2023-05-01 07:46:24 <raspbeguy> Hello, I try to connect to servers through my wireguard VPN using their DNS name on a private domain (home.arpa). The DNS is behind the VPN and had address 192.168.0.1.
2023-05-01 07:48:14 <raspbeguy> Hello, I try to connect to servers through my wireguard VPN using their DNS name on a private domain (home.arpa). The DNS is behind the VPN and had address 192.168.0.1.
2023-05-01 07:48:22 <raspbeguy> When trying e.g. ssh to a machine:
2023-05-01 07:48:22 <raspbeguy> ```
2023-05-01 07:48:22 <raspbeguy> ssh: Could not resolve hostname feed.home.arpa: Name does not resolve
2023-05-01 07:48:22 <raspbeguy> ```
2023-05-01 07:48:22 <raspbeguy> ssh root@feed.home.arpa
2023-05-01 07:48:23 <raspbeguy> but dig is working:
2023-05-01 07:48:23 <raspbeguy> ```
2023-05-01 07:48:25 <raspbeguy> ; <<>> DiG 9.18.13 <<>> feed.home.arpa
2023-05-01 07:48:25 <raspbeguy> dig feed.home.arpa
2023-05-01 07:48:27 <raspbeguy> ;; Got answer:
2023-05-01 07:48:27 <raspbeguy> ;; global options: +cmd
2023-05-01 07:48:29 <raspbeguy> ;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 2888
2023-05-01 07:48:29 <raspbeguy> ;; flags: qr aa rd ra; QUERY: 1, ANSWER: 1, AUTHORITY: 0, ADDITIONAL: 1
2023-05-01 07:48:31 <raspbeguy> ;; OPT PSEUDOSECTION:
2023-05-01 07:48:31 <raspbeguy> ; EDNS: version: 0, flags:; udp: 1232
2023-05-01 07:48:33 <raspbeguy> ;; QUESTION SECTION:
2023-05-01 07:48:33 <raspbeguy> ;feed.home.arpa.			IN	A
2023-05-01 07:48:35 <raspbeguy> ;; ANSWER SECTION:
2023-05-01 07:48:35 <raspbeguy> feed.home.arpa.		3600	IN	A	192.168.0.244
2023-05-01 07:48:37 <raspbeguy> ;; SERVER: 192.168.0.1#53(192.168.0.1) (UDP)
2023-05-01 07:48:37 <raspbeguy> ;; Query time: 73 msec
2023-05-01 07:48:39 <raspbeguy> ;; MSG SIZE  rcvd: 59
2023-05-01 07:48:39 <raspbeguy> ;; WHEN: Mon May 01 09:43:20 CEST 2023
2023-05-01 07:48:41 <raspbeguy> ```
2023-05-01 07:49:18 <raspbeguy> my DNS is present in my /etc/resolv.conf
2023-05-01 08:44:19 <Lord> this is probably related to the musl internal dns resolver which is quite special
2023-05-01 09:57:40 <quinq> raspbeguy, can all your servers in /etc/resolv.conf resolv home.arpa?
2023-05-01 10:16:51 <WhyNotHugo> If multiple repositories have the same pacakge (same pkgver and pkgrel), which one is installed? Is the order of repositories relevant?
2023-05-01 10:36:16 <quinq> WhyNotHugo, I think that you can find the answer to your question in apk-world(5)
2023-05-01 10:36:59 <quinq> I says “Specifying a package name with no other modifiers will install the latest version of that package from the first repository in which it is available.”
2023-05-01 10:39:02 <quinq> I have an issue with fstab
2023-05-01 10:40:22 <quinq> I added a line for an sshfs mount, in the form: user@host:/a/path /media/mountpoint fuse.sshfs various,options 0 0
2023-05-01 10:40:47 <quinq> But when I try to mount that (with $ mount /media/mountpoint), I get a message error:
2023-05-01 10:41:10 <quinq> mount: /media/mountpoint: bad option; for several filesystems (e.g. nfs, cifs) you might need a /sbin/mount.<type> helper program.
2023-05-01 10:41:42 <quinq> I can see in the path that there are: mount.fuse.sshfs  mount.fuse3       mount.sshfs
2023-05-01 10:41:53 <quinq> First one should match
2023-05-01 10:42:07 <quinq> Is there something I'm missing?
2023-05-01 10:43:13 <quinq> Or maybe that's an issue with mount and how it parses file-types/helpers
2023-05-01 10:45:54 <quinq> (note that just specifying sshfs works, I'm just curious about that, as the same line works on some other distros)
2023-05-01 11:35:11 <WhyNotHugo> quinq: Thanks! I'd only checked apk-repositories.
2023-05-01 11:36:39 <quinq> De nada ;)
2023-05-01 12:20:55 <raspbeguy> quinq: yes
2023-05-01 12:21:16 <raspbeguy> Wait, no, I misunderstood your question 
2023-05-01 12:21:46 <raspbeguy> Only one of them, which was added by networkmanager when connecting to the VPN
2023-05-01 12:21:54 <quinq> Most likely you get an answer first from one of those server that can't resolv feed.home.arpa
2023-05-01 12:23:10 <raspbeguy> Probably. But why doesn't it query all of them?
2023-05-01 12:23:24 <quinq> If you actually need to send some requests to some specific servers instead of global DNS, you should use and intermediary resolver made for that
2023-05-01 12:23:33 <quinq> That's the point, it does query all of them
2023-05-01 12:23:47 <quinq> But they don't all answer the same thing
2023-05-01 12:23:58 <quinq> And the faster to answer is one that says that this name doesn't exist
2023-05-01 12:24:51 <quinq> I mean that's my supposition, I don't actually know your setup :)
2023-05-01 12:30:17 <raspbeguy> Hum
2023-05-01 12:31:54 <raspbeguy> When my VPN is up I have 2 dns in my resolv.conf, the one in my man and the one in my vpn
2023-05-01 12:32:17 <raspbeguy> s/man/lan/
2023-05-01 16:44:55 <somercet1> anyone rec any TUI file managers? mc, xplr, nnn, &c?
2023-05-01 16:50:25 <psykose> WhyNotHugo: on apk master i don't think that behaviour is guaranteed anymore (first repo = first to get package)
2023-05-01 16:50:32 <psykose> i don't remember when the new behaviour would be
2023-05-01 17:00:30 <consus[m]> raspbeguy: thats because libc resolver does not do that. More conventional systems have systemd-resolved, but on Alpine you probably should do dnsmasq
2023-05-01 17:01:24 <raspbeguy> consus[m]: I used unbound as resolvconf has native support for generating unbound config
2023-05-01 17:05:17 <consus[m]> Neat
2023-05-01 17:11:37 <quinq> somercet1, not really a file manager, but ncdu is handy too
2023-05-01 17:12:44 <somercet1> thx, will look at. Checking out nnn now
2023-05-01 19:37:49 <stonks3141> somercet1: I use nnn, it's pretty good
2023-05-01 20:54:58 <atka> any known blockers preventing pure data from being packaged? kinda surprised it's not in the repo
2023-05-01 20:55:30 <atka> i'll see if I can get it to compile this weekend
2023-05-01 21:00:18 <quinq> What's “prue data”?
2023-05-01 21:00:43 <atka> a real-time graphical programming environment for audio, video, and graphical processing
2023-05-01 21:01:23 <atka> https://puredata.info
2023-05-01 21:01:27 <quinq> thanks :)
2023-05-01 21:48:32 <quinq> o/ atka 
2023-05-01 21:49:07 <atka> \o
2023-05-01 22:29:56 <Saijin_Naib> That sounds amazing! 
2023-05-01 22:30:54 <atka> Saijin_Naib: pure data?
2023-05-01 22:31:13 <Saijin_Naib> Yeah 
2023-05-01 22:32:50 <atka> https://www.youtube.com/user/acreil/videos
2023-05-01 22:33:16 <atka> I just started using it, I've been using hardware modular synths for a long time though
2023-05-01 22:33:32 <atka> the above video has some cool examples of what can be done audio wise
2023-05-01 22:33:42 <atka> s/video/channel
2023-05-01 22:39:30 <Saijin_Naib> I'm not near good enough for anything like this, about as far as I've gotten is poking about in SonicPi, LSDJ, and a bit more productive in FRACT: OSC, which has a pretty friendly DAW as a reward for beating the game 
2023-05-01 22:44:18 <atka> baby steps, simon hutchinson has really accessible tutorials on pure data
2023-05-01 22:44:21 <atka> https://www.youtube.com/@SimonHutchinson
2023-05-02 01:05:41 <Liver_K> Ohai atka 
2023-05-02 01:06:08 <atka> Liver_K: what are you doing here?!
2023-05-02 01:06:40 <Liver_K> Lol I've been here since March 31
2023-05-02 01:06:53 <Liver_K> Funny seeing you too though XD
2023-05-02 01:07:15 <Liver_K> atka: You an expert on alpine?
2023-05-02 01:07:32 <atka> I dunno about expert but I've been using it since 2015
2023-05-02 01:07:49 <Liver_K> Oh ofc you mentioned you use alpine before didn't you
2023-05-02 01:10:02 <Liver_K> Well I am so tantalizingly close to installing it instead of win10 now, just have to get everything where it should be perfectly before I start
2023-05-02 01:46:00 <atka> Liver_K: jumping right in to alpine eh
2023-05-02 01:46:15 <Liver_K> Oh no no
2023-05-02 01:46:29 <Liver_K> Much preparations lol
2023-05-02 01:47:12 <atka> you going full desktop, laptop replacement?
2023-05-02 01:48:30 <Liver_K> Well I am sacrificing this desktop yeah, but it's not going to be another desktop
2023-05-02 01:48:37 <Liver_K> It is going to be another server :)
2023-05-02 01:51:53 <atka> yeah, works well for both
2023-05-02 01:53:05 <Liver_K> atka: Do you have any diskless setups?
2023-05-02 01:53:38 <atka> yeah, just on rpi though, used to have one on a mac mini server
2023-05-02 01:54:43 <Liver_K> So diskless is ok for server setups?
2023-05-02 01:56:04 <ipcloud> \q
2023-05-02 01:59:22 <atka> Liver_K: depends on the use case but yeah I'd say so
2023-05-02 01:59:56 <atka> I've mainly used them for a NAS, running a btrfs raid array and hosting a few containers with podman
2023-05-02 02:00:12 <atka> I use guix for that now though
2023-05-02 02:01:06 <atka> I have a few servers, with alpine and containers running nvr software + wireguard
2023-05-02 02:01:10 <atka> but not diskless
2023-05-02 02:01:23 <Liver_K> Well... it is planned to be a general purpose personal server, so, among other things, BNC, NAS, HTTPD (lightweight), etc.
2023-05-02 02:03:04 <Liver_K> Basically as much as I can run functionally on this crappy hardware
2023-05-02 02:04:54 <atka> well, alpine will run better than win10 that's for sure
2023-05-02 02:05:55 <Liver_K> Yeah definitely lmao, I am not currently running any daemons on it right now
2023-05-02 02:06:17 <Liver_K> Perhaps my eyes are bigger than my stomach, but I do have plans for this hunk o metal :)
2023-05-02 02:08:34 <atka> nah, you can do it
2023-05-02 02:12:49 <Liver_K> I was briefly thinking about running it as a tor node too, since I have plenty of networking to spare
2023-05-02 02:13:15 <Liver_K> Eh, so many daemons and so little knowledge XD
2023-05-02 02:32:05 <Saijin_Naib> If my ext4 volume trashed itself due to a power outage but I can mount and chroot to it, how can I force apk to fix all when it throws an error about the state of the apk database? 
2023-05-02 02:49:11 <psykose> you mean other than `apk fix \*`
2023-05-02 02:51:34 <Saijin_Naib> Yep, can't even do that 
2023-05-02 02:51:43 <Saijin_Naib> It complains about the state of the apk database 
2023-05-02 02:53:31 <psykose> not actually sure except for deleting it and making a new one
2023-05-02 02:54:18 <psykose> /lib/apk/db is the db and `apk add --initdb ..` can make a new one i guess
2023-05-02 02:54:24 <psykose> not sure what it would lose you, if anything
2023-05-02 02:54:30 <psykose> the world file is all that generally matters
2023-05-02 02:57:05 <Saijin_Naib> Hmm okay, worth a shot, thank you
2023-05-02 02:57:16 <Saijin_Naib> Do I need to specify any path or anything for initdb
2023-05-02 03:00:34 <psykose> nope
2023-05-02 03:00:48 <psykose> defaults to / for root, otherwise you normally use it with --root etc
2023-05-02 03:03:38 <Saijin_Naib> Fuck, that wipes world
2023-05-02 03:03:49 <Saijin_Naib> Fuckity fuck fuck 
2023-05-02 03:04:05 <Saijin_Naib> Apk works now though, so thank you, haha
2023-05-02 03:05:51 <psykose> ah does it
2023-05-02 03:05:52 <psykose> oops
2023-05-02 03:05:58 <psykose> should've told you to copy it first
2023-05-02 03:06:12 <psykose> not great
2023-05-02 03:09:33 <Saijin_Naib> Yeah, was not expecting that. Was expecting to cp world to /lib/apk/db/installed
2023-05-02 03:10:25 <psykose> nah, you never touch things inside db yourself
2023-05-02 03:12:16 <Saijin_Naib> Ahh, so I just needed a copy of world... It'd be nice if initdb did that for you, or warned before you shot your foot off
2023-05-02 03:17:34 <atka> who needs feet anyway
2023-05-02 03:42:43 <Liver_K> 
ACTION raises hand

