <Saijin_Naib[m]> Whoever fixed evolution/evolution-data-server/evolution-ews, so many thank you
2023-11-01 05:21:18 <Saijin_Naib[m]> I've been using my fucking phone for the past month to do PIM stuff, ugh
2023-11-01 06:25:39 <ptrc> did anyone also stumble over an issue where launching pavucontrol with pipewire+\pipewire-pulse+wireplumber breaks audio output?
2023-11-01 06:26:17 <ptrc> not yet sure what causes this, but using something like helvum to disconnect the pavucontrol volume monitors fixes the issue
2023-11-01 09:36:43 <somercet1> anyone having lockup problems with FF 119?
2023-11-01 10:58:04 <lopid> hmm, maybe, but not on alpine
2023-11-01 11:14:04 <WhyNotHugo> I have completely borked the active tty (tty2) on my laptop, but I can still ssh into it...
2023-11-01 11:14:11 <WhyNotHugo> ... How can I switch the active tty to tty1 over ssh?
2023-11-01 11:22:14 <somercet1> WhyNotHugo, chvt
2023-11-01 11:24:43 <WhyNotHugo> somercet1: Thanks! 
2023-11-01 11:45:37 <Apachez> any of you with experience from vlan-aware bridges (with IP address set on the vlan interfaces) in linux?
2023-11-01 11:48:20 <dwfreed> ask your real question(s)
2023-11-01 12:00:25 <Apachez> so I have it setup and wonder if you are supposed to see two arp request be sent almost at the same time from the vlan interface with ip address set towards the egress?
2023-11-01 12:00:56 <Apachez> because right now when I tcpdump on remote I see 2 arp requests being sent at almost same time, arp reply returns but the linux host just ignore that
2023-11-01 12:01:05 <Apachez> so the arp on the linux says either INVALID or INCOMPLETE
2023-11-01 12:01:27 <Apachez> is there some sysctl which needs to be adjusted regarding arp to work properly?
2023-11-01 12:01:43 <Apachez> and why does the linuxbox with the vlanaware bridge configured send 2 arp requests at once?
2023-11-01 12:01:54 <Habbie> can you show your setup?
2023-11-01 12:02:31 <Apachez> https://forum.vyos.io/t/cant-make-untagged-vlan-to-work-with-bridge-br0-and-vrf/12613/3
2023-11-01 12:02:50 <Habbie> what OS are you on?
2023-11-01 12:02:54 <Apachez> I was hoping like "yeah I have seen that, you need to do X for that to work"
2023-11-01 12:03:04 <Apachez> linux 6.1.59 or so
2023-11-01 12:03:13 <Habbie> what distribution?
2023-11-01 12:03:22 <Apachez> vyos, based on debian 12 bookworm
2023-11-01 12:03:38 <Habbie> this channel is for alpine :)
2023-11-01 12:03:43 <Apachez> yes but alpine is linux
2023-11-01 12:04:12 <dwfreed> but how to do this in vyos may differ from how to do it in alpine
2023-11-01 12:04:17 <Apachez> and the main question is locate someone doing vlan-aware bridges who could tell if its normaly that they will send 2 arp requests at one and ignore arp replies?
2023-11-01 12:04:17 <dwfreed> this channel only supports alpine
2023-11-01 12:04:18 <Habbie> dwfreed, i am certain of it, even
2023-11-01 12:04:51 <Apachez> I understand if you dont have any experience of vlan-aware bridges and its ok to say so
2023-11-01 12:05:15 <Habbie> i have plenty of experience with vlan-aware bridges, which is why i can tell you that the way they are configured in a system matters -a lot-
2023-11-01 12:05:20 <Habbie> and thus asking in here is not useful
2023-11-01 12:05:21 <Apachez> and if you do have experience of it in alpine, how did you make it accept arp replies?
2023-11-01 12:08:06 <dwfreed> you need to ask your question in either a vyos channel, or a general network channel may be able to help
2023-11-01 12:14:34 <Apachez> yeah was hoping somebody in alpine was using linux but I guess I was wrong :(
2023-11-01 12:16:15 <Habbie> we are using linux; doesn't mean we can help
2023-11-01 12:17:22 <Apachez> yeah I figured that part out :(
2023-11-01 12:35:38 <dwfreed> Apachez: I mean, if you'd asked this in an appropriate channel, you probably would have had an answer already
2023-11-01 12:42:16 <minimal> Apachez: to be specific, Linux is the kernel, the rest of the OS is not Linux and differs between distros
2023-11-01 12:44:03 <dwfreed> and even though alpine and vyos both run on linux, vyos could be patching their kernel builds to change behavior in a way that alpine isn't that could be relevent here
2023-11-01 12:44:08 <dwfreed> or vice versa
2023-11-01 13:07:24 <Apachez> minimal: well sure but I would hope people use alpine to more than just install it and then nothing more
2023-11-01 13:07:44 <Apachez> hence its related to ask if someone have experience of that issue in an alpine channel aswell
2023-11-01 13:07:56 <Apachez> but again I was obviously hoping too much
2023-11-01 13:10:59 <minimal> Apachez: you're missing the point already made by someone else that how networking things (tools used, config file formats, etc) are setup differs between distros
2023-11-01 13:12:39 <Apachez> this is core linux same as alpine is using which is what you are missing with my question
2023-11-01 13:13:49 <minimal> "core linux"? you mean the kernel? as that's is Linux, nothing else
2023-11-01 13:14:31 <minimal> also again as dwfreed pointed out different distro are likely configuring the kernel differently and possibly/probably using different additional kernel patches
2023-11-01 13:16:34 <Apachez> this wasnt a question about kernel patches
2023-11-01 13:17:06 <dwfreed> kernel patches matter
2023-11-01 13:17:16 <dwfreed> and you're not getting the point
2023-11-01 13:17:24 <dwfreed> this channel is for *alpine* support
2023-11-01 13:17:31 <dwfreed> not *vyos* support
2023-11-01 13:19:48 <dwfreed> alternately, "/msg dpkg based on debian" and then just replace "Debian" with "Alpine" in the response, and it's the same idea
2023-11-01 13:23:19 <Apachez> yeah and noone could reply to how this behaves in alpine
2023-11-01 13:23:29 <Apachez> so I guess this channel isnt for alpine support either which is too bad
2023-11-01 13:26:40 <Habbie> it is for alpine support; but i don't understand why you keep arguing
2023-11-01 13:33:57 <Apachez> yeah and even my alpine related question was unanswered 
2023-11-01 13:36:59 <minimal> I assume you only wanted a reply if someone here could answer it, rather than everyone in the channel saying "nope, I don't know about that on Alpine" ;-)
2023-11-01 13:37:04 <Habbie> indeed
2023-11-01 13:37:19 <Habbie> and it was a -very- specific question
2023-11-01 13:37:30 <Habbie> it's very likely nobody that happened to be awake in fact had the relevant experience
2023-11-01 13:41:38 <Apachez> yeah bummer
2023-11-01 14:51:03 <alethkit> Does tiny-cloud require a specific version of NoCloud configuration for networking?
2023-11-01 15:00:56 <minimal> alethkit: you mean for network-config file? I don't believe it supports that
2023-11-01 15:03:01 <minimal> alethkit: though tomalok could give you a definitive answer
2023-11-01 17:49:39 <vkrishn> any thoughts of having these config values captivedetect.canonicalContent, captivedetect.canonicalURL, removed by default in alpine firefox install ?
2023-11-01 17:50:34 <vkrishn> those URLs kinda connects 3+ times a sec, had to remove it
2023-11-01 18:02:40 <Kladky> Considering that afaik desktop environments don't indecate that you need to login to a wifi network, firefox is the only way to actually notice that you need to do so.
2023-11-01 18:08:43 <vkrishn> giving search "update" in about:config shows more urls that it tries to connect to
2023-11-01 20:19:53 <alethkit> minimal: Yes. So would I have to set up some sort of local IMDS server?
2023-11-01 20:32:09 <minimal> alethkit: ? you said you're using NoCloud, rather than a metadata server
2023-11-01 20:33:35 <minimal> I'm not sure what you're trying to achieve, especially as you seem to assume tiny-cloud supports all/a lot of cloud-init compatible functionality whereas it supports a relatively small subset
2023-11-01 21:03:15 <alethkit> minimal: Oh, I thought NoCloud *supported* a metadata server
2023-11-01 21:03:32 <alethkit> https://cloudinit.readthedocs.io/en/latest/reference/datasources/nocloud.html method 3, in particular
2023-11-01 21:27:27 <minimal> alethkit: NoCloud-Net supports fetching YAML docs from a webserver, I wouldn't exactly call it a metadata server as a metadata server tends to be more than just a simple webserver
2023-11-01 21:28:00 <minimal> Also you're reading the cloud-init docs but asking about tiny-cloud - tiny-cloud is NOT cloud-init
2023-11-01 21:29:28 <minimal> so can you focus on whether you want to use tiny-cloud or cloud-init and then read the appropriate docs for whichever you want to use?
2023-11-01 21:57:50 <alethkit> I mean, I couldn't find any networking documentation on tiny-cloud, so I assumed it was cloud-init compatible
2023-11-01 21:57:57 <alethkit> or should I just read the source?
2023-11-01 22:29:30 <minimal> alethkit: I said earlier that I don't believe tiny-cloud supports NoCloud network configuration (i.e. it just uses dhcp)
2023-11-01 22:29:58 <minimal> you could look at the source or ask tomalok (the main dev) in the alpine-cloud channel
2023-11-02 03:59:13 <Apachez> finally solved my bridging issue
2023-11-02 10:08:52 <vkrishn> hi, pls help, how to undelete ext4 partition ?
2023-11-02 10:09:01 <vkrishn> is there tool like ext4magic ?
2023-11-02 10:10:35 <aron> how do you deleted the partition?
2023-11-02 10:15:34 <vkrishn> aron: https://tpaste.us/kXKw 
2023-11-02 10:17:00 <vkrishn> and https://tpaste.us/O8YX 
2023-11-02 10:18:24 <vkrishn> in https://tpaste.us/kXKw, i missed step mkfs.ext4 downloads.img
2023-11-02 10:20:05 <aron> it seems to me, you deleted the files, not the partition
2023-11-02 10:20:43 <vkrishn> then i immediately unmounted the .img and set chmod 444 to it
2023-11-02 10:20:58 <aron> does not matter
2023-11-02 10:21:03 <vkrishn> ok
2023-11-02 10:21:26 <aron> the file data is still on the disk, ofc
2023-11-02 10:21:34 <aron> but it's hard to recover them
2023-11-02 10:22:16 <vkrishn> i tried photorec, but is terrible when comes to recovering with original file names
2023-11-02 10:22:27 <aron> if your fs was fresh (newly formatted, unfragmented), then you can achieve a high success ratio, depending on the files you had
2023-11-02 10:22:44 <aron> hehe, yeah
2023-11-02 10:22:49 <aron> _what_ file names
2023-11-02 10:23:33 <aron> your file names are the first "victims" on file deletion
2023-11-02 10:23:44 <aron> this is "normal"
2023-11-02 10:23:45 <vkrishn> photorec seems to get all files, but file names and directory structure are all non understable
2023-11-02 10:23:55 <vkrishn> ah, ok
2023-11-02 10:23:58 <aron> yes, sounds absolutely reasonable
2023-11-02 10:24:35 <aron> that's all you can do
2023-11-02 10:24:43 <aron> what you already done
2023-11-02 10:24:54 <aron> you got your file data back
2023-11-02 10:25:04 <vkrishn> from https://ext4magic.sourceforge.net feature list , seems its doable
2023-11-02 10:25:20 <aron> in some extent
2023-11-02 10:26:07 <aron> you *may* get more data back, if some stars were at the right place
2023-11-02 10:28:31 <aron> you can build and use ext4magic, if you want
2023-11-02 10:28:58 <aron> it's not packaged on alpine
2023-11-02 10:29:03 <vkrishn> ok, will give a try
2023-11-02 10:39:46 <vkrishn> aron: can extundelete work ?
2023-11-02 10:40:40 <aron> maybe
2023-11-02 10:42:59 <aron> (it scans the journal, which is a good thing)
2023-11-02 10:43:30 <aron> never used these, tho
2023-11-02 10:44:18 <aron> similar scenario happened to me with ext2 (there were no ext3/ext4, nor journal at that time)
2023-11-02 10:44:42 <aron> but be aware, journal size is limited
2023-11-02 10:44:57 <aron> thus, the amount of information stored there is also limited
2023-11-02 10:53:26 <vkrishn> ok, thanks
2023-11-02 13:25:19 <xgpt> Does linuxserver.io rip out the alpine linux init system and replace it with something called 's6'? 
2023-11-02 13:26:53 <Habbie> are those docker images?
2023-11-02 13:27:28 <minimal> xgpt: no AFAIK they start from an Alpine base with no init and they add s6-overlay
2023-11-02 13:27:36 <xgpt> wait
2023-11-02 13:27:36 <xgpt> woah
2023-11-02 13:27:37 <minimal> so they're not ripping out anything
2023-11-02 13:27:40 <xgpt> you can ...
2023-11-02 13:27:46 <xgpt> you can have alpine without init?
2023-11-02 13:27:52 <xgpt> but how does that start a system?
2023-11-02 13:27:55 <xgpt> 
ACTION mind blown

2023-11-02 13:27:59 <minimal> yes of course
2023-11-02 13:28:23 <minimal> you do know the distinction between OS containers and Application containers?
2023-11-02 13:28:35 <xgpt> Habbie they are a collection of docker images, they are relatively standardized within their little garden of containers, and kept up to date as far as I can tell. 
2023-11-02 13:28:59 <xgpt> those terms look familiar but not really. minimal
2023-11-02 13:29:28 <minimal> you don't technically  nbeed an init if you're just running a single application inside a container
2023-11-02 13:30:23 <xgpt> ohhh, because docker can start  the command itself right?
2023-11-02 13:30:45 <minimal> I assume their base starts with using the Alpine minirootfs, which has no init or kernel
2023-11-02 13:31:04 <xgpt> so if...let's say I'm running a app container that's just a python hello world script, the PID 1 would be "python3 /data/hello_world.py" right?
2023-11-02 13:32:19 <minimal> the reason some sort of init is generally added is for signal handling, so that any child processes are shutdown when the parent process dies
2023-11-02 13:35:31 <xgpt> aah, but that seems like....and I feel like one of those unix neckbears for saying this, but a lot of 'overhead' for something as simple as a hello_world script
2023-11-02 13:35:39 <xgpt> is that the thinking behind not running the init?
2023-11-02 13:36:13 <minimal> docker comes with a minimal (lol) init to do just the basics
2023-11-02 13:36:26 <xgpt> what is that minimal init called?
2023-11-02 13:37:24 <minimal> strangely enough "docker-init" ;-) though there are other minimal (lol) alternatives like tini
2023-11-02 13:37:51 <xgpt> gotcha
2023-11-02 13:37:51 <minimal> look at the manpage for "docker run --init"
2023-11-02 13:38:28 <xgpt> so is docker-init something that's installed by default or that would need to be added to a base alpine container? this linuxserverio project seems to have standardized on s6 init
2023-11-02 13:38:33 <xgpt> which seems like a good minimal init system
2023-11-02 13:39:14 <minimal> no, they have standardised on s6-overlay, not on s6
2023-11-02 13:39:17 <ikke> xgpt: docker can inject it
2023-11-02 13:39:31 <ikke> docker run --init would inject an init binary in the container
2023-11-02 13:41:09 <minimal> I had a refresher look at the linuxserver Dockerfile for their Alpine base - they're using "apk" to create a base Alpine inside and then adding s6-overlay (and some other stuff) to that
2023-11-02 13:41:26 <xgpt> ikke inject?!
2023-11-02 13:41:26 <minimal> s/inside/install/
2023-11-02 13:41:39 <xgpt> minimal can you link me to that dockerfile?
2023-11-02 13:42:01 <minimal> lmgtfy?
2023-11-02 13:45:37 <Habbie> isn't docker-init tini?
2023-11-02 13:45:59 <ikke> I believe so
2023-11-02 13:49:12 <minimal> ah yeah it is (softlink to /sbin/tini-static)
2023-11-02 13:50:20 <minimal> well there's other alternatives like dumb-init
2023-11-02 13:52:48 <xgpt> oh
2023-11-02 13:53:21 <xgpt> minimal just wanted to make sure we're looking at the same Dockerfile :P thought you had it pulled up
2023-11-02 13:53:54 <minimal> https://github.com/linuxserver/docker-baseimage-alpine/blob/master/Dockerfile
2023-11-02 14:20:01 <vkrishn> is there something similar to "telinit 2" on al ?
2023-11-02 14:20:52 <vkrishn> usually on debian i do telinit 2 and then telinit 5 if desktop shows some problem
2023-11-02 14:22:49 <vkrishn> or just create new runlevel in /etc/runlevels ?
2023-11-02 14:30:04 <minimal> vkrishn: telinit is not enabled in busybox
2023-11-02 14:30:30 <minimal> Alpine uses busybox's init in combination with openrc
2023-11-02 14:31:52 <vkrishn> ok, thanks
2023-11-02 14:33:45 <minimal> also looking as Busybox's init source code I see "Busybox init doesn't support multiple runlevels. The runlevels field of the /etc/inittab file is complete ignored by Busybox init."
2023-11-02 14:35:50 <vkrishn> openrc does not support adding more /etc/runlevels ?
2023-11-02 14:36:53 <minimal> as I said Alpine uses a combination of bbinit+openrc, it's not soley bbinit or soley openrc
2023-11-02 14:37:22 <minimal> openrc in general has an init but it is not used by Alpine
2023-11-02 15:08:04 <vkrishn> ok, managed to setup and start xfce in AL 3.18.3 x86_64, how do i restart desktop ?
2023-11-02 15:14:14 <vkrishn> or shutdown just the desktop ?
2023-11-02 15:50:19 <vkrishn> hmm, i logged out of desktop, then did - "/etc/init.d/lightdm stop"
2023-11-02 15:50:44 <vkrishn> kinda works
2023-11-02 16:04:27 <stonks3141> vkrishn: just fyi, openrc has rc-service(1) so you can run "rc-service lightdm stop" instead
2023-11-02 19:41:16 <moat_> why doesn't linux-edge have CONFIG_DEBUG_FS=y
2023-11-02 19:43:00 <ikke> #14476
2023-11-02 19:43:16 <ikke> https://gitlab.alpinelinux.org/alpine/aports/-/issues/14476
2023-11-02 19:45:10 <moat_> oh no reason?
2023-11-02 19:45:24 <moat_> can i just make a pull request then?
2023-11-02 21:47:10 <Saijin_Naib[m]> Anyone using the new Gecko/FireFox based Midori 11? Is it worth checking out? Do we need an aport (I'm making a big assumption that it is easy enough for me to do) 
2023-11-02 21:47:56 <Saijin_Naib[m]> I did find an actively developed GTK/python/aria2 download manager I'm going to put into testing. Works excellent 
2023-11-02 21:50:36 <invoked> i'm confused why debugfs is needed for controlling the clock on nouveau.
2023-11-02 22:50:04 <jacksonchen666> hi, i'd like to know what i should do to make a package request (for opensearch), other than create a new issue at aport gitlab
2023-11-02 22:55:44 <ikke> Not a lot else, but just making a package request does not necessarily mean someone is going to packages it
2023-11-02 23:09:09 <minimal> I'd thought about packaging opensearch, but it's a big piece of work
2023-11-02 23:15:50 <minimal> especially as packaging opensearch would also imply packaging logstash and kibana
2023-11-03 04:12:14 <simple_utilisateur> hello world I created a simple distribution with the Linux kernel + busybox, I would now like to create universal applications (which contain all the dependencies statically or perhaps a solution that you will offer me)
2023-11-03 04:14:32 <FLOPS> app store, snaps or flatpacks
2023-11-03 04:18:39 <simple_utilisateur> the problem is the dependence of the libraries...
2023-11-03 04:18:39 <simple_utilisateur> but I only know how to compile normally, I don't know how to create flatpaks or snaps etc.
2023-11-03 04:18:39 <simple_utilisateur> Let's imagine that I retrieve the sources from Mozzilla, I would like to never use dynamic libraries for compilation but only static links.
2023-11-03 04:20:12 <simple_utilisateur> but in the end compiling and installing software that uses static libraries must certainly be universal... well I think
2023-11-03 04:22:43 <simple_utilisateur> could you guide me please?
2023-11-03 04:24:55 <fluix> this is offtopic for this channel and I don't know better places to ask (perhaps #linux on libera?), but this will be specific to different languages and projects
2023-11-03 04:25:27 <fluix> some will default to static linking, others to dynamic, many to a mix of both (libc being dynamic for instance)
2023-11-03 04:26:24 <quinq> simple_utilisateur, for C software, you just have to tell the linker to link statically (usually with -static)
2023-11-03 04:27:20 <quinq> No, the libc isn't dynamically linked by nature, especially musl, it'll link statically if you ask your toolchain to
2023-11-03 04:29:48 <simple_utilisateur> I thank you friends.
2023-11-03 04:29:48 <simple_utilisateur> So I try to approach the problem from a different approach.
2023-11-03 04:29:48 <simple_utilisateur> It's true that I'm looking for people like you with great knowledge of OS.
2023-11-03 04:29:48 <simple_utilisateur> I have already built several LFS, but I haven't learned anything about it.
2023-11-03 04:30:17 <quinq> Then I would recommend learning about compilers a bit, starting with simple things
2023-11-03 04:32:01 <simple_utilisateur> but technically it is possible to compile a program, with all these static dependencies?
2023-11-03 04:32:03 <quinq> What's an ABI, what are syscalls, elf format
2023-11-03 04:32:15 <quinq> Sure it is
2023-11-03 04:33:56 <simple_utilisateur> connais tu linux from scratch ? j'en ai construis plusieurs .
2023-11-03 04:33:57 <simple_utilisateur> Un gnu/linux universel.
2023-11-03 04:33:57 <simple_utilisateur> mais j'essaye de conceptualiser quelque chose de plus grand que moi.
2023-11-03 04:33:58 <quinq> Note that some program are actually using runtime dynamic libraries, with dlopen() for example, and you'd have to rewrite that for getting rid of dynamic linking
2023-11-03 04:34:24 <simple_utilisateur> do you know linux from scratch? I have built several of them.
2023-11-03 04:34:24 <simple_utilisateur> A universal gnu/linux.(google translate sorry :)
2023-11-03 04:34:24 <simple_utilisateur> but I'm trying to conceptualize something bigger than myself.
2023-11-03 04:34:52 <quinq> Well, you're always dependent on reality at some point, hardware
2023-11-03 04:35:21 <quinq> Your software built for an amd64 will not run out of the box on an arm system for example
2023-11-03 04:35:53 <simple_utilisateur> i can just build a toolchain like LFS non?
2023-11-03 04:36:20 <quinq> But also your software built on a recent linux might not work on older versions of the kernel because it could be missing some syscalls used by your program
2023-11-03 04:36:23 <quinq> etc.
2023-11-03 04:36:36 <simple_utilisateur> mhhm..i see...
2023-11-03 04:36:56 <simple_utilisateur> quinq ure awesome thx u a lots!
2023-11-03 04:37:06 <FLOPS> https://0install.net
2023-11-03 04:37:11 <quinq> A toolchain is what builds your program, LFS is a guide for building a linux kernel and software around (including a gcc or clang toolchain)
2023-11-03 04:38:08 <simple_utilisateur> yes its that
2023-11-03 04:38:40 <fluix> https://justine.lol/ape.html is fun too
2023-11-03 04:40:03 <simple_utilisateur> so my idea is suddenly impossible...
2023-11-03 04:40:03 <simple_utilisateur> I'm very sad lol
2023-11-03 04:42:44 <quinq> You can also study history, there are several “experimental” distribution that already deal or dealt with fully static linking
2023-11-03 04:43:24 <quinq> But also this is software, you can do *most* of anything really
2023-11-03 04:43:53 <quinq> That doesn't mean it's a practical thing to do outside of specific target or playing with bits
2023-11-03 04:44:57 <simple_utilisateur> for me static linking is the future.
2023-11-03 04:44:57 <simple_utilisateur> these are heavier programs, we agree on this point. but in terms of compatibility it would be the best.
2023-11-03 04:47:19 <quinq> Depends on the context and target
2023-11-03 04:47:53 <simple_utilisateur> quinq --> I am delighted to discuss with someone so cultured on this subject :)
2023-11-03 04:48:18 <quinq> Sure, it's simpler and faster to load for example, and you don't have to worry about missing dependencies
2023-11-03 04:49:07 <quinq> But on a whole system scale, you need a good and strong package system that would rebuild most things when some module is updated (any library)
2023-11-03 04:49:38 <quinq> Building can be speedup by keep build objects around for example
2023-11-03 04:50:03 <quinq> Anyway, start with smaller simpler things, and grow from that :)
2023-11-03 04:50:23 <quinq> simple_utilisateur, I'm not very cultured about it, I just try to share the few things I can
2023-11-03 04:50:42 <quinq> But it's bedtime now
2023-11-03 04:51:16 <quinq> Good luck, and stick around, you should get help from several people around here I guess
2023-11-03 04:51:29 <simple_utilisateur> you have a way of seeing things that goes in the same direction as me.
2023-11-03 04:51:29 <simple_utilisateur> and thank you again :)
2023-11-03 04:51:29 <simple_utilisateur> Good night.
2023-11-03 04:51:48 <quinq> cheers :)
2023-11-03 10:18:15 <arka> Is there a way to dump my current 'configuration' of sorts?
2023-11-03 10:18:50 <arka> Gonna try moving from this current NVR design I made to frigate, so I've gotta install some debian stuff instead
2023-11-03 10:19:03 <arka> Maybe I should have written more documentation
2023-11-03 11:49:36 <msiism> If I want to run conky with my personal configuration, do I just put a configuration file where I see it fit and then run `conky -c $conig_file` or is there some "standard" way of doing that?
2023-11-03 12:00:21 <msiism> Hm… apparently, ~/.conkyrc is what you use then.
2023-11-03 12:38:53 <nextgens> Hi, I am trying to understand why testing/dovecot-fts-flatcurve hasn't been rebuilt for aarch64. https://pkgs.alpinelinux.org/package/edge/testing/aarch64/dovecot-fts-flatcurve lists 0.3.4-r2 whereas most other archs are 0.3.4-r3. Any hint?
2023-11-03 12:40:13 <nextgens> https://pkgs.alpinelinux.org/packages?name=dovecot-fts-flatcurve
2023-11-03 12:41:32 <nextgens> the APKBUILD has an arch="all" ... https://git.alpinelinux.org/aports/tree/testing/dovecot-fts-flatcurve/APKBUILD
2023-11-03 12:42:14 <ikke> nextgens: the builder was stuck 
2023-11-03 12:43:36 <nextgens> ikke> ah. Is there any way you could poke it please?
2023-11-03 12:44:07 <ikke> I already did, just have to have a bit of patience 
2023-11-03 12:44:15 <nextgens> thank you!
2023-11-03 12:44:35 <ikke> qt5/6-webengine have been updated and take time to build 
2023-11-03 12:45:10 <nextgens> No problem, I'll just check back in a few days :)
2023-11-03 20:44:54 <freedomchaser[m]> when i start postgresql16 im getting this
2023-11-03 20:45:03 <freedomchaser[m]> https://bpa.st/2A6A
2023-11-03 21:21:03 <fluix> freedomchaser[m]: seems like something's wrong with your postgresql16 install. maybe try `apk fix`
2023-11-03 21:21:39 <freedomchaser[m]> i was on edge and reverted back to v3.18
2023-11-03 21:21:48 <fluix> apk upgrade -a
2023-11-03 21:21:51 <freedomchaser[m]> the thing is i did a fresh install
2023-11-03 21:22:16 <freedomchaser[m]> so the problem is something to do with probably the init script and the path variables
2023-11-03 21:22:27 <fluix> is this true: "sh: /usr/bin/initdb: not found"
2023-11-03 21:22:31 <fluix> does that file exist?
2023-11-03 21:23:25 <fluix> err, 3.18 doesn't have postgresql16
2023-11-03 21:23:27 <freedomchaser[m]> yeah it doesnt exist because its moved to /usr/libexec/postgresql16/initdb
2023-11-03 21:23:29 <freedomchaser[m]> and all the other stuff
2023-11-03 21:23:39 <freedomchaser[m]> yeah i downgraded to postgresl15 i dont need the stuff for 16
2023-11-03 21:23:44 <freedomchaser[m]> i jsut wanted something that worked
2023-11-03 21:24:02 <freedomchaser[m]> didnt want to use another distribution or bsd so chose alpine as it was for a small vps
2023-11-03 21:24:23 <fluix> 15 should work on 3.18 and 16 on edge, did you `apk upgrade -a` when changing between versions
2023-11-03 21:24:37 <fluix> *versions of alpine
2023-11-03 21:25:07 <freedomchaser[m]> yes
2023-11-03 21:25:44 <fluix> hm, maybe old versions of the config files
2023-11-03 21:26:22 <freedomchaser[m]> na i remembered to remove all the directories for postgresql
2023-11-03 21:26:24 <fluix> don't recall if there's some special command: you can try just rm'ing them or looking in `apk audit` to see what's different
2023-11-03 21:26:33 <freedomchaser[m]> its fine now though
2023-11-03 21:26:42 <freedomchaser[m]> i guess edge is not meant for production right
2023-11-03 21:26:53 <fluix> oh, gotcha
2023-11-03 21:27:07 <fluix> no guarantee of stability but people use it sometimes
2023-11-03 21:28:01 <freedomchaser[m]> yeah ofc i get it normally if i have my own hardware i do like going on bleeding edge on something like gentoo but thats because i can control a lot of things
2023-11-03 21:29:37 <fluix> looks like https://gitlab.alpinelinux.org/alpine/aports/-/issues/14652 perhaps?
2023-11-03 21:29:54 <fluix> not quite
2023-11-03 21:30:26 <fluix> I think packages _should_ work or be reported broken (either with a pending MR or an issue) in edge because stable releases are cut from it
2023-11-03 21:30:36 <fluix> (well, also because we don't want broken packages in edge)
2023-11-03 21:31:01 <fluix> please report it if you can reproduce it (and I may give it a try later)
2023-11-03 22:02:29 <tyler[m]123> <Saijin_Naib[m]> "I'm going to assume they must..." <- Tbh, I can't think of any, but I haven't participated in the development of the build system or anything. I would also like to know.
2023-11-03 22:04:46 <tyler[m]123> For example, Debian doesn't ship `.deb`s and install them on memory when you boot - that is slow and wastes too much memory. Instead it just loads the rootfs from the compressed squashfs file and makes an overlay.
2023-11-03 22:04:46 <tyler[m]123> Maybe I can suggest to a developer (or maybe even implement, if I have time) the alternative I said
2023-11-03 22:11:18 <tyler[m]123> Well, I mentioned Debian just as an example. In the end, seems to me that it's very efficient to ship a squashfs with a preconfigured rootfs and then just mount an overlay.
2023-11-03 22:11:43 <tyler[m]123> I made my own Linux + busybox build (with the help of Minimal Linux, on GitHub)
2023-11-03 22:12:13 <tyler[m]123> Before I added a squashfs + overlayfs, and everything was just on memory, my ~130MB ROM needed 1024MB of RAM to run
2023-11-03 22:12:30 <tyler[m]123> After squashfs + overlayfs, works even with 256MB (maybe lower, I havent tested)
2023-11-03 22:13:30 <tyler[m]123> As for performance, I expect it to be a little slower because disk is more often than not slower than memory. But to compensate, it takes a long time to install everything from disk into memory anyways, plus there is a significant reduction in amount of RAM required to boot the OS
2023-11-03 22:13:57 <tyler[m]123> * + busybox + other stuff build (with
2023-11-03 22:17:36 <tyler[m]123> I'm building my system rn to demonstrate this a little more
2023-11-03 22:18:27 <tyler[m]123> In the future I will also make an Alpine based version of this (install APK + other packages on squashfs, then mount overlay)
2023-11-03 22:40:58 <tyler[m]123> 
ACTION uploaded an image: (25KiB) < https://matrix.org/_matrix/media/v3/download/matrix.f1gh7.club/jdRfpBdMKxjKsORZVLpQCkae/2023-11-03_19-27-51.png >

2023-11-03 22:41:04 <tyler[m]123> I couldn't get standard Alpine Linux 3.18.4 to work with less than 200MB of RAM - anything less results in missing files or kernel panic.... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/pIRMByTlcrzSDMclZmtHdvnK>)
2023-11-03 22:42:07 <tyler[m]123> I will try with my build once it finishes for comparison. It should be bigger than Alpine. Although it's not really a 1:1 comparison for many reasons (different kernel, openrc is not used on my build, etc) - it should highlight the memory usage results between the two
2023-11-04 00:35:32 <tyler[m]123> 
ACTION uploaded an image: (7KiB) < https://matrix.org/_matrix/media/v3/download/matrix.f1gh7.club/tHqFjoRqtqVdvnGATkeNDhbM/2023-11-03_21-35-35.png >

2023-11-04 00:35:40 <tyler[m]123> For comparison:... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/zYHEjFKeJAWRYKgfQNaFlcQA>)
2023-11-04 00:36:37 <tyler[m]123> My other alpine based distro (470MB) needs 2GB of RAM just to boot (because of the apkovl installation). While this one, using squashfs with a sysroot, needs no more than 80MB to boot and function properly
2023-11-04 00:37:12 <tyler[m]123> I think this should be debated at least, it's a very good option IMO
2023-11-04 00:37:28 <tyler[m]123> Even though Alpine itself is small, it would make mkimage builds much more optimized
2023-11-04 00:43:33 <Apachez> and you can compress is somewhat 
2023-11-04 00:44:40 <Apachez> something like this can be handy if you want to optimize the size of your squashfs and not sacrifice too much performance: xz -Xbcj x86 -b 256k -always-use-fragments -no-recovery
2023-11-04 00:44:59 <Apachez> a blocksize of 1M would of course make even smaller image but the read performance would be worser than with 256k
2023-11-04 00:50:47 <tyler[m]123> <Apachez> "and you can compress is somewhat..." <- Yeah, on the kernel there are options to enable ZSTD, XZ, and many other compression methods for squashfs
2023-11-04 00:51:03 <tyler[m]123> in this case I just used gzip for no particular reason
2023-11-04 00:53:24 <tyler[m]123> I'll try to make a base alpine clone using this method
2023-11-04 01:00:20 <Apachez> try using xz
2023-11-04 01:00:30 <Apachez> zstd is claimed to be faster but it wont compress as much
2023-11-04 01:00:52 <Apachez> so if performance is a prio then zstd is the sweetspot today mainly because it can fully utilize multicore both for compression and decompression
2023-11-04 01:01:19 <Apachez> the others are often limited to singlecore at least for decompression such as xz while multicore will be utilized for compression with xz
2023-11-04 01:01:24 <Apachez> with mksquashfs that is
2023-11-04 01:01:50 <Apachez> rumours has it also there are issues if you try to do high compression with zstd
2023-11-04 01:01:59 <Apachez> like decompression memory usage will skyrocket
2023-11-04 01:02:20 <Apachez> but I dunno if thats some legacy claims as in if they are valid for a current kernel and the version of zstd that uses
2023-11-04 01:03:11 <Apachez> when it comes to squashfs you also have a couple of kernel compile settings you can tweak but I so far havent seen much change by changing them
2023-11-04 01:03:25 <Apachez> I think they will mainly affect if you use your squashfs for multi-io access
2023-11-04 01:03:51 <Apachez> if you use it as a rootfs most stuff will get cached anyway with the system filecache
2023-11-04 01:07:21 <tyler[m]123> I'm making a alpine clone that embraces the squash rootfs, will try using xz as well
2023-11-04 02:14:51 <tyler[m]123> I can't get /dev/sr0 to show on alpine initrd for some reason... Once I figure that out I have the final system
2023-11-04 02:18:24 <dwfreed> tyler[m]123: does the system have an actual CD drive?
2023-11-04 02:18:41 <tyler[m]123> it's an ISO, booting from qemu
2023-11-04 02:18:57 <tyler[m]123> In my other linux build, it just shows up after mdev -s
2023-11-04 02:19:12 <tyler[m]123> I'll try to find the original alpine initramfs and see what they're doing
2023-11-04 02:19:21 <tyler[m]123> *init script actually
2023-11-04 02:19:34 <dwfreed> it's possible to boot an iso off a "hard disk" too
2023-11-04 02:20:07 <dwfreed> so rather than looking for sr0 specifically, you should just be looking for wherever your media is
2023-11-04 02:20:26 <dwfreed> (because it could be sr1 even, not that dual CD systems are common anymore)
2023-11-04 02:22:09 <tyler[m]123> In this case I'm just running a test, not a final distro. So it's fine. But neither /dev/sr0 nor /dev/cdrom are showing up... so there's that
2023-11-04 02:22:59 <dwfreed> I mean, see point that qemu may not actually be presenting the iso as a cd drive
2023-11-04 02:24:52 <tyler[m]123> I'm passing it with -cdrom alpine-squash-rootfs.iso. I use this same command for the other Linux I built, and it shows on /dev/sr0. That linux is built with generic config + squashfs and overlayfs though. Alpine's kernel config is very different, so could be that.
2023-11-04 02:28:31 <tyler[m]123> I'm almost sure this is a module issue
2023-11-04 02:28:40 <dwfreed> alpine probably doesn't put the driver for it in initramfs, as it's not expecting to need it
2023-11-04 02:30:10 <tyler[m]123> It's not pure alpine, I downloaded the packages and made the ISO myself. The drivers are apparently here, "sr-mod", "cdrom", etc. But after modprobe + mdev -s, I still don't get /dev/sr0
2023-11-04 02:30:31 <tyler[m]123> Do you know where I can find the code for their init script?