2023-05-02 03:45:40 <abby> 
ACTION raises foot

2023-05-02 03:47:19 <atka> 
ACTION pulls out footgun

2023-05-02 03:59:00 <Liver_K> Nah kidding I don't need no foots. Got twos right here *looks down*
2023-05-02 03:59:14 <Liver_K> s/foots/feetses
2023-05-02 05:58:26 <frag> can grep search case insensitive using regex (and not the -i option)?
2023-05-02 05:59:50 <grawity> pcre supports (?i) to switch the pattern to case-insensitive mode, I don't think posix regexen have any equivalent short of [Hh][Ee][Ll][Ll][Oo]
2023-05-02 05:59:58 <grawity> which I believe is what -i does internally anyway
2023-05-02 06:00:15 <frag> hehe
2023-05-02 06:01:13 <grawity> so if you can use PCRE, then `grep -P '(?i)hello'`
2023-05-02 06:03:35 <frag> grawity: hmm weird the other regex dont have something similar, maybe i have to look more into pcre, ty!
2023-05-02 15:34:47 <pdw> I've just spun up an Alpine Linux VM (not a container) inside LXD:
2023-05-02 15:34:49 <pdw> lxc init images:alpine/edge alp --vm ; lxc config set alp security.secureboot=false ; lxc start alp ; lxc console alp
2023-05-02 15:35:00 <pdw> 1) How should I log in to this?  The obvious things (root/empty, alpine/alpine, etc.) don't work at the 'login:' prompt, and I don't think this is the "/cloud" image.
2023-05-02 15:35:10 <pdw> 2) Is there somewhere I could have looked up that answer to 1) for myself?  I spent half an hour or so searching online, but didn't find anything.
2023-05-02 15:51:22 <kwadroke[m]> pdw: perhaps you need to run a cloud-init/tiny-cloud setup to create your user & pass?  https://discuss.linuxcontainers.org/t/running-virtual-machines-with-lxd-4-0/7519
2023-05-02 15:52:27 <kwadroke[m]> I haven't tried using VMs with LXD/LXC as of yet. Didn't know you can do that until now
2023-05-02 16:26:11 <pdw> Thanks - I had not been aware of that thread.  Very helpful, in particular this comment which clearly states that there are no valid username/password combinations on their fresh images: https://discuss.linuxcontainers.org/t/running-virtual-machines-with-lxd-4-0/7519/111
2023-05-02 16:27:21 <pdw> If you've not seen VMs on LXD before (I hadn't before this week), you might find the command 'lxc console <VM> -t=vga' helpful.
2023-05-02 16:27:26 <Saijin_Naib> @psykose, I'm in and working, so thank you for the rescue! 
2023-05-02 16:27:35 <pdw> It opens up the GUI window onto the VM.
2023-05-02 16:28:17 <pdw> Thanks for the help, kwadroke[m].
2023-05-02 16:31:18 <psykose> Saijin_Naib: :)
2023-05-02 16:34:05 <Saijin_Naib> I had some stuff backed up from my dad only on there, so I was freaking out that I might not be able to get it 
2023-05-02 16:34:30 <Saijin_Naib> Now that I'm collected, I'm moving it all to my NAS 🤣
2023-05-02 17:00:18 <bbbhltz> Hi. I am having an issue viewing pdf files. I think it has something to do with poppler. Whenever I scroll through a pdf I am logged out. How would I go about finding the cause of the problem?
2023-05-02 17:03:06 <lopid> start x from console and see if there are any messages
2023-05-02 17:08:19 <psykose> sounds like what you view the pdf in crashes the wm/whatever
2023-05-02 17:15:31 <bbbhltz> Yep. I used zathura. Errors about lxqt.global_key_shortcuts.native
2023-05-02 17:25:03 <somercet1> bbbhltz, use mupdf with zathura. Better rendering and more stable, I find.
2023-05-02 17:25:26 <somercet1> zathura-pdf-mupdf
2023-05-02 17:25:58 <bbbhltz> That is what I switched to
2023-05-02 17:26:08 <psykose> zathura's mupdf backend is atrociously garbage
2023-05-02 17:26:25 <psykose> unless you have 100gb of memory i guess
2023-05-02 17:27:02 <psykose> they did slightly improve it last version
2023-05-02 17:27:55 <somercet1> mupdf is a lot smaller without embedded fonts. I think 15GB of libmupdf is embedded PDF-compatible fonts.
2023-05-02 17:28:17 <somercet1> zathura/mupdf is stable for me.
2023-05-02 17:28:37 <bbbhltz> I prefer poppler which is why I am trying to sort it out. It is probably a user problem because it only started doing this a few days ago
2023-05-02 17:57:34 <frag> when using 'tail -f' but not 'tail' adding |cat makes it display fewer lines (at least initialy), why? :S   tail -f $out|tr -d '\r'|sed -e 's/://1' -e 's/ \(.*\)!.*PRIVMSG \([^:]*\) :/ \2 \1: /1' -e '/ PING irc/d'|cat -A
2023-05-02 17:59:16 <frag> seems to happen even if i use something else than |cat (eg |grep -E 'somethig|$')
2023-05-02 18:31:05 <durrendal> shot in the dark, but does anyone here have any clue what Instance type "virtual-machine" is not supported on this server: QEMU failed to run feature checks means when trying to launch a vm with lxd? The error is somewhat opaque as to what checks are being run
2023-05-02 18:45:43 <minimal> insufficient info provided, online search shows several issues talking about this error message
2023-05-02 18:48:26 <WhyNotHugo> psykose: Can I prioritise packages from a given repo?
2023-05-02 18:48:35 <psykose> don't think so
2023-05-02 18:48:50 <psykose> you can force them to be from one repo by tagging a repo but that's about it
2023-05-02 18:49:17 <WhyNotHugo> Mostly, sometimes I build something that I push to aports, but I do want apk to download the replacement package when it's built upstream.
2023-05-02 18:49:36 <WhyNotHugo> I.e.: de-prioritise my local repo.
2023-05-02 18:50:04 <WhyNotHugo> Unrelated: I keep seeing a symlink in /boot/boot that points to .
2023-05-02 18:53:36 <minimal> WhyNotHugo: you mean a "boot" symlink in "/boot/" that points to "." ?
2023-05-02 18:54:41 <WhyNotHugo> `file /boot/boot`  prints `/boot/boot: symbolic link to .`
2023-05-02 18:57:30 <durrendal> minimal: sure I've spent a few days chasing this one and haven't gotten to the bottom of it. When running lxc init images:alpine/edge vmname --vm, that error is thrown. I've got 2 different alpine lxd clusters which throw the same issue setup the same way.
2023-05-02 18:57:44 <durrendal> https://lambdacreate.com/paste/sipo-world <- world from one of said clusters
2023-05-02 18:58:29 <durrendal> qemu-system-x86_64 runs just fine manually on the node, I can happily launch vm's manually, just not through lxd.
2023-05-02 18:59:38 <durrendal> it feels like I'm missing something. Worked fine before qemu got upgraded to 7.2.0, then I ended up with netdev monitor issues which were addressed in qemu 8.0.0, but haven't been able to launch anything since then. 
2023-05-02 19:00:19 <minimal> WhyNotHugo: yes "/boot/boot" is a symlink, not "I keep seeing a symlink in /boot/boot" which you said which indicates that /boot/boot was a directory which contained a symlink, rather than being a symlink itself
2023-05-02 19:00:52 <minimal> anyway, there's a reason for that symlink (don't remember exactly)
2023-05-02 19:01:12 <WhyNotHugo> I see it in this path/location. Not INSIDE this path. I see your point, but I wasn't sure how to express it non-ambiguously.
2023-05-02 19:01:56 <psykose> so if you have boot on a separate partition (and it's not fat32 and supports symlinks), /boot/.. resolves the same as when it's mounted to /boot
2023-05-02 19:02:10 <minimal> "/boot/boot is a symlink, why?"
2023-05-02 19:02:41 <WhyNotHugo> /boot has no special meaning on my setup. E.g.: it's not the boot partition, there is no boot partition, only an ESP in /efi.
2023-05-02 19:03:07 <psykose> something about bootloaders starting at boot partition and then doing /boot/vmlinuz, but then that is different from /boot/vmlinuz when mounted later
2023-05-02 19:09:08 <minimal> durrendal: probably best asking on the lxc channel on libera or raising a LXD issue, sounds specific to LXD
2023-05-02 19:16:35 <durrendal> fair enough
2023-05-02 19:47:17 <WhyNotHugo> Oh, this is why: https://gitlab.alpinelinux.org/alpine/aports/-/blob/master/main/mkinitfs/mkinitfs.trigger#L29-33
2023-05-02 20:58:32 <Saijin_Naib> Are we using dbus-broker already? 
2023-05-02 21:04:00 <psykose> no
2023-05-02 21:18:50 <bbbhltz> Re: scrolling pdf leads to crash... I messed around with mupdf and managed to cause the same problem. The errors I am getting are about libinput and touchjump. Then X crashes.
2023-05-02 21:19:31 <lopid> to rule out the obvious, it's not a corrupt pdf somehow?
2023-05-02 21:26:36 <bbbhltz> It would mean that I only have corrupt pdf files, and I have tried pdf files from different sources
2023-05-02 21:27:58 <psykose> having 'corrupt pdf files' that crash x is still an x bug
2023-05-02 21:34:55 <bbbhltz> Makes sense to me. Still a first though. 
2023-05-03 01:55:38 <wolf> goodnight. minidlna is horrible. If you install ffmpe5, mp4 appears, if you install ffmpeg5 some mp4 appears and mkv appears and mp3 does not appear if you remove ffmpeg5 only mp4 appears, horrible...
2023-05-03 03:34:52 <detly> I'm trying to find where the "--no-cache" option for apk is documented - I can't find it in the user handbook or the wiki entries for apk
2023-05-03 03:35:35 <detly> and also, is there an environment variable that turns that on?
2023-05-03 03:55:32 <psykose> man 8 apk says the same thing as add --help
2023-05-03 03:55:35 <psykose> and there isn't an env var no
2023-05-03 04:22:09 <detly> @psykose cheers, I didn't realise --help could go with subcommands
2023-05-03 04:22:25 <detly> and I don't have the man pages installed in my images typically
2023-05-03 04:49:24 <DeSantis> Don't let Micky Mouse diddle your little boy!  Mickey was busted by Chris Hansen on To Catch a Predator trying to buttscrew a 13 y/o boy in a sting operation!  Read all about it here:  https://pastebin.com/edvNf8kn   Ron Desantis 2024
2023-05-03 08:08:12 <cockliuser[m]> Why is this shit getting spammed everywhere 
2023-05-03 08:08:26 <cockliuser[m]> Is the FBI targeting libera now or something 
2023-05-03 08:11:35 <psykose> it's a test to see who interacts with spam instead of ignoring it
2023-05-03 09:17:17 <dwfreed> psykose: ^5
2023-05-03 09:26:40 <kode54> grats to them, my client inlines the pastebin
2023-05-03 09:30:38 <lopid> thus automatically giving them the clicks
2023-05-03 09:51:07 <Ermine> Previous times spammer just pasted contents here
2023-05-03 10:06:20 <quinq> kode54, what could go wrong anyway?
2023-05-03 10:06:37 <kode54> I mean, it's pastebin
2023-05-03 10:06:45 <kode54> they don't exactly inline things to execute
2023-05-03 10:06:58 <kode54> I don't even think they publish hit counts
2023-05-03 10:07:38 <kode54> I mean, other than that I have to see a few lines of some idiotic AI generated script
2023-05-03 17:56:07 <aep> whats the recommended way to use cgroups? this doesnt work, cgroups arent mounted at all https://wiki.gentoo.org/wiki/OpenRC/CGroups
2023-05-03 17:56:49 <aep> unfortunately googling for "alpine cgroups" isnt helpful because that is full of docker related things
2023-05-03 18:08:49 <Piraty> aep_: https://wiki.alpinelinux.org/w/index.php?search=cgroups&title=Special%3ASearch&wprov=acrw1
2023-05-03 18:19:59 <aep_> these also point to editing rc.conf which doesnt work
2023-05-03 18:21:57 <aep_> ah its rc-service cgroups. thanks
2023-05-03 19:30:38 <atpcsd> hey all.. /etc/network/interfaces.. I have a single physical interface.. and I am trying to get two vlans running on it.. ip commands work just fine, vlans work just fine.. I cannot figure out what I am doing wrong in the file for system networking to bring it up.. https://pastebin.com/rgcTF26W
2023-05-03 19:32:13 <atpcsd> I also tried the auto vlan172.. made no difference.. vlan172 did not come up, and the second ip on the eth0.69 vlan is not created either.. 
2023-05-03 19:33:51 <ikke> just setting a subinterface does not set vlan encapsulation
2023-05-03 19:34:35 <ikke> atpcsd: example that I use: https://tpaste.us/MR4D 
2023-05-03 19:34:59 <ikke> atpcsd: note that alpine uses ifupdown-ng
2023-05-03 19:35:14 <ikke> while it still accepts the old syntax, it's probably better to use the native syntax
2023-05-03 19:35:15 <atpcsd> https://pastebin.com/5GQBrahT
2023-05-03 19:35:30 <atpcsd> oh.. I got syntax at the wiki.. :P ;) 
2023-05-03 19:35:41 <ikke> yeah, probably still old
2023-05-03 19:35:57 <atpcsd> https://wiki.alpinelinux.org/wiki/Configure_Networking and https://wiki.alpinelinux.org/wiki/Vlan
2023-05-03 19:37:06 <ikke> that relied on the vlan package that automatically assigns the vlan id based on the subinterface name I suppose
2023-05-03 19:37:06 <atpcsd> I didn't have the vlan_id in my other attempts.. 
2023-05-03 19:37:21 <atpcsd> and then the address/cidr is valid? and gateway.. 
2023-05-03 19:38:04 <ikke> yes
2023-05-03 19:38:10 <atpcsd> auto eth0.69 iface eth0.69 address 1.2.3.4/20 gateway 1.2.15.254 vlan-raw-device eth0 vlan_id 69 
2023-05-03 19:38:48 <atpcsd> then iface eth0.69 inet static address 1.2.3.5/20 (for the secondary)? 
2023-05-03 19:39:09 <atpcsd> eth0.69 does work currently.. it's the only thing that does :/ 
2023-05-03 19:39:16 <atpcsd> but not the secondary.. 
2023-05-03 19:40:37 <ikke> You don't need seperate stanzas
2023-05-03 19:40:47 <ikke> add 2 address statemetns
2023-05-03 19:41:36 <atpcsd> ah.. (ty) 
2023-05-03 19:44:46 <atpcsd> https://pastebin.com/7Nx6xDBs
2023-05-03 19:45:30 <atpcsd> not sure if all that makes sense.. but seems to error out on the config.. 
2023-05-03 19:46:01 <ikke> Just to be sure, what version of alpine is this?
2023-05-03 19:46:12 <atpcsd> 3.17.3 
2023-05-03 19:46:45 <atpcsd> I also added the vlan pkg (fwiw) 
2023-05-03 19:47:30 <ikke> ah, that conflicts with ifupdown-ng
2023-05-03 19:47:38 <ikke> You don't need it for ifupdown-ng
2023-05-03 19:48:18 <atpcsd> I have busybox-ifupdown 
2023-05-03 19:48:36 <ikke> What I suggested assumes ifupdown-ng
2023-05-03 19:48:44 <atpcsd> (np) 
2023-05-03 19:48:54 <ikke> and it's the default for alpine since 3.14 or so
2023-05-03 19:49:28 <atpcsd> I built this box January of this year.. iirc 
2023-05-03 19:50:14 <ikke> installing vlan would cause ifupdown-ng to be deinstalled
2023-05-03 19:51:17 <atpcsd> (yes looks like it.. ) 
2023-05-03 19:51:53 <atpcsd> https://pastebin.com/bfCP5tv7
2023-05-03 19:53:00 <ikke> with ifupdown-ng, the 'inet static' part is redundant
2023-05-03 19:53:11 <ikke> (ie, it does nothing)
2023-05-03 19:53:41 <atpcsd> bingo that looks like it.. 
2023-05-03 19:53:46 <atpcsd> me adding that vlan package
2023-05-03 19:54:05 <ikke> I only know because I ran into it myself
2023-05-03 19:57:47 <atpcsd> soo.. where could I have read this myself? a man page? (thank you for this.. ) I have *tons* of ip add/route/lines.. that I wish I could have gotten to work
2023-05-03 19:59:07 <ikke> Yes, there are man pages
2023-05-03 19:59:10 <ikke> or https://github.com/ifupdown-ng/ifupdown-ng/blob/main/doc/interfaces.scd
2023-05-03 19:59:20 <ikke> https://github.com/ifupdown-ng/ifupdown-ng/blob/main/doc/ADMIN-GUIDE.md
2023-05-03 20:07:07 <atpcsd> . o ( no idea this existed.. ) - thank you.. 
2023-05-03 20:21:26 <ikke> atpcsd: no problem
2023-05-03 20:38:00 <blt[m]> I see https://pkgs.alpinelinux.org/package/edge/main/x86_64/musl 1.2.4 in edge and this is fixing a long term issue with DNS which is much appreciated : https://www.openwall.com/lists/musl/2023/05/02/1 . would it make it to 3.18 as we are in a freeze period? I feel it would deserve to include it...
2023-05-03 20:40:08 <pj> 3.18 is made from edge so yes?
2023-05-03 20:43:29 <ikke> indeed, everyting in edge up until release will be part of the next release
2023-05-03 20:44:57 <atpcsd> blt[m]: what resolver is being discussed? in openwall? 
2023-05-03 20:45:17 <ikke> musl 1.2.4 adds fallback to tcp for dns
2023-05-03 20:45:59 <atpcsd> like udp > 512 which then moves to tcp.. 
2023-05-03 20:49:19 <pj> there is this thing called "reading"
2023-05-03 20:49:26 <pj> which can grant knowledge
2023-05-03 20:49:51 <abby> how does one learn such magick
2023-05-03 20:50:31 <pj> usually by going to like kindergarten or sumthin
2023-05-03 20:51:02 <pj> dunno, don't care, ask chatgpt
2023-05-03 20:52:03 <atpcsd> I was trying to find out what the failure was.. as I have dnsdist for a few thousand chromebooks.. and was looking at tcpdump tcp port 53 and trying to figure out what the failure is.. 
2023-05-03 20:53:05 <atpcsd> reading might come with caring of what was being said.. but you understand the problem so no need. 
2023-05-03 20:54:40 <pj> if you don't care about conversation, why are you talking
2023-05-03 20:54:57 <pj> as to the problem, it's first sentence of the mail
2023-05-03 21:00:20 <atpcsd> yes.. that's what I was trying to understand from the reading.. I am sorry to have intruded; O great keeper of the silence. 
2023-05-03 21:00:46 <atpcsd> thank you for helping ikke maybe someone else will update wiki and keep more people out of here.. 
2023-05-03 21:00:59 <blt[m]> <ikke> "indeed, everyting in edge up..." <- Thanks and good news, I was not sure as we are close to the release I thought it was almost frozen
2023-05-03 21:01:16 <atpcsd> pj this could be you.. otherwise more people might keep coming in here.. 
2023-05-03 21:01:24 <atpcsd> thanks again. 
2023-05-03 21:01:56 <ikke> blt[m]: what is frozen is the toolchain
2023-05-03 21:02:28 <ikke> but musl was updated before the freeze
2023-05-03 21:05:31 <psykose> and after
2023-05-03 21:05:37 <ikke> heh
2023-05-03 21:26:25 <atpcsd> found a large request.. and you can see it.. 
2023-05-03 21:26:28 <atpcsd>  drill -t txt locations.publicdns.goog
2023-05-03 21:26:55 <atpcsd> dig +besteffort doesn't do any better.. but it does fail partially.. 
2023-05-03 21:27:23 <quinq> wow
2023-05-03 21:27:58 <atpcsd> write that down for the readers pjs and point them at the first sentence.. :P 
2023-05-03 21:39:46 <pj> who is pjs
2023-05-03 21:40:35 <quinq> pyjamas I suppose
2023-05-04 08:29:34 <Ermine> Apparently bash completions do not work for me at all. bash-completion is installed
2023-05-04 08:30:12 <lopid> wfm
2023-05-04 08:33:30 <Ermine> Cool
2023-05-04 08:33:53 <psykose> does it work in `bash -li`
2023-05-04 08:34:23 <Ermine> No
2023-05-04 08:37:27 <Ermine> complete prints nothing
2023-05-04 08:38:12 <lopid> do you have BASH_COMPLETION_VERSINFO?
2023-05-04 08:38:49 <Ermine> Also no
2023-05-04 08:39:10 <lopid> over to /etc/profile.d/bash_completion.sh then
2023-05-04 08:39:33 <lopid> put something in it to see if it's even sourced
2023-05-04 08:41:30 <Ermine> Oh, that file was missing somewhy. Thank you!
2023-05-04 08:42:20 <psykose> it's /etc/bash/bash_completion.sh and it's sourced from /etc/bash/bashrc
2023-05-04 08:42:25 <psykose> which is sourced by bash -l
2023-05-04 08:42:57 <lopid> not according to `apk info -L bash-completion | grep bash_completion.sh` here
2023-05-04 08:43:24 <psykose> changed between 3.17 and edge
2023-05-04 08:43:34 <lopid> :/
2023-05-04 08:46:39 <psykose> the issue seems to be that /etc/bash/bashrc is not actually sourced
2023-05-04 08:48:37 <psykose> that did use to work tho
2023-05-04 08:49:29 <psykose> ah
2023-05-04 08:50:07 <psykose> the fuck
2023-05-04 08:50:10 <psykose> it sources it for bash -i
2023-05-04 08:50:14 <psykose> and not for -l or -li
2023-05-04 08:50:57 <psykose> i guess it somewhat makes sense but
2023-05-04 08:51:29 <psykose> even just `bash` sources it
2023-05-04 08:51:40 <psykose> Ermine: do you spawn your shell with -l every time?
2023-05-04 08:52:00 <lopid> (and are you on edge)
2023-05-04 08:52:01 <Ermine> I think I don't
2023-05-04 08:52:10 <Ermine> And I'm on edge
2023-05-04 08:52:11 <psykose> it sounds like you do
2023-05-04 08:52:20 <Ermine> if tmux does
2023-05-04 08:52:50 <Ermine> Ah, it spawns -bash
2023-05-04 08:53:31 <psykose> that was the version that is -l iirc or something
2023-05-04 08:56:07 <psykose> yeah even just default tmux works with /bin/bash as configured login shell
2023-05-04 08:56:12 <psykose> i guess you set it to -bash somewhere
2023-05-04 09:04:47 <Ermine> Hm, I don't have any tmux conf files
2023-05-04 09:23:38 <psykose> what's your login shell
2023-05-04 09:25:54 <Ermine> bash too. It's in WSL
2023-05-04 09:42:18 <psykose> well it works in any configuration i try at all except bash -l, and nothing does bash -l anywhere by default when spawning shells
2023-05-04 09:42:35 <psykose> so i'm guessing there is some weird wsl shit with tmux
2023-05-04 09:45:02 <dwfreed> '-bash' is the same as 'bash -l'
2023-05-04 09:45:47 <dwfreed> historical peculiarities of how login works
2023-05-04 09:47:50 <dwfreed> the bash manpage notes that it only reads profile files when a login shell, and only reads bashrc files when a non-login shell; generally the profile files source the bashrc files when they detect that they're being sourced from bash
2023-05-04 09:49:08 <dwfreed> eg, debian's /etc/profile contains an if $BASH && $BASH != /bin/sh, source /etc/bash.bashrc
2023-05-04 09:49:17 <dwfreed> (which is where system bashrc lives there)
2023-05-04 09:56:17 <psykose> good point
2023-05-04 11:17:30 <psykose> Ermine: try baselayout 3.4.3
2023-05-04 11:17:50 <psykose> login bash should source it too since i did the same as above
2023-05-04 12:33:08 <WhyNotHugo> Firefox on Alpine has a built-in US dictionary. How can I get rid of it?
2023-05-04 12:33:23 <WhyNotHugo> It's not listed in its settings (only manually added ones are), and /usr/share/hunspell does not exist.
2023-05-04 13:08:50 <Ermine> weechat: that worked, thank you!
2023-05-04 13:17:37 <Guest495> Hi, I am having a problem trying to compile a C++ program with the zig C++ compiler. It is just a hello_world program that includes <iostream>, but I get: fatal error: linux/futex.h file not found. I understand that this is a kernel file that is missing. Is there some package that I must add in alpine in order to have this file?
2023-05-04 13:18:05 <Ermine> Try musl-dev
2023-05-04 13:18:21 <Guest495> Thanks, I'll try it
2023-05-04 13:19:09 <psykose> linux-headers
2023-05-04 13:20:15 <Guest495> Thanks! linux-headers worked
2023-05-04 13:20:33 <Ermine> psykose is clever, be like psykose
2023-05-04 13:30:13 <Quillith> I was not expecting Alpine Linux to be on reddit's frontpage but hey pleasant surprise! Awesome to see Alpine out in the wild https://old.reddit.com/r/ProgrammerHumor/comments/137e3qf/get_gnu_gnud/
2023-05-04 14:33:57 <frag> the up and coming linux distro
2023-05-04 14:48:10 <uniqNuuid> devilish gleam, ready to mansplain // lmao
2023-05-04 14:50:12 <uniqNuuid> i think he's refering Richard Stallman
2023-05-04 14:52:30 <uniqNuuid> or Richie Guix?
2023-05-04 15:14:17 <ncopa> alpine 3.18 rc3 was just pushed
2023-05-04 15:14:53 <solene> I can't run a downloaded pre-built binary for a program, is it a musl thing? ldd is supsicious https://pbot.rmdir.de/T16WsEslwnHsYtAIYe1Hfw
2023-05-04 15:15:20 <ncopa> libc.so.6 => /lib64/ld-linux-x86-64.so.2 (0x7f8b8a60b000)
2023-05-04 15:15:28 <ncopa> it is linked to gnu libc
2023-05-04 15:15:48 <solene> and, is that bad?
2023-05-04 15:15:58 <ncopa> if you run alpine, yes
2023-05-04 15:16:29 <solene> hmm, I'm trying to package it but I just get a segfault at runtime, and I'm not more lucky with the pre-built binary :/ I'll be sad
2023-05-04 15:16:44 <ncopa> you cannot run binaries compiled against gnu libc just like you cannot run windows executables
2023-05-04 15:16:51 <solene> make sense
2023-05-04 15:17:04 <ncopa> but like you can install wine (with the windows dlls) you can also try with gcompat
2023-05-04 15:17:13 <ncopa> it may or may not work
2023-05-04 15:18:10 <solene> I'll note this one down, doesn't help for musikcube (libmusikcore.so: fcntl64: symbol not found) but this may be handy one day
2023-05-04 15:21:31 <ncopa> fcntl64 should be relatively simple to fix in gcompat
2023-05-04 15:22:23 <ncopa> report it here: https://git.adelielinux.org/adelie/gcompat
2023-05-04 18:04:29 <solene> a simple doas.conf as "permit nopass solene as root" allows me to use doas, but ask for a password, is it disabled or not functional on linux?
2023-05-04 18:04:54 <solene> I mean, the "nopass" keyword isn't applied
2023-05-04 18:09:53 <pj> as root makes no sense
2023-05-04 18:10:35 <solene> pj: i took the simplest example
2023-05-04 18:10:48 <solene> raspbeguy just told me on alpine I should look at /etc/doas.d/doas.conf
2023-05-04 18:11:09 <solene> and it's working better when editing the right file, but doas.conf(5) only mentionned /etc/doas.conf
2023-05-04 18:11:21 <ikke> permit nopass solene
2023-05-04 18:12:11 <pj> because /etc/doas.d is stupid alpine-y thing
2023-05-04 18:12:19 <pj> it's patched into doas
2023-05-04 18:12:26 <pj> and /etc/doas.conf doesn't work
2023-05-04 18:13:19 <ikke> It's a compromise
2023-05-04 18:14:24 <pj> it's a stupid behaviour
2023-05-04 18:14:50 <ikke> if doas supported includes, it would not be necessary
2023-05-04 18:15:34 <solene> did you know if there have been an attempt to upstream the changes?
2023-05-04 18:15:45 <solene> s/did/do
2023-05-04 18:15:46 <pj> yes
2023-05-04 18:15:50 <pj> it's not merged
2023-05-04 18:15:57 <pj> since a year or so
2023-05-04 18:16:05 <solene> let me take a look
2023-05-04 18:16:30 <pj> 2 years
2023-05-04 18:16:31 <pj> https://github.com/Duncaen/OpenDoas/pull/71
2023-05-04 18:16:32 <ikke> https://gitlab.alpinelinux.org/alpine/aports/-/blob/master/main/doas/configuration-directory.patch
2023-05-04 18:16:50 <solene> ah right, that's alpine's upstream, I was thinking at OpenBSD
2023-05-04 18:17:14 <solene> we added support for .d directories for a few things, so it may be acceptable for doas as well
2023-05-04 18:17:14 <ikke> The openbsd one does not really work on linux
2023-05-04 18:17:31 <solene> I wonder if opendoas is still in sync with the openbsd one
2023-05-04 18:17:46 <pj> downstream yes
2023-05-04 18:17:47 <pj> upstream no
2023-05-04 18:17:56 <ikke> afaik Duncaen tries to remain close to openbsds doas
2023-05-04 18:18:03 <pj> yes
2023-05-04 18:18:13 <pj> that's why the .d patch got stale and never merged
2023-05-04 18:18:24 <pj> and there was much argument
2023-05-04 18:19:00 <ikke> It would help if openbsd doas starts supporting it
2023-05-04 18:19:36 <solene> I'd be happy to push in favor of that and make the commit if we don't have strong objection from other there ^^'
2023-05-04 18:21:44 <duncaen> not sure if this is actually a good idea
2023-05-04 18:22:29 <duncaen> the ordering of rules is important and with the drop in files its too easy to make mistakes and results in people/packages doing things that shouldn't happen
2023-05-04 18:22:29 <pj> I would rather want to see a better patch
2023-05-04 18:22:35 <pj> better designed* patch
2023-05-04 18:24:09 <duncaen> packages shipping config files like https://git.sr.ht/~mil/sxmo-utils/tree/master/item/configs/doas/sxmo.conf is why its bad
2023-05-04 18:25:05 <pj> no one forces you to install that though?
2023-05-04 18:34:31 <ikke> duncaen: the alternative would be peoplle writing post-install scripts appending config to /etc/doas.conf
2023-05-04 18:35:31 <duncaen> yea i see the use for this for a distribution, the problem is when external packages would start to ship drop in files that are not well understood
2023-05-04 18:41:57 <duncaen> maybe the directory shouldn't behave like usual drop in directories, instead you have one default /etc/doas.conf file and then drop in files that are allowed to be used with a flag that read alternative files from only that known directory
2023-05-04 18:42:43 <duncaen> the services/programs that rely on rules for them specifically would then have to specify the configuration file to use instead of applying all rules
2023-05-04 18:43:07 <ikke> duncaen: the usual way to handle that is to have a way to configure what to include
2023-05-04 18:43:17 <ikke> Then you can decide whether you want to include things or not
2023-05-04 18:45:12 <duncaen> yea that would be better, but then you still end up with many rules that compete against each other and you'll have to understand the whole include chain to know each rule does the right thing
2023-05-04 18:46:32 <duncaen> from looking at the files shipped with packages, the usual pattern seems to be to allow specific commands to be executed as specific user
2023-05-04 18:47:52 <duncaen> those rules being affected by other rules doesn't really make sense
2023-05-04 20:00:30 <minimal> re opendoas, I added a comment to the PR to see if there's any likelyhood of it being merged. I added doas support to Alpine's cloud-init some time ago but can't upstream it until opendoas merges the change
2023-05-04 20:01:23 <minimal> duncaen: ^^^
2023-05-04 20:02:55 <duncaen> but this is exactly the problem
2023-05-04 20:03:12 <duncaen> tools starting to support features that are not enabled by default or even supported by the origianl upstream
2023-05-04 20:03:18 <duncaen> s/support/depend on
2023-05-04 20:03:53 <duncaen> and it seems like people here are also not really happy with the patch
2023-05-04 20:04:35 <minimal> duncaen: I added this support to Alpine's cloud-init as Alpine were moving sudo from "main" to "community" and relying on doas by default and I was asked to add doas support as a result
2023-05-04 20:07:26 <minimal> I could have just appended rules to the /etc/doas.conf file instead of relying on this patch but that seemed messy
2023-05-04 20:09:31 <duncaen> yea seems like a bad idea but this is a bit of a problem, cloud-init seems to support openbsd which has a different doas
2023-05-04 20:09:54 <duncaen> the whole diverting from upstream is really bad for changes like this
2023-05-04 20:10:05 <minimal> duncaen: upstream cloud-init does not support doas AT ALL
2023-05-04 20:10:56 <duncaen> but if it would, and opendoas would divert from openbsd and the other freebsd port we end up with a mess
2023-05-04 20:11:09 <minimal> I would like to upstream my doas functionality to cloud-init, but this fragmentation is the issue
2023-05-04 20:12:33 <ikke> yes, it's less than ideal
2023-05-04 20:12:38 <duncaen> merging the patch would just solidify the fragmention, at the moment only alpine has this, merging the patch into opendoas would bring it to debian/ubuntu and other linux distributions, but not openbsd/freebsd
2023-05-04 20:13:45 <minimal> duncaen: so you think I should change the alpine cloud-init patch to just append rules to /etc/doas.conf?
2023-05-04 20:14:38 <abby> why are packages shipping doas rules in the first place
2023-05-04 20:15:37 <minimal> actualy that's not going to work as currently the alpine doas package doesn't create /etc/doas.conf, it creates/uses /etc/doas.d/doas.conf
2023-05-04 20:16:47 <minimal> abby: in "my" case it is not a packagge (cloud-init) shipping doas rules, rather with cloud-init when people specify user accounts to be created (in their user-data) on 1st boot they specify various related settings such as sudo and doas rules for those users
2023-05-04 20:17:08 <duncaen> abby: there are a few packages that basically allow themselves to restart their sevice or some specific service like postgres using rc
2023-05-04 20:18:47 <duncaen> https://pkgs.alpinelinux.org/contents?file=&path=%2Fetc%2Fdoas.d&name=&branch=edge
2023-05-04 20:19:32 <duncaen> I'm not sure if this is only 5 packages that need this because its not merged and maintainers don't want to depend on this or if its actually just 5 packages that would need this
2023-05-04 20:20:24 <duncaen> *don't want to depend on it until its merged, but would want to make use of doas.d
2023-05-04 20:22:20 <ikke> FYI, original issue behind this: https://gitlab.alpinelinux.org/alpine/tsc/-/issues/1
2023-05-04 20:23:56 <ikke> nagios-plugins uses it to be able to check service status
2023-05-04 20:24:32 <ikke> But rc-update works as normal user as well
2023-05-04 20:24:35 <ikke> er, rc-status
2023-05-04 20:25:26 <minimal> duncaen: I'll have a look at rewriting my cloud-init doas patch to not use this PR
2023-05-04 20:27:05 <XFaCiEer> minimal: Not if I rewrite it first!!
2023-05-04 20:27:16 <XFaCiEer> I mean... I'm not... but if I did...
2023-05-04 20:28:22 <minimal> once I fix another bit of cloud-init functionality that's been broken by an unrelated Alpine config file just being renamed...
2023-05-04 20:46:03 <pj> :^)
2023-05-04 21:00:15 <minimal> duncaen: the Alpine doas package currently doesn't handle /etc/doas.conf, it uses /etc/doas.d/doas.conf (doesn't even keep /etc/doas.conf as a symlink for compat purposes) :-(
2023-05-05 11:27:48 <engblom> I am trying to get alpine to boot using linux-edge kernel. I would like to temporary include every single kernel module into initramfs. What would be the easiest way to do it?
2023-05-05 11:30:09 <psykose> a everything.modules in /etc/mkinitfs/features.d/ that just has `kernel/` as the only line
2023-05-05 11:30:14 <psykose> and add everything to features=
2023-05-05 11:30:17 <psykose> then regen initfs
2023-05-05 11:39:20 <engblom> psykose: Thanks! I will try it.
2023-05-05 12:17:38 <oleovit> hi. i have a question. because its a lot of text: (you can curl it) https://transfer.sh/RxMG7v/msg
2023-05-05 12:20:09 <lopid> could you give us the gist so we don't potentially waste our time?
2023-05-05 12:28:26 <oleovit> i only needed to know, if i have to export these env vars in 20.locale.sh for fish. this commit is for bash, but made me think.. as described in my txt. this made me think:https://git.alpinelinux.org/aports/commit/?id=a5d457166b377270ac87a4a2cf938acdb6f56346
2023-05-05 12:28:33 <psykose> fish doesn't read /etc/profile*
2023-05-05 12:28:46 <psykose> none of that has any relevance
2023-05-05 12:29:44 <psykose> your logs also have a "OPENAI_API_KEY"
2023-05-05 12:31:07 <oleovit> @psykose so, LC_COLLATE et cetera like described, can be ommited? i only mean 20locale.sh 
2023-05-05 12:31:18 <psykose> i don't know what your question is
2023-05-05 12:31:47 <psykose> if you want to set them, set them
2023-05-05 12:31:50 <oleovit> i descriped it here: https://transfer.sh/RxMG7v/msg
2023-05-05 12:31:51 <psykose> if not, don't
2023-05-05 12:32:05 <psykose> none of those commits or anything in /etc/profile is relevant
2023-05-05 12:33:41 <oleovit> so, the env output in the text is valid and that's it, no need for the vars in 20locale.sh? and yes, this api key. i revoked it. my bad :(
2023-05-05 12:34:34 <psykose> probably not, you usually want C.UTF-8 at least
2023-05-05 12:36:09 <oleovit> so, to be on "the safe side", i could source 20locale.sh from my fish config? thats all i need to know, because fish cant do it
2023-05-05 12:36:53 <psykose> set -gx LANG C.UTF-8
2023-05-05 12:37:37 <user__> can someone help me? i'm confused by the documentation
2023-05-05 12:37:39 <oleovit> if LANG is everything i need to add, i'm happy. 
2023-05-05 12:37:53 <psykose> if not you'll eventually find out i guess
2023-05-05 12:37:57 <psykose> don't remember what reads the others
2023-05-05 12:40:48 <oleovit> thx psykose, like always. i will source locale i guess. 
2023-05-05 12:50:05 <user__> is it possible to reconfigure a traditional [sys] installation of alpine into a [data] state where it loads to the state to ram and changes are saved with lbu?
2023-05-05 12:59:07 <user__> hello?
2023-05-05 13:01:21 <pj> Bear in mind that people have their own life, so if someone doesn't reply in less than 10 minutes to your message, it might because no one is active or they don't know the answer
2023-05-05 13:02:14 <user__> understood
2023-05-05 13:02:20 <user__> thankyou anyway
2023-05-05 13:05:28 <psykose> or just have no idea what the answer is
2023-05-05 13:06:28 <user__> anyway the documentation is out of order
2023-05-05 13:08:39 <user__> i'm coming from debian. not sure if alpine is a good idea but you never know until you mess around
2023-05-05 18:36:37 <iggy> Is it possible to have "or" depends for a package? If so, does anybody know of a good package as an example?
2023-05-05 18:37:27 <iggy> my grep-fu failed me
2023-05-05 18:43:51 <ikke> iggy: Only by having both packages provide the same virtual package
2023-05-05 18:44:06 <ikke> In the dependencies you cannot specify an or relationship
2023-05-05 18:44:08 <iggy> I thought so
2023-05-05 18:44:21 <iggy> thanks for the confirmation
2023-05-05 18:46:02 <iggy> I doubt it's possible to have 2 packages provide the same virtual (i.e. there's a package that depends on either nftables or iptables+ip6tables)
2023-05-05 18:53:22 <ikke> You can, but you need to set provider_priority
2023-05-05 18:53:31 <ikke> psykose: that should work, right?
2023-05-05 18:57:15 <iggy> I'm beginning to think I should just open an issue describing the problem and let someone smarter than me fix it
2023-05-05 18:59:59 <ikke> iggy: why not make a merge request?
2023-05-05 19:01:11 <ikke> For my info, what is it that depends on either iptabels or nftables?
2023-05-05 19:17:40 <iggy> wg-quick
2023-05-05 20:04:50 <jacksonchen666> Minor package issue: The yggdrasil package leaves the sensitive config file readable by world and group, which is very undesirable. (Also, where could I report package issues like this?)
2023-05-05 20:05:31 <ikke> jacksonchen666: https://gitlab.alpinelinux.org/alpine/aports
2023-05-05 20:06:40 <ikke> jacksonchen666: the package does not seem to have any config files?
2023-05-05 20:06:49 <ikke> https://pkgs.alpinelinux.org/contents?branch=edge&name=yggdrasil&arch=aarch64&repo=community
2023-05-05 20:07:23 <jacksonchen666> ikke: post install script, config includes private/public key stuff which must be unique
2023-05-05 20:07:57 <ikke> ah I see
2023-05-05 20:08:03 <ikke> yggdrasil -genconf > /etc/yggdrasil.conf
2023-05-05 20:15:52 <ikke> jacksonchen666: !46825
2023-05-05 20:17:15 <jacksonchen666> ikke: thanks!
2023-05-06 08:28:27 <quinq> Hi
2023-05-06 08:28:57 <quinq> I'm trying to remove linux-firmware-qcom, (with "apk del linux-firmware-qcom"), but apk tells me:
2023-05-06 08:29:00 <quinq> World updated, but the following packages are not removed due to: linux-firmware-qcom: linux-firmware-any linux-firmware-other linux-firmware-any linux-lts zfs zfs-prune-snapshots zfs-auto-snapshot
2023-05-06 08:29:32 <quinq> I'm not sure I understand what verb or info the ':' is supposed to convey here, the "due to" is a bit unclear
2023-05-06 08:29:47 <quinq> Is that a conflict?
2023-05-06 08:29:56 <quinq> a missing thing?
2023-05-06 10:32:42 <pj> <quinq> "World updated, but the following..." <- it's required by linux-firmware-other
2023-05-06 10:33:49 <pj> so you most likely have linux-firmware-other in world file, thus apk will not remove a dependency which is linux-firmware-qcom from system (but it updated your world file to not include it there, if it was there)
2023-05-06 10:34:42 <quinq> Thanks for the help pj 
2023-05-06 10:34:58 <quinq> But is there a way of knowing that from apk del error message?
2023-05-06 10:35:32 <quinq> I mean why zfs would even be in that list for example
2023-05-06 10:35:48 <quinq> Does that mean that linux-firmware-qcom is a dependency of zfs (for example)?
2023-05-06 10:36:26 <pj> no
2023-05-06 10:37:50 <quinq> But then why would it be in that list?
2023-05-06 10:37:57 <quinq> My question is what is the meaning of the list
2023-05-06 10:38:26 <quinq> (I search within apk manuals but didn't find an answer to that, maybe I missed it though)
2023-05-06 10:38:44 <pj> I mean you could let me answer
2023-05-06 10:39:16 <quinq> Am I preventing you somehow? :/
2023-05-06 10:39:18 <pj> zfs is tied to linux-lts and linux-lts is tied to linux-firmware
2023-05-06 10:39:27 <quinq> ok
2023-05-06 10:39:50 <pj> so the error will tell about all packages in dep tree
2023-05-06 10:40:05 <pj> not necessarily that they are important
2023-05-06 10:40:25 <quinq> So it'll list all packages depending on linux-lts?
2023-05-06 10:43:01 <quinq> In any case, it works indeed if I remove linux-firmware-other too
2023-05-06 10:43:47 <quinq> What's strange is that linux-lts is also marked as a package depending on it ("apk info -r linux-firmware-qcom" lists it), but it doesn't seem to be a problem there
2023-05-06 10:44:12 <pj> zfs-lts depends on linux-lts and linux-lts depends on linux-firmware
2023-05-06 10:44:29 <pj> linux-firmware is provided by: linux-firmware-any linux-firmware-other linux-firmware-X
2023-05-06 10:46:11 <quinq> I see
2023-05-06 10:46:16 <quinq> Thanks again :)
2023-05-06 10:46:21 <pj> and linux-firmware-none
2023-05-06 10:46:58 <pj> there is 90 different firmware packages you can install and they all provide linux-firmware to satisfy linux-lts
2023-05-06 10:47:00 <quinq> Yeah, I just kept linux-firmware-rtlwifi and world is happy with it
2023-05-06 12:23:39 <uniqNuuid> there is also linux-firmware-none
2023-05-06 12:25:40 <pj> which I already mentioned
2023-05-06 12:26:33 <quinq> :)
2023-05-06 13:01:11 <JohannesJ[m]> Heh
2023-05-06 13:03:20 <JohannesJ[m]> Arch messed up again 🙈 im installing Alpine on my laptop again 😂
2023-05-06 13:05:50 <psykose> what this time
2023-05-06 13:06:36 <JohannesJ[m]> Hmm?
2023-05-06 13:07:05 <JohannesJ[m]> This time i think it was actually a pebkac thingy
2023-05-06 13:07:24 <psykose> ah
2023-05-06 13:07:26 <psykose> :)
2023-05-06 13:07:31 <JohannesJ[m]> Im trying to encrypt the whole thing.. somewhere along the line i made a mistake
2023-05-06 13:08:03 <JohannesJ[m]> Its another good excuse to try Alpine on the desktop again 🙈
2023-05-06 14:21:30 <uniqNuuid> JohannesJ[m]: stable, ~~fine~~
2023-05-06 14:22:03 <uniqNuuid> the alpine stable is really stable
2023-05-06 14:31:19 <uniqNuuid> how do you use mono fonts here? `this`?
2023-05-06 14:31:49 <uniqNuuid> i mean italic
2023-05-06 14:31:51 <abby> irc is already mono for most people
2023-05-06 14:32:14 <grawity> on the IRC side, I don't think there's a monospace format mode? italic exists but not sure how widely supported
2023-05-06 14:33:01 <abby> there is, but it's only used by a few clients
2023-05-06 14:33:23 <uniqNuuid> ^cool^
2023-05-06 14:35:09 <uniqNuuid> i hope this is correct
2023-05-06 14:35:27 <grawity> that's italics yeah
2023-05-06 14:36:37 <uniqNuuid> 02and blue02
2023-05-06 14:37:53 <abby> https://modern.ircdocs.horse/formatting.html
2023-05-06 14:39:02 <grawity> ah, that's a better one than the ancient mIRC docs
2023-05-06 14:39:17 <grawity> still does not mention irssi's blink though
2023-05-06 14:40:55 <uniqNuuid> i use this https://weechat.org/files/doc/stable/weechat_user.en.html#command_line_colors
2023-05-06 14:43:58 <uniqNuuid> well, weechat uses 34M memory here
2023-05-06 16:22:28 <qaqland[m]> when apk add xxx, "Executing mandoc-apropos-1.14.6-r6.trigger" takes a long time to continue
2023-05-06 16:23:22 <qaqland[m]> how can I solve this problem
2023-05-06 16:24:03 <psykose> you don't, it takes that long
2023-05-06 16:24:20 <psykose> man-db is faster i guess, instead of using mandoc, if you want that functionality
2023-05-06 16:25:15 <psykose> makewhatis (mandoc-apropos) doesn't update a db, i.e. every run just generates something fresh
2023-05-06 16:25:22 <psykose> it's not incremental and hence takes forever
2023-05-06 16:29:39 <qaqland[m]> 🙂
2023-05-06 18:30:49 <JohannesJ[m]> <uniqNuuid> "how do you use mono fonts here?..." <- Huh? Like *this* ? Its word btween *
2023-05-06 18:31:47 <JohannesJ[m]> So, does anyone know a guide to install alpine fully encrypted, that actually works? In the past i folled a few that somehow left me unbootable
2023-05-06 18:42:43 <reggiere[m]> <JohannesJ[m]> "So, does anyone know a guide..." <- I'd say mostly just follow archwiki, I dont think there is any stuff specific to alpine regarding encryption that you should worry about
2023-05-06 18:45:36 <JohannesJ[m]> Ohh nice :)
2023-05-06 18:45:37 <JohannesJ[m]> Thanks!
2023-05-06 18:53:06 <abby> JohannesJ[m]: irc does not support markdown-like markup
2023-05-06 18:53:44 <psykose> it does if you squint really hard with your friends
2023-05-06 18:54:01 <ikke> markdown is meant to be human readable :P
2023-05-06 18:54:18 <psykose> humans reading things is overrated
2023-05-06 18:54:35 <ikke> apparently
2023-05-06 18:56:51 <JohannesJ[m]> Oh, well im hardly on irc anymore these days
2023-05-06 18:57:29 <abby> you're on irc right now
2023-05-06 18:57:36 <ikke> :D
2023-05-06 19:02:04 <JohannesJ[m]> Sure :) im everywhere
2023-05-06 19:02:42 <JohannesJ[m]> I see markdown though
2023-05-06 19:03:37 <reggiere[m]> Johannes J.: because you are using a matrix bridge
2023-05-06 19:03:57 <JohannesJ[m]> …
2023-05-06 19:14:21 <minimal> JohannesJ[m]: what exactly do you mean by "fully encrypted"? Do you mean "full disk encryption"? Also is this for BIOS or UEFI system?
2023-05-06 19:18:26 <JohannesJ[m]> UEFI :) and i mean: root partition and swap partition encrypted
2023-05-06 19:18:51 <JohannesJ[m]> When i boot i need to enter a password and only then does everything get decrypted :)
2023-05-06 19:19:02 <JohannesJ[m]> Found a tutorial im following as we speak
2023-05-06 19:19:15 <minimal> JohannesJ[m]: with separate unencrypted /boot partition/fs? or /bot as part of / partition/fs?
2023-05-06 19:19:55 <minimal> using which bootloader? Grub?
2023-05-06 19:20:46 <JohannesJ[m]> Only a small efi partition with grub thats not encrypted
2023-05-06 19:21:21 <minimal> ESP partition is always unencrypted, UEFI does not support it being encrypted
2023-05-06 19:21:40 <minimal> I was asking if you also had a separate partition for /boot or not
2023-05-06 19:22:29 <JohannesJ[m]> No
2023-05-06 19:23:00 <minimal> ok, so in other words you want to use Grub's ability to access the encrypted rootfs in order to obtain the kernel and initramfs files
2023-05-06 19:23:25 <JohannesJ[m]> Exactly :) and i found a tutorial for that im now following :)
2023-05-06 19:23:50 <minimal> but you don't want to be prompted twice for passphrase (once by Grub, again by initramfs)
2023-05-06 19:24:18 <minimal> so you'll need to use a keyfile (for initramfs to use)
2023-05-06 19:24:51 <minimal> are you using LUKS v1 or v2?
2023-05-06 19:25:48 <satmd> if you cannot provide a plain kernel+initramfs to grub, you might instead be able to put a linux+efistub+initramfs ".efi" file into /boot/EFI
2023-05-06 19:26:03 <JohannesJ[m]> Yes, the tutorial i am following includes that :) let me pist the link
2023-05-06 19:26:06 <JohannesJ[m]> Pist
2023-05-06 19:26:09 <JohannesJ[m]> Post
2023-05-06 19:26:18 <satmd> that's how I solved it - albeit I did it on another linux distribution
2023-05-06 19:26:28 <JohannesJ[m]> This one: https://battlepenguin.com/tech/alpine-linux-with-full-disk-encryption/
2023-05-06 19:26:58 <minimal> satmd: yes using UKI file is an alternative method, but he mentioned (wanting to use) Grub
2023-05-06 19:27:46 <JohannesJ[m]> For now ill just use whatever the tutorial says :) i can later create a vm to play with :€
2023-05-06 19:28:15 <minimal> ok, that guide is using LUKS v1, as LUKS v2 doesn't work for "FDE" in current Grub release
2023-05-06 19:28:58 <minimal> also that guide is using LVM, which you didn't mention you were using ;-)
2023-05-06 19:29:06 <JohannesJ[m]> Sorry :)
2023-05-06 19:29:53 <minimal> and that guide uses a keyfile to avoid the double passphrase prompting I mentioned
2023-05-06 19:30:18 <JohannesJ[m]> Since its a thin laptop, i dont really need lvm but also im a bit done with all the thinkering :) so i’ll just go with it. Once ive learned enough about linux encryption i might do something myself and add it to the wiki instead :)
2023-05-06 19:30:37 <JohannesJ[m]> Yes, thats what i said :)
2023-05-06 19:31:29 <minimal> oh? I don't see anywhere that you mentioned double prompting or wanting to avoid it
2023-05-06 19:31:47 <minimal> also that guide sets up some stuff in a "strange" way
2023-05-06 19:32:04 <JohannesJ[m]> <JohannesJ[m]> "Yes, the tutorial i am following..." <- Well, with this reply i ment the keyfile :)
2023-05-06 19:32:23 <JohannesJ[m]> Sorry if that wasnt clear :)
2023-05-06 19:32:28 <minimal> I've never had to use GRUB_PRELOAD_MODULES when I've set such stuff up
2023-05-06 19:32:47 <JohannesJ[m]> Oh?
2023-05-06 19:37:59 <minimal> if the /etc/default/grub settings are configured correctly then grub-install (via grub-probe) should automatically workout the modules it needs, instead of manually specifying some of them via GRUB_PRELOAD_MODULES
2023-05-06 19:38:54 <JohannesJ[m]> Hmmz
2023-05-06 19:41:15 <minimal> I'm not saying that the steps in the article do not work - rather that I don't understand why some of them are being done
2023-05-06 19:42:26 <JohannesJ[m]> Okay :) well i’ll have to learn a bit more about it all to figure thst part out :)
2023-05-06 19:44:25 <JohannesJ[m]> If i dont include the preload line, and it turns out it doesnt work without it, i suppose i can boot into rescue mode and include it again?
2023-05-06 19:45:05 <minimal> well if you follow that guide you've disabled recovery/rescue mode haven't you? ;-)
2023-05-06 19:45:16 <JohannesJ[m]> Heh
2023-05-06 19:45:19 <JohannesJ[m]> Good point
2023-05-06 19:48:53 <minimal> besides you could check that the generated /boot/boot/grub.cfg contained "insmod" lines for "luks", "lvm" etc
2023-05-06 19:50:48 <JohannesJ[m]> True true
2023-05-06 19:51:28 <JohannesJ[m]> I must admit ive always installed debian :) it comes with a GUI installer 🙈
2023-05-06 19:51:50 <minimal> that's the "problem" with blindly following a "guide", it make configure things in a way you didn't expect
2023-05-06 19:51:58 <minimal> s/make/may/
2023-05-06 19:53:29 <JohannesJ[m]> True true
2023-05-06 19:53:51 <JohannesJ[m]> Though in big lines i know what it does :)
2023-05-06 19:57:14 <JohannesJ[m]> Clearly the lines are not big enough 🙈 it throws a “no bootable device”
2023-05-06 19:57:26 <JohannesJ[m]> Im beginning to think im not made for computers anymore
2023-05-06 20:16:19 <reggiere[m]> I do not know how to help with grub cause I dont use it but if you wanna troubleshoot stuff from this point instead of starting again from scratch you can still access the drive from bootable usb even if its encrypted
2023-05-06 20:16:58 <JohannesJ[m]> True :) i’ll just go to bed and see tomorrow :) thank you though :)
2023-05-06 20:21:22 <Guest702> Hello, I am trying to setup network printing with avahi + cups so that the printer can be connected to via it's local address, but it seems the method to get avahi to resolve the .local tld (via /etc/nsswitch.conf) is not supported by musl, so do any of you know any other solution to automatically discover the printer's ip address?
2023-05-06 20:21:24 <reggiere[m]> after booting you can... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/JInrsmxzGKoDluKnAxthXeHE>)
2023-05-06 20:23:56 <reggiere[m]> for example with the way my drive is setup I would do
2023-05-06 20:23:56 <reggiere[m]> `cryptsetup luksOpen /dev/mapper/vg0-root cryptroot` and then I would find my drive at /dev/mapper/cryptroot
2023-05-06 20:25:08 <reggiere[m]> as far as what to do afterwards I have no clue, as I mentioned I dont use grub, I boot directly using efibootmgr which is another can of worms
2023-05-06 20:25:56 <JohannesJ[m]> The problem is me really :) lately i seem to break everything that has a power cord
2023-05-06 20:26:15 <JohannesJ[m]> I need 6 months in the middle of nowhere with good books, enough food and no people 😂
2023-05-06 20:26:31 <reggiere[m]> thats fine, I caused lithium battery fire a week ago :3
2023-05-06 20:26:55 <reggiere[m]> not even joking, I was trying to repair a phone for the first time ever and it went the worst way possible
2023-05-06 20:28:36 <JohannesJ[m]> Well, i used to repair iphones, thankfully for me thats no longer possible 😂
2023-05-06 20:28:46 <JohannesJ[m]> Or at least, not as easily as it used to be
2023-05-06 20:29:03 <satmd> Guest702: maybe you have a chance with using a local dnsmasq or unbound as dns proxy, forwarding .local to 5353
2023-05-06 20:33:15 <Guest702> Doesnt seem like too bad of an idea, I'll give that a try and see how it goes, thanks!
2023-05-06 21:13:20 <JohannesJ[m]> minimal: you are correct, i left out the whole preload modules line in the grub config, and my encrypted setup boots just fine 🥰
2023-05-06 21:14:30 <JohannesJ[m]> I live Alpine again :)
2023-05-06 21:15:16 <JohannesJ[m]> Love.. i love Alpine again
2023-05-06 21:16:42 <minimal> well that's grub-install (via grub-probe) being "smart"
2023-05-06 21:17:59 <JohannesJ[m]> :) i actually deviated from the guide, i ram setup-alpine beforehand, and just stopped at the setup disk part
2023-05-06 21:18:48 <JohannesJ[m]> Now booting into plasma :) if all workd fine
2023-05-06 21:19:35 <JohannesJ[m]> It asks for a key just once :) then boots to plasma just fine 🥰
2023-05-06 21:22:17 <JohannesJ[m]> Funnily enough audio works out of the box now, the only thing i did different from the wiki is add hwdrivers and mdev tot the sysinit runlevel.. it complained about that but the wiki says nothing of it
2023-05-06 21:42:58 <Guest702> satmd I ended up using avahi2dns to bind avahi to dns instead, and it just worked.
2023-05-06 21:45:37 <pj> what could be the reason for linker to not find Scrt1.o even if it has -L with path that has it
2023-05-06 21:48:28 <pj> hm, it was looking for the one it was compiled with
2023-05-06 22:00:52 <Saijin_Naib> Is the pyc metapackage meant to behave like the doc one in that we can freely add or remove it without it really affecting any depends or installed programs? 
2023-05-06 22:01:20 <ikke> aahttps://pkgs.alpinelinux.org/packages?name=python3-pyc&branch=edge&repo=&arch=&maintainer=
2023-05-06 22:02:55 <Saijin_Naib> Is that a yes or a no? 
2023-05-06 22:03:29 <Saijin_Naib> Because if it is, you can't remove it after any packages with *-pyc are installed 
2023-05-06 22:03:48 <ikke> oh, it's called pyc
2023-05-06 22:03:55 <ikke> https://pkgs.alpinelinux.org/packages?name=pyc&branch=edge&repo=&arch=&maintainer=
2023-05-06 22:04:18 <Saijin_Naib> Yeah
2023-05-06 22:09:33 <psykose> no, docs aren't installed by default
2023-05-06 22:09:33 <psykose> that is
2023-05-06 22:09:39 <psykose> apk add !pyc would remove it
2023-05-06 22:10:54 <psykose> or !python3-pyc
2023-05-06 23:19:55 <JohannesJ[m]> YYYAAASSSSSS!! I got networkmanager to work with KDE
2023-05-06 23:20:08 <JohannesJ[m]> OMG! This kept me from using Alpine on my desktop FOR MONTHS!
2023-05-06 23:23:46 <JohannesJ[m]> Hmz.. when i copy something to the desktop, KDE crashes :( oh well! Thats for snother rainy day 🥰
2023-05-06 23:56:33 <pj> truly KDE experience
2023-05-07 00:05:03 <JohannesJ[m]> 😂😂
2023-05-07 00:05:11 <JohannesJ[m]> I still love it more then gnome
2023-05-07 00:48:47 <Saijin_Naib> <psykose> "apk add !pyc would remove it" <- So apk del pyc is expected to fail with dependencies? Alright, awesome, thank you for clarifying 
2023-05-07 00:49:19 <psykose> del doesn't delete things so yea
2023-05-07 00:49:19 <Saijin_Naib> <JohannesJ[m]> "I still love it more then gnome" <- No helping bad taste 🤣
2023-05-07 00:50:19 <Saijin_Naib> <JohannesJ[m]> "I still love it more then gnome" <- What themes and icons/cursors do you use? 
2023-05-07 00:50:55 <Saijin_Naib> psykose: So we are adding the packages without pyc with that, correct? As opposed to removing the PYC packages? 
2023-05-07 00:51:20 <psykose> it is removing
2023-05-07 02:30:25 <dalias> anyone have experience loading alpine on one of these? https://inovato.com/
2023-05-07 02:45:55 <atka> dalias: no experience here, looks like a neat little computer for $35. looks like it has mainline uboot support and is mostly supported by mainline linux
2023-05-07 02:46:27 <atka> I think it would work ok with generic arm images
2023-05-07 03:43:09 <Saijin_Naib> Thank you for putting that on my radar
2023-05-07 04:25:20 <qaqland[m]> dalias: it is a tx box
2023-05-07 04:25:28 <qaqland[m]> *TV
2023-05-07 05:16:56 <atka> is nix on alpine a thing?
2023-05-07 06:31:40 <kode54> GNOME 44 is great, dunno if this issue affects Alpine or not
2023-05-07 06:32:04 <kode54> xdg-desktop-portal-gnome prevents xdg-desktop-portal from working on any other desktop than GNOME
2023-05-07 06:32:16 <kode54> new to v44
2023-05-07 06:32:34 <psykose> they fixed that in 44.1
2023-05-07 06:32:37 <kode54> oh?
2023-05-07 06:32:48 <kode54> guess Arch didn't get the memo, because they shipped the broken one
2023-05-07 06:33:12 <psykose> arch has the working one
2023-05-07 06:33:22 <kode54> not here
2023-05-07 06:33:24 <kode54> let me check
2023-05-07 06:34:05 <kode54> maybe it only fixes KDE
2023-05-07 06:34:33 <kode54> if I start a wlroots desktop, it times out starting the gnome portal, then all my apps take forever to start, and the wlr portal never starts
2023-05-07 06:34:43 <psykose> which version do you have
2023-05-07 06:34:47 <kode54> 44.1-2
2023-05-07 06:34:53 <psykose> did you reboot since getting that
2023-05-07 06:34:55 <kode54> I'll report to arch
2023-05-07 06:34:58 <kode54> oh, reboot? lovely
2023-05-07 06:35:05 <kode54> I have rebooted since updating it too
2023-05-07 07:23:14 <livy_> Hello. Anybody uses OneDrive for Linux?
2023-05-07 07:23:36 <livy_> I have been using it for ages on Ubuntu. But now I have problem starting it on Alpine.
2023-05-07 07:23:57 <ikke> Are those pre-compiled binaries?
2023-05-07 07:24:23 <livy_> Yes, I downloaded it from repository. After installation, I used 'onedrive' command to authorize it with Microsoft.
2023-05-07 07:24:39 <livy_> Then I add it to OpenRC using: rc-update add onedrive
2023-05-07 07:24:43 <psykose> that doesn't work on alpine
2023-05-07 07:24:50 <psykose> it is however packaged
2023-05-07 07:25:05 <livy_> I mean I used 'apk add onedrive'. The boot log shows that the service indeed started.
2023-05-07 07:25:20 <livy_> But onedrive log told me that it failed to authorize.
2023-05-07 07:25:25 <livy_> Running it manually works, however.
2023-05-07 07:25:31 <livy_> Which leads me to the speculation that...
2023-05-07 07:25:41 <psykose> that's not what was meant by precompiled :p
2023-05-07 07:25:58 <livy_> I ran the command 'onedrive' as my own user and authorized it with my Microsoft account.
2023-05-07 07:26:06 <livy_> But OpenRC ran it as superuser.
2023-05-07 07:26:27 <livy_> And the superuser cannot use my Microsoft credentials.
2023-05-07 07:26:48 <ikke> s/superuser/root ;)
2023-05-07 07:26:51 <livy_> Which leads us back to the question: how can I auto-start onedrive as my own user?
2023-05-07 07:27:31 <livy_> On Linux Mint, I use: systemctl --user start onedrive;
2023-05-07 07:27:39 <livy_> The '--user' is crucial for it to work.
2023-05-07 07:27:41 <ikke> openrc has no user services
2023-05-07 07:27:55 <ikke> (you can run services as any user though)
2023-05-07 07:28:20 <ikke> But that still requires root to setup 
2023-05-07 07:29:37 <livy_> So, the easy way to auto-start onedrive is to run 'doas onedrive" to authorize it with Microsoft
2023-05-07 07:29:46 <livy_> using root and my Microsoft credentials.
2023-05-07 07:30:07 <livy_> Then 'rc-update add onedrive' to let OpenRC run it as root next time the computer boots.
2023-05-07 07:31:02 <livy_> Or should I auto-start it using cron?
2023-05-07 07:31:36 <livy_> I wonder if Busybox cron supports the '@reboot' option.
2023-05-07 07:31:37 <ikke> modify /etc/conf.d/onedrive and add command_user="youruser:yourgroup"
2023-05-07 07:31:43 <livy_> I think @reboot is from BSD cron.
2023-05-07 07:32:59 <livy_> @ikke Thank you. I will try your suggestion.
2023-05-07 07:33:26 <livy_> For now maybe it is fine. But for a multiple-users environment, it does not look good.
2023-05-07 07:33:46 <livy_> Because basically OpenRC does not support per-user services.
2023-05-07 07:33:59 <uniqNuuid> livy_: @reboot touch /test_at_reboot 
2023-05-07 07:34:05 <uniqNuuid> and reboot, and see if you see that file? 
2023-05-07 07:34:34 <livy_> I've tried the '@reboot' on OpenWrt. It used Busybox, just like Alpine.
2023-05-07 07:35:04 <livy_> And it did not work. The developer said '@reboot' is from BSD, and available on some distros such as Ubuntu.
2023-05-07 07:35:08 <livy_> But not on Busybox.
2023-05-07 07:35:38 <uniqNuuid> cronie supports @reboot
2023-05-07 07:37:05 <uniqNuuid> all you do is just test
2023-05-07 07:37:22 <uniqNuuid> with root user : crontab -e @reboot touch /something 
2023-05-07 07:39:33 <uniqNuuid> livy_: https://git.busybox.net/busybox/tree/miscutils/crond.c?id=051665ef69568cf16a445a86a43d5ae74d303add#n38
2023-05-07 07:39:54 <uniqNuuid> it has @reboot, so you just need to check alpine linux's busybox config 
2023-05-07 07:41:08 <uniqNuuid> and it is YES : https://git.alpinelinux.org/aports/tree/main/busybox/busyboxconfig#n785
2023-05-07 07:41:27 <uniqNuuid> so, alpine's busybox supports @reboot
2023-05-07 08:26:54 <kode54> https://bugs.archlinux.org/task/78413 :upsi
2023-05-07 08:27:09 <kode54> dammit, return sends even when I'm completing emoji
2023-05-07 08:27:20 <kode54> 🙃
2023-05-07 08:48:01 <livy_> @uniqNuuid Surprisingly Alpine Busybox does support '@reboot'.
2023-05-07 08:48:17 <livy_> Maybe OpenWrt disables that option when they compile BusyBox.
2023-05-07 08:56:19 <uniqNuuid> livy_: yeah, when in doubt, read source code. 
2023-05-07 08:57:33 <uniqNuuid> that's why it's called open source
2023-05-07 09:02:32 <livy_> I am sort of a hobbyist programmer, but it doesn't mean I can dig right into a big project written in C.
2023-05-07 09:02:55 <livy_> C is undoubtly the hardest programming language today.
2023-05-07 09:03:15 <livy_> In my country, they are lazy and always do C Sharp.
2023-05-07 09:03:39 <livy_> I have some experiences in C++ only.
2023-05-07 09:13:01 <uniqNuuid> livy_: i have no coding skills
2023-05-07 09:19:16 <livy_> I use cron '@reboot' and put 'onedrive --monitor' command inside a tmux session and it works now.
2023-05-07 09:19:38 <livy_> Better than running it as a system-wide OpenRC service.
2023-05-07 09:19:57 <livy_> Because each user may have their own onedrive service to synchronize their data.
2023-05-07 09:20:37 <livy_> It is a good practice to avoid using system service for such things.
2023-05-07 09:23:30 <livy_> Alpine is really a distro for "power user".
2023-05-07 09:23:56 <livy_> Imagine a new Linux using trying to use Alpine. They will soon give up because
2023-05-07 09:24:14 <livy_> they do not know why it doesn't work, and how to fix it.
2023-05-07 09:25:23 <livy_> # 'mousepad'			text editor
2023-05-07 09:25:23 <livy_> # 'ristretto'			image viewer
2023-05-07 09:26:45 <uniqNuuid> livy_: for that, i use /etc/local.d
2023-05-07 09:26:46 <livy_> Nevermind, this stupid Windoze pasted my clipboard when I mis-rightclick.
2023-05-07 09:27:10 <uniqNuuid> e.g. /bin/su -l USERNAME -c 'tmux new -d -s $(uname -n) weechat'
2023-05-07 09:28:14 <uniqNuuid> you need to enable : rc-service add /etc/init.d/local default
2023-05-07 09:28:53 <livy_> I've taken a look at the /etc/local.d/README
2023-05-07 09:29:25 <livy_> So basically that directory contains script/executable files with .start and .stop extension.
2023-05-07 09:29:27 <uniqNuuid> i compared @reboot  vs local.d before 
2023-05-07 09:29:55 <livy_> Do we need to chmod+x to scripts file inside /etc/local.d ?
2023-05-07 09:30:11 <uniqNuuid> i remember @reboot does not do correct $PATH after tmux start
2023-05-07 09:31:08 <uniqNuuid> while su -l USERNAME is able to pass SHELL and PATH to the command after -c 
2023-05-07 09:31:30 <uniqNuuid> livy_: yeah, chmod +x 
2023-05-07 09:32:34 <uniqNuuid> livy_: you can check your SHELL and PATH inside the tmux started by @reboot, see if it matches the users config in /etc/passwd?
2023-05-07 09:32:52 <livy_> "rc-service add /etc/init.d/local default" -- does 'default' runlevel work with onedrive?
2023-05-07 09:33:02 <livy_> I mean onedrive need the rootfs to be mounted.
2023-05-07 09:33:09 <livy_> I am new to OpenRC.
2023-05-07 09:33:16 <livy_> Do not quite understand its runlevel yet.
2023-05-07 09:33:28 <uniqNuuid> yes, rootfs is mounted by service in sysinit
2023-05-07 09:33:44 <uniqNuuid> default is way after sysinit
2023-05-07 09:33:56 <livy_> Oh then it shall be fine.
2023-05-07 09:36:00 <uniqNuuid> the other method is : https://0x0.st/HZ9L.txt livy_ 
2023-05-07 09:36:30 <uniqNuuid> so there are 3 methods, i think local.d is the best. 
2023-05-07 09:37:06 <livy_> I think this method has a problem: /bin/su -l USERNAME -c 'tmux new -d -s $(uname -n) weechat'
2023-05-07 09:37:18 <livy_> The 'local' service is still a system service started by OpenRC.
2023-05-07 09:37:32 <livy_> It will run no matter the user logs in or not.
2023-05-07 09:37:52 <livy_> While the 'onedrive' for that user only needs to run if the user log in.
2023-05-07 09:38:56 <uniqNuuid> so you want to run a tmux(onedrive) after user login in? 
2023-05-07 09:40:00 <livy_> Yeah, I think so.
2023-05-07 09:40:18 <uniqNuuid> why run the script in ~/.bashrc?
2023-05-07 09:40:23 <uniqNuuid> why not
2023-05-07 09:40:39 <livy_> I do not know if putting '@reboot' in 'crontab -e' can do that.
2023-05-07 09:40:52 <livy_> Because I mostly work with OpenWrt, which is a single user OS.
2023-05-07 09:41:59 <livy_> I do not see the file ~/.bashrc in my home directory. Maybe it is available after installing Bash?
2023-05-07 09:41:59 <uniqNuuid> you can put something like this in ~/.bashrc : tmux new -s $(uname -n) onedrive --options
2023-05-07 09:42:21 <uniqNuuid> livy_: oh, you are using ash? then put it in .profile
2023-05-07 09:42:54 <uniqNuuid> grep '^USERNAME' /etc/passwd to see the shell it uses
2023-05-07 09:43:02 <livy_> I use default Busybox ash.
2023-05-07 09:43:11 <uniqNuuid> ash sources ~/.profile 
2023-05-07 09:43:57 <livy_> Will the onedrive process be auto-terminated when I log out, if I put in in .profile?
2023-05-07 09:44:38 <livy_> I worry about the scenario when I log in/out multiple times. And it starts multiple instances.
2023-05-07 09:45:02 <uniqNuuid> if you start onedriver in tmux, it will not quit 
2023-05-07 09:45:38 <uniqNuuid> than you can use:  tmux new -ADXs $(uname -n) onedrive --options 
2023-05-07 09:46:59 <uniqNuuid> if you want to logout, then use ctrl + b d
2023-05-07 09:47:09 <uniqNuuid> that will detatch tmux and then you can exit 
2023-05-07 09:47:41 <uniqNuuid> next time you login, tmux new -ADXs $(uname -n) will attach it 
2023-05-07 09:50:45 <uniqNuuid> read tmux man pages about `new -ADXs`
2023-05-07 09:51:06 <uniqNuuid> every one of these options are important to make it work 
2023-05-07 09:51:25 <livy_> Thanks so much. I'll read them now.
2023-05-07 09:51:32 <livy_> So many things to learn today.
2023-05-07 09:51:49 <quinq> Learn something everyday
2023-05-07 09:52:14 <uniqNuuid>  The -A flag makes new-session behave like attach-session if session-name already exists
2023-05-07 09:54:26 <uniqNuuid> i think -D -X is not needed, those 2 are used to quit tmux sessions in another terminal, so you will always have only one session at any time
2023-05-07 09:55:47 <uniqNuuid> so, `new -As $(uname -n)` is good enough. 
2023-05-07 09:57:11 <uniqNuuid> using $(uname -n) is not needed, you can assign any `name`, it's just $(uname -n) is a good method to make the name  consistent and distinguish
2023-05-07 10:00:16 <uniqNuuid> anyway, i start tmux this way on SSH. https://0x0.st/HZ9x.txt
2023-05-07 10:01:40 <uniqNuuid> so when i ssh into remote machine, it will start or attach tmux for me. this makes sure i am always on tmux on ssh. never worried about ssh disconnection quiting my current jobs 
2023-05-07 10:37:48 <JohannesJ[m]> <Saijin_Naib> "What themes and icons/cursors do..." <- breeze :) but i used to have some theme that made it a bit nicer :) im trying to find it again :D
2023-05-07 10:38:43 <JohannesJ[m]> im really pleased with Alpine on my laptop now :)
2023-05-07 11:51:34 <livy_> @uniqNuuid "If -x is given, send SIGHUP to the parent process of the client as well as detaching the client, typically causing it to exit."
2023-05-07 11:51:55 <livy_> Looks like if we use -x, we do not need -d anymore.
2023-05-07 11:52:35 <livy_> "If -d is specified, any other clients attached to the session are detached."
2023-05-07 11:53:20 <livy_> So, the "safe" tmux invocation should be: tmux new-session -AX -s "<session-name>"
2023-05-07 14:18:23 <dalias> qaqland[m], i know that. but i'm asking about running alpine on it
2023-05-07 14:44:15 <uniqNuuid> livy_: you are right, -X includes -D
2023-05-07 14:47:48 <uniqNuuid> livy_: also in ~/.tmux.conf, you may want `bind d detach -P`, that will close shell when you manual detach
2023-05-07 15:53:38 <Saijin_Naib> <JohannesJ[m]> "im really pleased with Alpine on..." <- Yeah, Alpine continues to be a joy to use. Just keeps getting better
2023-05-07 15:56:41 <JohannesJ[m]> Well, iuts been a rather rocky road :D
2023-05-07 15:56:44 <latex> Except when the kernel breaks
2023-05-07 15:56:54 <latex> When the kernel breaks, you cannot downgrade on Alpine
2023-05-07 15:57:12 <JohannesJ[m]> no, but you can add different kernels?
2023-05-07 15:57:33 <latex> Just this fact alone makes me reconsider my choice of operating system
2023-05-07 15:58:00 <latex> Maybe BSD or something
2023-05-07 15:58:03 <minimal> you cannot install multiple versions of the same kernel package, e.g. in you are using linux-lts then when you update it the old version of it is removed
2023-05-07 15:58:10 <latex> Or Devuan
2023-05-07 15:58:23 <minimal> you can however install both linux-lts and linux-edge packages
2023-05-07 15:58:37 <JohannesJ[m]> but you can pin the old version, remove the new kernel, and install the old one again?
2023-05-07 15:58:38 <latex> linux-lts broke not too long ago
2023-05-07 15:59:01 <latex> 5.15.99 broke many Intel systems
2023-05-07 15:59:12 <minimal> JohannesJ[m]: in general Alpine repos only store a single version of each package
2023-05-07 15:59:24 <JohannesJ[m]> hmz
2023-05-07 15:59:35 <JohannesJ[m]> well, thats why we have good backups
2023-05-07 15:59:58 <minimal> latex: there has been talk in the past about addressing the kernel upgrade issue
2023-05-07 16:00:50 <JohannesJ[m]> i like talk. much more then listning.
2023-05-07 16:01:19 <uniqNuuid> latex: you can install linux-edge
2023-05-07 16:01:38 <minimal> uniqNuuid: I already mentioned that
2023-05-07 16:03:22 <uniqNuuid> it would be nice if there is linux-oldlts
2023-05-07 16:03:41 <uniqNuuid> so, 2 versions of lts and 1 version linux-edge
2023-05-07 16:04:05 <JohannesJ[m]> fair enough
2023-05-07 16:04:16 <minimal> and when the "old" version needs to be updated due to security issue then what?
2023-05-07 16:04:52 <uniqNuuid> for now, according to kernel.org : this would be  5.15 6.1 6.4
2023-05-07 16:05:01 <uniqNuuid> s/6.4/6.3
2023-05-07 16:05:17 <minimal> 6.3 is not LTS
2023-05-07 16:05:48 <uniqNuuid> read ^^
2023-05-07 16:06:09 <JohannesJ[m]> but this aplies only to the official repo right? so in theory one can run his own repo with multiple versions? or am i thinking wrong?
2023-05-07 16:06:58 <minimal> linux-edge's configuration is also completely different to that of linux-lts/linux-virt
2023-05-07 16:07:37 <uniqNuuid> well, i may download one working version of kernel to ~/ , and keep it there in case things break
2023-05-07 16:08:10 <uniqNuuid> just like what i did to apk-tools. i have install apk-tools-static 
2023-05-07 16:08:29 <minimal> uniqNuuid: als having both 5.15 and 6.1 versions of linux-lts doesn't fully resolve upgrade problems - if a kernel security issue requires updates to both linux-lts 5.15 and linux-lts 6.1 then what? you're back to the same issue of an update potentially breaking
2023-05-07 16:08:30 <uniqNuuid> in case apk-tools failed to work due to ssl issue? 
2023-05-07 16:09:10 <JohannesJ[m]> so, we're back to making backups again :)
2023-05-07 16:09:41 <uniqNuuid> minimal: so, i think it is nice to keep a version of linux-lts.apk somewhere on the disk is a good solutuon. 
2023-05-07 16:10:05 <uniqNuuid> that's is probably the easiest 
2023-05-07 16:10:37 <uniqNuuid> but well, there are few pkgs depends on the kernel, like zfs? 
2023-05-07 16:11:02 <minimal> JohnannesJ[m]: or fs with snapshots
2023-05-07 16:11:37 <JohannesJ[m]> true :)
2023-05-07 16:11:41 <minimal> yes zfs and some other kernel modules are out-of-kernel-tree and depend on the particular version of the kernel package they were built against
2023-05-07 16:11:55 <JohannesJ[m]> mind you, i dont have neither.. time to do something about that :)
2023-05-07 16:12:23 <minimal> JohnannesJ[m]: so which do you have?
2023-05-07 16:13:23 <JohannesJ[m]> neother. i dont do backups and i dont have snapshots :Di'm living on the edge
2023-05-07 16:16:22 <uniqNuuid> i am living on edge too. 
2023-05-07 16:17:22 <JohannesJ[m]> but ive always used debian, i never had a situation where things broke so badly i needed a backup :)
2023-05-07 16:18:54 <uniqNuuid> JohannesJ[m]: speaking of debian, i feel it is difficult to contribute or bug report or MR/PR, their system is kinda complex
2023-05-07 16:19:31 <minimal> JohannesJ[m]: I was picing up on your double negative ;-)
2023-05-07 16:20:00 <JohannesJ[m]> humor is not my best ;-)
2023-05-07 16:20:23 <JohannesJ[m]> well, ive only contributed to Alpine, so i wouldnt know ;-)
2023-05-07 16:49:17 <Nulo[m]> how do i use /etc/periodic? just drop a script in one of the folders and it'll run?
2023-05-07 17:03:12 <uniqNuuid> Nulo[m]: https://wiki.alpinelinux.org/wiki/Alpine_Linux:FAQ
2023-05-07 17:04:14 <Nulo[m]> uniqNuuid: i already read that. but it doesn't explicitly say "drop script in /etc/periodic/*/". i guess it is.
2023-05-07 17:04:15 <uniqNuuid> Nulo[m]: you'll look at `Why don't my cron jobs run`
2023-05-07 17:04:52 <uniqNuuid> Nulo[m]: did you read : `check the following`
2023-05-07 17:05:47 <Nulo[m]> yes, that's why i guessed it is. but idk, it could want a special format for the file or something, so i wanted to confirm
2023-05-07 17:08:21 <uniqNuuid> Nulo[m]: put some script `!/bin/sh` `touch /test_cron` in 15mins and chmod +x and see?
2023-05-07 17:10:04 <uniqNuuid> you'll have to check that file after 15min, i am afraid there is no way to let it run immidiately 
2023-05-07 17:11:26 <uniqNuuid> Nulo[m]: there is nothing so special, 1. it has a shebang !/bin/sh 2. it has +x bit 3. the file name does not contain `dot`
2023-05-07 17:11:46 <Nulo[m]> understood
2023-05-07 17:13:17 <uniqNuuid> Nulo[m]: also better use abosolute path like /usr/bin/some_command 
2023-05-07 17:13:53 <uniqNuuid> god know what default PATH it is 
2023-05-07 17:14:07 <bootloader> hey everyone. i'm having a tough time installing alpine on a macbook so, i'm curious on your thoughts
2023-05-07 17:15:23 <bootloader> is there anyway to write to /lib/firmware from a live-cd install? This is because I cant download necessary packages to install the system without first activating my wifi-card
2023-05-07 17:15:23 <uniqNuuid> Nulo[m]: but i usually test default path like this: crontab -e : * * * * * echo $PATH > /test_cron_path
2023-05-07 17:16:46 <bootloader> whenever i try to write to it, it says "read-only directory". i assume this is because its either reading from RAM or the live-cd
2023-05-07 17:16:50 <minimal> bootloader: are you using Standard or Extended ISO? I'm guessing Standard
2023-05-07 17:17:08 <bootloader> minimal :yes standard
2023-05-07 17:17:14 <minimal> try Extended then
2023-05-07 17:17:34 <bootloader> okay, ill give it a shot
2023-05-07 17:19:55 <uniqNuuid> bootloader: https://docs.kernel.org/driver-api/firmware/fw_search_path.html
2023-05-07 17:20:02 <uniqNuuid> /sys/module/firmware_class/parameters/path
2023-05-07 17:20:31 <uniqNuuid> You would echo into it your custom path and firmware requested will be searched for there first
2023-05-07 17:21:05 <uniqNuuid> so : mkdir /tmp/firmware ; echo '/tmp/firmware' > /sys/module/firmware_class/parameters/path
2023-05-07 17:21:14 <uniqNuuid> and download your firmware to /tmp/firmware ?
2023-05-07 17:22:47 <bootloader> uniqNuuid this is so beautiful, thank you. while the extended-version is downloading, i'll try this
2023-05-07 17:23:14 <bootloader> i really like the cut-down size of standard, so this is massive
2023-05-07 17:27:38 <minimal> bootloader: the Extended version comes with more firewire AFAIK
2023-05-07 17:27:50 <minimal> s/firewire/firmware/
2023-05-07 17:27:56 <uniqNuuid> bootloader: does it work? i never tried. i usually download the firmware to : /lib/firmware/updates/
2023-05-07 17:28:28 <minimal> bootloader: the "cut-down size of standard" is because it comes with fewer packages...
2023-05-07 17:29:27 <uniqNuuid> so, the logic is, when you load a e.g. wifi module, the wifi firmware will be searched by the kernel? 
2023-05-07 17:29:54 <uniqNuuid> so, all this can be done at runtime? 
2023-05-07 17:30:01 <minimal> modules define the firmware they depend on
2023-05-07 17:30:16 <minimal> do "modinfo <module name>" and look for firmware tags
2023-05-07 17:30:31 <uniqNuuid> that's it.
2023-05-07 17:30:35 <bootloader> uniqNuuid my internet is slow, so im still redownloading the firmware. but presumably, it should work
2023-05-07 17:30:40 <uniqNuuid> live loading modules and firmwares
2023-05-07 17:31:09 <minimal> from memory that is also used by mkinitfs when building a initramf to check each module it adds so see if it also needs to add firmware for it
2023-05-07 17:33:13 <uniqNuuid> bootloader: if it works, after installation of alpine linux, don't reboot, instead mount the Hard drive  and put the wifi firmware into /lib/firmware/updates/apple/ , and then reboot
2023-05-07 17:33:38 <uniqNuuid> create that apple dir first. 
2023-05-07 17:35:00 <bootloader> minimal does that mean alpine will naturally install the broadcom drivers?
2023-05-07 17:35:31 <uniqNuuid> this is how i install realtek ethernet firmware : https://0x0.st/HXnx.sh
2023-05-07 17:35:49 <minimal> bootloader: "naturally"?
2023-05-07 17:37:07 <bootloader> as in, with the extended version, will alpine attempt to install the drivers when it sees the wifi-card isn't working?
2023-05-07 17:38:13 <bootloader> i know ubuntu does that but, because of the minimalistic nature of alpine, i dint know that did searches for compatible drivers
2023-05-07 17:38:45 <minimal> uniqNuuid: so you're installing firmware from linux kernel git "master" which may not be intended to be used with the version of kernel you are actually running...rather than installing the linux-firmware-rtl_nic package...
2023-05-07 17:39:28 <uniqNuuid> minimal: i know what i am doing. that firmware was updated in 2013
2023-05-07 17:39:35 <uniqNuuid> last updated
2023-05-07 17:40:48 <minimal> bootloader: when the ISO boots it should look at hardware and load the driver - that's not the same as installing it though, the ISO runs from RAM, *then* you decide whether or not to install Alpine to disk, SSD, USB stick etc
2023-05-07 17:41:29 <minimal> uniqNuuid: you're "teaching"/showing other people bad practice though..
2023-05-07 17:43:09 <uniqNuuid> right, i should not. 
2023-05-07 17:43:25 <bootloader> minimal ah thanks for the clarification - installing was the wrong term
2023-05-07 17:44:08 <uniqNuuid> i think the extended ISO has firmware preinstalled 
2023-05-07 17:44:23 <bootloader> the broadcom driver is ignored by alpine, they suggest you download it manually via b43-fwcutter
2023-05-07 17:44:39 <minimal> uniqNuuid: yes, this is why I suggested bootloader download Extended
2023-05-07 17:45:23 <minimal> bootloader: ah, b43, did you mention that specific Broadcom device earlier? there are various Broadcom wifi devices
2023-05-07 17:45:52 <bootloader> minimal no i didnt. i forget its a special case
2023-05-07 17:46:19 <minimal> yeah there was some reason why Alpine doesn't have it - not sure if it is license related or just that it is old and unmaintained
2023-05-07 17:46:59 <bootloader> macbook man (T_T)
2023-05-07 17:48:00 <uniqNuuid> bootloader: what is the filename of firmware are you downloading? 
2023-05-07 17:48:06 <uniqNuuid> is it .sys?
2023-05-07 17:49:34 <bootloader> uniqNuuid i dont believe so, its `broadcom-wl-5.100.38.tar.bz2`
2023-05-07 17:49:48 <uniqNuuid> well, untar it 
2023-05-07 17:49:56 <uniqNuuid> see what is inside
2023-05-07 17:50:13 <uniqNuuid> i guess it is xxx.sys
2023-05-07 17:50:19 <bootloader> i'm also manually downloading the b43-fwcutter package and installing it locally
2023-05-07 17:50:41 <bootloader> i'll let you know asap
2023-05-07 17:51:43 <uniqNuuid> bootloader: well there is also : https://github.com/AdityaGarg8/Apple-Firmware/releases/download/Montery/firmware.tar.xz
2023-05-07 17:52:02 <uniqNuuid> it is already extracted by that fwcutter
2023-05-07 17:52:25 <bootloader> uniqNuuid youre just a life-saver man
2023-05-07 17:52:54 <uniqNuuid> i don't know if that's what you need, there are so many files. well just put them all in  /tmp/firmware
2023-05-07 18:18:35 <bootloader> uniqNuuid it worked. or at the very minimum, modprobe detected my b43 firmware
2023-05-07 18:19:46 <bootloader> that firmware list wiill definitely come in handy in the future. I checksum'd some of them and they seem legit
2023-05-07 18:20:43 <bootloader> I opted to use the b43-cutter this time around because there were an excess of unused files, but ill dig in there when needed
2023-05-07 18:22:52 <scorpion2185[m]1> can you hibernate with a swap file?
2023-05-07 18:25:29 <scorpion2185[m]1> does https://www.maartenbaert.be/simplescreenrecorder/ work/exist here?
2023-05-07 18:26:24 <Ermine> pkgs.alpinelinux.org
2023-05-07 18:26:50 <scorpion2185[m]1> apk doesn't find it and I have all repos
2023-05-07 18:27:08 <Ermine> Maybe it isn't in repos
2023-05-07 18:28:12 <scorpion2185[m]1> no aarch64 it seems
2023-05-07 18:29:00 <scorpion2185[m]1> I searched on arm64 device
2023-05-07 18:31:47 <scorpion2185[m]1> >disabled .. compling problems
2023-05-07 18:32:03 <Ermine> Apparently it doesn't work
2023-05-07 18:32:38 <scorpion2185[m]1> is there somethinff for arm64?
2023-05-07 18:35:08 <Ermine> pkgs.alpinelinux.org
2023-05-07 18:52:08 <zcrayfish> scorpion2185[m]1: yes, you can absolutely hibernate with a swap file. You will likely have to use both the resume and resume_offset kernel command line parameters.
2023-05-07 19:13:11 <scorpion2185[m]1> do you specify that on grub config? but it is  for aarch64 device with uboot
2023-05-07 19:14:39 <minimal> those are cmdline options that whichever bootloader you use (i.e. uboot) would pass to kernel
2023-05-07 19:18:35 <scorpion2185[m]1> i see I need to search how but with FDE I bet it won't be easy
2023-05-07 20:53:10 <Guest781> I'm having trouble reading a QR code. When I use zbarimg command, I get error: NoDecodeDelegateForThisImageFormat: 'PNG', but I cannot figure out how to fix this
2023-05-07 20:53:34 <Guest781> Does anyone else have an easy solution for decoding QR codes?
2023-05-07 21:01:33 <Saijin_Naib> Does it need to be CLI, or can you use a GUI app via Flatpak? 
2023-05-07 21:01:53 <Guest781> Nevermind, I realized that in order for zbarimg to work, I needed to install imagemagick
2023-05-07 21:01:58 <Guest781> I think it is working now
2023-05-07 21:02:09 <Saijin_Naib> Nice! 
2023-05-07 21:25:51 <lopid> dependency-not-a-dependency
2023-05-07 21:44:14 <bootloader> another question, i'm attempting to start iwd-daemon and it is being halted by netconfig
2023-05-07 21:45:50 <bootloader> `netconfig: network configuration is disabled`, i've looked through configs for iwd and enabledconfigs
2023-05-07 21:46:46 <bootloader> however the error persists and after looking through multiple config files and packages im not sure why this would be the case
2023-05-07 21:47:48 <bootloader> im am currently installing alpine 3.12, so I am working from a live-cd 
2023-05-07 22:21:47 <bootloader_> figured out the issue, it was a silly one, I didn't uncomment the config header (i.e [General]) so the config wasn't registered
2023-05-08 01:10:18 <minimal> bootloader: why are you installing Alpine 3.12?
2023-05-08 01:10:31 <minimal> rather than 3.17, the current release
2023-05-08 01:38:06 <bootloader> minimal 3.17 was giving me problems during the boot-process
2023-05-08 01:40:18 <bootloader> particularly when booting from the live-cd, it wouldn't complete the process saying `read only filesystem`
2023-05-08 01:41:39 <bootloader> so I figured I would grab a boot that was known to work on my device (3.12) and upgrade Alpine through APK
2023-05-08 01:44:05 <bootloader> For anyone that can answer, I have another question. Does anyone know where to find drivers for broadcom-wl with the PCI-ID of "14e4:43a0"?
2023-05-08 01:45:13 <bootloader> Unfortunately, it isn't supported by b43-fwcutter. And the Mac-drivers sent earlier were for macbook16,1 and are not backwards-compatible
2023-05-08 02:28:34 <bootloader> found the driver after some research, its STA(wl) from Broadcom themselves (seems to be free, but not open-source). unfortunately on alpine, this requires a manual installation - and I am not sure how the system-updates will treat the untracked driver
2023-05-08 02:29:42 <bootloader> this is for those using alpine on a pre-2018 macbook
2023-05-08 03:01:00 <bootloader> this link was insanely annoying to find, so ill put it here in its raw-form for reference `https://www.broadcom.com/site-search?filters[pages][content_type][type]=and&filters[pages][content_type][values][]=Downloads&page=1&per_page=10&q=BCM4360`
2023-05-08 03:01:49 <bootloader> that is where to obtain the broadcom-driver, with its README and all
2023-05-08 04:34:04 <eex0> greetings all, 18 yrs and I'm back on irc.. great things never die! 
2023-05-08 04:43:59 <psykose> don't they
2023-05-08 04:49:01 <eex0> I've watched you work @psykose .. (not a stalker) but you get things done, I've seen.
2023-05-08 04:49:20 <psykose> do i
2023-05-08 04:57:16 <pj> do you
2023-05-08 04:57:59 <psykose> nope
2023-05-08 05:26:26 <JohannesJ[m]> you do!
2023-05-08 05:35:54 <eex0> so humble
2023-05-08 06:09:11 <uniqNuuid> psykose: hats off 
2023-05-08 06:09:43 <uniqNuuid> you are the best bee i've ever seen
2023-05-08 06:10:35 <uniqNuuid> bootloader: you may want to write a mabook installation wiki? 
2023-05-08 06:15:09 <bootloader> uniqNuuid I would love to
2023-05-08 06:17:12 <bootloader> I'm likely going to release a shell script to perform an offline wifi-driver installation for Mac, on the non-zero chance I do this again
2023-05-08 06:18:15 <uniqNuuid> bootloader: i mean on wiki.alpinelinux.org 
2023-05-08 06:18:31 <uniqNuuid> shell script is good, instructions i better
2023-05-08 06:18:55 <bootloader> but another question, how do you download packages in apk and have them wait to install at the end?
2023-05-08 06:19:30 <uniqNuuid> apk has a dowonload only option i think
2023-05-08 06:19:43 <bootloader> uniqNuuid yeah the wiki would be for alpine. the shell part would be a git-repo somewhere
2023-05-08 06:21:39 <bootloader> alright neat, ill check through the apk to find it
2023-05-08 06:23:27 <uniqNuuid> apk fetch bootloader 
2023-05-08 06:28:07 <bootloader> uniqNuuid thank you, much appreciated
2023-05-08 07:34:25 <scara> How come I can't specify a port like in this alpine example for awall https://wiki.alpinelinux.org/wiki/Zero-To-Awall#SSH_to_another_Host ? I get the error "Type mismatch: 'filter.1.service' should be string, is table", my config is https://envs.sh/h9F.json and what I want to acomplish is the inbound blocking of port 8080
2023-05-08 07:35:20 <ikke> scara: might be a bug in the validation (strict validation is quite recent)
2023-05-08 07:35:44 <ikke> I think I ran into it as well 
2023-05-08 07:39:12 <scara> ikke: I'm going to try to add a workaround service but I couldn't find any real documentation, that alpine wiki articial is all I've got.
2023-05-08 07:43:45 <scara> No error but it doesn't work, port 8080 is still accessable. No idea where to begin debuging but here is my config https://envs.sh/h9t.json sorry that I'm kinda asking to be spoonfed but I really don't understand this documentation or lacking thereof, maby I'm looking the wrong place?
2023-05-08 07:44:04 <scara> I mean, there is no awall-doc package either.
2023-05-08 07:46:20 <eex0> excuse my ignorance but awall and ufw are pre-process tools to iptable rules? or meant to replace iptables in production?
2023-05-08 07:47:32 <pj> awall and ufw are sane frontends for iptables
2023-05-08 07:48:51 <eex0> thank you pj, I had that assumption but for clarification, had to ask
2023-05-08 07:49:50 <scara> Couldn't we just make iptables good then? Thats what OpenBSD did with their pf, kinda wish alpine would use pf (:
2023-05-08 07:50:17 <pj> you are free to go and make iptables good
2023-05-08 07:51:09 <scara> Sounds like too much work when I can't even get awall to work, I'd just hit ctrl+c ctrl+v on that pf code like thay say in the church.
2023-05-08 07:51:50 <scara> https://en.wikipedia.org/wiki/File:Kopimizm.svg
2023-05-08 07:52:13 <pj> eventually you might discover nftables
2023-05-08 07:52:29 <scara> What dat?
2023-05-08 07:52:59 <pj> "nftables is the successor of iptables, it allows for much more flexible, scalable and performance packet classification. This is where all the fancy new features are developed."
2023-05-08 07:53:52 <scara> Shouldn't awall be a frontend for nftables then?
2023-05-08 07:54:02 <scara> Or could I use nftables directly?
2023-05-08 07:54:17 <eex0> glancing at cloudflare code, bpf will rule the firewall
2023-05-08 07:54:28 <eex0> eventually
2023-05-08 07:54:32 <ikke> awall does not support nftables yet
2023-05-08 07:54:51 <scara> What about I rent a server on the other side of the world and use that as a firewall for this vps? It could pf ;)
2023-05-08 07:55:06 <pj> if you love pf so much, go ahead
2023-05-08 07:55:30 <eex0> pf = pfSense?
2023-05-08 07:55:45 <pj> packet filter
2023-05-08 07:55:48 <scara> Just jokeing, I want to learn alpine's utils because I run alpine now, it has its ups and downs, mainsly awall.
2023-05-08 07:55:49 <eex0> ah
2023-05-08 07:56:07 <pj> which pfsense builds on top of
2023-05-08 07:56:19 <pj> (hence the name)
2023-05-08 07:56:26 <eex0> love my netgate hardware, here!
2023-05-08 07:56:41 <scara> Dis pf https://man.openbsd.org/pf
2023-05-08 07:57:22 <scara> Someone pls tell me how I could get awall to tell me whats wrong.
2023-05-08 07:57:31 <scara> It wont block my ports.
2023-05-08 07:57:54 <mercenary> If you are used to working with pf, it makes more sense to work directly with iptables or nftables, instead of using these 'frontend' kludges
2023-05-08 07:57:57 <scara> Ohhhhhhh, Warning: firewall not enabled for inet
2023-05-08 08:01:39 <eex0> at a glance scara, awall is a whitelist firewall.. only "accept" in/out ports you require thus "blocking" everything else?
2023-05-08 08:05:21 <eex0> looks neat to me:
2023-05-08 08:05:25 <eex0>     	{
2023-05-08 08:05:25 <eex0>             "service": [ "dns", "http", "https", "ssh", "ntp", "ping" ],
2023-05-08 08:05:25 <eex0>             "in": "_fw",
2023-05-08 08:05:25 <eex0>     "filter": [
2023-05-08 08:05:25 <eex0>             "out": "internet",
2023-05-08 08:05:26 <eex0>             "action": "accept"
2023-05-08 08:05:29 <eex0> 	}
2023-05-08 08:08:00 <eex0> might be the easiest firewall I've seen.. kudos to dev's
2023-05-08 08:12:15 <ikke> eex0: you can set the default policy, so it can work either way 
2023-05-08 08:13:38 <quinq> “easiest”, that's debatable
2023-05-08 08:13:45 <eex0> I was just looking at that
2023-05-08 08:13:51 <eex0>   "policy": [
2023-05-08 08:13:51 <eex0>     { "action": "reject" }
2023-05-08 08:13:51 <eex0>   ]
2023-05-08 08:13:51 <eex0>     { "in": "internet", "action": "drop" },
2023-05-08 08:13:58 <quinq> You “just need” to write jsons for configuring your rules
2023-05-08 08:16:37 <Apachez> iptables -i internet -j drop       is easier :)
2023-05-08 08:18:35 <eex0> took me what seemed a month to setup pfsense but thats got ids and some crazy "if this" then "that" on "what" machine to which lan and rules for (coughs) you name it.. I'm scared to go back in there now..lol
2023-05-08 08:19:07 <quinq> I don't know pfsense, but that sounds crazy indeed
2023-05-08 08:19:56 <quinq> Isn't though “if this then that” just a regular match?
2023-05-08 08:20:32 <Apachez> dunno if it changed but doesnt both pf and ipf do "best match" rather than "first match" when it comes to rules?
2023-05-08 08:21:16 <Apachez> like if rule 1 is   srcip=192.168.0.0/24 action: allow     and rule2 is  srcip=192.168.0.1/32 action: drop   
2023-05-08 08:21:29 <Apachez> and packet arrives who is srcip 192.168.0.1 then it will be dropped by rule 2
2023-05-08 08:21:59 <Apachez> where iptables and cisco/arista/most other switches/routers acl's would have a hit on rule1 and allow the packet (and never examine rule2)
2023-05-08 08:23:24 <quinq> No, they do last match
2023-05-08 08:23:31 <quinq> (at least pf)
2023-05-08 08:23:50 <quinq> You can force a match whenever you want though
2023-05-08 08:24:24 <Apachez> yeah "best match"
2023-05-08 08:24:30 <Apachez> where /32 will win over /24 and so on
2023-05-08 08:24:35 <quinq> It's not best, it's last
2023-05-08 08:24:35 <Apachez> even if the /32 rule comes way later
2023-05-08 08:24:58 <eex0> pfsense: MATCH rule applies immediately, it doesn't stop the rule processing, therefore the evaluation continues till the end of the ruleset.
2023-05-08 08:24:59 <Apachez> so a rule 3 acting on 192.168.0.0/25 action: allow   would win ?
2023-05-08 08:25:14 <quinq> If it matches, yes
2023-05-08 08:25:23 <Apachez> well the packet is 192.168.0.1
2023-05-08 08:25:30 <Apachez> to my knowledge best match was being used but sure
2023-05-08 08:25:35 <Apachez> either way its retarded :P
2023-05-08 08:25:41 <Apachez> top-down is the way to do things =)
2023-05-08 08:25:41 <quinq> Again, there's no best match.
2023-05-08 08:26:10 <Apachez> troubleshooting ipf/pf rulesets (specially larger ones) is a nightmare
2023-05-08 08:26:23 <Apachez> compared to lets say iptables who does first match top down
2023-05-08 08:26:44 <quinq> Well, again, if you prefer that behavior you can do it
2023-05-08 08:26:48 <eex0> then there si that whole QUICK Floating rules thing taht twist on my skull
2023-05-08 08:27:12 <quinq> And there's a log rule that makes debugging easier
2023-05-08 08:27:25 <eex0> it's 1:30aam excuse the dyslexic behavior
2023-05-08 08:28:21 <quinq> It's all excused :)
2023-05-08 08:31:48 <Apachez> eex0: when you start to type like Duffy Duck its time for bed ;-)
2023-05-08 08:32:06 <Apachez> Daffy Duck
2023-05-08 08:32:26 <eex0> I'm heading that way now.. you all have a pleasant morning/day,evening where-ever you are
2023-05-08 09:55:51 <quinq> haha Apachez 
2023-05-08 12:55:34 <kn> hi, can someone point me at an APKBUILD that builds a kernel module?
2023-05-08 12:56:55 <minimal> kn: zfs-lts
2023-05-08 13:01:25 <kn> minimal: that looks great, thanks!
2023-05-08 13:19:47 <JohannesJ[m]> heh., even wireguard works with alpine/kde :O
2023-05-08 13:26:48 <scara> Does awall really not have the ability to block incomeing connections on a specific port?
2023-05-08 13:30:46 <JohannesJ[m]> heh, why wouldn't it?
2023-05-08 13:31:16 <scara> The port I want to block won't block.
2023-05-08 13:33:18 <JohannesJ[m]> https://wiki.alpinelinux.org/wiki/Zero-To-Awall see here, especially the SSH Service part, but the action would be drop instead of accept and instead of service and ssh you would put port and  i think, have you tried that?
2023-05-08 13:33:42 <scara> I think so.
2023-05-08 13:33:46 <JohannesJ[m]> atually, ifd you follow that page, drop would be the default action for all ports
2023-05-08 13:33:49 <scara> https://envs.sh/h9t.json
2023-05-08 13:34:36 <JohannesJ[m]> hm, that seems correct, but like i said.. follow the above wiki page, and all ports should be blocked anyway
2023-05-08 13:35:40 <scara> I will double check the wiki
2023-05-08 13:45:11 <scara> Best guess: modprobe: FATAL: Module ip_tables not found in directory /lib/modules/6.1.22-0-virt
2023-05-08 13:46:41 <JohannesJ[m]> ahhh
2023-05-08 13:46:53 <JohannesJ[m]> yo8u need to install ip tables i guess? let me see
2023-05-08 13:48:20 <minimal> nope, that's a kernel module that should be present, just not in that directory (it's in kernel/net/ipv4/netfilter below there)
2023-05-08 13:57:56 <scara> Well I did modprobe ip_tables
2023-05-08 13:58:04 <scara> and iptables is installed
2023-05-08 13:58:46 <minimal> scara: so where did the above error appear then?
2023-05-08 14:00:34 <JohannesJ[m]> oh! thankfully minimal is here :D
2023-05-08 14:22:44 <scara> minimal: did that modprobe, got that error, iptables installed, I know nothing else.
2023-05-08 14:24:45 <christian> Hi there, what is the correct and easy way to install pecl under Alpine Linux v3.17 ? 
2023-05-08 14:25:15 <christian> The php ds extension does not seem to be in the sources
2023-05-08 14:29:37 <ikke> You might need to reboot if you updated the kernels 
2023-05-08 14:31:08 <ikke> christian: apk add pho81-pear
2023-05-08 14:31:21 <ikke> php*
2023-05-08 14:31:50 <christian> @ikke: ok that was easy :)
2023-05-08 14:36:21 <christian> Thx ikke, worked like a charme. apk add php81-dev php81-pear && pecl install ds
2023-05-08 14:38:06 <AnyCreatureMorph[m]> How do I get gnome 44?
2023-05-08 14:40:34 <pj> by using edge or waiting for next alpine release
2023-05-08 14:51:08 <kn> following https://wiki.alpinelinux.org/wiki/Creating_an_Alpine_package i have an APKBUILD which builds a kernel module for -lts and -virt kernels, but 'abuild' fails with 'ERROR: foo-virt-6.1.72-r2.apk: package file format error' followed by '>>> ERROR: foo-lts: Failed to create index'
2023-05-08 14:51:49 <kn> i don't see the wiki going into this... are there any logs or ways to get more info out of abuild's failure?
2023-05-08 15:25:58 <somercet1> apk wants to install busybox-mdev-openrc, this caused problems before I removed it. Is it compatible with udevd, or do I need to switch, or...?
2023-05-08 17:05:37 <scara> I know its not recommended but is it possible to install alpine-linux without a swap partition and also fully encrypted?
2023-05-08 17:06:18 <atka> yes, you can also use a swap file while using FDE
2023-05-08 17:07:55 <scara> atka: I know but I do not want a swap partition.
2023-05-08 17:08:23 <atka> ok.. then don't use a swap partition
2023-05-08 17:09:38 <scara> atka: Last time I installed alpine it wasn't an option, I asume it isn't now either.
2023-05-08 17:10:33 <atka> set up your partitions manually or do a chroot install
2023-05-08 17:11:06 <atka> you don't have to use all of the setup-alpine scripts, or any of them if you don't want to
2023-05-08 17:12:40 <scara> Oh :)
2023-05-08 19:06:39 <a3e3a[m]> Hi! I have a joystick id like to connect to alpine linux to play games. It does not auto detect. Could someone tell me what i would need? Thank you.
2023-05-08 19:08:23 <ikke> a3e3a[m]: What kind of joystick is it?
2023-05-08 19:10:24 <a3e3a[m]> ikke: https://github.com/thnikk/GP2040
2023-05-08 19:20:14 <ikke> a3e3a[m]: do you see messages appear in dmesg when you connect it?
2023-05-08 20:11:10 <bootloader> hey everyone, I'm back today again, working on the wifi-driver install for pre-2018 macbooks. however, i've noticed that the STA(wl) driver is actually very outdated - effectively meaning I have to patch it myself
2023-05-08 20:12:17 <bootloader> now i'm always down for a good patch, but I really would prefer to avoid it as I am not very proficient in C 
2023-05-08 20:13:42 <bootloader> I say all this to ask, does anyone know where the updated STA(wl) driver would be? The one on the website is obsolete, and when I when to the latest ubuntu-package for the driver it had the same code
2023-05-08 20:16:01 <bootloader> I would imagine this shouldn't be the case, since the API-calls are straight  up wrong, but it seems like ubuntu has some other method of correcting it
2023-05-08 20:36:43 <a3e3a[m]> <ikke> "a3e3a: do you see messages..." <- I'll have to come back my stick is not responding. Thanks again!
2023-05-08 20:54:11 <iggy> scara: I'm pretty sure I just do -s0 or SWAP_SIZE=0 (env var) for setup-disk and it does the right thing... I will say though that the docs are wrong about some of the env var names for setup-disk (f.ex. setup-disk uses BOOT_SIZE and docs say BOOTSIZE)
2023-05-08 22:28:13 <smrtz> Heyo! I'm trying to use easy-rsa to generate an ed35519 set of creds, and it seems the latest openssl version in Alpine doesn't support ed25519?
2023-05-08 23:41:59 <minimal> scara: as iggy said you can run "SWAP_SIZE=0 setup-disk" and no swap partition will be created
2023-05-09 03:51:19 <uniqNuuid> bootloader: when in doubt, search AUR/archlinux
2023-05-09 03:52:50 <uniqNuuid> bootloader: https://raw.githubusercontent.com/archlinux/svntogit-community/packages/broadcom-wl/trunk/PKGBUILD
2023-05-09 03:54:16 <uniqNuuid> n2
2023-05-09 04:26:38 <bootloader> uniqNuuid oh that's massive, I'll check the package contents for the driver. Hopefully, their wl.ko package is compatible with my kernel version
2023-05-09 04:27:43 <bootloader> this whole mirror is a gem for drivers actually, niceeeee
2023-05-09 04:29:23 <bootloader> Thank you for the resource
2023-05-09 07:09:44 <scara> minima: :OOOOO thanks!
2023-05-09 07:27:25 <uniqNuuid> scara: setup-alpine.conf: DISKOPTS="-s 0 -m sys"
2023-05-09 08:24:24 <JohannesJ[m]> hmmz
2023-05-09 08:44:20 <consus[m]> Guys, is baremetal  autoinstall possible rn? Built-in or cloud-init perhaps.
2023-05-09 08:44:48 <JohannesJ[m]> alpine setup allows for an answer file
2023-05-09 08:45:24 <consus[m]> Sweet! Could you share a link to the uptodate documentation?
2023-05-09 08:46:23 <JohannesJ[m]> hang on :) im working atm :)
2023-05-09 08:46:39 <consus[m]> Thx :)
2023-05-09 08:47:07 <JohannesJ[m]> https://docs.alpinelinux.org/user-handbook/0.1a/Installing/setup_alpine.html
2023-05-09 08:50:15 <JohannesJ[m]> the handbook is a good start for everything :)
2023-05-09 13:03:42 <Habbie> is anyone aware of busybox plans for copying this madness? https://www.phoronix.com/news/GNU-Grep-3.8-Stop-egrep-fgrep
2023-05-09 14:22:53 <pj> Where is that madness? I cannot find it
2023-05-09 14:25:00 <frag> https://0x0.st/HZ0_.jpg
2023-05-09 14:27:40 <minimal> consus[m]: Not sure exactly what you're looking for - cloud-init is not itself an installer, it is a on-boot configuration tool
2023-05-09 14:28:26 <minimal> consus[m]: there is work in progress in Alpine for an automated installer using "cloud-init"-like meta-data/user-data files
2023-05-09 14:29:17 <minimal> consus[m]: myself, I create pre-prepared disk images for physical machines ready to be "dd"-ed to their disks/SSDs along with cloud-init for configuration upon 1st boot
2023-05-09 14:32:46 <minimal> frag: yes I used to be able to tell which baud rate was being used by listening to the initial modem handshake "screeches" ;-)
2023-05-09 14:33:09 <frag> hehe
2023-05-09 14:35:38 <minimal> too much time spent running an ISP dealing with 9.6, 14.4, 19.2, 28.8, 33.6, and 56 modems lol
2023-05-09 14:40:50 <minimal> consus[m]: didn't you previously ask this a year ago? ;-)
2023-05-09 14:45:23 <frag> BUSTED
2023-05-09 14:54:56 <consus[m]> Yeah
2023-05-09 14:55:19 <consus[m]> Just reiterating since alpine is now relevant again :)))
2023-05-09 14:55:32 <minimal> consus[m]: relevant again? what made it not relevant before?
2023-05-09 14:55:35 <consus[m]> (in my org)
2023-05-09 14:56:26 <minimal> consus[m]: the installer ISO "cloud-init"-like stuff *might* be in the 3.18 release due anytime now, there is some last minute ongoing on it
2023-05-09 14:59:57 <consus[m]> E.g. ununtu uses cloud-init instead of preseed these days
2023-05-09 15:00:04 <consus[m]> Currently I do the same and dd the image, however I would like to unify the image creation
2023-05-09 15:00:15 <consus[m]> Because all other images created with packer
2023-05-09 15:00:51 <consus[m]> And typing answers through vnc is hacky
2023-05-09 15:00:52 <minimal> consus[m]: Ubuntu uses curtin/subiquity (which then make use of cloud-init plus other stuff)
2023-05-09 15:01:50 <minimal> so technically it's not correct to say Ubuntu uses cloud-init - the YAML is a superset of cloud-init's
2023-05-09 15:02:41 <consus[m]> True
2023-05-09 15:03:15 <minimal> which is similar to how the intention is for Alpine to use a "cloud-init"-like YAML of its own
2023-05-09 15:03:59 <minimal> so once that is done then you could use Packer to boot an Alpine ISO and provide another ISO with the YAML files to drive the installation
2023-05-09 15:04:16 <consus[m]> Noice
2023-05-09 15:04:33 <consus[m]> Can i use http?
2023-05-09 15:04:52 <consus[m]> To provide the yaml file
2023-05-09 15:05:18 <minimal> still stuff is still work-in-progress, there has been some work for network retrieval but think current focus is on ISO
2023-05-09 15:05:38 <minimal> as I said this is work-in-progress, so give it time
2023-05-09 16:13:00 <JohannesJ[m]> i wonder if i can "squashfs" my desktp install, and then install it on other desktops?
2023-05-09 16:13:08 <JohannesJ[m]> like a crude GUI installer of some sort
2023-05-09 20:06:12 <ncopa> consus[m]: we added very basic support for tiny-cloud on release iso images. For now you can only set hostname and add ssh key to the default user (alpine), more or less
2023-05-09 20:06:57 <ncopa> well, you can provide an /etc/network/interfaces file, enable ntp, configure apk cache and apk repositories and install packages
2023-05-09 20:07:27 <ncopa> and run custom commands via `bootcmd` and `runcmd` as in NoCloud from cloud-init
2023-05-09 20:07:30 <ncopa> but thats it
2023-05-09 20:07:36 <ncopa> no support for disk install yet
2023-05-09 20:08:21 <ncopa> but you can hack around it by executing `setup-disk` from runcmd
2023-05-09 20:08:33 <ncopa> the user-data can also be a shell script, which will be executed
2023-05-09 21:16:09 <rsc> It seems like HAProxy in Alpine 3.17 is broken once it comes to fcgi-app and Unix socket, because TCP socket works. Is that already known?
2023-05-09 21:17:17 <ikke> rsc: Haven't heard about it and don't see any open issues
2023-05-09 21:21:11 <rsc> ikke: I see. Sounds like I should get a minimal reproducer then?
2023-05-09 21:21:22 <ikke> rsc: would be helpful
2023-05-09 21:23:19 <rsc> Btw, congrats to 3.18!
2023-05-09 21:23:59 <ikke> Thanks! The actual anouncement is ocming
2023-05-09 21:24:24 <ikke> https://www.alpinelinux.org/posts/Alpine-3.18.0-released.html
2023-05-09 21:37:22 <rsc> Btw, the lighttpd package could be more lightweight...
2023-05-09 21:40:13 <ikke> for the most lightweight function we use darkhttpd
2023-05-09 21:42:05 <rsc> But that one doesn't support fcgi, right?
2023-05-09 21:42:37 <ikke> correct
2023-05-09 21:42:50 <ikke> just a static webserver
2023-05-10 01:45:55 <Saijin_Naib> Sweet! Maybe time to drop my laptop back to latest-stable
2023-05-10 03:39:28 <uniqNuuid> upgrading to 3.18 
2023-05-10 03:40:06 <eex0> nice!
2023-05-10 03:41:15 <uniqNuuid> apk tools does not depend on openssl?
2023-05-10 03:41:58 <eex0> I had to "add" openssl on 3.18 edge
2023-05-10 03:42:37 <uniqNuuid> as i see the first 4 upgrades do not include openssl
2023-05-10 03:43:19 <eex0> recall seeing an issue where openssl while in cve was giving doubt? if memory serves correct
2023-05-10 03:44:16 <eex0> iow: insecurity
2023-05-10 03:46:00 <eex0> could hav been libressl on second thought that there was issue.. anyway.. "add" it if needed? 
2023-05-10 03:47:40 <uuidNuniq> done upgrading 
2023-05-10 03:47:54 <uuidNuniq> great, no issue what so ever
2023-05-10 03:48:25 <eex0> Ive been running edge for over a mnth and just can't get over how stable and few issues
2023-05-10 03:48:42 <uuidNuniq> well, edge is another story
2023-05-10 03:48:51 <uuidNuniq> it is a test bed
2023-05-10 03:49:17 <uuidNuniq> but the stable is really stable, damn, super stable. 
2023-05-10 03:49:18 <eex0> nods
2023-05-10 03:49:38 <eex0> I'll be putting that on a mchine tonight
2023-05-10 03:50:49 <uuidNuniq> when psykose pushes ncurses, you'll know
2023-05-10 03:54:34 <eex0> libdrm yesterday made me crazy for a brief minute.. come to find out, I had hacked my way around what it was meant to fix.. rolling back wasnt hard tho.
2023-05-10 06:33:02 <JohannesJ[m]> i wont upgrade just yet, im so glad i finally got alpine working perfectly :D
2023-05-10 07:33:28 <eex0> youtube video's were shredding my graphics, fans sounding like jet airplanes taking off until I discovered turning "ambient mode" off in the youtube video player itself. jeebus h. now why didn't i know this mnths ago..lol
2023-05-10 07:33:46 <eex0> sorry wrong channel .. meant for offtopic
2023-05-10 08:49:16 <bootloader> hey everyone, back again
2023-05-10 08:50:05 <bootloader> I've recently attempted using the arch-linux kernel module on alpine and got an "invalid module format"
2023-05-10 08:51:05 <lopid> quelle surprise
2023-05-10 08:51:14 <bootloader> I presume this was compiled on arch with the proper compiler-flags because its a kernel module in-and-of itself
2023-05-10 08:51:28 <lopid> presumably
2023-05-10 08:51:38 <bootloader> any suggestion to what may be causing this error
2023-05-10 08:52:03 <lopid> you just said it
2023-05-10 08:52:20 <lopid> use an alpine kernel module instead
2023-05-10 08:52:31 <bootloader> so the compiler-flags are off guaranteed?
2023-05-10 08:52:46 <lopid> it's a different kernel
2023-05-10 08:53:11 <bootloader> I'm using the broadcom-wl module, so unfortunately I can't get an alpine-version
2023-05-10 08:53:38 <lopid> is the source available?
2023-05-10 08:53:44 <bootloader> oh, I was under the impression that the underlying kernel was the same
2023-05-10 08:54:14 <bootloader> yeah, but its extremely outdated
2023-05-10 08:54:34 <lopid> arch and alpine are different distros. they may both use the linux kernel, may both be the same version even, but they're built separately
2023-05-10 08:54:40 <bootloader> I was self-patching it up until yesterday, when I got wind that the arch version may work on alpine
2023-05-10 08:55:36 <bootloader> okay thank you, its so painful (T_T)
2023-05-10 08:56:24 <lopid> there may be something i'm missing. who said the arch module may work on alpine?
2023-05-10 08:56:25 <bootloader> back to patching then
2023-05-10 08:57:29 <bootloader> uniqNuuid , but it was only a suggestion and not a guarantee
2023-05-10 08:58:33 <lopid> maybe they meant get the source from arch
2023-05-10 08:58:57 <lopid> use their build options as a guide, etc
2023-05-10 08:59:33 <bootloader> why didn't I think of that lol
2023-05-10 09:00:45 <bootloader> I'll check out the source then. I think it was that ubuntu/fedora had outdated sources - so my brain defaulted - but it was a lapse of judgement
2023-05-10 09:00:52 <bootloader> Thank you for the tip
2023-05-10 09:29:02 <kaathewise> Good day!  After I upgraded to 3.18, sound in firefox stopped working for me.  Previosly apulse would help, but now it does not work either.  Is there a guide for firefox sound somewhere?  I can't find anything on wiki unfortunatelly
2023-05-10 09:34:03 <Misthios> do u use pipewire? if yes pipewire-media-session was removed
2023-05-10 09:34:10 <ikke> kaathewise: do you get an error message?
2023-05-10 09:34:11 <kaathewise> Nope, only ALSA and nothing else
2023-05-10 09:34:40 <kaathewise> @ikke: apulse sends me away with `do_connect_pcm: can't open capture device "default". Error code -2 (No such file or directory)`
2023-05-10 09:37:05 <kaathewise> Hold up, my bad.  It seems all sound on my laptop died, not just firefox
2023-05-10 09:41:07 <kaathewise> Yep, I apologize, it was my bad.  After updating the ALSA config file got reset.  I fixed it and everything went back to normal
2023-05-10 09:42:02 <ikke> 👍
2023-05-10 09:42:07 <Misthios> 👍
2023-05-10 09:54:41 <piel> Hi, any suggestions how to get GRUB2 to recognise Alpine correctly? I have an old PC with 4 distros + win98 on it. os-prober finds Alpine kernel and adds it to the menu, but i have to add the initrd and rootfstype manually. I edited the grub menu but  i will only last to the next update this way. 
2023-05-10 12:24:27 <axx> I'm having trouble starting xorg on an Alpine VM, (using Xen) the error says MESA-LOADER: failed to open bochs-drm
2023-05-10 12:24:36 <axx> aditionally kmsro: driver missing
2023-05-10 12:25:13 <axx> any tips on what to do?
2023-05-10 13:10:54 <minimal> piel: os-prober is separate software from Grub. Assuming you're referring to a non-Alpine version of os-prober then it is likely it doesn't know how to "find" Alpine's kernel and initramfs files are they are named differently than in most distros
2023-05-10 13:30:19 <uniqNuuid> /etc/grub.d/40_custom
2023-05-10 13:32:16 <minimal> uniqNuuid: that doesn't seem relevant to his issue
2023-05-10 13:33:40 <uniqNuuid> does it work?
2023-05-10 13:33:45 <uniqNuuid> yes. it works
2023-05-10 13:34:34 <minimal> he has a "problem" with os-prober so he can fix os-prober to resolve that
2023-05-10 13:34:40 <uniqNuuid> alpine's vmlinuz and initramfs does not have a version attached, so we can use static grub config.
2023-05-10 13:34:59 <uniqNuuid> don't you know alpine's grub is patched? 
2023-05-10 13:35:23 <minimal> uniqNuuid: I'm well aware they don't have a version attached, that why I previously said "Alpine's kernel and initramfs files are they are named differently than in most distros"
2023-05-10 13:36:06 <minimal> uniqNuuid: "don't you know alpine's grub is patched?" What specifically are you referring to here?
2023-05-10 13:37:05 <uniqNuuid> i mean a normal grub won't be able to find alpine's kernel and initramfs pair.
2023-05-10 13:37:53 <uniqNuuid> so. it's better just use /etc/grub.d/40_custom
2023-05-10 13:38:19 <minimal> uniqNuuid: normally it is os-prober that finds other distros
2023-05-10 13:39:52 <uniqNuuid> yes, but it can't find initramfs. 
2023-05-10 13:40:05 <uniqNuuid> patching is it not easy
2023-05-10 13:40:23 <minimal> he mention he has 4 distros installed and so it is likely he is NOT using Alpine's Grub but rather Grub from one of the other distros, and he is already using os-prober, so it makes sense to fix that os-prober
2023-05-10 13:41:00 <uniqNuuid> os-prober can't find alpine's initramfs as he stated
2023-05-10 13:41:24 <uniqNuuid> then why not just use a static config
2023-05-10 13:41:39 <minimal> patching it IS easy, os-prober is a shell scripy
2023-05-10 13:42:12 <uniqNuuid> ok, shell script, what about next upgrade, will it get overwritten?
2023-05-10 13:42:52 <minimal> another "hacky" solution could be to create softlinks for os-prober to find
2023-05-10 13:43:25 <uniqNuuid> well, well, that depends on the filesystem of the `/boot`
2023-05-10 14:29:23 <patdk-lp> I'm having an issue using abuild on 3.18, worked on 3.17 and lower, and only with arm64, works fine with amd64. I'm rebuilding the unbound package without modifications, but abuild -r is failing for with: 13 0.166 ERROR: Unable to lock database: Permission denied
2023-05-10 14:29:24 <patdk-lp> but only for arm64, cannot really figure it out
2023-05-10 14:29:39 <defec7ed> How does the Alpine team officially build their `aports` tree? What I am asking is, how do the handle the compiling for other architectures? Do they cross-compile to ARM from x64? Do they build natively on ARM with a swarm of Raspberry Pi's? Do they build natively on ARM with AWS ARM servers? I want to know more about the hardware side of things so we can replicate your process
2023-05-10 14:29:42 <defec7ed> in our office.
2023-05-10 14:30:13 <minimal> defec7ed: no cross compiling is done, it is all compiled natively
2023-05-10 14:30:35 <defec7ed> that's what I believed given that we haven't been able to cross-compile; seems broken
2023-05-10 14:30:39 <minimal> as in aarch64 is compiled on a aarch64 server, doesn't mean that RPI aarch64 is compiled on RPIs
2023-05-10 14:30:43 <patdk-lp> I think he is asking if your using native cpu instructions, or translators (qemu)
2023-05-10 14:31:09 <defec7ed> ^ and where do you find a fast, native aarch64 system to do the compiling... AWS is my best guess
2023-05-10 14:31:37 <psykose> donated altra servers in a datacenter
2023-05-10 14:31:37 <minimal> the only exceptions are that armv7 and armhf is compiled on a aarch64 server emulating those other arm arches, and riscv is also emulated AFAIK
2023-05-10 14:31:44 <patdk-lp> there are many fast arm platforms, but they are costly, unless you go for a rk3588 based system
2023-05-10 14:31:56 <patdk-lp> you can get the aws arm systems, but they are almost 10k usd
2023-05-10 14:32:31 <defec7ed> patdk-lp: by that you mean purchase, not rent?
2023-05-10 14:32:36 <patdk-lp> yes
2023-05-10 14:33:15 <patdk-lp> I have been using qemu/docker to do all my arm builds currently, but that is broken now for me in 3.18 due to the above issue I posted
2023-05-10 14:33:32 <defec7ed> same here, except we are stuck at Alpine 3.14
2023-05-10 14:33:52 <patdk-lp> odd, mine is working from 3.9 to 3.17
2023-05-10 14:34:31 <defec7ed> so it sounds like renting an AWS ARM server (or serverS, plural) is the only real path forward if we want to walk in tandem with the Alpine team
2023-05-10 14:35:08 <minimal> I also use docker+qemu-user for Arm package builds on x86_64 and haven't had any real problems in the past (though probably haven't done an Arm build for a couple of months I guess)
2023-05-10 14:35:41 <minimal> defec7ed: why are you so fixated on AWS for Arm?
2023-05-10 14:36:11 <minimal> e.g., Oracle Cloud offer Arm, Hetzner started doing so recently, there are others like Scaleway as well
2023-05-10 14:36:51 <defec7ed> because docker+qemu is painfully slow and doesn't work for all versions. I am open to other providers, what I meant was it seems going cloud is the only feasible way to replicate Alpine's build process.
2023-05-10 14:36:52 <ikke> Armv7 and V6 is not emulated, but aarch64 in 32-bits mode
2023-05-10 14:37:03 <ikke> The cpu uses supports that 
2023-05-10 14:37:25 <ikke> Same with x86
2023-05-10 14:37:36 <ikke> rv64 is fully emulated 
2023-05-10 14:57:24 <uniqNuuid> patdk-lp: apple M1/M2 is cheap
2023-05-10 14:57:37 <uniqNuuid> and fast
2023-05-10 14:58:12 <psykose> >cheap
2023-05-10 14:58:24 <abby> ""cheap""
2023-05-10 14:58:40 <dwfreed> """cheap"""
2023-05-10 14:58:51 <uniqNuuid> mac mini?
2023-05-10 14:59:32 <uniqNuuid> oh btw, oracle ARM is 4core 24G ram, always free
2023-05-10 14:59:46 <uniqNuuid> that one is fast too
2023-05-10 14:59:57 <dwfreed> yeah, but then you have to deal with Oracle Cloud
2023-05-10 15:00:07 <dwfreed> which I've heard from many people is complete garbage
2023-05-10 15:00:15 <uniqNuuid> i've installed alpine on it, works fine 
2023-05-10 15:00:19 <abby> only 8GB ram/256GB/8-core for 600$, that's not great for a build machine
2023-05-10 15:00:48 <dwfreed> more like 4 useful cores and 4 wasted cores
2023-05-10 15:01:15 <uniqNuuid> dwfreed: oracle cloud is garbage, agreed, but you don't have to deal with it. install alpine once, then good to go
2023-05-10 15:02:52 <uniqNuuid> abby: is 6 ~ 10 minites building kernel and headers fast? that's M1 8core. i build archlinux kernel on it
2023-05-10 15:03:18 <abby> ram and storage are going to be limiting factors
2023-05-10 15:03:28 <abby> try building chromium
2023-05-10 15:05:29 <dwfreed> yeah, kernel doesn't require a lot of resources to build beyond CPU; chromium and firefox on the other thand
2023-05-10 15:05:33 <psykose> chromium is fine on it too
2023-05-10 15:05:33 <dwfreed> s/thand/hand/
2023-05-10 15:06:00 <psykose> that said, if you think it's cheap, send me one
2023-05-10 15:06:18 <patdk-lp> I haven't seen any m1/m2 being cheap
2023-05-10 15:06:39 <patdk-lp> cheaper than an arm server, but you can get something like the honeycomb that is as good and cheaper than m1
2023-05-10 15:07:28 <patdk-lp> I forget how long it took my rk3588 to build a new kernel, but it was quick
2023-05-10 15:08:52 <uniqNuuid> well, never build chromium 
2023-05-10 15:10:01 <uniqNuuid> patdk-lp: are you using qemu-user-static?
2023-05-10 15:10:14 <patdk-lp> for my docker builds, yes
2023-05-10 15:14:34 <uniqNuuid> patdk-lp: musl 1.24 has some changes, DT-xxxx thing, related?
2023-05-10 15:14:58 <patdk-lp> I don't know what DT-xxxx is
2023-05-10 15:15:02 <uniqNuuid> All packages for ppc64le, x86, and x86_64 was linked with DT_RELR. This should have reduced size of compiled binares
2023-05-10 15:15:41 <uniqNuuid> maybe latest qemu-user-static have compatible issue with this DT_RELR?
2023-05-10 15:18:45 <patdk-lp> https://0bin.net/paste/4GJM93UB#URQKtkVWmo-lL54EYhldu6ACsCxwc57+SAcocJI9TGK
2023-05-10 15:18:50 <patdk-lp> that is my build script
2023-05-10 15:19:59 <patdk-lp> actually that shouldn't be the issue would it?
2023-05-10 15:20:14 <patdk-lp> everything else runs and is fine, till it runs that abuild -r command, that is way down the list
2023-05-10 15:23:27 <patdk-lp> to me, it seems like sudo isn't working right for abuild
2023-05-10 15:26:45 <bbbhltz> Hi. What are the solutions when startup hangs at "loading initial ramdisk"?
2023-05-10 15:27:21 <uniqNuuid> bbbhltz: it is likely a console issue
2023-05-10 15:27:45 <uniqNuuid> what is your hardware environment
2023-05-10 15:28:58 <uniqNuuid> bbbhltz: virtual machine? bare metal? aarch64 ? x86_64? 
2023-05-10 15:29:24 <bbbhltz> Not virtual machine, x86_64
2023-05-10 15:29:37 <bbbhltz> Lenovo ThinkPad 13 2
2023-05-10 15:30:02 <JohannesJ[m]> heh, pipewire removed?
2023-05-10 15:30:08 <psykose> usually means the graphics failed
2023-05-10 15:30:15 <bbbhltz> No pipewire
2023-05-10 15:30:45 <dewdude> i need to remember to update all my systems before i leave
2023-05-10 15:30:58 <dewdude> 3.18 is now so i bet -edge needs updating too
2023-05-10 15:31:10 <bbbhltz> I am on Edge
2023-05-10 15:31:43 <dewdude> yeah...that shouldn't be too important for your problem unless it's some kind of major graphics issue...but alpine doesn't initialize VGA the way some distros do
2023-05-10 15:31:53 <dewdude> i take that back. i've never run alpine bare metal
2023-05-10 15:32:07 <JohannesJ[m]> ohhh i love lenovo! i run alpine with KDE on an ideapad
2023-05-10 15:32:24 <Mari[m]> hey i did the upgrade process on my alpine server
2023-05-10 15:32:24 <Mari[m]> and rebooted
2023-05-10 15:32:25 <Mari[m]> and it's still on 3.17
2023-05-10 15:32:44 <ikke> how did you upgrade, and how did you confirm you're still on 3.17?
2023-05-10 15:33:01 <uniqNuuid> bbbhltz: what's the gpu? 
2023-05-10 15:33:09 <bbbhltz> Intel
2023-05-10 15:33:14 <uniqNuuid> just intel?
2023-05-10 15:33:27 <psykose> bbbhltz: you can probably just pass nomodeset on kernel args to confirm that it works without that
2023-05-10 15:33:34 <Mari[m]> cat /etc/alpine-release
2023-05-10 15:33:43 <ikke> Mari[m]: does /etc/alpine-release.apk-new exist?
2023-05-10 15:33:45 <Mari[m]> and i followed the upgrade instructions on the wiki
2023-05-10 15:33:53 <bbbhltz> Right... So, it hd graphics 620
2023-05-10 15:34:14 <uniqNuuid> Mari[m]: run this: update-conf
2023-05-10 15:34:25 <Mari[m]> ikke: yeah
2023-05-10 15:34:44 <ikke> Mari[m]: that means it was modified and apk by default does not change files that have been updated in /etc
2023-05-10 15:34:48 <psykose> you upgraded fine, it just doesn't overwrite the file
2023-05-10 15:34:58 <Mari[m]> oh ok
2023-05-10 15:35:38 <bbbhltz> psykose: where to I add the nomodeset arg
2023-05-10 15:35:44 <psykose> kernel cmdline
2023-05-10 15:35:54 <dewdude> yeah. the thing to watch is the initial bootscreen and look at the versions
2023-05-10 15:36:05 <uniqNuuid> bbbhltz: grub menu, press "e", then add it on line "linux ...."
2023-05-10 15:36:30 <psykose> the initial boot screen just reads the same files pretty sure
2023-05-10 15:36:43 <uniqNuuid> dewdude: or dmesg -H
2023-05-10 15:37:06 <dewdude> yeah
2023-05-10 15:37:16 <dewdude> i gotta be honest...the amount i touch my alpine machines is almost zero
2023-05-10 15:37:28 <dewdude> i set em up, i keep them updated
2023-05-10 15:37:36 <dewdude> i forget they're there most of the time
2023-05-10 15:37:53 <ikke> That's how I like my machines
2023-05-10 15:38:09 <uniqNuuid> dewdude: same. super stable
2023-05-10 15:38:24 <JohannesJ[m]> well.. i guess living on the edge means upgrading as well :D
2023-05-10 15:38:32 <uniqNuuid> apk upgrade -aU and update-conf and reboot. 
2023-05-10 15:38:39 <uniqNuuid> never touch it again. 
2023-05-10 15:38:40 <bbbhltz> No-go on the nomodeset 
2023-05-10 15:38:43 <psykose> JohannesJ[m]: if you're referring to pipewire-media-session then it was removed yeah
2023-05-10 15:38:47 <psykose> just wireplumber now
2023-05-10 15:39:03 <dewdude> no, i go in and update them when i see the edge repo has new versions
2023-05-10 15:39:26 <uniqNuuid> bbbhltz: disable i915? add this:  module_blacklist=i915
2023-05-10 15:39:37 <dewdude> https://img.pickmy.org/f4mfifau.png
2023-05-10 15:39:42 <dewdude> all but 4 of those are alpine
2023-05-10 15:40:00 <dewdude> no. 3
2023-05-10 15:40:06 <dewdude> valutwarden is in fact alpine
2023-05-10 15:40:28 <JohannesJ[m]> i was ;-)
2023-05-10 15:40:29 <JohannesJ[m]> i installed wireplumber already itseems
2023-05-10 15:40:46 <uniqNuuid> dewdude: what is the software?
2023-05-10 15:41:08 <dewdude> which...the servers or the virtualization?
2023-05-10 15:41:15 <uniqNuuid> the vm manager
2023-05-10 15:41:20 <bbbhltz> uniqNuuid: as well as the nomodeset?
2023-05-10 15:41:21 <dewdude> I run Xen
2023-05-10 15:41:34 <dewdude> that's xcg-ng center for windows because i can't legally use xenorchestra right now
2023-05-10 15:41:37 <uniqNuuid> bbbhltz: you can try it alone. one a time
2023-05-10 15:41:55 <bbbhltz> That doesn't seem to work either
2023-05-10 15:42:20 <psykose> then it's probably not the gpu
2023-05-10 15:42:35 <uniqNuuid> bbbhltz: when you press e, what's on the screen, take a pic
2023-05-10 15:42:37 <JohannesJ[m]> dewdude you run a pbx on alpine? does it work?
2023-05-10 15:42:45 <dewdude> JohannesJ[m], yes.
2023-05-10 15:42:55 <dewdude> in fact...I self-compiled Asterisk on Alpine
2023-05-10 15:43:10 <bbbhltz> Prepare yourselves for a bad pic
2023-05-10 15:43:14 <dewdude> rather than use packaged version
2023-05-10 15:43:51 <dewdude> JohannesJ[m], that same pbx VM also runs like...6 icecast streams used for musiconhold
2023-05-10 15:44:13 <uniqNuuid> dewdude: what does sshbox do? 
2023-05-10 15:44:26 <dewdude> it's a public-facing ssh box
2023-05-10 15:44:28 <JohannesJ[m]> i am looking to replace 3CX after their latest shitshow..
2023-05-10 15:44:30 <JohannesJ[m]> oh? care to make a wiki article for it? ;-)
2023-05-10 15:44:56 <dewdude> sshbox has good keys and 2FA
2023-05-10 15:45:09 <dewdude> i basically ssh in to it and then can ssh in to the other network machines
2023-05-10 15:45:30 <dewdude> i can also tunnel over ssh in case the vpn isn't working
2023-05-10 15:46:06 <JohannesJ[m]> i suppoase an asterisk pbx with voicemail and dialplan on asterisk would be nice :)
2023-05-10 15:46:23 <dewdude> JohannesJ[m], I mean...i thought about it but i'm not sure how much interest there is. most people would probably want to run package version
2023-05-10 15:46:29 <JohannesJ[m]> :)
2023-05-10 15:46:33 <dewdude> which...is fine. Asterisk 18 is still supported
2023-05-10 15:46:54 <uniqNuuid> dewdude: that is a good idea managing ssh
2023-05-10 15:46:57 <psykose> if you test https://gitlab.alpinelinux.org/alpine/aports/-/merge_requests/45321 i'll happily merge it
2023-05-10 15:47:02 <dewdude> the thing about Asterisk and 3CX is the two are nothing alike
2023-05-10 15:47:09 <dewdude> uniqNuuid, thanks.
2023-05-10 15:47:30 <dewdude> there's literally nothing on that box either...it's just a bare bones install with stuff for SSH
2023-05-10 15:47:47 <dewdude> it runs in a whopping 256mb of ram
2023-05-10 15:48:11 <uniqNuuid> hmmm.. i rsycn my ~/.ssh everywhere to trusted machines 
2023-05-10 15:48:13 <dewdude> JohannesJ[m], if you've never done asterisk you'll find it daunting
2023-05-10 15:48:20 <JohannesJ[m]> there's always interest for it :)
2023-05-10 15:48:34 <dewdude> because people assume it does things for you that freepbx or 3cx do
2023-05-10 15:48:39 <dewdude> and asterisk literally does nothing
2023-05-10 15:48:46 <dewdude> you have to tell it from the ground up what it does
2023-05-10 15:49:15 <uniqNuuid> bbbhltz: i think it is a UEFI issue. 
2023-05-10 15:49:18 <dewdude> like..okay...you've got a phone connected over SIP...but it doesn't have an extension or can call anything until you build a dialplan that does that
2023-05-10 15:49:58 <dewdude> like my phone has about 16 different extensions that will ring it depending on where in the dialplan you get lost
2023-05-10 15:50:13 <dewdude> ...my ivr and menus are designed to not call me
2023-05-10 15:50:22 <uniqNuuid> bbbhltz: for some reason, the grub did not load 'linux' module. so you may need to write down what on the grub menu and start a grub commandline promot, and input stuff there 
2023-05-10 15:50:43 <dewdude> if my desk phone rings it means it's someone i both gave the number to *and* told them the menu code that rings me directly
2023-05-10 15:51:07 <dewdude> otherwise it just plays music and rickrolls you
2023-05-10 15:51:08 <bbbhltz> https://imgur.com/a/jqtyibd
2023-05-10 15:51:49 <dewdude> JohannesJ[m], if you're interested in building asterisk 20 on alpine; I did write this: https://git.pickmy.org/dewdude/asterisk-alpine
2023-05-10 15:51:54 <bbbhltz> I have a very crappy camera on my phone
2023-05-10 15:52:06 <patdk-lap> think I figured out the abuild issue, but no idea what to do about it
2023-05-10 15:52:08 <patdk-lap> "sudo: effective uid is not 0, is /usr/bin/sudo on a file system with the 'nosuid' option set or an NFS file system without root privileges?"
2023-05-10 15:52:33 <uniqNuuid> bbbhltz: so, on grub menu , press c
2023-05-10 15:53:23 <bbbhltz> Ok
2023-05-10 15:54:30 <uniqNuuid> bbbhltz: insmod linux
2023-05-10 15:55:22 <bbbhltz> uniqNuuid: done
2023-05-10 15:55:30 <uniqNuuid> and then : linux /TAB 
2023-05-10 15:55:42 <uniqNuuid> TAB is the tab key on your keyboard
2023-05-10 15:56:00 <uniqNuuid> do you see vmlinuz on the completion?
2023-05-10 15:56:07 <uniqNuuid> vmlinuz-lts
2023-05-10 15:56:23 <bbbhltz> Yes
2023-05-10 15:57:00 <uniqNuuid> yes?
2023-05-10 15:57:19 <bbbhltz> Yes, I see vmlinuz-lts
2023-05-10 15:57:35 <uniqNuuid> what else do you see?
2023-05-10 15:59:17 <bbbhltz> Grub efi config-lts initramfs-lts libutil.c32 extlinux.conf libcom32.c32 mboot.c32 menu.c32 vesamenu.c32 System.map-lts
2023-05-10 16:00:30 <uniqNuuid> linux /vmlinuz-lts root=UUID=xxxxx ro rootfstype=ext4
2023-05-10 16:00:36 <uniqNuuid> xxx is on your picture
2023-05-10 16:01:04 <uniqNuuid> after that, input : initrd /initramfs-lts
2023-05-10 16:01:18 <uniqNuuid> and then the third command: boot
2023-05-10 16:01:28 <uniqNuuid> so, total 3 lines
2023-05-10 16:01:54 <psykose> patdk-lap: what environment do you run that in (i.e. container, etc), how did you set it up
2023-05-10 16:02:02 <psykose> the error message is fairly self explanatory
2023-05-10 16:02:21 <psykose> but it's pretty impossible for sudo to be u-s so it would mean the way you set up the thing forbids suid
2023-05-10 16:05:54 <patdk-lap> psykose, I already posted all that info
2023-05-10 16:07:16 <bbbhltz> uniqNuuid: nothing... I'll try it again maybe
2023-05-10 16:07:27 <psykose> the dockerfile? you didn't paste how you run it
2023-05-10 16:07:42 <patdk-lap> docker build -t test .
2023-05-10 16:08:03 <psykose> so on aarch64 it's native?
2023-05-10 16:08:59 <uniqNuuid> bbbhltz: nothing after running `boot`?
2023-05-10 16:10:15 <uniqNuuid> the menu shows you've already installed 3.18, how did you install it? upgraded or fresh installed?
2023-05-10 16:10:37 <bbbhltz> Upgrade 
2023-05-10 16:10:46 <bbbhltz> Nothing after boot 
2023-05-10 16:12:15 <uniqNuuid> bbbhltz: well, press e on grub menu, apppend this `loglevel=7` on the line `linux ...`
2023-05-10 16:13:02 <bbbhltz> Loading initial ramdisk
2023-05-10 16:17:25 <blt[m]> anyone got the same issue after 3.18 upgrade : gpg: error getting version from 'scdaemon': No SmartCard daemon ?
2023-05-10 16:17:25 <blt[m]> I checked that pcscd is running and scdeamon is still ok but can't access to my yubikey anymore, any hint from what the problem could be ?
2023-05-10 16:18:01 <uniqNuuid> ®bbb
2023-05-10 16:19:35 <psykose> you do have gnupg-scdaemon installed right
2023-05-10 16:20:43 <uniqNuuid> bbbhltz: then, maybe you want to check your BIOS configuration? stuff like CSM DVMT
2023-05-10 16:21:15 <bbbhltz> Does the cryptroot thing have anything to do with this
2023-05-10 16:22:23 <blt[m]> yes I do like in 3.17
2023-05-10 16:23:15 <uniqNuuid> bbbhltz: what is that? you have full disk encryption?
2023-05-10 16:23:59 <blt[m]> it was working for a long time (since 3.15), I am using it 50 times a day and as soon as I did the upgrade yesterday my card is not seen, to be accurate it is seen by pcsc_scan but not on gpg
2023-05-10 16:24:18 <psykose> dunno what could be missing, probably related to changes in 2.4
2023-05-10 16:24:41 <psykose> but i know someone else with smartkeys has been using them fine
2023-05-10 16:27:22 <blt[m]> I read that "GnuPG 2.2.27. This version of GnuPG notably has trouble sharing the use of smart cards with other applications (which typically use the standard pcsc-lite daemon for smart card access, rather than GnuPG’s homegrown scdaemon). A newer version of GnuPG, 2.4.0, fixes this." perhaps that the way I was accessing the smartcard previously and how this has been fixed no in 2.4, I can try to remove pcsc-lite daemon
2023-05-10 16:28:08 <eex0> Q: I get this msg -> Gtk-Message: Failed to load module "canberra-gtk-module" .. what would be a "proper" fix when I have already explicitly included all of the modules it requires?
2023-05-10 16:29:31 <blt[m]> but I don't see an init with gnupg-scdaemon ?
2023-05-10 16:30:13 <psykose> it's a thing you run as your user pretty sure
2023-05-10 16:30:30 <psykose> eex0: that doesn't look like an error
2023-05-10 16:30:37 <bbbhltz> uniqNuuid: yes I have disk encryption 
2023-05-10 16:31:40 <blt[m]> psykose: I always used pcsc-lite and it ships with pcscd init but I just unistall it to see and so no more init
2023-05-10 16:32:08 <bbbhltz> uuidNuniq: CSM=Yes in bios
2023-05-10 16:32:20 <blt[m]> psykose: "The scdaemon is a daemon to manage smartcards. It is usually invoked by gpg-agent and in general not used directly" you're right
2023-05-10 16:32:20 <eex0> apologies, I did say "fix" which would indicate that it's an error but tbh, I can actually live w/o those in the app.. just a convenience to not have msg's
2023-05-10 16:33:12 <uniqNuuid> bbbhltz: i have no experiecne on disk encryption. i suggest set CSM=no
2023-05-10 16:37:03 <minimal> CSM is UEFI "BIOS compatibility mode", only used if you want to MBR-boot rather than ESP-partition boot
2023-05-10 16:37:05 <bbbhltz> Well, I guess I need to figure out how to fix this without losing my stuff. It has been working perfectly too... I guess these are the things that can happen. 
2023-05-10 16:37:11 <minimal> nothing to do with encryption
2023-05-10 16:38:33 <minimal> bbbhltz: you say you're using disk encryption? your previous screenshot of Grub settings didn't show any settings to do with encryption
2023-05-10 16:38:37 <psykose> eex0: as for what would give it, probably libcanberra-gtk3
2023-05-10 16:38:44 <psykose> that's what gives the module
2023-05-10 16:38:45 <uniqNuuid> i know  . i was hoping this may fix the blank display issue 
2023-05-10 16:38:50 <psykose> or -gtk2 for gtk2
2023-05-10 16:39:09 <bbbhltz> minimal: it is there but in the super unreadable part
2023-05-10 16:39:33 <minimal> bbbhltz: are you UEFI or BIOS (CSM) booting?
2023-05-10 16:40:15 <bbbhltz> minimal: uefi
2023-05-10 16:41:39 <eex0> @psykose, [smiles] yes, I used ldd to list all of the modules associated and took a deep dive into all of the modules associated and listed the modules in my app but I still get that "warning" msg.
2023-05-10 16:42:06 <psykose> ldd isn't relevant there
2023-05-10 16:42:16 <psykose> which app tho
2023-05-10 16:42:51 <minimal> bbbhltz: but you earlier said "CSM=Yes in bios"
2023-05-10 16:43:29 <bbbhltz> Uefi is selected and yes is the only choice
2023-05-10 16:43:30 <minimal> so which is it? you can't be UEFI booting if CSM is enabled
2023-05-10 16:44:52 <uniqNuuid> bbbhltz: csm=yes is the only option because you have other settings like WIN8/WIN7 mode or similar. i don't remember. 
2023-05-10 16:44:58 <eex0> what I am doing is writing strict apparmor rules for various applications, (firstly) internet facing apps. In the profiles I am listing file the app needs and deciding if they ccan either be allowed to function or deny them. and tbh, this really is the nutsehll answer. It's somewhat complex actually. 
2023-05-10 16:45:23 <psykose> is there an actual issue without the apparmor vaporware or nah
2023-05-10 16:45:46 <uniqNuuid> bbbhltz: anyway, explore your BIOS, and try to make that CSM option disappear
2023-05-10 16:46:16 <Ermine> As Ariadne once said, using selinux turns one problem into two. I guess the same goes for apparmour
2023-05-10 16:46:19 <bbbhltz> I have turned it off. Cycled through the options and now it says uefi and csm no
2023-05-10 16:46:47 <eex0> Im using hexchat w/ the apparmor profile that I created right now so, no, to answer your question @psykose just that one module that wants to grant me a msg.
2023-05-10 16:47:00 <uniqNuuid> bbbhltz: still can't boot?
2023-05-10 16:47:04 <bbbhltz> Nah
2023-05-10 16:48:18 <minimal> bbbhltz: you can't just "randomly" switch CSM on or off - when you installed Alpine (and any other OSes) on the machine then a decision was made as to whether to use UEFI or CSM for booting and the disk (and OSes on it) would have been set up for EITHER UEFI or CSM/BIOS booting
2023-05-10 16:48:34 <bbbhltz> So, I feel like I've had this problem before. What happened was I did an upgrade earlier today and did not reboot. Later I upgraded again and when I rebooted I had this problem. This happened to me before.
2023-05-10 16:48:57 <eex0> @psykose, an online search for the canberra modules msg turns up a ton of pages all saying install the pkg but, I have and still get the "warning"
2023-05-10 16:49:43 <uniqNuuid> bbbhltz: does it boot with 3.18 ISO?
2023-05-10 16:49:48 <uniqNuuid> make one and try
2023-05-10 16:50:32 <bbbhltz> minimal: in the bios I can choose UEFI and then change CSM. I can choose Legacy or Both as well. 
2023-05-10 16:50:56 <bbbhltz> uniqNuuid: the only thing I have is a 3.17.something
2023-05-10 16:51:20 <uniqNuuid> download from https://dl-cdn.alpinelinux.org/alpine/v3.18/releases/x86_64/alpine-extended-3.18.0-x86_64.iso
2023-05-10 16:51:27 <uniqNuuid> and make a usb boot, test it. 
2023-05-10 16:51:28 <psykose> eex0: i asked if it works without any apparmor
2023-05-10 16:51:31 <eex0> listed all the modules - $ ldd /usr/lib/libcanberra-gtk.so.0 [& also] $ ldd /usr/lib/libcanberra-gtk3.so.0 added every single one of them
2023-05-10 16:51:32 <psykose> and what application is using it
2023-05-10 16:51:43 <eex0> yes, it does
2023-05-10 16:51:46 <bbbhltz> minimal: no other computer...
2023-05-10 16:52:13 <eex0> hexchat is using it @psykose
2023-05-10 16:52:31 <minimal> bbbhltz: ok, so then CSM=Legacy is "BIOS booting", CSM=Both don't know how that behaves - does it try UEFI and then fallback to BIOS? What does your manual (for machine) say?
2023-05-10 16:52:38 <psykose> well, then the issue is in your apparmor :)
2023-05-10 16:52:38 <uniqNuuid> ok then boot with 3.17 usb and chroot, and try stuff like:  apk fix ?
2023-05-10 16:52:46 <bbbhltz> minimal: but the 3.17 boots
2023-05-10 16:52:54 <uniqNuuid> as well as update-conf
2023-05-10 16:52:56 <psykose> the actual modules are in /usr/lib/gtk-3.0/modules
2023-05-10 16:53:00 <psykose> see https://pkgs.alpinelinux.org/contents?file=&path=&name=libcanberra-gtk3&branch=edge&arch=x86_64
2023-05-10 16:53:25 <bbbhltz> uuidNuniq: chroot Huh? Got a guide for this
2023-05-10 16:53:31 <minimal> bbbhltz: "but the 3.17 boots"? you have both 3.17 and 3.18 installed on the SAME machine?
2023-05-10 16:53:34 <eex0> thank you, I will keep working thru it.. :)
2023-05-10 16:54:34 <bbbhltz> minimal: my usb with 3.17 boots. It's the only other install medium I have where I am
2023-05-10 16:54:48 <uniqNuuid> i don't know how to chroot into a encrypted disk... though. 
2023-05-10 16:55:12 <ikke> luksopen it, mount the device, chroot into it?
2023-05-10 16:55:47 <uniqNuuid> disk encryption is a thing i haven't tried yet ... so. 
2023-05-10 16:56:27 <uniqNuuid> but the other steps are same. mount the partions, and then mount dev,sys,proc,run,tmp ...
2023-05-10 16:56:29 <minimal> bbbhltz: your USB with 3.17 is using a different boot "setup", i.e. it uses the partitions and files on the USB stick, not on the HDD/SSD
2023-05-10 16:56:31 <uniqNuuid> and chroot
2023-05-10 16:56:34 <ikke> luksOpen will represent the unencrypted disk as a blockdevice
2023-05-10 16:56:59 <ikke> after that everything else is the same
2023-05-10 17:01:33 <uniqNuuid> i'll try that disk encryption some day, just don't feel liking needing it. 
2023-05-10 17:01:58 <bbbhltz> uniqNuuid: I've never used chroot before. I will give it a shot or I'll just cancel work tomorrow.
2023-05-10 17:05:11 <uniqNuuid> bbbhltz: here is a reference : https://wiki.gentoo.org/wiki/Chroot
2023-05-10 17:05:21 <uniqNuuid> ignore the gentoo related stuff 
2023-05-10 17:05:51 <minimal> bbbhltz: so if you boot the 3.17 USB stick and run "lsblk /dev/sda" (if you HDD/SSD if not "sda" then replace with correct name)
2023-05-10 17:05:57 <minimal> also run "blkid"
2023-05-10 17:06:11 <minimal> and then use a pastebin to show us the output of both commands
2023-05-10 17:06:41 <blt[m]> psykose: just for your info, I found the gulty ... in 3.16 or 3.17 with 'driver /usr/lib/libpcsclite.so' in .gnupg/scdaemon.conf it was working fine but not in 3.18 / gpg 2.4 so commenting the line (and still using pcscd) is fixing my issue
2023-05-10 17:06:54 <psykose> :)
2023-05-10 17:06:59 <psykose> good to hear
2023-05-10 17:08:41 <minimal> ah, you'd have to install it then lol. And blkid?
2023-05-10 17:09:43 <bbbhltz> Yes for blkid
2023-05-10 17:11:01 <bbbhltz> minimal: 3 partitions of type crypto_LUKS, swap and vfat
2023-05-10 17:11:57 <bbbhltz> minimal: which are my root, swap, and boot respectively 
2023-05-10 17:14:44 <patdk-lap> looks like my abuild issue is resolved, no idea what the problem was, reboot fixed it :(
2023-05-10 17:14:54 <minimal> right, so you are NOT UEFI booting, you are BIOS booting
2023-05-10 17:15:04 <minimal> bbbhltz: ^^^
2023-05-10 17:16:21 <bbbhltz> minimal: well, in the bios it is set to uefi. But, these are things I generally don't pay attention to. I can just type out that line if you want.
2023-05-10 17:16:36 <patdk-lap> vfat would be the uefi efi partition
2023-05-10 17:16:57 <minimal> patdk-lap: yes potentially/probably - but without seeing the output we're just guessing
2023-05-10 17:17:29 <uniqNuuid> bbbhltz: fdisk -l, if you see EFI or ESP on that disk, you are UEFI
2023-05-10 17:18:03 <bbbhltz> minimal: /dev/nvme0n1p1: UUID="xxx" TYPE="VFAT"
2023-05-10 17:18:26 <uniqNuuid> bbbhltz: that does not say anything. use fdisk -l
2023-05-10 17:18:35 <uniqNuuid> you want partition type 
2023-05-10 17:19:49 <uniqNuuid> apk add cmd:fdisk 
2023-05-10 17:19:57 <uniqNuuid> use the real fdisk
2023-05-10 17:20:04 <patdk-lap> fdisk -l /dev/nvme....
2023-05-10 17:20:15 <patdk-lap> oh, you said that :)
2023-05-10 17:20:56 <minimal> forgot setup-disk doesn't label filesystems
2023-05-10 17:24:40 <bbbhltz> uniqNuuid: what am I looking for with fdisk? Says it is 300 megs but doesn't say partition type
2023-05-10 17:26:02 <uniqNuuid> look at the line nvmr0n1p1
2023-05-10 17:26:17 <uniqNuuid> the last field 
2023-05-10 17:26:49 <uniqNuuid> fdisk -l | grep nvme0n1p1 
2023-05-10 17:31:38 <minimal> "sfdisk -l" and look at the "Type" field value for the relevant partition (is it "EFI System")
2023-05-10 17:32:27 <eex0> efibootmgr would list all bootables
2023-05-10 17:35:12 <bbbhltz> uniqNuuid: what did you mean by real fdisk?
2023-05-10 17:36:06 <bbbhltz> uniqNuuid: fdisk -l on that partition doesn't list a type
2023-05-10 17:36:43 <minimal> bbbhltz: sfdisk as I suggested above
2023-05-10 17:39:14 <bbbhltz> minimal: I am trying to install it. 
2023-05-10 17:49:31 <bbbhltz_> ok, managed to have access to another computer
2023-05-10 17:59:25 <bbbhltz_> @minimal sfdisk -l says that partition is an EFI system
2023-05-10 18:00:45 <minimal> ok, and does /sys/firmware/efi exist when you boot any Linux system?
2023-05-10 18:01:35 <bbbhltz_> minimal, when booting from the Alpine USB yes
2023-05-10 18:02:02 <minimal> so then that is booting via UEFI, not BIOS/MBR booting
2023-05-10 18:02:33 <minimal> and you were running "sfdisk" against the NVME device rather than the USB stick, correct?
2023-05-10 18:02:42 <bbbhltz_> yes
2023-05-10 18:04:37 <minimal> bbbhltz: and booting from the NVME was previously working?
2023-05-10 18:04:54 <bbbhltz_> minimal: yes, for the past few weeks, daily
2023-05-10 18:05:22 <minimal> and what changed before it stopped booting?
2023-05-10 18:05:38 <bbbhltz_> did update, upgrade -a
2023-05-10 18:05:43 <bbbhltz_> used it for a bit
2023-05-10 18:05:46 <bbbhltz_> rebooted
2023-05-10 18:06:03 <minimal> which *exactly* commands?
2023-05-10 18:06:04 <bbbhltz_> I think I did another upgrade before rebooting
2023-05-10 18:06:07 <bbbhltz_> ah
2023-05-10 18:06:13 <bbbhltz_> apk update
2023-05-10 18:06:16 <bbbhltz_> apk upgrade -a
2023-05-10 18:06:17 <minimal> you did or didn't change Alpine version?
2023-05-10 18:06:27 <bbbhltz_> I went from 3.17 to 3.18
2023-05-10 18:07:01 <bbbhltz_> using Edge
2023-05-10 18:07:35 <minimal> eh? "I went from 3.17 to 3.18" vs "using Edge" ?
2023-05-10 18:08:34 <bbbhltz_> I suppose that means there wasn't really a version change
2023-05-10 18:08:50 <minimal> you mean you went 3.17 -> Edge -> 3.18?
2023-05-10 18:09:32 <psykose> ran `apk upgrade` and was on edge all the time
2023-05-10 18:10:03 <bbbhltz_> Yes, I was on edge kept the system up to date
2023-05-10 18:10:44 <minimal> bbbhltz: "I went from 3.17 to 3.18" and "Yes, I was on edge" are contradictory
2023-05-10 18:11:13 <psykose> now u kno
2023-05-10 18:11:19 <psykose> anyway this was not related to the releases
2023-05-10 18:11:20 <psykose> moving on
2023-05-10 18:13:56 <bbbhltz_> time-wise, I should probably just reinstall, right?
2023-05-10 18:14:28 <bbbhltz_> or learn how to use chroot
2023-05-10 18:14:59 <psykose> i forget, did just an iso boot fine
2023-05-10 19:18:07 <JohannesJ[m]> <dewdude> "Johannes J., if you're intereste..." <- i will try this!
2023-05-10 19:26:06 <JohannesJ[m]> hmmz, im on 3.18
2023-05-10 19:27:13 <eex0> Great job on this release of Alpine dev's
2023-05-10 19:27:30 <eex0> Thank You!
2023-05-10 19:27:40 <JohannesJ[m]> indeed
2023-05-10 19:27:42 <JohannesJ[m]> now i want to help!
2023-05-10 19:28:30 <eex0> When I learn a bit more, I will too. :)
2023-05-10 19:28:32 <ikke> eex0: Thank you, and thanks to all the volunteers who helped
2023-05-10 19:30:52 <JohannesJ[m]> thank you! and you! and you! thank everyone! * goes into Oprah mode *
2023-05-10 19:31:24 <netbooter> Hey all!
2023-05-10 19:31:30 <eex0> waves
2023-05-10 19:31:46 <netbooter> Just wanted to ask if anyone could point me to the kernel config of the standard vs netboot kernel
2023-05-10 19:32:08 <netbooter> I guess the netboot kernel has the NIC drivers built in vs. supplied as a module for the standard flavor. right?
2023-05-10 19:32:32 <netbooter> Does anyone know where the sources are so I can compare them? I don't seem to be able to find them, sorry if it is super obvious...
2023-05-10 19:34:48 <eex0> does this help? https://docs.alpinelinux.org/user-handbook/0.1a/Installing/medium.html
2023-05-10 19:35:26 <netbooter> not really. those are the built variants. I'm looking for the "recipe" :)
2023-05-10 19:35:33 <netbooter> but thanks for trying to help anyway!
2023-05-10 19:35:38 <ikke> netbooter: I think it's the same kernel
2023-05-10 19:35:43 <bbbhltz> I also want to send out a thank you to everyone. Despite 10+ years of Linux under my belt I still find ways to make mistakes and I have had quicker replies from this community than from other. Less condescending as well. This time around I might not have been able to find an easy fix but your patience is remarkable. Exemplary work on all fronts.
2023-05-10 19:36:05 <ikke> netbooter: https://gitlab.alpinelinux.org/alpine/aports/-/blob/master/scripts/mkimg.netboot.sh
2023-05-10 19:37:48 <eex0> bbbhltz: this is the big leagues
2023-05-10 19:38:04 <JohannesJ[m]> bbbhltz you are right :) i too find the Alpine community a whole lot friendlier :)
2023-05-10 19:38:52 <ikke> We do our best to keep this a friendly and welcoming place
2023-05-10 19:40:08 <JohannesJ[m]> thats important :) with where windows and macos are going, more and more people are dipping their toes into linux
2023-05-10 19:40:53 <JohannesJ[m]> alpine is small, secure, its stable, and now i can say it even runs great on a laptop :)
2023-05-10 19:41:23 <bbbhltz> JohannesJ[m]: what ever do you mean? Is it the adds in the settings menus? 
2023-05-10 19:42:01 <JohannesJ[m]> windows you mean? yeah not just the ads in the windows settings, they're slowly apearing everywhere :)
2023-05-10 19:43:31 <JohannesJ[m]> i used to be an apple fanboy all the way! but a decade without Jobs is starting to show cracks as well :)
2023-05-10 19:43:39 <bbbhltz> It is amazing how they think of new places to put them
2023-05-10 19:44:25 <JohannesJ[m]> i traded my macbook for a lenovo ideapad and then looked at debian since im used to it for servers :) but then i wanted something that can run on my phone too, and on my tablet.. and then i found out about Alpine
2023-05-10 19:46:00 <JohannesJ[m]> im slowly learning Alpine :)
2023-05-10 19:46:47 <JohannesJ[m]> failed to run it on my laptop the first time around, so switched back to Manjaro. wasnt happy with that either so put my teeth in it :) and now im on KDE with network working perfectlty, even wireguard :)
2023-05-10 19:50:40 <netbooter> ikke: thank you! will check it immediately
2023-05-10 19:51:42 <bbbhltz> I am learning too. Little bit different than the Debiam and Arch installs I have used for years. I threw it on my rpi and it ran like a charm, so I tried it on an old pc and it also worked wonderfully so now I have it on my personal laptop. Well, I did mess it up somehow. But that will be fixed soon enough. Nothing wrong with a little backup and reinstall.
2023-05-10 19:52:25 <netbooter> ikke: yes I see now that those are added as "initfs_features" 
2023-05-10 19:52:30 <JohannesJ[m]> through fuckups we grow
2023-05-10 19:52:52 <netbooter> does this mean that the modloop is already loaded during the initramfs stage? I thought there is an actual init script which loads the modloop!
2023-05-10 19:53:46 <eex0> netbooter: that reminded me of yoda saying "thru mistakes, we grow" ..lol in my best yoda voice.
2023-05-10 19:54:41 <eex0> “The greatest teacher, failure is.” 
2023-05-10 19:54:49 <netbooter> that true! :D
2023-05-10 19:55:21 <JohannesJ[m]> :D
2023-05-10 19:57:06 <stonks3141> How can I set the default mem sleep state on boot?
2023-05-10 19:57:19 <stonks3141> the default is suspend2idle for me, and I want it to be deep
2023-05-10 19:57:21 <netbooter> then one more question guys: where could I compare the kernel config of the standard vs the virtualized?
2023-05-10 19:57:42 <netbooter> just wanna diff the Kconfig to see what has been omitted from the virt kernel
2023-05-10 19:57:58 <ikke> netbooter: https://gitlab.alpinelinux.org/alpine/aports/-/tree/master/main/linux-lts
2023-05-10 19:58:06 <minimal> netbooter: easiest way, unpack both linux-lts and linux-virt APK files and diff their config-lts/config-virt files ;-)
2023-05-10 19:58:07 <ikke> lts.*.config
2023-05-10 19:58:46 <netbooter> Thank you ikke!! <3 
2023-05-10 19:59:45 <minimal> well ok, diff-ing from the repo will show changes between lts and virt, both neither file is a full kernel config, they're both deltas against kernel's default config
2023-05-10 19:59:59 <minimal> s/both neither/though neither/
2023-05-10 20:17:29 <minimal> netbooter: you're doing netbooting? the netboot tarball actually comes with both the linux-lts and linux-virt config files...
2023-05-10 21:11:58 <eex0> I think mozilla has a typo in firefox .. "AudioIP~allback" @ peer ->signal ..
2023-05-10 22:12:48 <jim2312> I know you're not supposed to unlink dependancies but is it possible though? I would love to remove the gnome-tour but its a dependancy of gnome itself
2023-05-10 22:14:58 <pj> modify and build the package yourself
2023-05-10 22:15:50 <eex0> or apparmor/selinux deny 
2023-05-10 22:16:40 <pj> deny what
2023-05-10 22:17:18 <eex0> I'd have to look but compling would be best if size matters
2023-05-10 22:17:21 <jim2312> hmm, I'll stick to rm -rf'ing the doc files and .desktop file.. not really fancy rebuilding gnome from source just to get rid of a tour you take once and then is only in the way.. I was kind of surprised it was installed anyways, and even more surprised gnome itself depends on a tour.. alpine usually does not install any man/doc/etc stuff.. which I like
2023-05-10 22:18:22 <jim2312> I wonder if apk complains if I symlink the install location to /dev/null.. it works for some stuff lol
2023-05-10 22:19:19 <pj> I don't even know where do you see the docs in that
2023-05-10 22:19:25 <eex0> has anyone ever looked at that character file close?
2023-05-10 22:19:38 <eex0> "/dev/null"?
2023-05-10 22:20:29 <jim2312> oh gnome also depends on the gnome-background package.. (im trying to get my gnome more minimal)
2023-05-10 22:20:44 <jim2312> gnome-tour is a tour of gnome, its in /usr/share/help
2023-05-10 22:20:53 <jim2312> not considered documentation?
2023-05-10 22:21:04 <jim2312> I took it only once just to see what it is
2023-05-10 22:21:15 <jim2312> helpful if you're new to gnome but after that kind of useless
2023-05-10 22:22:00 <pj> if you want minimal gnome, then don't install gnome package?
2023-05-10 22:22:11 <eex0> smirks
2023-05-10 22:24:38 <pj> it's weird to demand from a metapackage to not include a specific package because you don't like it
2023-05-10 22:25:39 <eex0> opt-in v. opt-out springs forth in my mind.. I rant daily about that ..laughing
2023-05-10 22:31:45 <jim2312> I did remove the core gnome apps package
2023-05-10 22:32:19 <jim2312> just some apps gnome itself depends on like background and a tour which seem kind of odd to me so I was just looking for a way to remove them "properly" as I been just rm -rf'ing them lol
2023-05-10 22:32:34 <jim2312> they didn't get uninstalled when removing the core package
2023-05-10 22:32:57 <jim2312> but im sorry for troubling you with my gnome stuff, im merely a peasant too dumb for Sawyer hyperland right?
2023-05-10 22:32:58 <jim2312> bye bye
2023-05-10 22:46:28 <bootloader> hey everyone, I've reached the endgame on my journey to install the wifi-driver on mac
2023-05-10 22:47:02 <bootloader> however, there still are a few pinpoints I would like to ask about
2023-05-10 22:49:12 <bootloader> firstly, my wifi-card (wlan0) is being detected by `rfkill list` but not by `ifconfig`, `ip`, or `iwctl device list`. Any reason this may be the case?
2023-05-10 22:50:51 <bootloader> Secondly, `iwctl device list` hangs, I presume because of the same wifi-card error, could this be due to the wifi-card crashing or permissions?
2023-05-10 22:51:24 <rsc> Should I be able to use alpine:latest on riscv?
2023-05-10 22:52:41 <rsc> Oh no...there is no tag?
2023-05-10 22:54:18 <bootloader> Lastly, the `wl.ko` module was inserted into the kernel via `insmod` and is within a /lib/modules subfolder; but when I run `modprobe wl` it says that the module is undetected. Could this also be part of the error?
2023-05-10 22:54:34 <eex0> bootloader: are you using NetworkManager by any chance?
2023-05-10 22:54:53 <eex0> cause, that renames your device
2023-05-10 22:55:45 <bootloader> im not sure. how do you see if NetworkManager is active?
2023-05-10 22:56:44 <bootloader> eex0 I don't remember enabling it at any point, unless iwd calls NetworkManager
2023-05-10 22:58:05 <bootloader> but ill look for the command to find if its active
2023-05-10 22:58:23 <ovf> bootloader: is iwd up?
2023-05-10 22:58:35 <bootloader> yes it currently is
2023-05-10 22:58:56 <bootloader> iwctl also detects it
2023-05-10 22:59:01 <eex0> to check: nmcli
2023-05-10 22:59:14 <eex0> oh, I see you found it
2023-05-10 22:59:57 <ovf> bootloader: i.e. you see results for `iwctl adapter list`, but not `iwctl device list`?
2023-05-10 23:00:40 <bootloader> off both calls hang
2023-05-10 23:00:57 <bootloader> ovf *both calls hang
2023-05-10 23:01:01 <ovf> then what did you mean by 'iwctl also detects it'?
2023-05-10 23:01:12 <ovf> iwctl hanging sounds like a problem with dbus
2023-05-10 23:01:14 <bootloader> ioctl detects iAd I mean
2023-05-10 23:01:39 <bootloader> *iwctl detects iwd
2023-05-10 23:02:09 <bootloader> ovf oh thats a relief
2023-05-10 23:04:05 <bootloader> so its likely outdated package issue, or something around dbus? Ill reinstall dbus with the most-recent package and troubleshoot then
2023-05-10 23:04:49 <bootloader> if any other idea pop up, pls let me know. thanks for the assistance
2023-05-10 23:05:05 <ovf> let's look at `strace iwctl device list`
2023-05-10 23:07:01 <bootloader> I actually don't have it, Ill install the package now
2023-05-10 23:08:16 <bootloader> this process will unfortunately take awhile due to my current logistics
2023-05-11 00:09:46 <MrBubu> hello 
2023-05-11 00:10:43 <pj> rsc: no
2023-05-11 00:10:54 <MrBubu> is anyone here working/ed with any part of alpine linux documentation or wiki
2023-05-11 00:11:04 <pj> maybe better to state your problem first
2023-05-11 00:15:29 <MrBubu> @pj are you responding to someone else or me?
2023-05-11 00:16:09 <pj> you
2023-05-11 00:16:37 <MrBubu> oh ok. the problem is that the documentation is not very helpful
2023-05-11 00:17:33 <pj> is there anything not helpful or just overall it's not helpful
2023-05-11 00:17:33 <MrBubu> i was thinking on making some changes/contributions for the beginners but i don't want to mess it more than it is already
2023-05-11 00:20:28 <MrBubu> actually they become helpful after some experience or when you messed up things a few times
2023-05-11 00:21:21 <MrBubu> the 1st touch is re-ordering and later a few graphics (or maybe animated graphics)
2023-05-11 00:21:49 <pj> I hate literally no idea what you are referring to
2023-05-11 00:22:31 <pj> Wiki is made by Alpine community so its quality and quantity really depends on contributors
2023-05-11 00:24:07 <MrBubu> i am referring to the first steps (1 installing; 2 use-case configuration)
2023-05-11 00:26:17 <minimal> MrBube: which installation documentation specifically?
2023-05-11 00:26:34 <pj> I think it's https://wiki.alpinelinux.org/wiki/Installation
2023-05-11 00:26:53 <pj> for which I have no idea what you want to re-order or what kind of images are needed
2023-05-11 00:27:16 <pj> There is already one image that shows how setup-alpine looks like which is more than enough
2023-05-11 00:27:21 <minimal> MrBubu: ^^^
2023-05-11 00:28:33 <pj> The rest is just text, we don't provide graphical installers and it's all just running some kind of a shell script
2023-05-11 00:29:18 <minimal> MrBubu: which installation documentation specifically?
2023-05-11 00:29:59 <MrBubu> dear you are assuming every use case will need to run setup-alpine script while it is not true for all use cases
2023-05-11 00:31:22 <pj> There are separate wiki articles for standalone steps when not running setup-alpine
2023-05-11 00:31:48 <pj> But it would help if you pointed out which documentation you are referring to
2023-05-11 00:31:50 <minimal> MrBubu: no wish to clarify exactly which installation documentation you are referring to?
2023-05-11 00:32:06 <MrBubu> for example: my objective to create an external flash memory desktop that runs from ram and uses sometimes lbu commit to save changes back to the flash media
2023-05-11 00:32:45 <pj> There is a diskless section: https://wiki.alpinelinux.org/wiki/Installation#Diskless_Mode
2023-05-11 00:33:09 <MrBubu> the very 1st thing I should have done was to setup the flash memory partitions
2023-05-11 00:33:46 <MrBubu> the 2nd thing was setup-bootable and then reboot to the new bootable partition
2023-05-11 00:34:13 <pj> It doesn't sound like you want to contribute
2023-05-11 00:34:13 <MrBubu> then from there setup-cache
2023-05-11 00:35:11 <minimal> sign, no willing to tell us what existing documentation you are referring to, ok, I'm not playing this game...
2023-05-11 00:35:21 <MrBubu> you look like you have a lot of opinions mr pj
2023-05-11 00:35:22 <minimal> s/sign/sigh/
2023-05-11 00:35:40 <pj> Doesn't anyone?
2023-05-11 00:35:50 <pj> If you don't have opinions, it sounds like you are lost
2023-05-11 00:36:02 <MrBubu> ok
2023-05-11 00:36:12 <MrBubu> let's do it like this
2023-05-11 00:36:33 <pj> My opinion is that I need a link or some kind of information that allows me to pinpoint what "documentation" you are referring to
2023-05-11 00:36:43 <MrBubu> i'll make some graphics for the documentation section of the website and then you decide if you want to upload
2023-05-11 00:37:20 <MrBubu> there is no documentation for that case
2023-05-11 00:37:53 <pj> yes because it's not possible to document every niche set-up
2023-05-11 00:38:01 <MrBubu> you know what? 
2023-05-11 00:38:07 <pj> but there are parts of each set-up that are documented or can be documented
2023-05-11 00:38:08 <MrBubu> forget everything i said
2023-05-11 00:38:17 <MrBubu> excuse me
2023-05-11 00:42:13 <pj> just noticed i typo'd 'have' by writing 'hate' 👀
2023-05-11 00:47:38 <MrBubu> but it really fits into the conversation LOL
2023-05-11 00:47:58 <MrBubu> have a good day
2023-05-11 00:48:44 <pj> "uhm akchually it's 3am"
2023-05-11 00:49:00 <pj> damn, too late
2023-05-11 00:56:10 <bootloader> I went through all that trouble, to get the strace.apk, and I've found the issue
2023-05-11 00:56:41 <bootloader> the iwd didn't actually start, I was just under the impression it did
2023-05-11 00:58:18 <bootloader> I figured this out by just running `iwctl` rather than `iwctl device list`, and saw the `waiting for IWD` message. I found it weird that it would just hang, and that explains it
2023-05-11 00:59:51 <bootloader> so now the question becomes, why won't iwd actually start? It's strange because `rc-service` says that it exists and that it could run it
2023-05-11 01:01:09 <eex0> did you tell it when to start? like boot? excuse the dumass Q
2023-05-11 01:01:35 <bootloader> for context though, I am running iwd ***manually*** because it was installed manually. Is there a context or permission that may be missing causing a manual-run of iwd to fail?
2023-05-11 01:03:02 <bootloader> eex0 no the question is definitely warranted. I did tell it to start manually like `./iwd-manual start` from `root` user
2023-05-11 01:04:20 <minimal> iwd-manual? what package is that from?
2023-05-11 01:04:50 <bootloader> I also refreshed `sysctl,` using `rc-service sysctl restart`, so I don't think its a kernel issue - even though its definitely on the table
2023-05-11 01:05:45 <minimal> why restart sysctl? it just sets /sys/ settings
2023-05-11 01:07:45 <bootloader> minimal I manually imported the package, due to lack of internet, and used `apk add /path/to/iwd`. However, it didn't seem to do all the nice stuff alpine-packages usually do (like *.initd) so I copied the code from Alpine's `iwd.initd` into a file named `iwd-manual`
2023-05-11 01:08:33 <minimal> bootloader: that's because the init.d script comes from the SEPARATE iwd-openrc package
2023-05-11 01:09:10 <minimal> which would be automatically installed if you (were online and) did a "apk add iwd"
2023-05-11 01:10:51 <eex0> bootloade: have you tried this?  -> rc-update --update add [your_service] boot
2023-05-11 01:11:21 <bootloader> minimal yeah 100%, but because I'm not it didn't add the service in the right place to be used by `rc-service`. So I manually added it myself and ran it. And I thought it would work, because on my last-run iwd was working
2023-05-11 01:11:32 <minimal> eex0: that won't work as he doesn't have iwd-openrc package installed
2023-05-11 01:11:48 <eex0> duh me.. apologies
2023-05-11 01:12:36 <bootloader> minimal is this repo separate? (repo: `https://git.alpinelinux.org/aports/tree/community/iwd?h=3.12-stable`) 
2023-05-11 01:13:11 <bootloader> I assumed it was the same because of the name, but I see a world where this is not the case
2023-05-11 01:13:45 <minimal> why are you looking at a Alpine 3.12 repo?
2023-05-11 01:14:11 <bootloader> I'm currently on 3.12
2023-05-11 01:15:18 <bootloader> that's the version I had available. 3.17 wouldn't boot properly, so I fell back to a version that was tested to work
2023-05-11 01:15:27 <minimal> why? support for that ended 1 year ago. 3.18 was released YESTERDAY!
2023-05-11 01:16:19 <bootloader> minimal yeah, its not a good experience, but I will upgrade to 3.18 when my internet is back up
2023-05-11 01:16:42 <bootloader> im using sys-alpine
2023-05-11 01:16:45 <minimal> the last 3.12.x version was 3.12.12 from April 2021
2023-05-11 01:17:04 <minimal> what about 3.16? 3.15? 3.14? 3.13?
2023-05-11 01:17:54 <bootloader> yeah, I could've tested each 1-by-1 but it would've been a real-hassle, due to circumstances, so I went for the one that ppl 100% knew worked
2023-05-11 01:18:14 <minimal> also how did you test 3.17? by a direct upgrade from 3.12.x to 3.17?
2023-05-11 01:19:03 <bootloader> no I was booting 3.17 from a live-cd, and it refused to cooperate
2023-05-11 01:19:49 <bootloader> I threw the kitchen sink at it. but looking back, I think changing the dd write-type would've fixed it
2023-05-11 01:20:25 <minimal> so you're asking for help with whichever iwd version 3.12.x has?
2023-05-11 01:21:26 <bootloader> not necessarily. more of, "is there something that would block a manual run of xxxx-service?"
2023-05-11 01:22:43 <bootloader> I was on alpine-standard initially, and I got iwd to work using the same process. But since, I upgraded to alpine-extended, and the same process doesn't seem to be working
2023-05-11 01:23:28 <minimal> obvious answer might be "check the gitlab to see what known iwd package bugs/issues that were fixed in Alpine 3.13 -> 3.18" ;-)
2023-05-11 01:23:52 <bootloader> (T_T) not again
2023-05-11 01:23:57 <bootloader> this is so painful
2023-05-11 01:24:29 <bootloader> back to brazil then
2023-05-11 01:25:26 <minimal> why not focus on working out what is causing the 3.18 problem?
2023-05-11 01:26:15 <bootloader> oh yeah, I am. By "Brazil", I meant the trenches of git-patches
2023-05-11 01:26:38 <minimal> well 3.17 problem...starting by testing 3.18?
2023-05-11 01:26:45 <bootloader> I recently just did this with the STA(wl) wifi-driver
2023-05-11 01:27:26 <minimal> well that's a whole other matter
2023-05-11 01:28:04 <bootloader> oh no, I'm still on 3.12 Alpine. I'm going to upgrade when I confirm the wifi-card is working
2023-05-11 01:28:45 <minimal> you can't upgrade directly from 3.12 -> 3.17 (or 3.18), the package signing keys changed a couple of releases ago
2023-05-11 01:29:09 <minimal> unless you install the newer version of the keys package first
2023-05-11 01:29:27 <bootloader> im assuming 3.12-> 3.16 -> 3.18 will work?
2023-05-11 01:30:06 <bootloader> oh wait, that won't work either
2023-05-11 01:30:19 <bootloader> f*******************
2023-05-11 01:30:38 <minimal> you mean 3.12.x (whichever you're on) -> 3.16.5 -> 3.18.0 ?
2023-05-11 01:30:55 <bootloader> yeah, would that work?
2023-05-11 01:31:36 <minimal> it was 3.15.0 that introduced the new keys
2023-05-11 01:32:23 <minimal> so you'd need to have newer keys installed before upgraded to 3.15.0 or greater
2023-05-11 01:32:35 <bootloader> okay thank you
2023-05-11 01:32:51 <minimal> its mentioned in the release notes
2023-05-11 01:33:45 <bootloader> I think I'll just live-cd 3.1[56]. I've kind-of automated the process, so it should be a few button-clicks to get the wifi-module back
2023-05-11 01:35:04 <bootloader> first I'll check if there's a problem with my setup or the iid-package though, so ill be back
2023-05-11 01:35:05 <minimal> 3.15 is supported for another 6 months..
2023-05-11 01:35:10 <minimal> 3.16 for another 1 year
2023-05-11 01:38:16 <minimal> bootloader: ahhhh, iwd is in "community", not "main" lol
2023-05-11 01:40:26 <minimal> so those support timeframes don't apply ;-)
2023-05-11 03:16:19 <eex0> did libharfbuzz just roll back on 3.18 rel? libharfbuzz-icu.so.0.60730.0 went to libharfbuzz-icu.so.0.60720.0 and firefox stopped in its track
2023-05-11 03:16:53 <eex0> excuse me, libharfbuzz.so
2023-05-11 03:21:43 <eex0> I figured it out
2023-05-11 03:29:06 <rob[m]> wow, thanks for power8 support. Installed flawlessly and my gpu works. Excellent (IBM S822L) Linux power8 6.1.27-3-lts #4-Alpine SMP Wed, 10 May 2023 12:37:05 +0000 ppc64le GNU/Linux
2023-05-11 03:30:27 <rob[m]> I don't know if anyone tried to build libreoffice on ppc64le after https://gitlab.alpinelinux.org/alpine/aports/-/merge_requests/11136#note_184208, but I'm giving it a try right now
2023-05-11 03:40:54 <rob[m]> hmmm, Successful packaging process!, let's give it a try
2023-05-11 03:52:57 <bootloader> after a break, I came back and found the issue. if dbus can't write to `/run/dbus` it will not operate
2023-05-11 03:54:16 <bootloader> this happened because it was installed manually, and did not do automatic setup in the background. All to say ovf 's original point was correct
2023-05-11 03:56:02 <rob[m]> 
ACTION uploaded an image:  < https://matrix.org/_matrix/media/v3/download/matrix.org/sdglwTckxTrFBggDmDNYkAYJ/libreoffice_ppc64le.png >

2023-05-11 03:56:14 <rob[m]> works great, just got done compiling :D
2023-05-11 04:02:04 <ikke> minimal: fyi, alpine 3.12 should have the new keys already 
2023-05-11 04:03:38 <ikke> rob[m]: nice
2023-05-11 04:03:49 <ikke> We don't often get feedback around ppc
2023-05-11 04:04:33 <rob[m]> ikke: :D, well, from what I'm seeing, it works really well
2023-05-11 04:04:48 <rob[m]> even firefox-esr is packaged, so I can do some gui web browsing. Pretty good
2023-05-11 04:17:54 <rob[m]> I will go to sleep, I could make a proper pull request, but I have left this comment for now: https://gitlab.alpinelinux.org/alpine/aports/-/merge_requests/11136#note_306644
2023-05-11 04:24:32 <bootloader> okay, another question
2023-05-11 04:25:17 <bootloader> the wifi-card is up and running with iwd, however the connection is displaying as eth0 and says that it is linked by an ethernet
2023-05-11 04:26:43 <bootloader> I read that some drivers may call their driver an eth, but it doesn't explain the "ethernet link" - if that matters
2023-05-11 04:27:57 <bootloader> I can ping myself via loopback (127.0.0.1) so it is working, but I cannot ping my router - and I don't believe a firewall is up
2023-05-11 04:29:32 <bootloader> For additional context, my machine does not have an ethernet port. And rfkill shows my connection as wlan, while ip/ifconfig/iwctl show it as eth0
2023-05-11 04:29:40 <bootloader> Any thoughts?
2023-05-11 04:31:17 <ikke> bootloader: is it connected (associated) with a wireless network?
2023-05-11 04:32:45 <psykose> can you paste the logs of what you see
2023-05-11 04:40:15 <bootloader> ikke no I haven't done this. I thought the process was "wifi-driver (complete) -> iwd (complete) -> scan-for-networks (stuck) -> connect (incomplete)". How would you go about associating it?
2023-05-11 04:40:48 <ikke> bootloader: I'm not that familiar with iwd myself, so not sure how to do that with iwd
2023-05-11 04:41:21 <bootloader> psykose yeah I can, just a moment, ill use a usb to transfer
2023-05-11 04:42:52 <bootloader> ikke it says to use `iwctl device list` -> `iwctl station scan {result}` then connect. Alpine has a guide im following `https://wiki.alpinelinux.org/wiki/Wi-Fi`
2023-05-11 04:44:08 <bootloader> the part that has me stuck is that I can't scan for wireless-networks because it is only scanning eth0
2023-05-11 04:58:06 <bootloader> here are the logs `https://pastebin.com/16MAqs3B`
2023-05-11 04:59:45 <bootloader> the discrepancy between `rfkill` and everything else is strange
2023-05-11 05:02:18 <psykose> sus
2023-05-11 05:02:42 <eex0> <blinks>
2023-05-11 05:04:17 <bootloader> <strange music plays>
2023-05-11 05:05:23 <bootloader> *sussy music
2023-05-11 05:07:42 <eex0> what exactly is a sus? short sigh?
2023-05-11 05:09:53 <bootloader> if that was ironic, you've reached peak comedy. if not, its short for 'sus'pect
2023-05-11 05:10:30 <bootloader> which in turn means strange/weird
2023-05-11 05:14:24 <eex0> ah .. makes sense now
2023-05-11 05:49:27 <eex0> Q: what do you do when you've done nearly everything you can think of and hackers are still messing with you? lol
2023-05-11 06:11:31 <uniqNuuid> eex0: go live in Walden
2023-05-11 06:11:55 <uniqNuuid> by Henry David Thoreau
2023-05-11 06:13:07 <eex0> Massachusetts is too cold, I'm used to California weather
2023-05-11 06:30:49 <eex0> uniqNuuid: at this point walden is not safe
2023-05-11 07:06:33 <rsc> pj: any chance for a :latest tag? :)
2023-05-11 07:54:09 <bootloader> is `NL80211` not a driver on alpine?
2023-05-11 08:08:26 <pj> rsc: no, riscv is not supported, it's only available on edge
2023-05-11 08:14:31 <rsc> pj: ;-(
2023-05-11 08:14:37 <rsc> pj: even not with 3.18?
2023-05-11 08:15:41 <pj> edge only
2023-05-11 08:31:48 <bootloader> if anyone knows of a `NL80211` module or equivalent, please let me know
2023-05-11 08:40:21 <wyk72> hello 
2023-05-11 08:41:11 <wyk72> I'm experiencing a subtle problem with latest Alpine Linux: I can't get full speed out of a realTek RTL8125 2.5Gbit chip
2023-05-11 08:41:27 <wyk72> The kernel is 6.x (ALpine 3.18rc4)
2023-05-11 08:41:34 <wyk72> the driver is 8169
2023-05-11 08:41:44 <wyk72> chips gets detected properly, speed too
2023-05-11 08:42:13 <wyk72> but I get only 1.45Gbit speed on sending, full 2.35Gbit on receiving
2023-05-11 08:42:27 <wyk72> Now, it's a known problem of the r8169 driver, I've read
2023-05-11 08:42:51 <wyk72> and there is a Realtek-mantained version of the driver that fixes that
2023-05-11 08:43:32 <wyk72> it's called (aptly) 8125 instead of the 8169 catch-all driver
2023-05-11 08:44:15 <wyk72> I have the sources from Realtek, but I need some advice on how to compile them
2023-05-11 08:45:18 <ikke> wyk72: best would be to make an AKMS module
2023-05-11 08:46:41 <wyk72> What is  AKMS  ? Any pointers ?
2023-05-11 08:47:22 <wyk72> Oh I see it's  like DKMS but for Alpine
2023-05-11 08:56:44 <wyk72> does AKMS download kernel sources automatically ? 
2023-05-11 09:09:10 <wyk72> OMG
2023-05-11 09:09:22 <wyk72> it worked almost out of the box :)
2023-05-11 09:09:50 <wyk72> Jakub Jirutka is my new hero
2023-05-11 09:18:51 <Shiz> :)
2023-05-11 09:36:51 <wyk72> another question regarding modules
2023-05-11 09:37:30 <wyk72> there is a simple way to integrate akms into a run-from-ram Alpine usb key ?
2023-05-11 09:38:01 <wyk72> or to add a module to the .modloop without hacking it (i.e. unrolling it, rebuild, and disable signature verification) ?
2023-05-11 10:41:21 <ikke> wyk72 left
2023-05-11 11:20:39 <kaathewise> Good day!  Is `iw` package essential?  Alpine came pre-installed with it, but I don
2023-05-11 11:20:52 <kaathewise> *'t fully understand what it does and wherever it is safe to delete it
2023-05-11 11:20:59 <ikke> iw is for wireless
2023-05-11 11:21:14 <ikke> "nl80211 based CLI configuration utility for wireless devices"
2023-05-11 11:21:32 <kaathewise> Like bluetooth?  Because I use wpa_supplicant for wi-fi
2023-05-11 11:22:32 <ikke> It's a cli tool that you can use to check the status and change things
2023-05-11 11:22:38 <ikke> much like ip from iproute2
2023-05-11 11:22:56 <kaathewise> Ah, so if I do not use it, it is OK to delete the package?
2023-05-11 11:22:59 <ikke> Yes
2023-05-11 11:23:03 <kaathewise> TY!
2023-05-11 11:26:37 <ovf> even though psykose is enjoying some time off, anyone who would like to crash their firefoxes on wayland are welcome to do so: http://0x0.st/HNHy.html
2023-05-11 11:28:00 <ovf> this doesn't happen with the m.o build, so it's probably again some lld shenanigans
2023-05-11 11:29:34 <ikke> this.requestPointerLock();
2023-05-11 11:31:50 <ovf> that's something you will see in games/emulators, like https://copy.sh/v86/
2023-05-11 11:41:33 <boombim> hello everyone. How to make read -p works on alpine? which package should I install&
2023-05-11 11:41:35 <boombim> ?
2023-05-11 11:42:26 <ikke> boombim: read is a shell built-in
2023-05-11 11:42:36 <ikke> so you'd need to use a shell that supports it
2023-05-11 11:43:11 <ikke> boombim: but ash does support it
2023-05-11 12:03:26 <mbx_> apk update seems to fail on branch 3.15 atm. temporary error, please try again later
2023-05-11 12:04:20 <aron> mbx_: it's fine at me
2023-05-11 12:04:33 <aron> make sure you use http://dl-cdn.alpinelinux.org
2023-05-11 12:04:41 <ikke> mbx_: check dns / internet
2023-05-11 12:05:24 <mbx_> ye my bad, my LXC is offline, first time a new container can't call out, dang
2023-05-11 12:10:22 <mbx_> so tired, I forgot to set the gateway ahaha
2023-05-11 12:22:38 <boombim> How can I assign value to two variables in the right part of || operator?
2023-05-11 12:22:59 <boombim> oh sorry. missarddress
2023-05-11 13:47:51 <Nermalq>  Speedy Gonzalez and Jerry visit Barbra Streisand and masturbate her cooter before being expelled by her super powerful spray of kosher cum juices! https://justpaste.it/Barbra_Streisand_Speedy_Gonzalez
2023-05-11 13:52:29 <pj> ok but why
2023-05-11 13:52:41 <pj> ikke: could you kick the spam?
2023-05-11 13:54:14 <ikke> not that it matters that much
2023-05-11 13:59:19 <pj> now I got dm from them :))
2023-05-11 14:01:15 <ikke> fun
2023-05-11 14:48:28 <hodgewol> hi, i'm happily testing the netboot version of alpine but i'd like to customise the image it boots. Do i just need to create a customised chroot of alpine and then use mksquashfs to package it up and replace the original modloop-lts file?
2023-05-11 18:00:01 <JohannesJ[m]> The great Barbara doesnt masturbate :(
2023-05-11 18:04:43 <eex0> who's barbara, wondering cause, she might need a hand.. <smirks>
2023-05-11 20:03:42 <Newbyte> Why is it that backtraces without debug symbols are so limited in Alpine?
2023-05-11 20:16:44 <dwfreed> backtraces without debug symbols are pretty useless regardless of distro
2023-05-11 20:17:31 <ikke> I think what newbyte means is that you get only 2 entries or so 
2023-05-11 20:18:10 <Newbyte> yes, what ikke said. sorry for not making myself clear
2023-05-11 20:19:32 <dwfreed> usually happens when things are built without frame pointer enabled (which is not needed on x86, but makes it impossible for backtraces to actually delineate stake frames)
2023-05-11 20:19:48 <dwfreed> s/stake/stack/
2023-05-11 20:20:11 <Newbyte> thanks
2023-05-11 20:20:36 <ikke> hmm, stake frames
2023-05-11 20:21:10 <dwfreed> ikke: steak frames would be better
2023-05-11 20:21:14 <ikke> yes, I agree
2023-05-11 20:21:38 <minimal> especially if you're a vampire ;-)
2023-05-11 20:22:18 <Newbyte> Why does Alpine omit the frame pointer then? Smaller binaries?
2023-05-11 20:22:24 <dwfreed> yes
2023-05-11 20:22:39 <dwfreed> smaller binaries, less stack usage, frees up a register for other use
2023-05-11 20:23:02 <dwfreed> improves performance a small amount
2023-05-11 20:25:09 <Newbyte> Makes sense, thanks
2023-05-11 20:26:40 <ikke> export CFLAGS="-Os -fomit-frame-pointer"
2023-05-11 20:27:42 <dwfreed> yep, that would do it
2023-05-11 21:02:43 <Piraty> how exactly to make a service's stdout go to syslog?
2023-05-11 21:05:28 <eex0> good question tho I always (when looking for error) in apps do this: cd <dir or app> $ touch error.log then run it $ app 2>&1 > error.log
2023-05-11 21:06:31 <eex0> puts you in headless mode
2023-05-11 21:08:05 <eex0> I know I didn't directly answer that question and I apologies upfront
2023-05-11 21:10:59 <Apachez> -vomit-frame-pointer
2023-05-11 21:11:27 <Piraty> -fPUKE
2023-05-11 21:15:36 <rsc> pj: well...what's missing to get it supported?
2023-05-11 21:25:36 <ikke> Good server hardware 
2023-05-11 21:27:17 <ikke> We are now emulating it, which is terribly slow ( but still faster than the current hw)
2023-05-11 21:27:57 <eex0> O.o
2023-05-11 21:29:16 <rsc> ikke: but wasn't this the same issue for ARM, too?
2023-05-11 21:29:40 <ikke> No, we run arm natively 
2023-05-11 21:30:03 <ikke> The cpu supports 32-bits mode 
2023-05-11 21:31:06 <ikke> So we have a server with 80 cores for arm
2023-05-11 21:50:31 <thresh> hello.  when can we expect Amazon EC2 3.18 release images to be published?
2023-05-11 22:43:26 <minimal> thresh: probably best asking in the alpine-cloud channel
2023-05-11 22:58:45 <pj> rsc: it has to be affordable and actually powerful enough to do compute, currently all hw is shit (IMO), iirc there was #alpine-riscv or something alike, ikke can correct me about that
2023-05-11 23:50:15 <tomleb> Just realized I was on stable + testing, oops. How do you deal with staying on stable but with _some_ package from testing? Cherry-picking these package in a custom repository? 
2023-05-11 23:51:12 <minimal> tomleb: mixed packages between versions is not supported
2023-05-11 23:52:56 <tomleb> does the above solution counts as "mixed packages"? 
2023-05-11 23:54:16 <minimal> tomleb: yes. Basically most packages dependent on other packages (unless they're statically linked) and so any packages you use from testing are likely linked against newer versions of libs etc than those in whichever other "stable" release you're using
2023-05-11 23:55:29 <minimal> so if you use package X from testing that depends on libY and was built against version 2 of libY in testing and you have version 1 of libY in (for example) Alpine 3.17 then things are unlikely to work
2023-05-12 00:01:57 <tomleb> I see. Not sure how best to handle the case of dynamic library then. So far I just use statically linked packages so a non-issue. Otherwise, edge it is I guess?
2023-05-12 00:03:50 <minimal> tomleb: have you checked the packages you are using are statically linked? have you checked they don't rely on features in recent versions of programs than those in the "stable" repo you are using?
2023-05-12 00:06:03 <tomleb> no you're right some do use external programs like ffmpeg so there fould be incompatibility here
2023-05-12 00:06:31 <minimal> tomleb: did you check the programs you are using are indeed statically linked?
2023-05-12 00:06:37 <minimal> e.g "ldd"
2023-05-12 00:08:42 <tomleb> welp, ldd shows libsqlite3, libtag, a few others
2023-05-12 00:08:57 <minimal> and also musl?
2023-05-12 00:09:09 <tomleb> libc.musl yes
2023-05-12 00:09:21 <minimal> so it was linked against a particular version of musl...
2023-05-12 00:09:31 <tomleb> Looks like it
2023-05-12 00:09:55 <minimal> that was my point
2023-05-12 00:10:31 <tomleb> Yeah I see that, would you then suggest to use edge then?
2023-05-12 00:11:00 <minimal> yes, as I said initially, mixing Edge (including testing) with other releases is not supported
2023-05-12 02:08:20 <tomleb> Ended up staying on stable. I'll just rebuild the packages when they break or when a new minor version of alpine comes out. Something like REPODEST=~/packages/v3.18, etc
2023-05-12 04:12:50 <qaqland[m]> Is there any plan for /sbin/setup-wayland-base?
2023-05-12 04:14:17 <qaqland[m]> setup-desktop still use xorg
2023-05-12 07:52:13 <JohannesJ[m]> would benice
2023-05-12 07:52:19 <JohannesJ[m]> though im running KDE over wayland
2023-05-12 07:52:35 <JohannesJ[m]> though i didnt use setup-desktop
2023-05-12 08:02:12 <bbbhltz> good morning|afternoon|evening just a little question, I get warnings when upgrading. Not errors. About fgrep and os-prober. Should I change something? I haven't modified anything to do with grub since reinstalling 2 days ago.
2023-05-12 08:02:15 <bbbhltz> Warning: os-prober will not be executed to detect other bootable partitions.
2023-05-12 08:02:15 <bbbhltz> Check GRUB_DISABLE_OS_PROBER documentation entry.
2023-05-12 08:02:15 <bbbhltz> Systems on them will not be added to the GRUB boot configuration.
2023-05-12 08:03:27 <aron> bbbhltz: I always see this message, but I never experienced any problems, so I just ignore this
2023-05-12 08:04:31 <bbbhltz> aron: thanks, I didn't think it was a problem, so I am also ignoring it. Just never noticed it before this morning.
2023-05-12 08:05:55 <aron> some of my systems always printed this message, since I've installed them
2023-05-12 08:06:18 <aron> one of them is more than 2 years old
2023-05-12 08:06:49 <aron> I think they are EFI installs
2023-05-12 08:10:52 <uniqNuuid> bbbhltz: because you disabled os-prober in /etc/default/grub
2023-05-12 08:12:11 <uniqNuuid> GRUB_DISABLE_OS_PROBER=y?
2023-05-12 08:12:47 <uniqNuuid> it is disabled by default : Probing for other operating systems is disabled for security reasons
2023-05-12 08:13:17 <aron> oh, so this is the reason
2023-05-12 08:13:22 <aron> it's fine for me, tho
2023-05-12 08:13:57 <uniqNuuid> if you set : GRUB_DISABLE_OS_PROBER=false , you will not see that message 
2023-05-12 08:14:00 <uniqNuuid> you may test 
2023-05-12 08:14:08 <uniqNuuid> it does not harm your system for a test 
2023-05-12 08:14:29 <aron> (I never cared enough to investigate :P)
2023-05-12 08:15:48 <bbbhltz> uniqNuuid: OK, that line isn't even in my default/grub. I'll make a note to look at it someday, but I am ok with warnings. Same goes for the "fgrep is obscelent" thing. 
2023-05-12 08:16:20 <uniqNuuid> https://0x0.st/HNND.txt
2023-05-12 08:16:28 <abby> that message just comes from some script calling fgrep
2023-05-12 08:16:32 <uniqNuuid> ( grep GRUB_DISABLE_OS_PROBER /etc/default/grub ; sudo grub-mkconfig -o /boot/grub/grub.cfg ) 2>&1
2023-05-12 08:17:10 <uniqNuuid> and now : https://0x0.st/HNNk.txt
2023-05-12 08:17:31 <uniqNuuid> so, that's it. i've tested
2023-05-12 08:18:33 <bbbhltz> uniqNuuid: awesome, thanks
2023-05-12 08:19:49 <uniqNuuid> the fgrep thing will disappear oneday when they stop using fgrep
2023-05-12 08:19:55 <uniqNuuid> fgrep is not POSIX
2023-05-12 08:20:11 <uniqNuuid> fgrep is actually grep -F
2023-05-12 08:24:31 <uniqNuuid> alpine uses very old grub: https://ftp.gnu.org/gnu/grub/grub-$pkgver.tar.xz
2023-05-12 08:24:53 <uniqNuuid> that 2.06 is never updated 
2023-05-12 08:25:37 <uniqNuuid> it was released on 2021-06-08
2023-05-12 08:25:42 <uniqNuuid> 2 years ago
2023-05-12 08:30:26 <psykose> if they wanted us to use a new grub they would release a new one
2023-05-12 08:31:08 <uniqNuuid> arch linux uses : e67a551a48192a04ab705fca832d82f850162b64
2023-05-12 08:31:21 <uniqNuuid> which is 2023 04 13 
2023-05-12 08:31:38 <lopid> hence arch is what it is
2023-05-12 08:31:39 <uniqNuuid> yeah, it's no biggy, just saying. 
2023-05-12 08:31:57 <uniqNuuid> it just works :) psykose 
2023-05-12 08:32:11 <uniqNuuid> newer is not always better 
2023-05-12 08:40:08 <abby> i seem to remember arch using a git snapshot for grub breaking everyone's shit
2023-05-12 08:40:23 <abby> sometime in the past 6 months or so
2023-05-12 08:42:33 <aron> I never regret changing from arch to alpine :)
2023-05-12 08:42:42 <aron> also, alpine has armv7
2023-05-12 08:42:59 <uniqNuuid> abby: yes, it was due to fwsetup or stuff
2023-05-12 08:44:34 <uniqNuuid> https://archlinux.org/news/grub-bootloader-upgrade-and-configuration-incompatibilities/
2023-05-12 08:45:15 <uniqNuuid> aron: yes. multi arch is another +1
2023-05-12 08:45:31 <uniqNuuid> arch is x86_64 only, such a pitty. 
2023-05-12 08:46:17 <uniqNuuid> never realized this is an issue before until more and more arm cloud/devices coming into the market
2023-05-12 08:46:27 <aron> it had i686, but they removed it
2023-05-12 08:47:02 <uniqNuuid> so, for consistency, i'd use void/alpine/debian 
2023-05-12 08:47:22 <uniqNuuid> same update/upgrade policy on both arm and x86
2023-05-12 08:47:56 <aron> I wanted the "same" system on all of my machines (x86_64, armv7 and aarch64), I dislike debian, so I ended up with alpine
2023-05-12 08:48:47 <aron> I have a lot of custom packages
2023-05-12 08:48:51 <uniqNuuid> yeah, but musl malloc struggles on lower performent devices 
2023-05-12 08:49:08 <aron> musl is a +1 for me ;)
2023-05-12 08:49:22 <aron> glibc is such a spaceship
2023-05-12 08:49:31 <aron> have you ever looked into the implementation? :D
2023-05-12 08:49:57 <uniqNuuid> if musl gets better on malloc, it will be perfect 
2023-05-12 08:51:11 <uniqNuuid> just use llvm/clang's malloc, is a big improvement 
2023-05-12 08:51:30 <uniqNuuid> the mallocng is still not good. 
2023-05-12 08:52:40 <aron> this could be an application-specific issue, so if the program uses malloc very often, it might suffer more ;)
2023-05-12 08:53:44 <uniqNuuid> aron: try "vim /sbin/setup-disk" on your armv7 :) and scroll... 
2023-05-12 08:54:04 <aron> I'm an mc guy :P
2023-05-12 08:54:25 <uniqNuuid> oh and with syntax on
2023-05-12 08:54:37 <aron> I am (usually) able to quit from vi
2023-05-12 08:54:39 <uniqNuuid> then try mc with syntax on 
2023-05-12 08:55:08 <aron> no issue with syntax on
2023-05-12 08:55:15 <aron> just works as usual
2023-05-12 08:55:35 <uniqNuuid> aha, no laggy?
2023-05-12 08:55:42 <aron> nah
2023-05-12 08:55:58 <uniqNuuid> what about hjkl scrolling one line at a time 
2023-05-12 08:56:40 <aron> nothing
2023-05-12 08:56:47 <aron> setup-disk is only 40k
2023-05-12 08:57:01 <aron> I open much larger files with mc
2023-05-12 08:57:05 <aron> about 1M
2023-05-12 08:57:23 <uniqNuuid> oh shit, that means vim sucks 
2023-05-12 08:57:53 <aron> with syntax and jumping from here to there, mc just do it
2023-05-12 08:58:15 <psykose> does mc have actual syntax highlighting
2023-05-12 08:58:19 <psykose> because vim doesn't
2023-05-12 08:58:25 <psykose> that would explain it
2023-05-12 08:59:43 <uniqNuuid> syntax hl on vim uses lots of regex... to match {} and [] 
2023-05-12 08:59:49 <uniqNuuid> for shell scripts. 
2023-05-12 09:00:06 <aron> and that means a lot of malloc/free
2023-05-12 09:00:10 <aron> makes sense
2023-05-12 09:01:10 <uniqNuuid> aron: for mc you mean GNU Midnight Commander ?
2023-05-12 09:01:15 <aron> yep
2023-05-12 09:01:29 <uniqNuuid> it's a file manager ?
2023-05-12 09:01:50 <aron> yes, and it has a builtin text editor as well
2023-05-12 09:01:57 <aron> mc, mcedit, mcview
2023-05-12 09:02:26 <aron> hexeditor is not that great, but it works
2023-05-12 09:03:34 <aron> I keep dosens of mc instances on my armv7, so I would say, its memory footprint is not too large
2023-05-12 09:04:18 <abby> mcedit... great for making minecraft maps
2023-05-12 09:04:31 <aron> great for coding as well!
2023-05-12 09:07:22 <uniqNuuid> aron: i tryed mcview, it just gives a blue bg, the syntax is not on 
2023-05-12 09:07:44 <aron> mcview has no syntax, only mcedit
2023-05-12 09:08:01 <aron> don't ask why
2023-05-12 09:15:07 <uniqNuuid> ok. fine. i might try vis 
2023-05-12 09:15:15 <uniqNuuid> it is faster than vim
2023-05-12 09:15:46 <uniqNuuid> i am not getting used with anything other than vim's style of editing 
2023-05-12 09:15:59 <uniqNuuid> i don't even know how to use nano
2023-05-12 09:17:32 <aron> hehe
2023-05-12 09:17:50 <aron> it's the same with me and vi
2023-05-12 10:21:26 <ovf> psykose: https://0x0.st/HNHy.html
2023-05-12 10:21:49 <psykose> what about it
2023-05-12 10:24:28 <ovf> psykose: right, sorry, 0x0.st started sending a strange csp. https://planetk.org/pl.html -- another firefox crash
2023-05-12 10:24:49 <psykose> and what do you want me to do about that
2023-05-12 10:26:19 <ovf> 1. commiserate 2. consider putting at least the symbols in firefox-dbg
2023-05-12 10:26:57 <psykose> i'm not making a package for 100gb of debug symbols
2023-05-12 10:30:40 <ovf> just disabling stripping (i.e. retaining the non-debug symbols) is +46mb, and it's already the difference between 'something within libxul crashed' and 'this particular inlined overflow check got miscompiled/too aggressively optimised' (which seems to be the case this time -- upstream m.o builds work fine).
2023-05-12 10:35:11 <psykose> no, the issue is probably still https://bugzilla.mozilla.org/show_bug.cgi?id=1832173
2023-05-12 10:42:59 <ovf> sorry, i don't see how that could be relevant. if one had non-debug symbols for firefox, that crash would produce this backtrace: http://0x0.st/HNq1.txt which is inside pointer event-related telemetry code (which i incidentally can't seem to be able to disable in firefox prefs)
2023-05-12 11:29:08 <psykose> ovf: gave it a quick test, yeah just -g0 syms are roughly 40mb indeed
2023-05-12 11:29:09 <psykose> added
2023-05-12 11:30:15 <ovf> thanks!
2023-05-12 11:53:36 <uniqNuuid> what 100GB of debug symbols?
2023-05-12 11:54:25 <uniqNuuid> should i praise or suck at it. 
2023-05-12 12:28:33 <mbx> hey folks, how can I find the name of a service after installing its package? I.e. for openssh this would be sshd, how to find it if you don't know it, please?
2023-05-12 12:30:12 <psykose> apk info -L something-openrc
2023-05-12 12:32:22 <mbx> oh the -L flag is really good, so helpful. Thanks for the quick reply! <3
2023-05-12 12:33:12 <psykose> it's not an exact answer but it's very useful for a lot of stuff :p
2023-05-12 12:33:18 <psykose> it only works for installed things
2023-05-12 12:35:37 <somercet1> transmission-cli is an odd duck.
2023-05-12 12:35:40 <nu> for the other way around, to learn which package does a file belong to, you can try apk-file $file
2023-05-12 12:35:58 <somercet1> it contains transmission-remote, which is the main interface, and useless without, transmission-daemon
2023-05-12 12:37:19 <somercet1> Also, -create, -edit and -show should be common to all packages, like -cli, -remote, -gtk, &c.
2023-05-12 12:37:41 <mbx> nice
2023-05-12 12:38:11 <mbx> I wonder why php7-fpm has no service on Alpine Linux, when it has a service on Debian. I can't add it as service.
2023-05-12 12:41:50 <mbx> ok its called php-fpm7, but why it was not listed?
2023-05-12 12:42:18 <mbx> i need to learn reading, it was listed
2023-05-12 12:42:29 <mbx> sorry folks, apk info -L is awesome :)
2023-05-12 12:43:30 <psykose> somercet1: no? transmission-remote can connect to things on another computer, not just your current one
2023-05-12 12:44:11 <somercet1> then you're arguing for a package just for -remote.
2023-05-12 12:44:14 <psykose> no
2023-05-12 12:44:20 <psykose> the others are also optional
2023-05-12 12:45:00 <somercet1> but what does it have to do with -cli? That's a fully-fledged client for people who, I dunno, run all service daemons from tmux.
2023-05-12 12:45:42 <psykose> the "i dunno" is doing some heavy lifting there
2023-05-12 12:45:52 <psykose> anyway, stop inventing problems, and then you even won't have to solve them
2023-05-12 12:46:31 <somercet1> You think I'm a trouble maker?
2023-05-12 12:47:25 <psykose> where does it say that
2023-05-12 12:47:39 <somercet1> "stop inventing problems"
2023-05-12 12:47:52 <psykose> evidently we speak different languages i guess
2023-05-12 13:05:15 <userdocs_> but i can install remote on my local pc with no need for daemon, and use a remote daemon.
2023-05-12 13:05:19 <userdocs_> is that not the point?
2023-05-12 13:05:51 <somercet1> it's the bundling of -remote with -cli
2023-05-12 13:06:22 <somercet1> so if I install -daemon, I need to also install -cli to get -remote.
2023-05-12 13:06:57 <psykose> as opposed to -daemon and a theoretical -remote anyway?
2023-05-12 13:07:34 <userdocs_> i dont use transmission but i assume cli and be script with remote and used with a non local daemon.
2023-05-12 13:08:25 <userdocs_>  /cli.../cli can be scripted with..
2023-05-12 13:09:29 <psykose> nah, cli is a full torrent client to download something
2023-05-12 13:09:55 <userdocs_> just without the normal web ui?
2023-05-12 13:10:40 <psykose> it's a standalone cli client to download a .torrent or magnet etc
2023-05-12 13:11:12 <somercet1> userdocs_, it is almost literally wget for BT. I think it has an autoload dir to add torrents.
2023-05-12 13:11:31 <somercet1> Haven't used it since I first installed transmission ages ago.
2023-05-12 13:12:53 <userdocs_> the real moral os the story here is to use qbittorrent
2023-05-12 13:14:43 <userdocs_> but i think i see what you are saying, the cli package looks to be a collection of tools and the ui is possibly the odd one out?
2023-05-12 13:15:33 <userdocs_> but still as a community package, it also makes sense.
2023-05-12 13:18:58 <cockliuser[m]> btpd (BitTorrent protocol daemon) also works well for torrenting
2023-05-12 13:23:20 <psykose> lol
2023-05-12 13:24:10 <cockliuser[m]> why
2023-05-12 13:24:38 <cockliuser[m]> haven't had an issue with btpd, plus its pretty simple to use
2023-05-12 13:35:56 <oxfn> Hello! Does anybody know, why docker-compose has disappeared from 3.18 repository?
2023-05-12 13:36:15 <oxfn> Looks like yesterday it has been there
2023-05-12 13:37:32 <pj> use docker-cli-compose
2023-05-12 13:37:53 <pj> docker-compose has been deprecated for 2 years now
2023-05-12 13:38:26 <oxfn> Do you mean docker-compose package in APK registry?
2023-05-12 13:38:43 <pj> I mean docker-compose the docker software
2023-05-12 13:39:12 <pj> docker-compose has been removed from alpine 3 months ago
2023-05-12 13:39:54 <pj> did you just upgrade from 3.17 to 3.18?
2023-05-12 13:40:05 <psykose> instead of typing `docker-compose` type `docker compose` and use docker-cli-compose
2023-05-12 13:40:09 <oxfn> It is about compose v1 engine, not v2. It is very widely used and supported
2023-05-12 13:40:23 <pj> it's not
2023-05-12 13:40:30 <pj> it is abandoned for 2 years now
2023-05-12 13:41:01 <psykose> that's nice, did you actually run `docker compose` and have it not work however
2023-05-12 13:41:07 <pj> also psykose just added provides for docker-compose
2023-05-12 13:41:15 <minimal> oxfn: look at Docker's own info: https://docs.docker.com/compose/
2023-05-12 13:41:18 <pj> so it should not break or something
2023-05-12 13:41:24 <minimal> the "Important" box at the top of that page
2023-05-12 13:41:53 <oxfn> ok, got it, thanks
2023-05-12 13:42:31 <pj> also there is no such thing as compose engine
2023-05-12 13:42:32 <oxfn> separate script is deprecated I knew that. Thought you mean whole engine 
2023-05-12 13:42:40 <pj> it's literally just a cli frontend
2023-05-12 13:42:59 <pj> which they have been making compatible with old docker-compose since long time ago
2023-05-12 13:43:00 <oxfn> ok, not engine, poor mans orchestration tool :)
2023-05-12 13:43:58 <pj> it works fine <- (person who has been using docker-cli-compose for past 3 years or so)
2023-05-12 13:44:21 <pj> just docker-compose -> docker compose
2023-05-12 13:45:59 <oxfn> thanks ∑:)
2023-05-12 14:31:48 <raspbeguy> hello, is it permitted to propose aports with license Polyform-Small-Business-1.0.0 ?
2023-05-12 14:32:48 <psykose> 
ACTION googles