2023-11-04 02:30:44 <dwfreed> because you need the qemu cdrom drivers specifically
2023-11-04 02:30:46 <tyler[m]123> There's a lot of repos in the GitLab
2023-11-04 02:31:25 <dwfreed> start with aports, look at where it gets the files for the initramfs generator
2023-11-04 02:34:31 <tyler[m]123> I think they generate it dynamically with mkinitfs
2023-11-04 02:35:14 <dwfreed> yes, so look at what package contains mkinitfs, and then look at the package description for that package in aports, which will probably tell you what git repo it comes from
2023-11-04 02:35:27 <dwfreed> or it'll be a dependency of mkinitfs's package
2023-11-04 02:38:16 <tyler[m]123> Oof... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/gDtVPINKnlYUIxZnUVCczTsL>)
2023-11-04 02:38:25 <tyler[m]123> Found it but no idea what this is
2023-11-04 02:39:08 <dwfreed> $100 says it's a script
2023-11-04 02:40:40 <tyler[m]123> I think I might just use my own kernel lol
2023-11-04 02:40:53 <tyler[m]123> I don't understand what's going on in this init
2023-11-04 02:41:15 <dwfreed> it's really not complicated
2023-11-04 02:42:58 <tyler[m]123> This is just a section
2023-11-04 02:43:02 <tyler[m]123> The whole this is 880 lines
2023-11-04 02:43:03 <dwfreed> and I was wrong, it's actually C: https://gitlab.alpinelinux.org/alpine/mkinitfs/-/blob/master/nlplug-findfs/nlplug-findfs.c?ref_type=heads
2023-11-04 02:43:17 <dwfreed> not surprising, it's got to be pretty generic
2023-11-04 02:43:40 <dwfreed> there's a fair amount that goes into startup for alpine especially with the different modes like diskless
2023-11-04 02:44:35 <tyler[m]123> I'm gonna try with the kernel I built earlier just to make sure it's actually something to do with kernel and modules
2023-11-04 02:44:49 <tyler[m]123> Yeah, with my kernel it worked
2023-11-04 02:45:08 <tyler[m]123> Alpine switched a lot of things from built into the kernel into modules
2023-11-04 02:45:41 <tyler[m]123> Maybe the nlplug-findfs tool does the job of figuring out where the boot is, making a device and mounting it
2023-11-04 02:47:47 <tyler[m]123> InitRD is also very big due to the separate kernel modules in my case
2023-11-04 02:48:00 <tyler[m]123> That might be something to optimize also, along with the squashfs
2023-11-04 02:52:31 <tyler[m]123> The build script I made (not working yet bc of alpine kernel): https://github.com/rdbo/alpine-squash-rootfs/blob/master/build.sh
2023-11-04 10:02:07 <vova> Hi everybody, have any of you built ffmpeg successfully with --enable-mmal ? I'm trying since yesterday to do so, but everytimes I end up with `ERROR: mmal not found`, so I tried to install raspberrypi-userland-{dev,libs,static}, but I still get the same error (I'm building it on a raspberry pi 3) 
2023-11-04 10:35:30 <autonokia> Has anyone here run frigate in a container on alpine?
2023-11-04 10:35:47 <autonokia> (The NVR)
2023-11-04 10:36:52 <autonokia> Looks like some people have. Hell yeah.
2023-11-04 12:58:26 <Forza> Hi. Anyone here have experience with nfs-ganesha? 
2023-11-04 12:58:57 <Forza> I'm considering if we can get it on Alpine 
2023-11-04 13:00:17 <Forza> It's a user-space nfs server, often use with distributed filesystems. 
2023-11-04 13:22:39 <Forza> Do all ptython apps have to be named py3-xxx?
2023-11-04 13:22:52 <Forza> Id like to package https://github.com/supakeen/pinnwand
2023-11-04 13:41:38 <Forza> The only option to install is via pip, even via sources . Will that work on Alpine? 
2023-11-04 14:03:40 <ikke> Forza: python libraries should be prefixed with py3-. If it's an application, it would just use the original name
2023-11-04 14:04:53 <Forza> This is just an application.
2023-11-04 14:05:09 <Forza> Its a pastebin type server
2023-11-04 14:06:00 <Forza> But it only seem to have build/install instructions for pip in a venv setup
2023-11-04 14:46:14 <omni> looking at how other python packages are packaged may help you
2023-11-04 15:12:44 <Forza> omni: yes thank you. I've been going that haven't yet found a package that uses pip. Most use setup.py and some used some other thing
2023-11-04 15:13:34 <Forza> According to https://docs.python.org/3/installing/index.html#installing-index the pip style installer is preferred and that setup.py might get deprecated in the future
2023-11-04 15:21:52 <ikke> Forza: we use gpep517 style installations
2023-11-04 15:22:21 <ikke> I mean, pep517, which gpep517 is an implementation for
2023-11-04 15:22:57 <Forza> https://peps.python.org/pep-0517/ this?
2023-11-04 15:23:10 <Forza> I saw some abuilds using pep517
2023-11-04 16:29:31 <Forza> I started an apkbuild based on another package https://gist.tnonline.net/IU
2023-11-04 16:32:53 <Forza> Build fails and I'm not sure how to proceed https://bpa.st/ULFA
2023-11-04 16:34:17 <ikke> It expects *.pyc files, but for some reason they are not installed
2023-11-04 16:37:34 <Forza> Can i skip them
2023-11-04 16:40:56 <Forza> I remove removed subpackages="$pkgname-pyc"
2023-11-04 16:41:07 <Forza> Seems to work
2023-11-04 16:41:49 <ikke> It is an indication something is not working as expected though
2023-11-04 16:45:32 <donoban> anyone experienced segfault on sway? I've just had second on a week :(
2023-11-04 16:46:54 <donoban> where should it save the core dump? uhm, I think that was something like sway-server 
2023-11-04 16:50:10 <Forza> How can i see whats inside an apk file?
2023-11-04 16:50:21 <Forza>  pinnwand-1.5.0-r0.apk that was just built
2023-11-04 16:54:04 <ikke> Forza: apk manifest <filename>
2023-11-04 16:56:55 <Forza> https://gist.tnonline.net/IE
2023-11-04 16:57:06 <Forza> Looks empty
2023-11-04 16:57:13 <ikke> yes
2023-11-04 17:10:55 <Forza> Heres the whole aports build tree https://mirrors.tnonline.net/misc/pinnwand/abuild2.log
2023-11-04 17:11:11 <Forza> https://mirrors.tnonline.net/misc/pinnwand/
2023-11-04 17:55:51 <Forza> I think maybe pdm is missing
2023-11-04 17:56:45 <Forza> But it has some deps in testing only 
2023-11-04 18:57:18 <Forza> I'll give up for today :o. Can't figure out what next steps are so I made a request on upstream github instead 
2023-11-04 18:57:28 <Forza> Last attempt gave https://mirrors.tnonline.net/misc/pinnwand/abuild.log
2023-11-04 19:27:04 <Forza> This is how a successful install looks on anoter system using pip: https://bpa.st/VRUQ
2023-11-04 20:54:09 <papiris[m]> I'm porting snap-pac https://github.com/wesbarnett/snap-pac to Alpine / apk.... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/cLEmAJDLUElPGxGdwbFqrQhz>)
2023-11-04 21:28:02 <moat> i added more detail to my MR for debugfs on the edge kernel
2023-11-04 21:28:15 <moat> would be nice to get this merged!
2023-11-05 03:20:38 <Saijin_Naib[m]> Was anyone around involved with packaging/maintaining SeaMonkey Internet Suite?
2023-11-05 03:20:44 <Saijin_Naib[m]> I'm half-tempted to pick it back up for testing
2023-11-05 03:20:53 <Saijin_Naib[m]> I miss the stupid thing, and it does have some advantages
2023-11-05 06:06:55 <coc0nut> Saijin_Naib[m], what advantages? 
2023-11-05 07:45:39 <Kladky> papiris[m]: Is there a public repo for this?
2023-11-05 09:08:32 <papiris[m]> Yes, [here](http://gitlab.com/papiris/snap-apk).
2023-11-05 09:08:32 <papiris[m]> This will be my first package, so advice is very welcome
2023-11-05 09:15:26 <quinq> First, why not naming apk-snap instead? :)
2023-11-05 09:25:08 <nextgens> ikke> Hi, just to let you know, https://pkgs.alpinelinux.org/package/edge/testing/aarch64/dovecot-fts-flatcurve has not been rebuilt yet. Would you mind checking the status of the builder again please?
2023-11-05 09:38:24 <papiris[m]> <quinq> "First, why not naming apk-snap..." <- Sure, we can do that. Naming is tentative, I just replaced `pac` from pacman with `apk` :P
2023-11-05 09:38:24 <papiris[m]> It's a python3 program, but since it isn't a _module_, I guess we don't need the `py3-` prefix?
2023-11-05 09:38:24 <papiris[m]> Currently reading https://wiki.alpinelinux.org/wiki/Creating_an_Alpine_package  :)
2023-11-05 09:56:15 <quinq> papiris[m], I would break down the first commit a bit
2023-11-05 09:56:43 <quinq> separate the renaming stuff from the actual changes
2023-11-05 09:59:13 <quinq> Also the sh scripts could be improved imo
2023-11-05 09:59:28 <quinq> why making it an sh script at all if it's for calling python?
2023-11-05 09:59:51 <quinq> Why assuming env is in /usr/bin, and not sh being in /bin?
2023-11-05 10:00:24 <quinq> Why calling something in /usr/local/, as this isn't a standard installation path for apk (AFAIK)
2023-11-05 10:00:46 <quinq> It seems that the Makefile needs more rework
2023-11-05 10:01:11 <quinq> But I guess the work isn't finished :)
2023-11-05 10:22:05 <papiris[m]> <quinq> "why making it an sh script at..." <- The `apk_snap.py` script itself is rather massive, so I didn't want to put it in the `/etc/apk/commit_hooks.d/` dir :)
2023-11-05 10:22:05 <papiris[m]> Is there some cleaner way of achieving this?
2023-11-05 10:22:25 <papiris[m]> btw, should I move this convo over to alpine-devel?
2023-11-05 10:24:11 <quinq> papiris[m], you can make symlinks
2023-11-05 11:26:46 <papiris[m]> quinq thanks for your guidance :)... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/TDnGaeYvkMuIoKswjkrLAdZB>)
2023-11-05 11:31:34 <papiris[m]> I guess aliases for apk-snap pre and apk-snap post can be declared in ~/.profile, but I'd rather not mess with user files
2023-11-05 11:35:58 <quinq> Of course not, what's in ~ doesn't belong to packages :)
2023-11-05 12:07:56 <papiris[m]> <quinq> "separate the renaming stuff from..." <- commits are now split :) 
2023-11-05 12:50:23 <Ermine> Does apk have custom post-install hooks?
2023-11-05 12:56:09 <Guest5854> Hello, I am trying to spice-vdagent on Alpine linux installed from virt iso and it seems it is missing uinput. I wanted to rebuild the kernel but abuild seems to be building both linux-lts and linux-virt, while I just need the latter. Is there a way to tell it to build just virt?
2023-11-05 12:57:37 <papiris[m]> Ermine If you mean specific post-install hooks controlled by the individual package, then yes: https://wiki.alpinelinux.org/wiki/APKBUILD_Reference#$pkgname.post-install 
2023-11-05 12:57:37 <papiris[m]> If you mean custom global hooks, then yes also. Scripts placed in `/etc/apk/commit_hooks.d/` are executed on each commit. I'm not sure how to specify that a hook should only be executed before/after commit, so if someone could read this part of the `apk-tools` source code and tell us, that'd be highly appreciated :) https://github.com/alpinelinux/apk-tools/blob/3bf521ba06711394d6ac8a12e5e55b452a7e5107/src/commit.c#L232-L270
2023-11-05 13:08:29 <papiris[m]> moving discussion about porting of snap-pac to alpine-devel channel :)
2023-11-05 14:16:24 <Saijin_Naib[m]> coc0nut aside from all-in-one UX, providing Mail, Web, IRC, Matrix, News group, RSS in one is nice, and the SeaMonkey mail client allows for multi-select of folders which nothing else I know of does, and it handles picky servers better than Thunderbird with regards to renaming or deleting folders
2023-11-05 14:19:42 <Saijin_Naib[m]> I basically write email for a living, so stupid stuff like that is really important to me, unfortunately 🤣
2023-11-05 14:21:47 <minimal> write email for a living? so you're a spammer? ;-)
2023-11-05 14:25:36 <invoked> sounds like seamonkey is missing the following functionalities: music player, crypto wallet, video conferencing, and a compiler
2023-11-05 14:38:11 <Saijin_Naib[m]> minimal 🤣 almost. Support 
2023-11-05 14:38:38 <Saijin_Naib[m]> invoked it does still have an HTML editor though, I think 🤔 
2023-11-05 14:39:24 <minimal> Saijin_Naib[m]: don't you just have "Have you tried turning it off and on again" as a form template? lol
2023-11-05 14:41:25 <Saijin_Naib[m]> minimal I do, but that doesn't apply for most support cases as it is typically issues at the data layer or survey layer folks have, and that is almost always mostly project specific 
2023-11-05 15:10:38 <alpine> how does llvm config.guess detect alpine linux ?
2023-11-05 15:42:11 <shinfray> Hello, I was wondering, is it possible to use ToyBox instead of BusyBox in Alpine Linux?
2023-11-05 15:45:15 <CO2> join #firefox
2023-11-05 15:49:35 <minimal> shinfray: toybox is packaged for Alpine but I don't believe it can be used as a replacement however - only the "toybox" binary is installed, there is no infrastructure to put links in place for common utils like basename/cat/chmod etc
2023-11-05 15:51:07 <minimal> and toybox doesn't provide alternatives to everything that busybox does
2023-11-05 15:52:41 <shinfray> Thanks!
2023-11-05 18:23:24 <alpineuser> hello
2023-11-05 18:23:28 <alpineuser> is anyone around? 
2023-11-05 18:23:49 <alpineuser> I need a little help with installing alpine (im new to linux but can understand things a little)
2023-11-05 18:24:40 <alpineuser> basically I already installed alpine linux on virtualbox
2023-11-05 18:24:48 <alpineuser> I installed the virt variant of it 
2023-11-05 18:24:59 <alpineuser> I also installed guest additions for vbox
2023-11-05 18:25:44 <alpineuser> but it looks like the guest additions are not working... as I am not able to re-scale my window or go into full screen mode
2023-11-05 18:35:42 <Ermine> did you start them
2023-11-05 18:36:01 <alpineuser> i am not quite sure
2023-11-05 18:36:14 <alpineuser> I ran this command: 
2023-11-05 18:36:25 <alpineuser> VBoxClient --vmsvga
2023-11-05 18:36:34 <Ermine> rc-service virtualbox-guest-additions start
2023-11-05 18:36:43 <alpineuser> ohhhh
2023-11-05 18:37:02 <Ermine> rc-service virtualbox-guest-additions-drm start
2023-11-05 18:37:25 <Ermine> I believe those commands are specified in alpine wiki
2023-11-05 18:37:52 <alpineuser> https://wiki.alpinelinux.org/wiki/VirtualBox_guest_additions
2023-11-05 18:37:56 <alpineuser> Im following this wiki
2023-11-05 18:38:15 <alpineuser> The 2nd command failed for me, it looks like I don have that (guest-additions-drm) pkg/service installed
2023-11-05 18:38:30 <minimal> Ermine: isn't that rc-service virtualbox-drm-client? (assuming virtualbox-guess-additions-x11 is installed)
2023-11-05 18:38:55 <Ermine> minimal: yeah, you're right
2023-11-05 18:39:22 <minimal> there's an open MR to make some related changes: !54812
2023-11-05 18:41:28 <alpineuser> This one? https://gitlab.alpinelinux.org/alpine/aports/-/merge_requests/54812
2023-11-05 18:42:46 <alpineuser> Should I just run all those commands?
2023-11-05 18:42:54 <alpineuser> VBoxClient --seemless
2023-11-05 18:42:58 <alpineuser> if I want seemless?
2023-11-05 18:43:10 <Ermine> alpineuser: wait a sec please
2023-11-05 18:43:31 <Ermine> minimal: time to fullfil my duties apparently
2023-11-05 18:45:18 <minimal> Ermine: there's an unrelated package's changes also in the same MR
2023-11-05 19:14:17 <Ermine> alpineuser: I've updated the page, please follow its instructions
2023-11-05 19:15:29 <alpineuser> Ermine: got it! Let me give that a go :) Thanks!
2023-11-05 19:15:58 <Ermine> alpineuser: and for some time you'll need to run /usr/sbin/VBoxClient --clipboard --checkhostversion --seamless --drangaddrop --vmsvga-session in your terminal on each boot
2023-11-05 19:22:19 <alpineuser> Ermine: okay hmm seems like I cannot run all the flags at once, running each flag individually (VBoxClient --[flag]) doesnt seem to work either
2023-11-05 19:22:43 <alpineuser> Ermine: this is after running the commands from the updated wiki page
2023-11-05 19:23:25 <Ermine> Do these commands give you any output?
2023-11-05 19:23:50 <elagost> trying to install alpine on a rockpro64 and I can get it to start the kernel, but then it says "busybox: /sbin/acpid: No such file or directory" for a bunch of different commands. I've followed the Alpine on ARM wiki page but at no point does it instruct to add the basic filesystem stuff; where would I put that to get this to boot?
2023-11-05 19:25:16 <alpineuser> Ermine: mhm they gave some output, something like showing me the status of each service, seems like VBoxClient service is started but I am not sure if the services themselves are running - I do see process ID for each service so maybe that should indicate something.
2023-11-05 19:25:36 <Ermine> Do features work?
2023-11-05 19:26:10 <alpineuser> no the VBox features are not working
2023-11-05 19:26:41 <alpineuser> (Atleast when I try to change from VirtualBox´s View -> Fullscreen mode;  menu bar item
2023-11-05 19:26:58 <Ermine> can you please send me output of those commands (via any pastebin service)?
2023-11-05 19:27:17 <Ermine> Does e.g. shared clipboard work?
2023-11-05 19:28:26 <alpineuser> actually I dont have any DE, itś just terminal so not sure how to copy/paste on this (Shift + Left Arrow doesnt work) 
2023-11-05 19:29:03 <Ermine> Ah, resolution change won't work without any kind of GUI unfortunately
2023-11-05 19:29:31 <alpineuser> Ohhhhh 
2023-11-05 19:29:49 <alpineuser> I see.... Yeah I just tried pasting something from my host machine, doesnt seem to be working
2023-11-05 19:30:00 <Ermine> that won't work either
2023-11-05 19:30:05 <ikke> yes, it requires xorg (or wayland I suppoise)
2023-11-05 19:30:10 <alpineuser> So I guess the answer is I have to install a DE? :(
2023-11-05 19:30:18 <ikke> the ttys do not have a clipboard
2023-11-05 19:30:40 <ikke> A full DE is not required
2023-11-05 19:31:27 <Ermine> If you want bigger resolution, you can use VBoxManage on your host to add custom resolution
2023-11-05 19:32:07 <Ermine> This is described in https://wiki.archlinux.org/title/VirtualBox/Install_Arch_Linux_as_a_guest#Set_optimal_framebuffer_resolution
2023-11-05 19:33:05 <alpineuser> Ermine: thanks! I will check the page out. 
2023-11-05 19:42:26 <alpineuser> How would I check If I am running a BIOS or UEFI based boot-loader? 
2023-11-05 19:42:46 <alpineuser> It looks like I might need to know this so I can configure the kernel parameters
2023-11-05 19:46:13 <quinq> Internet says you can check for the existence of /sys/firmware/efi
2023-11-05 19:47:52 <alpineuser> apparently for alpine I may be using syslinux
2023-11-05 19:48:03 <alpineuser> https://wiki.archlinux.org/title/Kernel_parameter
2023-11-05 19:49:29 <Ermine> If you're talking about your virtualbox machine, it runs with BIOS unless you configured it to run with EFI
2023-11-05 19:54:31 <alpineuser> Ermine: I would be running extlinux right? I didnt configure anything but I do see the file /boot/extlinux.conf
2023-11-05 19:55:24 <alpineuser> and atleast if I understand from this page: https://wiki.syslinux.org/wiki/index.php?title=EXTLINUX, extlinux is a variant of syslinux... hm
2023-11-05 20:00:23 <alpineuser> Hm, setting the video= parameter as specified in this page https://unix.stackexchange.com/questions/443330/how-to-resize-alpine-linux-terminal-in-virtualbox doesnt work... 
2023-11-05 20:05:29 <Ermine> alpineuser: most likely you're booting with syslinux
2023-11-05 20:06:55 <Ermine> alpineuser: in /boot/extlinux.conf, kernel parameters are located on the line beginning with APPEND
2023-11-05 20:09:53 <alpineuser> Ermine: Do you know why my resolution doesnt change even when I add the kernel parameter to the APPEND line? (I added: video=1760x990 at the end of the APPEND line, after the modules are listed)
2023-11-05 20:10:14 <alpineuser> (This is after rebooting ofc!)
2023-11-05 20:10:48 <Ermine> Did you use VBoxManage to add this resolution to your VM?
2023-11-05 20:12:28 <alpineuser> mhm
2023-11-05 20:14:23 <alpineuser> running VBoxManage getextradata, heres what i have
2023-11-05 20:14:23 <alpineuser> Key: CustomVideoMode1, Value: 1760x990x24
2023-11-05 20:16:22 <alpineuser> I additionally tried to edit the extlinux conf file again, to modify the video parameter to 1760x990@24 
2023-11-05 20:22:03 <alpineuser> OKAY!
2023-11-05 20:22:10 <alpineuser> HUZZAH! I think I figured it out!! 
2023-11-05 20:22:14 <alpineuser> https://www.virtualbox.org/manual/ch09.html#customvesa
2023-11-05 20:22:30 <alpineuser> Apparently I should be passing in the parameters listed in that url above
2023-11-05 20:22:42 <alpineuser> so itś not video=something something
2023-11-05 20:22:53 <alpineuser> but vga=[some value]
2023-11-05 20:23:01 <alpineuser> Now I have a fullscreen terminal window :D
2023-11-05 20:28:16 <alpineuser> Ermine: Thank you for all the help and suggestions!
2023-11-06 05:40:25 <Fxzxmic> vaultwarden has a new version, and it's openrc conf needs to be updated too.
2023-11-06 05:44:55 <Fxzxmic> vaultwarden-web-vault weird name, why not vaultwarden-webvault
2023-11-06 05:45:42 <Fxzxmic> or just vaultwarden-webui
2023-11-06 12:07:46 <zcrayfish> Fxzxmic: it has been flagged already
2023-11-06 13:19:08 <Guest5952> hi. i neet to  turn on hugepages, but method writted here is not working https://xmrig.com/docs/miner/hugepages
2023-11-06 13:39:40 <Fxzxmic> zcrayfish: Yes, but the openrc conf needs to be updated too.
2023-11-06 13:41:21 <zcrayfish> it's flagged too
2023-11-06 13:41:30 <zcrayfish> so hopefully the maintainer will do something about it soon
2023-11-06 14:16:23 <singham> Which crons are on in alpine by default?
2023-11-06 14:21:31 <minimal> singham: did you check?
2023-11-06 14:24:11 <singham> No counter questions please
2023-11-06 14:24:49 <minimal> singham: I'm pointing out that you're asking a question regarding something you could figure out yourself in a minute
2023-11-06 14:26:08 <singham> I have changed my setup and have forgotten the initial setup
2023-11-06 14:26:15 <singham> I want to go back go it
2023-11-06 14:29:10 <frag> 'rewind 1w'
2023-11-06 14:29:18 <frag> and press y for yes
2023-11-06 14:29:21 <frag> jk..
2023-11-06 14:35:31 <singham> ?
2023-11-06 14:51:46 <jc> I want to boot a restricted Alpine OS on an exiting Ubuntu machine for financial transactions. Alpine OS would:
2023-11-06 14:52:10 <jc> 1. limited to memory only file system
2023-11-06 14:52:44 <jc> 2. limited packages (firefore, thunderbird, password manage)
2023-11-06 14:53:29 <jc> 3. have access to only single read/writeable USB stick
2023-11-06 14:53:58 <jc> So the questions is would this feasible?
2023-11-06 14:54:41 <Exagone313> jc: you can do this with a virtual machine, but the USB part probably makes it harder (not sure how USB passthrough works)
2023-11-06 14:54:47 <jc> Thanks!
2023-11-06 14:56:23 <Exagone313> jc: apparently virtualbox can do all this
2023-11-06 14:56:46 <jc> Could the bootable Alpine OS stick have an addition writeable partition?
2023-11-06 15:18:42 <tugrik> frag :)
2023-11-06 15:47:20 <linfan> Does Alpine linux have any bugs or duscussion forums?
2023-11-06 15:48:11 <linfan> Specifically on gnome wayland pipewire issues?
2023-11-06 15:51:15 <Exagone313> linfan: there is a link named "issues" on the top right of the website
2023-11-06 15:53:14 <linfan> i had to install xorg and xfce with pulseaudio to get clean sound out of the box, with gnome wayland and pipewire it just did not work.
2023-11-06 15:55:52 <Exagone313> linfan: if you install and run pavucontrol, what do you see?
2023-11-06 16:00:05 <linfan> don't remember, now as i said on xfce4 - after some 50 installs with gnome, kinda got fed up with it, on alpine
2023-11-06 16:02:53 <linfan> but as i recall, all seemed ok, but the sound was crackled and uneven and it only worked badly on vbox with ich97 set to default. No other combination of audio worked at all. It went totally silent with hda
2023-11-06 16:04:53 <linfan> as now on xfce4 i have windows direct sound and hda and pulseaudio and it just works out of the box
2023-11-06 16:05:45 <linfan> so it is either a gnome issue or a wayland issue or a pipewire issue or a combination of either.
2023-11-06 16:07:53 <linfan> but I can live without gnome, it surely is pretty, but quite bloated. Maybe the lack of systemd also plays some part in the malfuction of pipewire.
2023-11-06 16:10:39 <Apachez> then you would love swim or whatever its nammed nowadays
2023-11-06 16:12:04 <Apachez> https://swaywm.org/
2023-11-06 16:15:55 <linfan> not that keen on tiled desktops
2023-11-06 16:42:34 <Ermine> Tried to build my stuff with gcc -fsanitize=address, and it compains at missing -lasan. Which package should I install?
2023-11-06 18:29:29 <bpalu> Hi all, is k8s 'kind' currently working on alpine? https://pkgs.alpinelinux.org/package/edge/community/s390x/kind
2023-11-06 18:30:01 <bpalu> I can't get it to create a cluster.
2023-11-06 18:30:29 <bpalu> I read through the issues on github and it seems like alpine/openrc is currently not supported
2023-11-06 22:04:33 <Joe202311a> Hi everybody! Can anybody tell me what kind of penalties you do have to expect when using 32 Bit (x86) Alpinelinux instead of 64 Bit flavor? I'd like to install Alpinelinux in a VM with 512 MB of RAM, so both x64 and x86 is an option, but which one yields the better experience? No third party binaries planned. 32 Bit is more space efficient while still being sufficient for this
2023-11-06 22:04:35 <Joe202311a> amount of RAM, on the other hand x64 provides more cpu registers, additional cpu commands and maybe something I am not aware of right now. Are there security implications? What do you recommend? Are there any experiences to share? Thanks in advance!
2023-11-06 22:07:40 <ikke> Only implication I'm aware of is aslr being less effective on 32-bits due to less address space
2023-11-06 22:08:52 <ikke> You won't run into that quickly with only 512M memory, but max allocation for a process would be 4G
2023-11-06 22:09:13 <ikke> Other issue might be less support from upstream (more and more projects decide to drop support for 32-bits)
2023-11-06 22:18:55 <Joe202311a> @ikke, thank you. I will do some research on aslr.
2023-11-06 22:23:01 <omni> Joe202311a: I run x86_64 VMs (Xen "DomUs") with 160 megs
2023-11-06 22:25:24 <omni> and then I even use some of that for overlaytmpfs, but I set the size of the ramdisk to 32 megs, instead of the default half of available ram, with overlaytmpfsflags=size=32m on the cmdline
2023-11-06 22:27:26 <omni> depending on what you intend to run in the VM you may need more memory, I think 160 was the minimum I got away with for a VM that doesn't do too much
2023-11-06 22:28:22 <Joe202311a> omni: okay, IC, obviously there is no need for 32 Bit anymore. But may I ask WHY you chose the x86_64? Security concerns? Having an unified environment?
2023-11-06 22:30:17 <omni> yes
2023-11-06 22:30:35 <heat_> x86_64 is likely faster
2023-11-06 22:30:59 <omni> mostly security, I guess, but never felt a need to try with 32b, maybe you could get away with even less memory?
2023-11-06 22:32:50 <omni> one of several reasons why I got started with alpine was actually the desire to be able to run VMs with a small footprint wrt resourses like memory consumption
2023-11-06 22:33:39 <omni> to be able to have several to many VMs and be able to give the resource hungry ones as much as possible
2023-11-06 22:36:56 <Joe202311a> omni: I have very similar considerations, but I have a fixed amount of physical ram and a fixed amount of VMs i'd like to run, so it is okay to give each of them 512 megs. There is no need to save ram when there is no demand for it elsewhere. Security is very important to me, too, so I think I will go the x86_64 route...
2023-11-06 22:39:08 <Joe202311a> omni: I had the false assumption that 512 megs is very little for x86_64, but the numbers you gave me proved me better. Thanks!
2023-11-06 22:49:37 <omni> sure! =)
2023-11-06 22:50:26 <omni> Joe202311a: and now you know, so if one of your VMs would happen to need more you could take from one that needs less
2023-11-07 03:40:29 <thought> Hey there. I've got an nvr I've got set up and I'd like to avoid moving to edge if I can
2023-11-07 03:40:56 <thought> I have a wireless adapter I need drivers for, and I'm just learning about akms. Something online implied that you don't need the kernel building stuff to use it. Is that true?
2023-11-07 03:41:14 <thought> https://pkgs.alpinelinux.org/package/edge/testing/x86_64/rtl88x2bu-src This is the package I need, to my understanding.
2023-11-07 03:41:52 <thought> I'm basically just asking for a guide on using akms, as I can't find one.
2023-11-07 03:43:45 <thought> Or is there a way I can just steal a package that's not in main and add it to mine
2023-11-07 03:49:47 <thought> ah, --repository...
2023-11-07 03:52:19 <thought> Really fucked up when I read that this device was gonna be supported by linux lmfao; they tested on kernel 6.4
2023-11-07 03:52:24 <thought> always forget that alpine is pretty far behind
2023-11-07 06:45:00 <Saijin_Naib[m]> Far behind compared to what?
2023-11-07 06:45:33 <Saijin_Naib[m]> AKMS should just run based upon triggers in the package. You don't need to do anything
2023-11-07 06:53:55 <Saijin_Naib[m]> Anyone care about Ruffle? A Rust Flash emulator/player?
2023-11-07 10:19:39 <astor> Hi all, I have set up a PXE boot system for my lan with Alpine Netboot working all in memory. It works but I would like to add some files to the root filesystem now. I unpacked the initramfs, added the files and packed again it. The client did boot but I don't find the new files. I am missing something ?
2023-11-07 10:25:40 <Apachez> initramfs is mainly used during boot by the kernel
2023-11-07 10:25:53 <Apachez> that stuff is gone once the kernel is booted
2023-11-07 10:26:24 <Apachez> so with alpine best would be to create a new squashfs which you use as the root filesystem once booted
2023-11-07 10:27:07 <Apachez> using ipxe you could have it as two separate files without iso in between
2023-11-07 10:27:51 <Apachez> like one file for the vmlinuz and initrd thats loaded over tftp/ftp/ftps/sftp/scp/http/https/isci/etc which ipxe supports and one for the squashfs to be loaded and used as root
2023-11-07 10:33:05 <astor> Thanks Apachez. I assumed that the kernel "copied" the root fs from the initramfs into the regular root. Probably it's not in this way.
2023-11-07 12:36:57 <frojnd> Hi there.
2023-11-07 12:38:10 <frojnd> I think I'm missing some packages or configs? I don't see icon just [] for  
2023-11-07 12:39:01 <ikke> You mean for text (emoji) or for applications?
2023-11-07 12:39:23 <frojnd> For applications.
2023-11-07 12:39:46 <ikke> Then you probably need to install some icon pack
2023-11-07 12:39:57 <ikke> what graphical environment are you using?
2023-11-07 12:40:38 <frojnd> sway
2023-11-07 12:40:49 <frojnd> I have adawita-icon-pack installed
2023-11-07 12:50:40 <frojnd> Hm also hicolor-icon-theme
2023-11-07 12:58:51 <Apachez> astor: I dont think it works like that since all the root fs filesystems like squashfs I have seen have a complete set of /usr/bin, /bin and /sbin
2023-11-07 13:05:07 <frojnd> I tried to create /usr/share/icons/default/index.theme  and addded inside [Icon theme]; Inherits=Adawita but still can't see chars properly
2023-11-07 14:27:04 <kchibisov> Does rustup/nightly rust just works now with -crt-static  or they're some caveats I should be aware of?
2023-11-07 15:13:20 <frag> is this the best way to truncate a file? head -n20 $2 > /tmp/truncate && cat /tmp/truncate > $2
2023-11-07 15:14:35 <minimal> frag: without changing the fd? try ":> filename"
2023-11-07 15:15:01 <grawity> frag: util-linux truncate
2023-11-07 15:15:14 <grawity> the OS has an actual syscall for that
2023-11-07 15:15:28 <grawity> err, not util-linux
2023-11-07 15:15:30 <grawity> coreutils
2023-11-07 15:15:34 <frag> fd?
2023-11-07 15:15:36 <grawity> dunno if busybox has it
2023-11-07 15:15:41 <grawity> I expect it should
2023-11-07 15:15:51 <frag> busybox has it too, but it takes size as in kb mb etc
2023-11-07 15:15:56 <frag> not lines
2023-11-07 15:16:28 <minimal> frag: actually, what do you mean by "truncate"? "fd" = file descriptor. Sorry was thinking in terms of someone wanting to truncate/empty a (e.g. log) file but keep the same file descriptor
2023-11-07 15:16:35 <grawity> head -c 20 | wc -c
2023-11-07 15:16:43 <grawity> to get number of bytes
2023-11-07 15:16:52 <grawity> then give it to truncate
2023-11-07 15:16:58 <frag> :]
2023-11-07 15:17:01 <grawity> er, head -n
2023-11-07 16:20:04 <fluix> so I've got /run/user/1000 owned by the uid 1000, but /run/user itself is also 0700 except owned by root. dbus-run-session sway fails with "Unable to set up transient service directory: XDG_RUNTIME_DIR "/run/user/1000" not available: Permission denied"
2023-11-07 16:20:21 <fluix> should /run/user be 0755? if so, what creates it and how do I change that?
2023-11-07 16:23:38 <fluix> answer appears to be yes to the first question
2023-11-07 16:44:25 <frag> is there a small tool to convert xlsx?
2023-11-07 16:44:36 <somercet1> fluix, my /run/user is 1777, like /tmp
2023-11-07 16:44:53 <somercet1> my /run/usr/$pid is 0700
2023-11-07 16:45:16 <fluix> not sure 1777 is right. I've changed mine to be 0755 with a quick local.d script
2023-11-07 16:46:18 <somercet1> that's what I used. 0700 won't work, but 0701 should
2023-11-07 16:46:53 <fluix> 755 is what I have in Arch and what kennylevinson said so I figure I'll go with that
2023-11-07 16:46:58 <fluix> thanks though :)
2023-11-07 16:47:10 <somercet1> yup
2023-11-07 16:47:10 <ssm_> thank you alpine linux for being the only good linux distribution 🙏🙏🙏
2023-11-07 17:52:11 <frag> computer never went faster!
2023-11-07 19:59:02 <WhyNotHugo> fluix: /usr/user should be owned by root:root and 755.
2023-11-07 19:59:18 <WhyNotHugo> * /run/user
2023-11-07 21:07:12 <bz> whatever happened with alpine and sourcehut
2023-11-07 21:08:23 <ikke> not a lot
2023-11-07 21:09:15 <ikke> I assume you mean the mailing list integration?
2023-11-07 21:09:40 <bz> i thought alpine was going to like migrate to sourcehut or something at one point
2023-11-07 21:09:58 <user00> I want to ask about something not about alpine linux
2023-11-07 21:10:15 <ikke> bz: we never had any plans to migrate to source hut
2023-11-07 21:10:58 <bz> oh ok, i guess i just saw that somewhere
2023-11-07 21:11:27 <bz> thats good, because the guy who runs it is kind of a dick lol
2023-11-07 21:11:35 <ikke> bz: no need to get personal
2023-11-07 21:11:44 <ikke> Please adhere to our code of conduct
2023-11-07 22:28:37 <Kladky> bz: I think you might be thinking of postmarketOS, as they are an Alpine-based distro has(?) plans to move to sourcehut.
2023-11-08 02:40:43 <zcrayfish> Sertonix[m]: hey, are you around?
2023-11-08 05:27:38 <LACampbell> should networking be starting before or after your wireless service?
2023-11-08 05:28:07 <LACampbell> I had wpa_supplicant running first, and then networking ran, and the udhcpd thing could find a lease. but then I switched to iwd, and now it can't find a lease, and iwd starts after
2023-11-08 05:30:16 <grawity> well, it will never be able to find a lease if the network is /not connected yet/
2023-11-08 05:30:35 <LACampbell> ok right. so iwd should be first, that was my initial thought
2023-11-08 05:31:07 <grawity> for udhcp -- yes (although doesn't iwd have its own built-in DHCP as well?)
2023-11-08 05:31:22 <LACampbell> uh
2023-11-08 05:31:37 <LACampbell> if that were so, don't I still need networking service/ to bring up the devices..
2023-11-08 05:31:57 <LACampbell> let's delete the service and see what breaks
2023-11-08 05:32:29 <grawity> iwd itself has to bring the interface up in order to be able to start scanning and connecting
2023-11-08 05:33:00 <LACampbell> ok then why is networking... which seems to be hardbaked... trying to do udhcpd scans/
2023-11-08 05:33:02 <grawity> so by the time you're connected to the wifi network, it's already "up" and just missing IP addresses/routes
2023-11-08 05:33:41 <grawity> note I don't actually *know* how to use iwd's dhcp (or whether you should use it at all...), I just remember them being pretty proud of doing dhcp on their own
2023-11-08 05:34:57 <LACampbell> ha. still you've jogged my brain a bit
2023-11-08 05:35:11 <LACampbell> it looks like I have dhcp in /etc/network/interfaces. let's get rid of that and see what happens
2023-11-08 05:43:18 <Fxzxmic> Who can help to update the vaultwarden package?
2023-11-08 05:44:06 <Fxzxmic> It has been many days since the new version was released.