2023-05-12 14:34:04 <psykose> no
2023-05-12 17:07:57 <raspbeguy> psykose: ok thanks
2023-05-12 17:08:25 <raspbeguy> I guess I'll have to host my own private repos for that kind of stuff then
2023-05-12 17:29:36 <psykose> you could open an issue to ask the "Resident legal expert" but afaik OSI rejected it explicitly for the obvious reason of the money money stuff
2023-05-12 17:29:49 <ikke> Was just reading https://github.com/spdx/license-list-XML/issues/885
2023-05-12 17:30:05 <psykose> yeah, that thread is mostly just for spdx listing itself however
2023-05-12 17:30:19 <ikke> Yes, I'm aware
2023-05-12 17:30:22 <ikke> was just curious about it
2023-05-12 17:30:27 <psykose> as someone in there puts, it's not about politics
2023-05-12 17:30:32 <psykose> the actual final decisions are however
2023-05-12 17:30:35 <psykose> oh well :)
2023-05-12 17:30:50 <ikke> spdx is including more source-available licenses nowadays apparently
2023-05-12 17:31:04 <psykose> it's not a bad goal
2023-05-12 17:31:09 <psykose> merely a list of licences
2023-05-12 17:31:14 <ikke> no, I agree
2023-05-12 17:31:24 <psykose> to be clear, "final decisions" i mean on our side, not what spdx does
2023-05-12 17:31:43 <psykose> being a free software project is firmly a politics thing :p
2023-05-12 17:31:47 <ikke> We just cannot assume included in spdx == open source
2023-05-12 17:31:51 <psykose> yep
2023-05-12 17:36:04 <ikke> So PolyForm-Small-Business-1.0.0 has been included
2023-05-12 17:36:11 <ikke> (Just reached the end)
2023-05-12 17:41:01 <psykose> hope everyone is ready for some broken firewalls
2023-05-12 17:41:08 <psykose> :D
2023-05-12 17:41:10 <ikke> fun
2023-05-12 17:41:18 <psykose> works in my testing
2023-05-12 17:41:25 <psykose> you should give it a quick go too
2023-05-12 17:42:05 <ikke> it should just work with awall right? That's what kunkku mentioned
2023-05-12 17:42:19 <psykose> yeah, but to actually test it you have to reboot to not mix the commands
2023-05-12 17:42:28 <ikke> ok
2023-05-12 17:42:29 <psykose> and if it won't work and it's a machine that you need networking on...
2023-05-12 17:42:39 <ikke> I can test it on my laptop
2023-05-12 17:43:13 <psykose> package will take a bit, but you can ln -sfv xtables-nft-multi /sbin/iptables
2023-05-12 17:43:15 <psykose> and ip6tables
2023-05-12 17:43:17 <psykose> same thing
2023-05-12 17:56:00 <ikke> preliminary testing seems to work
2023-05-12 17:57:18 <psykose> awesome
2023-05-12 19:13:31 <shinobi57474858> where to find a good tutorial to setup sway?
2023-05-12 19:30:00 <raspbeguy> re-hello, I'm using KiTTY and most stuff is working (with terminfo installed) but for some reason the colors on rc-service output is missing. Ay idea where is comes from ?
2023-05-12 19:35:46 <psykose> i don't have them either
2023-05-12 19:35:51 <psykose> not sure why not, reading the code
2023-05-12 19:53:33 <asoija> howdy, I just updated to 3.18 on a laptop with an encrypted disk install (set up from the installer script in 3.17) and after rebooting it won't boot anymore. The bootloader loads but it seems that it can't "find" the encrypted disk to mount and continue from. It doesn't even show me the password prompt to decrypt
2023-05-12 19:55:11 <asoija> How can I fix this? A search around the web pointed me to this post, but the solution there (change the bootloader prompt from UUID to directly the /dev/mapper/root device) didn't seem to change anything https://dev.alpinelinux.org/~clandmeter/other/forum.alpinelinux.org/forum/installation/solvedcannot-boot-encrypted-system-after-install.html
2023-05-12 20:03:39 <umk3[m]> there is a forum? woah
2023-05-12 20:03:56 <psykose> nope
2023-05-12 20:05:43 <asoija> no, those are just the archived old posts
2023-05-12 20:09:23 <eex0> I see mozilla firefox changed an '?' to '@' and one named webcompat file and sent out a new release.. they stack trace if you deny their .xpi read access better now.. I thought this was funny
2023-05-12 20:11:06 <eex0> goes on record: Alpine as a desktop is blazing FAST! almost (to me) feels like I got a laptop upgrade to a faster cpu
2023-05-12 20:11:40 <eex0> Good job dev's
2023-05-12 20:46:24 <bootloader> Hey everyone, just updating in-case anyone was interested
2023-05-12 20:47:07 <bootloader> TL;DR = I’ve switched to Arch for this machine
2023-05-12 20:47:33 <bootloader> Long story = not every distribution is good for every task
2023-05-12 20:48:42 <bootloader> Love Alpine and I *really* love apk, but they both have their downsides when put into edge-cases like this
2023-05-12 20:50:00 <bootloader> For a blud, looking to get a low-cost workstation running, going to this extent isn’t worth it - when Arch is similar and already has it figured out
2023-05-12 20:52:12 <bootloader> I saw this mentioned the other day as well, but the Alpine Wiki is relatively baron. And I found myself using the ArchWiki 95% of the time. Obviously, this is the fault of no-one but, it is extremely frustrating
2023-05-12 20:54:10 <bootloader> I’m stubborn, so I will slowly but surely get broadcom-wl working on Alpine. But generally, if you’re also using a pre-2018 macbook, just use Arch
2023-05-12 20:58:08 <bootloader> If you’re ever interested in my progress check out my GitHub (username is fstrim). I’m still open to working on a wiki and also generally helping out in the Alpine space
2023-05-12 21:00:05 <bootloader> This OFTC username isn’t registered, and I won’t be here for a while. So until my return, and onwards, much love and thanks for all the help
2023-05-12 21:00:19 <psykose> safe travels
2023-05-12 22:59:34 <minimal> asoijs: how did you originally install Alpine?
2023-05-12 22:59:51 <minimal> asoija: ^^^
2023-05-13 00:24:27 <eex0> minimal: if it's not on video, it didn't happen .. lol
2023-05-13 02:46:10 <tomleb> I'm building a custom caddy package with the route53 plugin. What's the idiomatic way to make apk prefer that package over the caddy package in the community repo? Since my caddy package will end up with an older version than the community one.
2023-05-13 08:08:05 <uuidNuniq> tomleb: apk del caddy
2023-05-13 08:09:14 <uuidNuniq> just delete it, use your caddy instead, caddy is a statically linked binary. 
2023-05-13 08:10:02 <uuidNuniq> if you want to keep the /etc/init.d/caddy, cp it to a new name, e.g. /etc/init.d/caddy-custom
2023-05-13 08:32:46 <pj> apk add caddy=your_version
2023-05-13 09:41:05 <uniqNuuid> that's a pin pj
2023-05-13 09:41:30 <uniqNuuid> if one is gonna use a custom caddy, there is no need to keep alpine's caddy
2023-05-13 09:41:46 <uniqNuuid> just del it for god sake
2023-05-13 09:58:45 <oleovit> hi and short question; since this iptables change: https://git.alpinelinux.org/aports/commit/?id=f87a191922955bcf5c5f3fc66a425263a4588d48 , do i have to translate my classic iptables rules to nftables rules like descriped here:https://wiki.nftables.org/wiki-nftables/index.php/Moving_from_iptables_to_nftables , or are they "translated" by itself, due to the symlink. i have no clue about nftables
2023-05-13 09:59:51 <uniqNuuid> algitbot: no you don't have to
2023-05-13 10:00:56 <uniqNuuid> all you need probably is a reboot. if you don't reboot and you load the rules, you will end up with 2 pieces of rules, a legacy one and a nft one 
2023-05-13 10:01:15 <uniqNuuid> load the rules *again
2023-05-13 10:02:07 <oleovit> thx, yes, the reboot hint was during the upgrade. and i hab to save the rules again
2023-05-13 10:02:49 <uniqNuuid> you can check with this : iptables-legacy-save ; echo "========================" ; iptables-nft-save
2023-05-13 10:03:47 <uniqNuuid> that checks if you have 2 pieces of same rules on your system. 
2023-05-13 10:04:47 <uniqNuuid> if one of that command produces empty output, that means that ** variant ** is not used
2023-05-13 10:07:37 <oleovit> thx @uniqNuuid . anyway, time to learn/understand nftables syntax, even if not necessary. btw, iptables-legacy-save doesnt work, because i have no iptables-legacy pkg on the system 
2023-05-13 10:09:08 <uniqNuuid> oleovit: "doesnt work?" command not found? 
2023-05-13 10:09:41 <uniqNuuid> iptables provides both legacy and nft in one package. the differences is just a symlink
2023-05-13 10:10:03 <oleovit> yep, iptables-legacy is not installed
2023-05-13 10:10:34 <uniqNuuid> oleovit: iptables-legacy-save command not found?
2023-05-13 10:11:07 <oleovit> Unknown command: iptables-legacy-save
2023-05-13 10:11:11 <uniqNuuid> ok. 
2023-05-13 10:11:19 <uniqNuuid> maybe alpine is different. 
2023-05-13 10:11:40 <uniqNuuid> it was stripped before packaging 
2023-05-13 10:11:53 <uniqNuuid> on other systems, both exists 
2023-05-13 10:12:39 <uniqNuuid> oleovit: i tried to learn nft, but no luck, it's document is complex, there is no grammar, there is no way to learn. 
2023-05-13 10:12:56 <uniqNuuid> if they don't write a good man page for nft, then, i am not gonna learn it. 
2023-05-13 10:13:13 <uniqNuuid> i am not gonna run the nft commands which i don't understand. 
2023-05-13 10:14:55 <uniqNuuid> with iptables, you have a table a chain a match a target, it's syntax is very clear, it's doc is well written. but if you read `man nft`, i don't know what is is talking about the whole time. 
2023-05-13 10:15:16 <oleovit> maybe it has to do with this commit: https://git.alpinelinux.org/aports/commit/?id=7f294b874491e43a0badf26b476b1bc8beac39cb . and yep, right now, me too "i am not gonna run the nft commands which i don't understand.", even if iptables-restore-translate -f save.txt seems to translate the classic rules
2023-05-13 10:15:39 <uniqNuuid> maybe nft is good tech, but it's not user friendly, unless they write the doc like OpenBSD's pf.conf
2023-05-13 10:17:58 <uniqNuuid> but i use iptables-nft all the time. 
2023-05-13 10:18:16 <uniqNuuid> it is using the nft interface, but with iptables syntax. 
2023-05-13 10:19:28 <uniqNuuid> to avoid legacy and nft symlink on different systems. i replaced the command `iptables` to `iptables-nft` in all my scripts and commands. that makes sure i don't suffer from symlink
2023-05-13 10:19:39 <oleovit> yep, + i do not find an arch wiki entry. but this translation feature + the link in the 1st post -- it looks like it does the job it is intended to to, well. but i dont understand if it is properly translated, so i skip this part. you summed it up anyway: "it is using the nft interface, but with iptables syntax. " - thats all i need tu understand - for now
2023-05-13 10:20:08 <uniqNuuid> the translation command is half baked. useless
2023-05-13 10:20:30 <oleovit> iptables-nft -> thats what i will change in my function too
2023-05-13 10:21:31 <oleovit> half baked? anyway, i do not understand it anyway, so i will not implement it
2023-05-13 10:22:52 <uniqNuuid> simple ruls, it translate fine, complex matches, no, it doesn't work
2023-05-13 10:23:34 <oleovit> so, any iptables command in my function in order to generate rules, can be replaced with iptables-nft, in order "to call it directly"?
2023-05-13 10:23:48 <uniqNuuid> yes. 
2023-05-13 10:24:42 <oleovit> little be concerned about this, but that's the case with "new" things one is not used to i guess
2023-05-13 10:25:14 <uniqNuuid> there is almost no differences in syntax, legacy vs nft. if in rare cases it complains, you may need to refine your rules a little bit. 
2023-05-13 10:26:15 <oleovit> ok, lets try it and let's see. trial and error. thx again @uniqNuuid
2023-05-13 11:03:53 <oleovit> @uniqNuuid note: iptables-nft --help and iptables --help are identical. the usage of iptables-ntf in ones rules, is not existent. also, with iptables-nft-save, translation to nftables syntax will be done additionally if available. this i want to avoid for now. i execute rc-service iptables save anyway instead. i will leave it as it is and rely on the symlink. the legacy thing is obsolete: warnning 
2023-05-13 11:03:59 <oleovit> due the upgrade: the "iptables" commands now point to xtables-nft-multi instead of xtables-legacy-multi.      
2023-05-13 11:13:33 <tomleb> uniqNuuid: I was hoping I could keep it simple with the same package name so that the APKBUILD doesn't need the new name everywhere (eg: caddy-custom.initd, etc)
2023-05-13 11:14:57 <pj> uniqNuuid: clearly you did not understood they are making own package
2023-05-13 11:15:03 <pj> so pin is most appropriate
2023-05-13 11:15:29 <tomleb> pj: Anyway to do it with same package name + no pin?
2023-05-13 11:15:58 <pj> you can have tagged repo in etc/apk/repositories and add it via caddy@tag_name
2023-05-13 11:20:23 <tomleb> pj: forgot about tags, yeah that works great, thanks!
2023-05-13 11:32:55 <uniqNuuid> pj: tomleb, oh sorry. i use caddy in a different way. i build my own, i create my own init.d, i start it the way i liked. i create my own user account and file stroage directpty
2023-05-13 11:34:22 <uniqNuuid> oleovit: on a system iptables is liked to xxx-nft-multi, they are identical. but on some systems, iptables is linked to legacy. i use iptables-nft solely to avoid that thing 
2023-05-13 11:41:53 <uniqNuuid> btw, caddy the binary has a upgrade command :) try it, it is interesting 
2023-05-13 11:43:11 <uniqNuuid> and the caddy command can add plugins on the fly : caddy add-package github.com/somepkg[@branch|commit]
2023-05-13 11:54:16 <oleovit> @uniqNuuid i'm about to understand while reading what is available on the net. confusing at the first time, dealing with it, but maybe i get it. and yep, i understood your point in your last comment, thx for response
2023-05-13 12:22:49 <oleovit> have a nice weekend uniq and the whole rest and again thx
2023-05-13 16:52:29 <ziproot> I tried to install linux-firmware after booting the extended ISO. It returned errors like "failed to create /lib/firmware/intel/avs/apl/dsp-basetwl.bin" or "failed to rename /lib/firmware/intel/.apk.[long string of numbers] to /lib/firmware/intel/IntelStl.bin"
2023-05-13 16:53:37 <ziproot> How do I fix them?
2023-05-13 16:54:34 <ikke> ziproot: Are you still booted in the live iso??
2023-05-13 17:03:33 <ziproot> No, it's on the same computer as this one.
2023-05-13 17:04:16 <ikke> is it a disk install or run-from-ram?
2023-05-13 17:04:40 <ziproot> I think it's run from ram.
2023-05-13 17:05:30 <ziproot> Yeah, it's run from ram. When I reboot everything gets reset.
2023-05-13 17:06:28 <ikke> ok, that means the firmware is part of a modloop, which is read-only
2023-05-13 17:06:36 <ikke> but you can enable an overlay iirc
2023-05-13 17:08:06 <ikke> ziproot: set overlaysize in /etc/conf.d/modoop and restart the modloop service afterwards
2023-05-13 17:08:38 <ziproot> does the size matter?
2023-05-13 17:09:31 <ikke> should be large enough for the firmware
2023-05-13 17:09:55 <ziproot> OK, thank you.
2023-05-13 17:10:39 <ikke> make sure you persist that with lbu commit
2023-05-13 17:11:04 <ziproot> OK, I will.
2023-05-13 17:21:31 <ziproot> Hello, it's me again. /etc/conf.d/modloop had the overlay size as 0, which it said meant "50% of RAM." I have 16 GB RAM. When I changed it to a little more than 36.13 MB (size of linux-firmware-intel), the errors persisted. Trying to run "lbu commit usb" returned a resource busy error.
2023-05-13 17:52:51 <msiism> Something seems to be wrong with some icons or some such used on the wiki. Screenshot: https://www.msiism.org/fd/ik2Zyt/wikiwhoopsie.png
2023-05-13 17:57:37 <psykose> ah that thing again
2023-05-13 17:57:38 <eex0> dude/ette use a public image host might get someone to view that
2023-05-13 17:58:05 <eex0> I was wrong
2023-05-13 18:42:11 <psykose> don't actually remember what causes it
2023-05-13 18:45:43 <psykose> and it fixed itself
2023-05-13 18:45:43 <psykose> nice
2023-05-13 18:53:20 <ziproot> Hello. I downloaded the Alpine Linux "Extended" ISO and booted it from a USB. After booting, I tried to install linux firmware. I got errors like "failed to create /lib/firmware/intel/avs/apl/dsp-basetwl.bin" or "failed to rename /lib/firmware/intel/.apk.[long string of numbers] to /lib/firmware/intel/IntelStl.bin." I tried changing modloop overlay size but it was already 8GB before I changed it and that did not help. Furthermore, lsb 
2023-05-13 18:53:20 <ziproot> commit usb returned a network busy error. Is there any way I can get the firmware installed on the live ISO or do I have to continue with installation without internet?
2023-05-13 19:04:28 <eex0> ziproot: following the alpine installation doco? I noted: "after booting, I tried (x)" and that part doesn't follow any doco I've read. 
2023-05-13 19:05:19 <ziproot> Isn't that step 4 in "the general course of action?"
2023-05-13 19:05:23 <eex0> try again with proper installation technique
2023-05-13 19:05:29 <ziproot> "4. Booting the target computer from the prepared disk or storage device."
2023-05-13 19:14:20 <msiism> The LVM on LUKS guide on the wiki says that using GRUB with LUKS2 is still experimental. Is that still the case?
2023-05-13 19:15:40 <msiism> Ref: https://wiki.alpinelinux.org/wiki/LVM_on_LUKS#Luks2
2023-05-13 19:18:30 <eex0> msiism: "cryptsys" worked great for me
2023-05-13 19:20:31 <minimal> msiism: LUKS v2 does not work as "FDE" (i.e. /boot is encrypted) with the current Grub version, nothing to do with Alpine specifically
2023-05-13 19:21:24 <ziproot> So, is there a way to install Alpine Linux over WiFi or do I install without internet first, then add WiFi?
2023-05-13 19:21:58 <minimal> whenever Grub 2.12 (the next planned upstream release) comes out then LUKS2 should work for encrypted /boot
2023-05-13 19:22:19 <msiism> Okay, good to know.
2023-05-13 19:22:28 <minimal> as Grub git contains multiple patches to fix Grub's LUKSv2 support
2023-05-13 19:23:19 <eex0> minimal: maybe will fix the (long standing) cve for grub and jpg image  
2023-05-13 19:23:32 <minimal> that Wiki article's section appears to be trying to "kludge" around the LUKSv2 problems but it looks plain wrong to me (missing things)
2023-05-13 19:24:02 <minimal> eex0: why not look at Grub git to see?
2023-05-13 19:24:28 <msiism> So, maybe just using an unencrypted /boot and syslinus is the easiest solution to this.
2023-05-13 19:24:37 <msiism> s/linus/linux/
2023-05-13 19:24:44 <eex0> minimal: wasn't a question
2023-05-13 19:25:44 <minimal> eex0: it wasn't a statement either, and looking at the git history would change your "maybe" to either a "it does" or "it does not"
2023-05-13 19:26:27 <minimal> msiism: why use syslinux? you can use Grub with an unencrypted /boot
2023-05-13 19:26:46 <minimal> or you can use Grub with LUKSv1 and "FDE"
2023-05-13 19:26:56 <eex0> minimal: nods lots, thank you!
2023-05-13 19:27:53 <minimal> no idea when Grub 2.12 will actually be released - it was "aimed" for October last year, then a couple of months ago it was also "aimed", still no sign of it
2023-05-13 19:28:58 <msiism> Hm… I would have the opportunity to wait, actually.
2023-05-13 19:29:10 <msiism> At least for a few months.
2023-05-13 19:30:20 <msiism> But then, you can also convert LUKS1 to LUKS2… So, it's probably just fone to go with GRUB and LUKS1 for full disk encryption now.
2023-05-13 19:39:15 <ziproot> OK, let me rephrase the issue. After following step 4 of the installation guide: "booting the target computer from the prepared disk or storage device," I tried to run the command "apk add linux-firmware." I got errors like the ones documented above. I was told a possible fix was to enable an overlay, however, the fix given to enable the overlay involved setting overlay size, but overlay size was already set at 8GB (half of RAM). As a 
2023-05-13 19:39:15 <ziproot> result, this did not fix the issue. Is there any other way to fix the issue?
2023-05-13 19:41:24 <minimal> ziproot: which installation guide?
2023-05-13 19:41:31 <ziproot> https://wiki.alpinelinux.org/wiki/Installation#Basic_Installation_Step_Details
2023-05-13 19:41:49 <ziproot> Oh wait that was too far down: https://wiki.alpinelinux.org/wiki/Installation#Installation_Overview
2023-05-13 19:43:44 <minimal> that's a wiki doc, so written/modified by "random" people
2023-05-13 19:44:10 <ziproot> The wiki says that if it is wrong to look for a solution here.
2023-05-13 19:44:19 <minimal> if you do "linux-firmware" then you'll get lots of different firmware packages installed which could take up considerable space
2023-05-13 19:44:41 <ziproot> linux-firmware-intel produces the same errors, and I need that for wifi
2023-05-13 19:46:07 <ziproot> I am going to make an account on the wiki that is linked on alpinelinux.og and point to the official installation guide, which I just learned is not the same thing.
2023-05-13 19:48:25 <minimal> ziproot: perhaps I was not specific enough, installing ANY firmware typically results in ALL firmware packages being installed unless instead you do "apk add linux-firmware-none linux-firmware-intel"
2023-05-13 19:48:34 <ziproot> oh I see
2023-05-13 19:48:44 <minimal> installing "linux-firmware-none" stops ALL being installed
2023-05-13 19:50:23 <psykose> that's not true, just linux-firmware-intel would install only that
2023-05-13 19:50:37 <psykose> the default is -any that installs -all but if you install -intel you get just -intel
2023-05-13 19:50:39 <psykose> don't have to add -none
2023-05-13 19:51:09 <ziproot> I have updated the wiki to warn that the installation guide is unofficial and link to the official guide at docs.alpinelinux.org. Thank you for your assistance.
2023-05-13 19:53:19 <eex0> is "setup-alpine" to hard to type?
2023-05-13 19:53:32 <minimal> psykose: did that behaviour change at sometime? I got into the habit of installing linux-firmware-none to stop auto installation of unwanted firmware
2023-05-13 19:53:49 <psykose> not in years no
2023-05-13 19:54:03 <psykose> i mean, you do want -none because otherwise just `apk add linux-lts` gives you all the firmware, sure
2023-05-13 19:54:13 <psykose> but not if you add something specific there's no point
2023-05-13 19:54:24 <minimal> ah, that's what I'm thinking of, the kernel triggering it all
2023-05-13 19:55:39 <ziproot> setup-alpine is not the hard part, it's the "getting wifi working when wlan0 doesn't show up without intel firmware" part that's the hard part.
2023-05-13 19:57:01 <psykose> to save you some time i don't think anyone here remembers how to get a large packaged installed onto the ram-only system
2023-05-13 19:57:04 <eex0> reminds me, the other day it was discovered psykose about the wifi with no kernel driver.. remember? I am wondering if this is related
2023-05-13 19:57:19 <psykose> iirc you either have to make the modloop bigger or remount something rw or some shit, i don't remember
2023-05-13 19:57:29 <psykose> in general though it's easier to just use wired networking
2023-05-13 19:57:43 <frag_> is it bad to upgrade (incl kernel) and not reboot?
2023-05-13 19:58:21 <psykose> only if you expect to load more modules
2023-05-13 19:58:25 <psykose> otherwise not really
2023-05-13 19:58:34 <frag_> nice :]
2023-05-13 20:00:52 <ziproot> I can install with ethernet. However, this is bringing back memories of being forced to install Debian on ethernet, then installing intel firmware and it not working. However, it seems like it will be faster to just install over ethernet, so that is what I will do. Thank you.
2023-05-13 20:03:10 <asoija> minimal: sorry just saw your ping. I originally installed 3.17 by burning the x86_64 .iso into a USB stick and running setup-alpine. When it came to the disk setup part, I choose crypt and then sys. The install was flawless and it worked every single boot until I updated it.
2023-05-13 20:05:27 <asoija> I suspect it's something about the bootloader not being able to find the mountpoint / id of the encrypted disk to proceed with the booting? I could boot a live USB and fix /boot configs, but I don't know where to start or where to look.
2023-05-13 20:07:16 <fragstone> Hello
2023-05-13 20:08:29 <fragstone> can anyone give me a tip who i can upgrade a kernel on an raspi non hdd installation? apk --fix told me that he cannot write on /lib/modules during this is a ram disk
2023-05-13 20:11:09 <minimal> fragstone: I don't use ram-from-ram but I thought there were only two solutions for it: (1) install a newer version of Alpine, or (b) use update-kernel but it has some limitations/issues which someone else can probably explain
2023-05-13 20:12:12 <minimal> ^ install != upgrade, install from scratch to replace existing installation
2023-05-13 20:12:18 <ikke> yes, correct
2023-05-13 20:12:27 <fragstone> @minimal ah i don't know that script. that is a good hint 
2023-05-13 20:12:32 <ikke> I've heard update-kernel might require quite some memory though
2023-05-13 20:12:49 <minimal> ikke: yeah that's why I mentioned "issues" ;-)
2023-05-13 20:16:26 <fragstone> ok update-kernel doesn't work ... will also write to the ramdisk
2023-05-13 20:24:06 <scara> # apk add yggdrasil
2023-05-13 20:24:08 <scara> 1 error; 1343 MiB in 277 packages
2023-05-13 20:24:23 <ikke> scara: did you get an error message?
2023-05-13 20:25:00 <scara> No.
2023-05-13 20:25:01 <psykose> that means the error is from before you added anything if 'nothing happened but there is an error'
2023-05-13 20:25:09 <psykose> from the last time you ran something
2023-05-13 20:25:11 <ikke> apk fix
2023-05-13 20:26:04 <scara> Oh, that was easy, apk fix yggdrasil.
2023-05-13 20:26:59 <ikke> It tries to modprobe tun, if it cannot do that it will return an error
2023-05-13 20:27:56 <ikke> psykose fixed it already
2023-05-13 20:36:44 <scara> I think its the modules issue, I'm running an older kernel then the modules installed, on arch I have kernel-modules-hook installed that allows me to keep old modules installed so that I first update when i reboot.
2023-05-14 01:01:19 <somercet1> zathura-pdf-mupdf gave the following mupdf error: incompatible header (1.22.0) and library (1.22.1) versions
2023-05-14 01:01:47 <somercet1> zathura-mupdf needs a bump
2023-05-14 03:29:52 <Quillith> Has anyone used Ryujinx on alpine? I can't get it to install firmware, but when I try the same files on a PopOS computer it works fine
2023-05-14 03:31:07 <Quillith> specifically it gives me an error that 'the firmware file does not exist'
2023-05-14 04:18:45 <Mari[m]> how are you running it?
2023-05-14 04:18:52 <Mari[m]> i used the nix package and it worked
2023-05-14 07:36:52 <wyk72> hello everyone
2023-05-14 07:37:07 <wyk72> I have a question (maybe a dumb one)
2023-05-14 07:38:11 <wyk72> it's possible to use a different file system for an usb-key based Alpine ? i.e. booting from a f2fs disk instead of vfat/fat32 ?
2023-05-14 07:38:52 <wyk72> I'm hitting a problem with a 64Gb usb key: the apkovl file reached 4GB which is the maximun fat32 allows
2023-05-14 07:39:20 <wyk72> I know it's  a corner case, and maybe overall kind of stupid
2023-05-14 07:40:05 <wyk72> I can split the usb disk in two and use f2fs on second partition w some scripts, but I was thinking of using the whole disk with f2fs instead of vfat/fat32
2023-05-14 07:41:37 <wyk72> using lbu also on a second partition seems impossible without modifying the init script
2023-05-14 07:41:52 <wyk72> a thing I'd like to avoid
2023-05-14 07:45:15 <grawity> tbh I skipped all of that (partly because I didn't know apkovl etc. even exists) and just mkfs'd a partition on the USB stick, untarred the bootstrap tarball, and installed syslinux
2023-05-14 07:46:34 <wyk72> I can use thedisk in "sys" mode, and I did that once, but the USB key (old one) did not survive and faulted after 3 months (too many writes)
2023-05-14 07:47:01 <wyk72> The machine I use for those experiments is an old server w 64GB of ecc ram
2023-05-14 07:48:32 <wyk72> ..maybe it's just old "thinking" on my side, nvme disks are 20€ now on Amazon, I should just quit abusing USB keys
2023-05-14 07:54:26 <grawity> I don't see how that would be different between an overlay file and a partition...
2023-05-14 07:57:18 <wyk72> it's about minimizing writes to the usb drive
2023-05-14 08:01:53 <grawity> but like, stuff that isn't going to be written to, isn't going to be written to *either way*, no?
2023-05-14 08:14:43 <voltagex> what is the minimum viable apkovl.tar.gz? I know lbu creates them, the init script looks a bit strange to my eyes (run df -P against the .tar.gz itself?). I am thinking of writing a server that generates the apkovl on the fly.
2023-05-14 08:16:58 <dwfreed> when given a path, df will print only the information about the filesystem containing that path
2023-05-14 08:18:30 <voltagex> Ah, that makes sense
2023-05-14 08:18:40 <voltagex> I was getting an error with that line when generating the file myself
2023-05-14 09:56:12 <oleovit> sry me again. i switched to nftables. stateful firewall. but 2 lines are not applied. so i need the file, where my own ruleset is saved via rc-service nftables save. i cant find it. where is it located, so i can test it with some nft commands if its valid. except my last 2 rules, everything seems to be fine across reboots ( nft list ruleset , et cetera) 
2023-05-14 10:00:57 <oleovit> + i thought my own rules are stored in /var/lib/nftables/ according to the gentoo wiki - but there is only an empty state.nft file
2023-05-14 10:38:35 <psykose> /etc/nftables.nft pretty sure
2023-05-14 10:38:52 <psykose> and that state file
2023-05-14 10:38:55 <oleovit> it loaded the predefiened  /etc/nftables.nft , similiar to mine, so i thought, my rules applied. they didn't
2023-05-14 10:42:58 <psykose> where did you write your rules
2023-05-14 10:46:23 <oleovit> like this, in a function: https://transfer.sh/4mTKEx/code.txt
2023-05-14 10:46:51 <oleovit> i was used to it, due to iptables
2023-05-14 10:53:07 <psykose> ah
2023-05-14 10:53:18 <psykose> i think you might have to tweak some conf.d setting for nftables then, not sure
2023-05-14 10:53:24 <psykose> you might prefer to just write the rules by hand though
2023-05-14 10:53:27 <psykose> in /etc/nftables.d/
2023-05-14 10:53:36 <psykose> it gets loaded by .conf
2023-05-14 10:56:05 <oleovit> tried it again, def. the predefined rules with all the comments. yeah, maybe that's the only way to do it... i will try it, but it whould be cool to define it just like it was possible with iptables. let's take a look, thx psykose. mothers day.. i have a meeting before :(
2023-05-14 10:57:54 <psykose> good luck :)
2023-05-14 10:57:59 <psykose> it's probably possible, i never looked at it
2023-05-14 10:58:27 <oleovit> ahh, no time for this "meeting", thx ;)
2023-05-14 10:59:19 <oleovit> maybe there will be a change in the future, dont stress yourself, you have enough
2023-05-14 11:10:10 <felixphew> hi! if I have a key to decrypt a zfs root that lives in a file on /boot, is there an easy way to make the initramfs mount /boot before it tries to mount root (so the key is there)?
2023-05-14 11:51:38 <JohannesJ[m]> so, my next dip in the Alpine world is a phona that runs postmarketos
2023-05-14 11:54:29 <JohannesJ[m]> felixphew /boot is an UEFI partition?
2023-05-14 11:54:35 <JohannesJ[m]> or EFI
2023-05-14 11:54:37 <JohannesJ[m]> heh
2023-05-14 11:54:41 <felixphew> yes
2023-05-14 11:55:58 <JohannesJ[m]> see here: https://wiki.alpinelinux.org/wiki/Setting_up_disks_manually maybe this is of help
2023-05-14 11:56:37 <felixphew> thanks, I've read that :D
2023-05-14 11:56:43 <JohannesJ[m]> oh, sorry :(
2023-05-14 11:56:58 <felixphew> np :)
2023-05-14 11:57:13 <JohannesJ[m]> i use LUKS, and my /boot is an EFI partition too
2023-05-14 11:57:52 <felixphew> sure, but I assume you enter your LUKS key at boot? the idea here is that unattended reboots should be possible
2023-05-14 11:58:08 <JohannesJ[m]> ahh :) yes thats a difference
2023-05-14 11:58:52 <JohannesJ[m]> the whole idea of a fully encrypted disk in my case is that its not possible to access it upon a reboot :)
2023-05-14 11:58:52 <felixphew> (the kernel & the root partition key both live on an EFI partition on a USB drive; the internal drives are just ZFS)
2023-05-14 11:59:23 <psykose> the cryptkey= option is what you want i think
2023-05-14 11:59:28 <psykose> i don't know the mount order though
2023-05-14 11:59:30 <JohannesJ[m]> but doesnt initramfs mount /boot automaticly?
2023-05-14 11:59:45 <psykose> if you just have the efi it probably mounts it before reading cryptkey=
2023-05-14 11:59:52 <felixphew> idk, and would it mount it to /boot or /boot/efi by default?
2023-05-14 12:00:08 <JohannesJ[m]> good question
2023-05-14 12:00:16 <felixphew> the fstab inside the initramfs doesn't mention either; maybe I should expand the whole initramfs and have a closer look
2023-05-14 12:02:22 <JohannesJ[m]> looking here: https://wiki.alpinelinux.org/wiki/Installing_on_GPT_LVM i would say the partition is mounted to /boot
2023-05-14 12:03:05 <felixphew> hmm, OK; maybe it'll just work then!
2023-05-14 12:03:25 <JohannesJ[m]> you can always try :)
2023-05-14 12:05:23 <felixphew> well, I wanted to be relatively sure before I reboot and lose my convenient SSH session
2023-05-14 12:12:13 <felixphew> ah, lovely, mkinitfs built me an initramfs without zfs tools in it
2023-05-14 12:12:27 <felixphew> I'm sure I screwed something up to make that happen, but I'm not sure what
2023-05-14 12:15:12 <psykose> not having zfs in features=
2023-05-14 12:15:36 <felixphew> that'd be it!
2023-05-14 12:16:55 <felixphew> I'm used to dracut picking up on stuff like that automagically (I've been in fedora/rh/suse land lately)
2023-05-14 12:35:06 <felixphew> while we're at it, don't suppose you know why it spends a full minute "Spinning up disk..." for both sda and sdb (both SSDs)?
2023-05-14 12:35:18 <felixphew> after it times out with a "not responding..." they come up perfectly
2023-05-14 12:38:56 <felixphew> JohannesJ[m]: and no, it doesn't auto-mount /boot :'(
2023-05-14 12:39:04 <felixphew> at least now I can manually mount it
2023-05-14 13:10:09 <scara> Does apk support libalpm?
2023-05-14 13:10:17 <scara> libalpm hooks*
2023-05-14 14:11:27 <uniqNuuid> scara: no hooks. hook are built into package 
2023-05-14 14:12:27 <uniqNuuid> * it is possible to attach a hook to a package*, but no extra hook support at the moment. 
2023-05-14 14:14:07 <ikke> A package can declare triggers when certain file chaange
2023-05-14 14:14:09 <ikke> change*
2023-05-14 14:42:38 <minimal> felixphew: no initramfs' init does not mount /boot - the bootloader (assuming you're using Grub) does earlier in order to load the kernel and initramfs
2023-05-14 14:43:17 <felixphew> ah. I'm using the kernel stub, so that's not happening
2023-05-14 14:43:34 <minimal> the initramfs' init only mounts /
2023-05-14 14:43:34 <felixphew> (surely grub wouldn't pass the mounted filesystem over to the kernel, though, right?)
2023-05-14 14:44:19 <minimal> felixphew: no Grub does not "pass" a mounted filesystem, Grub has to mount "/boot" in order to get to the kernel and initramfs files
2023-05-14 14:45:14 <felixphew> yeah, exactly. so whether or not grub "mounts" it would make no difference to me, as I need the filesystem to be there at initramfs-init-run-time to load the key from
2023-05-14 14:45:33 <minimal> indeed and as I said initramfs' init only mounts /
2023-05-14 14:45:41 <felixphew> what do you think is the most sensible way to _make_ the initramfs mount it? is there a standard way to add hooks or scripts?
2023-05-14 14:47:10 <minimal> you want to use a cryptfile that is stored on "/boot" / ESP partition's fs?
2023-05-14 14:47:52 <felixphew> a zfs raw key stored on the ESP, yes
2023-05-14 14:48:04 <felixphew> keyfile*
2023-05-14 14:49:38 <felixphew> (if you're wondering "why?": key on a USB is to allow unattended reboots; and if there's going to be a USB in there at reboot time anyway, I may as well have that be my ESP so I can use the whole main drives for ZFS)
2023-05-14 14:50:48 <minimal> does initramfs' init support that? Looking at it's encrypted ZFS code I see it calling "zfs load-key"
2023-05-14 14:51:56 <felixphew> `zfs load-key` will pick up the file automatically, assuming it exists in the right place when the command is run
2023-05-14 14:52:10 <minimal> what "right place"?
2023-05-14 14:52:48 <felixphew> right now, /boot/<whatever>.key, but I can configure it to be whatever
2023-05-14 14:53:05 <minimal> felixphew: you can configure it how?
2023-05-14 14:53:26 <minimal> the initramfs' init code doesn't do any mounting of things like USB sticks in order to get a key there
2023-05-14 14:53:50 <felixphew> `keylocation` zfs property
2023-05-14 14:54:36 <minimal> but there's no code using that currently...
2023-05-14 14:54:43 <minimal> here's what it currently does: https://gitlab.alpinelinux.org/alpine/mkinitfs/-/blob/master/initramfs-init.in#L314-320
2023-05-14 14:55:05 <felixphew> yes, I know, I have that open too :) I mean `zfs load-key` uses that property
2023-05-14 14:55:24 <felixphew> the thing I need to work out how to do is modify that init to mount a filesystem for me
2023-05-14 14:55:45 <felixphew> but it doesn't look like "features" can add code to the init script, just files and kernel modules
2023-05-14 14:55:54 <felixphew> so I might need a custom mkinitfs
2023-05-14 14:56:17 <minimal> yes unless you work out the changes and submit a MR to get them applied
2023-05-14 14:56:50 <minimal> or you switch to a different initramfs
2023-05-14 14:58:04 <felixphew> I can't see a sensible way to add it in an upstreamable way
2023-05-14 14:58:22 <felixphew> code to e.g. copy the details of the filesystem to mount out of the host's fstab at initrd build time would be pretty janky
2023-05-14 14:58:56 <felixphew> hardcoding it to only do it for /boot doesn't seem generally useful - most people doing this sort of thing with a keyfile won't have it on /boot, it'll be on some other filesystem
2023-05-14 14:58:59 <minimal> isn't the hosts' /etc/fstab copied into the initramfs at creation time (from memory)?
2023-05-14 14:59:10 <felixphew> doesn't appear so
2023-05-14 14:59:26 <minimal> it's not listed in /etc/mkinitfs/features.d/base.files?
2023-05-14 14:59:42 <minimal> thought it was, would have to check
2023-05-14 14:59:48 <felixphew> the fstab in my initramfs is a generic Alpine one
2023-05-14 15:00:00 <felixphew> doesn't match my local fstab
2023-05-14 15:00:58 <felixphew> and no, it's not in base.files
2023-05-14 15:01:03 <minimal> ah, it's not in base.files, so add it to there and rebuild initramfs?
2023-05-14 15:01:11 <minimal> I think I did that in the past
2023-05-14 15:02:45 <felixphew> it seems to use /usr/share/mkinitfs/fstab, so I could modify that I suppose
2023-05-14 15:03:28 <felixphew> because in general I don't think you really want your host fstab in the initramfs unmodified (it'd work fine in my case)
2023-05-14 15:03:37 <minimal> NO, add "/etc/fstab" to base.files and then when the initramfs is (re)built it will be copied ibn
2023-05-14 15:04:10 <felixphew> yes, I see what you're saying
2023-05-14 15:04:11 <minimal> in general? I'm talking about you adding it to your base.files
2023-05-14 15:06:26 <felixphew> I know what you're suggesting, and I think it'll work for me. I'm just thinking about what a version of this that could be upstreamed would look like, and I think keeping a special fstab for the initrd and just customising that is maybe the way to go. I'm not trying things right now, it's late here.
2023-05-14 15:06:58 <minimal> well you already said you didn't see a sensible way to add it upstream...
2023-05-14 15:07:31 <felixphew> and now I'm reconsidering, now that you've brought my attention to how the fstab in the initrd gets copied in
2023-05-14 15:07:57 <minimal> for *LUKS* encryption if "cryptkey=yes" is passed on cmdline then /crypt_keyfile.bin (in initramfs) is used to unlock rootfs
2023-05-14 15:08:35 <minimal> if instead "cryptkey=/random/path/key.bin" is passed then /random/path/key.bin (in initramfs) is used instead
2023-05-14 15:08:56 <minimal> however for LUKS there is also no mechanism for mounting a device to get the key from
2023-05-14 15:10:53 <minimal> so there may be some interest in providing a mechanism to mount a specified fs for both LUKS and encrypted ZFS to use a keyfile from
2023-05-14 15:12:27 <felixphew> I'll ask about what people think the best interface for that is tomorrow. maybe a parameter `premount=/whatever` and then the details for the mount are taken from the fstab?
2023-05-14 15:13:19 <minimal> why from fstab?
2023-05-14 15:13:19 <felixphew> but I have to call it a night now. thanks for the help :)
2023-05-14 15:13:47 <minimal> also for USB devices the name can change depending on which ports things are plugged into
2023-05-14 15:14:28 <felixphew> because specifying a whole mount on the cmdline sounds kind of cumbersome? it's just the first idea that came to my head
2023-05-14 15:14:42 <minimal> e.g. you have both USB stick with OS and USB stick with key - if you plug OS stick into 1st port it might be called "/dev/sda" whereas if you plug it into another port it might be called "/dev/sdb"
2023-05-14 15:14:54 <felixphew> well, in an fstab we can use UUID=
2023-05-14 15:14:58 <felixphew> (or similar)
2023-05-14 15:15:47 <minimal> or specify UUID as value of "premount"?
2023-05-14 15:17:14 <felixphew> you need to specify a device (whether path or uuid or whatever) and a mount location and maybe even options. fitting that all nicely into one cmdline parameter sounds kind of clunky
2023-05-14 15:17:33 <felixphew> the mkinitfs init already knows how to parse an fstab
2023-05-14 15:19:26 <felixphew> getting an fstab in there seems easy (a feature with /etc/fstab in its .files is enough if you can use the host fstab)
2023-05-14 15:49:39 <ziproot> I was able to install Alpine Linux and get wifi working just fine, however, when I ran the setup-xorg-base command and rebooted I lost internet access, including access to ethernet.
2023-05-14 15:51:55 <ikke> Do you still see the interfaces?
2023-05-14 15:53:17 <ziproot> ip a returns lo only
2023-05-14 15:58:05 <Quillith> My connection broke after I asked this yesterday, so sorry if someone answered- has anyone had success using Ryujinx on alpine? It's a flatpak, but for some reason I get 'firmware file does not exist' on Alpine when i don't get that error using the same files on PopOS
2023-05-14 16:15:11 <ziproot> As I have a system I can boot into that is on my hard drive I am able to run commands into log files, mount them onto this system, and upload to pastebin, if necessary.
2023-05-14 16:17:12 <minimal> ziproot: does your /etc/network/interfaces file look ok? perhaps something like network manager was installed when you setup X?
2023-05-14 16:18:18 <ziproot> https://pst.moe/paste/otprmn
2023-05-14 16:20:42 <ziproot> minimal: it looks fine
2023-05-14 16:22:26 <minimal> and are you using iwd / wpa_supplicant / network manager?
2023-05-14 16:23:03 <ziproot> minimal: I am using wpa_supplicant
2023-05-14 16:25:42 <minimal> have you checked if wpa_supplicant is running? have you checked logfiles? have you tried running the /etc/init.d/networking script?
2023-05-14 16:26:49 <ziproot> let me reboot into alpine and run dmesg, I can output this to a pastebin in a few minutes
2023-05-14 16:28:27 <minimal> dmesg will show kernel-related messages, but other stuff may appear in /var/log/* files
2023-05-14 16:33:31 <ziproot> minimal: here is the output from dmesg https://pst.moe/paste/lcjfda
2023-05-14 16:33:41 <minimal> dmesg will show kernel-related messages, but other stuff may appear in /var/log/* files
2023-05-14 16:35:27 <ziproot> var/log/messages: https://pst.moe/paste/ifrqfn
2023-05-14 16:35:56 <minimal> your ESP filesystem needs a fsck
2023-05-14 16:37:43 <ziproot> May 14 07:33:45 localhost daemon.err /etc/init.d/wpa_supplicant[2475]: Could not find a wireless interface
2023-05-14 16:37:49 <minimal> your wpa_supplicant is failing
2023-05-14 16:38:26 <minimal> "Could not find a wireless interface"
2023-05-14 16:38:29 <ziproot> it seems like it tries to run, but fails because it can't find wlan0 interface, which checks out with "ip a" returning lo only
2023-05-14 16:39:28 <ziproot> I can try running fsck to see if that fixes or at least identifies the problem
2023-05-14 16:39:29 <minimal> possibly related to the udev errors
2023-05-14 16:39:39 <minimal> the fsck is unrelated
2023-05-14 16:39:45 <ziproot> oh ok nvm
2023-05-14 16:41:19 <minimal> "apk fix eudev" perhaps?
2023-05-14 16:42:23 <ziproot> how do I do that without internet?
2023-05-14 16:43:59 <ziproot> is it possible to install from local filesystem?
2023-05-14 16:44:28 <ziproot> oh wait it is, ok I will try that
2023-05-14 19:11:25 <raspbeguy> hello, having this error when trying to install gnome: https://termbin.com/3qth
2023-05-14 19:11:34 <raspbeguy> I'm on Edge
2023-05-14 19:14:47 <ziproot> Hello. I have fixed wifi, but xserver wouldn't run. I checked log and it was due to missing drivers which I added, however, now I get error "sna_option_accel_blt:18146 assertion 'sna->kgem.gen < 0120' failed."
2023-05-14 19:15:41 <ziproot> according to irclogs someone else had the same problem and it involved xf86-video-intel, which I have installed
2023-05-14 19:15:42 <minimal> ziproot: did you fix the udev error
2023-05-14 19:15:44 <minimal> ?
2023-05-14 19:15:51 <ziproot> yes, udev error is fixed
2023-05-14 19:16:51 <ziproot> the problem mentioned was actually only a few weeks ago, with edge version, and since then edge became v3.18 which I have installed
2023-05-14 19:16:56 <ziproot> (on irclogs)
2023-05-14 19:16:58 <ikke> raspbeguy: it apperently needs to be rebuilt
2023-05-14 19:17:55 <ikke> psykose: apparently gvfs-afc was skipped in the libplist rebuild
2023-05-14 19:18:22 <ziproot> my gpu is "Mesa Intel(R) Iris(R) Plus Graphics (ICL GT2)
2023-05-14 19:19:37 <ikke> raspbeguy: should be fixed in ~15 minutes
2023-05-14 19:19:44 <psykose> good catch
2023-05-14 19:19:45 <ziproot> also known as "Intel Corporation Iris Plus Graphics G7 (rev 07)"
2023-05-14 19:19:52 <ikke> or maybe longer, as the builders are busy
2023-05-14 19:19:58 <raspbeguy> Great
2023-05-14 19:19:59 <psykose> ikke: could you kill riscv
2023-05-14 19:20:06 <ikke> RIP
2023-05-14 19:20:23 <ikke> psykose: it's still continuing
2023-05-14 19:20:25 <psykose> yeah
2023-05-14 19:20:28 <psykose> but i want it to build gcc
2023-05-14 19:20:31 <ikke> oh
2023-05-14 19:20:32 <psykose> so i can do some gcc rebuilds
2023-05-14 19:20:51 <psykose> (gcc doesn't upgrade until main is done so i have to wait for it everywhere)
2023-05-14 19:21:04 <ziproot> ok I will try to apply advice for the other person, let me see if that works
2023-05-14 19:21:04 <psykose> shortcomings of build-base and no rootbld :p
2023-05-14 20:13:17 <Saijin_Naib> Is acf-awall the only graphical way to interact with awall, and is that better than gufw/ufw for in-home usage? 
2023-05-14 20:13:55 <Saijin_Naib> I need something easy, and gufw was that, but I'm trying to be a bit more Alpine instead of relying on old Debian/*buntu habits
2023-05-14 20:14:11 <pj> as if alpine has resources to develop multiple different graphical interfaces
2023-05-14 20:14:27 <Saijin_Naib> 🤷
2023-05-14 20:15:06 <Saijin_Naib> Maybe not, but a plug in to another front-end, or maybe it's just acf. Hence the question mark, instead of a period
2023-05-14 20:16:03 <scara> uniqNuuid: I'm just trying to figure a way to copy the kernel hooks I have on arch, to alpine, I want to beable to run an older kernel then the one installed.
2023-05-14 20:16:14 <scara> Like I want the /lib/moduals
2023-05-14 20:16:35 <scara> /lib/modules
2023-05-14 20:16:51 <minimal> scara: installing a new version of the kernel package automatically removes the old version
2023-05-14 20:22:10 <scara> minimal: Yes but I want to keep the old modules for as long as I'm running the old kernel.
2023-05-14 20:22:50 <minimal> right, I'm just pointing out that things (currently) are not designed to work that way
2023-05-14 20:23:54 <minimal> I thought ncopa or someone else recently talked about a way to potentially achieve that but can't find a reference to it
2023-05-14 20:25:41 <ziproot> When trying to run any window manager, I get errors of libraries not being found. Even dwm, a very basic manager, returns errors like "XftFontClose: symbol not found"
2023-05-14 20:27:32 <Saijin_Naib> @ziproot: did you run the setup-xorg script? 
2023-05-14 20:29:09 <ziproot> yes
2023-05-14 20:30:19 <abby> was that dwm compiled elsewhere
2023-05-14 20:30:47 <ikke> ziproot: not mixing repos (stable with edge)?
2023-05-14 20:31:11 <ziproot> no, only v3.18
2023-05-14 20:31:27 <ziproot> I tried both manually compiling dwm and installing it from apk and both had those types of errors
2023-05-14 20:36:00 <scara> minimal: What do you then do to fix it?
2023-05-14 20:39:11 <eex0> a firewall is a firewall regardless how or what you use to create it, pick your poison. tbh
2023-05-14 20:40:34 <ziproot> wait when I ran apk upgrade it said I was out of disk space
2023-05-14 20:40:54 <ziproot> but df says I used 3% spae
2023-05-14 20:42:01 <Saijin_Naib> Thanks, eex0
2023-05-14 20:43:16 <ikke> ziproot: and df -i?
2023-05-14 20:44:47 <ziproot> 1%
2023-05-14 20:46:56 <ikke> ziproot: how large is your disk?
2023-05-14 20:48:50 <ziproot> 47.4 GB
2023-05-14 20:50:34 <eex0> df -i reports Use% 0% for all fielsystems listed on my laptop.. interestingly
2023-05-14 20:53:26 <ncopa> minimal, scara i think there is an open issue. Maybe under apk-tools project 
2023-05-14 20:54:37 <ziproot> it says "failed to create boot/System-map-lts: no space left on device when trying to upgrade linux-lts
2023-05-14 20:55:04 <ziproot> oh wait I see, that's part of partition 1 which is 99% full
2023-05-14 20:55:16 <ncopa> It’s the boot part
2023-05-14 20:56:05 <eex0> yes, all 0 for me on /boot
2023-05-14 20:58:18 <ziproot> OK, I deleted the old version when upgrading and that gave me enough space
2023-05-14 21:02:05 <scara> ncopa: No but like really, what is the currently implemented solution? Reboot whenever you update your kernel? I run an IRC client on my VPS so I'm trying to keep downtime to a minimal, also weird quirks with if I wanted to make it possible to just automaticly start things back up, but thats not the point.
2023-05-14 21:02:25 <psykose> there is no implemented solution
2023-05-14 21:02:56 <ziproot> OK, I ran "for i in $(apk info); do apk fix $i; done" and then manually reinstalled gcc because that threw an error, and now dwm just says "cannot open display"
2023-05-14 21:03:03 <scara> psykose: So do you just tolerate that your software wont run?
2023-05-14 21:03:14 <psykose> ziproot: you can just `apk fix \*` for the same thing
2023-05-14 21:03:19 <psykose> scara: not sure what you mean
2023-05-14 21:03:33 <ziproot> psykose: oh, thank you
2023-05-14 21:03:41 <psykose> modules the kernel loads stay loaded, you just can't load new ones
2023-05-14 21:03:46 <scara> Some things just will not run without the modules installed.
2023-05-14 21:03:50 <psykose> like what
2023-05-14 21:04:09 <scara> In my case I need to setup a firewall so I need to modprobe ip_tables.
2023-05-14 21:04:19 <psykose> and then after you modprobe ip_tables, it stays loaded forever
2023-05-14 21:04:25 <psykose> and it can be deleted off disk and it won't matter
2023-05-14 21:04:41 <scara> But it hasn't been loaded in yet.
2023-05-14 21:04:47 <psykose> then load it when you boot the kernel
2023-05-14 21:04:50 <psykose> i'm sure you'll figure it out
2023-05-14 21:05:22 <scara> I guess right now, what I can do is manually download the old kernel and fill in the modules.
2023-05-14 21:07:41 <scara> Oh awesome, the mirrors ofcourse delete old packages, guess I'll just have to reboot, its not the end of the world.
2023-05-14 21:08:21 <eex0> I used to bitch before rtfm too.. once!
2023-05-14 21:08:30 <waterhead`> wow a chatter !
2023-05-14 21:08:32 <waterhead`> O M G
2023-05-14 21:09:04 <waterhead`> hi eex0
2023-05-14 21:09:09 <eex0> waves
2023-05-14 21:09:16 <waterhead`> OWOWW 597 chatters!
2023-05-14 21:09:28 <waterhead`> ok this is the biggest chann3el i been in 
2023-05-14 21:09:58 <eex0> stay tuned, it gets quiter..lol
2023-05-14 21:10:01 <waterhead`> eex0 are from england?
2023-05-14 21:10:02 <waterhead`> LOL 
2023-05-14 21:10:06 <eex0> quieter*
2023-05-14 21:10:15 <waterhead`> my first language was typo, i prefer it
2023-05-14 21:10:37 <waterhead`> I think almost everyone on oftc is english
2023-05-14 21:10:47 <waterhead`> bri'ish
2023-05-14 21:11:13 <eex0> waterhead: I'm in silicon valley and this particular channel is main, try #alpine-offtopic for chit-chat 
2023-05-14 21:11:29 <waterhead`> cool, i been to california
2023-05-14 21:11:43 <eex0> sunny a'f.. mhm
2023-05-14 21:11:47 <waterhead`> nice
2023-05-14 21:12:25 <waterhead`> i been to san diego, LA, hollywood hmm, languna beach , few other places, 3 different trips
2023-05-14 21:12:36 <waterhead`> fun in the sun
2023-05-14 21:13:04 <waterhead`> been to riverside too
2023-05-14 21:13:08 <waterhead`> just remembered
2023-05-14 21:13:30 <eex0> as I have visited england.. portsmith, plymouth
2023-05-14 21:13:46 <waterhead`> i've never been to england
2023-05-14 21:13:47 <waterhead`> would be nice 
2023-05-14 21:13:57 <waterhead`> i'm canadian
2023-05-14 21:14:14 <eex0> eh?
2023-05-14 21:14:17 <waterhead`> hahaha
2023-05-14 21:14:25 <waterhead`> yea
2023-05-14 21:15:03 <waterhead`> is there a limit of how many channels you can be on , on oftc?
2023-05-14 21:15:09 <waterhead`> wont let me join more than 3
2023-05-14 21:15:31 <waterhead`> ah ok , prolly invite only
2023-05-14 21:15:50 <eex0> no limit I know of
2023-05-14 21:16:06 <eex0> I'm on 8-10 myself
2023-05-14 21:16:11 <waterhead`> ok
2023-05-14 21:16:41 <waterhead`> i used to go t fishforums on oftc  over 10 years ago, its gone now, it was so busy , so many cool chatters
2023-05-14 21:16:51 <waterhead`> now all gone, no idea where the ar, but the channel still exists
2023-05-14 21:17:17 <eex0> games were born
2023-05-14 21:17:30 <waterhead`> i guess
2023-05-14 21:18:06 <eex0> irc @1994 I think
2023-05-14 21:18:15 <waterhead`> wow
2023-05-14 21:18:18 <waterhead`> been while
2023-05-14 21:18:27 <eex0> left for 18+yrs.. built games..lol
2023-05-14 21:19:06 <waterhead`> you mean , you did?
2023-05-14 21:20:04 <eex0> I still dable, make $ off of and troll secondlife
2023-05-14 21:20:15 <waterhead`> cool!
2023-05-14 21:20:24 <waterhead`> ever been to canada? 
2023-05-14 21:21:50 <eex0> I IM'd you waterhead
2023-05-14 21:22:14 <waterhead`> ok
2023-05-14 21:56:03 <eex0> smart bot!
2023-05-15 02:00:24 <codedroid> Switching to Alpine for server. Is there any spamass-milter present (SpamAssassin milter for Postfix) at all or provided anywhere?
2023-05-15 02:20:51 <rob[m]> <waterhead`> "i'm canadian" <- same here :D, well, a Mexican Canadian!
2023-05-15 02:26:12 <rob[m]> hmmm, anyone else on ppc64le? so far it's been great, but there are some minor issues... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/RFJymZslSmHGFuHtWaxqWnpe>)
2023-05-15 02:28:38 <rob[m]> context:
2023-05-15 02:28:39 <rob[m]> so, I'm vnc'ing into it temporally to do some homework on libreoffce that I have to do
2023-05-15 02:28:39 <rob[m]> normally I wouldn't be using vnc, but I'm having no choice right due to the shear amount of noise that my POWER server fan do (2U) 
2023-05-15 02:28:51 <rob[m]> picture for eye candy :D
2023-05-15 02:29:09 <rob[m]> 
ACTION uploaded an image: (161KiB) < https://matrix.org/_matrix/media/v3/download/matrix.org/batqpaBllAUiBkMjHQpGIebD/670_.jpg >

2023-05-15 02:30:11 <rob[m]> * context:
2023-05-15 02:30:12 <rob[m]> so, I'm vnc'ing into it temporally to do some homework on libreoffce that I have to do
2023-05-15 02:30:12 <rob[m]> normally I wouldn't be using vnc, but I'm having no choice right due to the shear amount of noise that my POWER server fans do (2U fans are loud as hell and they don't shut up) 
2023-05-15 06:15:18 <bootloader> so I know what I said before, but I feel like I can take my knowledge from arch and apply it
2023-05-15 06:16:00 <bootloader> does anyone know the packages that alpine uses to build itself `setup-alpine`
2023-05-15 06:16:18 <pj> alpine-conf
2023-05-15 07:27:17 <paul7439> I have just downloaded the VM edition and installed it in a virtual machine. I ran setup-alpine and after configuring the NTP client I received this error: "wget: can't connect to remote host: Operation timed out". My network is configured correctly and is working. I have seen a bug report about wget not working inside busybox, but I have not found a workaround. I'm also a bit surprised that this doesn't work given that it is the install script of a f
2023-05-15 07:27:17 <paul7439> resh install. Does anyone have any ideas?
2023-05-15 07:35:03 <psykose> are you sure that it's actually working if you ping a host and use wget manually
2023-05-15 07:42:43 <paul7439> Hi psykose! Yes, I see what might be the issue now; IPv4 works fine but ping -6 does not.
2023-05-15 07:53:52 <voltagex> how do I find out when/why a package was removed from Alpine?
2023-05-15 07:56:59 <psykose> usually git log
2023-05-15 07:59:17 <voltagex> thanks, seems like the package just got renamed (fts-dev)
2023-05-15 08:00:37 <psykose> ah, that was my mistake
2023-05-15 08:00:41 <psykose> should've added a fts-dev compat too
2023-05-15 08:01:07 <voltagex> no worries, was just trying to build dracut but I see it's packaged anyway
2023-05-15 08:02:38 <voltagex> aw, api.alpinelinux.org is gone.
2023-05-15 08:05:13 <voltagex> is the source code for pkgs.alpinelinux.org anywhere?
2023-05-15 08:05:17 <trim> im attempting to add packages to an index using `apk index`. but everytime I run it, it doesn't add packages
2023-05-15 08:05:54 <voltagex> what does apk index --verbose say?
2023-05-15 08:06:53 <trim> it say says "Index has 0 packages (of which 10 are new)"
2023-05-15 08:07:25 <trim> then when I run it with another package set, it says says "Index has 0 packages (of which 15 are new)"
2023-05-15 08:08:14 <psykose> voltagex: https://gitlab.alpinelinux.org/alpine/infra/aports-turbo/
2023-05-15 08:08:33 <voltagex> thanks psykose 
2023-05-15 08:08:33 <psykose> trim: what command do you run
2023-05-15 08:08:43 <psykose> verbatim
2023-05-15 08:09:42 <psykose> but i think it works
2023-05-15 08:09:48 <psykose> this looks like more of a visual bug
2023-05-15 08:09:50 <psykose> and a new one :p
2023-05-15 08:10:13 <psykose> it just always says 0, and new is the amount added (so the list of ./*.apk, or the diff with -x reused)
2023-05-15 08:10:41 <trim> `apk index -vU --allow-untrusted -o packages/APKINDEX.tar.gz packages/*.apk`
2023-05-15 08:11:40 <voltagex> seems like gitlab.alpinelinux.org is struggling a bit.
2023-05-15 08:12:29 <psykose> yeah, that works fine
2023-05-15 08:12:32 <psykose> it just says 0 anyway
2023-05-15 08:12:35 <trim> when I try to use it as a repo though the error "WARNING: opening /opt/packages: No such file or directory"
2023-05-15 08:13:00 <trim> is this because of something different?
2023-05-15 08:13:01 <psykose> it has to be sub-arch
2023-05-15 08:13:17 <psykose> packages/x86_64/*.apk, etc
2023-05-15 08:13:33 <trim> ohhhhhhh
2023-05-15 08:14:00 <trim> so I have to use that in my index call right?
2023-05-15 08:14:02 <psykose> it fetches the repo line at $repo/$(apk --print-arch)/APKINDEX.tar.gz
2023-05-15 08:14:09 <psykose> yeah just make a subfolder
2023-05-15 08:14:12 <psykose> put things in it
2023-05-15 08:14:16 <psykose> same thing but one more level
2023-05-15 08:14:52 <trim> can I put my index-file anywhere, or does it have to be in the same folder?
2023-05-15 08:15:14 <psykose> same
2023-05-15 08:15:29 <trim> okay thanks, ill try that out
2023-05-15 08:16:32 <voltagex> I wonder if the Alpine project would consider publishing the sqlite database behind pkgs.alpinelinux.org - it means tools like apk-file wouldn't have to parse HTML to find what package provides a certain file.
2023-05-15 08:20:05 <psykose> there's multiple and they're like over 10gb
2023-05-15 08:20:13 <voltagex> ...oh
2023-05-15 08:20:23 <voltagex> I wonder how apt-file does it for Debian then.
2023-05-15 08:21:30 <psykose> apt-file generates a db for you locally, and i guess the paths are fetched in some way that doesn't involve downloading every single .dev
2023-05-15 08:21:32 <psykose> .deb*
2023-05-15 08:21:53 <trim> pekoes it worked! you're a wizard
2023-05-15 08:22:10 <trim> *psykose
2023-05-15 08:22:39 <psykose> :)
2023-05-15 08:52:10 <voltagex> local alpine mirror go brrrrrrrrr
2023-05-15 08:53:01 <codedroid> Alpine has taught me spamass-milter is a rookies tool for rookie distros. Thank you Alpine.
2023-05-15 08:54:42 <psykose> if you mean spamassassin that exists
2023-05-15 08:55:11 <codedroid> Other distros build in a milter for spamassassin to use with stuff like postfix and I'm seeing now it looks unnecessary.
2023-05-15 08:56:00 <codedroid> I'm migrating configs from AlmaLinux to Alpine right now because Alma messed me up and lost some packages in its repos for no reason.
2023-05-15 08:56:26 <codedroid> Nobody wants to compile everything for their server.
2023-05-15 09:08:34 <WhyNotHugo> The package for official repos are not build on the CI pipeline for aports, are they?
2023-05-15 09:09:11 <psykose> no
2023-05-15 09:09:53 <WhyNotHugo> Where do they get built?
2023-05-15 09:10:26 <psykose> separate builders
2023-05-15 09:11:13 <psykose> on master changes a trigger gets called and the builders git pull and do stuff etc
2023-05-15 09:12:05 <WhyNotHugo> https://gitlab.alpinelinux.org/alpine/infra/docker/aports-build ?
2023-05-15 09:12:20 <psykose> there's no repo for it
2023-05-15 09:12:42 <psykose> aports-build is some ci stuff
2023-05-15 10:55:19 <wyk72> Hello ppl
2023-05-15 10:56:23 <wyk72> need a little help with LXC on Alpine 3.18: I am using a privileged LXC container with centos7 inside and I need  to use an OpenVPN tunnel inside it 
2023-05-15 10:56:54 <wyk72> but I always get the message "ERROR: Cannot open TUN/TAP dev /dev/net/tun: Operation not permitted (errno=1)" when I start the tunnel into the container (centos7 side)
2023-05-15 10:57:21 <wyk72> I tried many workarounds as stated on other forums
2023-05-15 10:57:43 <wyk72> but I still get this message and I'm puzzled on why it happens
2023-05-15 11:15:55 <wyk72> I solved it by adding the command "lxc-device add -n <containername> /dev/net/tun /dev/net/tun" in local.d script, but seems like a workaround
2023-05-15 11:17:45 <psykose> i mean it's more or less stuff like https://www.cloudhosting.lv/eng/faq/How-to-enable-tun-tap-inside-LXC https://discuss.linuxcontainers.org/t/no-dev-net-tun-inside-lxc-container/8344/12 etc
2023-05-15 11:18:22 <wyk72> both did not work at all
2023-05-15 11:18:38 <wyk72> "lxc-device add -n <containername> /dev/net/tun /dev/net/tun"  worked oob
2023-05-15 11:22:45 <wyk72> cgroups is  beautiful pain. Beautiful in some sense, but a pain.  And also centos7 is ancient and wants cgroups1 not cgroups2 because of that horror that is systemd. I should make myself happy with that workaround and that's it. I guess :)
2023-05-15 11:23:29 <psykose> systemd works with cgroups 2 by default
2023-05-15 11:23:39 <psykose> the reason starts and ends at 'centos7 is ancient'
2023-05-15 11:28:35 <wyk72> psykose: I 100% agree, but sometimes you have to suck it up and use a container for old software, horrible as it sounds
2023-05-15 11:29:13 <wyk72> or a VM
2023-05-15 11:29:57 <wyk72> but this VMWARE VPS is so small that a container fits better, and I  do not have (even) access to KVM
2023-05-15 11:30:08 <wyk72> anyway problem solved
2023-05-15 11:33:57 <psykose> :)
2023-05-15 13:57:43 <telmich_> hello...since today's update xorg does not start as a user anymore
2023-05-15 13:58:11 <telmich_> I saw an AIGLX error, not sure if that one is related
2023-05-15 13:59:28 <telmich_> ah, no, root's xorg log has the same entry
2023-05-15 14:02:28 <psykose> let me guess, you have xf86-video-intel installed
2023-05-15 14:08:08 <Nulo[m]> hi! when and why was initdb moved into /usr/libexec/postgresql15/initdb?
2023-05-15 14:08:13 <Nulo[m]> it was in /usr/bin before
2023-05-15 14:09:06 <psykose> it is in usr/bin, the pg_versions thing lets you pick the default
2023-05-15 14:09:26 <psykose> pg_versions set-default 15 and it should be in usr/bin
2023-05-15 14:10:07 <psykose> was changed around so they're all in their own dir and you pick the one you use
2023-05-15 14:10:39 <Nulo[m]> ah, but it doesn't get automatically set when installing postgres, no?
2023-05-15 14:11:40 <psykose> it does
2023-05-15 14:12:06 <psykose> if you just install postgresql15 on clean the trigger sets it as default as it's the only one that exists
2023-05-15 14:12:30 <psykose> https://img.ayaya.dev/vu0cGPL3A0GS
2023-05-15 14:12:43 <Nulo[m]> ah yeah. i'm installing as non-root (inside a container) so pg_versions fails
2023-05-15 14:40:50 <eex0> my hacker uses IxVeriWave .. disregard if you're NOT "my hacker" return to your regularly scheduled chan
2023-05-15 15:06:09 <eex0> well, it could be that wireshark took a dump and this bug is back .. from eons ago: In Wireshark 2.4.0 to 2.4.3 and 2.2.0 to 2.2.11, the IxVeriWave file parser could crash. This was addressed in wiretap/vwr.c by correcting the signature timestamp bounds checks.
2023-05-15 15:06:09 <eex0> Published: January 11, 2018
2023-05-15 15:06:46 <eex0> either way, I'll pitbull this thing
2023-05-15 15:10:47 <psykose> you good?
2023-05-15 15:11:10 <eex0> 3mg of binary data was sent from OFTC, encapsulated in tls 1.2 I might need some help, anyone know a cryptographer? 
2023-05-15 15:12:04 <eex0> not yet psykose, I will be though.. thanks
2023-05-15 15:12:22 <psykose> you sound like you took a questionable amount of psychoactive substances
2023-05-15 15:13:39 <eex0> I'm clear as a bell.. woke up in a good mood about 45 minutes ago to find wireshark left packet presents for me.
2023-05-15 15:13:57 <psykose> :3
2023-05-15 15:14:31 <eex0> don't think I'll be leaving irc on all night again tho, good thing I left wireshark on too
2023-05-15 16:17:59 <eex0> psykose tracking my ixveriwave incident is pointing to a possible OpenSSL Vulnerability 
2023-05-15 16:18:15 <eex0> working on it
2023-05-15 16:58:31 <wolf> good afternoon
2023-05-15 16:59:09 <wolf> Does not have the day package
2023-05-15 17:01:20 <psykose> the what
2023-05-15 17:14:50 <lopid> you know, the day package
2023-05-15 17:15:23 <lopid> something doesn't have it
2023-05-15 17:17:48 <psykose> permanent evening
2023-05-15 17:23:45 <wolf> I don't get it
2023-05-15 17:24:28 <quinq> Because it's missing
2023-05-15 17:27:58 <minimal> wolf: no one here knows which "day" package you are actually referring to
2023-05-15 17:38:58 <wolf> Thak's minimal
2023-05-15 17:39:43 <wolf> I refer to this: https://pastebin.com/AT7NpSDS
2023-05-15 17:41:14 <minimal> so "dia", not "day", wow someone could have easily figured out what you meant lol
2023-05-15 17:42:21 <minimal> so are you asking if Alpine has a package for "dia" or stating that Alpine does not have a package for "dia"? (in which case why are you stating it? are you expecting someone to suddenly package it for you?)
2023-05-15 17:45:02 <quinq> Although dia stands for dia-grams, not día
2023-05-15 17:47:56 <psykose> it used to be packaged
2023-05-15 17:51:34 <Ermine> package of the day
2023-05-15 17:52:07 <quinq> PotD
2023-05-15 18:12:30 <livy> Hello. I've done a fresh installation of Alpine 3.18, and followed the tutorial to install KDE Plasma.
2023-05-15 18:12:44 <livy> https://wiki.alpinelinux.org/wiki/KDE
2023-05-15 18:13:11 <minimal> livy: good for you. Did you perhaps have a question to ask?
2023-05-15 18:13:30 <livy> After the display manager 'sddm' started, it did not respond to mouse and keyboard.
2023-05-15 18:13:47 <livy> I could neither type my password, nor clicked the Shutdown/Restart button.
2023-05-15 18:14:29 <psykose> minimal: don't be rude
2023-05-15 18:14:41 <livy> Right now I am testing Alpine 3.18 inside VMware Workstation, but I do not think it matters.
2023-05-15 18:14:54 <livy> Command line interface works just fine.
2023-05-15 18:15:53 <livy> The only thing I can do now is moving the cursor.
2023-05-15 18:16:16 <quinq> Maybe you're not in the sufficient group (like input?)
2023-05-15 18:16:52 <livy> I'll try using root account and see. Right now I am using a normal account.
2023-05-15 18:17:06 <quinq> no no
2023-05-15 18:17:09 <quinq> Don't try that
2023-05-15 18:17:24 <livy> Yeah it doesn't work either. :(
2023-05-15 18:18:08 <psykose> the guide is missing the part where you have to rc-service svc start the services, not just add them
2023-05-15 18:18:20 <psykose> i guess you can add yourself to video/input groups anyway and logout and back int
2023-05-15 18:18:43 <livy> I know. I used 'rc-service <service_name> start'. I even hard reset the VMware VM.
2023-05-15 18:18:50 <livy> So the services are surely started.
2023-05-15 18:19:18 <livy> How to add my user to these groups? What are the group names?
2023-05-15 18:19:33 <quinq> In /etc/group
2023-05-15 18:19:46 <quinq> With: adduger $USER $group
2023-05-15 18:19:55 <quinq> https://wiki.alpinelinux.org/wiki/Setting_up_a_new_user
2023-05-15 18:21:20 <livy> grep video /etc/group
2023-05-15 18:21:29 <livy> video:x:27:root,livy,sddm
2023-05-15 18:21:43 <livy> Looks like my user is in video group by default.
2023-05-15 18:21:48 <quinq> Yeah but I talked about input
2023-05-15 18:21:54 <livy> I'll try to add it to input group now.
2023-05-15 18:21:55 <quinq> Video is for… video
2023-05-15 18:22:32 <quinq> Also you don't have to grep, you can also use "groups", which will tell you about the effective groups
2023-05-15 18:24:13 <livy> I've added my user to input group, then do a reboot.
2023-05-15 18:24:20 <livy> Still it doesn't work.
2023-05-15 18:24:47 <livy> I am stuck at the login screen and cannot type or click anything.
2023-05-15 18:26:09 <quinq> You said VMware, maybe you need some specific Xorg driver for those input (just a guess, I don't know how vmware works)
2023-05-15 18:27:01 <livy> It worked just fine before I started SDDM service.
2023-05-15 18:27:16 <livy> Maybe I'll test it tomorrow on QEMU, or on a physical machine.
2023-05-15 18:27:28 <livy> I am on my gaming laptop, and only Windoze. :((
2023-05-15 18:27:59 <livy> In the meantime, I load the snapshot with the vanilla Alpine installation and try to install Xfce and see.
2023-05-15 18:48:06 <livy> Xfce and lightdm combo works flawlessly.
2023-05-15 18:49:07 <livy> Tomorrow I'll try KDE again on QEMU pc-q35 when I go to work.
2023-05-15 21:04:30 <raspbeguy> hello
2023-05-15 21:04:56 <raspbeguy> I don't know how to change keymap on GDM
2023-05-15 21:09:34 <psykose> iirc you cannot
2023-05-15 21:09:50 <psykose> only after
2023-05-15 21:09:57 <psykose> https://gitlab.alpinelinux.org/alpine/aports/-/issues/14544#note_285138
2023-05-15 21:21:47 <raspbeguy> I'll check if there is an upstream issue
2023-05-15 21:22:14 <raspbeguy> This seems related to lack of systemd 
2023-05-15 21:22:21 <psykose> pretty much
2023-05-15 22:24:18 <eex0> lol .. pong!
2023-05-16 01:17:40 <eex0> tx?
2023-05-16 02:52:52 <iggy> anybody know why I'd be getting an error like: package mentioned in index not found (try 'apk update')
2023-05-16 02:54:41 <eex0> sync may be off in your repo?
2023-05-16 02:55:07 <iggy> I checked my package repo and the apk exists and it has an entry in the APKINDEX file (as seen by extracting the APKINDEX.tar.gz file and looking through it)
2023-05-16 02:56:56 <eex0> thats a new one for me
2023-05-16 02:58:00 <iggy> yeah, it's weird too because it's just the openrc package that fails (the main package installs fine)
2023-05-16 03:00:17 <uniqNuuid> iggy: apk fix
2023-05-16 03:59:54 <iggy> that gives the same error
2023-05-16 07:30:08 <iggy> apparently abuild is setting the arch to "noarch" for the openrc package, but the package is actually ending up in the x86_64 dir
2023-05-16 07:35:25 <iggy> I see the main repos the openrc packages all have arch=x86_64, but all of the openrc packages in my repo have arch=noarch... but this is the only package having issues even though all the other openrc packages are also noarch
2023-05-16 07:38:06 <iggy> scratch that, my openrc packages are all behind the main repos, so testing installing openrc packages from my repo didn't work
2023-05-16 07:44:46 <ikke> There is no dedicated dir for noarch
2023-05-16 07:47:18 <iggy> currently trying to figure out how to force an arch for a(n openrc) subpackage
2023-05-16 07:49:33 <ikke> Why?
2023-05-16 07:51:04 <iggy> because abuild is setting the arch to noarch which is then making apk look for something along the lines of /alpine/edge/main/noarch/caddy-ext-openrc-2.6.4-r1.apk because the APKINDEX says noarch for the openrc packages
2023-05-16 08:00:55 <iggy> interesting... the APKINDEX that is generated locally (as part of `abuild -r`) has arch=x86_64, it's only when I upload them to my package repo (which then runs `apk index -o APKINDEX.tar.gz ....apk`) that the APKINDEX ends up with arch=noarch
2023-05-16 08:05:58 <iggy> apparently, I need to --rewrite-arch all of them
2023-05-16 08:06:12 <iggy> thanks ikke and rubber ducky
2023-05-16 08:08:55 <ikke> quack
2023-05-16 08:27:22 <dwfreed> quack quack
2023-05-16 10:04:08 <indy> hi all, where is configuration place for extlinux.conf to prevent overwrite of console after every kernel update?
2023-05-16 10:05:48 <lopid> overwrite of console?
2023-05-16 10:06:07 <lopid> boot stuff is in /boot
2023-05-16 10:41:04 <indy> i know, but where i can set 'default' settings for APPENDm which is always overwritten by apk once kernel is updated
2023-05-16 10:41:33 <Cyanat> Hello, I try to create a script to install an Alpine Linux completely offline. So, on an online server, I have downloaded all needed apk using 'apk fetch', then copied these apk on my offline server, but when I try , for example 'apk add  --allow-untrusted ./curl-8.0.1-r0.apk', it fails because alpine try to get APKINDEX.tar.gz, and also lots of dependencies (readline, ca-certificatesn nghttp2-libs...). Is there a
2023-05-16 10:49:54 <pj> Cyanat: apk fetch -R
2023-05-16 10:51:24 <satmd> I'd believe that running an offline caching webserver with the packages might be a cleaner approach if you ever want to scale out
2023-05-16 10:56:04 <Cyanat> @pj that seems gret, thanks
2023-05-16 10:56:37 <Cyanat> @satmd, yes, it seems a good idea, I will probably try that in a second tim
2023-05-16 11:23:31 <indy> lopid, found it - /etc/update-extlinux.conf
2023-05-16 11:27:56 <lopid> oh, sorry
2023-05-16 12:17:13 <boombim> is screensharign broken again on edge?
2023-05-16 12:17:21 <boombim> doesn't work for me in firefox...
2023-05-16 12:33:14 <Cyanat> Sorry, me again with my offline instalaltion package. So, on my offline server, I have curl apk with all its dependencies. But installing curl continue to raise an error because dependencies are not installed, and because the server can't get them on internet. So I have to manually apk add libcurl-8.0.1-r0.apk, then nghttp2-libs-1.47.0-r0.apk, then... then curl. Is there a way to simply let alpine check a local fo
2023-05-16 13:07:02 <ikke> Your message was truncated 
2023-05-16 13:07:17 <nu> for having offline packages: mirroring repos with e.g rclone/sync works well. v3.18 main is just ~4GiB for x86_64. you can place it in a directory, and point /etc/apk/repositories to it. it will behave the same way as an online one
2023-05-16 13:08:15 <psykose> Cyanat: you have to actually put the apks in a folder that is in /etc/apk/repositories and not just in the same folder implicitly made for an existing http repo (which will never work without networking)
2023-05-16 13:28:22 <Cyanat> ok, it make sense. So I have stored all my apk in /test and chmod 777 this folder. In /etc/apk/repositories I've add an line "/test", but I don't know why, I get WARNING: Ignoring /test: No such file or directory
2023-05-16 13:29:16 <psykose> you have to put them in /test/$(apk --print-arch)/ and create an index too
2023-05-16 13:29:24 <psykose> and then also sign the index, and install the key you signed it with
2023-05-16 13:30:54 <Cyanat> Ok, not so easy as I thought... I think I will need to do some research on how creating our own repo :)
2023-05-16 13:55:29 <minimal> Cynat: as nu mentioned just rsync the relevant Alpine repo(s) to a local directory/server and use that
2023-05-16 14:03:36 <pj> minimal: you completely miss the point on what they are trying to do
2023-05-16 14:03:38 <pj> same as nu
2023-05-16 14:04:43 <pj> anyway, Cyanat: `apk index -o APKINDEX.tar.gz *.apk` in directory with .apk's
2023-05-16 14:15:58 <Cyanat> Ok, so to mirror the entire alpine 3.16 repository, it's something like that right?
2023-05-16 14:16:00 <Cyanat>  rsync --archive  --update --hard-links --delete --delete-after --delay-updates --timeout=600 rsync://rsync.alpinelinux.org/alpine/v3.16/ ./
2023-05-16 14:31:47 <minimal> pj: nope I didn't miss the point, I merely pointed out the easiest solution
2023-05-16 14:32:29 <minimal> Cyanat: That seems to look ok.
2023-05-16 14:33:09 <pj> yes, you are missing the point because suggesting completely different thing is not a solution
2023-05-16 14:34:08 <scara> Is this a bug? https://lainsafe.kalli.st/files/168424760873689.txt
2023-05-16 14:35:16 <psykose> probably
2023-05-16 14:35:47 <scara> Where is the most correct place to report it?
2023-05-16 15:10:40 <ikke> https://gitlab.alpinelinux.org/alpine/awall
2023-05-16 15:13:12 <jvoisin>  /usr/bin/hugo, 60M.
2023-05-16 15:13:16 <jvoisin> computers were a mistake.
2023-05-16 15:16:44 <leah2> the magic of static linking :p
2023-05-16 15:16:59 <psykose> that's quite average
2023-05-16 15:17:03 <psykose> half the cloud stuff is 100+
2023-05-16 15:18:23 <jvoisin> yeah, binaries at work are insane…
2023-05-16 15:18:29 <jvoisin> like a couple of GB
2023-05-16 15:28:50 <psykose> a couple gb for one executable? what?
2023-05-16 15:29:12 <psykose> i know google does link-the-world as a build strategy but the linker stills throw it all away and they do a bunch of other stuff
2023-05-16 15:29:18 <psykose> an actual multi-gb final binary seems fake
2023-05-16 15:29:34 <psykose> maybe the build folder, or all the debug data? :D
2023-05-16 15:34:52 <jvoisin> -rwxr-xr-x 1 root root 1076332672 Jan  1  2000 /usr/bin/XXX
2023-05-16 15:35:19 <jvoisin> this is on my workstation
2023-05-16 15:35:26 <jvoisin> prod binaries are bigger
2023-05-16 15:35:57 <psykose> yikes!
2023-05-16 17:16:33 <quinq> jvoisin, wth
2023-05-16 17:57:35 <frag_> alpine represented https://pr0gramm.com/top/5688840
2023-05-16 19:00:46 <handlerug> is there any documentation on apk the package manager's failure modes? like can I trust it to not corrupt the package database if it crashes, can it rollback transactions on recoverable and irrecoverable errors, etc
2023-05-16 19:01:11 <psykose> i don't think there are any docs no
2023-05-16 19:01:29 <handlerug> then I hope the answers are all yes :P
2023-05-16 19:01:30 <psykose> and if by crashes you mean 'it segfaults' then i don't think you can trust that for more or less any program, haha
2023-05-16 19:01:37 <psykose> but just general errors? it's safe
2023-05-16 19:01:46 <handlerug> well like atomic writes
2023-05-16 19:01:47 <psykose> it can leave an error state probably but not corrupted
2023-05-16 19:01:47 <Lord> there is no package database, no ?
2023-05-16 19:01:51 <psykose> there is
2023-05-16 19:01:58 <handlerug> thanks psykose
2023-05-16 19:02:04 <Lord> the world file ?
2023-05-16 19:02:04 <psykose> /lib/apk/db/
2023-05-16 19:02:31 <psykose> handlerug: overall though i'm not super sure
2023-05-16 19:02:42 <psykose> you could open an issue in apk-tools for clarification on what has already been thought of
2023-05-16 19:02:45 <psykose> it would be a better answer
2023-05-16 19:02:49 <handlerug> yeah I was asking particularly because you don't encounter it often
2023-05-16 19:02:54 <psykose> indeed
2023-05-16 20:21:50 <Armote[m]> alpine's chromium is glitched (amdgpu) after update. something has recently been afflicting chromium based browsers that causes glitches, but google-chrome update some 1-2 weeks ago fixed that. not sure if chromium is lagging behind on that. workaround is to use software renderer.
2023-05-16 20:22:21 <quinq> “glitched” meaning what?
2023-05-16 20:23:11 <psykose> i noticed the same
2023-05-16 20:23:15 <Armote[m]> the rendering is screwed, may start normal then deteriorate to boxes not rendering properly all over the window. gets worse for js heavy frontends.
2023-05-16 20:23:15 <psykose> hopefully todays upgrade has the fix
2023-05-16 20:23:23 <psykose> i noticed it for pwas
2023-05-16 20:23:28 <psykose> i.e. instantly on start skia is just broken
2023-05-16 20:23:29 <quinq> I see, thanks Armote[m] 
2023-05-16 20:25:00 <Armote[m]> terminal displays gl complication errors. but doesnt occur with llvmpipe.
2023-05-16 20:25:55 <Armote[m]> compilation*
2023-05-16 20:26:14 <psykose> yeah
2023-05-16 20:26:30 <psykose> check again in uhh 6 hours when it builds the upgrade
2023-05-16 20:26:36 <psykose> if not, well, not really sure why it's broken :p
2023-05-16 20:27:21 <psykose> hmm, let me read the quick log
2023-05-16 20:27:30 <psykose> unless you know which commit fixed it
2023-05-16 20:27:35 <Armote[m]> well the workaround for me is to disable access to /dev/dri, not sure what the normal way to trigger software rendering is. --disable-gpu perhaps?
2023-05-16 20:27:44 <psykose> ye
2023-05-16 20:27:54 <psykose> some flag like that
2023-05-16 20:29:23 <psykose> bunch of amd whatever rolls in the log
2023-05-16 20:29:25 <psykose> maybe it is
2023-05-16 20:51:25 <psykose> surprised anyone even uses the chromium :D
2023-05-16 20:52:01 <quinq> Only most of the world :/
2023-05-16 20:52:46 <quinq> And most likely in your pocket-computer
2023-05-16 23:47:49 <felixphew> anyone have any idea why the kernel would spend over a minute "Spinning up disk..." during boot? (the two disks in question are SSDs, and once this times out, they come up perfectly)
2023-05-16 23:48:14 <felixphew> this happens in the initrd, seemingly while nlplug-findfs is running
2023-05-16 23:50:30 <felixphew> I don't have all of alpine-base installed right now on this box; am I maybe missing maybe some hardware hints files or something that would normally be copied into the initramfs and would let it know to not do this?
2023-05-16 23:52:28 <psykose> easy way to find out is to add all of alpine-base and see if it fixes it, but i doubt it
2023-05-17 02:30:17 <felixphew> so on the assumption that doesn't work (I will try it, but can't until I get home), any other ideas?
2023-05-17 02:32:06 <psykose> idk
2023-05-17 02:32:23 <psykose> that is a kernel message indeed so it sounds like some device issue
2023-05-17 02:33:09 <psykose> when i had that it was a legit bios/drive/whatever issue
2023-05-17 02:33:41 <psykose> i think i had a faulty usb thing and when anything was plugged in it would hang on something for a bunch of seconds :D
2023-05-17 02:34:24 <psykose> it doesn't sound related to anything in the userspace boot process, unless the same kernel with a different initrd (from another distro etc) doesn't have the issue
2023-05-17 03:55:48 <psykose> Armote[m]: yep, it's fixed
2023-05-17 04:18:43 <felixphew> psykose: yep, it's a kernel message, but it didn't happen when booting the Alpine installer (not an identical kernel, but pretty dang close)?
2023-05-17 10:37:20 <gopher22> Hello, foes thelounge package vanished from alpine linux repos?
2023-05-17 10:38:52 <lopid> https://pkgs.alpinelinux.org/packages?name=thelounge&branch=v3.18&repo=&arch=x86_64&maintainer=
2023-05-17 10:39:55 <gopher22> yup, looks like is not available any more
2023-05-17 10:50:41 <pj> it was removed
2023-05-17 11:07:12 <gopher22> darn :( @pj perhaps you know if there is an archive page of https://pkgs.alpinelinux.org where at least I can see who is/was the maintainer of the package in question?
2023-05-17 11:07:51 <lopid> just change the branch on that page
2023-05-17 11:13:20 <donoban> gopher22: it seems removed long time ago Date:   Tue Nov 1 20:11:56 2022 +0000
2023-05-17 11:13:26 <donoban> "    package contains binaries prelinked against glibc"
2023-05-17 11:13:54 <donoban> Maintainer: Kay Thomas <kaythomas@pm.me>
2023-05-17 13:28:07 <gopher22> great stuff, thanks. I can see the last version that still had it was 3.16
2023-05-17 13:28:36 <gopher22> luckly I have a backup container with this image with thelounge preinstalled. this one :)
2023-05-17 13:31:16 <psykose> it's also in testing
2023-05-17 13:31:22 <psykose> i managed to make it 66% smaller
2023-05-17 13:31:38 <psykose> well
2023-05-17 13:31:40 <psykose> 'will be'
2023-05-17 13:33:05 <gopher22> it's in the testing repo? :o how did I miss it? :)
2023-05-17 13:33:12 <gopher22> brb, installing it now :)
2023-05-17 13:43:15 <psykose> when it builds in whenevver
2023-05-17 13:45:08 <gopher22> psykose looks like it is only available in the testing repo for ppc64le (?) architecture. is this intended?
2023-05-17 13:47:21 <psykose> should i say when it builds a few more times :p
2023-05-17 13:48:21 <gopher22> gotcha ;)
2023-05-17 13:55:35 <ikke> psykose: yes please 
2023-05-17 13:58:03 <psykose> ikke: the build knows where it is because it knows where it isn't
2023-05-17 13:58:54 <ikke> And therefore 
2023-05-17 14:05:39 <pj> it appears in places that is supposed to appear in
2023-05-17 14:37:17 <lemontree> hello
2023-05-17 14:39:59 <psykose> do oranges grow on the lemon tree
2023-05-17 14:43:27 <lemontree> sadly not, oranges are awesome
2023-05-17 14:44:37 <lemontree> but my question: i installed alpine linux on a thinkpad T400 which has libreboot installed. not the newest version of libreboot, but don't know the version currently. however, installation of e.g. fedora worked fine.
2023-05-17 14:46:05 <lemontree> the installation of alpine linux was successful but after rebooting, it doesn't boot. after starting the boot, it seems a grub menu shows up which i guess is the fault here, since i think alpine linux doesn't use grub by default (and syslinux instead?)
2023-05-17 14:47:07 <lemontree> i don't know 100% its grub but i think it is, at least i could press C to get to the commandline from grub and it looked similar to what grub does
2023-05-17 14:47:14 <lemontree> don't know that much about syslinux
2023-05-17 14:50:24 <minimal> lemontree: Alpine can use Grub as well as Syslinux
2023-05-17 14:50:34 <minimal> what errors did you see?
2023-05-17 14:50:49 <ikke> With UEFI, it even defaults to grub
2023-05-17 14:51:00 <lemontree> i don't see any errors and the everything before the reboot was successful
2023-05-17 14:51:32 <ikke> What do you see them?
2023-05-17 14:51:40 <ikke> Then*
2023-05-17 14:52:07 <lemontree> i just see a black screen after the grub thing is done. i know from e.g. when running fedora and when booting the alpine linux installer that booting this thing usually takes a long time because i think grub does things inefficiently
2023-05-17 14:52:23 <lemontree> but i waited like 5 to 10 minutes and it is still a black screen without errors
2023-05-17 14:52:52 <lemontree> if i remember correctly, this is seaBios variant, so no UEFI
2023-05-17 14:56:53 <minimal> lemontree: try removing "quiet" from the Grub entry's cmdline options? you should see some more info then during boot
2023-05-17 14:58:25 <minimal> lemontree: you're BIOS booting, not UEFI? then setup-alpine should have installed Syslinux rather than Grub. Are you sure you're using BIOS booting?
2023-05-17 14:58:46 <lemontree> seabios provides only bios booting
2023-05-17 14:59:21 <lemontree> but i wonder if seabios / or libereboot just assumes grub and therefore alpine assuming syslinux doesn't work
2023-05-17 15:00:49 <minimal> lemontree: well if BIOS is to be used for booting then Alpine's setup-alpine will install a boot sector on the disk/SSD (that of syslinux typically)
2023-05-17 15:01:25 <minimal> if UEFI is used for booting then when setup-alpine partitions the disk it will create an ESP partition and put a grub EFI in there for booting
2023-05-17 15:01:25 <lemontree> regarding removing quiet option: that helped a bit, but not that much: now i could see i think 3 log statements (loading the kernel and some more i couldn't read) but after that, it is still blank screen
2023-05-17 15:03:03 <Saijin_Naib> Is it responding to input? Can you toggle caps and see the indicator light change? 
2023-05-17 15:03:37 <Saijin_Naib> Didn't since older boards have issues with a change to how the framebuffer is started on more recent kernels? 
2023-05-17 15:03:40 <lemontree> nope, is not responding, toggling caps doesn't change anything
2023-05-17 15:03:56 <Saijin_Naib> Oh okay, I'm headed the wrong way then, sorry 
2023-05-17 15:04:25 <lemontree> (ah but i am sorry, the capslock doesn't have a light. but it responds to brightness changes)
2023-05-17 15:04:56 <lemontree> with brightness i mean the screen
2023-05-17 15:06:47 <Saijin_Naib> Oooh, okay. So it is alive at least... Can you ping or ssh from another machine?  
2023-05-17 15:06:48 <lemontree> does the alpine installer use grub or syslinux?
2023-05-17 15:07:06 <Saijin_Naib> To load itself you mean? 
2023-05-17 15:07:36 <lemontree> yes, when i boot up the installer, the image i burned on my usb stick
2023-05-17 15:12:59 <lemontree> Saijin_Naib: can't ping it, nmap found no ip adress
2023-05-17 15:18:17 <minimal> lemontree: the Alpine installer is an ISO image and so it typically uses isolinux (whether its on an actual CD/DVD or on a USB stick)
2023-05-17 15:18:33 <Saijin_Naib> So it isn't getting very far when initialized... 
2023-05-17 15:18:53 <minimal> it could be a display DRM/KMS issue or it could be something else
2023-05-17 15:19:26 <minimal> lemontree: try adding the "nomodeset" option to the cmdline in the bootloader?
2023-05-17 15:22:17 <lemontree> nomodeset doesn't change something
2023-05-17 15:22:51 <lemontree> but still my impression is that the problem is with libreboot
2023-05-17 15:25:22 <minimal> I suspect you'll need to ask some libreboot people
2023-05-17 15:27:22 <lemontree> yep, asked in the libreboot channel already
2023-05-17 15:27:37 <lemontree> but well, thank you for your patience and help with this issue, everybody :)
2023-05-17 18:22:08 <scorpion2185[m]1> where do I report a bug for https://git.alpinelinux.org/aports/tree/community/amazfish?h=master?
2023-05-17 18:22:30 <ikke> https://gitlab.alpinelinux.org/alpine/aports/
2023-05-17 18:43:42 <Saijin_Naib> https://www.freedesktop.org/wiki/Software/systemd/TheCaseForTheUsrMerge/
2023-05-17 18:44:00 <Saijin_Naib> What thoughts do folks have on this? 
2023-05-17 18:44:24 <Saijin_Naib> I'm not educated enough on this to tell if this is blowing smoke or not 
2023-05-17 18:44:28 <ikke> Some have very strong opinions on it
2023-05-17 18:45:59 <minimal> I thought this was (briefly) discussed here (or alpine-devel) a couple of months ago
2023-05-17 18:46:56 <ikke> it regularly comes up
2023-05-17 18:48:05 <quinq> What I don't get is the hegemonial agenda of this kind of things
2023-05-17 18:48:42 <ikke> simplificatiaon
2023-05-17 18:52:57 <quinq> Yeah, there are other ways of “simplifying” things
2023-05-17 18:53:07 <quinq> If by simplifying you mean the lack of alternatives
2023-05-17 18:53:46 <Saijin_Naib> So it might be blowing a bit of smoke? Good to know 
2023-05-17 18:53:54 <quinq> :)
2023-05-17 18:54:47 <Saijin_Naib> <ikke> "Some have very strong opinions..." <- Yeah, it is under the systemd article group, so that's just assumed 🤣
2023-05-17 18:55:06 <Saijin_Naib> From what I've seen in packaging, it sounds reasonable, no? 
2023-05-17 18:55:07 <ikke> To me the current distinction is kind of arbitrary
2023-05-17 18:55:13 <ikke> Yeah
2023-05-17 18:55:28 <Saijin_Naib> Meaning the split how it is now? 
2023-05-17 18:55:35 <Saijin_Naib> Ah, okay 
2023-05-17 18:55:36 <ikke> I think most controverse is the end state, not necessarily the change
2023-05-17 18:55:46 <minimal> the separation made sense in the early days of small disks and separate mountpoints
2023-05-17 18:56:17 <quinq> Also in networking systems, with remote mounts
2023-05-17 18:56:21 <ikke> The entire reason /usr was used was because the OS disk was full
2023-05-17 18:56:33 <ikke> /usr used to be what's now /home
2023-05-17 18:56:36 <quinq> Now everything is mostly monolothic single-disk system
2023-05-17 18:56:50 <Saijin_Naib> My misperception was that stuff I install goes into usr, and stuff from the distro goes into the higher level 
2023-05-17 18:57:09 <quinq> Not really
2023-05-17 18:57:10 <ikke> Saijin_Naib: the standard is /usr/local for things not managed by the OS
2023-05-17 18:57:30 <minimal> or /opt in some other Unix-type OSes
2023-05-17 18:57:48 <quinq> Well, /opt is for things too ugly to be properly integrated
2023-05-17 18:58:18 <Saijin_Naib> Is there documentation on this? Haha 
2023-05-17 18:58:53 <ikke> https://en.wikipedia.org/wiki/Filesystem_Hierarchy_Standard
2023-05-17 18:58:59 <minimal> Saijin_Naib: yes, the scars carved into the faces of long-time Unix/Linux admins ;-)
2023-05-17 18:59:04 <quinq> :)
2023-05-17 18:59:05 <Saijin_Naib> Many thanks 
2023-05-17 18:59:10 <ikke> https://refspecs.linuxfoundation.org/FHS_3.0/fhs/index.html
2023-05-17 18:59:22 <quinq> Regarding /usr/local/, that was for actually local things to the machine
2023-05-17 18:59:52 <quinq> Where you could share a /usr over several machines (ro), and install local modules in /usr/local
2023-05-17 19:00:07 <quinq> Now of course, it got a different meaning in practice, as ikke said
2023-05-17 19:00:23 <Saijin_Naib> minimal: Ah, greybeard wisdom.
2023-05-17 19:00:25 <eex0> redhat started all that sepeartion b.s
2023-05-17 19:00:46 <ikke> people started doing all kinds of post-rationalization of the hierarchy :)
2023-05-17 19:03:39 <minimal> eex0: nope, even before FHS older Unix systems had /bin vs /usr/bin and /sbin vs /usr/sbin separation, the Wikipedia article on FHS even mentions "hier" layout from 1979...
2023-05-17 19:31:25 <eex0> noticed a bad habit mistake I was doing, allowing users of gnome to have doas or sudo.. yikes, fixed!
2023-05-17 19:32:05 <quinq> :)
2023-05-17 19:40:13 <eex0> was a really bad habit of mine too.. and more things I have uncovered because Im tired of being hacked..lol
2023-05-17 19:43:14 <eex0> I noted a child could have hacked me and I thought my opsec was pretty decent, well, hah! mistakes only make us stronger eh.
2023-05-17 19:43:39 <pj> no
2023-05-17 19:49:01 <eex0> going to get rid of grub so, with efibootmgr I can use the script on alpine docs for a cryptsys setup drive? I read that but not sure I fully understand it yet. also efibootmgr says I have a memory mapped file that my biso loads in and sets it to 0000* so it would always be first to load until I changed that with efibootmgr.. now from what I read on coreboot that this would be the manufacturers splash screen but to me, that just looked suspect 
2023-05-17 19:49:01 <eex0> a'f so, I just changed the boot order perminantly.
2023-05-17 19:49:32 <eex0> biso = bios*
2023-05-17 19:50:28 <minimal> if you're using efibootmgr then you mean UEFI
2023-05-17 19:50:53 <eex0> nods yes
2023-05-17 19:50:54 <Ermine> There was a mail by Rob Landley on busybox mailing list on FHS
2023-05-17 19:51:59 <Ermine> (by FHS I mean all that separation)
2023-05-17 19:56:52 <Saijin_Naib> Have link, or is it recent so I can dig it out? 
2023-05-17 19:59:23 <ikke> http://landley.net/hg/aboriginal/file/be48c60f9edb/design.html#l288
2023-05-17 19:59:34 <ikke> http://lists.busybox.net/pipermail/busybox/2010-December/074114.html
2023-05-17 19:59:51 <ikke> (found it through https://lwn.net/Articles/477467/)
2023-05-17 20:00:23 <Ermine> thank you ikke 
2023-05-17 20:01:25 <Saijin_Naib> Ikke the librarian 
2023-05-17 20:01:53 <eex0> goes to show, once on the inet, always on the inet
2023-05-17 20:02:11 <eex0> dang 2010..lol
2023-05-17 20:02:15 <ikke> There is plenty of content disappearing
2023-05-17 20:03:54 <Saijin_Naib> Or buried increasing under waves of LLM-generated shit 
2023-05-17 20:07:00 <psykose> love to put my binaries in 9 places to serve no purpose other than to then have to put those 9 places into PATH
2023-05-17 20:07:54 <minimal> especially with busybox and "full fat" apps not putting themselves in the same place ;-)
2023-05-17 20:08:03 <psykose> love too use /usr/bin and regret my life choices because actually it was in /sbin, of course it was, why didn't you just guess that
2023-05-17 20:08:07 <pj> why not have a path in PATH for each program
2023-05-17 20:08:15 <pj> windows style
2023-05-17 20:10:16 <ikke> And then require reboot after installation, just the way we like it
2023-05-17 20:12:31 <pj> Here is protip: it doesn't actually require reboot
2023-05-17 20:12:49 <psykose> it does
2023-05-17 20:13:11 <psykose> if the env is immutable you have to reboot of course
2023-05-17 20:13:15 <pj> it doesn't
2023-05-17 20:13:41 <ikke> logout/login, but in practice they would just ask you to reboot
2023-05-17 20:14:05 <pj> none of that is required
2023-05-17 20:14:07 <psykose> the dbus stuff doesn't need you to relog for env updates anymore
2023-05-17 20:14:12 <psykose> on systemd at least
2023-05-17 20:15:00 <Ermine> VS asks for reebot, VMWare asks for reeboot, and I don't remember anything else asking for reboot these times
2023-05-17 20:15:21 <pj> VMware asks for reboot because it installs a driver
2023-05-17 20:15:35 <Ermine> Yes
2023-05-17 20:15:35 <psykose> classic driver copout /s
2023-05-17 20:18:05 <Ermine> Well, I guess NixOS doesn't add every single package dir to PATH?
2023-05-18 03:17:38 <Saijin_Naib> Was the UDF kernel option dropped from linux-lts because optical drives aren't super common or something?
2023-05-18 03:17:45 <Saijin_Naib> I just got bit by this trying to mount a flashdrive
2023-05-18 03:19:07 <Saijin_Naib> It also kind of sucks for my CD/DVD/Blu-Ray drive
2023-05-18 03:24:10 <XFaCiEer> Wait what? How does that work Saijin_Naib ?
2023-05-18 03:24:24 <XFaCiEer> It isn't dropped from non-LTS kernels?
2023-05-18 03:25:00 <Saijin_Naib> I dunno, it was present under linux-lts 5.15.x before the Alpine v3.18 upgrade and switch to linux-lts 6.1.x...
2023-05-18 03:25:19 <XFaCiEer> UDF support can't be dropped.
2023-05-18 03:25:25 <Saijin_Naib> I know we try to use mostly upstream defaults, so maybe upstream dropped it as default on or default module
2023-05-18 03:25:30 <XFaCiEer> That's ridiculous.
2023-05-18 03:25:35 <Saijin_Naib> Hm, then something else is amiss
2023-05-18 03:25:38 <XFaCiEer> It's like dropping FAT32.
2023-05-18 03:26:06 <XFaCiEer> UDF is way too ubiquitous for it to be just let go.
2023-05-18 03:29:39 <Saijin_Naib> Hm, no, it is indeed a module in our kernel config
2023-05-18 03:29:45 <Saijin_Naib> What the fuck is my machine doing now
2023-05-18 04:00:11 <Apachez> what do you mean?
2023-05-18 04:00:19 <Apachez> does it humping your dns-servers or something?
2023-05-18 04:00:45 <Apachez> screams at the other server to "look at ma packet size! look at it!!!" ?
2023-05-18 04:32:42 <mbx> heyjo, found a mistake in the docs, see https://wiki.alpinelinux.org/wiki/Upgrading_Alpine
2023-05-18 04:33:00 <mbx> it shows apk add --upgrade apk-tools ... instead of apk add --update apk-tools
2023-05-18 04:36:02 <mbx> no wait, --update is not a flag as well
2023-05-18 04:36:31 <mbx> had a type, it is
2023-05-18 04:36:34 <mbx> typo*
2023-05-18 04:36:55 <mbx> so yea, docs have a typo
2023-05-18 07:59:13 <powers> is anyone here familiar with the Linux kernel?
2023-05-18 07:59:40 <powers> If so, do you happen to know what was 'KERNEL_DS' replaced with after kernel-5.10?
2023-05-18 07:59:52 <powers> I see `get_ds` but its at weird file-path, so I wanted to confirm
2023-05-18 10:43:01 <psykose> mbx: apk add --upgrade is valid
2023-05-18 15:20:07 <blt[m]> any of you tried to install Alpine on Kobol Helios64 ? I got one running on armbian and wondering if it would work with alpine aarch64
2023-05-18 15:33:14 <Saijin_Naib> Oh, lucky you! That was my dream NAS with Alpine, but they ended orders right when I discovered them
2023-05-18 15:33:51 <Saijin_Naib> I think most of the hardware should be supported, but I have no idea how you'd have to install to have it boot 
2023-05-18 15:37:47 <blt[m]> 😇 yes I was lucky and too bad they stopped...
2023-05-18 15:37:47 <blt[m]> what I see is indeed nothing fancy, but the challenge is probably to boot on it, I am wondering if tweaking this https://wiki.kobol.io/helios64/install/preliminary/ with Alpine would work
2023-05-18 16:11:40 <Saijin_Naib> I guess give the ARM image a try using their microSD install instructions? 
2023-05-18 16:42:15 <blt[m]> That's exactly what I am going to try, yes thanks will post my findings in case
2023-05-18 17:12:57 <boxemall> hello there. just trying my 1st steps with alpine. i have it installed and ssh is running. i just can't ssh onto the machine even tho i have it's ip adress. currently logging as root user only. it's all for testing and learning currently so i don't hae a prob doing it the root user way. i have the standard install so i shouldn't have any firewall installed. any idea why i can't reach to the machine?
2023-05-18 17:13:34 <ikke> boxemall: How are you trying to login? By default password login for root is disabled
2023-05-18 17:14:25 <boxemall> ah okay. yeah was trying to log in as root since it is the only user i have atm. looks like i have to set up a user after all? or can i modify it so i have root access?
2023-05-18 17:14:47 <ikke> boxemall: adding an ssh key would help
2023-05-18 17:15:04 <ikke> otherwise change /etc/ssh/sshd_config PermitRootlogin yes
2023-05-18 17:15:16 <boxemall> uh news... let me look up the wiki. dunno hwo to do that
2023-05-18 17:16:39 <boxemall> just ran "ssh-keygen" it seems to be doing something...
2023-05-18 17:17:11 <boxemall> right,... i'm being asked in what file to save the key. anything i need to watch out for?
2023-05-18 17:19:28 <ikke> The default name should suffice
2023-05-18 17:19:44 <ikke> (if you rename it, you need to explicitly tell ssh to look for it)
2023-05-18 17:23:24 <boxemall> k
2023-05-18 17:23:44 <boxemall> do i need to enter a passphrase?
2023-05-18 17:24:05 <ikke> It's optional
2023-05-18 17:25:54 <boxemall> so no drawback for now. ty
2023-05-18 17:30:55 <boxemall> network error. timeout. still something seems missing. i see that it's on a different subnet. it's on 192.168.179 instead fo 178 could that be it?
2023-05-18 17:31:44 <ikke> Could be, I don't know your network
2023-05-18 17:35:58 <boxemall> how do i foce a different ip. kinda weird since my router won't let me cahnge it. it usually works 
2023-05-18 17:36:38 <ikke> What is the subnetmask / prefix length?
2023-05-18 17:41:34 <boxemall> nvm i got it. i connected to the "guest" network which does not have full access nor can it be reached by devices on the network or reach out to devices itself. it only has internet access. i did that, now i remember, since my standard wifi key is so long that you always have typos when u don't see what ur typing lol. i need to switch networks. easier way would be if alpine supported WPS. i guess it does not? or i 
2023-05-18 17:41:34 <boxemall> will have to try to make a text file on a usb device and copy it using nano?
2023-05-18 17:42:18 <ikke> Not aware of any WPS support
2023-05-18 17:42:31 <ikke> But maybe something like network manager supports it
2023-05-18 17:44:32 <boxemall> i'm only on shell. does network manager run on shell?
2023-05-18 17:46:47 <ikke> There is nmcli, but I don't know it very well
2023-05-18 17:51:09 <scorpion2185[m]1> man nmcli-examples nm-cli
2023-05-18 17:51:29 <boxemall> tried installing networkmanager. no such package it says. interesting... i double checked for typos. i guess invoking "nmcli" should start it on shell but it seems it's not installed either
2023-05-18 17:51:53 <boxemall> man page also not found
2023-05-18 17:51:56 <ikke> It's in the community repository, so make sure you have that enabled
2023-05-18 17:52:02 <boxemall> oh ok.
2023-05-18 17:53:19 <boxemall> right it was hashed "#" removed it. do i have to reboot or just try installing again?
2023-05-18 17:53:49 <boxemall> installing
2023-05-18 17:55:46 <ikke> yeah, no reboot required
2023-05-18 18:25:58 <boxemall> trying to "iwctl station wlan0 connect <SSID>" and i enter my SSID but it tells me invalid network. do i need to put it in brackets or something? <> doesn't work...
2023-05-18 18:26:15 <psykose> you can to scan first
2023-05-18 18:26:18 <boxemall> the network is found though, amogst others
2023-05-18 18:26:24 <boxemall> i did that
2023-05-18 18:26:37 <boxemall> i have a list and my network shows up
2023-05-18 18:27:09 <psykose> what is the SSID
2023-05-18 18:27:25 <boxemall> although it has spaces in its name like "this is my network"
2023-05-18 18:27:35 <ikke> Then you need to quote it
2023-05-18 18:27:37 <psykose> then you have to quote it
2023-05-18 18:27:40 <boxemall> aye
2023-05-18 18:28:23 <boxemall> nice this was it. now i just need to figure out a way to enter the key. gonna try opeing another shell. what was it again ctrl+f key?
2023-05-18 18:29:25 <ikke> ctrl + f2?
2023-05-18 18:30:58 <boxemall> it seems "CTRL+ALT+Fkey"
2023-05-18 18:31:17 <boxemall> i guess i need to mount the usb stick first?
2023-05-18 18:32:53 <ikke> order does not matter
2023-05-18 18:35:27 <boxemall> lol this is really stripped down. i coudl not even lsblk. installing now :-)
2023-05-18 18:36:39 <boxemall> hmm can't install. can't install libsmartcols
2023-05-18 18:36:47 <boxemall> getting a temp error
2023-05-18 18:37:44 <ikke> sounds like dns or another network issue
2023-05-18 18:40:10 <boxemall> weird since i was able to install everything so far. who knows... it gets stuck at 54%
2023-05-18 18:43:07 <boxemall> changed to another repo server. it installed flawlessly now
2023-05-18 19:00:19 <boxemall> ok i thought i could copy in one terminal and paste in another but that does not seem to work. is there another way of editing the wpa_supplicant maybe? it does not save the passkey in clear text (why would it). how do i get the crazy key i have to work without having a million retries due to typos...
2023-05-18 19:04:37 <dwfreed> wpa_supplicant can take the password in its original form, or it can take the result of the expansion function
2023-05-18 19:04:50 <dwfreed> it will do the expansion function if given the password in original form
2023-05-18 19:06:22 <dwfreed> wpa2 uses pbkdf2 to derive the actual network authentication key, as it always uses a 256 bit key; this is why your password can be 8 to 63 characters or 64 hex digits
2023-05-18 19:07:20 <dwfreed> often times tools that create the wpa_supplicant config will do the pbkdf2 expansion before writing the config; you can use wpa_passphrase to do the same yourself
2023-05-18 20:57:22 <blt[m]> Saijin_Naib: if I try the Generic ARM aarch64 it seems that it is missing a valid boot partition, but there is probably something else to do, not an expert on arm, first time I am playing with it. If I try standard aarch64, I can see the Booting 'Linux LTS' but got Mounting boot media failed. 
2023-05-18 20:59:46 <Saijin_Naib> I've never succeeded either on any ARM hardware, so I'm not going to be very useful 
2023-05-18 22:29:30 <ziproot> When I run startx it shows some kind of command line and then doesn't respond to most commands, including the command to kill x. I was able to open a virtual terminal which I am in now.
2023-05-18 22:31:15 <ziproot> By commands I mean keystrokes
2023-05-18 22:31:49 <ziproot> By command to kill x I mean ctrl+alt+backspace
2023-05-18 22:34:41 <dwfreed> ctrl+alt+backspace has to be given to X
2023-05-18 22:34:51 <dwfreed> the terminal that startx is running in would not handle it
2023-05-18 22:35:18 <Habbie> +1
2023-05-18 22:36:16 <ziproot> OK, thank you. So how do I get it to open up dwm instead of opening up the unresponsive terminal?
2023-05-18 22:36:39 <dwfreed> that's all dependent on your xinitrc
2023-05-18 22:36:41 <Habbie> can you show us what it showed?
2023-05-18 22:36:59 <ziproot> .statusbar.sh & \n exec dwm
2023-05-18 22:37:13 <ziproot> oh wait I see the problem
2023-05-18 22:37:20 <ziproot> I ported it from debian and it is trying to run dash which doesn't exist
2023-05-18 22:37:59 <ziproot> Thank you
2023-05-18 23:14:55 <ziproot> Hello, xserver still froze up, even with just "exec dwm" in startx file. I was able to kill it and prior to me killing it it had returned an error like "xf86EnableIO: faile to enable I/O port 0000-03ff (Operation not permitted)"
2023-05-18 23:15:26 <psykose> are you in the input/video groups
2023-05-18 23:15:42 <ziproot> I am in the video group, not the input group
2023-05-18 23:20:25 <ziproot> I ran "doas adduser ziproot input" but it "groups" still doesn't say that I'm in the input group, only ziproot, wheel, audio, video, and netdev
2023-05-18 23:20:44 <ziproot> oh wait nevermind, sorry about that
2023-05-18 23:23:44 <psykose> you have to relog after doing that
2023-05-18 23:30:03 <ziproot> Hello, I am now in the "input" and "video" groups and .xinitrc says "exec dwm" and I still get the same error.
2023-05-18 23:31:53 <psykose> do you have xf86-video-intel installed by accident
2023-05-18 23:32:17 <ziproot> no, only linux-firmware-intel
2023-05-18 23:32:48 <ziproot> I do have a secondary NVIDIA gpu
2023-05-18 23:33:27 <psykose> can you paste the xorg log
2023-05-18 23:34:10 <ziproot> OK, give me a minute
2023-05-18 23:37:02 <ziproot> https://pst.moe/adeoya
2023-05-18 23:38:13 <psykose> that just says not found
2023-05-18 23:39:42 <ziproot> oh wait https://pst.moe/paste/adeoya
2023-05-18 23:39:59 <ziproot> sorry about that
2023-05-18 23:42:03 <ziproot> and add ?format=irc to that url
2023-05-18 23:46:21 <psykose> there's no errors in that
2023-05-18 23:46:25 <psykose> dunno
2023-05-18 23:46:38 <psykose> prevent nvidia from even loading and maybe it works or maybe it doesn't
2023-05-18 23:46:44 <psykose> i assume you have mesa-dri-gallium too
2023-05-19 00:07:26 <ziproot> yes
2023-05-19 00:07:55 <ziproot> how do I stop nvidia from loading?
2023-05-19 00:08:38 <psykose> `blacklist nouveau` in /etc/modprobe.d/somefile.conf and reboot
2023-05-19 00:09:08 <ziproot> OK, thank you, I will try that.
2023-05-19 00:24:51 <ziproot> nouveau is blacklisted but I still cannot log in
2023-05-19 00:30:51 <psykose> does `apk add sway; sway` start
2023-05-19 00:31:50 <psykose> well, that also would require starting seatd, which is +1 step
2023-05-19 00:36:20 <ziproot> XDG_RUNTIME_DIR is not set in the environment, aborting
2023-05-19 01:02:14 <ziproot> psykose: the message "XDG_RUNTIME_DIR is not set in the environment, aborting" was printed when I tried to start sway
2023-05-19 01:02:40 <psykose> not to be offensive but that tells you exactly what the issue is and you can put it in google
2023-05-19 01:03:02 <ziproot> sorry, I was answering your question
2023-05-19 01:05:03 <qaqland[m]> Script from alpinelinux_wiki/wayland sometimes work, sometimes not. But you can try
2023-05-19 01:07:58 <ziproot> OK, sway starts, and it loads a gray screen with my cursor. I can move it around with my trackpad and not my mouse.
2023-05-19 01:09:48 <psykose> sure
2023-05-19 01:09:50 <abby> is your mouse plugged in
2023-05-19 01:09:57 <psykose> guess everything works then
2023-05-19 01:10:13 <psykose> xorg not working is a 'who even uses that anymore xd' :)
2023-05-19 01:14:19 <ziproot> OK, after unplugging and replugging in my mouse sway recognized it. Furthermore, I installed mega-gles, vulkan-loader, xcb-util-renderutil, and xcb-util-wm, which made dwm work. Thank you.
2023-05-19 09:38:27 <WhyNotHugo> Does anyone have a recipe for using shellcheck with APKBUILDs (but avoiding all the false positives?)
2023-05-19 09:38:57 <WhyNotHugo> I use shellcheck via vim (which is usually very useful), but I get a lot of things like "variable pkgdesc is unused", which don't make sense in this context.
2023-05-19 09:41:32 <ikke> Yes, our CI uses a shim, and Leo has a vim plugin, but that would also check apkbuild-lint 
2023-05-19 09:45:37 <WhyNotHugo> Thanks! Found it here: https://gitlab.alpinelinux.org/alpine/infra/docker/apkbuild-lint-tools/-/tree/master
2023-05-19 15:33:57 <eex0> I'd vote (when installing alpine) make grub optional. yes, no?
2023-05-19 15:34:39 <psykose> it already is
2023-05-19 15:35:52 <eex0> oh.. hrm, I didnt see that when I reinstalled 2-days ago tho, I may have done this so many times now I just dont notice anymore.
2023-05-19 15:37:41 <abby> ooh that shellcheck wrapper looks nice, i might steal it for void
2023-05-19 15:43:35 <psykose> yeah
2023-05-19 15:46:32 <minimal> eex0: make grub optional? in which specific installation scenarios?
2023-05-19 15:46:46 <minimal> in almost all scenarios you need a bootloader or some kind
2023-05-19 15:47:14 <minimal> already if you install Alpine in Sys mode using BIOS (rather than UEFI) then Syslinux is installed by default
2023-05-19 15:47:46 <eex0> efibootmgr can be setup to direct boot
2023-05-19 15:47:54 <eex0> from bios
2023-05-19 15:48:20 <minimal> yes but setup-alpine/setup-disk doesn't (currently) support efibootmgr, or limine, or some other bootloaders
2023-05-19 15:49:39 <eex0> nods and this is why I mentioned this again.. efibootmgr might be more secure than grub.. I'm still looking into all this myself.
2023-05-19 15:50:09 <minimal> so do "BOOTLOADER=none setup-disk" ?
2023-05-19 15:51:00 <minimal> eex0: it is unrealistic for the installation script(s) to support more than a limited set of bootloaders as then they'd become complicated to test and maintain
2023-05-19 15:51:28 <eex0> there is a script for efibootmgr .. use: grub? n,y [y]
2023-05-19 15:51:43 <minimal> what script?
2023-05-19 15:52:06 <eex0> moment, I'll have to hunt, it's on the alpine pages
2023-05-19 15:52:38 <eex0> https://wiki.alpinelinux.org/wiki/Bootloaders
2023-05-19 15:52:39 <minimal> if it is on the wiki then it not an official script delivered as part of alpine-conf
2023-05-19 15:53:42 <eex0> redhat writes efibootmgr, who writes grub and why is there still a cve for that in alpine?
2023-05-19 15:54:00 <eex0> Im up for options
2023-05-19 15:54:27 <minimal> you have the option of installing whichever bootloader you want
2023-05-19 15:54:44 <minimal> as to what the official setup scripts support, that is a different matter
2023-05-19 15:55:29 <eex0> nods, I wasnt trying to start a food fight or anything, shit!
2023-05-19 15:55:53 <eex0> just make some observation and ask a question
2023-05-19 17:02:13 <nuser> Hello everyone ;) I' wondering which distros are as light and well supported as alpine? I do have to get below 485MB with CLI only, alpine does that well but do you know any alternative? with binutils
2023-05-19 17:02:49 <Habbie> nuser, openwrt
2023-05-19 17:03:11 <Habbie> 485 is not very small - could maybe even fit debian in it, or many others
2023-05-19 17:04:12 <Habbie> the debian 12 docker image (not even the slim one) is 116MB
2023-05-19 17:04:26 <Habbie> that's without a kernel, of course
2023-05-19 17:05:33 <minimal> nuser: why are you considering alternatives to Alpine? what is your criteria? why does Alpine not meet that criteria?
2023-05-19 17:06:20 <nuser> really? how does that work without kernel? almost all standard iso cli install images install size is bigger than that 485mb.
2023-05-19 17:06:36 <ikke> docker images do not need a kernel
2023-05-19 17:07:33 <ikke> as they just run processes on the host
2023-05-19 17:07:36 <nuser> minimal: to diversify. but all is too big in size alpine 's about 200mb 
2023-05-19 17:07:50 <ikke> size of what?
2023-05-19 17:09:13 <nuser> ikke: size of disk rootfs and so on
2023-05-19 17:09:40 <ikke> You can safe a lot by only having the firmware that you need installed
2023-05-19 17:10:02 <minimal> nuser: alpine is about 200mb? that depends on exactly what you've installed
2023-05-19 17:10:08 <minimal> which again depends on your criteria
2023-05-19 17:10:17 <abby> iso images are usually larger than an installed system because they have extra tools in them
2023-05-19 17:10:23 <minimal> I have Alpine VMs that are around 60-70Mb in size
2023-05-19 17:10:39 <eex0> nice!
2023-05-19 17:10:41 <Habbie> VMs? so including kernel?
2023-05-19 17:10:47 <minimal> yes, linux-virt
2023-05-19 17:10:52 <Habbie> nice
2023-05-19 17:10:55 <nuser> that's tiny 60-70?!
2023-05-19 17:11:03 <Habbie> ah virt avoids a lot of hw drivers i guess
2023-05-19 17:11:10 <ikke> and kernel modules as well
2023-05-19 17:11:18 <minimal> Habbie: yes and avoids firmware too
2023-05-19 17:11:18 <Habbie> is what i meant!
2023-05-19 17:11:22 <Habbie> ah that too
2023-05-19 17:11:27 <ikke> right
2023-05-19 17:11:32 <ikke> firmware + kernel modules
2023-05-19 17:11:38 <Habbie> ack
2023-05-19 17:11:45 <minimal> and that's with "tailored" initramfs to only contain precisely what is required
2023-05-19 17:11:45 <Habbie> until you do pci passthrough i guess
2023-05-19 17:12:25 <minimal> nuser: which goes back to your criteria, what do you want to use Alpine, or another OS, for exactly?
2023-05-19 17:15:33 <nuser> just to play with it: i2pd, tor, apache, as ssh jumpproxy just to learn how to set things up to get it working, set up openvpn server and client at once  sthn' like ssh jumping that kind of stuff
2023-05-19 17:19:04 <nuser> I still don't get it how to get debian or arch,gentoo below 485mb all is in GB's ... could you point me some dovumentation or tutorials how to do it?
2023-05-19 17:19:21 <Habbie> where do you see GBs?
2023-05-19 17:20:32 <nuser> ex. my arch... 2.9gb alpine is only distro that I found below 485
2023-05-19 17:20:57 <Habbie> then i bet there's a lot installed on your arch that you don't need
2023-05-19 17:21:58 <nuser> not so much. just standard installation with aweso!e and openbox
2023-05-19 17:23:17 <nuser> that is why I ask for sthn' similar to alpine here. to diversify and try sthnig else
2023-05-19 17:24:15 <ikke> void perhaps
2023-05-19 17:24:58 <nuser> any ideas? or sould I try to make debian or arch as light? void linux?
2023-05-19 17:31:10 <nuser> ikke: could be fine rootfs around 35mb , thanks
2023-05-19 17:32:18 <minimal> nuser: would you go into a Ford dealership and ask them for advice as to what is the best Volkswagen or Fiat? ;-)
2023-05-19 17:32:32 <Habbie> no
2023-05-19 17:32:41 <Habbie> well
2023-05-19 17:32:47 <Habbie> you walk past the front desk to the mechanics
2023-05-19 17:32:49 <Habbie> and ask them ;)
2023-05-19 17:33:13 <minimal> Habbie: they'd still be in the dealership though lol
2023-05-19 17:33:18 <Habbie> ok fair
2023-05-19 17:34:11 <psykose> anyone can give me a 5mb rootfs image with every application
2023-05-19 17:34:42 <nuser> right, it was not the greatest idea :) how long will alpine be around? 
2023-05-19 17:35:02 <Habbie> what do you mean, how long?
2023-05-19 17:35:10 <ikke> Just use https://en.wikipedia.org/wiki/Sloot_Digital_Coding_System
2023-05-19 17:35:15 <Habbie> ahaha
2023-05-19 17:35:15 <ikke> It'll easily fit
2023-05-19 17:35:16 <nuser> psykose: how's that possible every app in 5mb?
2023-05-19 17:35:23 <Habbie> no, -one- app
2023-05-19 17:35:32 <Habbie> you can choose from many, but you pick one
2023-05-19 17:35:45 <ikke> nuser: alpine has been there for almost 20 years
2023-05-19 17:37:44 <nuser> I should definitely google stuff before I ask :)
2023-05-19 17:40:11 <nuser> I'm also wonderig why there is 10k apps difference between 3.15 and 3.18 repo? is it normal?
2023-05-19 17:40:32 <minimal> ikke: I'd never heard of Sloot. Sometimes "incredible" claims do have merit, I remember Fractal Image Compression back in the 1990s
2023-05-19 17:40:50 <ikke> There was a lot of controversy about that
2023-05-19 17:41:18 <ikke> They do say it's not compression but encoding
2023-05-19 17:41:28 <psykose> encoding is compression ;)
2023-05-19 17:42:31 <ikke> pre-shared dictionaries
2023-05-19 17:42:57 <minimal> ikke: Barnsley "cut his own throat" on that as he refused to license others to code it themselves, the only way to use it was to purchase his own library
2023-05-19 17:43:27 <minimal> it did however find interesting military applications though...
2023-05-19 17:58:38 <eex0> great for acid trips.. what I heard
2023-05-19 18:43:47 <follie> Hmm I probably should disable httpstate test of pulumi. It keeps failing on the builder
2023-05-19 18:44:34 <psykose> classic millisecond timing issue :D
2023-05-19 18:44:45 <ikke> Yeah
2023-05-19 18:44:51 <ikke> xpired token token-1 (7.395554ms past expiration)
2023-05-19 23:16:38 <felixphew> minimal: I've met Barnsley a few times; weird dude, great fun to talk to about maths stuff
2023-05-19 23:22:21 <felixphew> a funny Eccentric Wealthy Person Thing is that his bathroom is tiled with an infinite aperiodic tiling (https://goldenbpuzzles.com/wp-content/uploads/2016/11/golden-b-tiling.png)
2023-05-19 23:26:25 <minimal> in a horrible yellow/orange colour? lol
2023-05-19 23:30:19 <felixphew> I don't know, I haven't seen it myself
2023-05-20 01:20:33 <em6> hi, I cant find the grub file when running diskless and I cant find a solution online, any suggestions?
2023-05-20 01:27:19 <minimal> em6: what Grub file?
2023-05-20 01:28:44 <em6> I usually find it here: /etc/default/grub
2023-05-20 01:29:00 <em6> with sys install
2023-05-20 01:29:12 <minimal> ok, but you mentioned diskless
2023-05-20 01:29:48 <em6> yes I run diskless now with seperat disk for configs and lbu commit
2023-05-20 01:31:22 <minimal> booting via UEFI or BIOS?
2023-05-20 01:31:32 <em6> UEFI
2023-05-20 01:33:12 <em6> I was trying to find the grub file on the boot image but only found a folder called grub with a file called grub.cfg, but now I have no idea how this works
2023-05-20 01:33:44 <minimal> from memory the diskless installer just copies files off the ISO
2023-05-20 01:34:01 <minimal> why do you want to find /etc/default/grub?
2023-05-20 01:34:15 <em6> ok so I edit the ISO content?
2023-05-20 01:34:50 <minimal> diskless isn't really designed for changes like that though
2023-05-20 01:35:26 <em6> I want to add mitigations=off isolcpus=2-3 to GRUB_CMDLINE_LINUX_DEFAULT
2023-05-20 01:35:29 <minimal> e.g. AFAIK you don't really upgrade diskless, you download a newer version to use
2023-05-20 01:36:06 <minimal> you'd have to edit boot/grub/grub.cfg then...
2023-05-20 01:36:44 <minimal> though isn't the boot "media" an ISO filesystem? so not designed to be modified
2023-05-20 01:36:56 <em6> ok
2023-05-20 01:37:02 <em6> but is it safe then?
2023-05-20 01:37:51 <em6> do I add it to: /boot/vmlinuz-lts modules=loop,squashfs,sd-mod,usb-storage quiet initrd
2023-05-20 01:37:52 <em6> ?
2023-05-20 01:38:01 <minimal> I don't use diskless often so forget if it is indeed an ISO9660 fs
2023-05-20 01:38:38 <em6> ok
2023-05-20 01:38:53 <minimal> if you run "df" do you see a /media/cdrom (or /media/usb or /media/sda" mount?
2023-05-20 01:42:13 <em6> I have several things mounted!!?
2023-05-20 01:42:36 <em6> the iso is /media/sdc
2023-05-20 01:42:37 <minimal> I asked if "df" showed either of these mounts?
2023-05-20 01:43:02 <minimal> ok, so sdc is a USB stick or disk?
2023-05-20 01:43:07 <em6> then no
2023-05-20 01:43:17 <minimal> so what does "mount | grep sdc" show?
2023-05-20 01:43:24 <em6> usb stick
2023-05-20 01:45:59 <minimal> ?
2023-05-20 01:48:19 <em6> minimal: https://mclo.gs/TrXomxQ
2023-05-20 01:48:48 <em6> had to find a way to paste code
2023-05-20 01:48:52 <minimal> so its a iso9660 fs
2023-05-20 01:49:13 <minimal> which effectively is readonly
2023-05-20 01:49:58 <em6> ok I see
2023-05-20 01:50:50 <em6> thanks anyway
2023-05-20 01:54:17 <em6> do you think I can run a setup script to place the ISO on a disk and then have different write permisions? 
2023-05-20 01:58:27 <minimal> is is now about write permissions
2023-05-20 01:58:52 <minimal> a ISO9660 filesystem is "in theory" designed to be readonly
2023-05-20 01:58:58 <minimal> it was designed for CDROMs
2023-05-20 02:57:50 <monokrome> Hi, y'all. I'm running into an issue that I've had a bit of a tough time tracking down. I think this may be the best place to ask, but apologize if its not. Anyway, I've installed musl, musl-dev, gcc, linux-headers, etc. When I include sys/time.h, the `time` symbol is not defined. Am I missing some specific/obvious package that brings in a different variant or something?
2023-05-20 02:58:43 <monokrome> This isn't my code fwiw, its a package that seems to compile in other places: https://gitlab.com/mattwb65/srt-live-server/-/blob/master/slscore/common.cpp#L111
2023-05-20 03:00:04 <psykose> time is in <time.h> not <sys/time.h>
2023-05-20 03:00:30 <monokrome> hmm
2023-05-20 03:00:34 <monokrome> odd that it compiles elsewhere
2023-05-20 03:00:39 <psykose> not really
2023-05-20 03:00:46 <monokrome> but, yeah, that makes sense :}
2023-05-20 03:01:01 <monokrome> not really?
2023-05-20 03:01:37 <monokrome> adding time.h fixes the code
2023-05-20 03:01:48 <monokrome> thanks - guess I'll test it and see if its worth sending upstream
2023-05-20 03:01:53 <psykose> it is
2023-05-20 03:02:23 <monokrome> I'm confused. Why is not not really odd that it compile elsewhere? Also, wdym by "it is"?
2023-05-20 03:03:48 <psykose> musl is quite strict about include namespaces so you don't get things exposed without including the header for it (time is from time.h only)
2023-05-20 03:03:53 <psykose> it is worth sending it upstream
2023-05-20 03:04:07 <monokrome> Ah. Makes sense.
2023-05-20 03:04:24 <monokrome> Thanks for clarifying :)
2023-05-20 03:10:18 <iamthemcmaster> psykose: many thanks for hitting the update on suricata 6.0.12. i used to have notifications set up when they make new releases but i haven't gotten them set up again after changing jobs. i ought to get on that...
2023-05-20 03:10:37 <psykose> no worries :)
2023-05-20 04:55:11 <boxemall> i just did a "apk search auto-cpufreq" and it did not return any results. is there another addon that i could use to regulate CPU clocks when idle?
2023-05-20 04:57:11 <psykose> set the governor to powersave and forget about it
2023-05-20 05:09:55 <boxemall> has anyone successfully installed "iobroker" on alpine?
2023-05-20 06:46:32 <humm> You could CONT/STOP your evil processes just a few times each second.
2023-05-20 07:51:35 <powers> is there a way to download a package while avoiding non-essential package-upgrades?
2023-05-20 07:53:31 <powers> or better yet, is there a way to find the ids of older package-versions? That way I can tag the specific package i want to download/use
2023-05-20 07:57:36 <quinq> powers, apk upgrade $the_package_to_upgrade
2023-05-20 09:38:19 <sylvain31> hello, quite new to alpine linux, installed in VM, trying to install a gui desktop: PRETTY_NAME="Alpine Linux v3.18" does alpine-desktop (no such package) still exists?
2023-05-20 09:38:59 <psykose> nope
2023-05-20 09:40:03 <sylvain31> psykose, you confirm the the package alpine-desktop  doesn't anymore? So wiki should but updated I suppose. 
2023-05-20 09:40:21 <psykose> which part of it
2023-05-20 09:40:38 <sylvain31> https://docs.alpinelinux.org/user-handbook/0.1a/Working/post-install.html#_getting_a_graphical_environment
2023-05-20 09:41:24 <sylvain31> > Once that is done, you can install the recommended and supported graphical setup by installing (and thus running) apk add alpine-desktop. 
2023-05-20 09:44:49 <psykose> ah that
2023-05-20 09:44:53 <psykose> yeah
2023-05-20 09:44:59 <psykose> https://gitlab.alpinelinux.org/alpine/docs/user-handbook/-/issues/2
2023-05-20 09:45:07 <psykose> you can run setup-desktop xfce for the same thing as before i guess
2023-05-20 09:49:26 <sylvain31> psykose, yes, it did work with 'setup-desktop xfce` +  startx on the console.
2023-05-20 09:51:06 <sylvain31> that'ss the good issue thanks. May be just update the documentation so? May be the handbook is not a wiki though.
2023-05-20 09:51:31 <psykose> i just tried but pretty much the entire section is wrong
2023-05-20 09:51:43 <psykose> and i'm bad at writing docs and would rather not just rewrite the whole thing
2023-05-20 09:51:46 <psykose> anyone else can also do it
2023-05-20 09:51:47 <psykose> :p
2023-05-20 09:54:07 <sylvain31> psykose, found the repos: https://gitlab.alpinelinux.org/alpine/docs/user-handbook/-/blob/master/modules/Working/pages/post-install.adoc
2023-05-20 09:56:40 <sylvain31> what is the approval review process? I'm a bit new to Alpine-Linux to produce a valid, documentation, but may be just some cross link to find the package related and the command. I suppose the audience is not really Linux beginner for this distro?
2023-05-20 09:58:00 <psykose> indeed not
2023-05-20 09:58:23 <psykose> you just open a merge request and that's about it, same as anywhere else
2023-05-20 10:00:30 <sylvain31> psykose, OK. Basically I was testing the distro, to run as baremetal OS to run docker containers. It seems one of the lightweight distro that can do that well. 
2023-05-20 10:01:24 <sylvain31> now, I met a documentation mismatch may I can take some time to contribute by just patch some line of documentation. ;-) 
2023-05-20 10:01:43 <psykose> i mean it would need rewriting the whole section, i do mean the whole thing is wrong
2023-05-20 10:01:55 <psykose> so if you don't know how to do that then i'm not sure how you would change it
2023-05-20 10:02:00 <psykose> it's not a 2 word change
2023-05-20 10:04:41 <sylvain31> psykose, that's my point. It worked of my test, and I don't plan to really use the GUI on the distro. But to save other curious user the missing package message... May just remove the older command and put a link to https://wiki.alpinelinux.org/wiki/Alpine_setup_scripts#setup-desktop
2023-05-20 10:05:42 <humm> please not a link to the wiki
2023-05-20 10:11:28 <sylvain31> humm, what's wrong with the wiki?
2023-05-20 10:18:09 <humm> sylvain31, it’s good that the wiki exists. If not great, it is useful. But a user handbook should keep the pieces at one place. Having a page of the handbook describe what to do is preferable to having it point to the wiki—otherwise, there wouldn’t be a point in having a handbook.
2023-05-20 10:33:14 <sylvain31> humm, I agree. I've to go, the gitlab issue already reference it. May I could help, may be not. We will see. On a user point of view (here alpine linux is mostly skilled user, at least developer range or sysadmins) having to cut info seems not a problem to me. I got 3 peaces of info (probably much more in fact) 1. the hanbook, 2. the IRC, 3. the gitlab issue, 4. the wiki. Now I'm able to see the picture and the fix...
2023-05-20 12:41:52 <earboxer> Is there a way to use setup-bootable to create an alternate /boot device for an existing alpine install?
2023-05-20 12:42:48 <earboxer> (my /dev/sda1 fails to save the initramfs on kernel upgrades)
2023-05-20 13:27:25 <humm> Instead of just creating a filesystem, mounting it as /boot, installing kernel, and telling bootloader what to load?
2023-05-20 13:40:07 <earboxer> I need EFI on laptop, and sda2 is encrypted.
2023-05-20 13:50:35 <minimal> earboxer: what did you use to update the kernel?
2023-05-20 14:04:09 <earboxer> minimal: apk upgrade
2023-05-20 14:05:13 <minimal> earboxer: you're using Alpine in run-from-ram mode right? so that's not the way to do it
2023-05-20 14:05:51 <earboxer> I'm using sys mode
2023-05-20 14:06:24 <minimal> you mentioned setup-bootable though
2023-05-20 14:06:33 <earboxer> since the update, the system bootloops after 'loading initial ramdisk', which led me to try reinstall ramdisk from recovery
2023-05-20 14:07:06 <minimal> what ramdisk? you say you're using Sys mode
2023-05-20 14:09:45 <minimal> using setup-alpine to do the initial installation if you specify Sys mode then setup-alpine uses setup-disk, however if you specify run-from-ram setup-alpine uses setup-bootable instead
2023-05-20 14:10:27 <minimal> so I am confused as to why you would run setup-bootable on a Sys-mode system
2023-05-20 14:12:20 <earboxer> minimal: I haven't run it, I'm wondering if that would help me
2023-05-20 14:12:44 <minimal> ok, no, it is designed for run-from-ram systems
2023-05-20 14:13:16 <minimal> so what exactly is your problem? your boot partition is too small that after running "apk upgrade" mkinitfs runs out of space?
2023-05-20 14:13:50 <earboxer> No, my SSD is going bad
2023-05-20 14:14:04 <minimal> so replace it?
2023-05-20 14:15:05 <earboxer> I need a time machine to tell myself not to buy a macbook 8 years ago
2023-05-20 14:16:14 <minimal> what practical help are you looking for?
2023-05-20 14:18:11 <earboxer> I want to boot my (encrypted lvm ext4) system from an SD card
2023-05-20 14:19:33 <minimal> your encrypted LVM ext4 filesystem that is on a faulty SSD device?
2023-05-20 14:20:05 <minimal> if the device is faulty you should copy whatever you can off it and replace it
2023-05-20 14:23:11 <earboxer> There's a small part of me that says "only the [bytes assigned to] the first partition are bad", but you're probably right
2023-05-20 14:24:05 <minimal> SSDs remap blocks all the time, if it is faulty then you cannot assume only a particular "part" of your data on the SSD is at risk
2023-05-20 14:24:16 <psykose> i'd probably skip the booting it part and just take out the drive and try to get what's possible off it from another computer at this point
2023-05-20 14:24:26 <minimal> if you want to boot from SDcard then install Alpine onto a SDcard...
2023-05-20 14:24:47 <earboxer> Yeah, I got in and made a last backup
2023-05-20 14:58:14 <JohannesJ[m]> hmz, on 3.17 i could do "docker-compose up -d" 
2023-05-20 14:58:30 <JohannesJ[m]> now, on 3.18 that doesnt seem to work, even though docker-compose is installed
2023-05-20 14:58:41 <JohannesJ[m]> now i get -ash: docker-compose not found
2023-05-20 14:58:46 <ikke> JohannesJ[m]: `docker compose`
2023-05-20 14:59:04 <JohannesJ[m]> heh
2023-05-20 14:59:06 <JohannesJ[m]> thanks
2023-05-20 14:59:21 <ikke> docker-compose was rewritten to be a docker plugin instead
2023-05-20 14:59:22 <JohannesJ[m]> why the change? do you know? what did i miss?
2023-05-20 14:59:24 <ikke> that was already a while ago
2023-05-20 14:59:26 <JohannesJ[m]> ahh
2023-05-20 14:59:42 <ikke> docker-compose (v1) was written in python
2023-05-20 14:59:47 <ikke> v2 written in go
2023-05-20 15:00:16 <ikke> officially it can still be executed as docker-compose, but the package would need to include a symlink 
2023-05-20 15:00:46 <JohannesJ[m]> ahh :)
2023-05-20 15:00:52 <JohannesJ[m]> i need to pay attention more :)
2023-05-20 15:05:35 <pj> @Johannes J. because the python version was mess and hard to maintain/extend when moby and docker CLI are both written in Go (with massive amount of other container related libs)
2023-05-20 15:06:30 <JohannesJ[m]> ohh i like go more then python, this is true
2023-05-20 15:07:05 <JohannesJ[m]> i havent paid much attention lately because of private issues. so i missed that obviously :)
2023-05-20 15:07:40 <pj> it was also an opportunity to create better user interface
2023-05-20 15:08:38 <JohannesJ[m]> since most other commands start with docker and a space, why not compose too, heh
2023-05-20 15:13:20 <pj> 
ACTION goes back to migrating compose files to portainer

2023-05-20 15:14:28 <JohannesJ[m]> :)
2023-05-20 15:14:46 <JohannesJ[m]> doesnt portainer import them?
2023-05-20 15:14:56 <JohannesJ[m]> been a while since i played with that
2023-05-20 15:15:03 <psykose> 
ACTION looks at portainer

2023-05-20 15:15:14 <psykose> 
ACTION expands DONOTUSE list by 1 entry

2023-05-20 15:15:18 <pj> I'm migrating from portaienr web editor and the ones I have locally scattered over my server to git
2023-05-20 15:15:24 <JohannesJ[m]> ahh
2023-05-20 15:15:31 <pj> and then redeploying them via portainer from git
2023-05-20 15:16:54 <pj> * reconfiguring stuff and putting it behind traefik
2023-05-20 15:17:16 <JohannesJ[m]> i run everything behind Caddy :)
2023-05-20 15:17:23 <pj> I don't like caddy
2023-05-20 15:17:41 <pj> I'm not a fan of "recompile your webserver to add functionality"
2023-05-20 15:17:52 <JohannesJ[m]> fair enough :)
2023-05-20 15:18:18 <pj> additionally caddyfile is so great and basic it breaks on everything
2023-05-20 15:18:32 <JohannesJ[m]> never had a problem with it
2023-05-20 15:18:42 <pj> I had multiple
2023-05-20 15:18:46 <psykose> recompile is fun
2023-05-20 15:18:59 <pj> but traefik isn't without faults as well
2023-05-20 15:19:08 <pj> like hardcoded dashboard path
2023-05-20 15:19:19 <psykose> well then you better go catch it
2023-05-20 15:19:21 <pj> or the terrible dark theme
2023-05-20 15:19:32 <psykose> are there good dark themes
2023-05-20 15:19:38 <pj> or the advertisement you can't remove
2023-05-20 15:19:45 <JohannesJ[m]> thats for all software :) nothing is perfect, especially if its made by unperfect beeings :)
2023-05-20 15:19:52 <JohannesJ[m]> Alpine Linux comes close though :D
2023-05-20 15:20:11 <pj> I would disagree
2023-05-20 15:20:22 <pj> Alpine is very far
2023-05-20 15:21:03 <JohannesJ[m]> ;-)
2023-05-20 15:28:39 <minimal> Father Ted meme: "This distro is small, those distros are far away!" ;-)
2023-05-20 15:40:24 <JohannesJ[m]> though, withg 3.18 i have some issues i didnt have before :|
2023-05-20 15:43:23 <psykose> which ones
2023-05-20 16:23:44 <JohannesJ[m]> sometimes the screens flash black for a sec :( its a dock thing, but somehow i got it working under 3.17
2023-05-20 16:25:08 <JohannesJ[m]> and sometimes the whole system gets laggy.. to the point where it freezes and i have to force reboot it
2023-05-20 16:27:07 <JohannesJ[m]> nice..
2023-05-20 16:35:11 <jhjacobs[m]> hmz, my matrix needs some work still
2023-05-20 16:35:30 <jhjacobs[m]> in any case, i'll fix the errors with the docking station :) 
2023-05-20 16:39:39 <jhjacobs[m]> playing with matrix now :))
2023-05-20 16:39:49 <jhjacobs[m]> on Alpine, i might add
2023-05-20 17:01:50 <Misthios> "@psykose expands DONOTUSE list by 1 entry" what are the other items on there :)
2023-05-20 17:33:54 <powers> quinq to clarify, I mean for a package that has dependencies on other packages (i.e. linux-headers)
2023-05-20 17:36:35 <powers> To be more specific, even though something like gcc doesn't *need* to be updated, alpine-sdk updates it when you upgrade the package
2023-05-20 18:26:49 <powers> also, can you use an older linux-lts with a newer linux-lts-dev? For example, can I use linux-lts-6.1.27 with linux-lts-dev-6.1.29?
2023-05-20 18:27:55 <powers> No external API seems to have changed, only internals, but i dont know if thats enough to make it incompatible
2023-05-20 18:29:37 <minimal> why would you want to?
2023-05-20 18:49:02 <quinq> I must say that alpine consumes 4W less than the previous distro on that same laptop, that's nice
2023-05-20 19:26:12 <Ermine> quinq: how did you measure this?
2023-05-20 19:44:22 <quinq> I didn't, I just read what /sys/class/power_supply/BAT0/power_now tells me
2023-05-20 20:14:09 <jhjacobs[m]> hmz
2023-05-20 20:22:44 <aron> quinq: which distro was that?
2023-05-20 20:27:32 <quinq> Humm, void
2023-05-20 20:28:00 <quinq> Running another libc version
2023-05-20 20:28:19 <quinq> Maybe other kernel conf too
2023-05-20 20:29:25 <aron> thanks for the info
2023-05-20 20:30:00 <aron> my armv7 systems can run for quite a while from battery, but I was able to compare it with systemd-based systems
2023-05-20 20:30:14 <aron> (systemd takes a log of power)
2023-05-20 20:31:18 <aron> (at least based on what I experienced)
2023-05-20 20:32:53 <aron> s/log/lot/
2023-05-20 20:32:57 <aron> it's late
2023-05-20 20:33:07 <quinq> ;)
2023-05-20 21:03:25 <shafire> hi
2023-05-20 21:03:37 <shafire> why is there no libidl in 3.18?
2023-05-20 21:52:58 <jhjacobs[m]> Hmz, videos showing elementary is dead 🙈 THE KING IS DEAD! LONG LIVE ALPINE!
2023-05-20 22:34:57 <pj> shafire: it's unmaintained, outdated and unused
2023-05-20 22:41:08 <shafire> then I need to identify why virtualbox require that
2023-05-20 22:46:41 <pj> shafire: https://www.virtualbox.org/ticket/21169
2023-05-21 00:40:01 <shafire> pj: thanks
2023-05-21 00:40:08 <shafire> that will be a lot of work :(
2023-05-21 08:49:35 <quinq> Hi
2023-05-21 08:50:09 <lopid> hmmmmmmmmmmm
2023-05-21 08:50:14 <quinq> Today, my fstab sshfs entry doesn't work anymore (it tells me: fuse: unknown option(s): `-o defaults')
2023-05-21 08:50:50 <quinq> I'm trying to see when sshfs / fuse were updated for the last time, but I can't find any info about that with apk info, nor any log about apk operations
2023-05-21 08:51:14 <quinq> Two questions, is such information available with some configuration?
2023-05-21 08:51:38 <quinq> Am I the only one with that fuse issue? ^^
2023-05-21 08:52:50 <lopid> you could try the git repo or build logs: https://pkgs.alpinelinux.org/package/v3.18/main/x86_64/sshfs
2023-05-21 08:53:09 <quinq> Thanks lopid 
2023-05-21 08:53:24 <quinq> That works, though I was thinking about something on apk local side
2023-05-21 08:53:41 <lopid> yeah, i can't remember an xilog equivalent
2023-05-21 08:54:17 <lopid> maybe /var/cache/apk for the packages you've had
2023-05-21 08:55:18 <quinq> pretty empty here :>
2023-05-21 08:55:40 <quinq> But I see that fuse was updated at the start of the week
2023-05-21 08:56:18 <lopid> do you have something that cleans the cache automatically?
2023-05-21 08:56:58 <quinq> I don't think I activated cache
2023-05-21 08:57:37 <lopid> oh i thought it was just how it was by default
2023-05-21 08:57:55 <quinq> Maybe that's that: https://github.com/libfuse/libfuse/commit/dba6b3983af34f30de01cf532dff0b66f0ed6045 :/
2023-05-21 08:59:20 <quinq> Ah no, that's another fuse version
2023-05-21 09:01:26 <quinq> Though it seems there only has been a rebuild, no package update
2023-05-21 09:02:20 <quinq> Maybe that's not the right direction and the culprit is somehow util-linux
2023-05-21 09:14:23 <quinq> Reading sshfs changelog: “This is the last release from the current maintainer. SSHFS is now no longer maintained or developed.”
2023-05-21 09:14:35 <quinq> I see, so the developper decided to sabotage it
2023-05-21 09:15:00 <lopid> :-o
2023-05-21 09:15:41 <quinq> (that's from one year ago though ^^)
2023-05-21 09:41:54 <quinq> So yeah, very likely to be a mount (util-linux) change
2023-05-21 10:12:27 <WhyNotHugo> Is there any mechanism to change the console font during the mkinitfs stage?
2023-05-21 10:12:58 <WhyNotHugo> I'm using a larger-than-default font for the console, but it only kicks in _after_ cryptsetup.
2023-05-21 10:14:09 <lopid> you can change the keymap that grub uses, so you can probably change the font, too
2023-05-21 10:14:27 <lopid> (assuming grub)
2023-05-21 10:15:29 <WhyNotHugo> Not using grub.
2023-05-21 10:16:06 <WhyNotHugo> I'm talking about the linux console anyway, not the bootloader itself.
2023-05-21 10:17:34 <lopid> when you say "after cryptsetup", you mean after entering your luks passwd sucessfully?
2023-05-21 10:18:11 <lopid> or are you at the actual installer?
2023-05-21 13:38:44 <WhyNotHugo> lopid: right now the font is changed after I cryptsetup stage of the boot process. E.g.: after I'm prompted for the passphrase an the main partition is mounted.
2023-05-21 13:39:37 <WhyNotHugo> But I'm wonderig if I can change the console font earlier in the process. E.g.: during the initramfs phase before the cryptsetup prompt.
2023-05-21 14:06:06 <lopid> unfortunately i can't help with how, but i'd be surprised if it's not possible
2023-05-21 14:06:43 <lopid> not specific to alpine, so it'd be worth casting a wider net
2023-05-21 15:20:10 <minimal> quinq: please show your /etc/fstab entry for sshfs
2023-05-21 15:23:23 <quinq> minimal, user@host:path /media/net/ssh/$host/$path sshfs defaults,nosuid,nodev,noatime,noauto,_netdev,user,idmap=user,allow_other,reconnect 0 0
2023-05-21 15:23:47 <quinq> Removing defaults and users fixed the issue triggered by mount upgrade
2023-05-21 15:26:16 <psykose> sounds related to mount 2.39 changes
2023-05-21 15:26:33 <quinq> yeah, I confirmed that :)
2023-05-21 15:26:44 <minimal> quinq: "users"? you mean "user"?
2023-05-21 15:26:48 <quinq> Yes, typo
2023-05-21 15:27:37 <minimal> looks like the "general" option defaults is being passed to the specific sshfs code
2023-05-21 15:27:51 <quinq> It is
2023-05-21 15:28:01 <psykose> that said i'm not sure how that ever worked
2023-05-21 15:28:03 <quinq> Well, not to the code, to the exec
2023-05-21 15:28:13 <psykose> sshfs -o is a mixture of some mount options and some ssh options, defaults was never documented to exist for it
2023-05-21 15:28:14 <quinq> psykose, it has, for years
2023-05-21 15:28:27 <quinq> It wasn't passed before
2023-05-21 15:28:46 <psykose> i'd guess the older mount filtered unknown options somehow
2023-05-21 15:29:00 <quinq> I didn't have time to track the code change though, just looking at the executable though
2023-05-21 15:29:07 <minimal> psykose: isn't "defaults" a generic mount option?
2023-05-21 15:29:48 <quinq> It is yeah, but maybe mount was filtering that out for the sshfs/fuse mount helper for some reason?
2023-05-21 15:29:58 <psykose> kind of
2023-05-21 15:30:28 <minimal> mount manpage has it in the "FILESYSTEM-INDEPENDENT MOUNT OPTIONS" section
2023-05-21 15:30:33 <grawity> export LIBMOUNT_DEBUG=all
2023-05-21 15:32:39 <minimal> quinq: which Alpine release are you on now and what were you on when it was working?
2023-05-21 15:33:04 <quinq> edge
2023-05-21 15:33:14 <quinq> edge
2023-05-21 15:33:42 <quinq> (edge/amd64≥
2023-05-21 15:33:48 <quinq> s/≥/)/
2023-05-21 15:43:44 <minimal> yeah I see some mount option related changes in util-linux 2.39
2023-05-21 15:44:56 <minimal> "Before v2.39 this function ignored options specified by flags (see mnt_context_set_mflags()) before mnt_context_prepare_mount() call. Now this function always returns all mount options."
2023-05-21 15:45:01 <minimal> that might be related
2023-05-21 15:45:32 <minimal> function is mnt_context_get_options
2023-05-21 15:47:28 <psykose> you should probably report that to util-linux
2023-05-21 15:48:44 <minimal> WhyNotHugo: mkinitfs' init does not (currently) support fonts - you could add a new mkinitfs feature to add fonts & related tools to the initramfs and modify the initramfs' init to use them, and raise a MR to add that functionality to mkinitramfs
2023-05-21 15:54:28 <quinq> I can't see anything relevent in the git history related to "defaults" though, so maybe it's also a kernel change
2023-05-21 15:55:20 <minimal> quinq: well find the commit that added the above code comment I mentioned...
2023-05-21 15:57:13 <minimal> as a side note I see the mount manpage recommends "fuse.sshfs" be used rather than "sshfs", not sure if this applies to fstab entries
2023-05-21 15:57:31 <minimal> for the mount "-t" option I mean
2023-05-21 16:03:07 <minimal> quinq: **if** the comment I mention does relate to your problem then here's the relevant commit: https://github.com/util-linux/util-linux/commit/9a72da2c23d0bfdf8781abdf6d1da9e8dd914829
2023-05-21 16:03:34 <quinq> Yeah, I'm (partially) reverting it
2023-05-21 16:03:43 <quinq> Just to see, not as an actual fix
2023-05-21 16:03:57 <quinq> building now
2023-05-21 16:04:58 <quinq> Nope, doesn't seem to fix it.
2023-05-21 16:05:49 <minimal> well there seem to be quite a few options-related changes in util-linux 2.39
2023-05-21 16:06:00 <quinq> jeje yeah, to say the least
2023-05-21 16:06:49 <quinq> Hummm, it seems I'm still subscribed to the util-linux ml
2023-05-21 16:08:12 <minimal> Alpine's fuse package seems to be quite old
2023-05-21 16:09:08 <minimal> 2.9.9 vs 3.6.0 in github
2023-05-21 16:09:19 <quinq> Na, that's fuse3
2023-05-21 16:09:21 <psykose> fuse3
2023-05-21 16:09:47 <minimal> completely different and not just a newer version?
2023-05-21 16:10:26 <uniqNuuid> semiver
2023-05-21 16:11:31 <minimal> ah, I'd forgotten to put a "*" when I searched for "fuse" on pkgs.alpinelinux.org ;-)
2023-05-21 16:18:35 <quinq> psykose, minimal, sent
2023-05-21 16:21:28 <quinq> Well, grey-listed, should land soon
2023-05-21 16:24:01 <quinq> done
2023-05-21 16:27:05 <XFaCiEer> I concur with    your done.
2023-05-21 16:29:31 <quinq> XFaCiEer, what did you just done doing?
2023-05-21 16:34:05 <XFaCiEer> I done doing that   your doing is done.
2023-05-21 16:34:54 <quinq> 
ACTION walks into the setting sun