2023-11-08 05:50:31 <LACampbell> grawity: got it working. appreciate you jump starting my brain a bit
2023-11-08 05:50:55 <LACampbell> (hadn't turned on finding dhcp in iwd properly because I didn't uncomment [General] )
2023-11-08 06:34:57 <ikke> Fxzxmic: There is an open MR, but it fails to build at the moment
2023-11-08 06:35:09 <ikke> !54960
2023-11-08 06:46:23 <quinq> 3 is “many days ago”?
2023-11-08 06:46:54 <quinq> Is that a javascript thing?
2023-11-08 06:47:53 <Fxzxmic> I feel like I've been waiting for many days.
2023-11-08 06:48:27 <quinq> Not even three, just two and a half
2023-11-08 06:52:35 <Fxzxmic> I can't wait anymore.
2023-11-08 06:53:10 <Fxzxmic> And don't forget to update the openrc conf too.
2023-11-08 06:53:36 <quinq> Fxzxmic, that's some frustration you should canalyze into learning how to upgrave a package yourself
2023-11-08 06:54:30 <Fxzxmic> Do I need to register an account on your Gitlab?
2023-11-08 07:08:37 <quinq> I don't have a gitlab
2023-11-08 07:08:54 <quinq> I use git+stagit+email
2023-11-08 07:38:38 <Fxzxmic> strange name vaultwarden-web-vault
2023-11-08 07:39:16 <Fxzxmic> just vaultwarden-web or vaultwarden-webui is fine
2023-11-08 07:48:45 <Fxzxmic> git push seems needs a account
2023-11-08 08:22:32 <Fxzxmic> why there is a ssh key it still needs password
2023-11-08 08:23:04 <Ermine> did you add your ssh key into your account?
2023-11-08 08:23:46 <Ermine> Are you sure your remote is ssh?
2023-11-08 08:24:14 <Fxzxmic> yes
2023-11-08 08:25:11 <Ermine> and to which repo do you push?
2023-11-08 08:25:20 <Fxzxmic> ssh -T git@gitlab.alpinelinux.org
2023-11-08 08:25:22 <Fxzxmic> git@gitlab.alpinelinux.org's password:
2023-11-08 08:25:30 <Fxzxmic> It returns
2023-11-08 08:28:25 <Fxzxmic> I want to make commit to vaultwarden package on alpine
2023-11-08 08:29:19 <Ermine> why do you do ssh -T instead of git push
2023-11-08 08:30:00 <Fxzxmic> I know where's wrong
2023-11-08 08:33:58 <Fxzxmic> git push
2023-11-08 08:34:00 <Fxzxmic> git@gitlab.alpinelinux.org's password:
2023-11-08 08:34:05 <Fxzxmic> Oh no
2023-11-08 08:35:22 <Fxzxmic> Now I have no idea why.
2023-11-08 08:38:26 <Ermine> to which repository are you pushing
2023-11-08 08:39:50 <Fxzxmic> my fork
2023-11-08 08:47:05 <Ermine> are you sure? what is the output of git remote -v ?
2023-11-08 08:47:39 <Fxzxmic> origin  git@gitlab.alpinelinux.org:fxzxmic/alpine-aports.git (fetch)
2023-11-08 08:47:41 <Fxzxmic> origin  git@gitlab.alpinelinux.org:fxzxmic/alpine-aports.git (push)
2023-11-08 08:52:10 <Ermine> and which command do you use to push your changes?
2023-11-08 08:55:38 <Fxzxmic> git push
2023-11-08 08:57:56 <Ermine> Ugh, idk then what went wrong
2023-11-08 08:59:17 <Fxzxmic> push to github has never been wrong
2023-11-08 08:59:59 <Fxzxmic> whether alpine doesnt allow to push through ssh?
2023-11-08 09:01:07 <Ermine> Alpine allows pushing with ssh, I think it's bug either in your ssh client or in gitlab
2023-11-08 09:02:23 <Fxzxmic> push success through https
2023-11-08 09:05:46 <Fxzxmic> !55011
2023-11-08 09:15:57 <Fxzxmic> Maybe this is the reason? : Your account is authenticated with SSO or SAML. To push and pull over HTTPS with Git using this account, you must set a password or set up a Personal Access Token to use instead of a password. For more information, see Clone with HTTPS.
2023-11-08 09:16:19 <Fxzxmic> I login with github account.
2023-11-08 09:16:55 <Ermine> 'You must set a password'
2023-11-08 09:17:05 <Fxzxmic> github account cant push through ssh?
2023-11-08 09:17:28 <Habbie> yes
2023-11-08 09:17:30 <Habbie> if you add a key
2023-11-08 09:17:40 <Fxzxmic> I have set Personal Access Token
2023-11-08 09:18:17 <ikke> Did you add an ssh key?
2023-11-08 09:18:31 <Fxzxmic> Yes
2023-11-08 09:19:09 <Fxzxmic> but it still needs me to input password
2023-11-08 09:19:23 <ikke> Does ssh -vv -T git@gitlab.alpinelinux.org show that it is offering the ssh key?
2023-11-08 09:19:54 <Fxzxmic> let me try
2023-11-08 09:24:08 <Fxzxmic> seems stuck
2023-11-08 09:24:13 <Fxzxmic> debug1: Authentications that can continue: publickey,password,keyboard-interactive
2023-11-08 09:24:14 <Fxzxmic> debug1: Next authentication method: publickey
2023-11-08 09:24:15 <Fxzxmic> debug1: Offering public key: /home/Dev/.ssh/id_ed25519 ED25519 SHA256:XLRXUp9Xkzu7NC8Q+Po/vDaektLjlRpSGgUgJ0YzjBc explicit
2023-11-08 09:24:17 <Fxzxmic> debug2: we sent a publickey packet, wait for reply
2023-11-08 09:24:18 <Fxzxmic> debug1: Authentications that can continue: publickey,password,keyboard-interactive
2023-11-08 09:24:20 <Fxzxmic> debug2: we did not send a packet, disable method
2023-11-08 09:24:21 <Fxzxmic> debug1: Next authentication method: keyboard-interactive
2023-11-08 09:24:23 <Fxzxmic> debug2: userauth_kbdint
2023-11-08 09:24:24 <Fxzxmic> debug2: we sent a keyboard-interactive packet, wait for reply
2023-11-08 09:24:25 <Apachez> omg ?
2023-11-08 09:24:26 <Fxzxmic> debug1: Authentications that can continue: publickey,password,keyboard-interactive
2023-11-08 09:24:27 <Fxzxmic> debug2: we did not send a packet, disable method
2023-11-08 09:24:29 <Fxzxmic> debug1: Next authentication method: password
2023-11-08 09:24:30 <Apachez> SCHTAP!!
2023-11-08 09:24:30 <Ermine> Use pastebin please
2023-11-08 09:24:55 <Fxzxmic> oh I dont realize it is so much
2023-11-08 09:25:00 <Fxzxmic> sorry
2023-11-08 09:25:07 <Apachez> at least he didnt paste his privatekey =)
2023-11-08 09:33:37 <sertonix> zcrayfish: You wanted to talk?
2023-11-08 09:37:49 <Fxzxmic> config Github similarly, but there is no problem
2023-11-08 09:38:54 <Ermine> it's not similar
2023-11-08 09:39:16 <ikke> Fxzxmic: can you confirm you have a key in gitlab with the same fingerprint?
2023-11-08 09:43:46 <Fxzxmic> how do i check that?
2023-11-08 09:44:42 <Fxzxmic> using ssh-keygen, but it doesnt match
2023-11-08 09:46:33 <Fxzxmic> OH I AM STUPID
2023-11-08 09:49:04 <Fxzxmic> The public key I uploaded is from another develop env
2023-11-08 09:50:50 <Fxzxmic> So sorry to bother all of you.
2023-11-08 09:51:40 <ikke> No problem 🙂
2023-11-08 09:53:34 <Fxzxmic> !55011 my first merge request to alpine. LOL
2023-11-08 09:58:38 <zcrayfish> sertonix: yeah, I was hoping you could fix something on the wiki; because something is seriously wrong with my connection to it at the moment (page load times there are in excess of 2 minutes), but I managed to deal with it. 
2023-11-08 10:02:01 <sertonix> zcrayfish: ok, have you seen my backlight issue on the sway talk page?
2023-11-08 10:02:57 <zcrayfish> yes, I think you mentioned that a udev rule could probably fix it... You're probably right there.
2023-11-08 10:03:29 <zcrayfish> Hopefully later today whatever is causing me to barely be able to access the wiki will be corrected. (I think Starlink is having an issue)
2023-11-08 10:04:49 <sertonix> A udev rules defeats the purpus of seat managmant. Would you think that seatd maybe should be responsible for something like that?
2023-11-08 10:07:18 <zcrayfish> possibly
2023-11-08 10:12:21 <zcrayfish> https://git.sr.ht/~kennylevinsen/seatd-docs <-- good lord, is that IT for the docs?
2023-11-08 10:13:50 <sertonix> The docs are as minimal as seatd...
2023-11-08 10:14:28 <zcrayfish> oh it has some manpages too... not much
2023-11-08 10:14:31 <zcrayfish> https://git.sr.ht/~kennylevinsen/seatd/tree/master/item/man
2023-11-08 11:28:14 <sertonix> zcrayfish: I have send backlight support as feature request to the seatd mailing list. If you want to check it out: https://lists.sr.ht/~kennylevinsen/seatd-devel/%3CCWTE7EP5A45V.NA7HOHLMHPDX%40posteo.net%3E
2023-11-08 17:25:23 <ssm_> does alpine package any other unix utils besides coreutils and busybox? I was hoping to use something more bsd-like
2023-11-08 17:27:50 <Ermine> 'more bsd-like' is what
2023-11-08 17:28:35 <Ermine> Do you have any specific package in mind?
2023-11-08 17:52:09 <ssm_> Ermine: something like https://github.com/dcantrell/bsdutils
2023-11-08 18:22:21 <quinq> So you mean freebsd utilities
2023-11-08 18:24:18 <frag> have bsd-games, got over 8000 points in drop4
2023-11-08 18:24:35 <quinq> :D
2023-11-08 18:27:11 <frag> best game on alpine linux \m/
2023-11-08 18:32:19 <ssm_> quinq: I would prefer openbsd utils, but I don't think there's a portable version for those
2023-11-08 18:32:48 <quinq> Which ones?
2023-11-08 18:32:53 <quinq> There's jot for example
2023-11-08 18:33:14 <ssm_> quinq: I mean using them instead of busybox
2023-11-08 18:33:24 <quinq> I don't get the goal
2023-11-08 18:33:30 <ssm_> busybox is alright, but it's really rough around the edges (the ed implementation is next to useless)
2023-11-08 18:33:51 <quinq> Yeah, busybox is meh
2023-11-08 18:34:40 <quinq> But most standard utilities are more or less ok (I haven't extensively tested them)
2023-11-08 18:34:55 <quinq> For ed, you can install other implementation, sure
2023-11-08 18:35:04 <frag> i have no problem with busybox
2023-11-08 18:35:13 <quinq> ed, awk, etc
2023-11-08 18:35:13 <frag> use it instead of coreutils
2023-11-08 18:35:16 <quinq> frag, noted
2023-11-08 18:36:21 <quinq> But having a decent ed implementation isn't a justification of replacing the whole standard tools implemented by busybox, imo
2023-11-08 18:36:26 <quinq> s/of/for/
2023-11-08 18:36:54 <ssm_> there are other things that took me off guard, but I can't remember them off the top of my head
2023-11-08 18:37:41 <ssm_> the ps implementation is very non-posix from what I can tell
2023-11-08 18:37:48 <quinq> ssm_, fair enough, bugs exist
2023-11-08 18:37:51 <quinq> They should be reported though
2023-11-08 18:39:35 <quinq> indeed for ps, ssm_ 
2023-11-09 00:55:47 <Tommy[m]> Hello everyone... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/TXuzcUqYOvKiRePXsqKmSxmX>)
2023-11-09 07:11:40 <fluix> zcrayfish: kenny is active on IRC (#sway and related on libera) and probably happy to clarify things or accept patches on seatd docs
2023-11-09 07:12:33 <fluix> just an fyi, if you need it :)
2023-11-09 08:35:49 <dzilvys> hello all, a quick question. if I am trying to have the boot sequence from openrc thrown to /dev/ttyS0 how exactly would I go about that?
2023-11-09 08:36:28 <dzilvys> currently I have set rc_verbose="YES" on the local service, so when local service scripts are executed it shows it to a connected monitor.
2023-11-09 08:37:40 <dzilvys> However, I have a serial port in which I would like to echo certain stages of scripts. I have tried doing something like `echo "Flashing XXX" | tee /dev/ttyS0`. However, this fails to send anything to connected serial devices during the boot sequence 
2023-11-09 08:38:07 <dzilvys> However, if I wait until the boot sequence is done, and I can login. Once I login, I execute the same command and now its sent to the serial line
2023-11-09 08:38:52 <dzilvys> Is there a possibility that the serial line is not being established before local is finished, or is openrc doing something that ignores the `tee`?
2023-11-09 08:59:31 <grawity> dzilvys: what if you change your script to actually check whether /dev/ttyS0 exists at that point
2023-11-09 08:59:46 <grawity> if you want all of it, I'd probably boot with console=ttyS0, I believe that would redirect most things...
2023-11-09 09:00:10 <dzilvys_> a
2023-11-09 09:01:10 <dzilvys_> grawity: Where exactly would I put this and if I boot with console=ttyS0 will it stop directing everything to tty0? aka monitor or will it still do it?
2023-11-09 09:01:29 <dzilvys_> My end goal is to have the it displayed on all available outputs
2023-11-09 09:16:39 <dzilvys> actually reading https://wiki.alpinelinux.org/wiki/Enable_Serial_Console_on_Boot I see a section `Enabling two consoles during boot` but its confusing when it has the append in the code blocks so now Idk where to put this
2023-11-09 09:39:56 <grawity> it's a kernel commandline option that is specified through your bootloader
2023-11-09 09:40:10 <grawity> like, it says "Example syslinux.cfg" above the 1st code block...
2023-11-09 09:40:27 <grawity> you probably want to add it through the extlinux .conf though
2023-11-09 09:40:35 <dzilvys> GRUB_SERIAL_COMMAND="serial --unit=0 --word=8 --parity=no --speed 115200 --stop=1"```
2023-11-09 09:40:35 <dzilvys> sorry yeah I figured it out :), in the end I did this ```GRUB_CMDLINE_LINUX_DEFAULT="console=ttyS0,115200n8d console=tty0 modules=sd-mod,usb-storage,ext4 quiet rootfstype=ext4"
2023-11-09 09:40:42 <dzilvys> ugh format sucks
2023-11-09 09:40:49 <dzilvys> `GRUB_CMDLINE_LINUX_DEFAULT="console=ttyS0,115200n8d console=tty0 modules=sd-mod,usb-storage,ext4 quiet rootfstype=ext4"
2023-11-09 09:40:49 <dzilvys> GRUB_SERIAL_COMMAND="serial --unit=0 --word=8 --parity=no --speed 115200 --stop=1"`
2023-11-09 09:41:03 <dzilvys> You get the point I hope :D
2023-11-09 11:11:01 <ovf> i have a hunch i will need a windows vm in the near future. the last time i did that i used virtualbox which i think automagically sets things up. alpine doesn't have it for... reasons? what do you guys use? qemu? with a script?
2023-11-09 11:13:19 <donoban> ovf: I use qemu-openrc with some tweaks... but probably now I would switch to just a shell script per VM
2023-11-09 11:18:52 <chereskata> Does anyone use grpc-java?
2023-11-09 11:19:07 <chereskata> There are build problems and the maintainer is not active
2023-11-09 15:12:06 <SagestMage> Howdy all.  I've developed a dependency problem on a new install.  I can't seem to install many more things due to a conflicting dependency between icu-libs, icu-data-all.  It said something about harvbuzz-dev, so since I only needed that for a compile (now done) I removed it and am still in the same conflicted state.  I tried to put harfbuzz-dev back and it now is claiming the same dependency 
2023-11-09 15:12:12 <SagestMage> problem and won't install back.  I did the install of libreoffice a week or so ago on another machine with the identical setup.  It seems something funky is going on with the icu deps?  Is there any way to fix this, because it seems interlocked with most of the userspace system.
2023-11-09 15:12:40 <SagestMage> many thanks in advance for any help
2023-11-09 15:17:00 <ncopa> the problem is that we recently  bumped icu 74
2023-11-09 15:17:30 <ncopa> main repo got built pretty quickly, while the community repo rebuilds are taking time
2023-11-09 15:36:17 <SagestMage> ncopa: ok, no problem.  Should I just wait a little bit to get fixed in repos or do I have to do anything on my end?  I'm on edge also, I should have noted that.
2023-11-09 15:47:08 <omni> yeah, it will take some time still for everything to be rebuilt and then uploaded to package repos and synced to mirrors
2023-11-09 15:53:22 <ncopa> should just need to wait, and hope people dont push more big builds
2023-11-09 15:53:28 <ncopa> and hope the current builds doesnt fail
2023-11-09 15:53:46 <ncopa> we ran out of disk space earlier today and i pushed new kernels...
2023-11-09 15:54:36 <ncopa> one thing I have learned over the years, it is not possible to have too much compile power
2023-11-09 16:07:37 <SagestMage> no worries.  I hope the compiling goes well.
2023-11-09 16:11:52 <ikke> SagestMage: you can track the status on https://build.alpinelinux.org
2023-11-09 16:12:01 <SagestMage> on another note.  I used one of the wikis (xdg-mime setup) and I had some possible additions to the wiki.  How is the environment for just hopping on and adding it for review?  
2023-11-09 16:13:08 <ikke> The wiki is mostly community maintained, there is no formal process
2023-11-09 16:38:42 <witcher> gpg is failing to recognize my yubikey as a smart card. `gnupg-scdaemon` doesn't seem to provide the `scdaemon` binary, how do i get this working?
2023-11-09 16:39:46 <ikke> It's in a separate package
2023-11-09 16:48:58 <witcher> which one? i can't find a package providing `cmd:scdaemon`
2023-11-09 16:50:15 <telmich> Hey everyone
2023-11-09 16:50:45 <telmich> For some months now I am running on a custom linux-sof project based linux kernel, as I was not able to figure out what is different in the -lts of alpine linux that prevents sound from working
2023-11-09 16:51:37 <telmich> I would like to solve that mystery and also add sounds support for the dell xps 13 I have into alpine
2023-11-09 16:52:11 <telmich> I do have a working kernel tree and a working .config and was wondering if anyone else could share 2 eyes on it on where the difference might be?
2023-11-09 16:56:52 <telmich> Just added a README.txt on https://www.nico.schottelius.org/temp/xps-9320/README.txt for tracking it
2023-11-09 16:58:43 <ikke> And when it's not working, what happens?
2023-11-09 17:13:44 <telmich> aplay -l does not list any sound devices
2023-11-09 17:14:59 <telmich> lts, all variants looks like this: https://www.nico.schottelius.org/temp/xps-9320/aplay-l-alpine-lts.txt
2023-11-09 17:15:35 <telmich> the other ones look like this and provide sound output: https://www.nico.schottelius.org/temp/xps-9320/aplay-l-linux-sof.txt https://www.nico.schottelius.org/temp/xps-9320/aplay-l-ubuntu.txt
2023-11-09 17:18:53 <telmich> in the linux-sof .config I have CONFIG_SND_SUPPORT_OLD_API=y, whereas -lts does not have it set
2023-11-09 17:22:22 <telmich> As well as CONFIG_SND_HDA_INPUT_BEEP_MODE=1 and CONFIG_SND_HDA_PATCH_LOADER=y
2023-11-09 17:44:16 <consus> what is the established way to wait for mdev actually being done dealing with uevents?
2023-11-09 17:47:11 <em6> hello, pure data wont start after regular "apk add pure-data" and launging through dmenu in dwm. nothing happens. any ideas?
2023-11-09 18:03:09 <witcher> ikke: what package did you mean? i still can't figure out how to make it work, some help would be appreciated!
2023-11-09 19:02:28 <ikke> It has the same name as the bibary
2023-11-09 19:14:49 <Ermine> consus: mdev - most probably no, but mdevd has mdevd-coldplug which can wait
2023-11-09 19:19:28 <consus> thanks, i'll check it out
2023-11-09 22:33:10 <holgersson> Hi! Any pitfalls to know when using alpine for a small webserver that hosts gitea plus a database (most probably postgresql)? Should I go there for a "disk based install" or keep things in RAM and mount only all the data dirs?
2023-11-09 22:57:27 <WhyNotHugo> Trying to 'apk add !kwallet5' fails due to 'satisfies: webkit2gtk-4.1-2.40.5-r1[dbus:org.freedesktop.Secrets]'...
2023-11-09 22:57:41 <WhyNotHugo> I also have keepassxc installed, which also provides org.freedesktop.Secrets.
2023-11-09 22:58:04 <WhyNotHugo> I don't know at which point kwallet was pulled, but now I cannot remove it :/
2023-11-09 22:58:10 <Tommy[m]> Hello everyone... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/nwqIUNVyPPUmFgRCJZrGnpTf>)
2023-11-09 23:12:56 <minimal> holgersson: depends on how much RAM the machine has, how much is left after OS loads into RAM, how much RAM gitea and webserver and postgresql needs etc...
2023-11-10 00:09:39 <thresh> Howdy.  Is 3.15 EOL now?
2023-11-10 00:17:01 <minimal> thresh: did you check the Releases page?
2023-11-10 00:20:34 <thresh> minimal, I did.  It's confusing.  Says "End of support" on • 2023-11-01, but still receives security fixes.
2023-11-10 00:21:08 <minimal> so only security fixes will be made
2023-11-10 00:21:26 <minimal> what's not clear?
2023-11-10 00:21:27 <thresh> Until what time, and how is that different to 3.16, which has the same support.
2023-11-10 00:21:35 <minimal> also that's only "main"
2023-11-10 00:21:36 <thresh> The end date of said support is not clear
2023-11-10 00:22:59 <minimal> eh? "Support level" = "sec fixes" means *WHEN* it is supported it is only supported to make security fixes, no other fixes
2023-11-10 00:23:13 <minimal> when Support is finished, all support is finished
2023-11-10 00:23:38 <thresh> So maybe the bullet point should be red now, and Support level changed to "on request" ?
2023-11-10 00:23:39 <minimal> i.e. 3.16 still gets security fixes, 3.15 gets no fixes, 3.18 gets both bug fixes and sec fixes
2023-11-10 00:24:03 <minimal> also this only related to "main", not to "community"
2023-11-10 00:24:08 <thresh> It's fine
2023-11-10 00:25:11 <minimal> Community support is long since finished for 3.15
2023-11-10 00:25:47 <minimal> I assume no-one has updated that page since 1st of month
2023-11-10 00:26:20 <thresh> I assumed as much, which is why I asked here
2023-11-10 00:26:59 <thresh> Looking forward to remove 3.15 and add 3.19 once it's released from my build farms :-)
2023-11-10 03:04:33 <WhyNotHugo> Chrony blocks at system startup for multiple seconds. Can I disable this?
2023-11-10 04:14:57 <WhyNotHugo> Oh, /etc/conf.d/chronyd has a flag for this.
2023-11-10 06:18:20 <Saijin_Naib[m]> Does anyone here use NextCloud and have a recommendation for a remote system monitoring tool for it? Ideally GTK-based, but eah
2023-11-10 06:18:22 <Saijin_Naib[m]> I'm coming up empty
2023-11-10 07:05:48 <grawity> system monitoring?
2023-11-10 07:19:00 <Saijin_Naib[m]> https://github.com/nextcloud/serverinfo
2023-11-10 07:19:22 <Saijin_Naib[m]> Looks like some sort of JSON API endpoint is opened by NextCloud so tools can hook into it and visualize the data?
2023-11-10 07:19:26 <Saijin_Naib[m]> I was assuming this was some standard thing
2023-11-10 07:19:41 <Saijin_Naib[m]> JSON or XML, sorry
2023-11-10 10:31:21 <preshaa[m]> Did you get off the hook and now you're on edge? Have you been looking for her best possible way to score a good amount of any kind of ɖrųɠ since your neighbouring plug let you down ? Stay classy ✌🏻 🔥‼️and get across to the most vouched for HEAVILY from coast to coast & everything is tested ‼️🔥 -FREE SHIPPING AVAILABLE TO ALL OVER THE USA 🇺🇸... (full message at
2023-11-10 10:31:21 <preshaa[m]> <https://matrix.org/_matrix/media/v3/download/matrix.org/iKgDudLVcoLjPCLNPiJAeEEX>)
2023-11-10 10:38:06 <telmich> omg, that was the first time I saw such sophisticated drug spamming. Shocking and impressive.
2023-11-10 10:55:30 <fraolt> What is the best way to check if a package has already been updated? I'm waiting for qt6-qtbase and qt6-qtwebengine to be rebuilt after the icu 74 update on edge.
2023-11-10 10:58:22 <fraolt> I'm a bit nervous becaus https://gitlab.alpinelinux.org/alpine/aports/-/commit/4aefe0d7324e6f1b3c5b6bd1982586ce51cb62af was committed 72h ago and still no build. 😟
2023-11-10 10:58:38 <aron> try:   apk list -u $pkgname
2023-11-10 10:58:46 <aron> this yields empty if pkg is up-to-date
2023-11-10 11:10:12 <fraolt> Thx, I haven't installed the packages yet. But it shows qt6-qtwebengine-6.6.0-r2 not r4. So that's my indicator. That means I'll do `apk update && apk list qt6-qtwebengine` every once in a while.
2023-11-10 11:29:01 <ovf> fraolt: https://pkgs.alpinelinux.org/packages?name=qt6-qtwebengine might be a little more authoritative (rules out mirror sync delay etc.)
2023-11-10 11:29:12 <ikke> It uses the same source
2023-11-10 11:29:25 <ikke> At least, it uses dl-cdn as source
2023-11-10 11:30:06 <ikke> It syncs every 15 minutes, so in theory, checking locally can be quicker than pkgs.a.o
2023-11-10 11:33:12 <fraolt> Thank you both! I guess pkgs puts less load on the servers because I don't download the whole APKINDEX. I'll stick to that.
2023-11-10 11:33:41 <fraolt> s/don't/won't/
2023-11-10 11:33:41 <ikke> reloading pkgs.a.o is heavier than fetching an apkbuild file
2023-11-10 11:34:09 <ikke> We have numerous mirrors offering the repos, while we just have a single server hosting pkgs.a.o
2023-11-10 11:34:25 <fraolt> Ok, back to the terminal it is. :-)
2023-11-10 11:35:14 <fraolt> There's no "testing"-repo in the stable releases, right?
2023-11-10 11:35:47 <ikke> correct
2023-11-10 11:36:29 <fraolt> Makes sense. Thx.
2023-11-10 11:38:55 <ovf> ikke: oh. :-)
2023-11-10 12:44:42 <CcxWrk> dia is no longer packaged? Is there replacement?
2023-11-10 13:05:29 <nextgens> Hi! It looks like the buildbots have failed to rebuild rspamd-libs on x86_64 and aarch64 (https://pkgs.alpinelinux.org/packages?name=rspamd-libs&branch=edge). Can someone poke them please? ikke?
2023-11-10 13:06:50 <ikke> nextgens: the builders are still busy 
2023-11-10 13:07:12 <ikke> https://build.alpinelinux.org/
2023-11-10 13:07:31 <nextgens> thank you for checking and the link
2023-11-10 13:07:51 <nextgens> Is there a standardized tool to make a partial mirror with just the packages I require? I think I will end up pinning a specific version
2023-11-10 13:18:19 <nextgens> setup-apkcache looks like what I am after
2023-11-10 13:23:04 <ikke> That would cache installed packages 
2023-11-10 13:26:23 <nextgens> If I persist the directory where the cache is, I can use something like `apk add --repository=file:///tmp/apks/ ...` afterwards, right? Or will that always try to fetch different indexes than those cached there?
2023-11-10 13:27:18 <ikke> The cache is not a repo (it misses an apkindex)
2023-11-10 14:36:26 <minimal> WhyNotHugo: the chrony "block" at startup is intended so that the clock is correct then to ensure it does not change as other (later) daemons/services start. If you disable this "block" then the clock could start whilst other services are initialising which could affect them (i.e. if the clock goes backwards in time).
2023-11-10 15:09:00 <dwfreed> some services get mad if time goes backwards
2023-11-10 15:09:40 <dwfreed> dovecot is one I know of off the top of my head
2023-11-10 15:21:04 <txnintn> I'm using 3.16.7 Aarch64 with esp partition -> mounted /dev/sda1 /boot/efi, when I run setup-disk I get error "install: can't stat 'mnt/boot/efi/EFI/alpine/grubaa64.efi': No such file or directory.  suggestions
2023-11-10 15:39:50 <minimal> txnintn: if you're running setup-disk then I'd expect you would have the new rootfs mounted as /mnt and the ESP filesystem mounted as /mnt/boot/efi
2023-11-10 15:41:49 <minimal> dwfreed: yupe. It's the same way some people don't want the DHCP "delay" during boot yet other services may need a configured network interface...
2023-11-10 15:46:49 <txnintn> I am using using the raspberry vanilla uefi firmware and alpine 3.16.7 installed on sda1 and created a lvm root partition on sda2 and mounting them to /mnt/boot/efi and /mnt and then running setup-disk -m sys /mnt command.
2023-11-10 15:47:49 <nextgens> ikke> it doesn't look like the builders are making any progress; If I take 1/build-edge-aarch64, it has failed to build darktable and hasn't picked up anything new for the last few hours ... if I take 8/build-edge-x86_64, the three jobs haven't changed either: chromium doesn't give a build-log, py3-digitalocean and php81-spx are stuck on "Signing the index"
2023-11-10 15:48:15 <minimal> did you mount /mnt *before* /mnt/boot/efi ?
2023-11-10 15:48:22 <minimal> txnintn: ^^^
2023-11-10 15:48:44 <nextgens> ikke> can you help me understand please? The package I'm after was built on the 27th for other archs (rspamd-libs 3.7.3-r1)
2023-11-10 15:49:04 <minimal> also if sda is the device you're installing on then what is the device you booted off?
2023-11-10 15:49:53 <txnintn> yes, I create the lvm, then format it ext2 and run it before mounting the the esp partiton.
2023-11-10 15:50:09 <minimal> txnintn: also why are you installing Alpine 3.16.7 rather than 3.18.x ?
2023-11-10 15:50:51 <ssm> they were clearly born 03/16/2007
2023-11-10 15:51:34 <txnintn> Aarch64 3.16.7 has the raspberry pi support in it, 3.18 it doesn't appear to.
2023-11-10 15:52:14 <minimal> txnintn: there's a separate RPI install - the generic aarch64 install isn't designed for the RPI
2023-11-10 15:54:05 <minimal> "mainline" linux-lts may have *some* RPI support in it but it doesn't include the various patches that RPI Foundation make to properly support RPIs
2023-11-10 15:54:41 <txnintn> I believe I have ran into issues with the rpi version not getting along with the uefi firmware.
2023-11-10 15:55:09 <minimal> yes, because UEFI isn't supported by RPIs - you're using 3rd party software ;-)
2023-11-10 15:55:49 <minimal> well if you don't use the RPI specific kernel you'll likely have problems
2023-11-10 15:56:41 <minimal> Good luck! I'll leave you to it
2023-11-10 16:28:54 <vkrishn> dwfreed: i think lighttpd too(just stops)
2023-11-10 17:03:28 <frag> maybe a bit off topic, is it possible to use net bank apps and payment apps without google on mobile devices?
2023-11-10 17:04:18 <txnintn> FYI Minimal  RPI4 UEFI Vanilla Alpine with 3.18.3 aarch64 manual install with setup-disk works, wifi driver not loading.
2023-11-10 17:04:27 <frag> or even maybe on desktops..
2023-11-10 17:04:39 <ikke> frag: probably going to be difficult
2023-11-10 17:04:56 <frag> ikke: yeah :/
2023-11-10 17:54:23 <txnintn> Fyi version 3.18.4 std aarch64 loads wifi driver...testing continuing.
2023-11-10 17:59:32 <txnintn> Fyi minimal never mind not loading..
2023-11-10 18:02:28 <minimal> txnintn: not loading what?
2023-11-10 18:07:04 <txnintn> We talk quite a will back about 3.18.3 standard aarch64 not loading the rpi 4 wifi, you suggested the 3.16.7, it loads the wi-fi but grub is being picky.  I thought 3.18.4 loaded the wi-fi, my bad it didn't.  sorry to bother you over nothing
2023-11-10 18:19:38 <minimal> txnintn: I don't remember recommending 3.16.7, rather I asked why you were not using latest release (3.18.x). Also you're trying to install something (RPI4 + generic Alpine aarch64 download + UEFI) that not many people will have tried and is not supported.)
2023-11-10 18:23:56 <txnintn> I've been playing with this off and on for about 3 months, for my retirement house I'm building. I did a install doc on dual usb lvm raid with sd spare.
2023-11-10 19:29:44 <fraolt> Hi again, it seems the build pipline may have been cleared as now a very recent MR is being build: https://gitlab.alpinelinux.org/alpine/aports/-/jobs/1177334
2023-11-10 19:30:31 <fraolt> But py3-pyqt6-webengine is still not being rebuild. Is there something that needs to be re-triggered?
2023-11-10 19:32:35 <ikke> CI builds are separate from the builders
2023-11-10 19:32:43 <ikke> x86_64 is still busy, has not uploaded community yet
2023-11-10 19:35:53 <fraolt> Ah, ok, I see. I'll keep waiting then.
2023-11-10 19:36:32 <fraolt> Is aarch64 also busy?
2023-11-10 19:37:21 <ikke> yes
2023-11-10 19:37:53 <fraolt> ok, thanks.
2023-11-10 19:58:12 <txnintn> Minimal FYI UEFI Alpine Vanilla Standard 3.17.1 aarch64, Works wifi detected during setup-interfaces, grub worked with standard /mnt/boot/efi Yippie
2023-11-10 20:50:23 <TML> I have a system which has an x86_64 chip in it, but the motherboard is not EFI-capable. When I look at the available grub packages, it seems like I'm limited to i386-pc or x86_64-efi; is there somewhere an x86_64 grub *without* EFI that I'm not seeing?
2023-11-10 21:02:44 <txnintn> TML -> what brand mb?
2023-11-10 21:04:10 <TML> txnintn: EVGA
2023-11-10 21:05:06 <txnintn> So no settings to turn UEFI on?
2023-11-10 21:05:14 <TML> No
2023-11-10 21:05:51 <TML> it's an older Opteron CPU
2023-11-10 21:05:52 <txnintn> I have some intels and the setting were in really weird locations
2023-11-10 21:05:59 <TML> and the motherboard is equally old
2023-11-10 21:06:37 <dwfreed> despite the name, i386-pc works on all x86 systems newer than an i386 too
2023-11-10 21:06:54 <txnintn> Mine are about 2007 last of the LGA775, but they still work.
2023-11-10 21:07:08 <TML> This motherboard doesn't even support drives >2TB
2023-11-10 21:08:09 <txnintn> just a sec
2023-11-10 21:08:56 <dwfreed> >2TB should work just fine if they're GPT formatted (as the issue is MBR does not support >2TB)
2023-11-10 21:09:51 <TML> the bios won't even recognize the disks are attached, so they're not available to the OS
2023-11-10 21:10:14 <dwfreed> sounds like you should throw the system in the trash instead :P
2023-11-10 21:10:19 <TML> :)
2023-11-10 21:10:36 <txnintn> go cheap, scsi...hahah
2023-11-10 21:11:26 <TML> thankfully, it DOES have SATA1 ports (though it ALSO has IDE ports)
2023-11-10 21:11:49 <dwfreed> modern drives are faster than SATA1
2023-11-10 21:12:09 <dwfreed> even spinning rust these days can saturate a SATA1 link
2023-11-10 21:12:13 <TML> it's a venerable little low-power system that has served me well for like 20 years, I'm loathe to toss it out
2023-11-10 21:16:31 <txnintn> have you read this article -> https://wiki.archlinux.org/title/GRUB_Legacy
2023-11-10 21:17:03 <dwfreed> you don't need grub legacy; grub2 works just fine
2023-11-10 21:17:26 <dwfreed> as I said before, despite the name, i386-pc works on all x86 systems newer than an i386
2023-11-10 21:17:56 <dwfreed> (on UEFI systems, it does depend on CSM being available)
2023-11-10 21:18:04 <TML> dwfreed: is there any downside to using the i386 grub on an x86_64 system? It seems like it would create problems.
2023-11-10 21:18:11 <dwfreed> not at all
2023-11-10 21:19:07 <TML> great - I'll go that route, then. Thanks. I was just assuming that "Bad things" would ensue when using an i386 grub to install an x86_64 kernel
2023-11-10 21:23:17 <TML> can confirm everything boots just fine when installed with an i386 grub; thank you, dwfreed!
2023-11-11 10:47:39 <bananicorn> Hi, got a quick question about diskless install on an USB with apkovl and apk cache on the same usb
2023-11-11 10:49:16 <bananicorn> My usb is /dev/sdc and I see that the entire drive is mounted at /media/sdc - can I just create another partition (/dev/sdc3) and use it for apkovl/cache or will that create a problem since the whole drive is mounted? 
2023-11-11 10:55:24 <donoban> bananicorn: how did you create the image? "dd if..."?
2023-11-11 10:56:25 <bananicorn> I used gnome disks with "restore image" for a change, using the extended iso
2023-11-11 10:56:35 <bananicorn> I take it I should use dd instead?
2023-11-11 10:56:53 <donoban> no, it will behave like a cd-rom
2023-11-11 10:57:18 <donoban> I think that there are the right steps on the wiki
2023-11-11 10:58:30 <donoban> https://wiki.alpinelinux.org/wiki/Alpine_local_backup#Saving_and_loading_ISO_image_customizations
2023-11-11 10:58:41 <bananicorn> I'll have another look there then, thanks :)
2023-11-11 12:30:57 <WhyNotHugo> apk packages are signed, right? So an untrustworthy proxy cannot tamper them?
2023-11-11 12:34:51 <ikke> Both the index as the apk file itself are signed, yes
2023-11-11 12:44:20 <WhyNotHugo> Oh, nice, the index too. So no downgrade attacks either.
2023-11-11 14:57:40 <lnl> well, a mirror absolutely can serve you an outdated repository
2023-11-11 15:15:45 <quinq> freaking kernel overwrite
2023-11-11 15:31:55 <ikke> quinq: ?