2023-05-21 16:37:18 <quinq> Actually I have a question, is there some already set process for keeping current kernel environment from being overwritten by a new kernel update on Alpine?
2023-05-21 16:37:59 <quinq> Besides manually copying files, some helper tool of some sort (or something else)
2023-05-21 16:43:49 <psykose> nope
2023-05-21 16:44:32 <eex0> nop nop hello
2023-05-21 16:48:40 <minimal> quinq: what exactly do you mean by "current kernel environment"?
2023-05-21 16:49:55 <quinq> Mostly on-disk modules
2023-05-21 16:50:38 <quinq> (but also optionally actual kernel+initrd)
2023-05-21 16:52:06 <minimal> quinq: the modules are built for a specific kernel so you can't use them with a different (i.e. old) kernel.......or are you referring to have the modules for the *running* kernel remain after an upgrade (until you reboot to run the new kernel)? That has been discussed several times in the past
2023-05-21 16:52:20 <quinq> yes
2023-05-21 16:52:36 <minimal> so it has been discussed, but nothing implemented
2023-05-21 16:53:03 <quinq> And, sorry if I wasn't there for the discussions, what was the gist of the conclusions?
2023-05-21 16:53:22 <quinq> That it could be interesting to have, but somebody has to actually do it?
2023-05-21 16:53:47 <quinq> Or rather we are not interesting in supporting that?
2023-05-21 16:53:51 <minimal> as for having more than a single kernel package installed - either install linux-lfs AND linux-edge ;-)...or else wait to see if multiple versions of a specific kernel package (i.e. linux-lts) is ever implemented - no current likelyhood of that occurring AFAIK
2023-05-21 16:55:06 <minimal> conclusions? that it could possibly be done (for modules) but don't remember anyone actually saying they'd submit an MR to implement it
2023-05-21 16:55:19 <quinq> Allright, thanks
2023-05-21 16:55:50 <minimal> perhaps psykose remembers those discussions better?
2023-05-21 16:56:18 <psykose> the same thing everyone has said for the past 10 years
2023-05-21 16:56:28 <psykose> yes it would be nice to have
2023-05-21 16:56:33 <psykose> no there is no implementation for it
2023-05-21 16:56:37 <quinq> As an example, I could live with having /lib/modules/$kernel untouched and be responsible of cleaning it up
2023-05-21 16:56:44 <quinq> ok, thanks too psykose 
2023-05-21 16:57:46 <uniqNuuid> the stragegys is holding the kernel
2023-05-21 16:59:13 <minimal> uniqNuuid: how does that let you have multiple kernels installed?
2023-05-21 16:59:19 <quinq> Well, for the modules problem, that's one possible solution, but that doesn't provide booting old kernels
2023-05-21 16:59:51 <quinq> Anyway, I suppose all the pros and cons have been discussed, it's an inherent problem
2023-05-21 16:59:56 <uniqNuuid> to boot older kernels. apk fetch as many older kernels you want... 
2023-05-21 17:00:20 <uniqNuuid> everytime there is a kernel, fetch it, save it on your local disk 
2023-05-21 17:00:33 <quinq> Bottom point is that somebody has to implement it if they feel that they really need it :)
2023-05-21 17:00:34 <uniqNuuid> then you can install later 
2023-05-21 17:00:42 <minimal> uniqNuuid: their question was in the context of installing a newer version of the kernel package
2023-05-21 17:01:27 <uniqNuuid> yes? run apk fetch -> save the kernel -> you can downgrade later? 
2023-05-21 17:01:45 <quinq> Hard to downgrade if you can't boot though
2023-05-21 17:02:12 <uniqNuuid> when you can't boot, ISO to rescue
2023-05-21 17:02:24 <quinq> Sure, there's always a solution
2023-05-21 17:03:10 <uniqNuuid> or, you can check if someone made a rollback repo like archlinux did 
2023-05-21 17:03:13 <quinq> There's always a balance in those
2023-05-21 17:32:56 <eex0> I love see'ing dev tiffs .. system76 called out gnome and told to f'off and now their making their own DE cosmic and gnome put in a funny device security (secure boot) is off (no protection and now system76 was forced to change coreboot policy and sign code..lol.. hey, piss on me, I'll piss back..lol
2023-05-21 17:51:01 <quinq> I'm having trouble seeing the link between coreboot and gnome
2023-05-21 17:57:52 <eex0> system76 got on gnome dev's nerves and upset those folks talking smack about gnome DE.. system76 is creating a DE of their own now but in interrim gnome is saying system76 machines are insecure (secure boot) disabled and list several secure events related to having the coreboot system76 uses. long story really.
2023-05-21 17:58:40 <eex0> a history of junk talk iow
2023-05-21 17:59:59 <eex0> system76 refused to sign anything, modules included..
2023-05-21 18:01:07 <eex0> I'll never buy another one of thier paper weights
2023-05-21 18:01:09 <quinq> I still fail to see how secure-boot is any concern of Gnome layer
2023-05-21 18:01:21 <quinq> But if it's just about trash-talking the other, then yeah maybe…
2023-05-21 18:01:36 <ilya> hello?
2023-05-21 18:01:55 <eex0> thats what I meant, it's trash talk
2023-05-21 18:02:02 <eex0> hello ilya
2023-05-21 18:02:03 <quinq> ok :)
2023-05-21 18:02:07 <quinq> ilya, hello?
2023-05-21 18:02:10 <ilya> ahhahahah, hello
2023-05-21 18:02:14 <ilya> quinq, hello
2023-05-21 18:02:16 <quinq> hello eex0 
2023-05-21 18:02:24 <eex0> howdy
2023-05-21 18:03:26 <ilya> nice
2023-05-21 18:03:49 <ilya> does anyone use cli irc client?
2023-05-21 18:04:03 <ilya> i just joined with the first thing i had
2023-05-21 18:04:54 <quinq> I use irssi, like any nqqb
2023-05-21 18:05:11 <abob> oh, i also have irssi
2023-05-21 18:05:21 <abob> oh
2023-05-21 18:05:23 <abob> hello
2023-05-21 18:05:28 <maxim> hey
2023-05-21 18:05:29 <quinq> hello ilya
2023-05-21 18:05:51 <quinq> abob, like any other bob
2023-05-21 18:06:42 <maxim> abob is not bob)
2023-05-21 18:07:57 <quinq> Oh :>
2023-05-21 18:08:28 <maxim> does anyone dualboot alpine with other distros?
2023-05-21 18:11:47 <abob> well, bye then
2023-05-21 18:12:59 <scorpion2185[m]1> yes
2023-05-21 18:14:21 <maxim> with grub?
2023-05-21 18:19:06 <scorpion2185[m]1> yes
2023-05-21 18:20:45 <scorpion2185[m]1> os-proper didn't work nicely
2023-05-21 18:21:15 <scorpion2185[m]1> I like to use refind and open grub from there
2023-05-21 18:21:40 <maxim> i have never used refind, is it good?
2023-05-21 18:23:04 <scorpion2185[m]1> Yes but if you don't dual boot it can be useful to launch efi stub only
2023-05-21 18:23:51 <maxim> oh i see
2023-05-21 18:25:00 <maxim> i am a newbie in alpine and are here as many packages as in arch?
2023-05-21 18:25:59 <quinq> No, Arch Linux is a separate distribution from Alpine Linux, they have their own set of packages, so different count number
2023-05-21 18:26:53 <maxim> bye 
2023-05-21 18:27:00 <quinq> o/
2023-05-21 19:24:38 <psykose> quinq: there was also https://github.com/util-linux/util-linux/issues/2238 as a yet separate issue :D
2023-05-21 19:24:44 <psykose> seems they broke like 4 distinct things
2023-05-21 19:27:53 <quinq> Erf yeah
2023-05-21 19:28:39 <quinq> They were fast to fix the issue I had though, there's already a proposed patch
2023-05-21 19:40:01 <minimal> sounds like their testing was successful lol
2023-05-21 19:41:26 <psykose> quinq: yeah, i backported it2
2023-05-21 19:41:29 <psykose> 2/4 fixed
2023-05-21 19:41:36 <quinq> psykose, that's fast :)
2023-05-21 19:43:33 <Saijin_Naib> Psykose is a beast
2023-05-21 19:44:31 <ikke> 0
2023-05-21 19:45:42 <quinq> 1
2023-05-21 20:36:22 <eex0> you all are genius .. alpine used for desktop, ROCKS! "whoever said musl and alpine are slow" go suck a tail pipe..lol
2023-05-21 20:36:35 <lopid> who said that?
2023-05-21 20:37:04 <eex0> shrugs.. I read it some place
2023-05-21 20:37:41 <eex0> or maybe it was perl or py running on musl was slow.. I can't remember
2023-05-21 20:38:38 <quinq> Bah, there will always be some corner case where something is slower than something else
2023-05-21 20:38:49 <quinq> Doesn't mean that can be a source of general evaluation
2023-05-21 20:39:28 <eex0> I'm at home with alpine, other distro who? 
2023-05-21 20:42:00 <eex0> it took awhile for me to get my head around some things and 20 installs using different options but, this is so distro is solid as a rock!
2023-05-21 20:42:25 <eex0> f' dislexic today
2023-05-21 20:43:26 <quinq> Nice that it fits your needs :)
2023-05-21 20:46:19 <eex0> watched a guy the other day on youtube do an instal w/ gnome desktop, cool video tho he suggested using flatpak, not me, not knowing as much as I do now about canonical ltd. I'll use the kernel in-built apparmor by those canonical folks but not flatpak. I'm not much on cutting corners for coveniences.
2023-05-21 20:47:56 <quinq> That's the spirit
2023-05-21 20:48:06 <eex0> my opinion is as an on-looker is flatpak on a distro means it's ubuntu..
2023-05-21 20:49:04 <eex0> they can go sell someone elses data right quick..lol
2023-05-21 20:49:11 <quinq> :)
2023-05-21 20:51:15 <psykose> TIL alpine is ubuntu
2023-05-21 20:53:02 <eex0> don't get me started, I'll open up a Richard Stallman quick..lol
2023-05-21 20:53:32 <quinq> Let-me interject for a moment
2023-05-21 20:56:15 <eex0> I'd like to think the world is better off w/o folks injecting convenience code that makes everyone insecure.. I fight to try and stay secure the best I can because of that.
2023-05-21 21:00:08 <eex0> quinq?
2023-05-21 21:01:38 <quinq> That was just a mimic
2023-05-21 21:01:46 <quinq> Have a good night :)
2023-05-21 21:02:19 <eex0> lol. night
2023-05-21 21:07:34 <eex0> psykose if systemd is integrated by default, I'm out!
2023-05-21 21:31:07 <abby> too late
2023-05-21 21:31:14 <abby> i put systemd in your breakfast cereal
2023-05-21 21:31:32 <abby> it's integrating with you as we speak
2023-05-21 21:31:45 <lopid> https://imgflip.com/s/meme/Cereal-Guy-Spitting.jpg
2023-05-21 21:32:36 <abby> why does flatpak = ubuntu? ubuntu has snaps
2023-05-21 21:34:34 <eex0> canonical = ubuntu = flatpak 
2023-05-21 21:35:12 <abby> flatpak has nothing to do with canonical
2023-05-21 21:36:35 <abby> the original dev has been a red hatter since before flatpak existed
2023-05-21 21:38:52 <bl4ckb0ne> isnt alpine just a docker distro anyway? why would you need systemd
2023-05-21 21:38:55 <eex0> install flatpak, install something. look for ubuntu, canonical, all the flatpak's I looked at had those names even mozilla firefox and VSCode (microsoft) 
2023-05-21 21:39:33 <psykose> you sound like you do more drugs than me
2023-05-21 21:39:42 <eex0> I'm using it for a desktop
2023-05-21 21:40:18 <eex0> I read lots psykose and that twice you said I'm a drugy.. shameful
2023-05-21 21:43:29 <dwfreed> flatpak and canonical are opposites; canonical even recently took the stance that official flavors are not allowed to install flatpak by default
2023-05-21 21:45:01 <dwfreed> https://9to5linux.com/canonical-future-ubuntu-releases-wont-support-flatpak-by-default
2023-05-21 21:52:25 <eex0> install some flatpaks .. I'll presume they are created on ubuntu
2023-05-21 21:52:46 <dwfreed> they might be *built* on a ubuntu system, but that doesn't mean shit
2023-05-21 21:52:56 <Ermine> drugs are useful for dealing with linux desktop sometimes
2023-05-21 21:53:16 <dwfreed> and the only reason that happens is because Github Actions uses Ubuntu runners by default
2023-05-21 21:54:18 <eex0> like I said, anyone using flatpak is using ubuntu.. I digress
2023-05-21 21:56:01 <psykose> Ermine: pass em over
2023-05-21 21:56:07 <dwfreed> them's some good drugs you're on
2023-05-21 21:59:13 <AnyCreatureMorph[m]> Why is kde plasma Region & Language grayed out?
2023-05-21 22:03:44 <eex0> AnyCreatureMorph[m]: I fixed gnome by editing /etc/locale.conf adding: LANG=en_US.UTF-8 (reason) you'll have to read up on alpine's locale inference.
2023-05-21 22:09:28 <eex0> I'd like to stick around, maybe help out some but the drug heads who
2023-05-21 22:09:47 <eex0> possibly are headed towards homeless
2023-05-21 22:09:51 <eex0> are in charge
2023-05-21 22:10:05 <eex0> good day!
2023-05-21 22:10:44 <psykose> AnyCreatureMorph[m]: probably missing some specific dependency it needs
2023-05-21 22:10:58 <psykose> `apk add lang` is the generic thing that pulls a few things but it's probably not just that
2023-05-21 22:40:00 <AnyCreatureMorph[m]> <psykose> "`apk add lang` is the generic..." <- That fixed it
2023-05-21 22:45:04 <psykose> nice
2023-05-22 01:54:19 <em6> hey, I only have two repos in my list. how can I add "tesing"?
2023-05-22 01:56:32 <em6> my bad, I chenged to edge and now it works
2023-05-22 08:18:49 <kn_> hi, does mount.nfs from nfs-utils not support udp?  as client, whenever i pass proto=udp as option for a nfs3 share, it complains about invalid mount options
2023-05-22 08:34:27 <quinq> kn_, maybe you discovered the current 5th issue with new mount release :)
2023-05-22 08:35:43 <quinq> kn_, can you run your mount command prefixed with: LIBMOUNT_DEBUG=all
2023-05-22 08:35:56 <quinq> And grep for "optstr"
2023-05-22 08:53:23 <kn_> quinq: looks unhelpful.  'LIBMOUNT_DEBUG=all mount -t nfs -o proto=udp  localhost:foo .' reproduces the same failure, i.e. it fails super early parsing the options before doing any nfs
2023-05-22 08:54:56 <kn_> ^ output: https://termbin.com/k1s9
2023-05-22 09:01:12 <kn_> proto=udp seems to be the only culprit, it works if i copy the whole option string from 'mount -t nfs' output of a tcp mount. s/proto=tcp/proto=udp/ then fails.
2023-05-22 09:16:42 <kn> openbsd clients on the same network mount the same nfs3 share via udp just fine.
2023-05-22 09:20:00 <kn> but since alpine's mount already fails to parse proto=udp (like in the dummy localhost command above), this smells like some dumb issue way before network is involved
2023-05-22 09:20:31 <em6> hey, I am trying to start app with X11 forwarding to putty with admin rights, but I get an error message, any suggestions?
2023-05-22 09:22:06 <em6> https://mclo.gs/Uu4o1FW
2023-05-22 09:58:42 <kn> quing: which specific release are you referring to?  util-linux 2.39?
2023-05-22 09:58:51 <kn> quinq even, pardon.
2023-05-22 10:52:38 <kn> i bisected releases, 3.11 and 3.12 mount with proto=udp, >=3.13 fail with proto=udp
2023-05-22 10:55:34 <quinq> Sorry kn, I'm at werk :/
2023-05-22 10:56:58 <kn> no worries :)
2023-05-22 10:57:37 <quinq> kn, maybe mount.nfs tries to mount it as nfsv4, as it has no indication of that, but you need nfsv3, right?
2023-05-22 10:58:38 <quinq> “To mount using NFS version 3, use the nfs file system type and specify the nfsvers=3 mount option.”
2023-05-22 10:58:38 <kn> it should try v4 first, then negotiate to the smallest common version with the server.
2023-05-22 10:58:50 <quinq> udp isn't a valid nfsv4 option
2023-05-22 10:59:16 <grawity> should you even use nfs over udp these days, though?
2023-05-22 11:00:14 <kn> passing vers=3 makes no difference, it always chokes on proto=udp
2023-05-22 11:00:28 <quinq> That's nfsvers
2023-05-22 11:00:47 <grawity> does it recognize the bare 'udp' alias?
2023-05-22 11:00:52 <grawity> i.e. not proto=udp but just udp
2023-05-22 11:01:42 <kn> quinq: vers is compat for nfsvers as per nfs(8)
2023-05-22 11:02:19 <kn> nfs(5) even
2023-05-22 11:02:52 <quinq> ok but better not use compatibility options until you know that the base option works ^^
2023-05-22 11:03:04 <quinq> (just in case something has been dropped and not documented, for example)
2023-05-22 11:03:20 <kn> either way, same bug with vers/nfsvers and udp/proto=udp
2023-05-22 11:06:36 <quinq> I have to get back to the coal mine, but I can try testing tonight locally if you haven't had time to solve it yet (or maybe somebody else will too)
2023-05-22 11:07:47 <kn> with releases bisected, i can now bisect packages and/or kernels to see what exact version broke it
2023-05-22 11:11:42 <kn> i see tags/releases for x.y.z but binary packages are grouped by x.y, that means booting e.g. 3.13.0 ends up installing 3.13.12, no?
2023-05-22 11:14:40 <kn> installint 3.13.0 from jan 21 and 3.13.12 from aug 22 both ended up with the same 3.13.12 in /etc/alpine-release and checksum for the mount program, which confused me
2023-05-22 11:14:56 <kn> so i probably have to build packages myself if i want to bisect .z minor versions, right?
2023-05-22 11:33:44 <ikke> kn: we don't keep older versions of packages, so yes 
2023-05-22 11:58:30 <kn> ikke: thanks
2023-05-22 13:04:25 <minimal> kn: just to clarify, are you using "mount" from util-linux or "mount" from busybox?
2023-05-22 13:20:28 <minimal> kn: you said it worked for Alpine 3.12.x but failed for 3.13.x? both 3.12.12 and 3.13.12 come with the same version of util-linux (and therefore its' mount)
2023-05-22 13:57:05 <j_v> with iptables change to xtables-nft-multi, is the checkkernel function in /etc/init.d/iptables correct? /proc/net/ip_tables_names not exist and nf_tables modules is loaded. should checkkernel() be like the one from /etc/init.d/nftables?
2023-05-22 14:01:02 <j_v> at boot, iptables starts up fine, at shutdown, iptables stop fails on checkkernel
2023-05-22 14:54:51 <kn> minimal: whatever mount is shipped in a default install after only running 'apk add nfs-utils'
2023-05-22 14:58:21 <minimal> kn: try "which mount" and then "ls -l <whichever path for mount>" - does it link to /bin/bbsuid?
2023-05-22 14:58:55 <minimal> or try "mount --help" at look at the first few lines out its output
2023-05-22 16:40:36 <donoban> I have an alpine VM that used to work fine but now once I start gdm or sddm it seems that not recieves any input from my VNC client (remmina)
2023-05-22 16:41:11 <donoban> if I stop them I can write commands on the tty, so it doesn't look a problem with the VNC
2023-05-22 17:26:24 <j_v> Ok, well can't stay on all day. Pointed it out, will open an issue when I have time and after I research a bit more.
2023-05-22 17:42:35 <kn> minimal: it is busybox
2023-05-22 17:47:58 <kn> but mount(8) is just the wrapper, mount.nfs(8) yields the error, i.e. 'mount.nfs -o udp ...' is enough and doesn't involve either util-linux or busybox mount
2023-05-22 17:51:48 <kn> so i'll bisect nfs-utils between 3.12.12 and 3.13.12 now
2023-05-22 17:52:15 <quinq> I'm getting weirdness here
2023-05-22 17:52:16 <quinq> $ mount.nfs remote:/ /mnt/ -v
2023-05-22 17:52:16 <quinq> mount.nfs: failed to apply fstab options
2023-05-22 17:52:25 <quinq> I don't have any nfs fstab entry
2023-05-22 17:58:43 <quinq> Ah, it needs root access, otherwise strange error message
2023-05-22 17:59:22 <quinq> kn, mount.nfs remote:/ /mnt/ -v -o udp
2023-05-22 17:59:25 <quinq> That doesn't complain
2023-05-22 17:59:38 <quinq> Well, it complains that remote doesn't exist, but that's after udp has been parsed
2023-05-22 18:08:27 <minimal> kn: what was the *exact* error message you were seeing?
2023-05-22 18:08:54 <kn> # mount.nfs localhost:/ / -o udp
2023-05-22 18:08:54 <kn> mount.nfs: an incorrect mount option was specified for /
2023-05-22 18:10:09 <kn> same message, whether it is mount or mount.nfs, udp or proto=udp
2023-05-22 18:16:00 <minimal> kn: I thought you were trying to mount a NFSv3 export
2023-05-22 18:16:28 <kn> minimal: i am, sorry for the too minimized reproducer
2023-05-22 18:17:02 <kn> so strace shows mount(2) returning EINVAL, souds like mount.nfs is happy with the optstr but the kernel is not.
2023-05-22 18:17:57 <kn> i'm running -virt in a vm, was udp nfs perhaps dropped from this kernel config?
2023-05-22 18:21:10 <minimal> kn: yes EINVAL is what triggers mount.nfs to log "an incorrect mount option was specified"
2023-05-22 18:22:14 <minimal> kn: re "too minimised reproducer", it's best to be consistent in what you're testing as there could be multiple (whether related or unrelated) issues
2023-05-22 18:22:32 <eex0> discovery: edited: /etc/network/if-up.d/dad added: macchanger -r $IFACE to: has_flag() .. new mac on every ifup .. for interested parties.
2023-05-22 18:23:46 <minimal> kn: look at kernel config for Alpine Edge: CONFIG_NFS_DISABLE_UDP_SUPPORT=y ;-)
2023-05-22 18:26:17 <minimal> that's on linux-virt 6.1.29-0-virt
2023-05-22 18:26:24 <kn> ah linux-edge, i was looking at main/linux-lts as there's a config for virt as well and i saw the failure on stable/non-edge
2023-05-22 18:27:48 <minimal> no, not linux-edge, Alpine Edge with linux-virt
2023-05-22 18:29:22 <kn> i'm referring to the dirs/files in the aports repo, where i grepped the source
2023-05-22 18:29:38 <kn> community/linux-edge/config-edge.x86_64:CONFIG_NFS_DISABLE_UDP_SUPPORT=y
2023-05-22 18:30:12 <minimal> but that's linux-edge, I thought you were using linux-virt?
2023-05-22 18:31:52 <kn> i am using linux-virt, but the config knob you mentioned only turns up in community/linux-edge/ and testing/linux-{asahi,elm,gru}/
2023-05-22 18:32:25 <minimal> those config files are NOT full configs, they are "deltas" against the default config for that particular kernel
2023-05-22 18:32:48 <minimal> check /boot/config-virt on the machine Alpine machine you are testing on
2023-05-22 18:32:58 <kn> that explains the confusion, thanks, i'm not familiar with the alpine's build infrastructure
2023-05-22 18:34:15 <kn> ack, /boot/config-virt:CONFIG_NFS_DISABLE_UDP_SUPPORT=y
2023-05-22 18:36:53 <minimal> kn:L https://github.com/torvalds/linux/commit/b24ee6c64ca785739b3ef8d95fd6becaad1bde39
2023-05-22 18:37:30 <minimal> looks like it arrived in 5.6 kernel
2023-05-22 18:38:22 <kn> was just looking at it: https://github.com/torvalds/linux/blame/master/fs/nfs/Kconfig#L203
2023-05-22 18:38:34 <minimal> Alpine 3.12 is on 5.4, Alpine 3.13 is on 5.10.....so there you go
2023-05-22 18:41:28 <quinq> ohhh…
2023-05-22 18:42:05 <quinq> good catch
2023-05-22 18:43:13 <minimal> kn: so you could raise an Alpine Issue or MR to try and have this setting changes in linux-lts/linux-virt
2023-05-22 18:49:11 <kn> first i'll try a local kernel with that reenabled to see if it actually fixes my issue, only then i'd consider taking this to alpine upstream
2023-05-22 18:49:49 <eex0> decrypt https: https://www.trickster.dev/post/decrypting-your-own-https-traffic-with-wireshark/  .. for interested parties
2023-05-22 18:49:52 <minimal> well UDP isn't going to work at all without that changed, regardless of whether there may or may not be any other additional issues ;-)
2023-05-22 18:50:33 <minimal> eex0: alpine-offtopic channel perhaps?
2023-05-22 18:52:22 <kn> minimal: sure, but if udp doesn't help me, pushing it upstream is careless and possible dangerous for others (linux commit disabling it mentioned data corruption)
2023-05-22 18:54:04 <eex0> can shows https traffic of any application installed seemed on-topic to debug but yes, I'll drop it on offtopic, thank you
2023-05-22 19:14:18 <kn> is there a way to build only the -virt flavor of main/linux-lts?
2023-05-22 19:15:22 <ikke> kn: remove the -lts configs from the source
2023-05-22 19:15:46 <ikke> though, not sure if other things would complain
2023-05-22 19:41:18 <kn> ikke: didn't work.  now i just removed the -virt bits except for its config files, then i did s/lts/virt/ and removed the lts config files. now it definitely only builds -virt (:
2023-05-22 19:41:32 <ikke> heh
2023-05-22 21:19:13 <frag> can bluetooth keyboard/mouse be set up automatically from a script?
2023-05-22 21:21:27 <frag> i mean paired/connected to the specified adr
2023-05-22 21:26:48 <eex0> maybe wireplumber?
2023-05-22 21:44:05 <dwfreed> eex0: wireplumber is for audio
2023-05-22 21:44:38 <quinq> Have you never tried talking into your mouse?
2023-05-22 21:45:42 <kn> i'm sure many shouted at it ;)
2023-05-22 21:46:00 <dwfreed> frag: bluez's bluetoothctl can be used from scripts without issue
2023-05-22 21:46:47 <quinq> kn ;)
2023-05-22 21:47:17 <eex0> smh, if you need an enemy, thats fine but when wireplumber can control the use of bluetooth, audio or not I threw that out there. troll someone else
2023-05-22 21:47:30 <dwfreed> you're the one trolling
2023-05-22 21:47:48 <eex0> blocked!
2023-05-22 21:57:52 <frag> dwfreed: you would have to put the kb into pairing mode, and type in the pin i suppose(?)
2023-05-22 21:59:02 <dwfreed> depends on if the keyboard has a pin requirement; but yeah, in generally anything that's got a pin exchange requirement isn't going to be able to be completely automated
2023-05-22 22:01:28 <frag> maybe pair and backup the files where its stored? where is that btw :>
2023-05-22 22:03:51 <dwfreed> frag: https://github.com/bluez/bluez/blob/master/doc/settings-storage.txt
2023-05-22 22:10:58 <frag> dwfreed: ++
2023-05-22 23:05:17 <kn> udp nfs works, now for battle testing :)
2023-05-23 06:09:56 <powers> two questions
2023-05-23 06:10:35 <powers> 1. how would one manually sign a kernel-module in alpine?
2023-05-23 06:12:21 <powers> 2. how can i find the kernel config that determines if modules are signed on install? Its called "CONFIG_MODULE_SIG_ALL" and its not in sysctl
2023-05-23 06:16:14 <powers> ive realized this is more of a dev question, so i also asked there, but please feel free to answer
2023-05-23 06:31:15 <powers> ive found the answer to question-#1, linux-lts-dev package /scripts/sign-file. The apk search feature is really strong :)
2023-05-23 06:53:28 <grawity> the configs are not in sysctl in the first place, zgrep MODULE_SIG /proc/config.gz
2023-05-23 06:57:31 <ikke> That's what you get for criss posting 
2023-05-23 06:57:46 <ikke> That was already suggested in -dev
2023-05-23 06:57:59 <ikke> cross posting*
2023-05-23 06:58:29 <powers> yeah that was my bad
2023-05-23 06:58:52 <powers> i asked the question here, forgetting there was a -dev
2023-05-23 07:01:15 <powers> thanks again for the help though, saved me a ton of time
2023-05-23 07:20:54 <raspbeguy> I found out a solution for changing GDM login screen keymap https://social.gugod.fr/@raspbeguy/statuses/01H0X6BBYQBE6P07DDMBYNJVHA
2023-05-23 09:07:22 <WhyNotHugo> It's sad how GNOME is deliberatly trying to be less portable, but mad respect for the folks who keep figuring out how work around that.
2023-05-23 12:34:31 <WhyNotHugo> Can I use passwd to edit /etc/{passwd,shadow} on an alternative location? I basically have the disk for another devices plugged in and need to reset a forgotten password.
2023-05-23 12:34:54 <WhyNotHugo> I don't see any flags that would allow that... maybe there's some workaround?
2023-05-23 12:40:41 <WhyNotHugo> mkpasswd was the answer 
2023-05-23 12:41:22 <Ermine> WhyNotHugo: you can chroot as well.
2023-05-23 12:41:55 <WhyNotHugo> Different architecture
2023-05-23 12:42:34 <Ermine> Oic
2023-05-23 18:41:16 <raspbeguy> hello, I am thinking about getting a raspberry pi zero w and install alpine on it, is there any gotchas?
2023-05-23 19:45:38 <Apachez> yeah
2023-05-23 19:45:55 <Apachez> protip: use the ram edition of alpine where you run a command to sync down to the microsd
2023-05-23 19:46:13 <Apachez> and use an industrial grade microsd, personally I prefer the samsung industrial series for extended write cycles
2023-05-23 19:46:42 <Apachez> by using the "ram edition" you wont wear out that microsd compared to a regular linux installation
2023-05-23 19:47:08 <Apachez> since the only writes will be when you sync the directory to the permanent storage and if you prefer also keep logs there
2023-05-23 20:08:21 <vlp> Hello, does anyone deployed Ceph on Alpine ? I just started a manual deployment but I had to fix the rc file in order to add the --name parameter to my monitor
2023-05-23 20:08:51 <vlp> I would prefere a solution without modify the init script
2023-05-23 20:13:12 <minimal> vlp: modify the /etc/conf.d/ceph file?
2023-05-23 20:13:41 <minimal> there's a RADOSGW_WANT_NAME_PARM setting
2023-05-23 20:14:39 <vlp> y but this param is for RADOSGW not for a moniotor ? 
2023-05-23 20:14:53 <quarkyalice> why can't I set the locale on musl?
2023-05-23 20:16:44 <ikke> quarkyalice: what are you trying?
2023-05-23 20:17:03 <quarkyalice> adding it to /etc/locale.gen and running locale-gen
2023-05-23 20:18:01 <quarkyalice> hexchat just crashed
2023-05-23 20:18:13 <minimal> vlp: don't know, I don't use ceph. If that doesn't do what you want then work out changes to conf.d and init.d files and submit a MR to "fix" the ceph package...
2023-05-23 20:18:51 <minimal> quarkyalice: musl has limited locale support
2023-05-23 20:19:42 <ikke> There is https://pkgs.alpinelinux.org/package/edge/main/x86/musl-locales
2023-05-23 20:20:20 <minimal> quarkyalice: /etc/profile.d/20locale.sh defines some settings
2023-05-23 20:20:33 <quarkyalice> thanks!
2023-05-23 20:26:04 <psykose> there is also no such thing as locale.gen or locale-gen
2023-05-23 20:26:07 <psykose> on musl
2023-05-23 20:26:15 <psykose> you just need LANG= and that's about it
2023-05-23 20:26:40 <psykose> musl-locales for the libc translation specifically (not many) or just apk add lang for whatever ships translations
2023-05-23 21:06:36 <irreleph4nt> Hi. I am new to Alpine (but not new to Linux), trying to make use of it in a Dockerfile at work. I need help with solving the following problem, please:
2023-05-23 21:07:00 <irreleph4nt> The company I work for maintains an internal APK mirror. But whilst the packages get updated, the APKINDEX file may not. This leads to my problem: Trying to install perl-git, apk reports that it can't find the package in version 2.36.3 which is listed in APKINDEX on the mirror, because the actual APK version on the mirror is 2.36.6 already ...
2023-05-23 21:07:20 <irreleph4nt> I have tried solving this by creating a local APK repo which contains version 2.36.6 and all dependencies. The docker build, explicitly asking for 2.36.6 to be installed from the local repo fails, claiming that "can't select package 2.36.3 because it breaks world 2.36.6"
2023-05-23 21:07:35 <irreleph4nt> Do you have any suggestions on what I can do to solve this, please?
2023-05-23 21:08:17 <psykose> update the apkindex file
2023-05-23 21:08:34 <psykose> not sure how you would expect that to work otherwise lol
2023-05-23 21:08:40 <irreleph4nt> I would if I could but I can't. I am not the person controlling the mirror
2023-05-23 21:08:45 <psykose> then tell the people that are
2023-05-23 21:09:11 <psykose> you can do some hacky workarounds of `fetch the apk; apk add ./file.apk` i guess
2023-05-23 21:10:26 <irreleph4nt> I'll give that a try, thanks!
2023-05-23 21:11:49 <irreleph4nt> Is there a reason though for APK to behave this way? I mean my local mirror has the higher package version, I am explicitly asking for it with perl-git=2.36.3 --repository=<local> but it is still complaining it can select v 2.36.3. That doesn't really make sense to me.
2023-05-23 21:11:59 <irreleph4nt> my local repo*
2023-05-23 21:12:17 <irreleph4nt> it can't select v 2.36.3*
2023-05-23 21:12:53 <ikke> irreleph4nt: maybe something else keeping it back at 2.36.3
2023-05-23 21:15:46 <irreleph4nt> ikke: good point. I am building off a minirootfs and the image shows a total of 30 packages when running "apk list --installed". These include nginx and openssh-client, but I don't see anything stricking me as dependent on perl o.O
2023-05-23 21:16:40 <irreleph4nt> Does apk have something like a verbose switch to make it a bit more chatty about it's reasons to throw errors?
2023-05-23 21:17:34 <psykose> your local mirror doesn't have a higher package version if the apkindex doesn't say it does
2023-05-23 21:18:59 <psykose> "can't select package 2.36.3 because it breaks world 2.36.6" means you tried to add 2.36.6, the apkindex only has 2.36.3, and it can't select 2.36.3 from the repository because you specified 2.36.6 and not 2.36.3, and only 2.36.3 is in the apkindex, so it can't add what you told it to
2023-05-23 21:19:11 <irreleph4nt> let me download the mirror's apkindex and compare it to the one I created for my repo real quick ...
2023-05-23 21:19:38 <ikke> You would have to generate your own index, but I kind of assumed you did
2023-05-23 21:26:55 <irreleph4nt> ikke: I did create my own index for the packages in my local repo
2023-05-23 21:27:53 <irreleph4nt> psykose: Not correct. I checked and my index says P:perl-git, V:2.36.6-r0 whilst the mirror's index says P:perl-git, V:2.36.3-r0
2023-05-23 21:28:22 <psykose> where did you put your index and what did you add to /etc/apk/repositories
2023-05-23 21:30:03 <irreleph4nt> psykose, when I ran "apk index -vU -o APKINDEX.tar.gz *.apk" it registered the perl-git apk as "noarch" whilst the mirror lists it under x86_64 ... that's the only difference I can spot that may matter
2023-05-23 21:30:30 <ikke> --rewrite-arch
2023-05-23 21:31:13 <irreleph4nt> psykose, I ran the apk index command above, then realized three apks were identified as noarch, so I created the folders perl/x86_64 and perl/noarch, moving the apks to their respective folders
2023-05-23 21:31:30 <irreleph4nt> then I ran the apk index command again inside both folders, so both folders contain an apkindex file
2023-05-23 21:31:36 <psykose> that doesn't work, you need --rewrite-arch and one folder
2023-05-23 21:31:49 <irreleph4nt> let me give that a try real quick
2023-05-23 21:31:51 <irreleph4nt> thanks!
2023-05-23 21:32:05 <eex0> reminded me of multiarch for a moment
2023-05-23 21:32:47 <ikke> apk will just use the hardcoded arch to fetch packages from
2023-05-23 21:33:11 <ikke> (or whatever /etc/apk/arch contains)
2023-05-23 21:40:15 <irreleph4nt> psykose, that worked! thank you so much! :)
2023-05-23 21:40:18 <psykose> :)
2023-05-23 21:40:57 <psykose> now tell your office to fix their repo :p
2023-05-23 21:41:00 <psykose> how does anyone use it?
2023-05-23 21:41:10 <irreleph4nt> So do noarch packages exist for alpine? if I understand ikke correctly, the don't really as noarch is not an architecture in and of itself
2023-05-23 21:41:18 <psykose> they don't no
2023-05-23 21:41:43 <ikke> Yeah, in practice they don't exist
2023-05-23 21:41:44 <psykose> it was an abuild thing to start tagging some noarch stuff as noarch but it gets rewritten for the index and apk itself only handles one arch string
2023-05-23 21:41:57 <psykose> it does work with multiple indexes and symlinks though, theoretically
2023-05-23 21:41:59 <irreleph4nt> psykose, big company, so digging up a name for this will take a while. Meanwhile my own co-workers can proceed with our Alpine POC :)
2023-05-23 21:42:44 <psykose> i.e. you can have a noarch/ folder, put stuff in it, then in the x86_64 and aarch64 folder symlink to noarch and apk index --rewrite-arch on the symlinks and the index works and allows installing the noarch packages later (which also resolve over http fine)
2023-05-23 21:42:47 <psykose> i think i tested that before
2023-05-23 21:42:56 <psykose> there's more complications to deploy that in practice of course :p
2023-05-23 21:52:19 <irreleph4nt> Does anyone know off a pre-compiled apk for the nginx_stream_lua module for Alpine? Googling for it all I can find is Dockerfiles which compile whole nginxes. That's quite a bit more than what I need
2023-05-23 23:52:56 <umuench> Hello :)
2023-05-23 23:53:28 <umuench> Is here anyone who has a running wsl2 "Distro"?
2023-05-23 23:54:23 <umuench> How can I upgrade from 3.17 to 3.18 with Alpine from Windows Store?
2023-05-23 23:54:54 <umuench> Changing the repositories not working by me :(
2023-05-24 00:36:16 <minimal> umuench: maybe ask whomever looks after the Alpine WSL Microsoft Store package? That seems to link to github.com/agowa/WSL-DistroLauncher-Alpine
2023-05-24 00:50:14 <Ermine> I'd upgrade conventionally, like a real vm
2023-05-24 01:47:48 <dani-g5x[m]1> Hello there Alpine users
2023-05-24 04:07:03 <Saijin_Naib> It should just work.
2023-05-24 04:07:03 <Saijin_Naib> I can try and submit a PR to update it later, but yeah, it's just the base tarball for new installation, you can always change branches by editing APK repositories and doing an update and upgrade like usual 
2023-05-24 04:07:03 <Saijin_Naib>  
2023-05-24 04:08:32 <Saijin_Naib> 
ACTION uploaded an image: (89KiB) < https://matrix.org/_matrix/media/v3/download/matrix.org/KXEWmoccISEZnNCgGHSWFRaa/7mwtwr.jpg >

2023-05-24 07:45:59 <qaqland[m]> sudo rc-service tailscale status... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/ZFyZsHyDCUWwNZVfHWGyyQJO>)
2023-05-24 07:47:13 <qaqland[m]> Is there sth wrong? (x86_64 3.18)
2023-05-24 07:48:40 <psykose> works4me
2023-05-24 07:48:42 <Asmadeus> Is there a list of system users with their fixed id somewhere? all users aren't fixed but some like avahi are, others aren't, I'm not quite sure if there's any defined rule for this
2023-05-24 07:48:47 <ikke> Is there something in /var/log/messages?
2023-05-24 07:49:04 <ikke> Asmadeus: cat /etc/passwd
2023-05-24 07:49:30 <ikke> Everything bellow id 1000 is a system user
2023-05-24 07:49:32 <Asmadeus> Also, packages that have a adduser in their pre-install should depend on alpine-baselayout-data otherwise when installing in an empty root (with e.g. https://github.com/alpinelinux/alpine-make-rootfs ) it's possible the pre-install script runs too early and adduser fails
2023-05-24 07:49:41 <Asmadeus> ikke: I meant reserved by packages
2023-05-24 07:50:07 <psykose> that would be a quite silly amount of things to add dependencies to
2023-05-24 07:50:09 <ikke> We don't have a reservation system 
2023-05-24 07:50:11 <Asmadeus> for example https://gitlab.alpinelinux.org/alpine/aports/-/tree/master/main/avahi/avahi.pre-install explicitely requests id 86
2023-05-24 07:50:35 <Asmadeus> psykose: well, just failed for me on a new package I added, so I'll add it to my internal package at least :P
2023-05-24 07:50:42 <psykose> it makes more sense to 'install baselayout first then the rest' i guess
2023-05-24 07:50:44 <Asmadeus> might be possible to have abuild automate that though
2023-05-24 07:50:50 <Asmadeus> or that yeah
2023-05-24 07:51:28 <qaqland[m]> psykose: reboot and works 😛
2023-05-24 07:51:40 <Asmadeus> ikke: so if another package comes and decide they also want 86 it's up to reviewers to catch it? fair enough :P
2023-05-24 07:51:40 <ikke> Yeah, it's assumed to be present. setup-alpine explicitly adds it to world 
2023-05-24 07:52:18 <Asmadeus> hmm I guess I'm already patching alpine-make-rootfs, I can change it to install things in two steps to avoid the problem more globally
2023-05-24 07:52:24 <psykose> Asmadeus: i think the avahi ones are more the exception
2023-05-24 07:52:40 <psykose> usually things don't hardcode any id and just install whatever (so i actually don't know if it's stable at all, probably not)
2023-05-24 07:52:48 <Asmadeus> Right, now I've looked there aren't so many, I just happened to look at avahi first for example
2023-05-24 07:52:50 <psykose> so for that exception, no there's no list anywhere
2023-05-24 07:52:56 <psykose> www-data is another
2023-05-24 07:53:06 <psykose> i think anything shared between more than one package is probably hardcoded
2023-05-24 07:53:08 <psykose> hm
2023-05-24 07:53:54 <ikke> psykose: isn't www-data already present in baselayout?
2023-05-24 07:54:08 <psykose> it is, what about it
2023-05-24 07:54:37 <Asmadeus> I guess we can remove things like apache2/apache2.pre-install:3:addgroup -S -g 82 www-data 2>/dev/null
2023-05-24 07:54:46 <Asmadeus> (is what I think he meant)
2023-05-24 07:54:48 <psykose> ah
2023-05-24 07:55:30 <psykose> it doesn't seem to have it
2023-05-24 07:55:48 <psykose> er
2023-05-24 07:55:49 <psykose> wrong file
2023-05-24 07:55:50 <psykose> it does
2023-05-24 07:56:07 <psykose> why is there a pre-install for shadow and a group file with shadow in it already
2023-05-24 07:56:57 <Asmadeus> Ok so, 1/ change alpine-make-rootfs to install alpine-base first, 2/ fix my ugly scripts that copy systems over to copy users by name instead of uid as this is going to blow up on me someday... :D
2023-05-24 07:57:30 <psykose> my uid of 666 is now reserved in your system
2023-05-24 07:58:34 <Asmadeus> hope you don't have data in /var then
2023-05-24 08:00:30 <psykose> :D
2023-05-24 08:01:22 <satmd> but is that user part of group wheel?
2023-05-24 08:02:48 <psykose> only got a `tyre' sadly
2023-05-24 08:11:09 <satmd> thanks again for helping with fdreissig's chrome packaging bug recently! :)
2023-05-24 08:26:33 <Asmadeus> if anyone cares, alpine-make-rootfs update to install base packages first (requested by default): https://github.com/alpinelinux/alpine-make-rootfs/pull/26
2023-05-24 08:27:36 <Asmadeus> (any reason that's not on the gitlab btw? not alpinelinux looks official-ish enough to me.. ah, https://gitlab.alpinelinux.org/alpine/infra/infra/-/issues/10680 lists it)
2023-05-24 08:28:26 <Asmadeus> -not
2023-05-24 08:28:53 <ikke> It's maintained by jirutka, who prefers github
2023-05-24 08:30:21 <psykose> time 2 put mkinitfs on bitbucket and abuild in azure repos
2023-05-24 08:31:23 <abby> and aports on forgejo
2023-05-24 08:32:08 <psykose> abbygit2
2023-05-24 08:32:52 <Misthios> and something else on sourceforge? 
2023-05-24 08:34:29 <psykose> wow this really forged my source
2023-05-24 09:05:52 <ovf> really old sources ought to be hosted in fossil, naturally
2023-05-24 09:12:15 <lucidiot> how about pastebin
2023-05-24 12:32:16 <jvoisin> no love for bazaar?
2023-05-24 13:07:12 <pj> Migrate to launchpad
2023-05-24 16:29:00 <herkulessi[m]> How much do I break when I Ctrl+C apk during a major update? I would restart it a few seconds later, I just want to move it into a tmux session...
2023-05-24 17:25:56 <powers> does anyone know where the `apple_properties` driver is?
2023-05-24 17:26:46 <powers> it not in alpine's base-system module and im having trouble finding it through apk
2023-05-24 17:27:37 <powers> for reference `https://github.com/torvalds/linux/blob/master/drivers/firmware/efi/apple-properties.c`
2023-05-24 17:28:07 <powers> bad link: use this instead https://github.com/torvalds/linux/blob/master/drivers/firmware/efi/apple-properties.c
2023-05-24 17:35:35 <Saijin_Naib> <herkulessi[m]> "How much do I break when I Ctrl..." <- Nothing should break, because it should be atomic, but I've anecdotally had issues. Nothing APK fix didn't solve. I believe it was because I stopped it during a post install step. 
2023-05-24 17:48:43 <ovf> powers: it's built into the linux-edge kernel (CONFIG_APPLE_PROPERTIES=y), and just doesn't exist in the linux-lts kernel.
2023-05-24 17:50:16 <ovf> to find out, look at Kconfig/Makefile in the kernel source directory containing the file of interest to find the config option, then check against alpine's build system, e.g. https://git.alpinelinux.org/aports/tree/community/linux-edge/config-edge.x86_64 for edge
2023-05-24 17:51:56 <herkulessi[m]> <Saijin_Naib> "Nothing should break, because it..." <- Well, good to know... I just waited, and it finished before my ssh session died...
2023-05-24 17:53:29 <herkulessi[m]> On a related note: the Update broke my podman, because after the update crun was missing... Had me panicking for a second until everything came back up after manually installing crun...
2023-05-24 18:37:59 <powers> ovf thank you for the info
2023-05-24 20:07:44 <msiism> I'm trying to install Alpine with full disk encryption. However, running `cryptsetup luksFormat … /dev/sda1` tells me "Device /dev/sda1 does not exist or access denied." Not sure what to make of that.
2023-05-24 20:08:37 <msiism> I've just created a fresh DOS partition table with sda1 as the sole partition on that disk.
2023-05-24 20:09:38 <umuench> Hey :)
2023-05-24 20:17:21 <msiism> Here's a verbatim copy of what I did: http://termbin.com/eaat
2023-05-24 20:17:36 <msiism> This is on an Alpine live system.
2023-05-24 20:25:56 <msiism> Could this have anything to do with fdisk suggesting to start the partition at 2048 cylinders and me accepting that?
2023-05-24 20:27:17 <ikke> no
2023-05-24 20:27:34 <ikke> that's normal
2023-05-24 20:29:03 <msiism> Okay, good to know.
2023-05-24 20:52:05 <msiism> Well, indeed sda1 does not exist in /dev/. How would I make the system aware of it?
2023-05-24 23:22:38 <minimal> msiism: first of all, why are you not using setup-alpine/setup-disk to install Alpine? it can use LUKS Encryption if you select that?
2023-05-24 23:23:02 <minimal> Secondly, what exactly is /dev/sda? what does "dmesg" show for it?
2023-05-25 04:48:21 <uniqNuuid> LUKS is another complexity layer
2023-05-25 08:55:03 <powers> i would like to patch wpa_supplicant, however i am missing some details
2023-05-25 08:56:43 <psykose> what's there to patch
2023-05-25 08:57:17 <powers> 1. is there a way to extract source, patch, and rebuild an apk you have?
2023-05-25 08:57:27 <dwfreed> https://xyproblem.info/
2023-05-25 08:57:34 <dwfreed> apks do not contain source
2023-05-25 08:58:16 <powers> psykose it's a parculiar bug, but im getting `CTRL-EVENT-SCAN-FAILED ret=-22 retry=1` when running wpa_supp
2023-05-25 08:58:45 <dwfreed> $ errno 22
2023-05-25 08:58:45 <dwfreed> EINVAL 22 Invalid argument
2023-05-25 08:58:49 <powers> this bug, with a patch-fix has been documented here https://lists.infradead.org/pipermail/hostap/2022-January/040178.html
2023-05-25 08:59:45 <powers> dwfreed thanks for the info. is there a place to get apk-source code directly?
2023-05-25 09:01:20 <psykose> that link directly has no fix in it, that's a random person doing random things in some code
2023-05-25 09:01:38 <psykose> does go on to more interesting things tho
2023-05-25 09:01:49 <powers> psykose just to follow up on the bug report, `iw ... scan` is functional for my driver and someone else ran into the issue and used this patch to fix it here https://forums.gentoo.org/viewtopic-p-8713964.html
2023-05-25 09:02:34 <powers> psykose yeah its not an official-patch, or a longterm fix. more of a hotfix
2023-05-25 09:03:07 <dwfreed> there's a long term fix later in the thread
2023-05-25 09:04:06 <psykose> there's not, it's just yet some other patch
2023-05-25 09:04:12 <psykose> 'from debian'
2023-05-25 09:04:17 <psykose> nothing is commited to wpa_supplicant
2023-05-25 09:04:43 <ikke> powers: you would clone aports, include the patch in the relevant APKBUILD file, and then build ut from there 
2023-05-25 09:09:50 <powers> ikke thank you. i would like to clarify some things
2023-05-25 09:11:26 <powers> in APKBUILD, where it reads source=`https://w1.fi/releases/wpa_supplicant-$pkgver.tar.gz....`, i would replace that with my local patched source?
2023-05-25 09:13:05 <ikke> You'd add .patch files there which get automatically applied 
2023-05-25 09:13:38 <powers> yes, thanks, thats crystal clear
2023-05-25 09:15:05 <powers> one more question, may i have an example of what a well put-together patch-file looks like?
2023-05-25 09:15:57 <powers> i know im asking for a silver-spoon at this point, but im going to likely distribute this and want to do it properly
2023-05-25 09:19:32 <powers> or a better angle, would you consider this a good patch-file? https://git.alpinelinux.org/aports/tree/main/wpa_supplicant/unsafe-renegotiation-1.patch
2023-05-25 09:19:41 <ikke> Anything that is accepted by patch(1)
2023-05-25 09:20:03 <powers> ikke thanks again, good pointer
2023-05-25 09:20:20 <powers> appreciate all the help everyone
2023-05-25 09:45:21 <psykose> i also already applied the patch in question
2023-05-25 09:45:38 <psykose> if you're on 3.18 then upgrading would work
2023-05-25 09:45:44 <psykose> if you're on edge then in a few hours
2023-05-25 10:07:11 <powers> psykose awesome, much appreciated
2023-05-25 11:05:12 <raspbeguy> hello, i am sure this has been asked before, but is there a place with alpine linux merch (like hoodies, mugs, etc...) ?
2023-05-25 11:08:06 <psykose> nope
2023-05-25 11:08:31 <psykose> i'm sure there's some unofficial 3rdparty robotaiautoprintedstolentrademark merch on randomwebsite tho
2023-05-25 11:15:46 <raspbeguy> would it be acceptable if someone sold some merch and give the money to the Alpine project?
2023-05-25 11:16:42 <psykose> there's no one to give money to
2023-05-25 11:18:00 <psykose> i don't think it matters either, print whatever you want
2023-05-25 11:19:25 <raspbeguy> the project does have an infrastructure to maintin though
2023-05-25 11:19:43 <raspbeguy> someone has bills to pay I suppose
2023-05-25 11:20:03 <psykose> the sponsors of infrastructure do, but we don't pay anything out of pocket
2023-05-25 11:20:17 <psykose> (technically there's an exception to this but it doesn't matter for this case)[27;2;13~
2023-05-25 11:20:20 <raspbeguy> ok
2023-05-25 11:20:57 <raspbeguy> that's a good thing, and this is quite rare
2023-05-25 11:21:41 <psykose> well
2023-05-25 11:21:50 <psykose> it could stop being sponsored at any moment and then everything is fucked :D
2023-05-25 11:21:57 <raspbeguy> heh
2023-05-25 11:22:17 <psykose> a lot of it is quite shaky
2023-05-25 11:23:15 <psykose> ppc/s390x is some donated ibm hardware (this is probably the most stable one because i doubt they'd pull this)
2023-05-25 11:23:21 <psykose> the whole dl-cdn is fastly
2023-05-25 11:23:38 <psykose> a lot of misc infrastructure and x86_64 builders is equinix
2023-05-25 11:23:51 <psykose> more misc infrastructure on linode (which got acquired lol)
2023-05-25 11:24:17 <psykose> arm is altra servers on lease from.. arm(?) dc'ed at ungleich
2023-05-25 11:24:28 <psykose> that one is most shaky since idk if anyone gets that offer anymore
2023-05-25 11:24:33 <psykose> maybe it was equinix instead i forget
2023-05-25 11:25:04 <raspbeguy> Other question, is there any community effort to compose delegations for attending to some tech/foss events?
2023-05-25 11:25:27 <psykose> what's an attending delegation
2023-05-25 11:26:31 <raspbeguy> maybe I'm not expressing myself right. I meant have some groups of people to go to tech or floss conventions
2023-05-25 11:26:44 <psykose> well, you can go to tech conventions(?)
2023-05-25 11:27:07 <psykose> this isn't a company that sends employees to places, so i'm not sure i understand the framing
2023-05-25 11:27:17 <raspbeguy> I meant for like having a stand and stuff to present Alpine
2023-05-25 11:27:25 <psykose> hmm
2023-05-25 11:27:30 <psykose> no idea then
2023-05-25 11:27:48 <raspbeguy> like fossdem
2023-05-25 11:28:02 <raspbeguy> or french JDLL or CdL
2023-05-25 11:28:11 <psykose> yeah i kinda get what you mean
2023-05-25 11:28:18 <psykose> but no i haven't heard of anyone doing that
2023-05-25 11:28:56 <raspbeguy> hence my question about merch 🙂
2023-05-25 11:29:48 <raspbeguy> I wish there was some goodies to give to people at such events
2023-05-25 11:30:30 <raspbeguy> or sold them the rest of the time, but I wouldn't be confortable keeping the money
2023-05-25 11:30:37 <psykose> is it too pessimistic to insert my personal opinion into this that most merch is just cringe and i don't get it
2023-05-25 11:31:12 <psykose> that said, my opinion on what you're thinking of is to do whatever
2023-05-25 11:31:25 <psykose> nothing wrong with uh, printing shirts yourself, and doing whatever from there
2023-05-25 11:31:48 <psykose> unless you really want someone to explicitly give you permission (i don't think we actually register a trademark?)
2023-05-25 11:31:58 <psykose> nobody would care or sue you
2023-05-25 11:32:12 <ikke> We did get some hoodies donated 
2023-05-25 11:32:17 <psykose> and it's much less messy than trying to pay the money forward to other things
2023-05-25 11:33:51 <psykose> ignore all previous statements. what i mean to say is do whatever you want and then gift me a threadripper cpu*
2023-05-25 11:33:53 <psykose> /joke
2023-05-25 11:34:02 <raspbeguy> 😛
2023-05-25 11:40:06 <raspbeguy> ikke, I suppose they are not produced anymore?
2023-05-25 11:41:11 <ikke> raspbeguy: was just a one-off purchase i suppose 
2023-05-25 11:43:12 <psykose> if you print merch can you design it to be actually good
2023-05-25 11:43:25 <psykose> +1 if it's not just the logo in a corner on standard print
2023-05-25 11:43:30 <psykose> -1 if it says "alpine linux"
2023-05-25 11:45:07 <psykose> now i have to imagine wearing a sundress that just says "ALPINE LINUX" across it
2023-05-25 11:45:08 <psykose> curse you
2023-05-25 12:08:07 <raspbeguy> psykose, well, as organizing alpine delegations doesn't seem to be a thing, I guess if I do something that will be only printing one hoodie for myself 😛
2023-05-25 12:08:15 <psykose> :)
2023-05-25 12:09:12 <raspbeguy> so just to hear you cringe over the web now I have to make it with just a logo and the text "alpine linux"
2023-05-25 12:09:57 <psykose> groaan
2023-05-25 12:34:48 <raspbeguy> psykose, https://i.imgur.com/FiUKjJK.png watch out
2023-05-25 12:35:28 <psykose> raspbeguy: 0/10
2023-05-25 12:36:14 <raspbeguy> that was my best shot. the world is not ready for my talent
2023-05-25 12:49:06 <Iosif_S> Hello, I'm trying to install alpine-virt-3.18.0-aarch64.iso in Paralells, on an Apple M2, OSX Ventura 13.2, and i get this error: Mounting boot media: failed.
2023-05-25 12:49:22 <Iosif_S> After removing the quiet option from the grub settings, I notice that it does not find the hdd (Sata 0:1)
2023-05-25 12:49:33 <Iosif_S> Does the Virt kernel version not include the Sata drivers ?
2023-05-25 12:49:50 <psykose> does the standard iso work
2023-05-25 12:51:09 <Iosif_S> Yes, the standar iso works 
2023-05-25 12:51:27 <psykose> it also sounds a bit like https://gitlab.alpinelinux.org/alpine/aports/-/issues/14407 but that one is a freeze and not 'not finding an hdd'
2023-05-25 12:51:42 <psykose> what 'hdd' is it? is it something virtualised or
2023-05-25 12:52:37 <psykose> if it's actual drive passthrough then you'll have to use standard as virt intentionally doesn't have drivers for every device in the universe
2023-05-25 12:52:49 <psykose> if it's some standard virt drive thing that happens to be disabled, then that should be enabled i guess
2023-05-25 12:55:34 <Iosif_S> It's avirtual hdd, Paralells creates it automatically
2023-05-25 12:56:43 <psykose> do you know what driver it normally uses on linux? you can iirc check with lspci -k from a working boot and you see what is loaded for a start, though that's not the precise config option
2023-05-25 12:58:03 <nullbytes[m]> Anyone interested in create a desktop alpine?
2023-05-25 12:58:53 <Saijin_Naib> I'd buy an Alpine shirt or something if I knew proceeds went directly to the project, as a means to donate. Cringe? Sure. But you always need graphic tee shirts 🤷
2023-05-25 12:59:46 <Saijin_Naib> nullbytes: I think we'd need a graphic installer first to make that palatable for the target group, no? 
2023-05-25 13:00:36 <psykose> oh god no
2023-05-25 13:01:35 <raspbeguy> > oh got no -> psykose, is that for the shirt or for the graphic installer?
2023-05-25 13:01:46 <msiism> Can setup-disk really do LUKS now as someone suggested in here yesterday (after I had already left)?
2023-05-25 13:01:48 <Misthios> or both?
2023-05-25 13:01:59 <raspbeguy> both is what I am expecting
2023-05-25 13:02:18 <psykose> i am wearing an ipv6 sage shirt that i didn't even earn myself so i'm not judging
2023-05-25 13:03:26 <msiism> An ncurses-based installer is probably the best way to do things because it doesn't require a lot to run but could also be used for more complex setups.
2023-05-25 13:03:49 <raspbeguy> msiism, yes setup-disk can do LUKS now. I installed my laptop that way
2023-05-25 13:04:12 <msiism> Okay, then someone(TM) should really add that info to the official Alpine handbook.
2023-05-25 13:04:23 <msiism> Thanks for the info.
2023-05-25 13:05:04 <msiism> I mean, I'm writing my own Alpine handbook anyway. But I not sure it's ever going to go public.
2023-05-25 13:13:19 <minimal> msiism: yes it can setup LUKS
2023-05-25 13:13:20 <Iosif_S> I cant't tell, in dmesg shows up as: scsi 2:0:0:0:0 Direct-access ATA Alpine 18.0-1 SS WTB4 PQ: 0 ANSI: 5
2023-05-25 13:14:19 <minimal> Iosif_S: that appears to be a SATA drive. When you create the VM in Parallels can you specify a different type of disk, like Virtio, to use?
2023-05-25 13:15:43 <msiism> minimal: Hi, you also asked what /dev/sda is, in my case: It's a laptop's internal HDD.
2023-05-25 13:15:57 <minimal> msiism: I asked what "dmesg" reports for sda
2023-05-25 13:15:59 <psykose> can you screenshot the configuration of parallels for the vm or whatever it is
2023-05-25 13:16:10 <Iosif_S> It seems that I can only change the Sata potr 0:1...0:6
2023-05-25 13:16:55 <minimal> psykose: Parallels website doesn't give much info on what they support, I was trying to figure out their virtio support a month or so ago and struggled to find info
2023-05-25 13:17:12 <psykose> yeah i tried looking and have no idea what it's exposing for 'generic drive for vm'
2023-05-25 13:17:29 <msiism> minimal: `dmesg | grep sda` returns nothing, actually.
2023-05-25 13:18:01 <minimal> msiism: confused how it finds sdb then if there's no sda
2023-05-25 13:18:08 <psykose> so you'll have to play detective from linux side and find out which 'sata drive' it is and what driver it needs to find out what exactly is missing from -virt, and if it makes sense to enable or not
2023-05-25 13:18:18 <msiism> minimal: Good question.
2023-05-25 13:18:20 <psykose> but if that's too much effort that's fine, you can also just use -standard :D
2023-05-25 13:21:48 <Iosif_S> I'll upload an image in a minute
2023-05-25 13:25:44 <raspbeguy> Usually I dislike white hoodies (and shirts in general) but in this case this makes sense https://i.imgur.com/W8goHe6.png
2023-05-25 13:25:57 <Iosif_S> https://gitlab.alpinelinux.org/alpine/aports/-/issues/14407
2023-05-25 13:27:20 <minimal> Iosif_S: ok, but is that the issue you are having?
2023-05-25 13:27:59 <Iosif_S> https://imgbox.com/sGuYIgYZ
2023-05-25 13:28:50 <Iosif_S> I\ve pasted the wrong link initally
2023-05-25 13:31:59 <psykose> hm
2023-05-25 13:32:10 <psykose> so you create a vhd and expose it as sata, so yeah have to find what driver it is
2023-05-25 13:32:15 <psykose> good luck with the linux spyglass
2023-05-25 13:32:42 <minimal> Iosif_S: so it only lets you create SATA disks? wow, that's......not ideal lol. Anyway the linux-virt kernel does have SATA support builtin (though it shouldn't)
2023-05-25 13:33:17 <minimal> assuming SATA (in their "terminology") == AHCI
2023-05-25 13:34:09 <msiism> minimal: Settung up the disk using setup-disk works pretty nicely. The problem is that I'd like to use a different partition layout.
2023-05-25 13:34:29 <msiism> I'll have a closer look at how the script does things.
2023-05-25 13:35:12 <minimal> msiism: so you can partition the disk and format filesystems yourself and then call setup-disk (passing it the appropriate param to let it know you'd done that)
2023-05-25 13:35:37 <msiism> Yeah, I'll try that.
2023-05-25 13:38:47 <minimal> msiism: but that still doesn't explain how you ended up with sdb but not sda
2023-05-25 13:41:18 <msiism> Well, sda itself does exist in /dev/. But if I create a partition with fdisk, that then doesn't show up.
2023-05-25 13:41:24 <msiism> Though lsblk will show it.
2023-05-25 13:41:47 <minimal> msiisn: so why does "dmesg" not show sda then?
2023-05-25 13:42:03 <msiism> No idea.
2023-05-25 13:42:04 <minimal> has your dmesg buffer been truncated?
2023-05-25 13:42:44 <msiism> How would I check for that?
2023-05-25 13:44:58 <minimal> msiism: does the 1st line of dmesg output contain something like "[    0.000000] Linux version 6.1.29-0-virt"?
2023-05-25 13:45:10 <Iosif_S> <minimal> Yes, sata only. In dmesg shows uo as [sda] attached SCSI disc (that is in the standard distribution)
2023-05-25 13:47:02 <msiism> minimal: No, it starts with "[    1.557638] usb…".
2023-05-25 13:48:37 <minimal> msiism: ok, so it has been truncated (you don't see the early messages)
2023-05-25 13:49:45 <msiism> Okay.
2023-05-25 13:50:38 <msiism> So, looking into setup-disk, I'm seeing that, for the first partition, an offset of "1M" is used.
2023-05-25 13:50:49 <msiism> Is this really necessary? I'm on a BIOS system.
2023-05-25 13:51:04 <minimal> msiism: that's normal
2023-05-25 13:51:19 <msiism> And also recommended?
2023-05-25 13:51:49 <minimal> msiism: where do you think the boot sector (and any addition bootloader bit) is going to go?
2023-05-25 13:52:01 <msiism> Into that 1M.
2023-05-25 13:52:07 <minimal> assuming MBR partitioning
2023-05-25 13:52:24 <minimal> yes, do an offset really is necessary ;-)
2023-05-25 13:52:33 <minimal> s/do/so/
2023-05-25 13:52:45 <msiism> Okay, noted. ;)
2023-05-25 13:57:40 <minimal> Iosif_S: and does the "dmesg" output regarding the sda disc mention AHCI?
2023-05-25 13:58:08 <psykose> just pastebin the dmesg
2023-05-25 14:28:52 <msiism> Hm… I just used sfdisk to create a new MBR partition spanning the whole disk, with the default offset (1M). After doing that `ls /dev/` does not show sda1.
2023-05-25 14:30:22 <psykose> partprobe
2023-05-25 14:30:48 <msiism> Runng that did not change the situation, unfortunately.
2023-05-25 14:31:22 <psykose> could you paste the logs of what you did to make the partition? :)
2023-05-25 14:32:46 <msiism> I just ran sfdisk and then typed `,` into its shell and then `write`.
2023-05-25 14:33:19 <msiism> I'm seeing now that setup-disk as a line of code commented with "#create device nodes if not exist"
2023-05-25 14:33:36 <msiism> That line says `$MOCK mdev -s`.
2023-05-25 14:34:15 <msiism> No idea what $MOCK is, though.
2023-05-25 14:34:40 <psykose> that's not really related to this
2023-05-25 14:34:48 <psykose> just partprobe would show the new partitions
2023-05-25 14:34:55 <psykose> that's also not pasting the logs
2023-05-25 14:35:46 <msiism> I know.
2023-05-25 14:36:04 <msiism> I could, maybe, use script to have a log of the sfdisk session.
2023-05-25 14:37:03 <minimal> msiism: after you create the partition and exit sfdisk then "dmesg" output *SHOULD* show it "seeing" sda1
2023-05-25 14:37:33 <minimal> if nothing appears in "dmesg" output then that is normally where "partprobe" or other equivalent tools come in
2023-05-25 14:37:34 <msiism> Oh, yeah, it does.
2023-05-25 14:37:45 <msiism> It's in dmesg output.
2023-05-25 14:37:54 <minimal> so then /dev/sda1 should be created
2023-05-25 14:38:02 <minimal> are you using mdev or eudev?
2023-05-25 14:38:18 <msiism> Whatever the live system uses.
2023-05-25 14:38:46 <minimal> mdev then
2023-05-25 14:40:06 <msiism> `mdev -s` solved it.
2023-05-25 14:42:09 <minimal> which is what setup-disk does after it creates partitions...
2023-05-25 14:42:28 <msiism> Exactly.
2023-05-25 14:53:55 <uniqNuuid> mdev has issues
2023-05-25 14:54:11 <uniqNuuid> e.g. after partition, mkfs can't find the part
2023-05-25 14:54:50 <uniqNuuid> mdevd does not have this issue. `setup-devd mdevd` is the way to go
2023-05-25 14:58:33 <msiism> Good to know.
2023-05-25 15:33:45 <Piraty> what is required to bootstrap a mininal rootfs so `apk add -p miniroot foo` can work?
2023-05-25 15:35:23 <minimal> Piraty: use the static apk to install a base install that you can chroot into
2023-05-25 15:35:48 <Piraty> lol minimal sorry for unintended ping
2023-05-25 15:35:59 <minimal> well it doesn't need to be the "static" version of apk if you're doing that on an Alpine machine, the static version lets you do that on other distros
2023-05-25 15:36:18 <minimal> you didn't ping me, you spelt minimal wrong ;-)
2023-05-25 15:36:30 <Piraty> i don't care which one to use, but i doubt the static one does anything different than a properly installed apk
2023-05-25 15:36:53 <minimal> it lets you run apk on non-Alpine...
2023-05-25 15:36:55 <Piraty> spelt minimal wrong, so truth
2023-05-25 15:44:26 <uniqNuuid> Piraty: alpine already have a 2Mb minirootfs
2023-05-25 15:44:52 <Piraty> i shoved /etc/apk to the root, touch lib/apk/db/lock  and this seems to do it
2023-05-25 15:45:14 <Piraty> package fakeroot is not able to install unprivileged though, makes `apk add` segfault
2023-05-25 15:45:36 <uniqNuuid> this? https://dl-cdn.alpinelinux.org/alpine/v3.18/releases/x86_64/alpine-minirootfs-3.18.0-x86_64.tar.gz
2023-05-25 15:46:32 <Piraty> i want a build container bootstrapped from local repos
2023-05-25 15:47:09 <uniqNuuid> so? wget and tar -x
2023-05-25 15:48:36 <minimal> Piraty: did you use apk to create a base alpine install in a directory as I suggested?
2023-05-25 15:49:28 <Piraty> if that means `rsync -avP /etc/apk ./buildroot/etc` and touching that empty lock file, then yeah
2023-05-25 15:49:55 <Piraty> `apk -p path add foo`doesn't work on empty dir oob
2023-05-25 15:51:55 <minimal> Piraty: no it didn't mean that - how does using rsync match me suggesting you use "apk" to create a base alpine install?
2023-05-25 15:52:38 <Piraty> how should i use apk on an empty dir to create a base alpine install if it won't operate without setting up some minimal things
2023-05-25 15:52:41 <uniqNuuid> because keys are at /etc/apk minimal 
2023-05-25 15:52:43 <minimal> "apk --initdb --root <directory to use> --update-cache add alpine-base <any more packages>"
2023-05-25 15:53:05 <Piraty> minimal: you didn't suggest this before
2023-05-25 15:53:20 <minimal> I did, I just didn't specify the EXACTLY command params
2023-05-25 15:53:44 <Piraty> and assuming i could derive this from consulting the manpage and your writing within 15min is rude
2023-05-25 15:53:59 <minimal> "use the static apk to install a base install that you can chroot into" == run a apk command with options like above
2023-05-25 15:54:31 <Piraty> thank you then for dropping some wisdom on me
2023-05-25 15:54:43 <Piraty> --initdb makes sense
2023-05-25 15:56:02 <Piraty> `man apk` doesn't mention --initdb 
2023-05-25 15:56:44 <Piraty> apk-add has
2023-05-25 15:56:49 <Piraty> so your command is misleading
2023-05-25 16:02:54 <minimal> Piraty: how is it misleading? the example command I provides includes "add"
2023-05-25 16:03:59 <minimal> I'm sorry now that I tried to help you
2023-05-25 16:10:53 <Piraty> i'm sorry you're sorry, thanks for help
2023-05-25 16:10:53 <Piraty> due to command ordering
2023-05-25 16:22:02 <WhyNotHugo> libreoffice doesn't run on Wayland (it uses XWayland and has scaling issues).
2023-05-25 16:22:25 <WhyNotHugo> But upstream indicates that it supports Wayland. I think it's something related to build flags or dependencie?
2023-05-25 16:22:33 <WhyNotHugo> I can't see anything obviously wrong or missing.
2023-05-25 16:29:26 <Saijin_Naib> <uniqNuuid> "mdevd does not have this issue...." <- Even over eudev? 
2023-05-25 17:04:20 <psykose> WhyNotHugo: just 'libreoffice' doesn't install libreoffice-gtk4 which is what you probably want (and automatically will get used)
2023-05-25 17:04:47 <psykose> there's also -qt5 -qt6 and some env var to pick which windowing system, some kf5 kde integration thing, ..
2023-05-25 17:05:14 <psykose> it doesn't install it by default so someone can pick the others i guess, but yeah it's atrocious by default
2023-05-25 18:11:43 <WhyNotHugo> psykose: Thanks!
2023-05-25 18:12:02 <WhyNotHugo> What's the default windows system that it installs?
2023-05-25 18:13:54 <WhyNotHugo> libreoffice-gtk4 includes /usr/lib/libreoffice/program/libvclplug_gtk3lo.so o.O
2023-05-25 18:33:05 <psykose> nothing, whatever it's loading by default, the legacy widget thing
2023-05-25 18:33:10 <psykose> don't remember what it's named
2023-05-25 18:34:24 <psykose> and yeah there's both 3+4 in there
2023-05-25 18:34:33 <psykose> the whole thing is kinda weird
2023-05-25 20:27:12 <Lagopus> is alpine fixed and/or can you use the repo manager to roll it?
2023-05-25 20:27:50 <psykose> like a blunt?
2023-05-25 20:31:16 <Lagopus> yes like a blunt
2023-05-25 20:31:40 <jvoisin> roll your own blunts as much as you like, but don't roll your own crypto?
2023-05-25 20:32:08 <Lagopus> I can already tell from the banter that I should have installed this OS a long time ago.
2023-05-25 20:42:57 <psykose> more seriously if you're asking if it's stable or rolling then edge is rolling and stable is stable aside from fixes/security backports
2023-05-25 20:45:54 <Saijin_Naib> Does using the latest-stable repo then count as semi-rolling, like how DistroWatch classifies things? 
2023-05-25 20:46:06 <Saijin_Naib> And also yes, unreservedly, use Alpine 
2023-05-25 20:47:15 <psykose> no
2023-05-25 20:47:35 <psykose> unless we have some weird-extended sense of time, then yes :p
2023-05-25 21:01:32 <Lagopus> psykose: yes, that answers my question. Thanks
2023-05-25 21:02:00 <powers> ladies and gentlemen, im happy to announce that ive got the broadcom-wl driver functioning on Alpine 
2023-05-25 21:03:08 <abby> 
ACTION gets out the champagne

2023-05-25 21:03:51 <powers> as this was my first trip around the horsey, this likely took longer than it should've. but im excited to get this process streamlined and wiki'd
2023-05-25 21:04:31 <powers> i will do further testing over day, and post a preliminary git repo, so expect more info soon
2023-05-25 21:14:31 <psykose> sadly the actual driver probably won't be merged :p
2023-05-25 21:48:18 <gmalsack> hello all - can anyone tell me if I can upgrade my alpine linux from 3.1 to 3.18 by simply changing the repository info, running an apk update then an apk upgrade? Or will that break my device?
2023-05-25 21:49:19 <iggy> 3.1 all the way to 3.18 is probably a bit big of a jump. I generally have good enough luck doing 2 versions at a time
2023-05-25 21:49:42 <gmalsack> ok. thanks
2023-05-25 22:02:58 <powers> gmalsack i believe the signing-keys for alpine were changed somewhere around 3.15, and maybe another time before that
2023-05-25 22:03:44 <powers> gmalsack all to say you may need to upgrade to two other versions before 3.18
2023-05-25 22:04:53 <powers> gmalsack should look something like 3.1 -> 3.8 (if necessary) -> 3.15 -> 3.18
2023-05-25 22:05:43 <gmalsack> fml.... I just upgraded from 3.1 to 3.2 to 3.3 and all went well. Now that I've upgraded to 3.4 I'm diw. neither ssh nor web-gui is working. :-(
2023-05-25 22:07:17 <gmalsack> thankfully the openvpn client is still connecting, so I don't have to urgently drive out to the site and get it back up and running
2023-05-25 22:08:16 <powers> gmalsack could either be a dependency-issue (meaning that higher-versions of the packages require certain apk or so you dont have)
2023-05-25 22:09:32 <powers> an API issue, meaning if you didnt upgrade your apk, that the kernel on your version doesn't support the calls for ssh/web-gui
2023-05-25 22:10:47 <powers> or a signing-issue, where ca-certs weren't updated, so when trying to use networking/udhcpd your requests are being denied
2023-05-25 22:10:49 <gmalsack> I would have expected apk update then apk upgrade would have upgraded everything....
2023-05-25 22:11:33 <powers> gmalsack maybe. im new to alpine so my word isnt law, just some potential issues
2023-05-25 22:12:21 <gmalsack> I'm guessing something didn't upgrade properly from 3.3 to 3.4. like I said, ssh isn't running, web gives me this: https://snipboard.io/r9jtOG.jpg
2023-05-25 22:13:09 <gmalsack> nmap says only 80 is open. I'm guessing at this point, I'll have to put together a new sd card and drive over to the site and swap the cards.
2023-05-25 22:13:18 <powers> if possible i recommend backing up your configs and any other valuable data on a usb/sd-card, then live-booting and setting up 3.18
2023-05-25 22:13:38 <gmalsack> yup, that's the plan
2023-05-25 22:15:15 <powers> gmalsack , from what i see there its a module issue (API has changed). has your kernel itself actually upgraded?
2023-05-25 22:15:36 <powers> what is the output of `uname -r`?
2023-05-25 22:16:02 <gmalsack> don't know. can't get into it anymore.
2023-05-25 22:16:41 <psykose> you need upgrade -a between release but that's probably not the issue
2023-05-25 22:17:12 <psykose> you'd probably have better luck just going to 3.18 as fast as possible without checking every release
2023-05-25 22:17:22 <powers> gmalsack okay thats fair, best wishes on the move to 3.18
2023-05-25 22:18:27 <gmalsack> psykose when I change the repo version, I need to do apk update then apk upgrade -a?
2023-05-25 22:18:48 <gmalsack> Whoops. Didn't know that. Guess that might be what f'd up the router.
2023-05-25 22:19:03 <psykose> just apk upgrade -Ua is fine, but it's probably not the issue
2023-05-25 22:19:49 <psykose> the specific error looks like acf running with lua5.3 for some reason even though afaik it's only for 5.2
2023-05-25 22:19:55 <minimal> gmalsack: also to clarify, are these Sys-mode installations or run-from-ram?
2023-05-25 22:20:53 <gmalsack> dont know, but now I'm weiry about doing the other 3 routers.
2023-05-25 22:21:00 <gmalsack> minimal these are run-from-ram
2023-05-25 22:21:15 <gmalsack> minimal yes, I also ran lbu ci after the upgrade
2023-05-25 22:28:31 <powers> psykose just a question about merging the driver, i assume wl cant be merged due to it's licensing - which makes sense
2023-05-25 22:28:50 <psykose> i assume that's you that opened the mr
2023-05-25 22:28:54 <psykose> if not, crazy coincidence :-)
2023-05-25 22:29:15 <powers> is there any world this can make it into the community repo?
2023-05-25 22:29:41 <powers> psykose no its a coincidence, that's crazy lol
2023-05-25 22:29:43 <psykose> :)
2023-05-25 22:30:01 <psykose> https://gitlab.alpinelinux.org/alpine/aports/-/merge_requests/47458
2023-05-25 22:30:17 <powers> :0
2023-05-25 22:30:21 <powers> wtffffffff
2023-05-25 22:30:36 <powers> thats actually insane lmao
2023-05-25 22:30:38 <psykose> it works as is if that's the sort of thing you got working
2023-05-25 22:30:45 <psykose> or wanted to get working, i suppose
2023-05-25 22:31:12 <powers> i want to cry so bad, but im also happy its released
2023-05-25 22:31:15 <powers> idk man
2023-05-25 22:32:02 <powers> well i have a driver here too that's working, but ill use the test package
2023-05-25 22:34:22 <powers> well, ill still post what i have on git (T_T)
2023-05-25 22:44:19 <powers> i have some patches (for lts-6.1.29) that can be included in the driver version on testing/broadcom-wl, as it seem it only supports up to 5.10
2023-05-25 22:44:44 <powers> since its testing, should i go to their source or aports?
2023-05-25 22:45:41 <psykose> the driver supports up to 6.3 by what the repo says
2023-05-25 22:53:03 <powers> yes that's correct, i was on the main-branch instead of the patch-branch
2023-05-25 22:57:14 <powers> in the long-run this does save me the trouble of doing it myself, and i still have a non-akms version which a niche few might want. plus a small guide
2023-05-25 22:57:30 <powers> im coping, some may say seething, but thats okay
2023-05-25 22:58:00 <powers> im approaching the 7th stage of grief, my narvana
2023-05-25 23:08:18 <psykose> hate it when that happens
2023-05-25 23:08:19 <psykose> :-)
2023-05-26 01:08:49 <powers> is there anyway to remove ifconfig from busybox?
2023-05-26 01:09:02 <psykose> not without building your own busybox without it
2023-05-26 01:09:25 <powers> okay thank you
2023-05-26 01:09:59 <minimal> powers: why would you want to remove it? just don't use it if you don't like it
2023-05-26 01:10:09 <powers> ip seems to be the standard, but its likely for legacy reasons
2023-05-26 01:10:40 <powers> minimal true, i just wanted to know if it'd be possible
2023-05-26 01:11:17 <minimal> well yes, basically every "applet" in Busybox can be individually enabled/disabled when it is built
2023-05-26 01:14:25 <minimal> at some point in time when it's certain that nothing in Alpine is using "ifconfig" then in theory it could be disabled. However I'd guess in that scenario there'd be complaints as some people probably still use "ifconfig" manually out of habit rather than "ip"
2023-05-26 01:16:32 <powers> thanks for the response. that's also true, many legacy programs are probably using it as well, its just a general headache to remove
2023-05-26 01:18:02 <powers> i just wont use it, but its better for it to be here
2023-05-26 01:18:04 <psykose> from the busybox side specifically it probably weighs like 80kb
2023-05-26 01:20:57 <minimal> the idea behind "ip" was to combine stuff from "ifconfig" and some other related utils
2023-05-26 01:21:39 <powers> 80kb is pretty impressive actually. for me, i just keep using it instead of ip, out-of-habit. i wanted to disable it to force myself to adjust
2023-05-26 01:22:12 <powers> but i can also simply just not use it lol, so ill go that route
2023-05-26 01:22:16 <psykose> it actually starts at 12
2023-05-26 01:22:30 <psykose> +7 for status reporting
2023-05-26 01:22:46 <minimal> don't remember the last time I used ifconfig...
2023-05-26 01:22:54 <eex0> ip being more powerful , me either
2023-05-26 01:24:33 <powers> ip + iw is insanely strong
2023-05-26 01:24:53 <psykose> my ip is stronger than yours
2023-05-26 01:25:26 <powers> ifconfig is muscle memory for me because i used to use macos
2023-05-26 01:25:34 <minimal> oh dear is it Spargel season already? ;-)
2023-05-26 01:25:37 <powers> psykose lmao
2023-05-26 01:27:31 <eex0> pipes up!
2023-05-26 01:28:18 <powers> minimal is spargel good?
2023-05-26 01:28:38 <psykose> well then you better go catch it !
2023-05-26 01:29:10 <powers> i know im potentially walking into a landmine, but spargel is calling
2023-05-26 01:29:37 <powers> i really hope we're talking about the food
2023-05-26 01:29:41 <eex0> whats a spargel?
2023-05-26 01:30:21 <powers> it an asparagus dish, i think. i saw it way back when
2023-05-26 01:30:33 <psykose> isn't it just asparagus
2023-05-26 01:30:43 <powers> oh god
2023-05-26 01:30:59 <eex0> I had to google that.. add mustard and im in
2023-05-26 01:32:25 <powers> eex0 lets live in the timeline where we're refering to food (T_T)
2023-05-26 01:33:20 <eex0> I never play with my food, eh
2023-05-26 01:33:31 <powers> i dont know the other meaning, but with asparagus involved i dont have high hopes
2023-05-26 01:38:13 <Saijin_Naib> Anyone running minidlna successfully?
2023-05-26 01:38:33 <Saijin_Naib> This seems like the type of thing that should be NotThatHard(TM), and yet...
2023-05-26 01:38:40 <minimal> powers: it was a reference to psykose's "my ip is stronger than yours" - used to hate Spargel Season in Germany as the public toilets were "Spargel perfumed" lol
2023-05-26 01:39:34 <psykose> oof
2023-05-26 01:40:09 <minimal> eating it results in a strongly fragranced urine lol
2023-05-26 01:40:10 <eex0> I'll not be pulling fingers tonight
2023-05-26 01:43:59 <powers> minimal ohhhhh lml
2023-05-26 01:44:34 <powers> you dont know how glad i am to hear that, i went down a very different path
2023-05-26 01:46:52 <powers> Saijin_Naib ive never used the program, but whats the issue?
2023-05-26 01:47:52 <Saijin_Naib> Nothing that I can see. Service starts, no logs print, nothing in console even with verbose, have the ports open in ufw... 
2023-05-26 01:48:21 <Saijin_Naib> It just does nothing, doesn't scan the dirs I set, doesn't serve the page, doesn't become discoverable 
2023-05-26 01:48:53 <Saijin_Naib> Not sure if it is like xrdp which apparently is broken how we ship it due to a missing config or something 
2023-05-26 01:49:01 <Saijin_Naib> Or in missing a depends that wasn't declared
2023-05-26 01:49:37 <powers> can you strace it and send the results?
2023-05-26 01:51:58 <Saijin_Naib> Uuuuh
2023-05-26 01:52:08 <powers> strace -d [prog with options and arguments]
2023-05-26 01:52:10 <Saijin_Naib> Sure, I look up how to do that
2023-05-26 01:52:13 <Saijin_Naib> Thank you
2023-05-26 01:52:50 <powers> `apk add strace`, then `strace -d [prog with options and arguments]`
2023-05-26 01:54:08 <powers> so for example say i wanted to run `cd /home/folder` the strace would be `strace -d cd /home/folder`. this has to be done in terminal
2023-05-26 01:55:54 <psykose> run it yourself with `minildnad -d` in foreground
2023-05-26 01:57:21 <psykose> well
2023-05-26 01:57:31 <psykose> it also writes to /var/log/minildna/minildna.log
2023-05-26 01:57:44 <psykose> dlna
2023-05-26 01:59:31 <psykose> you also definitely don't want to do strace -d because it makes it like 6x more verbose with no actual info
2023-05-26 02:01:54 <powers> agreed, if a log for the program exists then its better to use that
2023-05-26 02:03:37 <powers> i `strace -d` for info on seg-faults or traps, but a logger for the program will give more concise info
2023-05-26 02:07:51 <psykose> you pretty much nevver want strace -d for any of that
2023-05-26 02:08:09 <psykose> don't know where you get that idea
2023-05-26 02:08:22 <psykose> love to have my nice strace log full of
2023-05-26 02:08:22 <psykose> strace: next_event: queued pid 655
2023-05-26 02:08:22 <psykose> strace: next_event: dequeued pid 655
2023-05-26 02:08:28 <psykose> how useful
2023-05-26 02:09:03 <psykose> that said it does print a bunch of wife stopped so i guess it's funny
2023-05-26 02:09:59 <powers> lmaoo
2023-05-26 02:10:21 <Saijin_Naib> It's not even writing to the log 🤷
2023-05-26 02:10:53 <Saijin_Naib> It's doing fuck all and doesn't say anything when launched in the Terminal interactively with -v for verbose
2023-05-26 02:11:16 <Saijin_Naib> Hmm.... Debug flag? Let me try that next
2023-05-26 02:11:27 <psykose> it backgrounds itself without -d
2023-05-26 02:11:30 <psykose> also it does write a log
2023-05-26 02:11:35 <psykose> you can't see it unless you are root
2023-05-26 02:11:42 <Saijin_Naib> 😑
2023-05-26 02:14:47 <eex0> idk about the spargel but these baby back ribs are d'lish
2023-05-26 02:15:11 <Saijin_Naib> You beautiful people... Thank you
2023-05-26 02:15:17 <Saijin_Naib> Can't write the database ou5
2023-05-26 02:17:35 <psykose> what's a database ou5
2023-05-26 02:17:54 <Saijin_Naib> Fat-fingerings of a touch keyboard, so nonsense 
2023-05-26 02:18:02 <Saijin_Naib> Can't write the database out
2023-05-26 02:19:27 <Saijin_Naib> Oh do I feel dumb. 
2023-05-26 02:19:34 <eex0> I put my fingers on a diet and they still do that, no worries!
2023-05-26 02:19:35 <Saijin_Naib> Thanks powers and psykose
2023-05-26 02:22:54 <powers> np!
2023-05-26 02:24:05 <eex0> I just realised you called me old inadvertantly back there 'powers
2023-05-26 02:24:15 <powers> psykose what do you use as a debug-tool when source is imploding?
2023-05-26 02:24:47 <eex0> good Q
2023-05-26 02:24:59 <eex0> I need to learn some stuff
2023-05-26 02:25:12 <powers> maybe strace is an old-world tool, if i can be more efficient then it would be for the best
2023-05-26 02:26:54 <psykose> i didn't say don't use strace
2023-05-26 02:27:01 <powers> eex0 wait i did, where? my bad, it wasnt my intention lol
2023-05-26 02:27:15 <psykose> strace -d tho is probably not what you want, emphasis
2023-05-26 02:27:28 <eex0> haha.. twist a towel for a quick snap
2023-05-26 02:28:48 <powers> psykose okay fair point
2023-05-26 02:29:45 <eex0> the trace stepping looked interesting, I've used trace facilities coding but not w/ strace before 
2023-05-26 02:42:01 <quarkyalice> has anyone gotten Parallels Tools working on Alpine in Parallels desktop on arm64?
2023-05-26 02:43:31 <eex0> I really need risc and arm64 machines to play with.. starts a wish list
2023-05-26 02:43:48 <quarkyalice> the M1 is a good choice, but it's expensive
2023-05-26 02:43:59 <quarkyalice> Apple M1
2023-05-26 02:45:14 <quarkyalice> an inexpensive option is the PineBook Pro but it's very slow and only has 4GB RAM
2023-05-26 02:46:30 <quarkyalice> most arm64 SoCs are designed for mobile and embedded usage so they are really slow, Apple Silicon is the best ones on the market currently
2023-05-26 02:47:36 <minimal> quarkyalice: I'm assuming Parallels Tools won't work as they are likely glibc-based binaries
2023-05-26 02:47:46 <psykose> no, you build them
2023-05-26 02:47:58 <minimal> oh? hmm, where's the source at?
2023-05-26 02:48:05 <quarkyalice> can't glibc binaries run using gcompat?
2023-05-26 02:48:22 <psykose> minimal: hm actually, not sure if you do
2023-05-26 02:48:28 <psykose> quarkyalice: generally no
2023-05-26 02:48:49 <minimal> psykose: found a webpage that states glibc 2.3.4 or later as a requirement
2023-05-26 02:48:59 <psykose> that is always meaningless and you know it :p
2023-05-26 02:49:21 <psykose> bet you need an 'internet connection' and a 'usb mouse' too
2023-05-26 02:50:06 <psykose> but yes can't even download this from anywhere to see that i can find
2023-05-26 02:50:09 <minimal> page doesn't mention anything about compiling them
2023-05-26 02:50:11 <psykose> weird mounted iso from inside vm
2023-05-26 02:50:14 <eex0> whats the scoop on glibc on Alpine, will there be another version like that voided distro?
2023-05-26 02:50:25 <psykose> no
2023-05-26 02:50:29 <eex0> good
2023-05-26 02:50:54 <quarkyalice> glibc on alpine would defeat the point wouldn't it?
2023-05-26 02:50:57 <abby> 
ACTION avoids

2023-05-26 02:51:08 <minimal> quarkyalice: they *might* run via gcompat, who knows
2023-05-26 02:51:18 <eex0> abby for president!
2023-05-26 02:51:29 <abby> i don't even use alpine
2023-05-26 02:51:31 <psykose> whats the scoop with this abby person on alpine
2023-05-26 02:51:33 <psykose> spy corpus ?
2023-05-26 02:51:54 <abby> im here to spread the tendrils of glibc
2023-05-26 02:51:55 <psykose> diplomatic mission ?
2023-05-26 02:52:13 <eex0> wtf is a tendril?
2023-05-26 02:52:27 <psykose> it's ten dril's from twitter
2023-05-26 02:52:32 <eex0> hahaha
2023-05-26 02:52:57 <eex0> no idea why I laughed, I didn't get that either
2023-05-26 02:53:21 <psykose> i laugh at my wall half the time
2023-05-26 02:54:23 <eex0> <~ recovering
2023-05-26 03:01:36 <quarkyalice> pine64 is working on some riscv hardware, that'll be interesting
2023-05-26 03:01:49 <quarkyalice> https://pine64.com/product/pinetab-v-10-1-8gb-128gb-risc-v-based-linux-tablet-with-detached-backlit-keyboard/
2023-05-26 03:03:06 <quarkyalice> this is development hardware
2023-05-26 03:45:21 <gmalsack> so I've upgraded a device to 3.16, but my /media/usb/apks folder still has the old 3.14.2 packages. Is there a way to replace that folder with the 3.16 files?
2023-05-26 03:46:31 <ikke> gmalsack: for run from ram systems, you need to update the boot medium to get new kernels 
2023-05-26 03:47:53 <ikke> gmalsack: to answer your last question, apk cache sync might help
2023-05-26 03:50:57 <gmalsack> ikke: is updating the boot medium while the unit is online an option? it's a vpn router and I'm remote. 
2023-05-26 03:51:43 <ikke> gmalsack: yes, though it can be tricky to get right
2023-05-26 03:51:53 <ikke> especially going from 3.1 to 3.18
2023-05-26 03:53:02 <gmalsack> hahaha that one is already f'd. I have to go replace that one. I have 3 more that are on 3.14. I upgraded one of those to 3.18 and found out the version of openvpn there doesn't support our current vpn router. so 3.16 is it for now until I can get the main vpn server replace
2023-05-26 03:53:37 <gmalsack> found that out after the upgrade left the unit off-line. had to drive 2 hours one way to get that unit back online.
2023-05-26 03:53:50 <gmalsack> whoops
2023-05-26 03:54:19 <psykose> is it the 'version of openvpn' or is it that since 3.17 it's openssl3 that disables tls1.1 and all the insecure protocols without an extra flag
2023-05-26 03:55:35 <gmalsack> psykose: no comment.... hahaha
2023-05-26 03:57:56 <ikke> quarkyalice: We're waiting for server-grade hw
2023-05-26 03:58:21 <psykose> only 6 more years
2023-05-26 04:07:10 <quarkyalice> ikke: server grade hardware is a long ways away but it's good seeing progress on developer boards for riscv
2023-05-26 08:16:02 <eggbean> I have a docker build which fails when I bump up the base image to alpine 3.18.0. longintrepr.h not found, but I have the python3-dev dependency. Don't know what's going on.
2023-05-26 08:21:45 <abby> maybe your include path is wrong?
2023-05-26 08:22:47 <Guest1395> abby: How do I check that? This is the relevant part of the log: http://sprunge.us/HCltZt
2023-05-26 08:23:08 <abby> in 3.17 it was python3.10/longintrepr.h and in 3.18 it's python3.11/cpython/longintrepr.h
2023-05-26 08:24:04 <abby> > Released: Apr 2, 2018
2023-05-26 08:24:06 <Guest1395> So what do I need to change in the Dockerfile?
2023-05-26 08:24:32 <abby> i'm gonna guess that that library isn't compatible with python 3.11
2023-05-26 08:24:49 <Guest1395> I see
2023-05-26 08:25:03 <Guest1395> So I just need to wait more probably?
2023-05-26 08:25:16 <Guest1395> s/more/most/
2023-05-26 08:25:28 <abby> I don't think waiting will help, the homepage of that package doesn't even resolve anymore
2023-05-26 08:25:37 <Guest1395> I see
2023-05-26 08:26:27 <Guest1395> Which package? python-potr?
2023-05-26 08:26:30 <abby> yes
2023-05-26 08:26:36 <Guest1395> Okay, thanks
2023-05-26 08:27:53 <abby> oh wait, I misread, pycrypto is what failed to compile
2023-05-26 08:28:15 <abby> yikes, last released in 2013
2023-05-26 08:28:35 <Guest1395> Oh
2023-05-26 08:29:54 <abby> you *might* be able to fix this by using pip install --global-option="-I/usr/include/python3.11/cpython/"
2023-05-26 08:30:19 <abby> but i don't think it's a good idea to be using 10-year-old crypto implementations :P
2023-05-26 08:30:29 <Guest1395> Ah, okay
2023-05-26 08:30:47 <Guest1395> Yes, I think I'll just drop support for python-potr then
2023-05-26 08:30:53 <abby> This software is no longer maintained.
2023-05-26 08:30:53 <abby> PyCrypto 2.x is unmaintained, obsolete, and contains security vulnerabilities.
2023-05-26 08:30:59 <abby> 😬
2023-05-26 08:32:08 <Guest1395> Have you got a link for the above? I'll add it to the commit message
2023-05-26 08:32:29 <abby> https://www.pycrypto.org
2023-05-26 08:32:32 <abby> top of the homepage
2023-05-26 08:32:33 <Guest1395> Thanks
2023-05-26 09:41:46 <j-k[m]> pycryptodome seems pretty drop-in
2023-05-26 10:22:38 <ncopa> my icons in thunar toolbar under xfce are broken
2023-05-26 10:22:50 <ncopa> "home" shows properly, but the others not
2023-05-26 10:23:11 <ncopa> i think this broke shortly before the 3.18 release. any idea?
2023-05-26 11:07:22 <donoban> ncopa: have you tried to look in strace for file opening errors?
2023-05-26 11:09:47 <ncopa> nope
2023-05-26 11:10:32 <donoban> probably is some missing theme/icon-theme
2023-05-26 12:23:08 <Cronix> hi, currently trying to create a docker container based on alpine 3.18 and x86_64 arch, i have some issues with libgcc_s.so.1, as it is x32 bit elf, no idea if that is intended or a bug?
2023-05-26 12:23:37 <Cronix> 1ef4343a4db4:/config# cat /etc/apk/arch  x86_64 1ef4343a4db4:/config# file /lib/libgcc_s.so.1  /lib/libgcc_s.so.1: ELF 32-bit LSB shared object, Intel 80386, version 1 (SYSV), dynamically linked, BuildID[sha1]=15a399d96350896c3cdb5545221674a8df9ab5bf, stripped
2023-05-26 12:24:10 <Cronix> 1ef4343a4db4:/config# cat /etc/issue  Welcome to Alpine Linux 3.18
2023-05-26 12:29:19 <Cronix> Error loading shared library libgcc_s.so.1: Exec format error (needed by foobar**
2023-05-26 12:39:10 <Hello71> alpine doesn't have any /lib/libgcc_s.so.1 so you must have installed it from somewhere else
2023-05-26 12:39:13 <Hello71> https://pkgs.alpinelinux.org/contents?file=libgcc_s.so.1&path=&name=&branch=edge
2023-05-26 12:44:34 <Cronix> thats where i got it from
2023-05-26 12:44:46 <Cronix> but its a 32bit binary library and not 64bit
2023-05-26 12:45:01 <Cronix> and i am confused by that, since my process wants a 64bit one XD
2023-05-26 13:13:07 <quinq> Cronix, from where?
2023-05-26 13:18:01 <Cronix> https://pkgs.alpinelinux.org/package/v3.18/main/x86_64/libgcc
2023-05-26 13:21:03 <dhruvin> It contains /usr/lib/libgcc_s.so.1
2023-05-26 13:22:00 <dhruvin> Can you try `file /usr/lib/libgcc_s.so.1` to see if it matches what you are looking for?
2023-05-26 13:22:27 <dhruvin> Cronix: ^
2023-05-26 13:23:12 <Cronix> that is exactly the file im talking about, but it is a x32 so file, despite the fact that i am on a x64 based alpine container
2023-05-26 13:23:27 <Cronix> file /lib/libgcc_s.so.1  /lib/libgcc_s.so.1: ELF 32-bit LSB shared object
2023-05-26 13:23:44 <dhruvin> No, /usr/lib/libgcc_s.so.1
2023-05-26 13:25:07 <dhruvin> See its contents here (i.e. no /lib/libgcc): https://pkgs.alpinelinux.org/contents?branch=v3.18&name=libgcc&arch=x86_64&repo=main
2023-05-26 13:26:28 <quinq> Cronix, you're mixing /lib/libgcc_s.so.1 and /usr/lib/libgcc_s.so.1
2023-05-26 13:26:50 <quinq> The first isn't distributed by any alpine packages, while the second is provided by libgcc
2023-05-26 13:26:50 <Cronix> hmm thats a good point
2023-05-26 13:27:06 <Cronix> sry, just got a bit blind there i guess
2023-05-26 13:27:08 <Cronix> lemme check
2023-05-26 13:27:19 <quinq> No problem, sharing a problem is what that's for ;)
2023-05-26 13:28:15 <Cronix> rubberduck engineering sometimes also helps yeah
2023-05-26 13:28:33 <dhruvin> Now I'm curious what base image are you using (if you can share) that provides that file :)
2023-05-26 13:28:46 <Cronix> steamcmd:alpine
2023-05-26 13:28:56 <dhruvin> 
ACTION is looking into it

2023-05-26 13:29:05 <Cronix> after cheking, yeah, they are indeed copying over a manually build so file
2023-05-26 13:29:24 <dhruvin> 
ACTION stopped looking into it

2023-05-26 13:29:43 <Cronix> https://github.com/steamcmd/docker/blob/master/dockerfiles/alpine-3/Dockerfile
2023-05-26 13:30:00 <Cronix> but its only the libstdc++ as far as i can tell
2023-05-26 13:32:26 <dhruvin> Cronix: one line above that libstdc++
2023-05-26 13:32:58 <Cronix> ah, ew
2023-05-26 13:33:16 <Cronix> just hardcore copying over all the x86 libs that are there
2023-05-26 13:33:19 <Cronix> jeez
2023-05-26 13:33:35 <Cronix> aac50db4e54c:/config$ file /usr/lib/libgcc_s.so.1  /usr/lib/libgcc_s.so.1: ELF 64-bit LSB shared object, x86-64, version 1 (SYSV), dynamically linked, stripped
2023-05-26 13:33:39 <Cronix> yeah, that looks better
2023-05-26 13:33:40 <quinq> That would copy to /lib/i386-linux-gnu, wouldn't it?
2023-05-26 13:33:44 <quinq> (I don't know dockair)(
2023-05-26 13:33:54 <dhruvin> You might want to report this to them, Cronix :)
2023-05-26 13:34:08 <Cronix> that they are stupid? :D
2023-05-26 13:34:14 <dhruvin> quinq: yes, copying from ubuntu to alpine
2023-05-26 13:34:29 <quinq> dhruvin, so that's not the issue here
2023-05-26 13:35:06 <dhruvin> /lib/libgcc comes from that step, right?
2023-05-26 13:35:36 <quinq> You just agreed that woud go into /lib/i386-linux-gnu, son /lib
2023-05-26 13:35:39 <quinq> s/son/not/
2023-05-26 13:35:55 <dhruvin> from /lib/i386-linux-gnu to /lib
2023-05-26 13:36:26 <quinq> Is that how that command works? :/
2023-05-26 13:36:33 <dhruvin> let me check
2023-05-26 13:40:16 <quinq> Yep dhruvin seems you're right
2023-05-26 13:40:17 <quinq> If <src> is a directory, the entire contents of the directory are copied, including filesystem metadata.
2023-05-26 13:40:21 <quinq> The directory itself is not copied, just its contents.
2023-05-26 13:43:21 <dhruvin> Thanks for confirming
2023-05-26 13:43:29 <dhruvin> :)
2023-05-26 13:57:19 <kn> is there a fully automated way of installing alpine, something akin to https://man.openbsd.org/autoinstall.8 ? https://docs.alpinelinux.org/user-handbook/0.1a/Installing/setup_alpine.html reads like i still have to manually invoke setup-alpine (and fetch/put the answer file there myself?)
2023-05-26 14:08:30 <minimal> kn: an automated installer is currently being worked on
2023-05-26 14:30:22 <xgpt> I'm looking for a load-once and run-from-ram bistro for raspberry pi that will run more or less as an appliance to talk to and listen to a few serial devices. It's going to be headless, but will have a need for audio out. I think alpine will be a great fit, but I'm struggling with the onboarding & installation. I believe I want a "disk"
2023-05-26 14:30:22 <xgpt> installation, as I will need to maintain some persistence once running for updates to the setup scripts. The scripts are entirely python at the moment and will need to run a few pip installs once before doing final configuration and then the system won't need any reconfiguration after it's set out into the wild. It will need to recover gracefully
2023-05-26 14:30:22 <xgpt> from power loss, which is why I'm excited about the 'Disk' installation method
2023-05-26 14:30:29 <xgpt> am I looking at this correctly?
2023-05-26 14:33:11 <minimal> xgpt: no, "Sys mode" (aka disk) is NOT run-from-ram, it is run from disk
2023-05-26 14:33:53 <minimal> the usual run-from-ram install does provide persistence via the "lbu" command
2023-05-26 14:36:57 <xgpt> not sys mode
2023-05-26 14:37:34 <xgpt> sorry, not "disk" but "data"
2023-05-26 14:38:02 <xgpt> although I don't want data to persist that I don't tell it specifically to persist
2023-05-26 14:38:35 <xgpt> I basically want to boot into a Ramdisk image, download my packages, write my python script, write a boot script that launches my python script, commit that to disk, and then never write to disk again
2023-05-26 14:39:05 <xgpt> are you saying I don't want "data" but the regular alpine linux install and "lbu" will achieve that?
2023-05-26 14:41:08 <minimal> xgpt: nope, I didn't mention "data" mode at all. Have a read here as it covers all 3 modes: https://wiki.alpinelinux.org/wiki/Installation
2023-05-26 15:31:24 <wk> hey guys wassup 
2023-05-26 15:31:40 <wk> I have a orange pi zero , do you think I could port Alpine linux ? 
2023-05-26 15:36:33 <minimal> wk: you don't really need to "port" Alpine to it, but you'd need to sort out a bootloader for it and create a SDcard image for it
2023-05-26 15:36:57 <minimal> u-boot-sunxi appears to support the Orange Pi 3, not sure about Orange Pi Zero
2023-05-26 15:38:21 <minimal> also the kernel package might need some tweaks for Orange Pi Zero specific drivers
2023-05-26 15:39:33 <wk> https://github.com/moonbuggy/orange-pi-zero-alpine 
2023-05-26 15:39:44 <wk> I found that page But i think is outdated
2023-05-26 15:47:04 <wk> minimal: do you think I could boot with that Image and than just update the system ? 
2023-05-26 15:52:08 <minimal> no idea
2023-05-26 16:27:30 <kn> minimal: great to hear!
2023-05-26 17:19:24 <scoko> hi all, need some help configuring a private alpine linux mirror
2023-05-26 17:19:58 <scoko> i keep getting: "WARNING: updating and opening http://my.mirror.local/alpine/v3.18/main: No such file or directory
2023-05-26 17:20:40 <scoko> i have rsync'ed both v3.17 and v3.18 for the main branch, x86_64
2023-05-26 17:20:58 <scoko> i use nginx with the configuration given on https://wiki.alpinelinux.org/wiki/How_to_setup_a_Alpine_Linux_mirror
2023-05-26 21:19:26 <xgpt> What's the preferred way to start a program on reboot? I'm used to systemd unit files and don't know what I should be using in Alpine for custom scripts I need to run on reboot
2023-05-26 21:21:29 <lopid> just "a program" or a managed service?
2023-05-26 21:23:06 <lopid> see /etc/local.d/README for a simple case
2023-05-26 21:24:02 <xgpt> Idk, I usually start programs as a managed service I think? 
2023-05-26 21:24:06 <xgpt> Not sure lopid 
2023-05-26 21:24:51 <lopid> not sure why you're asking me what you need
2023-05-26 21:24:53 <xgpt> I think not local.d
2023-05-26 21:25:00 <xgpt> started this way would be needed, but may easily fail, e.g. if the .start script had been executed several times.
2023-05-26 21:25:00 <xgpt> Per Gentoo wiki This infrastructure should not be used to start other scripts or programs in the background. If the /etc/init.d/local service script is restarted several times, then those scripts or programs will be started in the background several times, possibly resulting in race conditions. A .stop script for terminating background processes
2023-05-26 21:25:08 <xgpt> In regards to local.d
2023-05-26 21:25:38 <xgpt> So I guess what's the preferred way to start a python script that's the entire point of this os install for me
2023-05-26 21:25:50 <xgpt> Like the entire point of the box is to run a python script 
2023-05-26 21:26:20 <minimal> xgpt: create a init.d script for it and enable it?
2023-05-26 21:28:10 <lopid> are you going to be stopping and starting the local service all the time?
2023-05-26 21:28:11 <stonks3141> Is it safe to remove eudev and eudev-dev and immediately add eudev=3.2.11 and eudev-dev=3.2.11?
2023-05-26 21:29:22 <minimal> stonkd3141: why would you want to do that?
2023-05-26 21:29:43 <stonks3141> minimal: eudev broke mysteriously for me and I want to see if it's because I updated
2023-05-26 21:30:09 <minimal> but the update would also have likely update their dependencies too...
2023-05-26 21:30:46 <minimal> what did you update from/to ?
2023-05-26 21:30:52 <stonks3141> I assume that would be fine since its a patch version and nothing in my system depends on 3.2.11
2023-05-26 21:31:17 <stonks3141> I'm not sure what version I was on before, probably 3.2.11 since I apk upgrade fairly often
2023-05-26 21:31:22 <stonks3141> I updated to 3.2.12
2023-05-26 21:31:34 <minimal> I'm asking about *alpine* versions
2023-05-26 21:31:40 <stonks3141> oh, edge
2023-05-26 21:32:14 <minimal> from 3.18 to Edge? or 3.17 to Edge?
2023-05-26 21:32:38 <stonks3141> no, I've always been on edge
2023-05-26 21:33:14 <stonks3141> I updated eudev from 3.2.11 to 3.2.12 recently and my system broke on next restart
2023-05-26 21:33:22 <stonks3141> specifically libinput
2023-05-26 21:33:35 <stonks3141> I'm trying to narrow down the problem
2023-05-26 21:35:02 <minimal> eudev depends on at least 6 other packages
2023-05-26 21:35:20 <minimal> previous versions of eudev may depend on previous versions of those packages
2023-05-26 21:35:55 <minimal> also does the Edge repo even still have the previous eudev package available?
2023-05-26 21:36:10 <stonks3141> I checked the apkbuild, it doesn't pin any of the dependencies to a specific version
2023-05-26 21:36:14 <stonks3141> could it break anyway?
2023-05-26 21:36:59 <stonks3141> I don't know if edge has 3.2.11, I just want to figure out if updating eudev was responsible for my problem
2023-05-26 21:37:01 <minimal> the previous version of eudev may wel have been compiled against previous versions of other stuff, so yes it "could" break
2023-05-26 21:37:02 <stonks3141> is that possible?
2023-05-26 21:41:20 <stonks3141> so it's not possible to downgrade eudev and I should try something else?
2023-05-26 21:42:09 <minimal> generally speaking Alpine repos only have a single version of each package
2023-05-26 21:42:46 <minimal> you could try to downgrade (if you can find the APK file(s)), if it works it works, if it does then it doesn't...
2023-05-26 21:43:03 <minimal> s/does then/doesn't then/
2023-05-26 21:43:16 <stonks3141> yeah I think I'll just try to figure it out another way
2023-05-26 21:43:23 <minimal> so what exactly is the issue?
2023-05-26 21:44:07 <stonks3141> I restarted, and after boot my wm (dwl) failed to start, saying that libinput was unable to find any devices
2023-05-26 21:44:14 <stonks3141> libinput list-devices has no output
2023-05-26 21:45:08 <stonks3141> and 'libinput debug-events --device /dev/input/mouse1' dies with 'udev device never initialized' after I move the mouse
2023-05-26 21:45:34 <minimal> so does /dev/input/ contain any/many entries?
2023-05-26 21:46:17 <stonks3141> yes, it has entries, their group is 'input' like it should be, and I can read from them with cat
2023-05-26 21:46:56 <powers> anyone familiar with iwd?
2023-05-26 21:47:15 <stonks3141> the last time I had this issue udev wasn't setting the group properly, but it is this time
2023-05-26 21:50:24 <minimal> stonks3141: did you check logfiles?
2023-05-26 21:50:25 <powers> iwd is not picking up any networks, even though iw and wpa_supplicant are
2023-05-26 21:53:11 <minimal> stonks: in Edge libinput package shows last buildtime as 25 March, whereas eudev-libs (which it depends on) shows last build 16 May. Perhaps libinput needs as rebuild as it seems it wasn't rebuilt since eudev version was updated
2023-05-26 21:53:22 <stonks3141> minimal: nothing out of the ordinary afaict in dmesg | grep '(input|udev|device)' or in udev's debug log
2023-05-26 21:53:43 <stonks3141> update: my wireless mouse works fine
2023-05-26 21:53:56 <stonks3141> it's just the trackpad and keyboard that aren't being detected
2023-05-26 21:55:07 <eex0> powers: there is only a small list of intel drivers fully supported by the kernel, perhaps that's why'iwd' isn't picking that card out?
2023-05-26 21:55:15 <minimal> stonks3141: ^^^
2023-05-26 21:55:43 <stonks3141> yeah I saw, is there anything I can do or just wait for rebuild?
2023-05-26 21:56:21 <minimal> psykose: what do you think? is a libinput rebuild required with the eudev version update?
2023-05-26 21:56:43 <aldcor>  powers i tried iwd on voidlinux and always had problems so I don't use it at all. Just write supplicant
2023-05-26 21:59:44 <powers> when i ran `iwmon` i saw two interesting things. 1 -> iwd seems to fail after one "Trigger Scan Response" but there is no error info. 2 -> iwd is only scanning 2.4GHz.
2023-05-26 22:00:57 <powers> eex0: does iwd only support intel cards? ive seen broadcom card support on ubuntu and arch, but the alpine build may be different
2023-05-26 22:02:06 <powers> aldcor: yeah wpa_supplicant works well, im just trying to get iwd working for people who may not have an option
2023-05-26 22:02:08 <eex0> I had just assumed that it was intel only by definition: iwd (iNet wireless daemon) is a wireless daemon for Linux written by Intel
2023-05-26 22:02:36 <psykose> minimal: no, that doesn't make any sense
2023-05-26 22:03:07 <psykose> nor is it how any of that works
2023-05-26 22:03:57 <powers> yeah iwd is an intel-made daemon but it interfaces with nl80211. i believe that means it is compatible with all cfg80211 drivers
2023-05-26 22:04:39 <psykose> i don't know where you come up with this hilarious stuff like 'iwd is only for intel' but it's certainly wrong
2023-05-26 22:05:16 <eex0> puffs & passes
2023-05-26 22:06:21 <stonks3141> psykose: the current build of libinput statically links eudev-dev-3.2.11 but eudev is at 3.2.12, how would that not be an issue?
2023-05-26 22:07:39 <psykose> no it doesn't
2023-05-26 22:07:41 <eex0> I'm done answering anything in this channel with even a Q? mark
2023-05-26 22:07:47 <psykose> where do you even come up with this stuff
2023-05-26 22:08:22 <stonks3141> um okay
2023-05-26 22:08:28 <psykose> libinput definitely does not "statically link" "eudev-dev-3.2.11"
2023-05-26 22:09:59 <psykose> you are the only person with this specific issue since the upgrade out of hundreds
2023-05-26 22:10:11 <psykose> have you considered that something in eudev itself doesn't configure the touchpad properly
2023-05-26 22:10:39 <psykose> like the new hwdb quirks or anything else :-)
2023-05-26 22:10:42 <stonks3141> yes, and minimal suggested that the version mismatch might cause issues so I took their word for it
2023-05-26 22:11:24 <psykose> well, it could also be complete magic, yes :p
2023-05-26 22:11:28 <minimal> stonks3141: yes I *suggested* that, but I never said anything about static linking
2023-05-26 22:12:11 <stonks3141> my mistake, it's not statically linked, but the build of libinput in the repos is from before eudev 3.2.12 was in aports
2023-05-26 22:15:37 <powers> you know? i really have to ask, does *anyone* here use iwd, is it functional?
2023-05-26 22:15:54 <stonks3141> powers: I do, it works better than wpa_supplicant
2023-05-26 22:16:05 <stonks3141> as in breaks less often
2023-05-26 22:16:47 <powers> because speaking of libraries and dependencies, i noticed that the offical iwd-repo depends on ell, but ell isn't listed as a dep in aports
2023-05-26 22:17:33 <psykose> because it statically links an ell it ships itself
2023-05-26 22:17:34 <powers> stonks3141: do you happen to have ell downloaded?
2023-05-26 22:17:50 <powers> psykose ah okay
2023-05-26 22:18:52 <powers> because when i was rummaging through the arch-packages i saw ell listed as a dep. alpine is a different distro, so everything doesnt have to be 1-to-1, but it was interesting
2023-05-26 22:19:53 <psykose> stonks3141: upgrade and find out
2023-05-26 22:20:01 <psykose> my money is on 'nothing changes' :p
2023-05-26 22:20:09 <psykose> or it fixes itself on rebooting regardless of anything else
2023-05-26 22:20:58 <stonks3141> I've rebooted multiple times
2023-05-26 22:23:33 <stonks3141> psykose: no change, thanks for rebuilding it anyway on the off chance it was broken
2023-05-26 22:23:41 <psykose> yeah
2023-05-26 22:23:49 <psykose> gotta account for magic sometimes
2023-05-26 22:24:47 <psykose> could you paste `rc-update` somewhere
2023-05-26 22:29:26 <stonks3141> psykose: https://envs.sh/hMR.txt
2023-05-26 22:30:08 <psykose> that's mdev and not eudev
2023-05-26 22:30:12 <psykose> setup-devd udev
2023-05-26 22:30:15 <psykose> you're welcome
2023-05-26 22:31:20 <stonks3141> rc-status -a says both mdev and udev are started
2023-05-26 22:32:14 <psykose> not relevant
2023-05-26 22:32:17 <psykose> please just run that command
2023-05-26 22:32:18 <psykose> ty
2023-05-26 22:34:46 <stonks3141> that fixed it, but I had a working system for 3 months before now
2023-05-26 22:34:48 <stonks3141> very strange
2023-05-26 22:36:16 <psykose> :)
2023-05-26 22:36:45 <stonks3141> can you explain what that did and why it was necessary?
2023-05-26 22:36:50 <stonks3141> I just want to learn
2023-05-26 22:43:42 <psykose> it made you use udev, which has multiple parts to it
2023-05-26 22:44:00 <psykose> before you had mdev with something else also starting 'udev'(not really) and it just happened to work by magic
2023-05-26 22:45:04 <stonks3141> interesting, thanks for explaining!
2023-05-26 22:47:49 <psykose> the reason it's multiple things is because it takes multiple parts to do to make it asynchronous (you can see in the new rc-update output)
2023-05-26 22:48:07 <psykose> theoretically it can just be one service but then you wait (many seconds) for all devices to get configures
2023-05-26 22:48:41 <psykose> so 'udev starts', trigger starts, then it waits for all netlink to finish, then does something after (postmount), so 4
2023-05-26 22:48:48 <psykose> could be 1 if you wait forever on 1
2023-05-26 22:48:55 <psykose> i really don't know how it worked before though
2023-05-26 22:49:16 <psykose> nothing would make it process things without the trigger, so mdev itself was probably doing good enough to configure your input
2023-05-26 22:53:52 <stonks3141> good to know :)
2023-05-26 23:15:33 <powers> very strange. iwd only processes three bands (all being 2.4GHz), while iw processes all 150+ bands
2023-05-26 23:16:36 <psykose> for scan?
2023-05-26 23:16:42 <powers> wpa_supplicant also performs a context-switch after brief failure that iwd is lacking. i know this works on arch, so its stranges
2023-05-26 23:16:48 <powers> yeah for scan
2023-05-26 23:16:55 <psykose> that seems like something else is wrong
2023-05-26 23:17:04 <psykose> i ran 5ghz-only at some point and iwd had no issues
2023-05-26 23:18:15 <powers> i want to believe its a driver conflict, but iw is performing and iwd is doing basic processing of my card
2023-05-26 23:18:47 <powers> iwmon shows the scan process concluding without failure
2023-05-26 23:19:31 <powers> and iwctl also can parse my driver capabilities, type, and mac-addr
2023-05-26 23:22:20 <powers> genuinely stumped, there's an error in the /var/log `Could not register frame watch type 00d0: -95` but this is expected and wpa_supp handles it well
2023-05-26 23:23:10 <powers> only thing i can think of is that this error is cardinal sin to iwd, making it refuse to work
2023-05-26 23:24:09 <powers> but if that were the case, why would scanning return a no-error?
2023-05-26 23:27:44 <powers> its also extremely unfortunate that the documentation for iwd is lagging because there likely is a setting equivalent to wpa_supp's `ap_scan=1`, but there's genuinely no way of telling other than going into source
2023-05-26 23:46:57 <psykose> ap_scan=1 is the default in wpa_supplicant
2023-05-26 23:47:09 <psykose> the others also aren't even for linux
2023-05-26 23:47:22 <psykose> ah
2023-05-26 23:47:25 <psykose> no, read some wrong docs
2023-05-26 23:48:06 <psykose> regardless, not related to anything here
2023-05-27 00:14:56 <wk> hey guys is there somebody here using alpine on Desktop that could share with us your experience? 
2023-05-27 00:21:15 <Saijin_Naib> Three yearsish now, across three different machines. Won't go back to any other distro 
2023-05-27 00:21:41 <Saijin_Naib> setup-desktop script fixes all the trouble I had setting up a DE from scratch before it existed
2023-05-27 00:22:14 <Saijin_Naib> Have used MATE, GNOME, XFCE, and LXQT 
2023-05-27 00:24:31 <XFaCiEer> Gotta catch them    all.
2023-05-27 00:25:15 <Saijin_Naib> Yeah, except Plasma 🤷
2023-05-27 00:25:20 <Saijin_Naib> Missing Budgie and LXDE
2023-05-27 00:25:22 <Saijin_Naib> And Cinnamon... 
2023-05-27 01:07:54 <xgpt> I see that raspberry pi images are available on the /downloads page, but I'm struggling to find good guides in either the /wiki or /docs of the website. i see a bunch of hacks and youtube videos. where should i go for the correct procedure to install alpine on a raspberry? I've been trying this on and off all day long and it's getting really
2023-05-27 01:07:54 <xgpt> frustrating :( 
2023-05-27 01:12:21 <xgpt> Just found this...3 days old...  Looks good?
2023-05-27 01:12:22 <xgpt> https://www.youtube.com/watch?v=jhtXjLUZB9g
2023-05-27 01:13:31 <minimal> xgpt: what's the issue? you download the Alpine RPI image, write it to a SDcard and boot, then run setup-alpine
2023-05-27 01:13:36 <Saijin_Naib> I just use XFCE now, but DE hopping was fun and worthwhile, I think 
2023-05-27 01:14:04 <minimal> actually, forget that, it's ready to run from the SDcard image isn't it
2023-05-27 01:27:18 <xgpt> oh wait
2023-05-27 01:27:24 <xgpt> is it that simple?
2023-05-27 01:27:50 <xgpt> so does it boot the entire system to ram, and then I can safely repartition the disk as desired from within setup-alpine?
2023-05-27 01:27:52 <xgpt> minimal
2023-05-27 01:28:40 <psykose> pretty much
2023-05-27 01:29:15 <psykose> you format an sdcard as fat32 then write the .tar.gz to it
2023-05-27 01:29:22 <minimal> from memory you'd have to unmount the SDcard before running setup-alpine to install onto SDcard (whether the same card or a new one)
2023-05-27 01:29:39 <psykose> ye
2023-05-27 01:29:56 <psykose> if you have a non-scuffed setup you could install to a usb instead of using the sdcard i guess, it's what i did
2023-05-27 01:30:01 <xgpt> I'm just getting a weird hang on the rainbow screen :(
2023-05-27 01:30:08 <xgpt> scuffed?
2023-05-27 01:30:28 <xgpt> but setup-alpine doesn't require that the drive that booted it be mounted?
2023-05-27 01:30:49 <xgpt> so it can install the system entirely from the in-ram system? that's such a neat party trick
2023-05-27 01:31:04 <psykose> if it hangs on rainbow you have some other issues i guess, i don't remember how that works
2023-05-27 01:32:36 <xgpt> I'll try flashing it again, what filesystem does it need? mkfs.msdos mkfs.vfat or mkfs.fat?
2023-05-27 01:32:56 <xgpt> oohhhhhh
2023-05-27 01:33:01 <xgpt> I forgot the -F 32!
2023-05-27 01:33:45 <minimal> depending on which RPI you have it can be fussy about which FAT version
2023-05-27 01:39:53 <xgpt> ugh, no bueno
2023-05-27 01:39:54 <xgpt> :(
2023-05-27 01:43:12 <xgpt> it's saying 'resource busy' when i try to umount /media/mmcblk0
2023-05-27 01:43:44 <minimal> xgpt: from memory becase there is *another* mount layered on top of that
2023-05-27 01:48:10 <xgpt> so how would I umount it?
2023-05-27 01:48:26 <minimal> unmount the mount "on top" first
2023-05-27 01:48:54 <xgpt> blargh, so I should probably have one partition that is of the appropriate FAT32 and then have a giant ext4 with the rest of the disk for my 'disk' install ?
2023-05-27 01:48:55 <minimal> or perhaps you've got a shell in the /media/mmcblk0 directory?
2023-05-27 01:49:04 <xgpt> not sure
2023-05-27 01:49:09 <xgpt> have to start over though :(
2023-05-27 01:49:21 <minimal> disk install? you mean "Sys"? or "Data"?
2023-05-27 01:49:39 <xgpt> i mean data
2023-05-27 01:49:54 <xgpt> and I'm starting to think I've got a reading issue because I've been saying the one thing and meaning another all day
2023-05-27 01:59:03 <eex0> powers: thank you for the idea I got earlier to write apparmor profiles for network listing/configuration utilities allowing for only certain users on the machine to have the tools function for.  
2023-05-27 02:01:59 <eex0> ifconfig, iwd, macchanger, etc allowed for a single doas user so now other users non-doas those tools become useless for. I may attempt binutils this way aswell.. again, thank you
2023-05-27 02:04:31 <powers> eex0: np! though i didnt really do anything, you crafted the idea on your own
2023-05-27 02:06:02 <powers> just a question though, do programs not have that functionality currently through group privileges?
2023-05-27 02:08:04 <powers> for example, if your a not a superuser wanting to interact with iwd, you would be put in the netdev group
2023-05-27 02:10:27 <powers> unless you are referring to ifconfig, ip, and iw - in which case, i do not think a feature like that currently exists
2023-05-27 02:12:26 <powers> i see a world where people would use that in a high-security multiuser networking-stack
2023-05-27 03:08:41 <eex0> I dub'thee latest firefox "beans-spilt" normandy wants all!
2023-05-27 03:10:29 <psykose> do less drugs
2023-05-27 03:11:02 <eex0> if you say this once more, I won't be pleasant toward you! 
2023-05-27 03:11:10 <eex0> 20:05:44.202 1685156744202	app.normandy.recipe-runner	ERROR	Error checking filter for "Heartbeat: Daily Viewpoint Survey (es-ES)". Filter: [(normandy.channel in ["release"]) && (normandy.locale in ["es-ES"]) && (['rolling-viewpoint', normandy.userId]|bucketSample(19468 + normandy.request_time / (24*60*60*1000) + 7, 7, 7000))]. Error: "TypeError: NetworkError when attempting to fetch resource." Log.sys.mjs:722
2023-05-27 03:11:32 <eex0> and tomns more, dont need posting, find it yourslef
2023-05-27 03:13:09 <eex0> when you get me angry typo typing Ive had enough!
2023-05-27 03:13:49 <eex0> I am a recovering addict, remind me again please psykose!
2023-05-27 03:53:24 <xgpt> I have my mmc drive partitioned as such, mmcblk0p1 == (vfat) /boot, but only 512MB. mmcblk0p2 == (ext4) with the intention of using it for a 'data' install.....
2023-05-27 03:53:24 <xgpt> but I can't figure out how to get setup-alpine or setup-disk to see the second partition
2023-05-27 03:53:35 <xgpt> what am I doing wrong and where is the right guide for me?
2023-05-27 04:10:35 <powers> where are udhcpc configs read from?
2023-05-27 04:11:02 <powers> ive tried /etc/udhcpc/*.conf but to no avail
2023-05-27 04:11:19 <powers> is it /var/lib/udhcpc?
2023-05-27 04:20:11 <powers> forget it, /etc/udhcpd.conf is the actual location
2023-05-27 04:20:25 <powers> i thought it was a dummy file
2023-05-27 04:33:22 <psykose> that's udhcpd not udhcpc
2023-05-27 04:33:34 <psykose> /etc/udhcpc/udhcpc.conf is read
2023-05-27 04:42:47 <powers> ive set desired nameservers in both /etc/udhcpd.conf and /etc/udhcpc/udhcpc.conf, but when i reboot it seems to use neither
2023-05-27 04:43:53 <psykose> the former is literally not used
2023-05-27 04:43:58 <psykose> what did you put in the latter
2023-05-27 04:44:18 <powers> okay so it was a dummy file
2023-05-27 04:44:20 <psykose> the correct thing is something like dns="1.2.3.4"
2023-05-27 04:44:26 <powers> ooooooo
2023-05-27 04:44:37 <xgpt> okay, figured out the entire process
2023-05-27 04:44:42 <powers> i put `opt dns 1.2.3.4`
2023-05-27 04:44:45 <psykose> haha
2023-05-27 04:44:58 <xgpt> am struggling with how to edit usercfg.txt from within alpine linux though :(
2023-05-27 04:45:13 <psykose> you just edit it
2023-05-27 04:47:11 <xgpt> it appears as though /media/mmcblk0p1/ is read only?
2023-05-27 04:54:16 <powers> thank you for the help btw
2023-05-27 04:57:59 <xgpt> psykose I opened vim /media/mmcblk0p1/usercfg.txt and it's not letting me save. how do I fix this permissions error?
2023-05-27 04:58:17 <xgpt> I'm not even sure why it's mounted if the system is truly running from RAM
2023-05-27 04:58:56 <psykose> you would have to remount it rw yes
2023-05-27 05:00:20 <xgpt> oh, gotcha
2023-05-27 05:00:49 <xgpt> just got that actually, I'm quickly learning that my gaps in linux knowledge are not alpine's fault :P
2023-05-27 05:01:10 <xgpt> and I get better returns on queries that replace "how do I X linux" for "how do I X alpine"
2023-05-27 05:02:16 <xgpt> although I am not sure how to get audio working on alpine on the raspberry pi
2023-05-27 05:02:25 <xgpt> any tips on guides?
2023-05-27 05:02:40 <xgpt> alpine:~# alsamixer
2023-05-27 05:02:40 <xgpt> ALSA lib confmisc.c:855:(parse_card) cannot find card '0' << 
2023-05-27 05:04:34 <psykose> good question
2023-05-27 05:04:37 <psykose> i never tried that
2023-05-27 05:05:14 <xgpt> ugh, I'm building a serial interface that uses 'espeak' to communicate with a blind user and I just bought a pi zero w because it's so low demand
2023-05-27 05:05:23 <xgpt> I really need sound :(
2023-05-27 05:05:30 <xgpt> I'm not even using a monitor
2023-05-27 05:06:51 <xgpt> who should I reach out to for help with this? Will the raspberry pi forums be helpful to someone who doesn't want Debian?
2023-05-27 05:07:31 <xgpt> I've actually gotten it working on their raspbian but it's a little laggy and leaves little headroom for other software I'd like to use. I think it will fly on alpine though
2023-05-27 05:13:39 <psykose> based on random shit i find you have to configure it
2023-05-27 05:14:46 <xgpt> https://pastebin.com/JJLSjd7P 
2023-05-27 05:14:58 <xgpt> that's a log of what I get when I run 'espeak test'
2023-05-27 05:15:20 <psykose> yes, there's no audio card
2023-05-27 05:15:21 <xgpt> plus uname-a and cat/etc/os-release
2023-05-27 05:15:26 <xgpt> but there is...
2023-05-27 05:15:29 <psykose> apparently you need dtparam=audio=on
2023-05-27 05:15:53 <xgpt> alpine:~# cat /media/mmcblk0p1/usercfg.txt
2023-05-27 05:15:53 <xgpt> alpine:~#
2023-05-27 05:15:53 <xgpt> dtparam=audio=on
2023-05-27 05:18:01 <psykose> are you sure config.txt has a `include usercfg.txt` (it should)
2023-05-27 05:18:09 <psykose> and that you rebooted after putting that
2023-05-27 05:18:14 <psykose> i just tried it and now it shows up
2023-05-27 05:18:29 <xgpt>  yes
2023-05-27 05:19:44 <psykose> and it's in /media/mmcblk0p1/config.txt right
2023-05-27 05:19:55 <psykose> (same folder)
2023-05-27 05:20:23 <xgpt> [all]
2023-05-27 05:20:23 <xgpt> include usercfg.txt
2023-05-27 05:20:23 <xgpt> alpine:~#
2023-05-27 05:20:49 <xgpt> yeah, didn't monkey with that file yet
2023-05-27 05:21:31 <xgpt> do you think it's because there's no hdmi plugged in??
2023-05-27 05:22:13 <psykose> idk what configuration the 0w has
2023-05-27 05:22:27 <psykose> you normally have to pick between the jack and the hdmi, but 'cannot find card' is before that
2023-05-27 05:22:35 <psykose> i.e. separate
2023-05-27 05:22:43 <psykose> ah
2023-05-27 05:22:47 <psykose> did you `apk add alsa-utils`
2023-05-27 05:23:08 <psykose> well, yeah you would have
2023-05-27 05:23:09 <psykose> hmm
2023-05-27 05:23:10 <psykose> not sure then
2023-05-27 05:23:45 <psykose> just worked for me on an rpi4b (didn't test if it actually outputs audio but it's all detected so probably works)
2023-05-27 05:23:49 <xgpt> yeah I did
2023-05-27 05:23:59 <xgpt> I mean, I did ask add alsa-utils
2023-05-27 05:24:24 <psykose> the next obvious thing to try is to put it in config.txt and not usercfg.txt i guess, but based on the layout that won't make a difference
2023-05-27 05:24:25 <xgpt> maybe I need to....lemme try leu committing that and rebooting 
2023-05-27 05:24:48 <psykose> if you rebooted and the modification was still there, then it was in effect
2023-05-27 05:24:53 <xgpt> lbu**
2023-05-27 05:24:57 <psykose> if it was gone, then no
2023-05-27 05:25:08 <xgpt> the modification was there but alsa-utils wasn't there
2023-05-27 05:25:19 <psykose> ah
2023-05-27 05:25:21 <xgpt> the config line was but the utils weren't? maybe...it needs them on boot?
2023-05-27 05:25:29 <psykose> low chance but perhaps
2023-05-27 05:25:36 <xgpt> nope
2023-05-27 05:25:54 <psykose> you might also want to run `setup-devd mdevd` just to use the better device manager but i doubt it matters either
2023-05-27 05:27:46 <xgpt> I'll try that next
2023-05-27 05:30:23 <xgpt> scanning
2023-05-27 05:31:22 <xgpt> nope
2023-05-27 05:33:25 <psykose> can you paste dmesg?
2023-05-27 05:33:55 <psykose> this is mine, on an rpi4: https://img.ayaya.dev/oGF0ri3N4J04
2023-05-27 05:34:03 <psykose> you can see bcm2835_audio being loaded
2023-05-27 05:35:20 <psykose> hm
2023-05-27 05:35:47 <psykose> in /media/mmcblk0p1/config-rpi, do you see CONFIG_SND_BCM2835=m
2023-05-27 05:36:07 <psykose> it might be default-off for some reason on the armhf one (in the default config for it)
2023-05-27 05:36:22 <psykose> assuming this is even the audio on that hardware
2023-05-27 05:36:24 <psykose> not sure what it is
2023-05-27 05:38:24 <xgpt> psykose https://termbin.com/gzcm
2023-05-27 05:39:37 <xgpt> it's absent
2023-05-27 05:40:18 <xgpt-ZzzSLEEPzzZ> I Idle on here, feel free to ping me and I'll respond :)
2023-05-27 05:41:19 <psykose> wait
2023-05-27 05:41:27 <psykose> does the 0w even have an audio jack?
2023-05-27 05:41:41 <psykose> what were you trying
2023-05-27 05:41:50 <psykose> from the look of it it does not
2023-05-27 05:42:13 <psykose> so it needs more configuration for whatever addons that would give one
2023-05-27 06:54:24 <powers> is there a mechanism is alpine to mount an encrypted-disk and sign-in at the same time?
2023-05-27 06:56:17 <powers> *assuming that your mount and account passwords are the same
2023-05-27 07:45:03 <powers> btw, do you all use awall or iptables?
2023-05-27 07:46:13 <powers> awall looks nice, but im not sure if its preferred
2023-05-27 07:58:43 <psykose> usually nft
2023-05-27 09:05:59 <uniqNuuid> awall firewalld ufw those are distro specific software, they are not general. iptables if you are already familiar with it, nft if you want to future ready 
2023-05-27 10:07:45 <Ermine> ufw and firewalld are not distro-specific.
2023-05-27 10:12:42 <dwfreed> ^^^
2023-05-27 14:44:08 <dalias> i'm trying to boot alpine on a libre computer le potato. anyone know if the generic arm kernel is capable of working here?
2023-05-27 14:44:22 <dalias> i'm getting no output after u-boot passes off execution to kernel
2023-05-27 15:50:43 <fdfkdfhkhdfkjgh> I was unable to install alpine linux on a flash drive. At the stage from the image losst.pro/wp-content/uploads/2019/11/12.png, the terminal does not accept sdb or sdb2 as a response. How to do?
2023-05-27 15:52:41 <psykose> why are you installing 3.10
2023-05-27 15:53:08 <donoban> uhm, another install problem here
2023-05-27 15:54:05 <psykose> wait is that image actually from 2019
2023-05-27 15:54:28 <psykose> that's certainly one way to be funny i guess
2023-05-27 15:54:31 <donoban> https://postimg.cc/jL1z5wwC
2023-05-27 15:55:01 <psykose> donoban: what did you run
2023-05-27 15:55:03 <donoban> this is a fresh 3.18 virt-img, booted, sys-install, rebooted and run setup-desktop with gnome
2023-05-27 15:55:26 <donoban> it seems that gdm was not installed
2023-05-27 15:55:35 <psykose> sure, it's missing some deps
2023-05-27 15:55:42 <psykose> well
2023-05-27 15:55:47 <psykose> no, gnome adds gdm
2023-05-27 15:55:53 <donoban> with my other vm I can't get VNC work when I start both gdm/ligthdm
2023-05-27 15:56:17 <donoban> nothing relevant on logs :(
2023-05-27 15:56:47 <donoban> let try to debug the setup-desktop script
2023-05-27 15:56:55 <psykose> just post the entire log and not 3 lines
2023-05-27 15:57:01 <psykose> idk why everyone does this thing where they hide all the info
2023-05-27 15:57:22 <donoban> well, it's on VNC so not easy to copy/paste
2023-05-27 15:57:36 <donoban> and honestly I didn't expect it fails :D
2023-05-27 15:59:25 <donoban> now I retried it and it wants to install another 496 pacakges
2023-05-27 15:59:46 <psykose> please just post the logs
2023-05-27 16:00:00 <fdfkdfhkhdfkjgh> @psykose, i started alpine-extended-3.17.2-x86_64
2023-05-27 16:00:13 <psykose> no you didn't because your screenshot says 3.10
2023-05-27 16:00:35 <donoban> there is an issue, each time I run setup-desktop it adds another http://eu.edge.kernel.org/alpine/v3.18/community
2023-05-27 16:00:42 <donoban> to /etc/apk/repositories
2023-05-27 16:00:57 <psykose> sounds like the sed thing in setup-xorg-base
2023-05-27 16:02:38 <donoban> now it workded  * service gdm added to runlevel default
2023-05-27 16:02:52 <donoban> I could repeat the whole process to check if it's reprucible
2023-05-27 16:03:13 <fdfkdfhkhdfkjgh> @psykose, this screenshot is an analogy just to show at what stage the problem appeared
2023-05-27 16:04:11 <donoban> looking at 'setup-desktop' the repo enabling problably comes from setup-xorg-base
2023-05-27 16:04:16 <psykose> fdfkdfhkhdfkjgh: so it's useless? glad we agree
2023-05-27 16:04:26 <psykose> donoban: yeah it does
2023-05-27 16:04:38 <donoban> ok so it's already known/fixed?
2023-05-27 16:04:41 <psykose> the only failure case i can think of is the community/ repo failed to add correctly so the apks were missing
2023-05-27 16:04:42 <psykose> no
2023-05-27 16:04:44 <psykose> i mean why it failed
2023-05-27 16:04:56 <psykose> the only way the thing fails is that the repo didn't get added, so there was only main
2023-05-27 16:04:58 <psykose> hmm
2023-05-27 16:05:02 <psykose> but then apk add would also fail
2023-05-27 16:05:03 <psykose> shrug
2023-05-27 16:05:09 <psykose> you didn't post the logs which would be really nice
2023-05-27 16:05:17 <donoban> let me retry, I will produce a full log
2023-05-27 16:05:31 <donoban> but first let check if input works on gdm
2023-05-27 16:05:51 <donoban> I would like to test pam :D
2023-05-27 16:06:04 <donoban> yeahh
2023-05-27 16:06:10 <donoban> it works!
2023-05-27 16:06:18 <psykose> :)
2023-05-27 16:06:20 <donoban> I don't understand what's wrong with my other vm
2023-05-27 16:06:28 <donoban> I updated it to edge and stills the same problem
2023-05-27 16:06:47 <donoban> but well, it was a vm just for testing pam
2023-05-27 16:07:03 <donoban> well let's to reinstall from scratch again
2023-05-27 16:07:13 <psykose> fdfkdfhkhdfkjgh: the actual issue is probably that you don't have an sdb, but you didn't post anything else to know for sure :-)
2023-05-27 16:07:18 <psykose> nobody can read your screen
2023-05-27 16:09:45 <donoban> I was wondering if the fast repository detection could be paralelized
2023-05-27 16:10:19 <psykose> i don't think it matters
2023-05-27 16:10:20 <donoban> running ping on the background/subshell and storing the result on /tmp/xxxx
2023-05-27 16:10:35 <psykose> that also inherently makes it more noisy
2023-05-27 16:10:45 <psykose> if anything it should be reworked to not recommend mirrors as much
2023-05-27 16:10:57 <donoban> relly on the cdn domain?
2023-05-27 16:11:06 <fdfkdfhkhdfkjgh> @psykose, if I didn't have sdb then I wouldn't be able to run the mount /dev/sdb1 /media/sdb1 command
2023-05-27 16:11:29 <donoban> if you don't know what to do f) sounds the more reasonable option
2023-05-27 16:11:40 <donoban> but it's sloooooooow
2023-05-27 16:11:54 <psykose> i.e. instead of printing a huge list that goes off your screen, it's one more modal
2023-05-27 16:11:59 <psykose> 1) pick dl-cdn
2023-05-27 16:12:08 <psykose> 2) pick 3rd-party mirror (current options)
2023-05-27 16:12:09 <psykose> idk
2023-05-27 16:13:04 <donoban> dl-cdn already routes you with some near geolocation?
2023-05-27 16:13:37 <fdfkdfhkhdfkjgh> What now?
2023-05-27 16:14:42 <donoban> fdfkdfhkhdfkjgh: you should know what do you want to mount :\, it's a USB pendrive? try to look on dmesg after plugging it
2023-05-27 16:16:01 <donoban> 
ACTION stills pinging apk mirrors

2023-05-27 16:17:52 <psykose> donoban: yeah, it's geolocated
2023-05-27 16:18:32 <donoban> great! problem happened again
2023-05-27 16:19:12 <donoban> maybe this is enough for understand what happens
2023-05-27 16:19:14 <donoban> https://tpaste.us/4az4
2023-05-27 16:19:48 <donoban> main failed?
2023-05-27 16:19:50 <donoban> uh
2023-05-27 16:20:23 <donoban> well
2023-05-27 16:20:45 <donoban> here is full: https://tpaste.us/MRwm
2023-05-27 16:20:47 <psykose> urgh
2023-05-27 16:20:55 <donoban> I have the feel that it installed packages from setup-xorg-base
2023-05-27 16:21:00 <donoban> but don't frm setup-desktop
2023-05-27 16:21:05 <psykose> nah
2023-05-27 16:21:12 <psykose> you're thinking of it wrong, it's one apk add down the chain
2023-05-27 16:21:17 <psykose> just read the scripts, it's <100 lines
2023-05-27 16:21:34 <psykose> basically setup-desktop passes stuff to setup-xorg-base, which enables community, apk update, apk add
2023-05-27 16:21:38 <psykose> and the update fails
2023-05-27 16:21:41 <psykose> to all of main is missing
2023-05-27 16:21:42 <psykose> that's all
2023-05-27 16:21:45 <psykose> it doesn't handle the update failing
2023-05-27 16:21:55 <psykose> you should open an alpine-conf issue that it.. should
2023-05-27 16:21:58 <psykose> at least || exit 1 is better
2023-05-27 16:22:14 <donoban> what means
2023-05-27 16:22:17 <donoban> v3.18.0-391-g76d02640033 [http://dl-cdn.alpinelinux.org/alpine/v3.18/main]
2023-05-27 16:22:35 <psykose> it's the older version
2023-05-27 16:22:44 <donoban> uhM
2023-05-27 16:22:47 <psykose> but you can see it then does '1 stale;'
2023-05-27 16:22:48 <donoban> strange
2023-05-27 16:22:54 <psykose> and the new apk-tools ignores stale repos during transaction
2023-05-27 16:22:56 <donoban> on the second run it shows the same error/warning
2023-05-27 16:23:00 <donoban> but it installs more packages
2023-05-27 16:23:04 <psykose> paste it
2023-05-27 16:23:07 <donoban> wtf
2023-05-27 16:23:18 <donoban> I created the img 2G instead 20G
2023-05-27 16:23:22 <donoban> :@
2023-05-27 16:23:34 <donoban> noooooo
2023-05-27 16:23:43 <donoban> I can't scrollback enough now
2023-05-27 16:23:46 <psykose> :D
2023-05-27 16:23:59 <donoban> beh
2023-05-27 16:24:07 <donoban> it's fast to retry if I select cdn mirror :)
2023-05-27 16:28:27 <donoban> ouhhchh
2023-05-27 16:29:17 <donoban> https://i.postimg.cc/Zq46jmZW/screenshot-2023-05-27-182833.png
2023-05-27 16:30:03 <donoban> what failed now :(
2023-05-27 16:31:11 <donoban> I unmounted sda, and run setup-disk again
2023-05-27 16:31:14 <donoban> it worked fine this tme
2023-05-27 16:31:55 <donoban> I feel that something with apk is unreliable
2023-05-27 16:32:22 <donoban> like fails without error and the calling script continues..
2023-05-27 16:33:05 <psykose> looks like random networking that doesn't abort in the middle
2023-05-27 16:33:51 <donoban> setup-desktop starts the same way
2023-05-27 16:33:57 <donoban> first run installs 147 packages
2023-05-27 16:34:09 <psykose> yes and i already told you what to report there
2023-05-27 16:34:25 <donoban> well I want to post the 2º run log
2023-05-27 16:35:01 <donoban> https://tpaste.us/EJ5R
2023-05-27 16:35:21 <donoban> uhMm, this time seems that main didn't failed
2023-05-27 16:36:16 <donoban> I think that the other time it failed but meh, probably I was wrong
2023-05-27 16:36:21 <msiism> Having thought about it more, it now seems to me that the easiest way to have Alpine with full disk encryption is to set up the disk manually and then use setup-alpine with a modified setup-disk command.
2023-05-27 16:36:53 <msiism> Like, e.g., `BOOTLOADER=grub setup-disk -m sys /mnt` or some such.
2023-05-27 16:37:00 <donoban> so this "WARNING" WARNING: updating http://dl-cdn.alpinelinux.org/alpine/v3.18/main: temporary error (try again later)
2023-05-27 16:37:20 <donoban> should make setup-desktop fail, or should fail apk in any case? uhM
2023-05-27 16:49:52 <donoban> psykose: maybe the problem is the subshell apk info
2023-05-27 16:50:06 <donoban>  setup-xorg-base ${BROWSER:-firefox} $(apk info --quiet --depends gnome gnome-apps-core) "$@" 
2023-05-27 16:50:29 <donoban> that will expain what the world didn't had anything about gnome
2023-05-27 16:51:59 <donoban> uhM
2023-05-27 16:52:09 <donoban> if setup-xorg-base enables community repository
2023-05-27 16:52:20 <psykose> oh
2023-05-27 16:52:21 <donoban> but apk info wants to find gnome depencies before enabling it..
2023-05-27 16:52:23 <psykose> yeah you're right
2023-05-27 16:52:23 <donoban> that's it :D
2023-05-27 16:53:12 <psykose> open an issue
2023-05-27 16:53:28 <donoban> I've already opened one but I'm gonna edit it
2023-05-27 16:53:46 <psykose> thanks
2023-05-27 16:53:57 <donoban> :)
2023-05-27 17:02:11 <quinq> psykose, it seems my messages don't reach you on Libera
2023-05-27 17:03:28 <quinq> (if you're the same person)
2023-05-27 17:07:50 <msiism> Is it normal for `setup-keymap` to fail because it can't find available keyboard layouts on the Alpine extended image?
2023-05-27 17:13:03 <msiism> Okay, setup-timezone also isn't going anywhere here.
2023-05-27 17:13:03 <donoban> msiism: no, is there some error?
2023-05-27 17:13:27 <donoban> it's a fresh iso boot? looks that something failed on start
2023-05-27 17:13:29 <msiism> Yeah, not sure how to get a verbatim copy now, though…
2023-05-27 17:13:49 <msiism> Yeah, it's the stable extended ISO I just put on a USB key.
2023-05-27 17:13:58 <msiism> I could try rebooting
2023-05-27 17:14:22 <donoban> I suppose that is not memory constrained
2023-05-27 17:14:46 <msiism> In what way? The machine has 2GB memory.
2023-05-27 17:15:52 <donoban> uhmm.... that should be enough but I'm not 100% sure for extended image
2023-05-27 17:15:52 <msiism> So, let me reboot.
2023-05-27 17:17:02 <donoban> uhmm. 840MB..
2023-05-27 17:17:57 <msiism> There are no errrors when booting.
2023-05-27 17:17:58 <donoban> please if you reboot check that nothing failed on start due not enought space left
2023-05-27 17:18:09 <donoban> anyway let me try to boot it on a VM
2023-05-27 17:19:07 <msiism> `free` after login says I have 1.7G free memory.
2023-05-27 17:19:20 <donoban> and df -h ?
2023-05-27 17:20:32 <msiism> Reports 100% used for /dev/sdb (the USB key) and /dev/loop0. All other are either 0% or 1% used.
2023-05-27 17:20:47 <donoban> well I booted it on a VM with 2GB, setup-keymap works fine
2023-05-27 17:21:10 <msiism> OH, and it does now for me…
2023-05-27 17:21:16 <msiism> I should have recorde the error…
2023-05-27 17:21:25 <donoban> the problem must be another thing
2023-05-27 17:21:40 <msiism> It said something about not being able to find dependencies.
2023-05-27 17:21:48 <msiism> Some apk-ish stuff.
2023-05-27 17:21:58 <msiism> And then the needed files didn'T exist.
2023-05-27 17:22:28 <msiism> The image should be okay. I verified the checksum and the signature.
2023-05-27 17:22:41 <donoban> try
2023-05-27 17:24:26 <donoban> apk add --virtual .setup-keymap-deps kbd-bkeymap
2023-05-27 17:26:02 <msiism> Okay, that returns an error.
2023-05-27 17:27:24 <donoban> uhM, not sure if is the problem
2023-05-27 17:28:42 <msiism> Let me setup netwokring and then pastebin the log.
2023-05-27 17:29:53 <donoban> can you just do: apk add kbd-bkeymap ?
2023-05-27 17:29:57 <msiism> http://termbin.com/497m
2023-05-27 17:30:28 <msiism> Says no such package.
2023-05-27 17:30:40 <msiism> But setup-keymap works now.
2023-05-27 17:30:45 <msiism> After the reboot, I mean.
2023-05-27 17:30:53 <msiism> Same for setup-timezone.
2023-05-27 17:30:55 <donoban> lol, this is pretty weird
2023-05-27 17:31:05 <donoban> I also had similar error trying to debug it
2023-05-27 17:31:19 <msiism> Cosmic rays, maybe.
2023-05-27 17:31:35 <psykose> bkeymaps not bkeymap
2023-05-27 17:31:52 <msiism> Oh, ok, that woked.
2023-05-27 17:32:07 <donoban> ouch
2023-05-27 17:33:35 <msiism> Okay, so I'll use that system to set up my disk now.
2023-05-27 17:40:35 <msiism> Uhhm… cryptsetup has terrible error reporting for invocation errors, by the way.
2023-05-27 17:41:14 <donoban> ;(
2023-05-27 17:41:53 <msiism> I just got "invalid numeric value" because I forgot the first hyphen in "--iter-time". ;)
2023-05-27 17:50:14 <donoban> :)
2023-05-27 17:55:31 <quinq> msiism, most likely you passed the "ter-time" argument to option -i
2023-05-27 17:55:41 <quinq> And that's not a numeric value
2023-05-27 18:07:15 <msiism> I see.
2023-05-27 18:07:40 <msiism> Well, error messages should include a context field for that very reason.
2023-05-27 18:07:56 <msiism> invalid argument -- 'ter-time'
2023-05-27 18:08:01 <msiism> That would be very telling.
2023-05-27 18:08:18 <msiism> numeric argument, even.
2023-05-27 18:11:04 <quinq> Yeah, or rather the flag that failed to be parsed correctly
2023-05-27 18:11:08 <quinq> (here -i)
2023-05-27 18:13:21 <msiism> Yeah, like: invalid argument to option '-i' -- 'ter-time'
2023-05-27 22:35:15 <thought> Gamers.
2023-05-27 22:35:18 <thought> I have no clue what I'm doing wrong.
2023-05-27 22:35:28 <thought> My script is sometimes running in cron, and sometimes nope.
2023-05-27 22:35:45 <thought> When it doesn't run, I get: user nvr: process already running: /usr/local/bin/ipcam.sh 8106 IPCam6
2023-05-27 22:36:11 <thought> I've looked this up and literally only get insane blogspam about running a process even though it already IS running.
2023-05-27 22:36:16 <thought> Is there some config I need to look for?
2023-05-27 22:37:23 <psykose> that happens when the script didn't exit from the previous run?
2023-05-27 22:38:14 <thought> psykose, are you asking me?
2023-05-27 22:38:29 <psykose> gamers
2023-05-27 22:39:06 <thought> Does it have to exit to be run again? To my understanding, cron usually just does it anyway, right?
2023-05-27 22:40:24 <thought> Oh, also, I tried to then add >> /home/nvr/output.log, and no, I don't get anything, it makes a file, though.
2023-05-27 22:40:36 <thought> That's in the script, of course.
2023-05-27 22:40:39 <psykose> it's pretty easy to test, i don't actually know what the per-crond behaviour is
2023-05-27 22:40:42 <thought> Maybe ffmpeg is eating it.
2023-05-27 22:40:46 <psykose> and you didn't post what you run
2023-05-27 22:41:01 <thought> psykose, I'll post it if someone's interested, but I didn't want to wave it around if it was something obvious.
2023-05-27 22:41:15 <psykose> more info > less info
2023-05-27 22:41:53 <thought> https://paste.ee/p/6DED2
2023-05-27 22:43:16 <thought> I assume ffmpeg isn't actually giving me anything to actually output. I could test that manually.
2023-05-27 22:43:44 <psykose> why do you run this every 5 minutes instead of just constantly
2023-05-27 22:44:45 <thought> psykose, mostly because if something goes wrong, I want 5 minute increments of the video. I had the same idear.
2023-05-27 22:46:03 <thought> I'd do it constantly if things like network interruptions wouldn't just kill ffmpeg, and writing a whole watchdog seemed like eighty times harder than this.
2023-05-27 23:02:32 <dwfreed> if you start it in an initscript, openrc's supervise-daemon can keep it alive
2023-05-27 23:02:42 <dwfreed> no need to do your own watchdog
2023-05-27 23:03:09 <psykose> you'll have to tweak the death limit tho
2023-05-27 23:03:16 <dwfreed> sure
2023-05-27 23:03:16 <psykose> well
2023-05-27 23:03:18 <psykose> actually no
2023-05-27 23:03:25 <psykose> ffmpeg can just write periodic itself
2023-05-27 23:03:48 <psykose> (so you don't have to deathloop it)
2023-05-27 23:03:49 <psykose> iirc
2023-05-27 23:05:32 <psykose> the -f segment -strftime 1 -segment_time "strftime-string-out" already does that
2023-05-27 23:05:39 <psykose> you just have it locked to -t
2023-05-27 23:05:45 <psykose> if you merely remove it it works without dying lol
2023-05-27 23:06:36 <psykose> not exactly what i'd call super super robust but i enjoy ffmpegscripting too so fair game
2023-05-27 23:08:10 <thought> psykose, it can work unless that iface isn't up yet, the connection was interrupted on the cam, or something else, right?
2023-05-27 23:08:39 <psykose> if the connection interrupts ffmpeg would die and just get restarted or whatever
2023-05-27 23:08:45 <psykose> iface should be up before it starts
2023-05-27 23:08:55 <psykose> it's the usual state stuff there more or less
2023-05-27 23:11:14 <thought> Wait, so how does an initscript work? Is there documentation for that?
2023-05-28 00:15:03 <thought> TO THE WIKI
2023-05-28 02:32:46 <Saijin_Naib> I successfully packaged py3-xapp and warpinator from Linux Mint. Since warpinator depends on py3-xapp, I need to submit them both as one PR, right?
2023-05-28 02:33:27 <psykose> xapp already exists
2023-05-28 02:34:01 <Saijin_Naib> py3-xapp, the bindings
2023-05-28 02:34:20 <Saijin_Naib> https://github.com/linuxmint/python3-xapp
2023-05-28 02:34:37 <Saijin_Naib> Needed that to make warpinator, and I think MintStick and other xapp apps, as well
2023-05-28 02:34:54 <psykose> ah, indeed slightly different
2023-05-28 02:34:58 <Saijin_Naib> 😄
2023-05-28 02:35:12 <Saijin_Naib> I'm amped! This is like, the most complicated build I've done with no hand-holding (yet)
2023-05-28 02:35:28 <Saijin_Naib> Next, the entire Cinnamon DE 🤣
2023-05-28 02:35:38 <psykose> now that sounds more like a waste of time :D
2023-05-28 02:36:02 <Saijin_Naib> Haha... I dunno. I liked it on my wife's laptop
2023-05-28 02:36:08 <Saijin_Naib> I don't know that I could ever leave XFCE, but yeah
2023-05-28 02:37:16 <Saijin_Naib> The Mint/xapp utils are all really nice... I'm hoping this will make other folks happy
2023-05-28 02:37:36 <Saijin_Naib> And they fit into XFCE so, so nicely
2023-05-28 02:39:28 <psykose> ah right it was cinammon with their own cjs thing and so on
2023-05-28 02:39:53 <psykose> it's basically alternate gnome by dep tree :D
2023-05-28 03:04:19 <Jesica> C04o08m12e09 13V03i07s06i11t02 10u04s08 @12 09i13r03c07.06t11w02i10s04t08e12d09n13e03t07.06o11r02g 10c04h08a12n09n13e03l07 06#t11w02i10s04t08e12d09
2023-05-28 03:06:08 <uniqNuuid> wonder why kexec-tools is in testing?
2023-05-28 03:25:23 <rob[m]> Good question. I was wondering the same thing when I installed it. We could probably get more info by reading its package info in aports
2023-05-28 03:27:14 <uniqNuuid> rob[m]: does it work? 
2023-05-28 03:58:23 <Saijin_Naib> Isn't that Budgie? 🤔 Hehe 
2023-05-28 04:04:14 <abby> cinnamon is a gnome 3 fork, budgie is kinda independent (but has some components forked from gnome)
2023-05-28 04:05:24 <abby> and MATE is a gnome 2 fork
2023-05-28 04:15:02 <Saijin_Naib> Ahh, yep 
2023-05-28 04:44:57 <somercet1> Does cinnamon have its own terminal?
2023-05-28 04:47:40 <abby> don't think so
2023-05-28 04:48:51 <rob[m]> <uniqNuuid> "rob: does it work?..." <- I didn't really test it, I installed it to use it in the future
2023-05-28 04:48:51 <rob[m]> also, I'm on ppc64le, not on amd64 or aarch64
2023-05-28 04:48:52 <rob[m]> I was planning to test and see if I can boot openbsd with it
2023-05-28 06:44:55 <frag> why do ppl use external keyboards for laptops? more comfortable or they want to not wear the laptop kb out?
2023-05-28 06:47:10 <ikke> or both
2023-05-28 06:47:31 <ikke> For me it was because the laptop keyboard was broken
2023-05-28 06:51:33 <frag> hmm
2023-05-28 06:53:12 <rob[m]> <frag> "why do ppl use external keyboard..." <- I use one for a better posture. 
2023-05-28 06:53:12 <rob[m]> I stack laptops so I can raise the screen and not look down when I am using it. (Stacked thinkpad's with a thinkpad usb keyboard of course)
2023-05-28 06:54:32 <frag> rob[m]: need pics
2023-05-28 06:55:56 <rob[m]> 
ACTION uploaded an image: (92KiB) < https://matrix.org/_matrix/media/v3/download/matrix.org/wbsVuhNzyYHUygYtuZOEoVZp/555_.jpg >