2023-11-11 15:40:38 <quinq> You know, there's a kernel update and then you lose all modules from the current kernel :)
2023-11-11 15:40:54 <quinq> And then you're forced to reboot if you need to load one
2023-11-11 15:59:46 <qaqland> My screen is 1920x1280 but /sys/class/drm/card0-HDMI-*/modes only have 1920x1080
2023-11-11 16:00:34 <qaqland> what should I do to make it have the correct resolution?
2023-11-11 17:34:10 <sertonix> Where could I propose changes to the Alpine logo svg?
2023-11-11 18:08:41 <dewdude_> well..my alpine honeymoon is over. not supporting network devices is a deal breaker: https://img.pickmy.org/lrq4p8ec.png
2023-11-11 18:19:02 <dwfreed> dewdude_: the module is built in linux-lts, at least: https://github.com/alpinelinux/aports/blob/master/main/linux-lts/lts.x86_64.config#L1250
2023-11-11 18:21:28 <dwfreed> the lack of google results really isn't meaningful
2023-11-11 18:21:58 <quinq> It just means your web search engine sucks
2023-11-11 18:22:15 <quinq> Or that's not the place to look for that kind of information :)
2023-11-11 18:25:02 <dwfreed> https://pkgs.alpinelinux.org/contents?file=r8152.ko.gz&path=&name=linux-lts&branch=edge&repo=main&arch=x86_64
2023-11-11 18:27:59 <minimal> perhaps he doesn't have linux-firmware-rtl_nic installed...
2023-11-12 09:41:09 <frag> if you want to send some semi-sensitive information to yourself, is it more secure to just save it as a draft in email account, and go to the other computer and download the draft than to send it as normal email?
2023-11-12 09:41:27 <frag> is email still unsecure? whos likely to get it/read it?
2023-11-12 09:42:52 <quinq> Why not using ssh?
2023-11-12 09:44:44 <frag> hmm its from a smarthphone .. i guess it has ssh too maybe ..
2023-11-12 09:46:05 <frag> ugh, kinda back to syncing from phone o,o
2023-11-12 09:46:29 <frag> trying to use smartphone not too much/advanced setup
2023-11-12 09:55:35 <quinq> I'm not sure why people still insist on calling their pocket computer “phones”
2023-11-12 09:55:46 <quinq> It's like calling computers “My typewriter”
2023-11-12 09:56:29 <quinq> But yeah, those devices are still a problem, locked behind limited operating system
2023-11-12 09:56:43 <quinq> There are ssh client/daemons though, AFAIK
2023-11-12 09:57:05 <quinq> And/or just SFTP clients
2023-11-12 10:04:09 <donoban> frag: maybe 'croc' is useful for you
2023-11-12 10:24:24 <frag> donoban: ++
2023-11-12 10:28:10 <donoban> is it an android phone? I think that you need f-droid for install the android client
2023-11-12 10:28:53 <frag> i prefer f-droid :]
2023-11-12 10:30:03 <donoban> :)
2023-11-12 15:51:34 <dewdude_> dwfreed, after sleeping on it the problem is like that I'm using a -virt kernal...since it's virtualized; so it likely won't have the network drivers built in since it's unusual to have a usb nic directly hooked to a VM
2023-11-12 15:52:32 <dewdude_> this wasn't an issue in the past since I was just passing disks
2023-11-12 15:55:10 <Habbie> remind me - alpine doesn't use kernel modules at all?
2023-11-12 15:55:30 <Habbie> hmm, no, not simply true
2023-11-12 16:21:25 <dewdude_> yup. that's all it was.
2023-11-12 16:21:31 <dewdude_> foiled by my own laziness
2023-11-12 18:29:39 <dwfreed> lol
2023-11-12 18:42:03 <holgersson> minimal: well, thanks. I'll go for a RAM installation w/ /var mounted on disk or sth. like that.
2023-11-12 18:47:14 <minimal> holgersson: you'll have the running OS loaded into RAM then, reducing available RAM
2023-11-12 18:54:58 <holgersson> minimal: yes, I know
2023-11-12 18:56:42 <minimal> holgersson: you may also hit issues updating the kernel when using run-from-RAM
2023-11-12 18:57:04 <holgersson> minimal: erm, how do I actually install a VM setup using tmpfs for /? Would I have to boot always from the ISO/ro image?
2023-11-12 18:57:31 <holgersson> So far I used alpine only on RPis (or in conttainers)
2023-11-12 18:57:34 <holgersson> *containers
2023-11-12 19:12:59 <minimal> holgersson: you boot a VM from the ISO where the VM also has a storage device and then you run setup-bootable to install run-from-ram Alpine to the storage device and then just boot VMs using that device in future?
2023-11-12 19:15:56 <holgersson> minimal: ah, that was the missing piece for me. thanks!
2023-11-12 19:24:08 <minimal> holgersson: at some point I'll get around to scripting run-from-RAM VM creation, I only do that for Sys-mode (run from disk) currently
2023-11-12 19:28:11 <holgersson> minimal: do you maintain the setup-* scripts?
2023-11-12 19:28:57 <minimal> holgersson: nope, I've got mym own developed scripts for creating Alpine physical/VM/Cloud disk images
2023-11-12 19:29:05 <holgersson> ah
2023-11-12 19:29:36 <minimal> https://github.com/dermotbradley/create-alpine-disk-image
2023-11-12 19:50:54 <holgersson> ah nice, thanks
2023-11-13 00:56:09 <omni> I use https://github.com/alpinelinux/alpine-make-vm-image
2023-11-13 10:59:33 <bananicorn> Hi, I'd love to use incus (lxd community fork) on alpine but don't have the experience to package it, does anyone else perhaps have time and interest? :)
2023-11-13 11:00:52 <bananicorn> Pinging fcolista since you are the current lxd maintainer, have you looked into incus at all?
2023-11-13 12:07:02 <Crocodillian> Hi, can I install alpine on a laptop using a different live stick, like nixos/ubuntu/fedora, also, does alpine support zfs?
2023-11-13 12:12:52 <caskd> yes you can install it using other live sticks however it takes significantly more effort
2023-11-13 12:13:26 <caskd> zfs should work just fine however i do not know if it is actively supported
2023-11-13 12:14:54 <Crocodillian> great thank you
2023-11-13 12:16:05 <Crocodillian> my problem with the alpine installer stick is that the font is too tiny on my crappy windows tablet that I'm installing it to, also there are no tools included that I need for the install
2023-11-13 12:18:21 <caskd> you may use setup-interfaces in combination with setup-apkrepos to prepare network and package repos to install further tools available on upstream repos
2023-11-13 12:18:54 <Crocodillian> cool
2023-11-13 12:19:03 <Crocodillian> you know if there's a way to change the console font?
2023-11-13 12:19:32 <caskd> off the top of my head, not really since i never required it but the arch wiki seems to hint this may be possible
2023-11-13 12:19:38 <caskd> ihttps://wiki.archlinux.org/title/HiDPI#Linux_console_(tty)
2023-11-13 12:20:01 <Crocodillian> great, thanks for your help, I will try this
2023-11-13 12:21:23 <caskd> alternatively it may be possible to install openssh-server on the live iso and ssh into the device from another which has a full desktop session
2023-11-13 13:06:39 <fcolista> bananicorn, no I didn't. 
2023-11-13 14:43:41 <WhyNotHugo> So /etc/openvpn/up.sh is not the upstream script; it's one included in aports. But it has comments and details that apply only to Debian or Gentoo.
2023-11-13 15:27:47 <minimal> bananicorn: I'd thought about packaging incus, but first I'm looking into differences between lxd and lxd-feature packages
2023-11-13 16:18:50 <bananicorn> minimal: Thanks for the update :)
2023-11-13 16:28:30 <tretinha> hey, i've been rocking sway + alpine for a few days here as my main workstation - setting dbus-update-activation-environment variables at first line of my sway config helped with screensharing and all so it's apparently all good in that front
2023-11-13 16:29:02 <tretinha> however, while trying to use virt-manager i keep getting errors related to dbus 
2023-11-13 16:29:16 <tretinha> i'm starting sway with a sway-run script after i log in at the tty 
2023-11-13 16:29:52 <tretinha> apparently my xdg runtime is correct and all... so i'm a bit lost on what to do to fix this 
2023-11-13 16:30:09 <tretinha> i'm also unable to upload self-signed pfx certs to firefox or chromium, for no apparent reason :P 
2023-11-13 16:30:35 <tretinha> i know this is all a little bit weird but i would appreciate any directions or thoughts
2023-11-13 16:30:36 <tretinha> thanks!
2023-11-13 16:38:52 <minimal> tretinha: perhaps you should detail what you tried and how it failed/error messages...
2023-11-13 17:22:32 <tretinha> I tried to elaborate here a bit on the certificate part, but it's exactly the same as this one: https://gitlab.alpinelinux.org/alpine/aports/-/issues/12965
2023-11-13 17:24:22 <tretinha> I unfortunately get the same result as the OP when I try to run pk12util, and when I try directly via firefox it just errors out with a message that provides no useful information whatsoever
2023-11-13 17:24:24 <tretinha> :/ 
2023-11-13 17:24:49 <tretinha> "The PKCS #12 operation failed for unknown reasons." for reference
2023-11-13 23:12:30 <bpalu> Hi, I have switched to alpine edge a while ago but today checked /etc/alpine-release and it still says 3.16.5
2023-11-13 23:13:13 <bpalu> as far as i can tell i am actually running edge, for example apk update also loads the package indexes from .../edge/...
2023-11-13 23:13:47 <bpalu> now i am confused if i might have misconfigured something because i would expect /etc/alpine-release to show a different version
2023-11-13 23:14:05 <Habbie> hi bpalu 
2023-11-13 23:14:12 <Habbie> can you tell me how you switched?
2023-11-13 23:14:23 <Habbie> i've never done this but i can do the experiment here and see what happens :)
2023-11-13 23:15:07 <bpalu> I changed the entries in /etc/apk/repositories
2023-11-13 23:15:28 <bpalu> and then I executed apk update && apk upgrade --available
2023-11-13 23:15:51 <bpalu> https://wiki.alpinelinux.org/wiki/Upgrading_Alpine
2023-11-13 23:16:08 <Habbie> ok, that doesn't update alpine-release here
2023-11-13 23:17:12 <bpalu> huh okay thank you. so i guess it just doesn't change the alpine-release and os-release files when you upgrade?
2023-11-13 23:18:53 <Habbie> ~ # apk info -W /etc/alpine-release 
2023-11-13 23:18:56 <Habbie> ERROR: /etc/alpine-release: Could not find owner package
2023-11-13 23:19:05 <Habbie> maybe!
2023-11-13 23:19:21 <Habbie> i don't know much
2023-11-13 23:27:09 <bpalu> thank you :)
2023-11-13 23:27:37 <bpalu> could anybody else confirm that this is expected?
2023-11-13 23:38:44 <poieen> hello guys, does anyone know how to prevent udhcpc or a specific OpenRC runlevels output to be printed to console? I'm using a stacked runlevel to start networking after/while logging in, which obviously makes the login screen all garbled up.
2023-11-13 23:39:27 <poieen> or rather not login screen but login prompt
2023-11-14 00:01:50 <poieen> adding the following line to /etc/network/interfaces seems to disable the output that was ruining the login prompt: 'udhcpc_opts &> /dev/null'
2023-11-14 00:02:20 <poieen> i'll look into better ways of doing this but ig this is fine thus far
2023-11-14 00:35:25 <minimal> bpalu: /etc/alpine-release is owned by the alpine-release package strangely enough ;-)
2023-11-14 00:40:31 <minimal> do you have this package installed?
2023-11-14 03:51:47 <qaqland> when would we use the new service manager instead of openrc? :p
2023-11-14 04:42:54 <ssm> why would you want to switch away from openrc? it rules
2023-11-14 04:43:09 <ssm> it's the closest thing I've used to bsd rc on linux
2023-11-14 05:46:41 <WhyNotHugo> pipeline for new aport fails because dependency is not found: https://gitlab.alpinelinux.org/alpine/aports/-/merge_requests/55355
2023-11-14 05:46:51 <WhyNotHugo> the dependency is part of the MR. Do I need to create a new MR for each dependency?
2023-11-14 05:49:13 <WhyNotHugo> Oh, the dependency actually failed to build, nvm
2023-11-14 06:04:36 <ikke> Yeah, the pipeline should build in dependency order 
2023-11-14 06:05:57 <WhyNotHugo> I had missed that the first packages had also failed. MR is ready now
2023-11-14 06:14:14 <rand0_> Hello, all. I am trying to build a custom driver for a Veikk tablet I have using https://github.com/jlam55555/veikk-linux-driver/. When I run `make`, I get `https://github.com/jlam55555/veikk-linux-driver/`. I have installed all of the required packages, such as linux-headers,  alpine-sdk, and linux-lts-dev. I am wondering if anyone can point me to a solution.
2023-11-14 06:15:17 <ikke> rand0_: the 2nd link is the same as the first, I guess a cp error?
2023-11-14 06:20:21 <rand0_> Aw, I don't know if anyone replied as my connection dropped. :(
2023-11-14 06:22:31 <ikke> I replied:
2023-11-14 06:22:32 <ikke> rand0_: the 2nd link is the same as the first, I guess a cp error?
2023-11-14 06:23:14 <rand0_> Hm. I don't recall which that may be. This is the driver https://github.com/jlam55555/veikk-linux-driver
2023-11-14 06:23:24 <rand0_> I thought that was the only link I shared.
2023-11-14 06:23:47 <ikke> rand0_: You said: "When I run `make`, I get: <link>'
2023-11-14 06:23:52 <ikke> that link points to the driver
2023-11-14 06:24:16 <ikke> It sounds like that should be a paste of the output you got
2023-11-14 06:24:19 <rand0_> Oh, sorry. I meant to paste:  /lib/modules/6.1.60-0-lts/build: No such file or directory
2023-11-14 06:24:23 <ikke> ah :)
2023-11-14 06:24:49 <ikke> rand0_: what does ls /lib/modules/ return?
2023-11-14 06:25:54 <rand0_> I am using "/lib/modules/`uname -r`" in the BUILD_DIR to get the right one, but the ls output is `6.1.60-0-lts/  6.1.62-0-lts/  6.5.9-0-edge/  6.6.1-0-edge/`
2023-11-14 06:33:06 <rand0_> I've never built drivers before, so I am unsure as to what I am missing but it seems like it should be pretty straightforward. *shrug*
2023-11-14 06:34:28 <Apachez> what could possibly go wrong? ;-)
2023-11-14 06:36:01 <rand0_> lol
2023-11-14 06:37:20 <ikke> rand0_: there is no package providing /lib/modules/*/build
2023-11-14 06:40:02 <rand0_> Yeah, I was foolishly hoping https://www.reddit.com/r/AlpineLinux/comments/14xz3v4/how_to_build_drivers/ was correct.
2023-11-14 06:40:14 <rand0_> Not really sure what else to look into as I've never done this before.
2023-11-14 06:49:24 <rand0_> Think I see why. I am currently on 6.1.60-0-lts, which lacks the build dir, but 6.1.62-0-lts does have it. Guess I need to restart and use the newer kernel.
2023-11-14 06:49:50 <rand0_> Thanks, @ikke
2023-11-14 06:50:09 <ikke> Right, I was thinking about that, but I could not find a package providing the builddir, so I discounted that 
2023-11-14 06:50:30 <rand0_> Odd. Guess I will try that now!
2023-11-14 06:51:04 <quinq> Feels like we had that exact same conversation a few months ago
2023-11-14 06:51:06 <quinq> bot?
2023-11-14 06:51:45 <rand0_> This is the first I've asked of it.
2023-11-14 06:52:15 <quinq> :)
2023-11-14 08:02:23 <rand0_> I've only just realised that my Grub configuration isn't using the latest LTS or Edge kernels I have installed. Looking into trying to set that up, I am hitting a wall. http://sprunge.us/4qGgTG my current config (pardon the messiness, I have some commented out lines present)
2023-11-14 08:03:44 <rand0_> By config I mean /etc/default/grub. My /boot/grub/grub.cfg is http://sprunge.us/lvAVMA
2023-11-14 08:05:16 <rand0_> Ideally, I would like to automatically target latest LTS and latest Edge, with edge as default. None of the entries have clearly identifiable kernel indicators, however. And the latest change I made logs me in but now shows a message about sparse something or other. 
2023-11-14 08:50:43 <bpalu> minimal: i confirmed that i have the right version of alpine-release.
2023-11-14 08:50:47 <bpalu> # apk list -I alpine-release
2023-11-14 08:50:49 <bpalu> alpine-release-3.19_alpha20230901-r0 x86_64 {alpine-base} (MIT) [installed]
2023-11-14 08:51:36 <bpalu> i tried reinstalling alpine-release but /etc/alpine-release still contains 3.16.5
2023-11-14 08:52:47 <bpalu> I mean this is not a real problem, i'm just curious why the file is not updated
2023-11-14 08:56:32 <bpalu> i reinstalled alpine-release with 'apk -v fix alpine release' output: https://paste.debian.net/plain/1298115
2023-11-14 08:59:31 <bpalu> but it did not change anything
2023-11-14 11:48:50 <CcxWrk> Where/how can I get debug symbols for libstdc++.so ?
2023-11-14 11:49:40 <ikke> We don't ship them by default, so you would have to rebuild the package with a -dbg subpacakge
2023-11-14 11:49:42 <ikke> subpackage*
2023-11-14 11:53:36 <CcxWrk> Hrm, that's annoying one. I thought splitdebug was commonly available for core libraries.
2023-11-14 11:54:23 <ikke> it is for musl
2023-11-14 11:54:28 <ikke> and some other packages
2023-11-14 11:55:46 <qaqland> @ssm: there is news that alpine linux would use a new one
2023-11-14 11:56:44 <ikke> No decission has been made so far. At the moment, there is nothing to switch to yet
2023-11-14 13:20:24 <Guido[m]1234> 
ACTION uploaded an image: (2509KiB) < https://matrix.org/_matrix/media/v3/download/matrix.org/aKznBgdMlBebxPfJWSdjLBSp/20231114_141754_7516268922570070401.jpg >

2023-11-14 13:23:57 <Guido[m]1234> Guido[m]1234: Ho guys, i'm continusly running into this issue while installing alpine in data mode via setup-alpine script and i can't figure out how can i go on, someone have any idea?
2023-11-14 13:24:15 <Guido[m]1234> sent an image.
2023-11-14 13:24:15 <Guido[m]1234> Hi guys, i'm continusly running into this issue while installing alpine in data mode via setup-alpine script and i can't figure out how can i go on, someone have any idea?
2023-11-14 13:24:15 <Guido[m]1234> * In reply to @gaumeloth:matrix.org
2023-11-14 13:29:21 <Guido[m]1234> * In reply to Guido... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/azkmosWieIpPgLQLsaFOxGJS>)
2023-11-14 13:29:51 <Guido[m]1234> * In reply to Guido... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/dCRelzMGxNwWOsORgzqHMktC>)
2023-11-14 13:31:39 <Guido[m]1234> In reply to @gaumeloth:matrix.org
2023-11-14 13:31:39 <Guido[m]1234> sent an image.
2023-11-14 13:31:39 <Guido[m]1234> Hello everyone, I keep encountering this issue when trying to install Alpine in data mode using the setup-alpine script. I'm unsure how to proceed. Does anyone have any suggestions
2023-11-14 13:32:25 <Guido[m]1234> * .In reply to @gaumeloth:matrix.org
2023-11-14 13:32:25 <Guido[m]1234> Hello everyone, I keep encountering this issue when trying to install Alpine in data mode using the setup-alpine script. I'm unsure how to proceed. Does anyone have any suggestions
2023-11-14 13:32:25 <Guido[m]1234> sent an image.
2023-11-14 15:32:14 <WhyNotHugo> powerctl has been crashing at startup for two weeks: https://gitlab.alpinelinux.org/alpine/aports/-/merge_requests/54364
2023-11-14 17:27:08 <ssm> qaqland: link? If they're going to change it to something stupid (systemd) I'm done with alpine
2023-11-14 17:28:53 <ssm> I really don't understand what practical reason there'd be to switch from openrc
2023-11-14 17:30:43 <Habbie> systemd? on musl? unlikely
2023-11-14 17:32:12 <ikke> ssm: openrc has its issues. We won't be switching to systemd
2023-11-14 17:32:34 <Habbie> i can think of plenty of reasons to switch from openrc, indeed, but there's plenty of sane options
2023-11-14 17:33:53 <ssm> ikke: something more specific?
2023-11-14 17:34:29 <ssm> I'm browsing the mailing lists, but I can't find anything about switching rc
2023-11-14 17:36:04 <frag> s6 last i heard
2023-11-14 17:40:46 <ssm> looks... complicated
2023-11-14 17:41:20 <ssm> yeah maybe I'm done with alpine, but I have no idea what I'd even switch to
2023-11-14 17:41:24 <bananicorn> Just read about it today, there is still work to be done on that to support a simpler declarative configuration along the lines of systemd
2023-11-14 17:41:34 <Habbie> procd is neat too
2023-11-14 17:43:12 <bananicorn> http://skarnet.com/projects/service-manager.html
2023-11-14 17:43:21 <bananicorn> Here you can read more about the vision for s6
2023-11-14 17:45:12 <Habbie> that should come with dates :)
2023-11-14 17:45:29 <Habbie> but thanks, will read, seems interesting
2023-11-14 17:47:37 <bananicorn> It is a few years old, last thing I read the author was held up with some other work, so no idea when any of this is coming
2023-11-14 17:47:57 <bananicorn> And I have no clue about the technical side, so I can't speak to how promising it is :)
2023-11-14 17:48:19 <bananicorn> Maybe someone else here can say more
2023-11-14 17:54:20 <OdaNobunaga[m]> Last thing I read on Reddit was that Bercot needed to update his whole s6 stack. And gaming seems to distract him, which I can understand only too well. XD
2023-11-14 19:50:30 <pj> s6 in alpine? probably in next millenium
2023-11-14 19:52:37 <Habbie> bananicorn, thanks
2023-11-14 19:58:27 <Habbie> the table in that article, while i don't agree with everything in it, does explain why some people might want to move away from openrc
2023-11-14 20:23:04 <ssm> I don't really want something that works like systemd no matter how small or efficient it is personally
2023-11-14 20:23:07 <ssm> maybe I'm in a minority there
2023-11-14 20:23:48 <Habbie> 'works like systemd' is not very specific
2023-11-14 20:25:11 <ssm> Habbie: "The last piece of a full alternative to systemd" from the website you linked
2023-11-14 20:25:18 <ssm> sorry, bananicorn, not you
2023-11-14 20:26:04 <Habbie> so what aspect bothers you?
2023-11-14 20:30:13 <ssm> the fact I have a hard time describing what some systems of s6 do whereas I can understand how openrc works in 5 minutes
2023-11-14 20:30:29 <ssm> I don't want a process supervisor, or auto healing daemons
2023-11-14 20:31:13 <ssm> I'm primarily an openbsd user, and I like alpine for its simple bsd-style systems
2023-11-14 20:31:18 <ikke> s6 at the moment is mostly building blocks
2023-11-14 20:31:35 <ikke> so you should not judge it as a full init system at the moment
2023-11-14 20:31:44 <ssm> I can judge its project goals
2023-11-14 20:33:45 <dalias> you don't have to think of it as "auto-healing daemons"
2023-11-14 20:34:00 <dalias> a supervisor is just the right way to be able to command start/stop/restart
2023-11-14 20:34:16 <dalias> rather than hacky race conditions with pidfiles and manually killing and starting
2023-11-14 20:34:33 <Habbie> ssm, right - i need to admit here, when i ran freebsd 20 years ago, the first thing i installed was daemontools :D
2023-11-14 20:34:46 <Habbie> and indeed, it is about control, and consistent environments, for me
2023-11-14 20:34:49 <Habbie> automatic restarts are a choice
2023-11-14 20:36:59 <invoked> openrc's got a lot of issues but at least we know what those issues are.
2023-11-14 20:37:48 <invoked> i don't think anybody gets into any init system without stepping on a rake at some point.
2023-11-14 20:47:16 <ssm> can parallel service loading not be disabled in openrc to prevent race conditions?
2023-11-14 20:47:48 <ssm> I assume that's where the race conditions would be, anyways
2023-11-14 20:48:09 <ssm> and there are boot targets anyways, which is supposed to mitigate that
2023-11-14 20:50:14 <invoked> if it were easy enough to patch out, it would have been by now.  anyway: https://github.com/OpenRC/openrc/issues?page=2&q=is%3Aissue+is%3Aopen
2023-11-14 20:53:55 <ikke> ssm: the default for openrc is serial loading, not parallel
2023-11-14 20:54:59 <ssm> ...so where's the race condition
2023-11-14 20:55:18 <ssm> invoked: your link returns no results
2023-11-14 20:56:22 <invoked> no idea.
2023-11-14 20:56:24 <ikke> ssm: that's in the section "Supports readiness notification", so it's per service, not accross all services
2023-11-14 20:56:53 <ikke> ie, can a service notify that it's actually ready (not just the process has started)
2023-11-14 20:57:40 <ikke> (fyi, I'm not familiar enough with openrc to say where the race conditions are)
2023-11-14 20:57:52 <ssm> ready?
2023-11-14 20:58:07 <ssm> if the process started it has a PID
2023-11-14 20:58:45 <ikke> No, that's not when it's ready
2023-11-14 20:59:09 <ikke> It's ready when it's able to perform it's task, which is not per definition when the process has started
2023-11-14 21:00:02 <ssm> that sounds like an abstraction that not every daemon fits into
2023-11-14 21:00:07 <ikke> A webserver is ready when it's able to accept http requests, a db is ready when it's able to answer queries
2023-11-14 21:00:28 <ikke> ssm: sure, not every daemon has to provide readiness, but it's nice to be able to use it when it's usefull
2023-11-14 21:00:46 <ikke> Because something has a feature does not mean everything needs to use it
2023-11-14 21:02:01 <ssm> sounds like a slippery slope, what other hooks should I poke into daemons to see what state they're in?
2023-11-14 21:02:09 <ssm> spawning a PID should be the only anchor for any usecase imo
2023-11-14 21:03:39 <ssm> this is how you get gnu extension hell, which I thought alpine was trying to avoid
2023-11-14 21:04:20 <Ermine> There are no other hooks as far as process supervision is concerned. Every 'other hook' is called monitoring and is out of scope of process supervision and s6 author refused to implement such stuff
2023-11-14 21:05:23 <Ermine> and your way of thinking is how you get non sequiturs
2023-11-15 01:37:24 <alpine> what does `unsupported relocation type 37` mean
2023-11-15 02:07:23 <WhyNotHugo> ssm: pids are racey, and an ugly hack. A supervisor runs the daemon as a child, it can signal it easily withou race conditions and knows exactly when it exits.
2023-11-15 02:07:55 <WhyNotHugo> readiness notifications are opt-in, for services where it is relevant (this helps when serviceB depends on serviceA being *ready*).
2023-11-15 02:08:42 <WhyNotHugo> automatic restarts are, as far as I can tell, opt out
2023-11-15 02:09:26 <WhyNotHugo> You can't have parallel services without readiness notification unless NONE of your services have any dependency ordering
2023-11-15 03:12:29 <elly> lnl: thank you for working on the chromium aport!
2023-11-15 05:36:10 <qiaoran> hello!
2023-11-15 05:37:51 <qiaoran> DDoS
2023-11-15 05:39:02 <link> help
2023-11-15 05:40:31 <quinq> s/link/zelda/
2023-11-15 12:12:54 <lnl> elly: no problem :3
2023-11-15 12:51:55 <jonesv> Hello :). I am considering moving to Alpine Linux (I was on Arch for 10 years, I tried QubesOS for a while, and now I want to try Alpine).
2023-11-15 12:52:30 <jonesv> Is it a bad idea to go for Alpine as a desktop distro? I use it in containers already, but I never tried for my main OS.
2023-11-15 12:53:37 <ikke> There are plenty of people who use alpine on the desktop, including ne
2023-11-15 12:53:42 <ikke> Me*
2023-11-15 12:58:23 <jonesv> ikke: are there some things that clearly don't work really well? E.g. I would like to do Android dev (so I need IntelliJ/Android-Studio)
2023-11-15 12:58:44 <jonesv> I don't think I need a lot other than that, I mostly use the terminal, and I want to try Sway (I was on i3 on X11)
2023-11-15 12:59:25 <jonesv> Also I'm happy to leave systemd and try something else :-). I was interested in s6 but I understand Alpine is still on OpenRC (which is fine)
2023-11-15 13:05:06 <ikke> Pre-compiled binaries that rely on glibc won't work by default 
2023-11-15 13:08:25 <telmich> A note from my side, also using Alpine on notebooks for some years now: if you really depend on glibc apps, using docker on alpine might be a workaround, depending on which apps you need to use
2023-11-15 13:08:49 <jonesv> I am not sure how much glibc is in Android-Studio. I guess a bit?
2023-11-15 13:09:03 <jonesv> I tried to search for "Android-Studio on Alpine Linux" but I really haven't seen much
2023-11-15 13:12:37 <ikke> Flatpak is often also used
2023-11-15 13:16:48 <jonesv> Does flatpak help with libc?
2023-11-15 13:20:02 <ikke> It would, similar to docker 
2023-11-15 13:21:46 <jonesv> oh, and instead of flatpak could I do it manually with bubblewrap, just for Android-Studio for instance?
2023-11-15 13:23:20 <fluix> don't see why not
2023-11-15 13:33:35 <jonesv> Oh I think it is done here: https://wiki.alpinelinux.org/wiki/Running_glibc_programs#Chroot_+_Bubblewrap
2023-11-15 13:47:29 <PureTryOut> Is there anything I can configure to make IO starving less of a problem? Almost always when I build a package on my laptop the whole machine slows down while installing all the dependencies to the point that music starts stuttering. I'd like to be able to use the machine when I build packages 😅 This is a problem even though the machine runs on a NVMe SSD
2023-11-15 13:53:07 <ikke> Do you have enough memory? That's most of the time the cause of stuttering/freezing for me
2023-11-15 13:59:04 <PureTryOut> 16GB of which ~5.6GB is in use while installing those deps
2023-11-15 13:59:06 <PureTryOut> I'm pretty sure it's IO
2023-11-15 14:55:36 <ikke> You can try ionice to give it less priority 
2023-11-15 15:18:49 <jonesv> Do you know the state of s6 for alpine? I read that Alpine was considering transitioning to it a couple years ago, but I haven't heard about it recently
2023-11-15 15:20:02 <Habbie> that came up yesterday as well
2023-11-15 15:20:03 <ikke> jonesv: as per skarnet, s6 is not ready yet as a replacement for openrc 
2023-11-15 15:20:38 <ikke> So before we can even consider switching, that would first need to happen 
2023-11-15 15:57:10 <andym48[m]> What is wrong with openrc?
2023-11-15 16:06:34 <durrendal> supervision seems to be the issues most people have with it. It's just not particularly awesome at it
2023-11-15 16:06:58 <durrendal> but it isn't systemd, so you win some you lose some
2023-11-15 16:23:39 <andym48[m]> Yes - it isn't systemd!
2023-11-15 16:24:05 <andym48[m]> Can you explain 'supervision' for me?
2023-11-15 16:26:03 <ikke> andym48[m]: keeping an eye on the status of the process and restarting it if necessary / desired
2023-11-15 16:28:53 <andym48[m]> Thank you
2023-11-15 17:28:50 <jonesv> Right, thanks for the news :)
2023-11-15 17:40:48 <bz> it seems like a lot of newer software that is meant to run as a background process has no functionality for daemonizing itself or process supervision, presumably because systemd provides that functionality and devs dont feel any need to do it anymore
2023-11-15 17:41:13 <bz> which is fine for people using oses that use systemd but annoying for everyone else
2023-11-15 17:43:38 <bz> one of the many unfortunate side effects of having a huge monolithic feature-creep master control program like systemd becoming the de-facto standard on most linux distributions
2023-11-15 17:45:28 <bz> but thats why openrc has supervision functionality and why freebsd has its own daemonizing program and things like that
2023-11-15 17:48:38 <bz> to deal with software that doesn't daemonize itself or create a pid file even though it should
2023-11-15 17:51:32 <ssm> bz: you can daemonize a command with an asynchronous list (&)
2023-11-15 17:52:21 <bz> thats how ive done it on netbsd which doesnt have a daemonizing utility in the base system (that i know of, anyway)
2023-11-15 17:52:40 <bz> but i dont know if thats like the "proper" way to do it or not, but its worked for me in the past
2023-11-15 17:54:13 <bz> and logging to a logfile with the > operator and making the rc script create the pid file and all that
2023-11-15 18:16:46 <ssm> can't you just program supervision into an openrc script with `false; while [ $? -gt 0 ]; do run_daemon; done`
2023-11-15 18:17:02 <ssm> unless the daemon exits normally, it will restart
2023-11-15 18:17:53 <Hello71> systemd absolutely did not invent supervision, didn't invent forkless daemons, and doesn't claim to have invented it. daemontools had supervision and forkless daemons more than 20 years ago
2023-11-15 18:18:49 <Hello71> and s6 inherited that from daemontools, not systemd
2023-11-15 18:19:13 <ssm> I don't care who inherited what, a bad extension is a bad extension
2023-11-15 18:19:41 <ssm> people don't know how to use unix, so they make extensions to overcome hypothetical shortcomings
2023-11-15 18:19:59 <Hello71> this is a dumb reactionary take. stuff is not bad because it's from systemd, it needs to be evaluated on its own merits
2023-11-15 18:20:12 <ssm> <I don't care who inherited what>
2023-11-15 18:20:32 <invoked> software being specifically dependent on systemd is a strong time-saving clue that it's probably crap
2023-11-15 18:21:26 <alpine> https://github.com/google/sanitizers/issues/1080   ;-;
2023-11-15 18:21:47 <alpine> [Feature request] musl support · Issue #1080 · google/sanitizers · GitHub
2023-11-15 18:47:07 <ksandstr> systemd mythology also used to ascribe the invention of dependency-ordered booting to it. seems reasonable to take grains of salt with all of that
2023-11-15 19:06:32 <dari> PureTryOut: on which kernel you have freezes? -lts or -edge?
2023-11-15 19:07:19 <PureTryOut> default, so -lts
2023-11-15 19:09:07 <dari> PureTryOut: with nvme your scheduler is probably "none" I never got luck with it, you could try with queue/scheduler = mq-deadline
2023-11-15 19:11:08 <dari> also setting vm.dirty_background_bytes and vm.dirty_bytes to low values like 8MB helped me get rid of such a weird freezes 
2023-11-15 19:30:12 <Saijin_Naib[m]> I have had an amazing improvement on eMMC with BFQ, though supposedly NVME doesn't need a scheduler due to native multiple queue architecture 
2023-11-15 19:31:07 <Saijin_Naib[m]> Kyber, mq-deadline, none, etc all were ass on eMMC and lead to lockups and stuttering under even minimal load. 
2023-11-15 19:32:02 <Saijin_Naib[m]> I lost like, 2% max IO speed using BFQ vs none, but it is far more responsive and consistent with response time 
2023-11-15 19:32:40 <Saijin_Naib[m]> I'm using it on my SATA3 SSDs as well, same pattern there, though they never locked up unlike my eMMC 
2023-11-15 19:34:00 <ssm> I thought the general consensus was NOOP was best for flash storage, but it's been years since I looked up anything about that
2023-11-15 19:35:51 <Saijin_Naib[m]> Noop is not multithread, so it's been replaced by none since a few yards 
2023-11-15 19:35:55 <Saijin_Naib[m]> Years, even 
2023-11-15 19:36:52 <Saijin_Naib[m]> And yes, folks give and repeat advice dogmatically, but sometimes it's not actually good (like in my testing for my machines ) 
2023-11-15 19:39:47 <Saijin_Naib[m]> Use the ol' eyeball benchmark to get a feel for the schedulers, then get actual data with GNOME Disk Utility or similar to fact check your eyeball benchmark 
2023-11-15 22:42:06 <jonesv> > software being specifically dependent on systemd is a strong time-saving clue that it's probably crap
2023-11-15 22:42:15 <jonesv> Unfortunately there is a lot of crap out there :D
2023-11-15 22:43:11 <jonesv> Or "crap" got redefined. I sometimes feel like the goal today is productivity, and profitable software is "good" software. Which I don't fully agree with, to say the least :)
2023-11-15 22:57:34 <WhyNotHugo> Why does usage of gpep517 typically use '--output-fd 3 3>&1'. Isn't this functionally equivlent to '--output-fd 1'?
2023-11-15 23:11:11 <OdaNobunaga[m]> <ikke> "jonesv: as per skarnet, s6 is..." <- Has anyone experience with Artix or Obarun? They have been using s6 for some time.
2023-11-15 23:39:13 <ptrc> WhyNotHugo: nope, because some build scripts are *bad* and might, for example, close stderr or mess with it
2023-11-15 23:40:05 <ptrc> in this case, we let them close fd 3 while still keeping our stdout and stderr intact
2023-11-15 23:42:00 <Tommy[m]12> Hello everyone... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/xPVoHMyLbbozvuENJBRELuIP>)
2023-11-16 01:44:36 <bz> personally i'm of the belief that an init system should just be an init system and software should just focus on doing one thing for the sake of simplicity when you have an os composed of lots of different small things written by different people that all need to work together as seamlessly as possible
2023-11-16 01:46:37 <bz> and lots of different distributions of said os that do things differently
2023-11-16 01:48:59 <bz> but i dont have any evangelical reasons for this, it just seems like the least confusing way to do it in the long run
2023-11-16 03:29:26 <Tommy[m]12> Hello everyone... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/qtZHksBFoQYDOWEjXCTIuwhp>)
2023-11-16 03:36:55 <ssm> imagine ever ordering spambot drugs off irc
2023-11-16 03:38:21 <Daanct12> matrix*
2023-11-16 03:38:24 <Daanct12> :P
2023-11-16 04:15:27 <charlatan_> Hi there! I have some questions about setting up Alpine as my home firewall. I am planning to use netfilter.
2023-11-16 04:16:49 <charlatan_> First q - is it acceptable to setup a user other than root when installing? I am aware that some other firewalls setup cert authentication and then remove the root account.
2023-11-16 05:24:06 <grawity> "setting up another user" and "removing root" are very different things
2023-11-16 05:24:59 <grawity> as are "another user" vs "cert authentication"
2023-11-16 05:26:46 <grawity> I believe it's *always* acceptable to create a non-root account for adminning via sudo/doas – alpine is a general purpose distro, not a sealed firewall appliance – although in some cases (like a system with only 1 admin who only ever edits /etc/nftables.conf and nothing else) it might not give advantages
2023-11-16 05:27:58 <grawity> SSH key-based authentication can be used with either root or non-root, and it should be set up regardless of what you do with the root account
2023-11-16 10:41:34 <donoban> anyone has experience booting a luks root with a keyfile on a usb or similar? I suppose that it will ignore /etc/crypttab, maybe passing all needed as kernel cmdline?
2023-11-16 10:57:22 <donoban> the mission is a little more difficult, its RAID-1 btrfs with two LUKS partitions. I suppose that btrfs will refuse to mount if it misses one and I don't see how to specify multiple devices for unlock
2023-11-16 11:39:20 <Apachez> drop btrfs and use bcachefs - not that it will fix your issue but still ;)
2023-11-16 11:40:32 <aron> I wonder, how accurate is alpine's position on distrowatch?
2023-11-16 11:43:56 <donoban> well I could consider if decided to reinstall :)
2023-11-16 12:00:41 <dok> hi, just installed alpine on a desktop machine, I struggled to get Xorg working without starting X as root, finally what I needed was for the user to be in the input group. Any idea why this is not the default ?
2023-11-16 12:01:50 <dok> (maybe because i created the user before running setup-xorg-base)
2023-11-16 12:02:55 <ikke> Because that's not the only way to achieve that (some users uses something like seatd
2023-11-16 12:15:31 <Ermine> Because wayland compositors generally use seatd/elogind, and having user in input group is security issue in this case)
2023-11-16 12:18:40 <dok> okay, and can I use seatd without wayland ?
2023-11-16 12:43:46 <Ermine> With software which supports seatd
2023-11-16 13:17:25 <telmich> Has anyone tried to compile / get teleport running on alpine linux? Seems "just go code", so in theory it should work
2023-11-16 13:18:11 <Ermine> what is teleport
2023-11-16 13:18:44 <telmich> it's a proxy-ish software that allows node to register to a proxy (read: website) and allows to connect to nodes via that website and/or via cli
2023-11-16 13:18:55 <telmich> it's a bit more of a fancy ssh that allows auditing
2023-11-16 13:19:23 <Ermine> please give a link to it
2023-11-16 13:23:12 <telmich> main website of teleport is: https://goteleport.com/
2023-11-16 13:23:32 <telmich> installlation docs are on https://goteleport.com/docs/installation/
2023-11-16 13:39:03 <Ermine> Well, it's go, so it shouldn't be problematic. At least you can fall back on their docker images
2023-11-16 13:41:09 <telmich> I actually tried that
2023-11-16 13:41:17 <telmich> And it's not really usable, if you need the client side
2023-11-16 13:41:29 <telmich> Because tsh and kubectl work together and interact with each other
2023-11-16 13:42:07 <telmich> So if you use teleport for proxying to k8s, kubectl uses tsh that generates credentials inside the container
2023-11-16 18:05:55 <frag> when i log one netbank with browser, i need official id number, password, and a one-time-password from a code brick, to acces netbank through mobile app, i need a 4 digit pin code ...
2023-11-16 18:10:12 <Habbie> was this also true the first time you installed the app?
2023-11-16 18:15:09 <frag> no, hehe, then you have to log in like through a browser
2023-11-16 18:19:54 <Habbie> ok
2023-11-16 18:20:05 <Habbie> likely it identified your phone at that point and trusts it now
2023-11-16 18:23:39 <frag> hmm yes :/
2023-11-16 18:24:00 <frag> > we know EXACTLY who you are, where you are, when you go to bathroom etc..
2023-11-16 23:52:35 <Quillith> Hi, I've really enjoyed using Alpine as desktop this past year or so! Question- has anyone had success modding Baldur's Gate 3? Usually I manually move mod files around since even Linux mod-managers aren't usually compatible with Alpine, but I can't find how for this game
2023-11-16 23:57:43 <Quillith> Closest thing I've found is Lamp, a mod manager for Linux. I found almost all of its listed requirements on the Alpine Linux packages page except for p7zip-full & p7zip-rar , so close but no cigar
2023-11-17 00:15:48 <dari> Quillith: package "7zip"
2023-11-17 00:17:11 <Quillith> ah, thank you! I'll give building this a go then
2023-11-17 00:57:14 <Quillith> ahh, unfortunately this linux program seems to require glibc. my search resumes
2023-11-17 01:17:03 <invoked> ?
2023-11-17 01:29:53 <Quillith> Alpine isn't compatible with programs that use glibc I thought
2023-11-17 01:30:39 <ssm> Quillith: it is, if you use runtimes (flatpak/appimage)
2023-11-17 01:31:31 <ssm> as much as runtime container nonsense is usually bloat, it's actually quite handy on alpine, especially for proprietary glibc software like steam
2023-11-17 01:46:24 <Quillith> Oh yeah, I use flatpak for Steam! But Steam is an existing flatpak. Can I move a program into a pre-existing flatpak or would I need to build a flatpak from scratch for a program?
2023-11-17 01:46:54 <Quillith> like if I already have a Steam flatpak, can I just move this program into its directory? Or is that silly
2023-11-17 01:53:35 <ssm> I don't know exactly how flatpak works internally, but if you could create a glibc chroot it should have the same effect
2023-11-17 01:54:01 <ssm> chances are it may not work just trying to run your software chrooted into the flatpak, as it would likely lack the correct dependencies
2023-11-17 01:54:55 <ssm> also, it'll potentially break every time you update the flatpak
2023-11-17 01:57:12 <Quillith> hm sounds like a headache down the line. Still, making it a flatpak is a great idea. If I can pull it off it'd be great. I'll start giving it a go, thanks for the advice!
2023-11-17 01:57:43 <Quillith> I found some documentation on making programs into flatpaks, so why not try? lol
2023-11-17 02:08:50 <jerkwad_robot> Thanks grawity for the feedback
2023-11-17 03:23:03 <alpine> somehow chrony has become out of sync o.o
2023-11-17 03:23:38 <alpine> the rest of the world is 1 pm but chronyc makestep reports 10 am
2023-11-17 03:26:37 <dwfreed> what is the output of 'chronyc tracking'
2023-11-17 10:55:08 <Fxzxmic> What should I do to split bash-completion into a separate subpackage?
2023-11-17 10:56:09 <Fxzxmic> https://pkgs.alpinelinux.org/package/edge/community/x86_64/xca
2023-11-17 10:57:27 <ikke> Fxzxmic: usually it's sufficient to declare a $pkgname-bash-completion subpkg 
2023-11-17 10:57:59 <Fxzxmic> The package maintainer did not do this, but I think it would be better to split it into a separate subpackage.
2023-11-17 10:58:31 <Fxzxmic> ikke: Nothing else?
2023-11-17 10:58:37 <ikke> Eirher create an MR or an issue (in that order of preference)
2023-11-17 10:59:00 <ikke> abuild already has default split functions for that 
2023-11-17 10:59:33 <Fxzxmic> Oh, that's great.
2023-11-17 11:40:38 <Forza> Which will be the next alpine release? 3.18.5 or 3.19?
2023-11-17 11:41:07 <Forza> I saw kernel 6.6 will be next LTS release. Will that be used with next alpine? https://www.phoronix.com/news/Linux-6.6-Goes-LTS
2023-11-17 11:41:58 <dari> Forza: most probably 6.6 will be in next
2023-11-17 11:42:44 <dari> Forza: now you can try use linux-edge 6.6 
2023-11-17 11:47:39 <Forza> :) will do
2023-11-17 11:52:32 <Fxzxmic> https://gitlab.alpinelinux.org/alpine/aports/-/merge_requests/55567  Does this look okay?
2023-11-17 14:40:59 <jonesv> This should get me into a functional Alpine system, shouldn't it? https://openzfs.github.io/openzfs-docs/Getting%20Started/Alpine%20Linux/Root%20on%20ZFS.html
2023-11-17 14:41:39 <jonesv> I followed it all, but upon reboot I am now stuck in an initramfs emergency recovery shell that says: "cannot import 'rpool': no such pool available
2023-11-17 16:50:48 <pj> Forza: patch releases contain only bug/security fixes
2023-11-17 17:00:11 <Fxzxmic> When a MR has been approved, how will it be merged?
2023-11-17 17:00:30 <ikke> Fxzxmic: Someone with permissions will eventually come to it and merge it
2023-11-17 17:00:58 <ikke> but we are now working in the 3.19 release, so less focus on merging MRs unless they fix some broken package
2023-11-17 17:01:55 <Fxzxmic> Ok, understood
2023-11-17 23:42:08 <Forza> pj: of course. Forgot that. But if we get 3.19, it'll probably use the 6.6 as it's the lts kernel? 
2023-11-18 06:13:42 <ptrc> ..did anyone else had an issue recently with sdl2_ttf rendering boxes instead of characters on aarch64?
2023-11-18 06:14:02 <ptrc> can reproduce it with VIDEODRIVER wayland and kmsdrm, but only on aarch64
2023-11-18 06:14:24 <ptrc> the font is loaded manually from a file, the exact same files (NotoSans-Regular.ttf, etc.) work on other architectures
2023-11-18 06:14:30 <ptrc> this might be the weirdest bug yet
2023-11-18 09:13:16 <WhyNotHugo> I see several tree-sitter packages in aports. Is anyone using these with neovim? Upstream support for tree-sitter relies on pinned versions, but I'm curious if anyone has worked around that.
2023-11-18 09:37:21 <Kladky> nvim-treesitter, right? Seems strange to me that it would require pinned versions, especially considering how barebones nvim is compared to something like helix.
2023-11-18 10:18:40 <pj> tree-sitter grammars rely heavily on tree-sitter library version
2023-11-18 10:19:02 <pj> that's universal thing across all editors
2023-11-18 11:50:36 <freem> Hello. I was wondering what is the policy of alpine when a daemon is installed, that is, is it automatically started at any boot or not, and if yes, is there is a way to prevent that? I am asking because I am quite tired of debian starting daemons from packages I install automatically while I usually (~99% times) would prefer to integrate them into the daemon "supervision" I use (runit) and the names and behavior tend to
2023-11-18 11:50:40 <freem> version of the system
2023-11-18 12:00:32 <freem> I tried searching the documentation, but I have not found the answer to this question, might not have searched correctly, but the documentation seems rather lightweight
2023-11-18 12:00:33 <Kladky> They are not started at boot by default
2023-11-18 12:00:38 <freem> oh great, thanks
2023-11-18 12:00:49 <freem> this is perfect
2023-11-18 12:00:55 <freem> (and the sane way)
2023-11-18 12:01:07 <Kladky> Very much so
2023-11-18 12:01:18 <Kladky> You can start services at boot using rc-update
2023-11-18 12:03:35 <ikke> indeed, we do not start any service on installing a package or enable it at boot, that's at the users descretion
2023-11-18 12:23:27 <freem> well, if I switch my vps to alpine (I've been thinking about it since a long while now, just some change resistance issue from me to fix) I'll likely use either busybox's runit implementation, runit itself since I'm more familiar with daemontools and already have stuff to work with it. Maybe even give a shot at s6, that one seems interesting, as it fixes some issues from the usual daemontools without bringing much comple
2023-11-18 13:21:17 <freem> BTW, apparently, https://docs.alpinelinux.org/user-handbook/0.1a/Installing/manual.html is not up to date, it says:
2023-11-18 13:21:30 <freem> from the primary mirror of the latest release version:  http://dl-cdn.alpinelinux.org/alpine/v3.13/main
2023-11-18 13:22:01 <freem> but on http://mirrors.ircam.fr/pub/alpine/ there's a 3.19.
2023-11-18 13:23:31 <ikke> 3.19 has not been released yet, it's still being built
2023-11-18 13:24:02 <freem> oh, so I should target 3.18 then?
2023-11-18 13:24:15 <ikke> Yes
2023-11-18 13:24:17 <freem> thx
2023-11-18 13:24:24 <tzeentch[m]> freem: using latest-stable
2023-11-18 13:24:47 <tzeentch[m]> and it will update to stable until 3.19 release
2023-11-18 13:25:12 <freem> I kind of like frozen versions, as they don't surprise when updating
2023-11-18 13:26:06 <freem> although I guess the number of changes to do would make it obvious that a version change occurred... I'll think about it when I'll have finished setup that VM :)
2023-11-18 17:20:16 <ChrisGoldapp> Hello everyone! :)
2023-11-18 17:21:29 <ChrisGoldapp> I am having an issue with installing Alpine on a ARM 32 bit QEMU guest on my 64 bit x86-64 Linux host. 
2023-11-18 17:22:26 <ChrisGoldapp> I downloaded the virt ARMv7 image, and invoked QEMU like: 
2023-11-18 17:22:27 <ChrisGoldapp> qemu-system-arm -M virt -cpu cortex-a15 -m 8G -hda alpine-arm.img -cdrom ~/Downloads/alpine-virt-3.18.4-armv7.iso -boot d -net nic -net user -nographic -serial mon:stdio
2023-11-18 17:23:03 <ChrisGoldapp> I expected to see the Alpine installer on the CLI, but nothing happened.
2023-11-18 17:23:44 <ChrisGoldapp> Nothing happened as in: the command seems to run, but not output appears.
2023-11-18 17:23:48 <ChrisGoldapp> Do you know where I went wrong?
2023-11-18 17:25:22 <ChrisGoldapp> I have noticed an entire CPU core is pegged to 100%, maybe I just need to wait a bit.
2023-11-18 18:02:01 <minimal> ChrisGoldapp: how are you expecting it to boot? BIOS is PC only. That ISO appears to be using UEFI for booting and so you'd need to pass qemu-system-arm the correct arguments for it to use UEFI
2023-11-18 18:11:20 <ChrisGoldapp> minimal: Thank you, I didn't know that. I expected that virt automatically sorted this out, but it seemed I was mistaken.
2023-11-18 18:14:03 <ChrisGoldapp> I've found this Stackoverflow post: https://superuser.com/questions/1397991/running-alpine-linux-on-qemu-arm-guests
2023-11-18 18:14:20 <ChrisGoldapp> So it appears I need to specify additional arguments to get it to run.
2023-11-18 18:34:01 <minimal> ChrisGoldapp: unlike PCs there is no standardised booting method for "random" Arm physical devices, some use U-Boot, some use UEFI, some use proprietary bootloaders (like Raspberry Pi)
2023-11-18 18:34:16 <minimal> this is nothing specific to Alpine, its a general Arm issue
2023-11-18 18:36:20 <minimal> that superuser.com article is bypassing the use of a bootloader altogether by passing qemu the kernel and initramfs files directly
2023-11-18 18:36:36 <ksandstr> ie. "arm is not a platform"
2023-11-18 18:38:01 <minimal> so you have 2 options, either (a) extract the kernel and initramfs from the ISO and do similar as that article, or (b) run qemu-system-arm and pass it arguments to UEFI firmware to use in order to then boot via grub-efi (which is what the armv7 virtual ISO has configured)
2023-11-18 18:43:09 <ChrisGoldapp> minimal: Okay, thank you. I'll look into that some more.
2023-11-18 18:45:42 <ChrisGoldapp> Okay I downloaded the kernel and initramfs from that repo, but the armv7 version. It works now!
2023-11-18 18:45:49 <ChrisGoldapp> Thank you so much, minimal!
2023-11-18 20:08:15 <elly> is anyone here particularly wise in debugging X problems? my desktop, running edge, has an AMD Radeon 6750, and when I start X (with no config file), the X server hangs with some junk on the screen
2023-11-18 21:14:25 <bananicorn> Hi, is there a place where I can find the pubkeys for the repos on the alpine websites?
2023-11-18 21:17:04 <ikke> bananicorn: there is, but there is no index
2023-11-18 21:17:44 <bananicorn> Never mind, I should just use the ones from the package I suppose :)
2023-11-18 21:17:45 <ikke> bananicorn: https://alpinelinux.org/releases.json lists them
2023-11-18 21:18:19 <bananicorn> Oh that's great to know though, thank you ikke
2023-11-19 01:17:13 <ChrisGoldapp> Hello everyone. It's me, again.
2023-11-19 01:17:36 <ChrisGoldapp> I managed to get Alpine Linux running inside a ARMv7 32-bit QEMU guest earlier.
2023-11-19 01:18:33 <ChrisGoldapp> But unfortunately, it won't start up right and it says "IP request but no network device found", and consequently the install fails.
2023-11-19 01:18:56 <ChrisGoldapp> I suppose this is more of a QEMU problem but maybe someone of you knows the solution.
2023-11-19 01:19:06 <ChrisGoldapp> qemu-system-arm \
2023-11-19 01:19:07 <ChrisGoldapp>   -net nic -net user \
2023-11-19 01:19:07 <ChrisGoldapp>   -M virt -m 4G -cpu cortex-a15 \
2023-11-19 01:19:07 <ChrisGoldapp>   -kernel armv7-vmlinuz-lts -initrd armv7-initramfs-lts \
2023-11-19 01:19:07 <ChrisGoldapp>   -append "console=ttyAMA0 ip=dhcp alpine_repo=http://dl-cdn.alpinelinux.org/alpine/edge/main/" \
2023-11-19 01:19:07 <ChrisGoldapp>   -hda alpine-armv7.qcow2
2023-11-19 01:19:25 <ChrisGoldapp> I ran this command, and confoundingly this works if I use aarch64 (changing everything appropriately)
2023-11-19 01:19:39 <tzeentch[m]> ChrisGoldapp: leaks netdev
2023-11-19 01:21:52 <ChrisGoldapp> tzeentch[m]: I tried setting netdev explicitly, but only found a solution for x86_64 guests
2023-11-19 01:23:18 <tzeentch[m]> -device virtio-net,netdev=net0 \
2023-11-19 01:23:18 <tzeentch[m]>  -netdev user
2023-11-19 01:23:23 <ChrisGoldapp> Removing the option still doesn't work, even though QEMU sets up working networking by its own for x86/x64 guests
2023-11-19 01:23:26 <ChrisGoldapp> Ah, thank you
2023-11-19 01:23:27 <tzeentch[m]>  * ```
2023-11-19 01:23:27 <tzeentch[m]> -device virtio-net,netdev=net0 
2023-11-19 01:23:27 <tzeentch[m]> -netdev user
2023-11-19 01:23:29 <ChrisGoldapp> I'll try that
2023-11-19 01:23:35 <tzeentch[m]>  * ```... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/KKQBlKYvdOqqjRuuUoAqakkN>)
2023-11-19 01:30:14 <ChrisGoldapp> Hmm, it wants me to specify an id, and if I do that it can't find the network device net0
2023-11-19 01:30:40 <ChrisGoldapp> I'm looking through my /dev folder for a network device to use
2023-11-19 01:31:35 <tzeentch[m]> ChrisGoldapp: nope
2023-11-19 01:31:48 <tzeentch[m]> id=net0,it used in qemu only
2023-11-19 01:33:36 <WhyNotHugo> pj: The main issue is that neovim doesn't use upstream tree-sitter grammars, but requires custom ones.
2023-11-19 01:36:53 <ChrisGoldapp> tzeentch[m]: So unfortunately this command still gives me an error: https://gist.github.com/ChristianGoldapp/e39c81cd27989ead3942596e7153b26c
2023-11-19 01:37:36 <ChrisGoldapp> Still "IP requested but no network found"
2023-11-19 01:37:50 <WhyNotHugo> I set my battery to only charge up to 85% (to avoid extra wear) by writting to /sys/class/power_supply/BAT0/charge_stop_threshold
2023-11-19 01:37:59 <WhyNotHugo> What's a good approach to apply this after reach reboot?
2023-11-19 01:38:03 <ChrisGoldapp> Funnily enough the most recent gist before I created this was by someone else about QEMU networking
2023-11-19 01:39:30 <ChrisGoldapp> https://gist.github.com/junftnt/3103650874892cafeb14af9de26d020b
2023-11-19 01:41:36 <ChrisGoldapp> Ah, this looks like the right trail: https://wiki.qemu.org/Documentation/Platforms/ARM#Generic_ARM_system_emulation_with_the_virt_machine
2023-11-19 01:42:28 <WhyNotHugo> Sounds like this is a good approach to my question: https://lists.alpinelinux.org/~alpine/devel/%3CCAF-%2BOzABh_NPrTZ2oMFUKrsYmSE5obOadKTAth1HU5_OEZUxPQ%40mail.gmail.com%3E
2023-11-19 02:08:34 <ChrisGoldapp> Okay, so I asked ChatGPT for help and it actually worked
2023-11-19 02:08:47 <ChrisGoldapp> I added the line "  -device virtio-net-device,netdev=net0"
2023-11-19 02:08:55 <ChrisGoldapp> Where earlier it was only "virtio-net"
2023-11-19 04:25:10 <felixphew> hi! has anyone here managed to boot alpine as a Xen PVH dom0? it appears to start to work for me, but hangs partway through boot (right after adding PCI devices)...
2023-11-19 04:25:29 <felixphew> I have logs if anyone's interested
2023-11-19 04:25:36 <felixphew> same config works fine as a PV dom0
2023-11-19 04:34:20 <felixphew> https://dpaste.org/83fmY
2023-11-19 06:29:41 <mjt> what does the "Step 2" of "Prepare" stage in https://wiki.alpinelinux.org/wiki/Bootstrapping_Alpine_Linux supposed to do?
2023-11-19 06:30:34 <mjt> it looks like a left-over from some older version
2023-11-19 06:32:43 <mjt> (this is my first expirence with alpine ever, - it's interesting the very first thing I faced is some error :) )
2023-11-19 12:12:52 <frag> are there good alternaties to pdf for invoice and similar? and/or is it possible to get them super small? seems 12 or 16 KB is the smallest even if its just a bit of text .. using groff
2023-11-19 12:17:20 <felixphew> it's possible to produce really small PDFs, if you're careful how you produce them and how they're encoded... but it becomes exponentially more fiddly the smaller you try and go
2023-11-19 12:17:30 <vkrishn> anyone know of a good guide of does/don't of switch_root during linux boot ?
2023-11-19 12:18:34 <felixphew> frag: what formats count as an alternative depends entirely on who you're sending these to! in some situations, you might be able to get away with a plain text invoice, even. it really depends on the expectations of whoever you're invoicing.
2023-11-19 12:19:02 <felixphew> (but PDF is probably the only thing you can _assume_ _anyone_ will be able to deal with)
2023-11-19 12:19:23 <DuClare> frag have you tried postscript?
2023-11-19 12:20:26 <frag> i guess invoices need to be (a bit) hard to change, has to be static ... dunno if images are even a valid option, but feels unproffesional
2023-11-19 12:20:37 <DuClare> forget hard to change
2023-11-19 12:20:39 <DuClare> There is no such thing
2023-11-19 12:21:03 <frag> DuClare: i think i considered it, not sure why i dropped it as an option, maybe reconsider
2023-11-19 12:21:11 <felixphew> the _illusion_ of being hard to change might be relevant, but I think in most scenarios it won't be
2023-11-19 12:21:19 <felixphew> rather, what people are used to / know how to deal with will
2023-11-19 12:21:33 <frag> DuClare: hehe true, but i guess it just have to be hard enough for "normal" ppl to do ..
2023-11-19 12:21:51 <frag> felixphew: true
2023-11-19 12:22:19 <frag> dunno if pdfs have a date stamp when they are generated(?)
2023-11-19 12:23:23 <frag> postscript is 12KB (with just minimal text)
2023-11-19 12:23:42 <DuClare> I'm pretty sure you can make it much smaller than that
2023-11-19 12:23:48 <felixphew> what's the main concern with filesize? is this just something you care about on principle, or do you have a very slow network connection or smth?
2023-11-19 12:23:51 <DuClare> But it depends on what you want to include and how
2023-11-19 12:24:39 <frag> felixphew: hmm just like to keep it small/tidy/effective :]
2023-11-19 12:24:48 <frag> simple
2023-11-19 12:25:39 <frag> i just do 'groff file > file.ps' and its 12KB
2023-11-19 12:25:54 <frag> i guess ppl cant view ps files easily
2023-11-19 12:29:21 <felixphew> if you're going ps -> pdf with ghostscript, have a look at some of the options it has to compress and subset fonts
2023-11-19 12:29:35 <frag> hmm, maybe i included more than i though, with _only_ clean text i can get it down to 8KB
2023-11-19 12:30:17 <felixphew> if you're using roff, actually, you might be using "base 14" fonts, which _technically_ means you can avoid embedding fonts at all
2023-11-19 12:30:35 <felixphew> as any PDF implementation is supposed to have some version of those 14 fonts built-in
2023-11-19 12:34:40 <frag> i use groff :]
2023-11-19 12:35:12 <frag> not sure what font im using, not setting any
2023-11-19 12:42:25 <felixphew> most default groff macro packages will use base fonts (iirc, usually helvetica and/or times)
2023-11-19 12:45:12 <frag> i use times roman/times bold 'pdffonts' tell me .. 'pdfinfo' shows creation/modification dates also (prob can be manipulated but yeah)
2023-11-19 12:45:40 <frag> i dont use any macro packages
2023-11-19 12:50:28 <frag> seems pdf1.3 can make the pdf a bit smaller :S
2023-11-19 14:27:29 <invoked> frag: just curious, what's your constraint, and why?
2023-11-19 14:28:14 <invoked> are you trying to get something below 10k?
2023-11-19 14:28:48 <frag> invoked: just as small as possible, could potentially be a lot of files, and in general try to keep it small
2023-11-19 14:29:40 <invoked> for AP, i'd suggest sticking with pdf otherwise it'd probably be more trouble than it's worth
2023-11-19 14:29:53 <invoked> just because of what the other side expects
2023-11-19 14:30:39 <invoked> AR i mean.
2023-11-19 14:30:50 <invoked> 
ACTION gets coffee

2023-11-19 14:31:51 <invoked> your problem is interesting, because storage has gotten so cheap. you must be generating a ridiculous number of invoices.
2023-11-19 14:32:37 <frag> mhmhm//   ar?
2023-11-19 14:33:30 <invoked> accounts receivable
2023-11-19 14:37:06 <frag> aha
2023-11-19 18:24:52 <fluix> Migadu (email provider) sends html files for invoices
2023-11-19 18:27:28 <frag> the one line invoice
2023-11-19 23:18:18 <jonesv> Is there someone here using zfs-on-root?
2023-11-19 23:19:34 <dwfreed> I'm sure *somebody* is; ask your real question
2023-11-19 23:20:22 <felixphew> jonesv: I am?
2023-11-19 23:20:40 <jonesv> I followed this: https://openzfs.github.io/openzfs-docs/Getting%20Started/Alpine%20Linux/Root%20on%20ZFS.html
2023-11-19 23:20:48 <jonesv> But upon reboot I am now stuck in an initramfs emergency recovery shell that says: "cannot import 'rpool': no such pool available
2023-11-19 23:21:06 <jonesv> I haven't seen any weird warning or error during the tutorial, though
2023-11-19 23:34:19 <jonesv> felixphew: any idea what I may be missing?
2023-11-19 23:35:04 <felixphew> from the initramfs shell, what's the output of `zfs import`?
2023-11-19 23:35:13 <felixphew> sorry, `zpool import`
2023-11-19 23:37:56 <jonesv> felixphew: "no pools available to import"
2023-11-19 23:38:26 <felixphew> ok. can you see the devices that you created the pool on? are they still there?
2023-11-19 23:39:17 <jonesv> felixphew: you mean the /dev/disk/by-id/<name>?
2023-11-19 23:39:29 <felixphew> sure, if that's what you used
2023-11-19 23:40:01 <jonesv> felixphew: yeah that's what they do on that link above. But `/dev/disk` does not exist from the initramfs
2023-11-19 23:53:54 <felixphew> hmm... that aside, surely the pool should show up as importable, even if its devices are only present under the wrong names?
2023-11-19 23:56:25 <jonesv> felixphew: you mean that `zpool import` should not say "no pools available to import"? xD
2023-11-19 23:57:15 <felixphew> yeah. (my own zfs setup is very different from yours, though, so it's hard to test)
2023-11-19 23:57:25 <felixphew> hang on... you're sure the zfs modules are loaded, right?
2023-11-19 23:57:49 <felixphew> make sure they show up in `lsmod`
2023-11-19 23:58:30 <jonesv> Let me reboot, I was just checking the BIOS to see if secure boot is disabled, but that was a complete guess
2023-11-19 23:59:05 <jonesv> (it is disabled)
2023-11-19 23:59:32 <jonesv> felixphew: "zfs" shows up in `lsmod`
2023-11-19 23:59:58 <jonesv> And a bunch of modules are "Used by zfs"
2023-11-20 00:00:06 <felixphew> well, I'm out of ideas. it's probably something relatively straightforward but I didn't get a lot of sleep :/
2023-11-20 00:00:13 <jonesv> like zzstd, zlua, etc
2023-11-20 00:00:43 <jonesv> felixphew: could it be that I screwed up and chose the wrong disk in /dev/disk/by-id?
2023-11-20 00:00:55 <felixphew> um... maybe
2023-11-20 00:01:12 <felixphew> have a look at the disks that you think the pool should be on using `fdisk`
2023-11-20 00:01:25 <jonesv> There were two, one that started with like NVM- and that looked like a name, and one that looked like a hash. But it felt like those were the same (with a different name) somehow
2023-11-20 00:01:46 <felixphew> it's common for things to have two IDs, yeah
2023-11-20 00:02:08 <jonesv> not sure how to run fdisk... like I don't have a /dev/disk to run it on... ^^
2023-11-20 00:03:02 <felixphew> ok; so you're not sure what the linux names of the disks you used are
2023-11-20 00:03:26 <felixphew> I don't think lsblk is in the initrd, unfortunately
2023-11-20 00:03:37 <jonesv> felixphew: they were very long names. So I took the nvme one, but I don't remember the full name
2023-11-20 00:03:50 <felixphew> oh, the id name would be long
2023-11-20 00:04:00 <felixphew> but that name is going to be a symlink to the "normal" linux device
2023-11-20 00:04:21 <felixphew> if it was an nvme disk, it'll be something like /dev/nvmeXnY
2023-11-20 00:04:26 <jonesv> I don't see much in /dev actually. Like no sda or nvme or anything that I would recognize as a disk
2023-11-20 00:04:41 <jonesv> there are /dev/lopp
2023-11-20 00:04:53 <jonesv> there are /dev/loopX*
2023-11-20 00:05:19 <jonesv> some /dev/ramX, /dev/ttyX, dev/vcsX (not sure what vcs is)
2023-11-20 00:06:02 <felixphew> vcs-es are the other half of the tty devices; don't worry about them, they're not relevant here
2023-11-20 00:06:39 <jonesv> Right
2023-11-20 00:06:43 <felixphew> specifically the virtual consoles, I mean
2023-11-20 00:06:50 <jonesv> Yeah it feels like the disks are not there in /dev
2023-11-20 00:07:04 <jonesv> Maybe something wrong with the grub patches I had to apply?
2023-11-20 00:07:13 <felixphew> ok, so the disks not being there is the key issue here
2023-11-20 00:07:17 <felixphew> that's why zfs can't see it
2023-11-20 00:07:54 <jonesv> oh, `dmesg` actually ends with "Mounting root: failed"
2023-11-20 00:08:06 <felixphew> as for why they're not there, idk; maybe you're missing the nvme drivers? did you add nvme to mkinitfs.conf?
2023-11-20 00:09:05 <felixphew> I don't think mkinitfs will add nvme drivers by itself (iirc it doesn't do any of the "magic" that e.g. dracut does where it tries to figure out for you what drivers are needed to mount your root)
2023-11-20 00:09:30 <jonesv> I did not. Can I do that from the initramfs shell?
2023-11-20 00:09:49 <felixphew> no, because that's on your root filesystem, which is on a drive that your computer can't talk to
2023-11-20 00:09:55 <felixphew> because it doesn't have drivers :-P
2023-11-20 00:09:59 <jonesv> hahaha right
2023-11-20 00:10:08 <felixphew> you'll need to boot into the live environment again
2023-11-20 00:10:16 <jonesv> So would I do that at the very end, while in the chroot?
2023-11-20 00:10:40 <felixphew> yep! then rebuild the initramfs
2023-11-20 00:10:44 <felixphew> https://wiki.alpinelinux.org/wiki/NVME
2023-11-20 00:11:11 <jonesv> ok let me try this
2023-11-20 00:16:52 <jonesv> felixphew: those are the "two" nvme disks I see: /dev/disk/by-id/nvme-WDC_PC_SN720_SDAQNTW-1T00-1001_192901800127 and /dev/disk/by-id/nvme-eui.e8238fa6bf530001001b448b440ad4cb. I assumed they were the same
2023-11-20 00:17:13 <felixphew> ls -l /dev/disk/by-id
2023-11-20 00:17:36 <felixphew> should show you where the symlinks point
2023-11-20 00:17:45 <felixphew> if they point to the same place, then they're the same :3
2023-11-20 00:18:02 <jonesv> They do :)
2023-11-20 00:18:13 <felixphew> probably /dev/nvme0n1, right?
2023-11-20 00:18:20 <jonesv> Also I get a "Warning: The resulting partition is not properly aligned for best performance: 1989923504s % 2048s != 0s", I hope it's not a big problem
2023-11-20 00:18:38 <jonesv> probably /dev/nvme0n1, right? -> yes!
2023-11-20 00:19:05 <jonesv> (the warning comes from step 1 in https://openzfs.github.io/openzfs-docs/Getting%20Started/Alpine%20Linux/Root%20on%20ZFS.html#system-installation)
2023-11-20 00:19:49 <felixphew> jonesv: as long as it's just the _end_ of the last partition that's misaligned, it shouldn't be a big problem, at last as far as I know
2023-11-20 00:20:04 <jonesv> got it, that's what I was hoping :)
2023-11-20 00:23:09 <jonesv> Do I need to reinstall grub or something after I run mkinitfs?
2023-11-20 00:27:11 <felixphew> embarrasingly, I don't know; I've avoided putting grub on anything for years now and I forget how it works
2023-11-20 00:27:14 <felixphew> I _think_ not
2023-11-20 00:27:33 <felixphew> because it should generate the new initramfs in place of the old one
2023-11-20 00:30:11 <jonesv> What do you use instead of grub?
2023-11-20 00:31:28 <jonesv> Also do you use dracut instead of mkinit?
2023-11-20 00:34:11 <jonesv> felixphew: oh it worked! The nvme trick! Now it boots :). Thank you so much!
2023-11-20 00:36:37 <jonesv> If I may still ask: right before the login ("Enter passphrase for ..."), it says: "cannot import '2341241234214322134': a pool with that name already exists, and "use the form 'zpool import <pool | id> <newpool>' to give it a new name. It says that for 3 pools (two with long numbers and one called "rpool"). Do you know why it could be?
2023-11-20 01:07:34 <felixphew> that message is, I think, normal - the init scripts are trying to import the pool when it's already been imported
2023-11-20 01:08:11 <felixphew> I do use mkinitrd on alpine, just often with customised scripts and options. I was a fedora user for a long time (and still am on some systems) so I'm a little more familiar with dracut
2023-11-20 01:08:58 <felixphew> what boot setup I use depends on the system. my servers have really nice EFI support, so I just add the kernel as an entry directly with efibootmgr - no bootloader required
2023-11-20 01:09:26 <felixphew> on (non-alpine) desktop and laptops I generally use systemd-boot (formerly gummiboot)
2023-11-20 01:10:00 <felixphew> s/mkinitrd/mkinitfs
2023-11-20 01:19:31 <jonesv> felixphew: right, that's interesting. Thanks again :)
2023-11-20 01:20:07 <jonesv> felixphew: oh and last thing: do you use zfs on a laptop? I think I read that hibernation is not really possible, do you know about that? 
2023-11-20 01:21:00 <felixphew> I did, I don't currently - and hibernation is possible, but to do it most reliably, you want swap _not_ on ZFS
2023-11-20 01:21:07 <felixphew> I never really used hibernation tbh
2023-11-20 01:21:24 <jonesv> felixphew: so you either suspend or poweroff?
2023-11-20 01:21:35 <felixphew> yep - mostly suspend
2023-11-20 01:22:01 <jonesv> Right. Yeah that's what I was planning to do
2023-11-20 01:34:22 <invoked> felixphew: if only uefi implementations were uniform across hardware...
2023-11-20 01:37:00 <invoked> manufacturers take tianocore and then just start removing stuff, so it's hard to know what's actually going to work, except on rack servers, maybe.
2023-11-20 02:00:46 <heat> removing? you're too hopeful :)
2023-11-20 02:04:50 <invoked> i'd be fine with it if these gimped efis worked right. i've seen all kinds of weird, broken stuff... just so they can make room for useless crap
2023-11-20 02:05:44 <invoked> i get the feeling they don't know what they're doing, and they don't test. except on servers.
2023-11-20 02:07:08 <heat> i'm yet to see vendor EFI that isnt either full of "value-add" (garbage) or tianocore but NIH'd ("lets take it all, but change it in small ways everywhere so we'll take 10 years to sync with upstream!")
2023-11-20 02:07:51 <invoked> right.
2023-11-20 02:07:53 <heat> AMI aptio even has a custom build system for who-knows-why
2023-11-20 04:05:04 <Saijin_Naib[m]> I'm really hoping to get hands on a StarLabs and see what their coreboot is like! Seems like they are good OSS citizens, so I'm hopeful 
2023-11-20 05:04:16 <invoked> yeah, although coreboot on amd is still new-ish. but i understand that on those you can switch between coreboot and aptio whenever you want.
2023-11-20 10:02:25 <anon_> hey
2023-11-20 11:29:21 <Apachez> any of you remembers what the dongles are named to make sure that a vga output actually outputs vga after a reboot since many boxes will disable vga output if none is detected during boot?
2023-11-20 11:31:42 <Apachez> nailed it, like random order of words: Dummy Plug VGA Display Emulator Headless
2023-11-20 12:00:03 <lopid> who you calling headless
2023-11-20 12:18:13 <DuClare> dummy.
2023-11-20 15:15:35 <jonesv> Should `doasedit` come with the `doas` package? Doesn't seem so, for me :/ 
2023-11-20 15:17:38 <minimal> no, it comes with the doasedit package
2023-11-20 15:18:54 <jonesv> minimal: uhoh, so I'm missing something: `apk search doasedit` does not see it
2023-11-20 15:19:26 <minimal> which Alpine version are you on? doasedit is in testing repo which is Edge only
2023-11-20 15:20:06 <minimal> you do know you can use https://pkgs.alpinelinux.org to check packages/package contents?
2023-11-20 15:21:50 <jonesv> minimal: I downloaded v3.18-extended and installed from there. Is Edge similar to Arch Linux (rolling releases)?
2023-11-20 15:22:22 <minimal> Edge is not a release, its the development that becomes the next release
2023-11-20 15:22:38 <minimal> however anything in testing repo will NOT be part of any release
2023-11-20 15:23:25 <jonesv> So if I want doasedit, I need to both add the testing repo and be on Edge (because the testing repo is only on Edge)?
2023-11-20 15:23:29 <minimal> so 3.19 is currently being prepared based on Edge but 3.19 will NOT contain doasedit (unless someone moves doasedit from testing to either main or community before then)
2023-11-20 15:24:19 <minimal> jonesv: testing is only supported if you're running Edge
2023-11-20 15:24:47 <jonesv> minimal: and is it common to use Edge for a personal laptop? I'm tempted to just try that
2023-11-20 15:25:19 <minimal> as Edge is for in-development activity there is always a risk that Edge could break at any point in time
2023-11-20 15:26:57 <jonesv> right. Well for now I'll stay out of edge, and I'll just `doas vim` instaed of `doasedit` :)
2023-11-20 15:26:58 <jonesv> Thanks!
2023-11-20 15:32:31 <quinq> Though using edge shouldn't be a problem in most cases
2023-11-20 15:34:49 <jonesv> quinq: somehow the wiki saying that I could lose data is a bit scary xD. Also I'm using ZFS, I don't know how risky that is
2023-11-20 15:35:12 <invoked> you probably should have backups, either way.
2023-11-20 15:35:49 <invoked> i mean, if you're concerned about losing data...
2023-11-20 15:37:37 <jonesv> invoked: yeah that too
2023-11-20 15:37:39 <jonesv> Are you all on edge?
2023-11-20 15:38:00 <invoked> i'm not but that's because i don't need anything from edge.
2023-11-20 16:01:11 <minimal> jonesv: no, I'm quite calm ;-)
2023-11-20 16:03:25 <raspbeguy> jonesv, well, in my opinion Alpine Edge fits desktop use as much as Debian Unstable
2023-11-20 17:28:10 <Cyanatide> Hi! Doc https://wiki.alpinelinux.org/wiki/Custom_network_interface_names states that I should be able to change naming of nic in /etc/mactab, however, it seems that this documentation in outdated... What would be the best way to rename interface?
2023-11-20 17:28:49 <Cyanatide> in fact, I just want to change the eth order in order to have physical order coherent with logical order
2023-11-20 17:33:22 <lopid> you want to use predictable nic names?
2023-11-20 17:33:50 <lopid> or actually name them yourself?
2023-11-20 17:34:28 <Cyanatide> name them by myself
2023-11-20 17:35:15 <lopid> that might not be specific to alpine
2023-11-20 17:35:51 <lopid> udev, apparently: https://search.brave.com/search?q=linux+change+ethernet+interface+name
2023-11-20 17:38:28 <Cyanatide> But file /etc/udev/rules.d/70-persistent-net.rules doesn't exist. I should have somewhere a file defining that eth1 correspond to mac xx:xx:xx:xx:xx:xx, don't you think?
2023-11-20 17:39:14 <lopid> i think the kernel will pick a name already
2023-11-20 17:39:43 <minimal> Cyanatide: yes, and that also assumes you're using eudev rather than mdev
2023-11-20 17:39:59 <minimal> busybox has nameif which reads /etc/mactab
2023-11-20 17:41:04 <dari> Cyanatide: I'm using udev rules with eudev
2023-11-20 17:41:35 <dari> Cyanatide: SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="xx:xx:xx:xx:xx:xx", ATTR{type}=="1", KERNEL=="eth*", NAME="eth30"
2023-11-20 17:42:54 <dari> you can change eth30 to for example lan0 net10 etc etc
2023-11-20 17:42:55 <Cyanatide> Ok, I'm using alpine 3.16 which use by default mdev. What would be the proper solution to manually set the interface naming with mdev?
2023-11-20 17:43:27 <minimal> via /etc/mactab as you mentioned?
2023-11-20 17:45:28 <Cyanatide> ok I will try, in fact i was afraid that creating a file mactab would create a conflict with another file defining that eth1 correspond to mac xx:xx:xx:xx:xx:xx
2023-11-20 17:50:36 <minimal> you'll have to add a line to /etc/mdev.conf to call nameif (like that article shows)
2023-11-20 17:54:15 <Cyanatide> I see, indeed it seems to work like that.
2023-11-20 17:59:57 <Cyanatide> Ok, thanks for your help then! It works, and if it's the right process to rename interface, then I im happy :)
2023-11-20 18:00:42 <minimal> it would probably make sense for that line to be added to the packaged mdev.conf
2023-11-20 18:03:26 <mjt> alpine doesn't use udev, does it?
2023-11-20 18:10:21 <Cyanatide> @minimal, I agree
2023-11-20 18:10:49 <Cyanatide> @mjt, not by default on alpine 3.16 at least. Maybe it can be configured
2023-11-20 18:11:42 <mjt> there's eudevd, whatever it is
2023-11-20 18:28:29 <dari> script "setup-devd" doing the job
2023-11-20 18:30:25 <dari> https://gitlab.alpinelinux.org/alpine/alpine-conf/-/blob/master/setup-devd.in
2023-11-20 18:43:04 <minimal> mjt: Alpine can use eudev (which is basically udev)
2023-11-20 18:43:47 <minimal> however the initramfs (created by mkinitfs) still uses mdev even if you enable eudev (or mdevd) in the OS
2023-11-20 18:44:12 <mjt> oh. they created mdevd too? :)
2023-11-20 18:44:36 <minimal> mjt: who is "they"?
2023-11-20 18:44:38 <mjt> hmm. I thought it's busybox.
2023-11-20 18:44:46 <mjt> already tried to find it in there.. nope :)
2023-11-20 18:45:24 <minimal> mjt: Busybox's mdev daemon is disabled in Alpine
2023-11-20 18:45:54 <minimal> https://git.alpinelinux.org/aports/tree/main/busybox/busyboxconfig#n660
2023-11-20 18:46:32 <mjt> that's what I was looking for, - I didn't know there's a daemon in busybox.
2023-11-20 18:46:54 <mjt> fun stuff
2023-11-20 18:46:55 <minimal> there isn't in Alpine, as it's disabled ;-)
2023-11-20 18:47:18 <mjt> it's been so long ago since I messed up with this
2023-11-20 18:47:30 <minimal> I don't rmeember if booster or dracut (as alternative initramfs') can make use of eudev instead of mdev
2023-11-20 18:49:31 <mjt> I use my own home-grown initrd which predates dracut for good amount. It used to use mdev, but some years ago I switched to udev
2023-11-20 18:56:08 <sertonix> Can anybody tell me how 'rc_add -k' translates to rc-update?
2023-11-20 19:01:56 <ikke> what does rc_add -k do?
2023-11-20 19:04:09 <sertonix> It was the way to enable an openrc service before rc-update. I can't find documentation on it thought. I am not sure thought what the -k option does
2023-11-20 19:05:50 <minimal> why does it matter if it is no longer present in OpenRC (and therefore in Alpine)?
2023-11-20 19:06:28 <sertonix> It is still in the wiki. I wanted to update it
2023-11-20 19:06:48 <ikke> What wiki?
2023-11-20 19:07:22 <sertonix> wiki.alpinelinux.org
2023-11-20 19:07:57 <ikke> I mean, what page
2023-11-20 19:08:17 <quinq> Yes jonesv, running edge here, no problem yet (using encrypted zfs root)
2023-11-20 19:08:28 <sertonix> For example: https://wiki.alpinelinux.org/wiki/Protecting_your_email_server_with_Alpine
2023-11-20 19:08:52 <ikke> "we can start it at boot"
2023-11-20 19:08:56 <ikke> rc-update add postfix
2023-11-20 19:09:56 <sertonix> Thought it could also be 'rc-update add postfix boot'. But might not be that important
2023-11-20 19:10:41 <ikke> The default runlevel is sufficient for most things
2023-11-20 19:11:10 <ikke> postfix has nothing to do with booting the machine, so it does not need to be in the boot runlevel
2023-11-20 19:12:34 <ikke> sysinit > boot > default
2023-11-20 19:12:45 <ikke> https://github.com/OpenRC/openrc/blob/master/user-guide.md#runlevels
2023-11-20 19:21:01 <scorpion2185[m]> is it possible that Alpine will have dinit one day?
2023-11-20 19:22:00 <sertonix> I think it shouldn't be too hard to port it but nobody has done it yet
2023-11-20 19:24:03 <scorpion2185[m]> it has also user services with a separate instance
2023-11-20 19:25:26 <ikke> Apparently https://davmac.org/projects/dinit/alpine-demo/ exists
2023-11-20 19:26:09 <sertonix> Yes but I haven't found sources to that
2023-11-20 19:26:22 <sertonix> I will try to package dinit when I have time
2023-11-20 19:26:26 <scorpion2185[m]> dinit 10 alpine 3.13
2023-11-20 19:26:32 <scorpion2185[m]> *3.12
2023-11-20 19:54:29 <Saijin_Naib[m]> invoked yeah, i am after their Intel kit, which seems mature. Probably StarLite MKIV
2023-11-20 19:57:15 <Saijin_Naib[m]> jonesv i use it, and it is totally fine minus some fuss leading up to tagging a new release, which basically means don't obsessively update multiple times a day and you won't even notice
2023-11-20 20:00:22 <sertonix> scorpion2185[m]: here you go: https://gitlab.alpinelinux.org/alpine/aports/-/merge_requests/55804
2023-11-20 20:10:00 <Ermine> minimal: CONFIG_FEATURE_MDEV_DAEMON=y .Whoops.
2023-11-20 20:23:41 <scorpion2185[m]> nice! changing init is not easy I did it once on some distro and it was a mess. user services can be tested easily
2023-11-20 20:24:08 <scorpion2185[m]> unless you have a very small instllation and you need service files
2023-11-20 20:25:24 <minimal> Ermine: doh! I misread that somehow, anyway I guess it should be disabled as it's not being used
2023-11-20 20:27:58 <sertonix> scorpion2185[m]: Since I am currently setting up s6 I think I might be able to also get a dinit system running
2023-11-20 22:18:31 <felixphew> I'm running an alpine box with s6-linux-init + s6-rc, and yeah, it's fiddly to set up, and lots of common tasks are very manual unless you write your own scripts for them
2023-11-20 22:19:36 <felixphew> I can't wait for alpine to actually ship with s6-rc (or anything that's less of a mess than openrc, really)
2023-11-20 22:42:23 <DuClare> is openrc a mess?
2023-11-20 22:50:22 <pj> it is
2023-11-20 22:50:32 <pj> but s6 is also a mess
2023-11-20 22:54:40 <DuClare> time to my own init
2023-11-20 22:54:53 <DuClare> just right after my own bootloader and kernel...
2023-11-20 23:02:11 <pj> i mean, we don't even use openrc init
2023-11-20 23:02:18 <pj> it's busybox
2023-11-20 23:04:45 <quinq> openrc-0.51-r0 x86_64 {openrc} (BSD-2-Clause) [installed]
2023-11-20 23:14:11 <vanveldt> there are a huge bunch of inits. But alpine onyl supports openrc, isn't it?
2023-11-20 23:14:50 <minimal> no, as pj pointed out Alpine's openrc does *not* provide the init, that is provided by Busybox
2023-11-20 23:15:11 <minimal> openrc works together with Busybox init
2023-11-20 23:16:00 <vanveldt> minimal: aha!
2023-11-20 23:16:37 <minimal> https://git.alpinelinux.org/aports/tree/main/openrc/APKBUILD#n107
2023-11-20 23:17:34 <quinq> Ah, you mean the init binary
2023-11-20 23:18:36 <minimal> isn't the init binary the init? ;-)
2023-11-20 23:18:43 <minimal> literally....
2023-11-20 23:19:31 <minimal> ps auxww|grep init
2023-11-20 23:19:40 <minimal>     1 root     0:01 /sbin/init
2023-11-20 23:19:43 <quinq> Well, not literally, otherwise it would be “init binary” :)
2023-11-20 23:19:49 <minimal> ls -l /sbin/init
2023-11-20 23:20:11 <minimal>  /sbin/init -> /bin/busybox
2023-11-20 23:20:17 <pj> the init is busybox, but openrc does own stuff to start and manage system
2023-11-20 23:20:17 <quinq> There's an init system
2023-11-20 23:20:30 <minimal> init is PID 1
2023-11-20 23:20:35 <quinq> built around busybox init binary and openrc init files
2023-11-20 23:20:44 <quinq> Yes, the init process
2023-11-20 23:21:33 <minimal> on Unix/Linux the init is PID 1
2023-11-20 23:21:38 <pj> when we talk about init, it's almost always the init process, the actual binary that handles that
2023-11-20 23:21:53 <minimal> which in this case is Busybox
2023-11-20 23:21:56 <quinq> we being you and minimal? :)
2023-11-20 23:22:06 <quinq> Sorry I didn't know
2023-11-20 23:22:31 <pj> anything that doesn't handle init and rc together
2023-11-20 23:23:51 <minimal> https://en.wikipedia.org/wiki/Init
2023-11-21 08:07:36 <ctrlaltdelete> hi
2023-11-21 08:07:45 <ctrlaltdelete> im compiling my own kernel
2023-11-21 08:07:51 <ctrlaltdelete> should i get it from linux.org
2023-11-21 08:07:57 <ctrlaltdelete> or do i have to use alpine kernel ?
2023-11-21 09:41:45 <felixphew> either will ultimately work; it depends what you want to get out of compiling your own kernel
2023-11-21 10:08:43 <Ermine> alpine kernel is basically vanilla
2023-11-21 15:22:51 <jonesv> Should elogind somehow set XDG_RUNTIME_DIR? I installed it as per https://wiki.alpinelinux.org/wiki/Wayland, but I still need to manually set XDG_RUNTIME_DIR in order to run sway
2023-11-21 15:38:28 <jonesv> Oh, maybe I just don't need elogind :)
2023-11-21 19:06:59 <Ermine> Seatd will be enough
2023-11-22 03:01:00 <oksomethingelse> my USB device shows up in dmesg, but I cant find it using fidks -l nor lsblk. anyone have ideas about why?
2023-11-22 03:24:20 <oksomethingelse> tried with another usb drive, and it worked fine. 
2023-11-22 03:24:29 <oksomethingelse> shows up with lsblk
2023-11-22 03:24:39 <oksomethingelse> the drive is detected on another OS
2023-11-22 03:25:45 <ssm> does it have some weird partition format or filesystem that fdisk and lsblk don't recognize maybe? does dmesg show the disk letter?
2023-11-22 04:21:02 <oksomethingelse> maybe. it's a portable backup drive: https://www.amazon.com/dp/B0B787PNL7  
2023-11-22 04:21:07 <oksomethingelse> it claims to be linux compatible
2023-11-22 04:22:14 <oksomethingelse> ev/disk4 (external, physical):
2023-11-22 04:22:17 <oksomethingelse>    #:                       TYPE NAME                    SIZE       IDENTIFIER
2023-11-22 04:22:19 <oksomethingelse>    0:      GUID_partition_scheme                        *4.0 TB     disk4
2023-11-22 04:22:21 <oksomethingelse>    1:                        EFI EFI                     209.7 MB   disk4s1
2023-11-22 04:22:23 <oksomethingelse>    2:       Microsoft Basic Data slime                   4.0 TB     disk4s2
2023-11-22 04:22:25 <oksomethingelse> when mounted on macos
2023-11-22 04:23:05 <oksomethingelse> but this is pretty standard filesystem and format 
2023-11-22 04:34:08 <quinq> oksomethingelse, what's interesting is what dmesg tells you when you plug the USB in
2023-11-22 04:54:00 <oksomethingelse> new high-speed USB device number 8 using ehci-pci
2023-11-22 04:54:03 <oksomethingelse> New USB device found, idVendor=0634, idProduct=5602, bcdDevice= 1.10
2023-11-22 04:54:05 <oksomethingelse> New USB device strings: Mfr=1, Product=2, SerialNumber=3
2023-11-22 04:54:07 <oksomethingelse> Product: CT4000X8SSD9
2023-11-22 04:54:09 <oksomethingelse> Manufacturer: Micron
2023-11-22 04:54:11 <oksomethingelse> SerialNumber: 2336E6F7EDBA
2023-11-22 04:54:13 <oksomethingelse> yeah, this is what I get
2023-11-22 04:56:16 <oksomethingelse> so it shows up as a USB device, just not a storage device
2023-11-22 05:10:05 <Hello71> sounds like you updated kernel and didn't reboot
2023-11-22 05:14:06 <oksomethingelse> that is indeed something that has happened
2023-11-22 05:14:19 <oksomethingelse> but why does it detect my other USB drive?
2023-11-22 05:14:36 <oksomethingelse> [/t/r/m/tmp]$  uptime
2023-11-22 05:14:39 <oksomethingelse>  21:14:27  up 279 days  9:02,  0 users,  load average: 2.94, 2.56, 2.45
2023-11-22 05:14:41 <oksomethingelse> wheee
2023-11-22 05:22:04 <oksomethingelse> I have a smaller USB drive that detects just fine
2023-11-22 05:22:18 <oksomethingelse> so wouldn't the lack of rebooting play a part here too?
2023-11-22 05:25:05 <oksomethingelse> i suppose I'll reboot anyway. it's time
2023-11-22 05:25:15 <oksomethingelse> I'll be back if it's not solved
2023-11-22 05:33:33 <eydaimon> Hello71: rebooting did work. thanks. 
2023-11-22 06:52:49 <Justin[m]> Is there any discussion or consideration behind a longer support cycle for Alpine. 2 years is good, but if it went to 5 years on every x release that would be great!
2023-11-22 06:59:56 <fluix> it's already 3 years on request, 5 years is quite a bit of time
2023-11-22 07:01:21 <fluix> but I'm also young haha
2023-11-22 07:17:27 <hyau[m]> Like theoretically it's possible, cause the lts kernel is really lts like to this day 4.14.330-lts is supported which is around 5 ish years old even more
2023-11-22 07:19:05 <hyau[m]> But it's not like this because it would be hell to maintain old versions for the Alpine Devs for more than 2-3 years. And server Devs are more likely to update to newest main in 2~ish years time to get the newest security patches
2023-11-22 07:20:47 <ikke> Yes, and the lts kernel support will be reduced as well 
2023-11-22 07:21:36 <ikke> And trying to backport fixes is not without risk either 
2023-11-22 07:22:10 <hyau[m]> Yeah also that this could also bring a packaging issue, keeping the edge channel latest-stable and a previous-stable and a even older-stable would probably bring some problems
2023-11-22 07:23:38 <hyau[m]> Mainly dumb people accidentally moving theirs latest stable to the oldest stable ever
2023-11-22 07:24:45 <hyau[m]> If you want something that has around 5~ years support use the previous debian releases those are as stable as possible
2023-11-22 07:26:08 <hyau[m]> But I also think there is a distro that has even longer support I don't remember now what was it called.
2023-11-22 07:26:39 <hyau[m]> Slackware also has a long support window
2023-11-22 07:32:04 <hyau[m]> Also I have question is alpaquita linux the only Alpine linux based distro? Or are there more?
2023-11-22 07:40:14 <ikke> PostmarketOS
2023-11-22 07:41:55 <quinq> 06:33:33  eydaimon$ Hello71: rebooting did work. thanks.
2023-11-22 07:42:03 <quinq> Classic unsolved problem
2023-11-22 08:36:39 <hyau[m]> Any thoughts on LURE? Haven't tried it myself, but from what I saw it's pretty good.
2023-11-22 11:26:56 <jonesv> What should I install to get the sound working on Alpine? I am playing a video on YouTube and have no sound. I tried to install pipewire, alsa, pulseaudio, but it seems like I am missing some configuration...
2023-11-22 11:31:38 <Ermine> check out pipewire page on wiki
2023-11-22 11:32:01 <jonesv> I did, also the alsa and pulseaudio, also on the Arch wiki and Gentoo
2023-11-22 11:33:43 <jonesv> I just really don't know much about how audio works on Linux (seems like pipewire is a "frontend" and alsa/pulseaudio are "backends"? Not sure what e.g. Firefox uses. Also it seems like dbus is involved, but the big picture is still a bit unclear to me)
2023-11-22 11:34:58 <Ermine> Did you follow the instructions? What issues did you get?
2023-11-22 11:36:25 <pj> alsa talks to sound cards, pulseaudio/jack/pipewire are sound servers that use alsa and handle various stuff like mixing etc.
2023-11-22 11:39:02 <jonesv> Ermine: I installed a bunch of stuff (pipewire, wireplumber, pipewire-pulse, pipewire-jack, pipewire-alsa, dbus-openrc, alsa-utils, alsa-conf, pulseaudio, pulseaudio-alsa), then I tried to run /usr/libexec/pipewire-launcher but I get a bunch of errors and still no sound in Firefox
2023-11-22 11:39:22 <jonesv> Ermine: when I run alsamixer, I don't see any sound card there (not sure if I should)
2023-11-22 11:40:39 <jonesv> pj: oh right, and so apps talk to one of the sound servers (either pulseaudio or jack or pipewire) who then send the sound to alsa?
2023-11-22 11:50:00 <Ermine> jonesv: which errors do you get?
2023-11-22 12:07:50 <jonesv> Ermine: whether dbus is running or not, I get "pw_rtkit_bus_get()] Failed to connect to session bus: /usr/bin/dbus-launch terminated abnormally without any error message"
2023-11-22 12:08:53 <jonesv> Ermine: and some "pipewire__module_init()] could not get system bus: Connection refused". Wait, let me find a pastebin
2023-11-22 12:09:30 <Ermine> 'dbus running or not' - you mean system dbus or session dbus? The first one is ran with rc-service dbus start
2023-11-22 12:13:06 <jonesv> Ermine: the first one (I ran `rc-service dbus start`), I'm not really clear what the second one is
2023-11-22 12:14:23 <jonesv> Ermine: here is the error: https://paste.sh/S0H6c3CS#Mn7T9TYQyr0P0-Xo8aWXi-th
2023-11-22 12:17:25 <Ermine> what happens if you run dbus-launch yourself?
2023-11-22 12:18:20 <jonesv> Ermine: hmm "command not found: dbus-launch"
2023-11-22 12:18:45 <Ermine> I think you need dbus-x11 package
2023-11-22 12:18:45 <jonesv> Ermine: I am running Wayland, I only see a package called "dbus-x11" which does not seem right
2023-11-22 12:18:59 <Ermine> it is actually right, despite the name
2023-11-22 12:19:02 <jonesv> oh
2023-11-22 12:19:45 <Ermine> I don't know who introduced that package, but renaming it now is not an option
2023-11-22 12:20:49 <jonesv> Ermine: got it. It goes further I think, now it complains about xdg-desktop-portal
2023-11-22 12:21:20 <jonesv> Ermine: https://paste.sh/S0AUvUw8#18nT-cNIBxu2wQpvvKkh2vmw
2023-11-22 12:21:43 <jonesv> Ermine: which is a package I don't have, let me try to install it
2023-11-22 12:22:21 <jonesv> Ermine: or on Wayland should I go for `xdg-desktop-portal-wlr`?
2023-11-22 12:22:57 <Ermine> depends on your DE/compositor
2023-11-22 12:24:09 <Ermine> There's x-d-p package for everything, and -gnome, -kde and -wlr packages for GNOME, KDE and wlroots-based compositors respectively
2023-11-22 12:24:44 <Ermine> (and -gtk, idk why it's needed, but you may install it as well)
2023-11-22 12:26:48 <jonesv> Ermine: I am on Sway, I think I want the -wlr
2023-11-22 12:26:58 <Ermine> Yes
2023-11-22 12:27:13 <jonesv> Now I need to understand how to "start" the backend (doesn't seem like a service)
2023-11-22 12:27:23 <Ermine> You don't need
2023-11-22 12:27:47 <Ermine> Once you have pipewire and wireplumber running, sound will work
2023-11-22 12:29:21 <jonesv> Ermine: pipewire-launcher still says "Is xdg-desktop-portal running?" like in my second paste above
2023-11-22 12:29:45 <jonesv> Ermine: I just tried to set XDG_CURRENT_DESKTOP=wlr but it doesn't seem to help
2023-11-22 12:31:24 <Ermine> =sway
2023-11-22 12:31:48 <jonesv> oh
2023-11-22 12:31:53 <jonesv> Also I just saw this: https://github.com/emersion/xdg-desktop-portal-wlr#running
2023-11-22 12:32:37 <jonesv> Oh, probably I need `exec dbus-update-activation-environment --systemd WAYLAND_DISPLAY XDG_CURRENT_DESKTOP=sway`
2023-11-22 12:32:53 <jonesv> Except I don't have systemd -_-
2023-11-22 12:32:56 <Ermine> and sway page proposes running sway as dbus-run-session sway
2023-11-22 12:33:05 <Ermine> without --systemd I guess
2023-11-22 12:33:54 <jonesv> Oh let me restart sway with `dbus-run-session-sway`
2023-11-22 12:35:04 <jonesv> Ermine: where did you see that? I was using `exec sway` from https://github.com/swaywm/sway/wiki
2023-11-22 12:36:37 <Ermine> on sway wiki page
2023-11-22 12:37:42 <jonesv> Ermine: another wiki than the one I linked, I presume? I don't find dbus-run* there
2023-11-22 12:37:51 <Ermine> alpine wiki
2023-11-22 12:38:12 <Ermine> https://wiki.alpinelinux.org/wiki/Sway
2023-11-22 12:40:05 <jonesv> Right, yeah `dbus-run-session -- sway`. Still the same "Is xdg-desktop-portal running?" issue though
2023-11-22 12:47:28 <Ermine> try to run xdg-desktop-portal &
2023-11-22 12:49:07 <jonesv> Ermine: that's the thing: `xdg-desktop-portal: command not found`. Same for `xdg-desktop-portal-wlr`
2023-11-22 12:49:21 <grawity> those are usually in lib, not in $PATH
2023-11-22 12:50:04 <Ermine> So they are supposed to be activated?
2023-11-22 12:50:14 <jonesv> There is a /usr/share/xdg-desktop-portal/portals/wlr.portal`, and I read somewhere that `XDG_CURRENT_DESKTOP=<value>` would somehow load `<value>.portal
2023-11-22 12:50:50 <grawity> yeah, they're dbus-activatable on demand
2023-11-22 12:50:55 <jonesv> which is why I tried XDG_CURRENT_DESKTOP=wlr
2023-11-22 12:51:07 <tretinha> jonesv: what I did was (1) installed superd, (2) ran pipewire-pulse, pipewire, wireplumber, xdg-dpw, and xdg-dp through it, (3) exec'd the dbus-update-activation-environment in the first line of sway's config
2023-11-22 12:51:09 <Ermine> grawity: But why then it doesn't run?
2023-11-22 12:51:09 <grawity> that works slightly differently
2023-11-22 12:51:12 <tretinha> superd services reference: https://git.sr.ht/~whynothugo/superd-services/tree
2023-11-22 12:51:22 <grawity> with the old config style, portals/*.portal has UseIn=<DESKTOP>
2023-11-22 12:51:25 <tretinha> I added them to ~/.config/services/enabled/
2023-11-22 12:51:37 <grawity> with the new config style, <DESKTOP>-portals.conf lists which portals to use
2023-11-22 12:51:57 <grawity> so first it'll be looking for wlr-portals.conf, then for any portals/*.portal that have UseIn=wlr
2023-11-22 12:52:21 <grawity> Ermine: based on scrollback I'd say "because the entire dbus session bus is messed up"
2023-11-22 12:53:12 <jonesv> grawity: I just installed dbus and started it with `rc-service dbus start` :/
2023-11-22 12:53:23 <grawity> that's not the session bus!
2023-11-22 12:53:53 <grawity> system services – anything you mangae via `rc-service` – are the opposite of your login session
2023-11-22 12:53:53 <Ermine> grawity: session bus is here because dbus-run-session
2023-11-22 12:54:08 <Ermine> system bus is also here
2023-11-22 12:54:38 <Ermine> maybe it needs rtkit and polkit?
2023-11-22 12:54:40 <jonesv> grawity: I run sway with `exec dbus-run-session --sway`. Isn't that right?
2023-11-22 12:54:58 <grawity> it would be right if you had a space after the '--'
2023-11-22 12:55:09 <Ermine> btw did you set XDG_RUNTIME_DIR?
2023-11-22 12:55:13 <jonesv> grawity: yes, I do have a space, sorry :(
2023-11-22 12:55:34 <jonesv> Ermine: yes, it points to /tmp/1000-runtime-dir
2023-11-22 12:55:55 <Ermine> well, it was dumb question, sway wouldn't run otherwise
2023-11-22 12:56:04 <jonesv> Ermine: I set it manually in ~/.zprofile
2023-11-22 12:56:09 <grawity> does dbus-run-session log anything to its stdout/stderr or to syslog when the portal service is requested?
2023-11-22 12:56:15 <jonesv> Ermine: yeah I needed it to run sway in the first place :)
2023-11-22 12:57:09 <jonesv> grawity: hmm let me learn how to read the syslog first (I'm new to openrc :) )
2023-11-22 12:57:27 <grawity> or, for example, if you run `gdbus introspect -e -d org.freedesktop.portal.Desktop -o /org/freedesktop/portal/desktop`...
2023-11-22 12:58:50 <jonesv> grawity: nothing appears in /var/log/messages when I run /usr/libexec/pipewire-launcher and see the error
2023-11-22 12:59:40 <grawity> what error does pipewire output?
2023-11-22 13:00:23 <Ermine> jonesv: try to install rtkit and polkit and start them. Maybe that will work
2023-11-22 13:00:27 <jonesv> grawity: your command shows: https://paste.sh/G97k2dzO#4QLRJyMFjePMffef3U_2gQPC
2023-11-22 13:00:44 <jonesv> grawity: the error is: https://paste.sh/S0AUvUw8#18nT-cNIBxu2wQpvvKkh2vmw
2023-11-22 13:00:53 <grawity> well, that shows the portal "frontend" working, at least
2023-11-22 13:01:45 <jonesv> grawity: the frontend being xdg-desktop-portal and the backend being xdg-desktop-portal-wlr?
2023-11-22 13:01:47 <grawity> yes
2023-11-22 13:01:49 <jonesv> Ermine: let me try that
2023-11-22 13:01:52 <grawity> but what error does pipewire-launcher output *now*, after you've fixed the dbus session bus?
2023-11-22 13:02:14 <jonesv> grawity: I think it's the same, let me paste it again
2023-11-22 13:02:46 <jonesv> grawity: https://paste.sh/SJW3z3tN#2ms_6lftRU5DOnNqkZ9kebVB
2023-11-22 13:03:16 <grawity> okay
2023-11-22 13:03:23 <jonesv> grawity: oh wait it is not the same
2023-11-22 13:03:48 <grawity> so half of it is asking for rtkit to be running (system service), as that's how it gets realtime cpu priority
2023-11-22 13:03:57 <grawity> technically optional, I suppose
2023-11-22 13:04:15 <jonesv> I can install and add rtkit
2023-11-22 13:04:48 <grawity> the rest, regarding xdg-portal PermissionStore, looks camera-only
2023-11-22 13:04:48 <jonesv> I am confused as to when the message changed. All I did was your `gdbus introspect -e -d org.freedesktop.portal.Desktop -o /org/freedesktop/portal/desktop`, which I don't think should do anything, right?
2023-11-22 13:05:14 <grawity> well, it *might* have started the .p.Desktop frontend on demand, but I don't see why pipewire wouldn't have caused the same
2023-11-22 13:05:26 <grawity> I think it was just that your previous paste was from before you fixed dbus-run-session in general
2023-11-22 13:05:39 <grawity> well, not sure
2023-11-22 13:06:37 <grawity> for the record
2023-11-22 13:06:48 <jonesv> grawity: I don't think so, because I am still in the same sway session (I fixed the dbus-run-session by running `exec dbus-run-session -- sway`, I think)
2023-11-22 13:06:58 <tretinha> just to make sure, did you test your audio again after these last changes? haha always good to check 
2023-11-22 13:07:00 <grawity> if you want to see what services are currently running on the bus, you can do: gdbus call -e -d org.freedesktop.DBus -o /org/freedesktop/DBus -m org.freedesktop.DBus.ListNames | grep -o "'[^:' ]*'" | sort
2023-11-22 13:07:09 <jonesv> tretinha: good point :). Let me check
2023-11-22 13:07:27 <grawity> and if you want to see what services are available for automatic activation, .ListActivatableNames
2023-11-22 13:07:52 <jonesv> tretinha: no sound :(. Also I kind of fear that it is muted somewhere and I wouldn't notice
2023-11-22 13:08:13 <jonesv> grawity: so runnig rtkit does not remove the warnings in the error log (the ones about RTKit)
2023-11-22 13:08:15 <tretinha> do you have pavucontrol or something installed to check?
2023-11-22 13:08:44 <grawity> yeah, pavucontrol, pactl list sinks, wpctl status (wireplumber)
2023-11-22 13:09:12 <grawity> overall it ought to still work even without rtkit and without the portal bits (those are mainly for camera and mic)
2023-11-22 13:09:41 <tretinha> yea i recommend that you use mozillas test page to check if everything else is correct too: https://mozilla.github.io/webrtc-landing/gum_test.html
2023-11-22 13:09:48 <ham5urg> I've heard that Alpinelinux-people are planing to switch to runit. Is this correct?
2023-11-22 13:10:01 <jonesv> tretinha: I have the firefox sink: https://paste.sh/V_yn7wDJ#IKOFacpDi4pjSjLBYDJ8T_hf
2023-11-22 13:10:20 <grawity> "Dummy Output"
2023-11-22 13:10:33 <grawity> so just out of curiosity, do you have r/w rights on /dev/snd/*
2023-11-22 13:11:56 <jonesv> grawity: hmm I have /dev/sdn/timer with crw-rw---- 1 root audio
2023-11-22 13:12:10 <Ermine> ham5urg: no
2023-11-22 13:12:15 <grawity> jonesv: *only* timer?
2023-11-22 13:12:22 <grawity> jonesv: do you have any /dev/snd/control* or /dev/snd/pcm* devices?
2023-11-22 13:12:28 <jonesv> grawity: my user is in groups wheel audio video seat
2023-11-22 13:12:32 <jonesv> grawity: yes only timer
2023-11-22 13:12:54 <grawity> then, uh, your problems start way before pipewire
2023-11-22 13:12:57 <jonesv> grawity: that's all there is in `/dev/snd`
2023-11-22 13:13:10 <jonesv> grawity: I probably forgot to install something :D
2023-11-22 13:13:26 <grawity> missing kernel drivers, or something? is udev running?
2023-11-22 13:13:47 <grawity> ultimately whether you're using pipewire or pulseaudio or "raw alsa", they all access the sound card via /dev/snd/pcm* (which is handled by the kernel half of ALSA)
2023-11-22 13:14:19 <jonesv> grawity: udev is running
2023-11-22 13:14:54 <grawity> what kind of machine is this running on?
2023-11-22 13:15:10 <jonesv> grawity: I installed from the alpine-extended liveusb, following https://openzfs.github.io/openzfs-docs/Getting%20Started/Alpine%20Linux/Root%20on%20ZFS.html#system-installation
2023-11-22 13:15:23 <jonesv> grawity: a Lenovo Thinkpad X1 carbon
2023-11-22 13:15:35 <grawity> that sounds new enough that it probably needs SoF firmware to be installed
2023-11-22 13:15:44 <grawity> https://pkgs.alpinelinux.org/package/edge/community/x86_64/sof-firmware
2023-11-22 13:16:25 <jonesv> grawity: should I install and reboot or something?
2023-11-22 13:16:33 <grawity> yes
2023-11-22 13:16:56 <dari> jonesv: did you use "setup-devd udev" ?
2023-11-22 13:17:45 <jonesv> dari: yes, but I don't know how persistant this is. I did it at least during the installation
2023-11-22 13:18:06 <jonesv> grawity: now I have a bunch of stuff in /dev/snd!
2023-11-22 13:18:13 <jonesv> (thanks to sof-firmware)
2023-11-22 13:18:15 <tretinha> :) test your audio 
2023-11-22 13:18:32 <dari> it replace mdev so very persistant :)
2023-11-22 13:18:48 <jonesv> dari: right :). But it's not a bad thing, right?
2023-11-22 13:19:12 <jonesv> So now I get the "Is xdg-desktop-portal running?" again, let me try grawity's gdbus line
2023-11-22 13:19:18 <dari> imho if gonna use it as desktop then better use eudev
2023-11-22 13:19:52 <jonesv> dari: how can I check what I have now? And is it safe to change?
2023-11-22 13:20:21 <dari> jonesv: if you ran above command then for sure you are using eudev
2023-11-22 13:20:29 <jonesv> tretinha: still the same: no sound on firefox, pactl list sinks says "Dummy Output" for Firefox
2023-11-22 13:21:14 <jonesv> dari: oh, maybe I set it to udev during installation, and then eudev following some tutorial (I remember reading the descriptions and one of them (presumably eudev) sounded like the one for a desktop)
2023-11-22 13:21:49 <dari> udev=eudev
2023-11-22 13:22:10 <jonesv> grawity: so now all the stuff in /dev/snd has permissions `crw-rw---- 1 root audio`. And my user is in the group "audio", so it doesn't look wrong to me
2023-11-22 13:22:22 <jonesv> dari: oh, right :)
2023-11-22 13:22:25 <Ermine> heck this carbon is super expensive
2023-11-22 13:22:30 <dari> jonesv: check what you firefox trying to use, paste it in firefox address bar: about:support#media
2023-11-22 13:22:32 <grawity> jonesv: right so now it should at least detect them in `pactl list sinks`
2023-11-22 13:22:42 <jonesv> Ermine: I get it for 50% through my school :)
2023-11-22 13:22:53 <Ermine> lucky you!
2023-11-22 13:24:14 <jonesv> grawity: hmm looks the same to me: https://paste.sh/NZ6FQ4hY#dwdHbUiP72LyWPpDagG06M0p
2023-11-22 13:24:58 <jonesv> dari: "Audio Backend" says "pulse-rust", then in "Output Devices" I have "Dummy Output"
2023-11-22 13:25:16 <jonesv> Should I maybe reinstall firefox?
2023-11-22 13:25:25 <Ermine> no need
2023-11-22 13:25:34 <dari> jonesv: so firefox is fine, looking for default device which in your case is dummy..
2023-11-22 13:26:39 <jonesv> oh so one thing that changed (maybe with sof-firmware) is that in `alsamixer`, now I get a "Master" with some volume level. And in the sound cards there is a "default:0 sof-hda-dsp"
2023-11-22 13:28:48 <jonesv> Can I ignore this? "Failed to open module /usr/lib/wireplumber-0.4/libwireplumber-module-logind: Error loading shared library /usr/lib/wireplumber-0.4/libwireplumber-module-logind.so: No such file or directory"
2023-11-22 13:29:00 <jonesv> (still in https://paste.sh/SJW3z3tN#2ms_6lftRU5DOnNqkZ9kebVB)
2023-11-22 13:30:19 <Ermine> this can be ignored
2023-11-22 13:32:29 <jonesv> So everything looks fine, except that Firefox doesn't pick up the right sound card?
2023-11-22 13:33:47 <jonesv> oh! Killed and restarted firefox and now it works \o/!!!
2023-11-22 13:34:34 <Ermine> \o/
2023-11-22 13:35:00 <jonesv> Let me reboot and try again :D
2023-11-22 13:36:59 <grawity> I think the logind module in WP tells it when to "release" devices if you switch to another user's tty
2023-11-22 13:37:10 <grawity> at least that's what logind integration did in pulseaudio
2023-11-22 13:37:18 <grawity> so it should work fine without
2023-11-22 13:37:35 <jonesv> Ok for some reason I don't need grawity's magic gdbus introspect after a reboot, I just need to manually run pipewire-launcher
2023-11-22 13:38:08 <grawity> yeah pipewire ought to be able to trigger autostart in the same way
2023-11-22 13:38:21 <grawity> as there's nothing actually magic in the gdbus command
2023-11-22 13:38:33 <jonesv> grawity: right
2023-11-22 13:38:51 <jonesv> What's the preferred way to have pipewire-launcher start automatically now? Seems like it's not a service
2023-11-22 13:39:06 <grawity> it's a user-level thing, so definitely not rc-service
2023-11-22 13:39:22 <grawity> maybe just run it from your sway startup script?
2023-11-22 13:39:27 <jonesv> grawity: ~/.zprofile?
2023-11-22 13:39:51 <grawity> no, it should be run "inside" dbus-run-session
2023-11-22 13:39:52 <jonesv> grawity: I run sway manually with `exec dbus-session-run -- sway`. You mean this should be a script? :D
2023-11-22 13:39:53 <dari> add exec to sway/i3 config
2023-11-22 13:40:05 <jonesv> dari: oh, in the sway config I can do that probably
2023-11-22 13:40:19 <grawity> so if you do 'dbus-run-session sway' currently, then having sway run the rest will be easiest
2023-11-22 13:40:38 <jonesv> grawity: right
2023-11-22 13:40:57 <grawity> the aforementioned 'superd' looks handy in general (it would be also run from sway the same way)
2023-11-22 13:43:39 <jonesv> grawity: yes, I wanted to look into superd, but in general I like starting minimalistic: if I start having a lot of services in my sway config, then it will be time to try superd
2023-11-22 14:27:59 <jonesv> Thank you so much grawity, dari, Ermine, tretinha for all the help! Now I get the sound working at boot, it's great :-). 
2023-11-22 14:28:22 <jonesv> Next step: get Android-Studio running on Alpine :)
2023-11-22 14:28:36 <jonesv> (I think it will involve some bubblewrap for glibc)
2023-11-22 14:33:53 <jerkwad_robot> I need some help. I am struggling with PPPOE and getting Alpine setup as my firewall. DLS modem is in bridge mode. I setup Eth0 and PPPOE in /etc/network/interfaces manually. Both auto, dhcp. I have ifupdown installed. However, I am not getting an IPv4 address.
2023-11-22 14:34:31 <jerkwad_robot> Eth0 is the port I connect to the dls modem.
2023-11-22 14:34:39 <jerkwad_robot> Eth1 is my internal network.
2023-11-22 14:35:51 <jerkwad_robot> It could be the kernel modules... do they get loaded when I install PPPOE?
2023-11-22 15:45:01 <Forza> Hi. Do we have a naming convention for packages? I want to make a new package for a script that helps set up dm-cache. Thinking I should call it dm-cache or dmcache. 
2023-11-22 15:51:07 <Forza> https://wiki.alpinelinux.org/wiki/Package_policies
2023-11-22 16:16:50 <hyau[m]> Are you using Alpine firewall or an another one, jerkwad?
2023-11-22 16:17:10 <ikke> I use awall
2023-11-22 16:17:15 <ikke> (Which is the alpine fw)
2023-11-22 16:18:21 <ohdienge> I'm running alpine-linux with distrobox. I have installed helix (apk add helix). If I do "whereis" I don't see a binary. Where do I find the binary of the helix editor ?
2023-11-22 16:18:32 <hyau[m]> Yeah I think it has a problem with can you try ufw? I think my friend runs bridged too but uses ufw instead of wall.
2023-11-22 16:18:32 <ohdienge> hi all :)
2023-11-22 16:18:56 <hyau[m]> Have you tried running hx?
2023-11-22 16:18:59 <pj> ohdienge: it's hx, not helix
2023-11-22 16:19:12 <ohdienge> pj: ah thanks!
2023-11-22 16:19:33 <ikke> ohdienge: apk info -L helix would show you all the files in the package
2023-11-22 16:20:41 <ohdienge> it starts. thanks :)
2023-11-22 20:57:53 <vkrishn> should be a good read, https://wiki.insteps.net/Mobile/Install-alpinelinux-via-adb , in ~120 loc
2023-11-22 20:58:27 <vkrishn> working on video, should attach it soonish
2023-11-23 09:21:36 <frojnd> Hm.. I'm on 3.18 and would like to install calibre. Problem is that calibre is in testing, so according to https://wiki.alpinelinux.org/wiki/Repositories#Testing I had to add `@testing https://dl-cdn.alpinelinux.org/alpine/edge/testing` to repositories. But when trying to actually add it with apk I get ERROR: unable to select packages: https://termbin.com/u7axv
2023-11-23 09:22:05 <frojnd> How can I install calibre on 3.18? Without braking too much python packages
2023-11-23 09:24:31 <frojnd> Maybe I could build it myself locally
2023-11-23 09:31:22 <frojnd> forgot to add @testing after calibre but I still have a few packages missing now: https://termbin.com/yncx
2023-11-23 10:21:40 <zcrayfish> Another option, if you can't get the one in testing to work, is to use the calibre flatpak from flathub.
2023-11-23 10:59:50 <frojnd> Noted.
2023-11-23 10:59:52 <frojnd> Hmm...
2023-11-23 10:59:59 <frojnd> I have some issues with openconnect vpn
2023-11-23 11:00:32 <frojnd> It seems my /etc/resolv.conf is not properly setted and also some address are unavailable
2023-11-23 11:00:46 <frojnd> Maybe I missed something when installing alpine or openconnect network related
2023-11-23 11:21:13 <frojnd> My /etc/resolv.conf is not being populated with dns after I restart iwd service
2023-11-23 11:21:33 <frojnd> I've removed everyting from /etc/resolv.conf and file is absolutely empty :D
2023-11-23 11:21:37 <frojnd> So I guess I have missed something
2023-11-23 11:49:40 <Ermine> Is iwd supposed to populate resolv.conf
2023-11-23 12:07:50 <Forza> I want to prevent mdev from adding a by-uuid device link.
2023-11-23 12:08:10 <Forza> For specific uuids only
2023-11-23 12:09:47 <Forza> Maybe i can add a new file in /lib/mdev/ that removes it?
2023-11-23 12:11:35 <frojnd> Ermine: not sure?
2023-11-23 12:11:38 <frojnd> I jsut use iwd
2023-11-23 12:18:50 <dari> iwd on systemd systems when setup with "EnableNetworkConfiguration=true" will use systemd resolver
2023-11-23 12:19:30 <Forza> I think i get it. Mdev matches only one line in mdev.conf so i have to do something like: vd[a-z].*   root:disk 0660 */lib/mdev/persistent-storage; /opt/script.sh
2023-11-23 12:19:54 <dari> in Alpine it will be ignored until tell what else to use, for example: resolvconf
2023-11-23 12:34:25 <frojnd> So I should install resolvconf? Which is inside conman and oprenresolv ?
2023-11-23 12:34:46 <frojnd> So I should try and install maybe openresolv?
2023-11-23 12:36:44 <frojnd> dari: I have enabled EnableNetworkCOnfiguration=true because it then fires up dhcp
2023-11-23 12:37:32 <dari> dunno, I don't like all those weird managers for doing such a basic stuff
2023-11-23 12:37:55 <dari> frojnd: so then you need to tell iwd what should use to get dns servers
2023-11-23 12:38:45 <dari> frojnd: https://www.mankier.com/5/iwd.config#Settings-Network "NameResolvingService"
2023-11-23 12:45:28 <frojnd> Ok so I just isntall openresolv probably and tell it to use NameResolvingService=resolvconf
2023-11-23 12:47:28 <dari> ye and then check if/how openresolv work
2023-11-23 13:15:03 <frojnd> Seems like it
2023-11-23 13:15:22 <frojnd> It looks like it works
2023-11-23 13:15:48 <frojnd> On the other side I also have: IPSec VPN client compatible with Cisco equipment installed with openconnect
2023-11-23 13:17:11 <frojnd> And /etc/resolv.conf looks like: https://termbin.com/081d so not sure if vpnc updated the /etc/resolv.conf file or openresolv :D
2023-11-23 13:43:03 <dari> if resolv.conf before was empty and after openresolv installation something is happening then looks like vpnc detected resolvconf and used it
2023-11-23 14:00:36 <frojnd> Yeah looks like, have disconnected openconnect vpn's session and /etc/resolv.conf is updated
2023-11-23 14:00:38 <frojnd> Nice
2023-11-23 14:09:38 <dok> hi, I have a segfault when executing blender, it fails very eatly on, during some dlopen
2023-11-23 14:10:40 <dok> how can i help to find the issue? I am trying to rebuild blender with abuild from aports (so I could enable more debug info)
2023-11-23 14:28:35 <ikke> ulimit -c unlimited
2023-11-23 14:28:41 <ikke> should give you a coredump
2023-11-23 14:28:48 <ikke> (helps to use a version with debug symbols)
2023-11-23 14:28:59 <ikke> dok: also make sure you add a -dbg subpackage to blender
2023-11-23 14:37:14 <dok> ikke: i am not sure that there is a -dbg package, but maybe the default already has debug symbols... idk
2023-11-23 14:39:43 <ikke> dok: abuild strips debug symbols by default
2023-11-23 14:39:57 <ikke> if you add a -dbg subpackage, it won't and they will be collected in that subpackage
2023-11-23 14:41:52 <dok> uhmm ok, is it normal that apk doesn't list -dbg packages ?
2023-11-23 14:44:00 <ikke> It would if they exist
2023-11-23 14:49:06 <dok> okay
2023-11-23 14:51:33 <ikke> Not many packages are built with a dbg subpackage as they tend to take up a lot of space
2023-11-23 15:23:30 <frojnd> Hm
2023-11-23 15:25:27 <frojnd> I've added some android rules to /etc/udev/rules.d/ since I got permission denied for a device. Then I restarted: `doas udevadm control --reload-rules` and `doas rc-service udev restart` and created adbusers group and added my user to it. after relogin I can't even see permission denied, since I don't see any android device when adb devices
2023-11-23 15:26:35 <Ermine> I've read that group as 'abusers'
2023-11-23 15:28:56 <frojnd> :D
2023-11-23 15:29:05 <frojnd> system reboot helped, odd.
2023-11-23 15:29:20 <frojnd> I thought that saying was still active: be root don't reboot.
2023-11-23 15:39:21 <dari> frojnd: I think you shouldn't restart udev and after "udevadm control --reload-rules" enough is "udevadm trigger"
2023-11-23 15:42:55 <Habbie> frojnd, i prefer "root often, reboot often"
2023-11-23 15:59:37 <frojnd> :)
2023-11-23 15:59:56 <frojnd> How can I run a binary that is compiled agains gcc on an alpine?
2023-11-23 16:00:01 <Habbie> you mean against glibc?
2023-11-23 16:00:51 <frojnd> Yeah glibc: heimdall suite (a tool for samsung phones to install lineage os) I've just tired on a arch and it opens it
2023-11-23 16:04:10 <frojnd> Well don't have time currently gonna do it on another machine
2023-11-23 16:04:20 <frojnd> But def gonna have to solve this issue
2023-11-23 16:04:22 <frojnd> In the future
2023-11-23 16:37:04 <Forza> Sigh
2023-11-23 16:37:32 <Forza> Seems my udev rule from gentoo doesn't work with eudev on alpine
2023-11-23 16:41:06 <Forza> It says: invalid SYMLINK operation
2023-11-23 16:42:55 <Forza> The rule https://bpa.st/3I4Q
2023-11-23 16:46:53 <Forza> Seems there isnt a SYMLINK- option, only SYMLINK+
2023-11-23 17:18:38 <Forza> I did this https://bpa.st/CXWQ
2023-11-23 18:33:26 <grawity> Forza: OPTIONS{link_priority}="9001"
2023-11-23 18:46:08 <Forza> Grawity where?
2023-11-23 18:48:16 <mjt> hmm. the same people.. am I in #btrfs or in #systemd? :)
2023-11-23 18:51:40 <Forza> :o you're in #alpine-linux 
2023-11-23 18:54:06 <mjt> trying to get hold on this little thing after being a debian developer for some 15+ years (and after using redhat before and other unixes even before that). Obviously being old-scool myself. But alpine feels so.. stone-age.. I dunno why, it's all the stuff I used to and love
2023-11-23 18:56:57 <mjt> even busybox is one of my favority tools which I rely on and fix and maintain it in debian
2023-11-23 18:57:07 <mjt> favorite*
2023-11-23 19:09:43 <Forza> Stone age is good?
2023-11-23 19:10:05 <Forza> Alpine is efficient and logical
2023-11-23 19:10:16 <Forza> (Maybe not mdev... lol)
2023-11-23 19:10:39 <Ermine> Yes for some people
2023-11-23 19:11:01 <mjt> well, I used mdev before and liked it
2023-11-23 19:11:12 <Forza> I ended up pushing the dm-cache stuff to github. I think i need to re-think how to package it properly for Alpine
2023-11-23 19:11:17 <ikke> boring technology
2023-11-23 19:11:28 <mjt> ..and wrote some patches for mdev too
2023-11-23 19:11:29 <Forza> https://github.com/Forza-tng/dm-cache/
2023-11-23 19:11:47 <Ermine> stone age is using MAKEDEV and no device manager
2023-11-23 19:12:04 <Forza> Heh no devfs
2023-11-23 19:12:34 <mjt> devfs was fun
2023-11-23 19:12:54 <mjt> /dev/MAKEDEV was not :)
2023-11-23 19:13:14 <Ermine> some people would say it was
2023-11-23 19:15:24 <Forza> Ikke i closed the MR. I have too much headache so I'm making mistakes and taking time. Maybe we can look at it again another day? 
2023-11-23 19:16:55 <ikke> sure
2023-11-23 19:17:10 <ikke> Maybe take some rest :)
2023-11-23 19:18:17 <Forza> Yes. 
2023-11-23 19:18:39 <Forza> 
ACTION has chronic headaches, but lately is extra bad :(

2023-11-23 19:18:47 <ikke> sad to hear
2023-11-23 19:19:31 <Ermine> I know this feeling bro
2023-11-23 19:24:20 <Forza> 🙏
2023-11-24 02:00:26 <joseAlvarez> Hello, good evening from Mexico. Can someone help me fix my Alpine Linux server?
2023-11-24 02:04:51 <zcrayfish> what's wrong with it?
2023-11-24 02:06:26 <joseAlvarez> ok: I transformed a Ubuntu VPS into Alpine Linux and it worked excellently. Today, I enabled the edge repository and installed the python and poppler-utils packages. Poppler-utils didn’t work for me, so I uninstalled it, returned the repositories to default, and reinstalled poppler-utils. It worked, but the image from the PDF came out without text, so I installed the font-misc-misc package. Again, I got an image without 
2023-11-24 02:08:09 <joseAlvarez> I have access to the serial console and I can see the boot menu without options and it doesn’t go beyond that. 
2023-11-24 02:09:16 <joseAlvarez> I’m thinking of booting from a virtual DVD but I don’t know how to fix the boot menu without losing my data
2023-11-24 02:58:45 <joseAlvarez> I restarted my Alpine Linux machine and it won’t boot, how can I fix it without losing data? Thank you.
2023-11-24 06:35:50 <mjt> joseAlvarez2: do you have access to the boot console? What you describe so far, says nothing about your boot issues
2023-11-24 10:28:52 <dok> I have more time to debug why blender is having a segfault, i would like a rebuild the package but with debug info, how can i do that ?
2023-11-24 10:30:55 <ikke> dok: like I mentioned before, by including a $pkgname-dbg subpkg 
2023-11-24 10:31:18 <dok> hummm, so I only have to modify the APKBUILD right ?
2023-11-24 10:31:35 <dok> sorry, I am very new to alpine-linux
2023-11-24 10:31:46 <ikke> Yes
2023-11-24 10:32:15 <dok> and where do I found built packages ?
2023-11-24 10:32:29 <ikke> You also may need to look at the cmake build type 
2023-11-24 10:32:35 <dok> /var/cache/apk ?
2023-11-24 10:32:48 <ikke> They are placed in ~/packages 
2023-11-24 10:32:51 <dok> okay
2023-11-24 10:33:31 <dok> ... I don't see any dir ~/packages, but I believe I already built this yesterday
2023-11-24 10:36:07 <ikke> Did you build it with abuild -r?
2023-11-24 10:36:33 <dok> yes, and also with -rk
2023-11-24 10:37:21 <dok> maybe that's because I've not ceated the ~/packages dir before
2023-11-24 10:42:22 <ikke> You don't need to create it manually 
2023-11-24 10:45:23 <dok> humm... okay... actually getting blender sources are going to take a while, they apparently got DDoS
2023-11-24 10:46:31 <dok> maybe it didn't built yesterday
2023-11-24 10:47:08 <dok> yeah that's more likely, i built another package and it can be found in ~/packages/...
2023-11-24 10:47:22 <dok> thanks
2023-11-24 10:48:03 <dok> (I am going to "mirror" the source tarball locally)
2023-11-24 10:49:03 <ikke> It can probably also be found on https://distfiles.alpinelinux.org/distfiles
2023-11-24 10:51:05 <dok> ok
2023-11-24 10:52:26 <dok> much better, thanks again
2023-11-24 11:13:51 <singham> Folks, I'm unable to install alpine aarch64 on rpi3b+
2023-11-24 11:14:01 <singham> Nothing lights up
2023-11-24 11:14:37 <singham> Made bootable fat32. flashed all files in aarch64 tarball 
2023-11-24 11:14:44 <singham> no prompt on screen
2023-11-24 11:18:48 <dok> singham: strange I've also installed alpine on rpi3b+ and it went fine
2023-11-24 11:19:18 <singham> Which tarball?
2023-11-24 11:21:03 <dok> I think that was the 3.17 release, and the aarch64 "raspberry pi" download
2023-11-24 11:21:18 <dok> which includs all rpi dtbs
2023-11-24 11:28:37 <singham> yes. There is no bootloader?
2023-11-24 11:29:40 <dok> i think there is, it should be u-boot
2023-11-24 11:35:18 <dok> maybe not. but I have bootcode.bin and boot/ in the root of partition mmcblk0p1
2023-11-24 11:36:47 <dok> hummm and it's a rpi 3B not 3B+
2023-11-24 11:44:06 <singham> ok. 3.18 armv7 or aarch64 is simply not working!
2023-11-24 11:48:18 <dok> singham: have you tried with 3.17 ?
2023-11-24 11:54:40 <singham> I don't even get the rpi led lights on!
2023-11-24 11:55:13 <singham> there should be u-boot I think
2023-11-24 12:48:26 <slm> aer
2023-11-24 14:56:30 <telmich> is it possible that we don't have mesa drivers for i965? xorg complains with AIGLX error: dlopen of /usr/lib/xorg/modules/dri/i965_dri.so failed (Error loading shared library /usr/lib/xorg/modules/dri/i965_dri.so: No such file or directory) and I cannot find the file in edge on https://pkgs.alpinelinux.org/contents?file=&path=%2Fusr%2Flib%2Fxorg%2Fmodules%2Fdri%2Fi965_dri.so&name=&branch=edge&repo=main&arch=x86_64
2023-11-24 15:06:14 <telmich> found the solution: remove the xf86-video-intel package
2023-11-24 17:11:29 <mjt> that smells a wrong solution.. it just uses generic vesa driver instead of real one..
2023-11-24 17:51:52 <zcrayfish> mjt: It's the correct solution for recent Intel hardware, which uses modeset
2023-11-24 19:39:13 <sertonix> Does anybody know what might cause the a battery_supply device to not be detected? I only had that occur on alpine linux.
2023-11-24 19:39:25 <sertonix> The device is an surface pro 7
2023-11-24 19:40:40 <ikke> sertonix: perhaps some kernel module missing?
2023-11-24 19:41:35 <sertonix> Perhaps. But I don't know which could change something. I had tried some I found in the internet but no success
2023-11-24 19:45:14 <ikke> And you don't have an installation remaining where it works?
2023-11-24 19:45:57 <frag> date +%s  # gasp, didnt notice before now ..
2023-11-24 19:46:12 <ikke> you mean 17*?
2023-11-24 19:46:17 <frag> ikke: yees
2023-11-24 19:46:30 <frag> a new era begins...
2023-11-24 19:46:31 <frag> jk..
2023-11-24 19:46:34 <ikke> Happened passed Tuesday
2023-11-24 19:46:47 <ikke> (this happens hevery 3.x years
2023-11-24 19:47:06 <frag> hmm..
2023-11-24 20:14:51 <lopid> nobody can predict when it will happen next
2023-11-24 20:18:35 <quinq> It's like San Diego, its signification was lost eons ago
2023-11-24 20:22:14 <sertonix> After which strategy is the linux kernel config updated?
2023-11-24 20:22:43 <sertonix> For me CONFIG_BATTERY_SURFACE=m is missing but I have the feeling that there might be a lot more missing
2023-11-24 20:23:14 <ikke> sertonix: modules are enabled on request
2023-11-24 20:26:58 <sertonix> ikke: there are a lot of surface related kernel configs. I can only guess what configs would be benefitial for my device.
2023-11-24 20:27:30 <sertonix> Wouldn't it be better to use kernel defaults unless there is a reason not to?
2023-11-24 21:28:31 <jonesv> Is someone here using gpg with a smartcard (in my case Yubikey) here? I get this annoying "gpg: selecting card failed: No such device" error and I can't find what's missing
2023-11-24 21:40:29 <mjh[m]> why switching to an alternative mirror might cause apk search to fail?
2023-11-24 21:52:38 <jonesv> mjh[m]: how is it failing? Timeout?
2023-11-24 22:07:39 <fluix> jonesv: have you tried `strace`ing it to see what it's failing at
2023-11-24 22:10:28 <jonesv> fluix: I was just thinking about that, but then I didn't know how it would help. Since you mention it, let me try :). Do you think something glibc-related?
2023-11-24 22:10:43 <fluix> no idea, just a shot in the dark
2023-11-24 22:10:59 <fluix> hopefully just gets you a bit more information than the gpg error
2023-11-24 22:15:04 <jonesv> fluix: hmm I don't see anything special there: https://paste.sr.ht/blob/635519e6fdd8e3d59ca31b52dee493a448242302
2023-11-24 22:18:09 <fluix> looks like gpg-agent is the one returning an error that gpg is just printing out. you could try strace that instead
2023-11-24 22:18:28 <jonesv> fluix: hmm how would I run gpg-agent then? Does it appear in the strace?
2023-11-24 22:19:30 <fluix> perhaps you can just run it manually? if not, you'll have to modify wherever it gets run and pipe the strace output to a file (you may want the -f flag to follow forks and a larger -s <value> to print more of the strings)
2023-11-24 22:20:02 <fluix> you can see a connect(... call and then read and write calls which send and receive data, at the end receiveing "ERR 100696144 No such device ..."
2023-11-24 22:21:05 <jonesv> fluix: oh I had not realized that `connect(...` was in gpg-agent.
2023-11-24 22:21:53 <fluix> well, it looks like a connect *to* gpg-agent
2023-11-24 22:24:10 <jonesv> fluix: yeah I'm not completely sure how it works. I tried to `gpgconf --kill gpg-agent` and then run `gpg-agent --server -v`, which shows "OK Pleased to meet you". From another terminal I ran `gpg --card-status` and was expecting it to contact my gpg-agent, but it seems like it did not
2023-11-24 22:26:00 <fluix> perhaps there's gpg-agent logs you can look at?
2023-11-24 22:26:25 <jonesv> fluix: where would you look? I don't see anything in /var/log/messages
2023-11-24 22:27:03 <jonesv> fluix: (I am a bit new to openrc, I was on Arch + systemd before...)
2023-11-24 22:32:29 <fluix> which yubikey is this btw?
2023-11-24 22:33:16 <fluix> it may not log to syslog, so it would be wherever (if anywhere) it's configured
2023-11-24 22:34:23 <jonesv> fluix: Yubikey 5c NFC. It works on another Linux, with gpg, so I'm pretty sure it's supported. However I am maybe missing a package on Alpine, I don't know
2023-11-24 22:35:02 <fluix> alright, no idea if this is involved at all, but this file doesn't list it: https://gitlab.alpinelinux.org/alpine/aports/-/blob/master/main/gnupg/60-scdaemon.rules
2023-11-24 22:35:48 <jonesv> fluix: one thing is that in the strace, I see those protocol messages ("OK", "RESET"), and I would expect my `gpg-agent --server -v` to show that if the client connected to it. At least I was hoping for that.
2023-11-24 22:36:06 <fluix> oh, do you have gnupg-scdaemon installed?
2023-11-24 22:36:35 <jonesv> I have the 60-scdaemon.rules file on my system, from the gnupg-scdaemon package, but then it's not completely clear to me who runs scdaemon (it's not listed as an openrc service)
2023-11-24 22:36:37 <fluix> that would make sense, maybe it's connecting to a different one
2023-11-24 22:37:03 <fluix> guessing gpg-agent will spin it up as necessary, but not sure
2023-11-24 22:37:12 <jonesv> Like I created ~/.gnupg/scdaemon.conf like describe here, but I don't know who picks it up: https://support.yubico.com/hc/en-us/articles/360013714479-Troubleshooting-Issues-with-GPG
2023-11-24 22:37:49 <fluix> maybe you need a udev rule to add that same gnupg group to the yubikey??? really don't know here
2023-11-24 22:37:58 <fluix> also gotta run in a bit, so can't help further. good luck though!
2023-11-24 22:39:30 <jonesv> fluix: thanks for the help!
2023-11-24 22:39:46 <jonesv> I don't understand the group of the yubikey. How would the yubikey have a group?
2023-11-24 22:40:02 <fluix> see the udev rules file
2023-11-24 22:40:14 <jonesv> On my other system that works, my user is not in the "gnupg" group
2023-11-24 22:40:24 <fluix> nothing to do with your user I don't think
2023-11-24 22:40:51 <fluix> also maybe worth a try, install libusb-compat as suggested in *Arch*: https://wiki.archlinux.org/title/GnuPG
2023-11-24 22:40:55 <jonesv> So are you saying that the scdaemon.rules file may be missing a rule for my specific yubikey?
2023-11-24 22:40:59 <fluix> yes
2023-11-24 22:41:17 <jonesv> I compared the file to my other computer and it seems similar, but let me make an actual `diff`
2023-11-24 22:41:47 <jonesv> libusb-compat does not seem to magically fix it :(
2023-11-24 22:41:52 <fluix> ah well
2023-11-24 22:41:56 <fluix> anyways, yeah gotta run
2023-11-24 22:41:57 <fluix> gl!
2023-11-24 22:42:39 <jonesv> yeah thanks again!
2023-11-24 22:47:38 <joseAlvarez> Hello, can someone help me fix the boot of my Alpine system?
2023-11-24 22:48:39 <ptrc> joseAlvarez: don't ask to ask :p what's your issue?
2023-11-24 22:49:05 <ptrc> sorry, maybe bad wording; what's the issue with your Alpine system's boot?
2023-11-24 22:50:04 <joseAlvarez> Thank you! I restarted and it didn’t boot again, it just stays on the boot menu.
2023-11-24 22:52:34 <joseAlvarez> Right now I have the computer booted with Ubuntu and I have already backed up my data, I would like to know what commands I should use to try to fix the system loader.
2023-11-24 23:16:31 <ptrc> oh well
2023-11-24 23:17:00 <ptrc> why do people always lack the patience to just.. leave their client in an irc channel when they need help
2023-11-24 23:29:36 <jonesv> ptrc: sometimes I guess there is a need to reboot regularly. Also I guess many people don't use a bouncer, for some reason
2023-11-24 23:45:24 <jonesv> I'm going nowhere with my gpg issue :(. Time to sleep
2023-11-24 23:45:47 <jonesv> I hope it's not a glibc issue with smartcards. I really need to use my yubikeys
2023-11-25 00:05:07 <jonesv> fluix: I added "disable-ccid" and now it works! \o/ No clue why, though
2023-11-25 11:48:26 <freem> jonesv: has to be noted that IRCv3 does have a draft feature to get rid of the bouncer requirement, but I think most IRC networks do not implement it, and same for IRC clients
2023-11-25 11:49:05 <freem> which is a pity as the bouncer requirement is really annoying
2023-11-25 11:50:39 <freem> (out of laziness I choose to switch to quassel, but the history feature of this is really poor, and I should instead install back that good old hexchat and find motivation to setup a more traditional bouncer, wich does not stores thing in a binary database so that I could grep history easily)
2023-11-25 14:51:06 <frag> is /home included as default in lbu commit now?
2023-11-25 14:52:53 <ikke> I think only /root/.ssh
2023-11-25 14:53:39 <frag> hm.. 
2023-11-25 15:08:16 <frojnd> Hi there.
2023-11-25 15:10:08 <frojnd> I have a linux computer with luks2 encrption. I have to return the hardware. Now I want to use that software even though I'll return the hardware. Should I dd hard drive (boot from live usb) and dd nvme hard drive to another alpine machine (in the same local network) ? And then on that alpine machine I could somehow access all the data?
2023-11-25 15:10:31 <frojnd> luks2 encryption is Archlinux OS. 
2023-11-25 15:11:38 <scorpion2185[m]> yes you can acess that data if you have the password
2023-11-25 15:11:59 <scorpion2185[m]> you can copy it like any other data after opening luks
2023-11-25 15:14:24 <frojnd> But can I dd directly to another machine ?  I would probably have to setup some sort of network drive
2023-11-25 15:15:28 <quinq> I'd rather use tar
2023-11-25 15:15:47 <quinq> And yes, you can pipe that into an ssh link for example
2023-11-25 16:14:42 <frag> hmm my home dir was added to /etc/apk/protected_paths.d/lbu.list , dunno how :S
2023-11-25 16:16:14 <frag> oh well im gonna over write it with my script anyway
2023-11-25 16:46:52 <papiris[m]> I've got a script (`apk-snap`) that takes snapper snapshots on every apk commit, and I'd like to make the snapshot descriptions more... descriptive :p
2023-11-25 16:46:52 <papiris[m]> Does apk "announce" what package changes will be committed, in an API; or is my best bet parsing the stdout during commit?
2023-11-25 16:46:52 <papiris[m]> i.e. "upgraded pkgs: emacs, emacs-doc"
2023-11-25 17:13:15 <fluix> what's an apk commit?
2023-11-25 17:13:52 <txnintn> do you mean lbu commit?
2023-11-25 17:14:20 <ikke> committing a transaction
2023-11-25 17:14:33 <ikke> ie, applying the changes to the system after determining the new state is valid
2023-11-25 17:25:14 <jerkwad_robot> Hello all. Thanks for the awesome distro!
2023-11-25 17:41:23 <quinq> Nice, jerkwad_robot :)
2023-11-25 17:42:17 <jerkwad_robot> So, I need some more help :). I am running alpine as my home firewall. I am connecting to the internet via pppoe and that works fine for the most part. My issue is with some websites. I can resolve the names but the page won't show on my workstation browser.
2023-11-25 17:42:20 <jerkwad_robot> it just hangs.
2023-11-25 17:43:14 <jerkwad_robot> I've been logging dnsmasq DNS and it 'looks' fine on the surface, and I can dig from my workstation and get a response.
2023-11-25 17:43:59 <quinq> Most likely a MTU issue
2023-11-25 17:44:09 <jerkwad_robot> This wouldn't be a huge deal, but I am trying to setup a vm using qemu (not a fan of boxes) and the QEMU.org site won't load.
2023-11-25 17:44:30 <jerkwad_robot> ^ Ah, yeah. So, i am using nftables and I found this ....
2023-11-25 17:45:33 <jerkwad_robot> https://k1024.org/posts/2023/2023-04-16-nftables-tcp-clamp-mss/
2023-11-25 17:45:41 <quinq> Page title sounds about right
2023-11-25 17:46:20 <quinq> I don't use linux for firewall so can't really help much more, but I'd advise into trying that :)
2023-11-25 17:46:39 <jerkwad_robot> Yeah, I added those rules they mention. 
2023-11-25 17:47:29 <jerkwad_robot> Thanks quinq
2023-11-25 17:47:53 <jerkwad_robot> I also added the MTU setting to my pppoe settings.
2023-11-25 17:48:01 <quinq> But you still get those issues?
2023-11-25 17:48:17 <jerkwad_robot> yeah
2023-11-25 17:48:39 <jerkwad_robot> I must admit, I am a noob on nftables, but I think I have it set correctly.
2023-11-25 17:50:11 <jerkwad_robot> presntly, the MTU on my pppoe interface shows 1492
2023-11-25 17:50:24 <quinq> 18:47:53  jerkwad_robot$ I also added the MTU setting to my pppoe settings.
2023-11-25 17:50:33 <quinq> What setting? You shouldn't have to set anything there
2023-11-25 17:50:47 <quinq> Only clamp passing traffic to the pppoe tube MTU
2023-11-25 17:52:32 <quinq> I'd say paste your effective nftables config so that somebody more knowledgeable about it around here could help you
2023-11-25 17:59:27 <jerkwad_robot> I added the MTU setting in the ppp provider file
2023-11-25 17:59:33 <jerkwad_robot> Will do. 
2023-11-25 18:00:41 <quinq> But what MTU?
2023-11-25 18:10:47 <jerkwad_robot> https://pastebin.com/3kY0dY6z
2023-11-25 18:11:56 <jerkwad_robot> 1492
2023-11-25 18:27:21 <quinq> Why would you clamp incoming traffic?
2023-11-25 18:27:24 <quinq> It's already here
2023-11-25 18:27:32 <quinq> That's not the issue though
2023-11-25 18:27:39 <quinq>  ct state vmap { established : accept, related : accept, invalid : drop }
2023-11-25 18:27:49 <quinq> Isn't that matching and stopping the rule to apply?
2023-11-25 18:29:11 <jerkwad_robot> Hmm
2023-11-25 18:29:43 <jerkwad_robot> Let me re-read my nftable policy :)
2023-11-25 18:32:16 <jerkwad_robot> I may need to rebuild that. That might explain why, even if I disable the fw, those same websites don't work.
2023-11-25 18:32:43 <jonesv> What is not working really well for Desktop users on Alpine? I'm new to Alpine as my main distro (I used it for containers before), and it all just works. I just neede to install gcompat and then Android-Studio just worked, I did not even have to make a bubblewrap for glibc. Is there something in Alpine that is still a bit tricky because of "not glibc"?
2023-11-25 18:32:46 <jerkwad_robot> Frankly, I took it from the template they provide for a home firewall
2023-11-25 18:33:54 <jerkwad_robot> @jonesv, I haven't looked hard, but I am not sure what "office suites" are available for offline use.
2023-11-25 18:34:18 <quinq> jonesv, any ugly proprietary software I guess
2023-11-25 18:34:35 <jonesv> jerkwad_robot: there are libreoffice packages, at least
2023-11-25 18:35:08 <jonesv> quinq: haha right. Ugly proprietary softare that is not package in an ugly ElectronJS executable, I guess xD
2023-11-25 18:36:24 <quinq> It's a bit complex to say what would and wouldn't work without an actual list though
2023-11-25 18:36:55 <quinq> But any software that don't heavily depends on glibc-isms wouldn't cause problems
2023-11-25 19:29:46 <jonesv> quinq: right. Yeah it's just that a few years ago I started systematically trying to build C++ projects I contributed to with musl. Many needed a few tweaks, usually nothing bad, and then I contributed a musl worker in the project CI so that maintainers did not forget about it.
2023-11-25 19:30:34 <jonesv> quinq: And now I must say I haven't had a single issue yet. Maybe also that's because the projects I use that are not already packaged I have fixed in the past, I don't know. But I'm pleasantly surprised
2023-11-25 19:31:15 <quinq> I guess that the projects themselves are good enough :)
2023-11-25 22:10:19 <OdaNobunaga[m]> jonesv: "Is there something in Alpine that is still a bit tricky because of 'not glibc'?"
2023-11-25 22:10:19 <OdaNobunaga[m]> Browsers.
2023-11-25 22:43:31 <quinq> I guess it's about web browsers
2023-11-25 22:43:33 <quinq> But no
2023-11-25 23:03:08 <jonesv> OdaNobunaga[m]: oh? I am on Firefox, I haven't had an issue. Which browser has an issue?
2023-11-25 23:11:19 <modloopxtract> hey everyone
2023-11-25 23:11:39 <modloopxtract> im trying to do a custom install
2023-11-25 23:12:05 <modloopxtract> and i would need to extract the drivers from the modloop onto the ext4 rootfs into /lib
2023-11-25 23:12:29 <modloopxtract> does an alpine package contain the modloop? apk search modloop did yield only alpine-conf which is 252kb
2023-11-25 23:12:45 <modloopxtract> or maybe there is another way to do this what I am trying to achieve
2023-11-25 23:12:56 <modloopxtract> any help would be appreciated, thank you!
2023-11-26 04:09:58 <uuidNuniq> modloop is on alpine mirror
2023-11-26 08:39:45 <OdaNobunaga[m]> jonesv: The secure browsers (LibreWolf, Ungoogled Chromium etc.) are not available in Alpine. They can be installed via Flatpak though.
2023-11-26 08:39:45 <OdaNobunaga[m]> There are a lot of other programs Alpine is missing, like Tomb, Bottle and SMPlayer. Some are on Flathub, some not.
2023-11-26 09:19:57 <andym48[m]> I do miss SMPlayer 😀
2023-11-26 14:08:03 <jonesv> I am very happy with mpv :)
2023-11-26 14:08:16 <jonesv> OdaNobunaga[m]: Tomb, you mean this? https://github.com/dyne/Tomb
2023-11-26 14:08:33 <OdaNobunaga[m]> Yes.
2023-11-26 14:08:55 <jonesv> I am personally not a big fan of Flatpak/AppImage/snap :(
2023-11-26 14:14:04 <jonesv> OdaNobunaga[m]: oh, Tomb is not something to be compiled, it's "just" a shell script apparently, so it should not be too hard to package. Now it uses `sudo`, and I have `doas`. Also I am running on ZFS and it seems like it does not support that? Not sure.
2023-11-26 14:14:45 <jonesv> But yeah if I was to use tomb, I think I would try to make a package myself :)
2023-11-26 14:15:04 <jonesv> OdaNobunaga[m]: if I may ask: what's the point of using Tomb on top of a full disk encryption?
2023-11-26 14:15:19 <jonesv> OdaNobunaga[m]: is it to then share the file around (e.g. on a cloud or something)?
2023-11-26 14:16:36 <OdaNobunaga[m]> I wouldn't use it on an encrypted disk.
2023-11-26 14:18:18 <OdaNobunaga[m]> I was interested in OpenZFS with native encryption, but I found several articles claiming that the encryption from OpenZFS is not really finished and might eat all data. Someone else mentioned tomb as an alternative to the native OpenZFS encryption.
2023-11-26 14:23:48 <jonesv> Oh, I hope I did not make a mistake by going for OpenZFS then xD
2023-11-26 14:30:18 <OdaNobunaga[m]> Someone has linked this: https://docs.google.com/spreadsheets/d/1OfRSXibZ2nIE9DGK6swwBZXgXwdCPKgp4SbPZwTexCg/htmlview
2023-11-26 14:30:18 <OdaNobunaga[m]> And there's Github: https://github.com/openzfs/zfs/issues?q=is%3Aopen+is%3Aissue+label%3A%22Component%3A+Encryption%22
2023-11-26 14:42:37 <jonesv> OdaNobunaga[m]: thanks!
2023-11-26 17:04:44 <invoked> i don't know what "finished" means with respect to encryption that's been part of zfs for years. anyway, zfs on luks is still fine (and is slightly more private than zfs native encryption). 
2023-11-26 18:13:59 <Soni> hello! are you familiar with ipv6-only networks, and the benefits of ipv6-only kernels for ipv6-only networks?
2023-11-26 18:42:56 <satmd> I don't think you can even configure a kernel to be ipv6-only, because IPv6 is a subsection of IP in the kernel config.
2023-11-26 18:43:27 <satmd> Info might be old, but I've tried that some months ago :D
2023-11-26 18:48:20 <Soni> hmm, that seems like a solvable problem, it is just code after all...
2023-11-26 18:54:28 <Soni> anyway, DHCP(v4) has this option 108, which claims to disable the IPv4 stack. in practice it doesn't disable the IPv4 stack but merely disables the DHCP client, instead leaving it up to a CLAT to implement ipv6 <-> ipv4 translation. the benefit of an ipv6-only kernel is removing the ipv4 stack altogether and doing the translation at the libc/API level.
2023-11-26 18:55:32 <telmich> <Soni> "hello! are you familiar with ipv..." <- quite a bit, yes. We are running all of our servers in ipv6 only networks
2023-11-26 18:55:46 <ikke> telmich: I was hoping you would pop in :)
2023-11-26 18:55:57 <telmich> :-)
2023-11-26 18:56:10 <telmich> Soni: What is the problem that you are trying to solve?
2023-11-26 18:56:28 <Soni> telmich: mostly wanna show folks that ipv6-only kernels are viable
2023-11-26 18:56:44 <telmich> There is no such thing under Linux, let me find the notes
2023-11-26 18:57:08 <Soni> it would be an ABI break. but maybe a worthwhile ABI break.
2023-11-26 18:57:25 <telmich> Found it
2023-11-26 18:57:26 <telmich> https://ungleich.ch/en-us/cms/blog/2019/01/09/die-ipv4-die/
2023-11-26 18:57:32 <Soni> telmich: how are you getting CLATs/BRs to work?
2023-11-26 18:57:53 <telmich> There is even a 300 CHF bounty, if you are able to build and show that linux works w/o IPv4: https://ungleich.ch/u/projects/jobs-hacks-bounties/
2023-11-26 18:58:07 <telmich> Soni: we are not, we are using nat64
2023-11-26 18:58:42 <Soni> ah :<
2023-11-26 19:00:01 <telmich> You only need clatd if you want to support IPv4 on the host and we don't need IPv4 in our networks, hence IPv6 only
2023-11-26 19:00:44 <Soni> yes. uh wait do you not have SIIT-DC?
2023-11-26 19:01:02 <telmich> We do as well
2023-11-26 19:01:38 <telmich> Generally speaking, I put both stateful and stateless both under NAT64, even though SIIT is the more correct term
2023-11-26 19:03:39 <Soni> do you do illegal DNSSEC-breaking things?
2023-11-26 19:05:55 <telmich> Sure
2023-11-26 19:06:21 <telmich> That's fine, because if sites don't want us to, they just need to enable IPv6
2023-11-26 19:06:28 <Soni> :/
2023-11-26 19:06:38 <Soni> DNS64 shouldn't exist
2023-11-26 19:06:53 <Soni> in fact this would be something handled by ipv6-only kernels
2023-11-26 19:07:35 <telmich> How would an IPv6 only kernel handle that in your opinion?
2023-11-26 19:07:38 <Soni> (or rather, by moving the CLAT into libc)
2023-11-26 19:08:42 <telmich> But if your kernel is IPv6 only, you cannot use IPv4 anymore, the sockets/syscalls would not function
2023-11-26 19:09:01 <Soni> when you get rid of the ipv4 stack but you still wanna support ipv4 apps, you can use libc-based CLAT/API translation (a la bump in the API, but stateless just like a CLAT)
2023-11-26 19:09:14 <Soni> (RFC 3338)
2023-11-26 19:09:41 <Soni> so while your source code would look like ipv4 API calls, it would really be calling ipv6 APIs under the hood
2023-11-26 19:10:00 <Soni> including synthesizing IPv6 addresses based on the NAT64/SIIT prefix
2023-11-26 19:10:21 <telmich> That is kinda of what is being done with getaddrinfo() vs. gethostbyname
2023-11-26 19:10:44 <Soni> but it wouldn't use DNS64
2023-11-26 19:11:09 <telmich> I am not sure if your approach could work, as there are a lot of IPv4 related (old) syscalls, but hypothetically, mocking them in the libc would be an interesting project. 
2023-11-26 19:11:26 <Soni> again: ABI break, recompile your stuff
2023-11-26 19:11:53 <Soni> whether it's a worthwhile ABI break... well it would allow eliminating the ipv4 stack and solving the "why does DNS64 still exist" problem
2023-11-26 19:12:34 <telmich> Aside from this nice hypothetical idea, did you actually have an alpine linux specific question or are you more interested in a general ipv6 (only) discussion?
2023-11-26 19:13:07 <telmich> Because for the former, this is a good room, but for the latter, the discussion might be more appropriate in something like the IPv6 matrix room, #ipv6:ungleich.ch 
2023-11-26 19:13:14 <Soni> well, alpine is known for being different from other distros... presumably it would be more open to such ABI breaks
2023-11-26 19:13:59 <telmich> Soni: Alpine does not exactly maintain the libc, but it does use the musl libc
2023-11-26 19:14:14 <telmich> So if you were to change musl, alpine would automatically get the changes
2023-11-26 19:14:19 <Soni> kernel ABI breaks
2023-11-26 19:15:19 <telmich> Gotta go, time for dinner. But generally speaking, if you want to create an IPv6 only linux distro, I'm all ears, but I can promise you there is quite a bit of work to be done
2023-11-26 19:15:20 <Soni> (specifically the kernel<->userland ABI, also known as the syscall ABI?)
2023-11-26 19:18:31 <Soni> alright
2023-11-26 21:24:58 <jonesv> Have you guys ever seen "Gdk-Message: 22:20:36.810: Unable to load hand2 from the cursor theme" with Firefox? Basically I don't get the "hand" cursor on links, somehow it seems like it is missing from my system
2023-11-26 21:39:19 <ikke> Maybe you need to give it a hand? 🙃
2023-11-26 21:41:31 <TangentDelta> Hello
2023-11-26 21:45:49 <TangentDelta> I have a couple of SBCs that use the TI AM3352 SoC, which is in the same family as the SoC used on the BeagleBone boards. I've been trying to get Alpine to boot on one, but haven't been having the best luck.
2023-11-26 21:48:45 <jonesv> ikke: xD
2023-11-26 21:48:59 <TangentDelta> I think that the version of U-Boot that they come with is the source of the problems, but I don't know enough about booting Linux with U-Boot to know for sure. I can load the kernel, initramfs, and DTB into RAM off of the SD card just fine, but when I try to start the kernel up I get an vague error saying that the initramfs isn't the correct format.
2023-11-26 21:50:57 <ikke> jonesv: did it work before or has it never worked on that system?
2023-11-26 21:51:55 <TangentDelta> Some Googling seems to suggest that I need to rebuild the RAMdisk image in a different, "legacy" format
2023-11-26 21:59:07 <TangentDelta> Hmm, I'll give that a shot.
2023-11-26 22:31:38 <TangentDelta> Ah! That was it!
2023-11-26 22:32:37 <TangentDelta> Kernel booted, but got an error mounting the rootfs and got kicked to an emergency shell. Progress!
2023-11-26 22:50:23 <jonesv> ikke: it's a very new system, and I never used Alpine on desktop before. So for me it never worked
2023-11-26 23:59:16 <TangentDelta> Hmm, I need to build the device tree for this board :\
2023-11-27 02:06:08 <uuidNuniq> TangentDelta: a suggestion: chainload a newer u-boot on top of the stock u-boot
2023-11-27 02:07:35 <uuidNuniq> something like this: start_emmc_autoscript=dcache flush ; if ext4load mmc 1 0x1000000 uboot; then go 0x1000000; fi; if fatload mmc 1 0x1000000 uboot; then go 0x1000000; fi;
2023-11-27 02:08:39 <TangentDelta> Thanks! I think I need to build my own U-Boot for this board. I tried the generic AM335x U-Boot meant for the BeagleBone, but it just runs off into the weeds as soon as I jump into it.
2023-11-27 02:09:31 <uuidNuniq> the u-boot source code has many presets, find one that's as close to your board as possible 
2023-11-27 02:10:13 <TangentDelta> I was able to compile the dtb, using the dts provided by Netgate for this board, but the ethernet PHY is acting really weird. I have to pin my switch's port at 100M FDX or else I get one-way traffic.
2023-11-27 02:11:48 <uuidNuniq> well, you need to find out the compatible keyword on ethernet from the DTS, and load related modules earlier like at initramfs stage
2023-11-27 02:14:05 <uuidNuniq> TangentDelta: e.g. : https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/tree/arch/arm64/boot/dts/amlogic/meson-gxl.dtsi?h=v6.6.2#n778
2023-11-27 02:14:17 <uuidNuniq> this is the module you need to load for example.
2023-11-27 02:14:35 <uuidNuniq> gxl-mdio-mux in this case.
2023-11-27 02:15:22 <TangentDelta> Ah, I see
2023-11-27 02:16:06 <uuidNuniq> a lot to do, i think. 1) a newer u-boot from 2023 2) a dtb 3) load ethernet module  in alpine's initramfs 
2023-11-27 02:21:06 <uuidNuniq> TangentDelta: can you find the dts from here? https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/tree/arch/arm/boot/dts/ti/omap?h=v6.6.2
2023-11-27 02:21:15 <uuidNuniq> which one is for your board?
2023-11-27 02:23:29 <TangentDelta> Unfortunately, I have been unable to find a dts for this board for Linux. I think it used a board file for all of the defs. I verified this dts against some similar boards, and compiled it using the Linux kernel source https://github.com/freebsd/freebsd-src/blob/main/sys/dts/arm/ufw.dts
2023-11-27 02:45:59 <TangentDelta> Thanks for taking the time to help me with this uuidNuniq, I think you've given me a good direction to head in. I'll work more on it tomorrow.
2023-11-27 03:03:38 <uuidNuniq> TangentDelta: you're welcome. it's just i happen to have some experience on alpine on ARM. my device is a Amlogic S905D SoC (4 core A53)
2023-11-27 12:43:47 <vkrishn> found this, https://kernal.eu/posts/alpinedroid/, for those who have su installed
2023-11-27 13:35:25 <TangentDelta> I've never cross-compiled for ARM on my AMD64 desktop before. That was easier than I thought it would be.
2023-11-27 13:36:01 <TangentDelta> Just had to get the toolchain package and then specify the prefix that the toolchain uses.
2023-11-27 16:29:14 <blt[m]1> I am trying to bring up my eth0 and get dhcp to run automatically when I connect my laptop to my dock but adding 'allow-hotplug eth0 iface eth0 inet dhcp' in /etc/network/interfaces it doesn't work, is there another way or am I missing something ?
2023-11-27 17:58:08 <joseAlvarez> How can I clone my configuration to another computer? In case I lose the first one, have an operational copy.”
2023-11-27 18:21:30 <bl4ckb0ne> is there a way to do a `zfs load-key` on boot? I have an encrypted pool that requires manual mount 
2023-11-27 18:30:50 <lopid> perhaps /etc/local.s/READM
2023-11-27 18:30:53 <lopid> local.d
2023-11-27 18:30:58 <lopid> README
2023-11-27 18:31:15 <lopid> /etc/local.d/README  jees that disaronno hits fast
2023-11-27 18:31:27 <lopid> quickly
2023-11-27 18:32:11 <lopid> manual mount? so you need to provide it a password?
2023-11-27 18:39:03 <bl4ckb0ne> i need to enter my user password to load the key
2023-11-27 18:45:02 <bl4ckb0ne> looks like I was missing zfs-load-key on runlevel boot
2023-11-27 21:10:18 <quinq> Hello
2023-11-27 21:10:34 <quinq> Is mtp access to devices working for you on edge?
2023-11-27 21:10:52 <quinq> Here it doesn't detect devices anymore (6.1.63-0-lts)
2023-11-27 21:23:59 <quinq> simple-mtpfs tells me: LIBMTP PANIC: Trying to dump the error stack of a NULL device!
2023-11-27 21:24:03 <quinq> Nothing else
2023-11-28 01:00:01 <micahrl> does alpine initramfs-init support switching into a squashfs root? it looks like no from my reading of it, but I am wondering if there's some blessed way to do that that I'm missing.
2023-11-28 01:35:21 <TangentDelta> Hmm. Well, I got a newer version of U-Boot to compile and run on the board, but there's something wrong with the config somewhere.
2023-11-28 02:09:28 <TangentDelta> It looks like I can't just create the config and dts and expect it to work. There's stuff in the board/ti/am335x/board.c that I have to modify.
2023-11-28 02:10:28 <TangentDelta> There's a bunch of if statements for making sure the board and the config match up and I have to add my new board to that, for example.
2023-11-28 03:59:46 <jscript> what containers work on alpine linux itself
2023-11-28 04:12:15 <jscript> does flatpack work in alpine
2023-11-28 04:13:00 <zcrayfish> Yes, flatpak works just fine in alpine.
2023-11-28 04:14:45 <jscript> alright ^^
2023-11-28 04:15:01 <jscript> does docker work as well?
2023-11-28 04:15:36 <jscript> also snapcraft and appimage do not work right?
2023-11-28 04:17:33 <jscript> hmm https://youtu.be/3YHIa3le_1k?si=iJLguusuUZgHTMEw
2023-11-28 04:20:49 <jscript> Things! (Each Has Its Purpose)"
2023-11-28 04:20:49 <jscript> "Snaps, Flatpaks and Applmages Do Very Different 
2023-11-28 05:28:11 <Saijin_Naib[m]> Docker and podman work fine
2023-11-28 05:28:29 <Saijin_Naib[m]> no snapd, so no snaps, and appimage only use glibc currently, so yes, they also do not work
2023-11-28 05:28:45 <Saijin_Naib[m]> I know I asked this before, but we do not package anything that can unrar, right?
2023-11-28 05:28:59 <Saijin_Naib[m]> I have 7zip from Windows in WINE, but I'd rather use XArchiver where Ic an
2023-11-28 05:39:59 <pj> Saijin_Naib: bsdtar?
2023-11-28 05:41:27 <pj> anything that uses libarchive should be able to unpack rar archives
2023-11-28 05:41:36 <pj> bsdtar is part of libarchive-tools package
2023-11-28 05:55:24 <Saijin_Naib[m]> Let me see if I have that all, and if I do, I guess it is Issue time with XArchiver for not picking them up for usage
2023-11-28 05:56:16 <Saijin_Naib[m]> Yep, Issue time. Thanks!
2023-11-28 05:56:25 <Saijin_Naib[m]> that or we need to compile our 7zip to support unrar
2023-11-28 05:57:23 <grawity> does alpine have a 'nonfree' repository where such things go?
2023-11-28 05:57:34 <Saijin_Naib[m]> Hello71: I see you disabled unrar in 7zip due to it being non-free, but we have it in libarchive... Can we enable it in 7zip?
2023-11-28 05:57:50 <Saijin_Naib[m]> grawity: I think ages ago, but that got dropped long before unmaintained even
2023-11-28 05:57:56 <Saijin_Naib[m]> Maybe 3.10 era or so?
2023-11-28 05:59:06 <grawity> afaik the difference is that libarchive has its own implementation of the rar format, while 7zip ships the "official" unrar code
2023-11-28 06:07:24 <Saijin_Naib[m]> <grawity> "afaik the difference is that..." <- Oh... Oh, that is a bit less clear. Damnit
2023-11-28 06:08:47 <grawity> okay, in other words, it's not the overall RAR support that is non-free, it's specifically the RARLAB-provided code that is non-free
2023-11-28 06:13:38 <pj> yes
2023-11-28 06:14:26 <pj> grawity: non-free was removed since we do not want to promote non-free software but also because it was mostly unmaintained and no one really cared about it
2023-11-28 06:15:47 <Saijin_Naib[m]> Oh, sorry, I meant less clear as in what path to take forward, but thank you for the explanation regardless
2023-11-28 06:32:11 <ajhalili2006> Speaking of channel logs, do the devel channel also logged or only the main channel?
2023-11-28 06:32:23 <ikke> https://irclogs.alpinelinux.org/
2023-11-28 06:32:24 <ajhalili2006> (oh nevermind, just checked)
2023-11-28 12:52:56 <frojnd> I think I need a "normal" mount command and not busybox: mount: mounting /dev/mapper/wd-red on /mnt/encrypted/ failed: Invalid argument
2023-11-28 12:53:07 <frojnd> Problem is that I don't know how to install "normal" mount?
2023-11-28 12:53:22 <frojnd> /dev/mapper/wd-res is ext4 formatted thoguh
2023-11-28 12:53:30 <frojnd> Or do I have to isntall ext4 on alpine?
2023-11-28 12:53:59 <ikke> frojnd: mount -t ext4
2023-11-28 12:54:24 <frojnd> ikke: yes that way it worked
2023-11-28 12:54:26 <frojnd> Thank you
2023-11-28 12:54:50 <ikke> It means the ext4 module was not loaded yet 
2023-11-28 12:58:42 <frojnd> I have an .img file (my entire system from old laptop) can I somehow run this .img with qemu?
2023-11-28 12:59:21 <frojnd> I've just dded to img an entire nvme disk that I now have on /mnc/encrypted/system.img
2023-11-28 13:36:09 <TangentDelta> I was looking at the device tree and configuration for a Marvell Armada-based system that I have a couple of, and it's so simple and straight-forward compared to this AM335x-based board, lol
2023-11-28 13:37:31 <sewn>   how can i ask apk to re-launch a package hook?
2023-11-28 13:37:39 <TangentDelta> They use some sort of "Marvell MBUS" bus, which is PCI-like
2023-11-28 14:41:41 <Hello71> frojnd: losetup -fP img
2023-11-28 16:25:46 <caskd[m]> -Pf *
2023-11-28 16:26:04 <caskd[m]> -f is the flag that takes img as argument
2023-11-28 17:21:26 <ham5urg> I try to ssh -X ... vncviewer ... onto a alpine-box but I get an "Unable to find fonts. Check your FontConfig configuration.". What is the name of the font package I need to install?
2023-11-28 17:26:09 <ham5urg> I can't find the meta package which pulls in the fonts under /usr/share/fonts
2023-11-28 17:47:00 <quinq> ham5urg, what are you trying to run?
2023-11-28 17:50:35 <ham5urg> quing, I tried vncviewer on a console-only box. I got it to run after I installed ttf-dejavu, freetype, xauth
2023-11-28 17:55:02 <quinq> ok so problem solved? :)
2023-11-28 18:45:57 <ham5urg> quing, problem solved. I will ask the vnc-people about the min. depenencies for vncviewer-headless.
2023-11-28 19:40:30 <quinq> ok, cool :)
2023-11-28 20:16:27 <ham5urg> How do I 'shutdown -h now' with busybox?
2023-11-28 20:17:14 <quinq> You poweroff
2023-11-28 20:19:04 <ham5urg> Thanks
2023-11-28 20:21:04 <quinq> :)
2023-11-28 23:13:27 <pjs> anyone know of a useful tool to monitor disk usage and notify of strange usage increases?
2023-11-28 23:13:53 <pjs> Not sure why but since switching to Alpine linux a few months ago I lose 300-600mb a day of storage and I can't for the life of me figure out where it's all going
2023-11-28 23:13:55 <ssm> pjs: former can be done with the df utility (in base)
2023-11-28 23:14:10 <ssm> ncdu package may also be of interest
2023-11-28 23:14:20 <ssm> ncurses version of the du command
2023-11-28 23:14:20 <pjs> ssm: right, basically I want something to pattern where the usage is growing
2023-11-28 23:14:35 <ssm> sort by modification time maybe?
2023-11-28 23:14:49 <pjs> yea maybe that's a good idea
2023-11-28 23:47:19 <Tommy[m]12> Hello everyone... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/kAINmAZLUuoDKdkmKCdUlkze>)
2023-11-29 01:09:01 <Hello71> caskd[m]: -f changes the mode, it doesn't actually take an option argument. for example, losetup -f file1 -f file2 is invalid
2023-11-29 03:49:56 <qaqland> search in https://pkgs.alpinelinux.org/ down? 
2023-11-29 03:50:38 <qaqland> no result about "terminus-font"
2023-11-29 03:53:56 <qaqland> oh, it changes name from "terminus-font" to "font-terminus" :)
2023-11-29 06:06:25 <caskd[m]> Hello71: ohh okay
2023-11-29 07:42:00 <aa__> Hi all, thank you for making this amazingly lightweight operating system. I'm a front-end developer running home server with AlpineLinux and LXD in it. However, I got multiple kernel panics related to memory free even in freshly installed instances. I believe that the problem is in cgroup and sub{u,g}id because lxcfs and lxc service still fail to start. From the following issue, I could make my containers wake up but still the error present: https://gitla
2023-11-29 07:42:01 <aa__> b.alpinelinux.org/alpine/aports/-/issues/15290
2023-11-29 07:44:12 <aa__> My LXD setup is from AlpineLinux wiki: https://wiki.alpinelinux.org/wiki/LXD ; and none of additional configs were touched since then. Since I'm not a Linux developer, it's pretty difficult for me to debug everything and resolve the issue. If someone can explain me what's happening, I'll be really appreciate of it.
2023-11-29 07:45:47 <aa__> By default /etc/sub{u,g}id file have same content; root:1000000:1000000000\nroot:100000:65536. Lxcfs service says: chown: invalid user: '1000000100000:1000000100000'.
2023-11-29 08:58:27 <suffix> \quit
2023-11-29 09:05:34 <Tach> Hi guys, why does this gives issues on Alpine ?? sed -i '/^### foo$/a bar' /home/myfile.txt
2023-11-29 09:05:46 <Tach> sed: unsupported command '
2023-11-29 09:16:52 <ikke> Tach: busybox sed is more limited. Install gnu sed to get full support 
2023-11-29 09:17:28 <Tach> ikke yeah I read about that but it seems it works fine on cli but not from bash
2023-11-29 09:20:37 <quinq> busybox sed supports that -i strange option
2023-11-29 09:21:11 <quinq> Maybe you broke your IFS in your bash session
2023-11-29 09:21:30 <quinq> Maybe you have a different PATH and not actually using busybox sed
2023-11-29 09:21:48 <Tach> quinq yeah investigating!
2023-11-29 09:21:50 <Tach> good ones
2023-11-29 09:21:58 <quinq> Maybe what you're showing us is not exactly what you're trying
2023-11-29 09:22:01 <quinq> Who knows ^^
2023-11-29 09:25:27 <mjt> busybox sed definitely works with that expression
2023-11-29 09:26:05 <mjt> unsupported command ' ??  where does this ' comes from?
2023-11-29 09:26:49 <mjt> sed -i "'/^### foo$/a bar'" t
2023-11-29 09:26:49 <mjt> sed: unsupported command '
2023-11-29 09:28:42 <mjt> so.. you have to investigate who is interpreting (or not) your quotes
2023-11-29 10:45:05 <Tach> mjt from my command I thought... going to test the double quotes again
2023-11-29 12:45:50 <ham5urg> Does a alpine.tar.gz exist to just unpack, configure and boot? Installation without the installer.
2023-11-29 12:47:30 <ikke> Only for some platforms, and then only as a run-from-ram system 
2023-11-29 13:10:41 <grawity> on x86 I've used the bootstrap tarball for that purpose
2023-11-29 13:10:52 <grawity> the tiny one with basically busybox
2023-11-29 13:11:41 <ikke> minirootfs 
2023-11-29 13:11:50 <grawity> yeah I think that
2023-11-29 13:12:09 <grawity> it worked well enough, though I had to learn openrc the harder way
2023-11-29 13:12:19 <ikke> But you need to take care of installing the bootloader, the kernel and firmware, openrc and anything else yourself 
2023-11-29 13:14:07 <grawity> sure but that's like installing archlinux >_>
2023-11-29 13:21:23 <stingray_> Hello! I have a problem with installing alpine. At the moment of choosing for a repository, there is nothing in the list. Editing the file myself doesn't fix the problem... Could someone help me?
2023-11-29 13:21:54 <ikke> stingray_: do you have a working network connection?
2023-11-29 13:27:02 <stingray_> @ikke no... I thought I have because all my VMs work, but when I ping with alpine it doesn't work. I'm on UTM on Mac.
2023-11-29 13:28:00 <ikke> It fetches the mirror list online, so without a connection it cannot 
2023-11-29 13:30:27 <stingray_> How is it possible than other linux distribution work but alpine don't ? Is there something to do to activate the network? I followed the setup-alpine lux script
2023-11-29 13:32:33 <ikke> Does `ip link list` return interfaces?
2023-11-29 13:33:08 <stingray_> yes eth0
2023-11-29 13:56:35 <amnesia_> hi
2023-11-29 13:57:03 <amnesia_> how are you all?
2023-11-29 14:02:36 <ok> Hello. I would like to ask if some has a recommendatation how it can be found from which package is header? For example, I have compilation error:... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/UxOnncNgGjThKlUIroIRnAEw>)
2023-11-29 14:03:35 <ok>  * Hello. I would like to ask if some has a recommendatation how it can be found from which package is header? For example, I have compilation error:... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/uuzsLVKhiMQuhrPXrwXwPzaD>)
2023-11-29 14:06:53 <ok> Also i tried: apk search -e types.h w/o any  results
2023-11-29 14:06:54 <ok> s/results/result/
2023-11-29 14:08:47 <ok> ok, installation of musl-dev helps me but why 'apk search -e types.h' doesn't help to find musl-dev?
2023-11-29 14:14:17 <Hello71> the command help says: "apk search searches all repositories for packages matching at least one pattern. If no pattern is given, it lists all packages in the repository. A pattern matches if it is a case-sensitive substring of the package name." and "-e, -x, --exact       Match package names exactly"
2023-11-29 14:14:32 <Hello71> so "musl-dev" string does not contain "types.h", why would it be printed?
2023-11-29 14:15:38 <ikke> apk has no index of what package contains what files 
2023-11-29 14:16:19 <ok> thanks !
2023-11-29 14:34:57 <free_mind> Can someone tell me how many days are left before release 3.19 will show up?
2023-11-29 14:35:42 <ptrc> free_mind: don't think so - there isn't a specific release date scheduled afaict
2023-11-29 14:38:56 <free_mind> I believe in Nov there shows up a new stable branche; see: https://alpinelinux.org/releases/ "Each May and November we make a release branch from edge."
2023-11-29 14:40:58 <ikke> There is no fixed date. The first rc has been tagged and it's mostly testing now
2023-11-29 14:43:12 <ok> am i right that there is no riscv64-linux-gnu-gcc in Alpine, only none-elf?
2023-11-29 15:23:34 <free_mind> ikke: any estimate how many days the testing takes? One week, two weeks...?
2023-11-29 15:31:51 <invoked> free_mind: just out of curiosity, is there something specific from 3.19 you're after?
2023-11-29 15:37:44 <free_mind> invoked: no not really, just like to avoid installing a new system and then upgrading within a week time since a new release comes out...
2023-11-29 15:38:37 <free_mind> invoked: I plan to install a new laptop system with alpine in RAM.
2023-11-29 16:07:15 <invoked> roger that
2023-11-29 16:09:24 <ikke> free_mind: whenever ncopa decides it's stable enough to release
2023-11-29 16:09:29 <ikke> I cannot tell exactly when that is
2023-11-29 16:10:59 <free_mind> ikke: I respect that. Just thought maybe someone can estimate it, so I know I should/can wait a few days.
2023-11-29 16:13:44 <ncopa> free_mind: depends a bit on how much we want fix/test before the release
2023-11-29 16:14:05 <ncopa> i would guess that we have a rc2 this week, and the final release next week
2023-11-29 16:14:26 <ncopa> doing some "risky" late minute changes, like bumping the kernel to 6.6
2023-11-29 16:14:34 <ikke> free_mind: take a look at the historical releases dates to get an idea: https://alpinelinux.org/releases/
2023-11-29 16:15:44 <ncopa> but there are some bumps. zfs does not build with 6.6 on aarch64 due to some license changes, etc
2023-11-29 16:15:55 <ncopa> all those minor things takes time to investigate and deal with
2023-11-29 16:16:38 <ncopa> right now i am investigating if its worth to do a separate build for the rpi 5 with 16k pages
2023-11-29 16:17:59 <ikke> ncopa: related: https://social.treehouse.systems/@marcan/111493984306764821
2023-11-29 16:18:26 <ncopa> looks like its not worth it: https://www.raspberrypi.com/news/benchmarking-raspberry-pi-5/
2023-11-29 16:19:40 <free_mind> ncopa: In my/this case, the cpu whould be x86_64 (not aarch64), but I respect your efforts and opinion/judgement on this matter. Maybe I just go for latest stable now and upgrade later on.
2023-11-29 16:20:10 <ikke> free_mind: FYI, alpine linux upgrades are typically very trivial
2023-11-29 16:20:13 <ncopa> free_mind: yeah, it may take a week or two, but hopefully not more than that
2023-11-29 16:22:26 <free_mind> ikke: That is true; I've been using Alpine for many years on quite a few RPI's simultanious. Running smooth. ;)
2023-11-29 16:23:56 <free_mind> ikke: Background for this case is a tablet and/or laptop that just need to make a RDP connection over a wireguard vpn.
2023-11-29 16:24:21 <free_mind> ikke: booting with usb stick and running in RAM.
2023-11-29 16:27:14 <ncopa> rpi upgrade maybe little more than trivial this time. the kernel names are being simplified, so no more linux-rpi4 or linux-rpi2
2023-11-29 18:18:08 <Stef> hello
2023-11-29 18:18:29 <Stef> I've got an issue where passwd doesn't work. looking for advise
2023-11-29 18:19:06 <TangentDelta> I was able to get Alpine installed on an EspressoBIN v7. I had to replace the version of U-Boot in flash with a 2022 version meant for Armbian, lol
2023-11-29 18:19:44 <TangentDelta> Stef: The `passwd` command to change a password?
2023-11-29 18:19:57 <TangentDelta> How is it not working?
2023-11-29 18:21:19 <Stef> @TangentDelta yeah. it's like not taking stdin I think. when I run it doesn't let me enter a passwd and just blows through the prompts; almost like hit enter twice without typing a password
2023-11-29 18:22:02 <Stef> passwd: password for stef is unchanged
2023-11-29 18:23:20 <TangentDelta> So you're logged in as the "stef" user, and issuing the `passwd` command with no parameters?
2023-11-29 18:24:28 <Stef> no, I'm running it under root so "passwd stef", if I su to stef and then run passwd I get "passwd: can't change locked password for stef"
2023-11-29 18:25:08 <Stef> the stef account is a new account and adduser had the same reaction for setting the passwd and the passwd command
2023-11-29 18:25:57 <Stef> I'm running this over ssh if that can affect anything
2023-11-29 18:28:23 <TangentDelta> Hmm, I've never encountered something like that before :\
2023-11-29 18:28:32 <Stef> yeah me either
2023-11-29 18:28:56 <Stef> is it possibly a bug in passwd?
2023-11-29 18:29:41 <TangentDelta> Are you sshing in as root?
2023-11-29 18:30:00 <Stef> yeah
2023-11-29 18:30:16 <Stef> I have no other account to login with lol
2023-11-29 18:30:36 <TangentDelta> Can you try logging in locally on the console insead of ssh?
2023-11-29 18:30:54 <Stef> it's a pi and I don't have those hdmi cables it needs
2023-11-29 18:31:13 <Stef> but that would rule out ssh as being the trouble or now
2023-11-29 18:31:16 <Stef> not*
2023-11-29 18:32:06 <TangentDelta> I've seen sort of similar behavior with old machines that had the wrong terminal parameters set.
2023-11-29 18:33:54 <TangentDelta> You could try setting the password this way, too. https://askubuntu.com/questions/80444/how-to-set-user-passwords-using-passwd-without-a-prompt
2023-11-29 18:35:07 <Stef> thanks. I believe that's what I've used in the past but couldn't find it on google
2023-11-29 18:40:49 <Stef> it's ssh for sure. I did "ssh -T" to "Disable pseudo-tty allocation" and it worked. sadly the shell looks really ugly  this way
2023-11-29 18:41:17 <TangentDelta> Hmm that's odd
2023-11-30 00:13:46 <jonesv> I am suspending on LID close using pm-utils as described here: https://wiki.alpinelinux.org/wiki/Suspend_on_LID_close
2023-11-30 00:14:22 <jonesv> But I would like to run swaylock, too. There is this `before-sleep` in sway which seems to work great for that, but... it requires systemd. I have OpenRC.
2023-11-30 00:14:28 <jonesv> How do you guys deal with that?
2023-11-30 01:08:33 <micahrl> Is there a way to request that Alpine load an apkovl from a filesystem with a particular label via the kernel cmdline? I see apkovl= supports Linux device names, but I don't always know the correct Linux device name at boot time.
2023-11-30 02:26:18 <buinb> jonesv, have you tried man zzz? :)
2023-11-30 11:02:28 <rycwo> hello
2023-11-30 11:02:57 <rycwo> I'm having trouble running setup-interfaces on a new laptop... doesn't seem to find my wireless interfaces--any ideas?
2023-11-30 11:06:56 <jonesv> buinb: I am using pm-utils, I hadn't realized that maybe zzz has some documentation about sway's before-sleep without systemd. Let me check :)
2023-11-30 11:09:03 <jonesv> buinb: actually now I remember, I did actually read `man zzz`. How does it solve my problem better than pm-utils here? I still don't know how to have "sleep" to be triggered with OpenRC
2023-11-30 11:09:51 <jonesv> I guess my problem is that I don't know how to setup my alpine such that after some idling time it will `pm-suspend` (or zzz, or whatever) instead of letting the computer run indefinitely
2023-11-30 11:31:42 <rycwo> nevermind found the issue: https://gitlab.alpinelinux.org/alpine/aports/-/issues/14892
2023-11-30 12:39:15 <tiotags> hello, I've received an email that my aports package has to either be moved to community or removed, I was wondering what are the requirements of moving to community ?
2023-11-30 12:40:04 <ikke> tiotags: The package needs to be generally in shape / working properly, not much else
2023-11-30 12:40:41 <tiotags> hello ikke
2023-11-30 12:42:42 <tiotags> I think it works properly, would anybody be kind enough to give a peer review ?
2023-11-30 12:43:54 <ikke> tiotags: Feel free to make an MR
2023-11-30 12:44:56 <tiotags> ok, I have a merge request for an update, should I merge the 2 or make a new mr ?
2023-11-30 12:46:51 <ikke> I guess it's !56324
2023-11-30 12:47:25 <tiotags> yes
2023-11-30 12:47:43 <ikke> I have merged it, you can open a new MR for the move
2023-11-30 12:48:03 <tiotags> thank you
2023-11-30 15:03:47 <jscript> flatpak's work on musl regardless of libc type right?
2023-11-30 15:04:29 <kelvium> yes
2023-11-30 15:05:44 <jscript> Yay
2023-11-30 17:21:16 <somercet1> ah, Firecrash
2023-11-30 22:54:02 <k294> Hi, I'm running 3.10.9, when killing a process I see "punt!". What is the purpose of this?
2023-11-30 22:54:52 <k294> Running strings on nc and kill shows it