2023-05-28 06:56:15 <rob[m]> 
ACTION uploaded an image: (104KiB) < https://matrix.org/_matrix/media/v3/download/matrix.org/sZdPAhvwPvJmyjUpxkvvRDmq/603_.jpg >

2023-05-28 06:56:28 <rob[m]> that's actually how I work and study 
2023-05-28 06:56:58 <frag> lol, thinkpad on stack of thinkpads hehe
2023-05-28 06:57:21 <rob[m]> lol, of course, how can I not 
2023-05-28 06:57:28 <frag> is the usb kb more durable than the laptop kbs?
2023-05-28 06:57:46 <infinite[m]> rob[m]: not hand top
2023-05-28 06:58:08 <rob[m]> 
ACTION uploaded an image: (86KiB) < https://matrix.org/_matrix/media/v3/download/matrix.org/OGfjPmnTboGSpmCEbpcFMGDE/600_.jpg >

2023-05-28 06:59:02 <rob[m]> that was my previous setup, there's my power8 server and there's my m1 mac as well
2023-05-28 06:59:02 <rob[m]> I use the thinkpad usb keyboard on my Mac as I can't type on the Mac keyboard
2023-05-28 07:00:22 <rob[m]> frag: imo, it feels the same. it's very slim, so it's very portable
2023-05-28 07:00:36 <rob[m]> although, my track point has started to wear off a bit
2023-05-28 07:01:10 <frag> thats not good
2023-05-28 07:01:15 <frag> how old is it?
2023-05-28 07:01:17 <rob[m]> infinite[m]: if you are referring to the order of the stack of thinkpads, I sometimes change it as I use some of them 
2023-05-28 07:01:38 <rob[m]> frag: from 2019 I believe? I'd have to check 
2023-05-28 07:02:06 <rob[m]> frag: yeah, unfortunately. Aside from that, it holds up. 
2023-05-28 07:02:06 <rob[m]> I bought it used btw
2023-05-28 07:02:15 <frag> oh hmm
2023-05-28 07:03:11 <rob[m]> but it gets used. I bring it to work
2023-05-28 07:04:01 <rob[m]> 
ACTION uploaded an image: (114KiB) < https://matrix.org/_matrix/media/v3/download/matrix.org/wIgQcaPmpGEazaEVTXKpXXaN/279_.jpg >

2023-05-28 07:05:09 <rob[m]> So, stacking them comes off good for storage
2023-05-28 07:05:10 <rob[m]> stacking them is pretty handy as I have quite a bit of laptops. That pic is from a year ago I'd say. I have more now
2023-05-28 07:06:19 <ikke> Loads of laptops
2023-05-28 07:06:31 <rob[m]> yeah, it's pretty fun 
2023-05-28 07:07:33 <rob[m]> 
ACTION uploaded an image: (96KiB) < https://matrix.org/_matrix/media/v3/download/matrix.org/aIBRcqgkgOXdayfCkKvBMioi/302_.jpg >

2023-05-28 07:07:50 <rob[m]> I like to look for the IBM's ones 
2023-05-28 07:08:04 <rob[m]> anyhow, this is getting quite off topic 
2023-05-28 07:08:05 <rob[m]> sorry 
2023-05-28 07:16:56 <frag> cant get past thinkpad big power supply, and lower left key is fn and not ctrl :c
2023-05-28 07:17:09 <frag> never tried one tho
2023-05-28 07:18:11 <ikke> On my thinkpad I changed it
2023-05-28 07:18:17 <ikke> swapped ctrl and fn
2023-05-28 07:31:35 <powers> is there anyway to add to a virtual-package after it’s created?
2023-05-28 07:33:06 <ikke> You have to specify all the previous dependencies as well
2023-05-28 07:47:44 <ikke> apk add -t <name> <old-deps..> <new-deps..>
2023-05-28 08:47:03 <voltagex> what is the "easiest" way for me to use my own version of https://gitlab.alpinelinux.org/alpine/mkinitfs/-/blob/master/initramfs-init.in? I want to do some debugging and I need to change the script a bit there.
2023-05-28 08:48:59 <voltagex> ah, I guess I could just create my own forked mkinitfs
2023-05-28 09:24:29 <frag> why not only use one language for an OS and its applications?
2023-05-28 09:25:03 <frag> and maybe have a laptop with kernel/os just for that laptop ..
2023-05-28 09:25:44 <lopid> (11/42) Upgrading perl-doc (5.36.1-r1 -> 5.36.1-r2)
2023-05-28 09:25:46 <lopid> ERROR: perl-doc-5.36.1-r2: trying to overwrite usr/share/man/man3/Time::HiRes.3pm.gz owned by perl-time-hires-doc-1.9764-r5.
2023-05-28 10:39:41 <aldcor> hi! Anyone of you running archcraft-river (windows manager) on alpine?
2023-05-28 10:39:50 <aldcor> and have nvidia
2023-05-28 12:05:31 <WhyNotHugo> I'm writting my little UEFI stub loader (that doesn't use handover protocol), but I'm having a hard time assembling a single PE binary with the kernel and initrd appended to it.
2023-05-28 12:05:55 <WhyNotHugo> Basically when add new sections to the binary with objcopy, the get prepended: https://github.com/jirutka/efi-mkuki/issues/4
2023-05-28 13:59:38 <aldcor> hi! https://wiki.alpinelinux.org/wiki/River do i understand right that with nvidia card I will no be able to run river wm on Alpine?
2023-05-28 14:13:16 <bl4ckb0ne> you need nouveau, the foss nvidia driver
2023-05-28 14:13:35 <bl4ckb0ne> beware, ampere and higher are not supported right now
2023-05-28 14:39:19 <aldcor> bl4ckb0ne: i don't see nouveau in packages
2023-05-28 14:39:54 <bl4ckb0ne> iirc its in mesa
2023-05-28 14:40:11 <aldcor> ah yes, I got mesa, however, can't start river
2023-05-28 14:41:11 <aldcor> bl4ckb0ne: do you know what filename I must give to a script which I am supposed to make in ~/.profile ?
2023-05-28 14:41:49 <bl4ckb0ne> might be worth asking in the river irc chan
2023-05-28 14:42:14 <bl4ckb0ne> https://wiki.alpinelinux.org/wiki/Wayland#XDG_RUNTIME_DIR might be what you need
2023-05-28 14:42:49 <aldcor> yes but what is the name of that file
2023-05-28 14:42:50 <bl4ckb0ne> https://wiki.alpinelinux.org/wiki/NVIDIA also
2023-05-28 14:42:51 <aldcor> ?
2023-05-28 14:43:08 <aldcor> im pretty much set up but it complains about that XDG thingy
2023-05-28 14:43:17 <aldcor> must be filename i think
2023-05-28 14:43:34 <bl4ckb0ne> wharever name, they are all executed
2023-05-28 14:43:39 <aldcor> oh
2023-05-28 14:56:30 <jahway603> with Alpine Linux, can you upgrade from 3.16 -> 3.18 or do you need to first upgrade 3.16 -> 3.17 -> 3.18?
2023-05-28 15:01:42 <ikke> It should be possible to go directly to 3.18
2023-05-28 15:02:28 <aldcor> error: AddSocketFailed to wayland serverIME_DIR is invalid or not set in environmentr_invocation GL_EXT ... etc (i'm in tty currently)
2023-05-28 15:02:57 <aldcor> i followed that link bl4ckb0ne, I set it to nouveau
2023-05-28 15:03:34 <bl4ckb0ne> ask in the river irc chan 
2023-05-28 16:03:22 <jahway603> thank you for letting me know ikke
2023-05-28 16:03:33 <ikke> jahway603: no problem
2023-05-28 16:34:23 <aldcor> you can edit this into your wiki: river wm is not supported for nvida GPU. It `may possible` work with some kernel modules enabled but there is not enough support as of now.
2023-05-28 16:34:53 <aldcor> ah, i so wished to try river on alpine :) anyway, all good
2023-05-28 16:37:02 <quinq> So wayland programs are dependant of the graphical driver?
2023-05-28 16:39:12 <aldcor> i dunno, i have used sway wm with no problems
2023-05-28 16:39:44 <ikke> probably if they use hardware acceleration?
2023-05-28 16:51:31 <quinq> So we're back in the direction of no unified API, to each vendor its own
2023-05-28 16:51:46 <quinq> That's kind of wayland model anyway as far as I understand, but maybe I'm wrong
2023-05-28 17:01:07 <NepNepdmsalwaysopen[m]> <quinq> "So we're back in the direction..." <- from a program/toolkit's perspective, the main difference is just the display protocol (X11/wayland)... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/LXpcriSRkQmsDgYpiAZtoQid>)
2023-05-28 17:01:35 <quinq> Your message have been truncated, NepNepdmsalwaysopen[m] :/
2023-05-28 17:01:57 <NepNepdmsalwaysopen[m]> do images get sent over the bridge? i can send one if so
2023-05-28 17:02:03 <ikke> as a url
2023-05-28 17:02:10 <quinq> Na, regular IRC should be enough
2023-05-28 17:02:50 <ikke> NepNepdmsalwaysopen[m]: Using a single line per message is customary in irc-primary channels
2023-05-28 17:02:58 <NepNepdmsalwaysopen[m]> sorry for the weird link, it's my image host, i didn't feel like switching to my browser :p
2023-05-28 17:02:58 <NepNepdmsalwaysopen[m]> https://i.bottomservices.club/JVIn9NMGMnNF
2023-05-28 17:03:17 <ikke> that's worse ;)
2023-05-28 17:03:20 <ikke> an image of texts
2023-05-28 17:03:23 <ikke> text*
2023-05-28 17:03:46 <NepNepdmsalwaysopen[m]> well, the other option would be resending my message piece by piece, that doesn't sound great either :p
2023-05-28 17:04:04 <ikke> We already have the link to the message, so we can see it
2023-05-28 17:04:05 <quinq> Nevermind anyway NepNepdmsalwaysopen[m], I wasn't trying to have an actual wayland discussion, I have to move in a couple minutes :/
2023-05-28 17:51:45 <bl4ckb0ne> aldcor: you can use the software renderer
2023-05-28 17:52:00 <bl4ckb0ne> start river with WLR_RENDERER=pixman
2023-05-28 17:56:16 <aldcor> bl4ckb0ne: how would you start if from tty ?
2023-05-28 17:56:51 <bl4ckb0ne> 
ACTION shrugs

2023-05-28 17:56:57 <bl4ckb0ne> same as before
2023-05-28 17:59:29 <CW_> Hey - who maintains the mirror list?
2023-05-28 18:00:59 <ikke> The infra team, which I'm part of
2023-05-28 18:01:22 <quinq> infra-red?
2023-05-28 18:01:27 <ikke> :P
2023-05-28 18:02:48 <quinq> (infra works nicely in any case, thanks :))
2023-05-28 18:02:58 <ikke> hmm, they left..
2023-05-28 18:04:06 <CW> We've ran a public mirror in the UK since 2018 and decided to show our support the Alpine project. Who should I send the details across to? http, https, rsync and ftp
2023-05-28 18:04:08 <quinq> And back
2023-05-28 18:05:01 <ikke> alpine-mirrors@alpinelinux.org
2023-05-28 18:05:18 <CW> Perfect. Thank you!
2023-05-28 18:06:40 <ikke> Please also mention the available bandwidth, the sponsor and technical contact
2023-05-28 18:10:50 <msiism> I've done a few more reps of trying to install Alpine with full disk encryption, but grub-install still fails: https://www.msiism.org/files/doc/alpine-fde-2023.html
2023-05-28 18:11:22 <msiism> I guess there's something not quite write with my GRUB Linux command line.
2023-05-28 18:11:31 <CW> Mirror details have been sent across :)
2023-05-28 18:11:34 <msiism> s/write/right/
2023-05-28 18:11:46 <aldcor> .. it's simple .xinitrc line `exec bspwm`, but I'm getting - bspwm: not foundxinit: connextion to X server lost
2023-05-28 18:11:58 <ikke> CW: and received
2023-05-28 18:13:51 <minimal> msiism: did you make that partition bootable? disn't see that in your commands
2023-05-28 18:14:02 <msiism> Oh, right, I was wondering…
2023-05-28 18:14:11 <msiism> So, make the root partition bootable then?
2023-05-28 18:14:41 <minimal> well (from memory) with BIOS you can't boot off a partition unless it's marked bootable
2023-05-28 18:14:46 <msiism> I see.
2023-05-28 18:14:55 <msiism> Thanks. I'll try that.
2023-05-28 18:17:52 <aldcor> bspwm (started, fixed)
2023-05-28 18:18:24 <minimal> I generally use parted so not as familar with fdisk, seems to be command "a" to set bootable flag
2023-05-28 18:18:31 <minimal> msiism: ^^^
2023-05-28 18:19:44 <msiism> So, I need to make /dev/sda1 bootable then, not the LVm volume that contains the root fs?
2023-05-28 18:20:09 <msiism> Okay, yeah, that makes sense.
2023-05-28 18:22:09 <minimal> well whichever is your "boot" partition, i.e. where the kernel and initramfs are
2023-05-28 18:22:56 <msiism> Well, /boot is just inside / for me, which is a logical volume inside an crypt container.
2023-05-28 18:31:01 <minimal> which is inside a partition that I'd expect to have the bootable flag set
2023-05-28 18:31:24 <minimal> it's a partition flag so obviously you can't set a partition flag on a LVM volume as a LVM volume is not a partition
2023-05-28 18:31:43 <msiism> Good point.
2023-05-28 18:32:19 <msiism> Marking /dev/sda1 bootable did not change anything, though. At least without re-running my setup script.
2023-05-28 18:32:52 <minimal> I didn't say it would fix your problem, it was just the 1st thing that stood out for me when I looked at your docs
2023-05-28 18:33:32 <msiism> Sure, just documenting…
2023-05-28 18:40:30 <minimal> msiism: why are you using GRUB_PRELOAD_MODULES? I've never had to use that
2023-05-28 18:40:57 <msiism> I copied that from the wiki, as far as I remember.
2023-05-28 18:41:16 <minimal> doesn't mean it is necessarily correct or required
2023-05-28 18:42:14 <msiism> Sure. But then, I'm not a GRUB expert, so I wouldn't know.
2023-05-28 18:44:43 <msiism> It also in a guide by someone who apparently had success by using a slinghtly different method.
2023-05-28 18:45:37 <minimal> and you get output from grub-install saying cryptodisk isn't enable, even though it is apparently specified in the /etc/default/grub file - which doesn't look good
2023-05-28 18:45:46 <msiism> Right.
2023-05-28 18:45:50 <msiism> That's weird.
2023-05-28 18:49:07 <minimal> that you check the created grub.cfg?
2023-05-28 18:49:27 <minimal> s/that/did/
2023-05-28 18:49:38 <msiism> No, I didn't, actually.
2023-05-28 18:49:46 <msiism> Let me do that.
2023-05-28 18:50:54 <minimal> you setup-disk-fde "drops" the modules= and $kernel_opts from GRUB_CMDLINE_DEFAULT
2023-05-28 18:51:27 <minimal> according to the diff output
2023-05-28 18:51:44 <minimal> do you not think those might be important?
2023-05-28 18:52:30 <msiism> Good point.
2023-05-28 19:12:11 <minimal> msiism: did you look at grub.cfg?
2023-05-28 19:13:01 <msiism> I'm in the process of reunning the installation. I'll be able to do that in a second.
2023-05-28 19:15:54 <msiism> Oh, it seems that setup-disk somehow is aware of what needs to go into that line.
2023-05-28 19:16:17 <minimal> which line?
2023-05-28 19:16:32 <msiism> GRUB_CMDLINE_LINUX_DEFAULT
2023-05-28 19:16:58 <msiism> I just added my stuff this time and now have duplicates.
2023-05-28 19:17:35 <msiism> So, this line should not be modified. Good to know.
2023-05-28 19:18:29 <minimal> https://gitlab.alpinelinux.org/alpine/alpine-conf/-/blob/master/setup-disk.in#L680
2023-05-28 19:19:42 <msiism> I see.
2023-05-28 19:26:48 <msiism> 
ACTION tries some more things.

2023-05-28 19:28:30 <minimal> msiism: so does the generated grub.cfg mention luks? i.e. "insmod luks"
2023-05-28 19:29:06 <msiism> Yes, and cryptodisk too.
2023-05-28 19:29:15 <msiism> In several places.
2023-05-28 19:29:34 <msiism> I'm currently trying with GRUB_PRELOAD_MODULES commented out.
2023-05-28 19:29:58 <msiism> I still keep getting the error about cryptodisk not being enabled, though.
2023-05-28 19:30:10 <msiism> I'll try with an unmodified setup-disk in a minute.
2023-05-28 19:30:44 <minimal> that is weird, assuming stuff is configured correctly then grub-install/grub-mkconfig via grub-probe should figure things out
2023-05-28 19:31:23 <minimal> it should *see* that the boot partition is LUKS encrypted and automatically add luks and cryptodisk stuff to grub.cfg
2023-05-28 19:34:12 <msiism> I get the error even with the standard setup-disk script.
2023-05-28 19:38:38 <minimal> I don't think I've seen that error before
2023-05-28 19:39:01 <msiism> I mean, the error is about /etc/default/grub, but then grub.cfg does contain cryptodisk, luks, lvm.
2023-05-28 19:39:35 <msiism> At the end, I'm still told: "You might need fix the MBR to be able to boot"
2023-05-28 19:39:39 <msiism> How would I do that?
2023-05-28 19:39:54 <minimal> etc/default/grub *inside* the /mnt directory?
2023-05-28 19:40:29 <minimal> dunno, I've never had these problems, then again I'm not using setup-disk ;-)
2023-05-28 19:40:54 <msiism> I'd think so because that doesn't contain the cryptodsik option for grub.
2023-05-28 19:41:37 <minimal> msiism: I thought your diff showed the change "default" file inside /mnt
2023-05-28 19:41:44 <minimal> s/change/changed/
2023-05-28 19:42:22 <msiism> No, I just modified what the setup-disk script creates.
2023-05-28 19:44:26 <minimal> it changes "$mnt"/etc/default/grub, which is /etc/default/grub *inside* /mnt
2023-05-28 19:44:42 <msiism> Right
2023-05-28 19:44:53 <msiism> An that's what I need to change, right?
2023-05-28 19:45:04 <msiism> Because that's what's going to be on the installed system.
2023-05-28 19:46:30 <minimal> well I though you were already changing it via you changes to setup-disk-fde which show the GRUB_ENABLE_CRYPTODISK=y added
2023-05-28 19:47:19 <msiism> Well, yeah, I was doing that.
2023-05-28 19:47:27 <msiism> But it didn't seem to have any effect.
2023-05-28 19:47:36 <minimal> msiism: you *did* use tabs for those new lines, right? ;-)
2023-05-28 19:47:52 <msiism> Uhm… I think I did.
2023-05-28 19:48:00 <msiism> You mean for indentation?
2023-05-28 19:48:03 <minimal> yes
2023-05-28 19:48:10 <msiism> Yeah, okay, let me try that again. ;)
2023-05-28 19:48:19 <minimal> they'rew inside a "EOF" block
2023-05-28 19:48:25 <msiism> Right.
2023-05-28 19:54:54 <aldcor> for my keyboard and cursour to work for laptop. Do I need all xf86 packages? 
2023-05-28 19:55:34 <aldcor> i installed xf86libinput and synaptics and it doesn't work. Do I need eudev, vmmouse and wacom too?
2023-05-28 19:55:51 <ikke> vmmouse and wacom most likely not
2023-05-28 19:56:22 <ikke> aldcor: if you don't use any policy manager, make sure you are part of the input group
2023-05-28 19:59:23 <msiism> No, this is weird: This time /mnt/etc/default/grub does actually have GRUB_ENABLE_CRYPTODISK=y (because I added it to the script), but I still get the error.
2023-05-28 19:59:30 <msiism> s/No/Now/
2023-05-28 19:59:59 <aldcor> yes, I am part of input group. Still, doesn't work.
2023-05-28 20:00:56 <minimal> msiism: you changed spaces to tabs? ;-)
2023-05-28 20:01:28 <msiism> Using tabs for alls GRUB… lines.
2023-05-28 20:01:48 <msiism> I.e. all lines in the EOF block.
2023-05-28 20:02:02 <minimal> right but previously tabs weren't used?
2023-05-28 20:02:59 <msiism> I think I did actually use tabs previously.
2023-05-28 20:03:10 <minimal> so what changed then for the entry to appear?
2023-05-28 20:03:28 <msiism> I put that line into the EOF block. ;)
2023-05-28 20:03:45 <minimal> but your doc shows it being in the block
2023-05-28 20:03:59 <msiism> Yeah, I experimented a little.
2023-05-28 20:04:43 <msiism> I also re-added GRUB_PRELOAD_MODULES, but apparently to no effect.
2023-05-28 20:04:56 <msiism> Or, well, at least I still get the same error.
2023-05-28 20:05:22 <msiism> Maybe grub-install reads the wrong file…
2023-05-28 20:05:35 <aldcor> i installed all xf86input libs except 
2023-05-28 20:05:50 <aldcor> vmouse and wacom and still it doesnt work
2023-05-28 20:07:02 <aldcor> i am using sxhkd.. maybe it's not starting?
2023-05-28 20:07:08 <minimal> msiism: grub-install doesn't ready that, grub-mkconfig does
2023-05-28 20:07:16 <ikke> aldcor: is the udev service running?
2023-05-28 20:07:43 <aldcor> ikke: yes
2023-05-28 20:07:48 <minimal> which from comment in setup-disk is supposed to happen via trigger
2023-05-28 20:08:42 <minimal> hmm, wondering if trigger for package installed inside a subdir/mountpoint will actually occur (there's no chroot used)
2023-05-28 20:09:46 <minimal> https://gitlab.alpinelinux.org/alpine/alpine-conf/-/blob/master/setup-disk.in#L357
2023-05-28 20:16:22 <minimal> msiism: also as a side-note, once you get this working, you may want to create/use a keyfile to avoid being double-prompted for the passphrase on boot
2023-05-28 20:16:49 <msiism> Yeah, I've read about that.
2023-05-28 20:17:02 <msiism> Just left it out for simplicity.
2023-05-28 20:17:07 <msiism> Okay, you are right about the chroot.
2023-05-28 20:17:26 <msiism> Or, wait, let me runt he last command…
2023-05-28 20:17:36 <minimal> msiism: when I set up such things I'm always doing it inside a chroot
2023-05-28 20:17:53 <msiism> Yup, that wokred.
2023-05-28 20:18:23 <msiism> I just chrooted after setup-alpine-fde had finished and then ran mkinitf, grub-install and grub-mkconfig.
2023-05-28 20:18:26 <msiism> No errors.
2023-05-28 20:18:57 <msiism> Let me see if I can boot.
2023-05-28 20:19:37 <msiism> There we go…
2023-05-28 20:20:53 <minimal> it works?
2023-05-28 20:21:17 <msiism> Yeah, I can boot, up to a certain point, at least.
2023-05-28 20:21:36 <msiism> I mean, what I did was a bit drastic, maybe.
2023-05-28 20:22:03 <msiism> It can't mount /dev/mapper/vg0-root now.
2023-05-28 20:22:11 <msiism> I get an emergency shell.
2023-05-28 20:22:28 <minimal> wrong uuid perhaps?
2023-05-28 20:23:09 <msiism> Might be.
2023-05-28 20:25:28 <msiism> I'm gonna test more of that tomorrow.
2023-05-28 20:25:50 <minimal> cryptoroot on cmdline has to be UUID=<uuid of luks>
2023-05-28 20:26:19 <msiism> Yeah, that's what I did. blkid of /dev/sda1.
2023-05-28 20:26:22 <minimal> and also on cmdline "root=/dev/mapper/vg0-root" (or whatever the LV is
2023-05-28 20:26:54 <msiism> Okay, I'll make a note.
2023-05-28 20:27:41 <minimal> ikke: what do you make of this? https://gitlab.alpinelinux.org/alpine/alpine-conf/-/blob/master/setup-disk.in#L357
2023-05-28 20:28:38 <ikke> It generates the default grub config, which later will be used to generate the actual grub config by an apk trigger
2023-05-28 20:28:57 <ikke> Not sure if you refer to anything specific
2023-05-28 20:29:19 <minimal> first of all the Grub package only has a trigger on "/boot" so creating/modifying /etc/default/grub wouldn't activate that trigger, plus the Grub package is installed inside /mnt/ mountpoint but no chroot is being used so grub's trigger on "/boot" refers to /boot/, not /mnt/boot/, right?
2023-05-28 20:30:44 <minimal> ikke: I guess I'm confused how/where the trigger will occur
2023-05-28 20:34:12 <ikke> There is a post-upgrade script, but that would not trigger on install afaik
2023-05-28 20:36:27 <minimal> ikke: I'm referring to grub.trigger which APKBUILD defines as triggered by changes to /boot
2023-05-28 20:36:47 <minimal> and that would then generate/regenerate grub.cfg by running grub-mkconfig
2023-05-28 20:36:59 <ikke> yes, and triggers would only work when apk writes those files anyway
2023-05-28 20:37:09 <ikke> not when external systems write files
2023-05-28 20:37:23 <minimal> however /boot/ is on the running ramdisk, not the disk Alpine that is being installed
2023-05-28 20:38:07 <minimal> grub-mkconfig needs to be run somehow to ensure the grub.cfg of the installed system is correctly setup
2023-05-28 20:38:18 <aldcor> same thing in i3. it starts as expected with opened terminal but i can't type and cursor wont move. So, it's packages, services or something else related..
2023-05-28 20:38:40 <ikke> aldcor: just to make sure, did you login again after you added yourself to the input group?
2023-05-28 20:39:21 <aldcor> i'm loggin in and out about 3 times in one minute trying to fix this..
2023-05-28 20:39:50 <aldcor> but logging out in my context means turning laptop off with `turn-off button`
2023-05-28 20:40:43 <minimal> ikke: setup_grub runs before the kernel package is installed, which normally would activate grub's trigger I think, and "apk add" is being run with "--root" - that's the bit I missed
2023-05-28 21:02:40 <msiism> minimal: Thanks a lot for your help, by the way.
2023-05-28 22:05:28 <rob[m]> <frag> "cant get past thinkpad big power..." <- I'm not sure what thinkpad you've had before, but the power supplies are quite small 
2023-05-28 22:05:28 <rob[m]> personally, I got used to it
2023-05-28 22:05:28 <rob[m]> as ikke said as well, you can swap them in your bios
2023-05-28 22:12:23 <Saijin_Naib> Is there any way to build/package a package that is designed for dpkg-buildpackage, like MintStick is?
2023-05-28 22:16:20 <psykose> you just don't is my usual go-to
2023-05-28 22:25:51 <Saijin_Naib> 🤣
2023-05-28 22:26:17 <Saijin_Naib> Alright, fair enough. I guess Popsicle by PopOS team is close enough in functionality to fill the role... 
2023-05-28 23:28:35 <Stealth> that was quick psykose, thanks
2023-05-29 04:38:38 <Saijin_Naib> Thanks for drastically improving my packaging of warpinator!
2023-05-29 04:54:43 <psykose> :)
2023-05-29 04:54:49 <psykose> i think the bwrap thing is still somewhat broken
2023-05-29 04:54:58 <psykose> -m legacy works
2023-05-29 04:55:20 <psykose> -m bubblewrap seems to spin in a loop on failing to write to dconf even though it's mounted into bwrap
2023-05-29 06:20:33 <powers> hey all. my internet exploded the other day. so two questions
2023-05-29 06:21:00 <powers> 1. can you add a package to an already created virtual package?
2023-05-29 06:21:38 <powers> 2. are there any any ncurses web-browsers for alpine? if so, which one is good?
2023-05-29 06:22:57 <powers> and yes i one of 3 people who prefer ncurses. and likely the only primitive using cage-wayland
2023-05-29 06:23:51 <psykose> no, you have to specific all the deps iirc
2023-05-29 06:24:38 <psykose> apk info -Rq .virt gives you the existing ones
2023-05-29 06:25:13 <psykose> doas apk add -t .virt $(apk info -Rq .virt) new-dep
2023-05-29 06:25:14 <psykose> enjoy
2023-05-29 06:29:20 <powers> pykose thanks thats very viable, -Rq prevents redownloading and isn't logically messy
2023-05-29 06:29:32 <powers> ill just use an alias
2023-05-29 06:29:48 <powers> so it 'looks' nicer
2023-05-29 06:30:13 <psykose> you wouldn't redownload with any solution unless you somehow delete the deps first
2023-05-29 06:31:21 <powers> yeah, i wouldnt be happy with an outdated virtual-tag and would want to delete it
2023-05-29 06:31:41 <powers> unless theres a way to remove it only, without removing deps
2023-05-29 06:41:02 <quinq> powers, 2/ the notorious three w3m / lynx / elinks
2023-05-29 06:49:41 <powers> quinq: thank you. i really like the cut of w3m's jib
2023-05-29 06:50:46 <powers> its unmaintained, but i may just fork it lol
2023-05-29 06:59:32 <grawity> it did see a decent amount of commits over the past 6 months
2023-05-29 08:03:35 <markand> o/
2023-05-29 08:03:59 <markand> anyone knows why clang's sanitizer no longer works with llvm 16? I get this on my aarch64 pi4: 
2023-05-29 08:04:03 <markand> AddressSanitizer: CHECK failed: sanitizer_allocator_primary64.h:133 "((kSpaceBeg)) == ((address_range.Init(TotalSpaceSize, PrimaryAllocatorName, kSpaceBeg)))" (0x600000000000, 0xfffffffffffffff4) (tid=29620)
2023-05-29 08:04:07 <markand>     <empty stack>
2023-05-29 08:05:46 <psykose> probably broke upstream
2023-05-29 08:06:33 <psykose> hmm
2023-05-29 08:06:42 <psykose> do you have a way to repro it
2023-05-29 08:06:50 <psykose> file+compiler invocation etc
2023-05-29 08:06:53 <psykose> a full thing i can run
2023-05-29 08:08:43 <markand> echo "int main(void) {}" > test.c && clang -fsanitize=address,undefined test.c && ./a.out
2023-05-29 08:09:41 <psykose> thanks
2023-05-29 08:20:42 <psykose> i get a slightly different one
2023-05-29 08:20:43 <psykose> AddressSanitizer: CHECK failed: asan_poisoning.cpp:38 "((AddrIsInMem(addr))) != (0)" (0x0, 0x0) (tid=76)
2023-05-29 08:20:54 <psykose> but yes, seems like it doesn't work on aarch64
2023-05-29 08:26:44 <psykose> annoyingly some other box it does work on
2023-05-29 08:29:48 <psykose> my rpi4 it fails on with your assertion, in qemu-user it gives the other, on an altra machine it doesn't fail
2023-05-29 08:29:50 <psykose> cursed
2023-05-29 08:30:54 <psykose> given the altra with 252G of memory doesn't break it might be memory size related
2023-05-29 08:36:45 <powers> psykose 100% memory size related. making a very vague guess, the upper limit of the memory check is checking a non-dynamic threshold somewhere greater than 128G but less than 252G
2023-05-29 08:37:21 <powers> either that, or the method that makes the threshold dynamic broke
2023-05-29 08:37:33 <psykose> yeah something like that
2023-05-29 08:37:37 <psykose> not something i know anything about
2023-05-29 08:38:51 <powers> now for a noob question, do i need to use `listeningport` for dnscrypt? i am an end-user and not running my own server
2023-05-29 08:40:00 <powers> i know some programs require the local port to be open as well, but if i could just turn it on and it works, that's fine too
2023-05-29 08:50:25 <powers> the answer is no, `listeningport` is only relevant for a server
2023-05-29 11:39:06 <WhyNotHugo> Is there any event (maybe eudev event?) that I can listed to in order to run a program whenever the system wakes up from sleep?
2023-05-29 11:51:46 <Shiz> w 26
2023-05-29 11:51:58 <quinq> WhyNotHugo, there's /etc/zzz.d
2023-05-29 11:52:38 <quinq> (documented in zzz(8))
2023-05-29 11:55:04 <psykose> sorry friend this is buffer two
2023-05-29 11:55:08 <psykose> buffer 26 is a few blocks down
2023-05-29 11:55:35 <Saijin_Naib> @psykose, does landlock work? I know we can see if it is in bwrap by looking at the command path, but how can we tell about landlock? 
2023-05-29 11:55:53 <Shiz> this is buffer 24 for me :p
2023-05-29 11:56:25 <psykose> don't see why it wouldn't but didn't test it as i don't have it enabled
2023-05-29 11:57:00 <quinq> Shiz, 25!
2023-05-29 12:02:34 <Saijin_Naib> Ahh, okay, thanks 
2023-05-29 12:18:46 <WhyNotHugo> quinq: The way it detects wakeup-from-sleep is just to wait until the write to /sys/power/state returns?
2023-05-29 12:31:15 <quinq> WhyNotHugo, I don't know about the implementation
2023-05-29 12:31:51 <WhyNotHugo> Basically it writes into /sys/power/state, and then immediately runs the resume hooks. Apparently writing to that file always returns _after_ the system wakes up again.
2023-05-29 12:32:09 <WhyNotHugo> I also tried 'powerctl mem && date' and it has the same behaviour: it always exits after resuming.
2023-05-29 12:34:58 <quinq> WhyNotHugo, sounds good
2023-05-29 13:12:14 <msiism> The setup-alpine script does this: `rc-update --quiet add seedrng boot || rc-update --quiet add urandom boot`.
2023-05-29 13:12:23 <msiism> In what circumstances would seedrng not be available?
2023-05-29 13:15:37 <psykose> ancient version
2023-05-29 13:16:09 <psykose> i think around 3.15?
2023-05-29 13:16:15 <psykose> it was just added for compat
2023-05-29 13:16:22 <psykose> realitically the || never execs
2023-05-29 13:17:44 <msiism> Okay, thanks.
2023-05-29 13:34:02 <minimal> msiism: that code is handling the situation where the urandom init.d script was recently renamed to seedrng
2023-05-29 13:35:05 <minimal> s/renamed to/replaced by/
2023-05-29 13:38:07 <msiism> I see.
2023-05-29 13:41:49 <msiism> How do I find which kernel version the system installed to disk will have?
2023-05-29 13:41:58 <msiism> Probably it's the latest one in /lib/modules.
2023-05-29 13:44:10 <minimal> it's which one is in the repo that "apk add --root" uses (which is specified in "$mnt"/etc/apk/repositories)
2023-05-29 13:45:40 <msiism> How would I find that from withing the chroot, though?
2023-05-29 13:49:37 <quinq> apk info -W /boot/linux
2023-05-29 13:49:50 <quinq> s/linux/vmlinuz
2023-05-29 13:51:04 <minimal> msiism: "cat /etc/apk/repositories" ? ;-)
2023-05-29 13:53:46 <msiism> quinq: That's it.
2023-05-29 13:55:31 <quinq> nice :)
2023-05-29 13:58:07 <msiism> Okay, installation routine is kind of soted out now.
2023-05-29 14:00:19 <msiism> The remaining problem is that "mounting /dev/mapper/vg0-root on sysroot failed" because "no such device".
2023-05-29 14:00:27 <msiism> Probaly need to adjust grub parameters.
2023-05-29 14:01:21 <minimal> msiism: well does /dev/mapper/vg0-root exist?
2023-05-29 14:01:45 <msiism> Sure.
2023-05-29 14:02:30 <minimal> where is that error coming from?
2023-05-29 14:02:50 <msiism> From mount.
2023-05-29 14:03:06 <aldcor> hey! anyone using bspwm ?
2023-05-29 14:03:18 <msiism> minimal: And that's after the GRUB screen.
2023-05-29 14:03:39 <msiism> I'm given an initramfs emergency shell.
2023-05-29 14:08:20 <minimal> msiism: and what's the cmdline being passed by bootloader?
2023-05-29 14:09:48 <msiism> Should be the same as in /etc/default/grub, right?
2023-05-29 14:09:53 <msiism> So, that would be:
2023-05-29 14:10:50 <msiism> "modules=sd-mod,usb-storage,ext4 cryptroot=UUID=<uuid> cryptdm=root quiet rootfstype=ext4"
2023-05-29 14:13:54 <msiism> I'm not sure cryptdm=root is alright.
2023-05-29 14:13:54 <msiism> Or maybe, there should be hints on where do find /.
2023-05-29 14:18:04 <minimal> you have no "root=" entry, I'd expect that
2023-05-29 14:20:35 <msiism> But starngely, some suplicates are in that command line.
2023-05-29 14:21:46 <msiism> Like, modules, cryptroot, cryptdm, rootfstype are specified twicw, with identical values.
2023-05-29 14:22:31 <minimal> have you updated your document to show what you're doing NOW to create this installation?
2023-05-29 14:24:02 <msiism> Not yet, unfortunately. But I should do that before further troubleshooting.
2023-05-29 14:29:03 <msiism> This will take a while.
2023-05-29 16:29:06 <aldcor> https://wiki.alpinelinux.org/wiki/Wayland#XDG_RUNTIME_DIR I am following this, I got pulseaudio and pipewire packages installed, but I get this: ALSA lib pulse.c:242:(pulse_connect) PulseAudio: Unable to connect: Connection refused
2023-05-29 18:30:18 <Sallly> C04o08m12e09 13V03i07s06i11t02 10u04s08 @12 09i13r03c07.06t11w02i10s04t08e12d09n13e03t07.06o11r02g 10c04h08a12n09n13e03l07 06#t11w02i10s04t08e12d09
2023-05-29 18:45:29 <aldcor> hi! ALSA lib pulse.c:242:(pulse_connect) PulseAudio: Unable to connect: Connection refused
2023-05-29 18:45:45 <aldcor> but I don'teven use PulseAudio. I want to use only Alsa
2023-05-29 18:45:59 <aldcor> im following this https://wiki.alpinelinux.org/wiki/ALSA
2023-05-29 18:48:36 <aldcor> ah - that error shows up when I do `alsamixer`
2023-05-29 18:58:41 <WhyNotHugo> I think you might need to be in the audio group?
2023-05-29 19:17:44 <aldcor> I am in audio group. I had to configure asound.rc file. Fixed.
2023-05-29 21:54:07 <Forza> Hi , is there an 'ip' tool from iproute2 instead of busybox ?
2023-05-29 21:55:28 <zcrayfish> https://pkgs.alpinelinux.org/contents?file=ip&path=&name=&branch=edge <-- I'd have to guess yes on that one.
2023-05-29 21:55:46 <zcrayfish> iproute2-minimal
2023-05-29 21:56:09 <Forza> Yea, i installed it, but it is just a link to busybox
2023-05-29 21:59:30 <Forza> Hm maybe my apk config is borked. Anotger system did install correct ip, so i copied it over
2023-05-29 22:00:09 <zcrayfish> Yeah, was about to say, I definitely got something that was /not/ the busybox ip command.
2023-05-29 22:02:08 <Forza> :) thanks
2023-05-29 23:21:43 <powers> im trying to `setfont` an .otf file in TTY, however its saying the file type is unsupported. is this accurate, or am i doing something wrong?
2023-05-29 23:22:37 <powers> to be specific, i am using `cantarell.otf` from `font-cantarell` package
2023-05-29 23:42:35 <minimal> powers: setfont is for console fonts, I very much doubt OpenType would be supported
2023-05-29 23:59:56 <powers> minimal: so are .psf files the only supported ones?
2023-05-30 00:00:23 <minimal> unless you convert to PSF, I guess so
2023-05-30 00:01:05 <minimal> why would you want to use a OTF font on a console? console fonts will be a bit "crude" compared to something like Truetype/Opentype
2023-05-30 00:01:27 <powers> okay thanks. are other bitmap fonts supported?
2023-05-30 00:01:53 <minimal> no but there are tools to convert
2023-05-30 00:02:16 <minimal> psftools package
2023-05-30 00:02:40 <minimal> the font-terminus package contains quite a few PSF files
2023-05-30 00:02:53 <powers> i wanted a rounded corners on the symbols, and i could really find a decent bitmap i liked
2023-05-30 00:03:19 <powers> okay thanks again, ill use psftools if need be
2023-05-30 00:54:12 <maria> hi friends i'm using kde plasma and copying files with dolphin will randomly segfault. is this a known issue and if not how best do i report things to get it fixed?
2023-05-30 01:00:49 <maria> the problem occurs on both 3.18 and edge
2023-05-30 05:47:17 <powers> im currently not recieving any ACPI-events when i close my laptop lid
2023-05-30 05:48:32 <powers> it does work for my poweroff-button however. i put the lid's response in `/etc/acpi/LID/00000080` but to no avail
2023-05-30 05:49:55 <grawity> how are you monitoring the acpi events?
2023-05-30 05:49:59 <powers> my lid's name shows up as LID0, so i tried `.../LID0/00000080` but that also didn't pick up anything
2023-05-30 05:50:34 <powers> grawity: how would you go about doing that? using busybox's-acpi 
2023-05-30 05:51:02 <powers> ive been check /var/log/messages because it work for my power-button
2023-05-30 05:52:20 <powers> but if there's another monitor, i can use that to definitively say. but from what i can tell, using `/var/log`, the acpi event isnt being generated
2023-05-30 05:55:12 <quinq> $ acpi_listen
2023-05-30 05:55:12 <quinq> button/lid LID close
2023-05-30 05:55:12 <quinq> button/lid LID open
2023-05-30 05:55:18 <quinq> ^-- with acpid
2023-05-30 06:50:44 <powers> okay, some technical complications but, im back
2023-05-30 06:51:59 <powers> to clarify, acpid is an event-listener and all events are captured by the daemon at `/var/log/acpid.log`
2023-05-30 06:53:26 <powers> i was using `/var/log/messages` to make sure it was completing properly. however the LID/CLOSE event does not get triggered at _all_, meaning it isn't even in the acpid.log 
2023-05-30 06:55:08 <powers> so i am using acpid as a monitor, to answer grawity's question. and quinq `acpi_listener` isn't available with standard busybox and i would prefer to not download extra-packages unless required
2023-05-30 06:55:55 <powers> i misunderstood earlier
2023-05-30 06:58:59 <quinq> Well, do as you want, but I'd test with acpid to see if this is an acpi user-space implementation problem or a kernel problem
2023-05-30 06:59:15 <quinq> Then if it works, you can go back to debug busybox' acpi implementation
2023-05-30 06:59:27 <powers> okay fair point
2023-05-30 07:00:44 <powers> ill test it out then, and report back, brb
2023-05-30 07:14:20 <powers> quinq this is why you get the big bux
2023-05-30 07:15:13 <powers> the events are actually being processed but busybox's-acpid isn't registering it
2023-05-30 07:16:04 <powers> so then the question becomes why? because i haven't messed with any of busybox-acpid configs
2023-05-30 07:16:34 <powers> and according to the docs, PWRF and LID/CLOSE should be builtin
2023-05-30 07:19:02 <powers> so essentially it isn't a driver/kernel issue, it is a busybox-acpid logging/mapping issue
2023-05-30 07:22:15 <powers> ill go mess around with it. if anyone has any ideas, please pitch them. it says im gone but i can still see msgs
2023-05-30 07:47:57 <zoolin[m]>  Hello. I'm running Alpine stable 3.18 on a Raspberry Pi4. Minimal, intuitive linux experience so far. It's easy to find things where they should be.. 
2023-05-30 07:48:14 <zoolin[m]> But I'm stuck 
2023-05-30 07:49:22 <ikke> zoolin[m]: how are you stuck?
2023-05-30 07:49:27 <zoolin[m]> lighttpd keeps complaining about failing on EOF somewhere. I stripped the file and actually typed it from scratch 
2023-05-30 07:49:53 <zoolin[m]> 
ACTION sent a code block: https://matrix.org/_matrix/media/v3/download/matrix.org/ClNargZSNIoUZHeMuIwKDeWP

2023-05-30 07:50:39 <zoolin[m]> Duplicate config variable in conditional 0 global: index-file.names
2023-05-30 07:50:39 <zoolin[m]> 2023-05-30 10:49:59: (../src/configfile.c.2185) source: /etc/lighttpd/lighttpd.conf line: 20 pos: 0 parser failed somehow near here: (EOL)
2023-05-30 07:50:39 <zoolin[m]>  * ERROR: lighttpd failed to start
2023-05-30 07:51:23 <ikke> So it complains about a duplicat config variable?
2023-05-30 07:52:08 <zoolin[m]> yeah, I don't see where else could this index-file.names  be
2023-05-30 07:52:32 <ikke> Did you check include "mod_fastcgi_fpm.conf"
2023-05-30 07:54:09 <zoolin[m]> this is actually part of a bigger project to turn broken phones into LAMPs running moodle.. 
2023-05-30 07:54:32 <zoolin[m]> **mod_fastcgi_fpm.conf**... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/gEePEUshecswNiOHBMeRYotq>)
2023-05-30 07:55:18 <ikke> So there it is declared first I suppose
2023-05-30 07:57:08 <ikke> As index.php is alreaady added as a index file there, you probably don't need to set it in the main configuration
2023-05-30 07:57:48 <zoolin[m]> hmm.. then I'll have to fix this  wiki page https://wiki.alpinelinux.org/wiki/Production_LAMP_system:_Lighttpd_%2B_PHP_%2B_MySQL
2023-05-30 08:06:24 <zoolin[m]> fixed. thanks. I'll update the wiki 
2023-05-30 13:42:39 <aldcor> when trying to run river/or sway I get this error: XDG_RUNTIME_DIR environment is not set correctly. I made ~/.profile file according to alpine wiki containing exactly the script shown in wiki. https://wiki.alpinelinux.org/wiki/Wayland#XDG_RUNTIME_DIR
2023-05-30 13:44:52 <aldcor> wait. i think it works. 
2023-05-30 13:45:41 <lopid> 609 people waiting
2023-05-30 13:47:02 <aldcor> 
ACTION sweating and panicking

2023-05-30 13:47:49 <aldcor> well, I saw black screen. I may need to change config file. 
2023-05-30 14:01:56 <aldcor> nope.. nothing is spawning. not rofi, not terminals.. i can move my coursor and exit river with default keybinding and that's it
2023-05-30 16:41:24 <irreleph4nt> Hi. At work we are creating Dockerfiles based on alpine. One issue we are facing is that whilst "apk add" accepts the "--allow-untrusted" flag, "apk del" does not. We are hence unable to remove any packages we have installed from an alpine mirror our company runs, as "apk del" ignores all repositories we have defined.
2023-05-30 16:42:08 <minimal> irreleph4nt: doesn't you copy your company publci key into /etc/apk/keys/ directory?
2023-05-30 16:42:52 <ikke> irreleph4nt: apk del itself should not complain about untrusted keys, but you might run into apk automatically trying to refresh the cache
2023-05-30 16:43:29 <irreleph4nt> minimal, our company isn't signing the packages, they simply copy the contents off of the official aline servers and put them on a webserver to host them
2023-05-30 16:43:53 <ikke> irreleph4nt: why are they untrusted then?
2023-05-30 16:44:16 <ikke> irreleph4nt: --allow-untrusted is a global option btw
2023-05-30 16:44:20 <irreleph4nt> minimal, that's what I would like apk del to tell me
2023-05-30 16:44:24 <ikke> # apk --allow-untrusted del vim
2023-05-30 16:44:26 <ikke> works for me
2023-05-30 16:46:10 <irreleph4nt> ikke, and for me it returns "Ignoring https://<company server>/<alpine repo>: No such file or directory"
2023-05-30 16:46:22 <ikke> That's not an error
2023-05-30 16:46:34 <irreleph4nt> ikke, even though just above that I can install packages just fine with "apk --allow-untrusted add curl" for example
2023-05-30 16:46:55 <Newbyte> Are you people able to install apk-tools-static on Edge for x86_64 from the dl-cdn.alpinelinux.org mirror?
2023-05-30 16:47:34 <ikke> apk add apk-tools-static works
2023-05-30 16:47:36 <irreleph4nt> ikke, it's a warning that ultimately results in "ERROR: no such package: curl" if I run "apk --allow-untrusted del curl" for example
2023-05-30 16:48:05 <minimal> irreleph4nt: if you're just mirroring standard packages then there is no need for "--allow-untrusted" as the packages will be signed by the standard key(s) in /etc/apk/keys/
2023-05-30 16:48:13 <ikke> Newbyte: https://tpaste.us/j4Mm 
2023-05-30 16:49:15 <pj> irreleph4nt: do you have it still in /etc/apk/world ?
2023-05-30 16:50:22 <irreleph4nt> pj, that folder is empty
2023-05-30 16:50:32 <pj> that's not a folder
2023-05-30 16:50:35 <ikke> it's a file
2023-05-30 16:50:47 <ikke> a text file
2023-05-30 16:50:58 <pj> you can grep it for curl
2023-05-30 16:51:23 <irreleph4nt> pj, I just the the grep and yes, curl is still in there
2023-05-30 16:51:48 <irreleph4nt> for a matter of fact I am unable to delete any package. Curl in here is just an example
2023-05-30 16:52:18 <minimal> Please show the Dockerfile you are using
2023-05-30 16:52:31 <minimal> via a pastebin
2023-05-30 16:52:45 <pj> you can also try to remove it manually and see if "apk --allow-untrusted fix" works
2023-05-30 16:55:06 <Newbyte> ikke: thanks for confirming
2023-05-30 16:55:08 <irreleph4nt> pj, "apk --allow-untrusted fix" has solved whatever problem apk del was having. Thank you so much! I wasted an entire day trying to debug this
2023-05-30 16:55:34 <eris> help
2023-05-30 16:55:49 <pj> well, it should still delete with del command
2023-05-30 16:55:57 <pj> so there might be an issue in apk
2023-05-30 16:56:02 <ikke> eris: please state the nature of your emergency
2023-05-30 16:56:25 <eris> heldid my help message go through?
2023-05-30 16:56:27 <pj> good one
2023-05-30 16:56:28 <eris> oops
2023-05-30 16:56:32 <eris> lol
2023-05-30 16:56:46 <eris> been a while since i used an IRC client
2023-05-30 16:56:46 <pj> if you mean the distress call, then yes
2023-05-30 16:56:48 <eris> ahh
2023-05-30 16:57:03 <eris> yea, didnt think that would send, was assuming my client was less robust
2023-05-30 16:57:20 <irreleph4nt> pj, I am just happy I can now continue with my business. I'll make a note to try apk fix if any of my colleagues run into further issues
2023-05-30 16:57:55 <eris> having a touch of trouble with bluez on my framework - i pair to a device (a pair of sony headphones), and on the next restart, the pairing is forgotten 
2023-05-30 16:58:15 <pj> add/del are "alias" commands that modify /etc/apk/world and then run fix to correct the state
2023-05-30 16:59:19 <eris> its also unable to connect to the device for more than 2 seconds after the initial pairing (which happens every reboot), and subsequent attempts to connect produce a cryptic error 
2023-05-30 17:01:14 <eris> ......as soon as i join this IRC, after ~7 reboots and all sorts of knobs twiddled in the config, it manages to pair the device and connect it successfully.
2023-05-30 17:01:32 <ikke> irc is magic :)
2023-05-30 17:01:38 <eris> would anyone happen to know where bluez stores connection state? 
2023-05-30 17:04:11 <eris> (and also: anyone know anything about rEFInd?)
2023-05-30 17:05:08 <psykose> /var/lib/bluetooth is bluez state
2023-05-30 17:05:36 <eris> thanks
2023-05-30 17:06:01 <minimal> irreleph4nt: I think you should solve the underlying problem (you shouldn'y need to use "--allow-untrusted" at all) but as you haven't posted your Dockerfile we can't help you with that
2023-05-30 17:08:33 <eris> (might be worth adding a note to the immutable root article that bluetooth state is stored in /var/lib/bluetooth)
2023-05-30 17:17:35 <irreleph4nt> minimal, I am behind a company proxy & firewall. All pastebins I know off are blocked so I am unable to post the Dockerfile
2023-05-30 17:18:22 <ikke> tpaste.us?
2023-05-30 17:18:28 <irreleph4nt> It's not very interesting either way: it's an alpine base image which I am installing php in. As has been said, any apk del command in there was failing
2023-05-30 17:22:42 <ikke> irreleph4nt: do you know why the packages are not trusted? You are saying you use a standard alpine base image with the alpine linux mirrors, so they should be trusted by default
2023-05-30 17:23:27 <eris> would anyone know why refind automatically assumes that the initrd and loader parameters are located on the EFI partition?
2023-05-30 17:23:55 <irreleph4nt> eris, because it's an EFI bootloader?
2023-05-30 17:24:04 <eris> that would make sense
2023-05-30 17:24:15 <eris> but what is the point of being able to specify volume to a non-EFI partition 
2023-05-30 17:24:27 <eris> and the BTRFS and EXT and soforth drivers 
2023-05-30 17:25:08 <eris> if refind can't boot from those filesystems (or any external partition), why include filesystem drivers for things that can't work on an EFI partition?
2023-05-30 17:25:18 <irreleph4nt> ikke, it's not a standard alpine base image. It's an image build in-house which extracts the alpine minirootfs (nothing else, really). I am building php on top of that ...
2023-05-30 17:25:55 <irreleph4nt> ikke, the mirror our company runs appears to be a download-externally-and-upload-internally kind of setup
2023-05-30 17:26:02 <ikke> irreleph4nt: how do you think the alpine docker image is created?
2023-05-30 17:26:22 <irreleph4nt> it has all sorts of quirks. APKINDEXes are not refreshed for example even though new packages are added to the directory
2023-05-30 17:26:33 <ikke> irreleph4nt: oh, I do recall you mentioning that before
2023-05-30 17:26:53 <ikke> so stale apkindex files
2023-05-30 17:27:30 <irreleph4nt> ikke, yeah. I turned to this chat last week when I could not figure out how to work around the stale apkindex files :)
2023-05-30 17:27:33 <ikke> But it's strange, apk should be able to delete packages, even if they are not part of the apkindex anymore
2023-05-30 17:35:55 <minimal> irreleph4nt: I suspect an issue with the keys in your base image, either no keys, or out-of-date keys
2023-05-30 17:36:54 <ikke> maybe they still have the pre 3.15 keys
2023-05-30 17:39:33 <minimal> from the mention of "quirks" etc sounds like they've got a franken-mirror and a franken-alpine-base-image
2023-05-30 17:40:32 <minimal> eris: "EFI partition" I assume you mean ESP partition. All partitions on a EFI-partitioned device are EFI partitions ;-)
2023-05-30 18:13:09 <pj> idk why are you ignoring the fact that apk del fails
2023-05-30 18:18:15 <minimal> pj: if you're referring to me then I'm not ignoring that, rather I asked for the Dockerfile contents to understand in general how the image is being created in the first place
2023-05-30 22:24:58 <eris> hi docmax
2023-05-30 22:25:01 <eris> having a fun time?
2023-05-30 22:39:19 <eris> hey, anyone here good at bluetooth? having another issue 
2023-05-30 22:39:28 <eris> and opening IRC hasn't fixed it 
2023-05-30 22:40:17 <eris> i have a bluetooth mouse, whenever i move it, it causes bluetoothctl logs to fill with connected:yes connected:no, but it never actually moves the mouse 
2023-05-30 22:40:28 <eris> it pairs successfully, but a bluetoothctl connect <device> doesn't work
2023-05-30 22:40:52 <eris> it fails with br-connection-create-socket
2023-05-30 22:46:50 <psykose> sounds similar to something but i forget what fixes it
2023-05-30 22:47:32 <eris> there's probably a config option hidden somewhere
2023-05-30 23:07:14 <eris> no progress from just leaving it :) 
2023-05-30 23:08:09 <eris> also: whenever my laptop moves across networks, it manages to connect to a new wifi network, but it does not form an internet connection. restarting `networking` (presumbly, udhcpcd) allows it to connect to the internet once again after a move
2023-05-30 23:08:29 <eris> is there a way to automate this? do other DHCP clients have better functionality?
2023-05-30 23:08:46 <psykose> don't remember if bb udhcpc handled that well at all
2023-05-30 23:09:15 <psykose> dhcpcd probably works alright (forget how well it worked to just install it, and keep using `networking`)
2023-05-30 23:12:36 <eris> i'm also running iwd, but no networkmanager or wpasupplicant 
2023-05-30 23:13:11 <psykose> i do the same
2023-05-30 23:13:20 <psykose> i think i had installed dhcpcd and just forgotten about it eons ago
2023-05-30 23:13:22 <eris> can't iwd provide the udhcpc functionality, or am i misremembering? i used to use KISS linux and we never had udhcpcd running, iirc 
2023-05-30 23:14:15 <eris> christ, this lack of a mouse is killer 
2023-05-30 23:14:17 <psykose> there is a
2023-05-30 23:14:18 <psykose> [General]
2023-05-30 23:14:18 <psykose> EnableNetworkConfiguration=true
2023-05-30 23:14:27 <psykose> which uses the builtin dhcp client iirc
2023-05-30 23:14:33 <eris> yea, i tried that and it did.... jack shit 
2023-05-30 23:14:59 <eris> doesn't even show up in iwctl for some reason
2023-05-30 23:15:47 <eris> just checked: iwctl reports 'NetworkConfigurationEnabled: disabled'
2023-05-30 23:17:42 <eris> what's in the networking service?
2023-05-30 23:19:10 <psykose> i don't think it interacts with any of that, just starts dhcp clients and configures other interface stuff
2023-05-30 23:19:33 <psykose> after you change the config you have to restart iwd for it to read it, aside from that not sure why it would remain disabled
2023-05-30 23:19:36 <psykose> lemme see
2023-05-30 23:20:20 <eris> i've restarted a bunch
2023-05-30 23:20:30 <eris> like a solid 20 full restarts since i ticked off that option
2023-05-30 23:20:31 <eris> lol
2023-05-30 23:20:48 <eris> (i'm using the immutable root config, which means a LOT of rebooting) 
2023-05-30 23:21:30 <eris> hmmm, maybe with the bluetooth thing it was hid2hci?
2023-05-30 23:21:53 <psykose> i dunno what that does
2023-05-30 23:22:19 <psykose> could be, i've used bluetooth mice without installing that
2023-05-30 23:22:30 <eris> mmmmmmm
2023-05-30 23:22:47 <eris> yea, i was running this /exact/ mouse on my previous laptop 
2023-05-30 23:22:57 <eris> didnt have to do any fancy setup
2023-05-30 23:23:00 <eris> was using alpine too 
2023-05-30 23:23:31 <eris> different hardware though
2023-05-30 23:24:55 <psykose> yeah seems to just work with the networkconfiguration thing set
2023-05-30 23:25:13 <eris> odd
2023-05-30 23:25:46 <eris> iwctl says its disabled. ill reboot with 'networking' and 'dhcpcd' disabled, see if that causes it to kick in
2023-05-31 01:57:07 <ezntek[m]> is it wise to daily-drive alpine linux as a workstation OS?
2023-05-31 01:57:22 <psykose> no
2023-05-31 01:57:40 <ezntek[m]> noted
2023-05-31 02:05:58 <bl4ckb0ne> ezntek[m]: i do that and i cant recommend it, glib binaries wont work 
2023-05-31 02:06:05 <bl4ckb0ne> glibc*
2023-05-31 02:06:19 <bl4ckb0ne> also no multiarch, i cant enjoy my steam games
2023-05-31 02:28:55 <Saijin_Naib> Steam Flatpak and Lutris are things and they work great 
2023-05-31 02:29:08 <Saijin_Naib> ezntek: I vote yes. 
2023-05-31 02:30:13 <bl4ckb0ne> joke aside, been running alpine as daily driver on both workstation/laptop/server and its absolutely great
2023-05-31 02:30:40 <bl4ckb0ne> if what you need requires glibc, then its probably not worth it
2023-05-31 02:30:51 <bl4ckb0ne> and flatpak lutris/steam/bottles whatever works great
2023-05-31 02:31:11 <psykose> the 32-bit wine is almost done for that stuff
2023-05-31 02:31:20 <psykose> latest 8.9 actually opens more stuff without a black square screen
2023-05-31 02:31:24 <psykose> some mod installers now work
2023-05-31 02:31:36 <psykose> ..for the games you'd play in flatpak steam anyway, but yknow
2023-05-31 02:55:40 <ezntek[m]> i dont really run glibc binaries
2023-05-31 02:55:52 <ezntek[m]> i ran musl on my workstation for quite a while and I had no problems, really
2023-05-31 02:56:02 <ezntek[m]> if flatpaks are supported, then I'm down
2023-05-31 03:04:47 <invoked> i daily-drive alpine as a workstation
2023-05-31 03:04:51 <invoked> 
ACTION hides behind a plant

2023-05-31 03:13:09 <psykose> but what kind of plant?
2023-05-31 03:14:47 <quinq> A clover
2023-05-31 03:17:55 <invoked> a rubber tree
2023-05-31 03:18:28 <psykose> :3
2023-05-31 03:21:51 <ezntek[m]> how's apk
2023-05-31 03:22:01 <ezntek[m]> is it actually fast and usable
2023-05-31 03:22:11 <ezntek[m]> (i have trauma from dpkg and apt)
2023-05-31 03:22:17 <ezntek[m]> (arch kinda cured it)
2023-05-31 03:23:10 <bl4ckb0ne> sometimes you have to whack it on the side to behave
2023-05-31 03:23:31 <psykose> that's usually just a skill issue
2023-05-31 03:23:33 <bl4ckb0ne> otherwise its good, as long as psykose doesnt touch the ncurses package
2023-05-31 03:23:51 <psykose> fucking hate the ncurses package
2023-05-31 03:24:17 <ezntek[m]> wow
2023-05-31 03:24:21 <ezntek[m]> i guess ill have to see for myself, then
2023-05-31 03:33:55 <Saijin_Naib> Same hardware an Alpine setup will be installed and updated in like 15min tops, Debian-derived is 45min or more 
2023-05-31 03:34:44 <Saijin_Naib> APK is brutal fast, and has never failed a major version upgrade with held depends unlike Debian-derived 
2023-05-31 03:35:22 <Saijin_Naib> Alpine is honestly a revelation coming from Debian for me 
2023-05-31 03:35:57 <Saijin_Naib> The setup-desktop script really helps with first start where I struggled prior
2023-05-31 04:08:36 <psykose> bl4ckb0ne: did you make vulkan 50x faster yet
2023-05-31 07:32:42 <oleovit> hi. abort https://git.alpinelinux.org/aports/commit/?id=23c937e88c909a4c8e627db8448b78ee0b8a83d6 . i can't upgrade https://0x0.st/HbbG.txt
2023-05-31 07:36:39 <psykose> ah right
2023-05-31 07:36:43 <oleovit> apk update is showing me these firmware pkg's, but apk upgrade doesnt install them
2023-05-31 07:38:29 <psykose> fixed
2023-05-31 07:39:07 <oleovit> i will wait, just a info. dont want to force every pkg with apk add --upgrade . ah, fixed, thx like always @psykose
2023-05-31 07:39:25 <psykose> you can't force it
2023-05-31 07:39:59 <oleovit> i will not even try it ;)
2023-05-31 07:40:12 <oleovit> ok, thank you!
2023-05-31 09:40:02 <iggy> I'm getting a bunch of conflicts with with linux-firmware*
2023-05-31 10:04:14 <ikke> possibly related to http://dup.pw/alpine/aports/e917f162a41a
2023-05-31 10:08:54 <ikke> iggy: what conflicts do you get?
2023-05-31 10:17:39 <eris> yo
2023-05-31 11:18:54 <ptrc> does anyone also have issues with latest chromium on edge being flagged as suspicious in google searches?
2023-05-31 11:22:55 <ptrc> nevermind, it's just tunnelbroker ipv6 being always flagged
2023-05-31 11:34:20 <quinq> ptrc, out of curiosity, flagged by who, how, to who?
2023-05-31 11:34:48 <ptrc> by google as 'suspicious', so you have to fill out a few captchas whenever you google something
2023-05-31 11:35:15 <quinq> For regular search?
2023-05-31 11:35:47 <Newbyte> Is there any more "proper" way to clean the apk cache than doing `rm /var/cache/apk/*`?
2023-05-31 11:36:44 <quinq> Newbyte, man-page says apk cache purge
2023-05-31 11:36:58 <ptrc> quinq: yes, looks something like this: https://ptrc.gay/qvDsfPOE
2023-05-31 11:37:19 <quinq> Thanks ptrc 
2023-05-31 11:37:28 <quinq> First time I see this, yet another horrible web thing :(
2023-05-31 11:37:34 <ptrc> the more captchas you fill out, the easier they get, and sometimes there isn't one at all
2023-05-31 11:37:46 <ptrc> well, just another reason to set ddg/startpage as default search engine :)
2023-05-31 11:37:54 <Newbyte> quinq: thanks!
2023-05-31 11:38:03 <quinq> yeah :)
2023-05-31 11:38:48 <quinq> Sadly this is creepfeature that is seen a lot on the web nowadays, more and more people using this cloudfare thing
2023-05-31 12:54:17 <jindam> when i visit gitlab alpine aports, says about issue with arm infrastructure
2023-05-31 12:54:41 <jindam> is that the reason why we dont have openjdk for arm versions
2023-05-31 13:21:11 <invoked> quinq: recaptcha is google
2023-05-31 13:23:01 <invoked> cloudflare introduced turnstile last year but i've rarely seen it. before that they just used recaptcha or hcaptcha
2023-05-31 13:24:40 <quinq> invoked, you just have to open any website behind cloudfare without previous fingerprint there and you get it
2023-05-31 13:26:29 <invoked> it's something other than that. it's seen when people do successive searches (as ptrc mentioned)
2023-05-31 13:27:08 <reggiere[m]> I despise cloudflare, I open a site and "Please enable javascript so we can verify you are a human :3"
2023-05-31 13:27:09 <invoked> it's not perfectly reproducible. for eg, your first 1-2 searches might go fine, no captcha. then on the 3rd, it captchas you
2023-05-31 13:27:33 <reggiere[m]> I am paraphrasin cause I dont remember how exactly the message displayed goes
2023-05-31 13:27:46 <reggiere[m]> *paraphrasing
2023-05-31 13:35:43 <quinq> Exactly reggiere[m] 
2023-05-31 13:36:04 <quinq> invoked, it's not something other than “that” (not sure what's your that :))
2023-05-31 13:36:35 <quinq> It's about how the user is placed behind a javascript captcha barrier
2023-05-31 13:38:41 <invoked> my point is it's not (just) fingerprint. ptrc's complaint was about chromium using google search, and getting recaptcha'd intermittently
2023-05-31 13:39:30 <invoked> that's all google's backend making some kind of decisions when to captcha. it's not cloudflare.
2023-05-31 14:21:33 <humm> psykose: NetBSD Curses when? @ fucking hate the ncurses package
2023-05-31 14:31:50 <Armote[m]> apk upgrade seems to be failing.  linux-firmware-20230515-r1   conflict with dozens of packages.
2023-05-31 14:50:24 <ikke> Armote[m]: it has been fixed, but the builder is still building
2023-05-31 14:50:30 <ikke> (or should have been fixed)
2023-05-31 16:39:15 <PureTryOut> Uh the linux-firmware update seems to have broken linux-firmware-amdgpu, Steam doesn't launch complaining about failing to initialize my GPU
2023-05-31 16:39:21 <PureTryOut> (Steam running in Flatpak)
2023-05-31 16:52:16 <duckl1ng[m]> I think there is something wrong with your ci servers, building timeouts on connecting to gitlab.alpinelinux.org https://gitlab.alpinelinux.org/duckl1ng/aports/-/jobs/1041708
2023-05-31 17:19:40 <ikke> PureTryOut: there should be a fix underway, after chromium has been built 🤐
2023-05-31 17:20:19 <PureTryOut> Seeing the failures that might take a bit lol
2023-05-31 17:20:39 <PureTryOut> But I reverted the linux-firmware changes myself and nothing changed sadly. Testing a revert of Mesa now too
2023-05-31 17:39:12 <PureTryOut> even eglinfo reports failures. GBM platform: eglinfo: eglInitialize failed
2023-05-31 17:48:24 <ikke> duckl1ng[m]: probably a temporary network glitch
2023-05-31 17:48:28 <ikke> succeeded now
2023-05-31 17:50:56 <PureTryOut> Mesa revert didn't help either...
2023-05-31 17:51:27 <PureTryOut> amdgpu_device_initialize: amdgpu_query_info(ACCEL_WORKING) failed (-38)
2023-05-31 18:13:43 <eris> yo
2023-05-31 18:21:02 <ikke> eris: /o
2023-05-31 18:21:43 <eris> does anyone else here use the immutable root + atomic upgrades config?
2023-05-31 18:53:13 <msiism> I've finally updated my rough write-up on installing Alpine with full disk encryption: https://www.msiism.org/files/doc/alpine-fde-2023.html
2023-05-31 18:53:21 <msiism> This version also has a happy ending.
2023-05-31 18:54:16 <msiism> Not sure what I did wrong all the other times I tried to install more or less exactly like this. But the root fs is now being found.
2023-05-31 18:57:21 <msiism> I'm also wondering why I can only `grub-install --target=i386-pc …`. Target `x86_64` was not available.
2023-05-31 18:57:53 <msiism> I did use the x86_64 extended image, thoiugh.
2023-05-31 19:38:58 <lopid> why are there tildes before the hashes?
2023-05-31 19:42:50 <msiism> That's what `script` does.
2023-05-31 19:42:55 <lopid> (or squigglies befoer octothorpes)
2023-05-31 19:42:59 <lopid> oh
2023-05-31 19:43:03 <msiism> I used `script` to record stuff.
2023-05-31 19:43:12 <msiism> It does a terrbile job, but it does the job.
2023-05-31 19:43:40 <msiism> Had to do a lot of editing afterwards.
2023-05-31 19:44:03 <lopid> good job, nonetheless (you)
2023-05-31 19:44:03 <ikke> Isn't that PWD it prints before #?
2023-05-31 19:44:44 <msiism> Probably.
2023-05-31 19:44:58 <msiism> Maybe there's a way to configure it better.
2023-05-31 19:45:15 <lopid> impossible to edit the web page?
2023-05-31 19:45:31 <msiism> Oh, well, sure, I could do that.
2023-05-31 19:45:39 <msiism> May it's less confusing then.
2023-05-31 19:45:45 <msiism> s/May/Maybe/
2023-05-31 19:46:23 <lopid> it's a shame that root's conventional prompt contains a comment marker, but then, that is intentional…
2023-05-31 19:46:33 <msiism> Oh, it is?
2023-05-31 19:46:35 <msiism> I was wondering.
2023-05-31 19:46:53 <lopid> it could be a myth, but it de-risks copypasting
2023-05-31 19:48:06 <msiism> Yeah, that seems to make sense. But then, people probably didn't really copy-paste that much when that was invented, did they?
2023-05-31 19:48:16 <lopid> right
2023-05-31 19:49:05 <lopid> probably one of those things that has all sorts of post implementation justifications, but ultimately came down to the author picked a character at random
2023-05-31 19:50:17 <msiism> Okay, the tildes are gone.
2023-05-31 19:51:08 <lopid> is it WIP because of it asking for the passphrase twice?
2023-05-31 19:52:51 <msiism> Yes, and also because it is a really rough version.
2023-05-31 19:53:14 <msiism> It's stripped down to what is needed to find and correct errors in how I install.
2023-05-31 19:53:33 <msiism> The actual guide has quite a bit more to it.
2023-05-31 19:53:59 <lopid> looks better than some of the official wiki's pages. perhaps it could be linked to from there
2023-05-31 19:54:28 <msiism> I might copy the final version over to the wiki.
2023-05-31 19:54:59 <msiism> My guide is going to have the same license as pages on the wiki.
2023-05-31 19:56:23 <msiism> I'd also really like to know why the GRUB target is i386-pc and not x86_64.
2023-05-31 19:57:53 <lopid> 32-bit more compatible than 64-bit?
2023-05-31 19:58:04 <msiism> Possibly.
2023-05-31 20:12:54 <WhyNotHugo> What's with the FOURTEEN mounts in /sys/fs/cgroup? They kinda make the output of `mount` a pain to read. Are these even really used?
2023-05-31 20:30:00 <msiism> Okay, just gotta add "cryptkey" to GRUB_CMDLINE_LINUX_DEFAULT.
2023-05-31 20:58:49 <frag> whats a good thinkpad, or are most/all good?
2023-05-31 21:05:08 <jvoisin> x230 über alles.
2023-05-31 21:05:12 <eris> msiism: lol, i could've done with that two days ago wehn i was setting up my new laptop :) 
2023-05-31 21:05:21 <eris> > BIOS/MBR 
2023-05-31 21:05:23 <eris> Ah
2023-05-31 21:05:26 <eris> never mind 
2023-05-31 21:06:22 <eris> 'frag: why restrict yourself to a thinkpad? :P 
2023-05-31 21:07:29 <humm> lopid: octothorpe as root prompt is older than octothorpes as comments in the shell, even
2023-05-31 21:08:55 <msiism> eris: I don't own a UEFI system. But when I ever get one, I could also write a guide for that.
2023-05-31 21:09:50 <frag> eris: ehh, actually ordered a pinebook pro, but kinda want to try a thinkpad too
2023-05-31 21:10:05 <frag> why,  what do you recommend eris ?
2023-05-31 21:10:07 <eris> oh, ive not heard good things about the pinebooks 
2023-05-31 21:10:23 <eris> i used a dell XPS up until recently, where it stopped powering on :D 
2023-05-31 21:10:39 <eris> and im using a Framework now, which so far as been my most solid option, in the whole two days that i've used it 
2023-05-31 21:10:59 <frag> dell xps seems nice ja
2023-05-31 21:11:04 <eris> other than the fact that the laptop's screen brightness can't be changed by xbacklight....... (but XFCE's works for some reason?) 
2023-05-31 21:11:20 <eris> frag: i'd never reccomend you buy one though, considering mine doesn't power on anymore completely out of the blue 
2023-05-31 21:11:41 <frag> eris: was it old?
2023-05-31 21:11:51 <eris> no
2023-05-31 21:12:03 <eris> its still under warranty 
2023-05-31 21:12:14 <eris> apparantly a common problem with this model - the 13 inch ones 
2023-05-31 21:13:25 <frag> wow...
2023-05-31 21:14:00 <minimal> msiism: still don't understand why you are setting GRUB_PRELOAD_MODULES...
2023-05-31 21:15:32 <msiism> It's in the file already. I didn't add it.
2023-05-31 21:19:30 <quinq> eris, as an alternative from xbacklight, brightnessctl could work (sorry just barging in since having read the full backlog)
2023-05-31 21:19:44 <quinq> s/from/to/
2023-05-31 21:20:11 <eris> what package is brightnessctl provided by? 
2023-05-31 21:20:23 <eris> i don't think it'd work, regardless
2023-05-31 21:20:31 <eris> the sysfs files dont seem to have any effect 
2023-05-31 21:20:33 <quinq> By brightnessctl package
2023-05-31 21:25:12 <minimal> msiism: strange. setup-disk creates/overwrites that file but only puts 4 lines it, none of which are PRELOAD
2023-05-31 21:27:06 <msiism> But isn't PRELOAD_MODULES required to make the encrypted /boot work?
2023-05-31 21:27:45 <minimal> msiism: so after step 9 there should only be 4 lines in that file, so how then in step 10 are there more lines present?
2023-05-31 21:28:43 <minimal> msiism: nope, I mentioned before that PRELOAD_MODULES should not be required for cryptodisk, indeed I believe during your testing the other day you removed that line and it made no difference
2023-05-31 21:29:15 <msiism> Hm… okay, no that you say it, I think I vaguely remember that, yeah.
2023-05-31 21:29:41 <msiism> Well, doesn't mkinitfs generate a GRUB config file too?
2023-05-31 21:29:51 <minimal> and there's nothing in setup-disk to add the GRUB_ENABLE_CRYPTODISK line either, so how did it get there?
2023-05-31 21:31:26 <minimal> msiism: mkinitfs? nope it is generic regardless of bootloader
2023-05-31 21:32:18 <msiism> Oops, a little connection break…
2023-05-31 21:35:27 <eris> woah
2023-05-31 21:35:33 <eris> brightnessctl works
2023-05-31 21:36:20 <quinq> :)
2023-05-31 21:38:11 <quinq> I honestly don't remember the code path there, but they fallback on severeal APIs, had a quick look when I was bothered that xbacklight didn't work while brightnessctl did, but it worked
2023-05-31 21:40:28 <minimal> msiism: re grub-install "i386-pc" just means its a MBR/BIOS setup whereas "x86_64-efi" means UEFI (there's another value for 32bit UEFI, forget what it is)
2023-05-31 21:41:20 <msiism> Oh, so 'x86_64' doesn't even exist as a target then?
2023-05-31 21:41:37 <eris> hmm, time to figure out how to specify my timezone
2023-05-31 21:41:48 <eris> im an hour behind....
2023-05-31 21:42:01 <minimal> msiism: have you looked at the grub-install manpage?
2023-05-31 21:43:42 <msiism> Not yet. I can't seem to find it.
2023-05-31 21:43:58 <msiism> Oh, gotta install docs…
2023-05-31 21:44:59 <msiism> Uhhm… this thing doesn't seem to have a man page.
2023-05-31 21:45:17 <minimal> msiism: so have you figured out where GRUB_PRELOAD_MODULES and GRUB_ENABLE_CRYPTODISK are coming from?
2023-05-31 21:45:46 <minimal> msiism: ok, info file then (I think Debian created their own grub-install manpage)
2023-05-31 21:46:03 <msiism> grub-install --help has the info. Just reading it.
2023-05-31 21:47:29 <msiism> No idea yet about GRUB_PRELOAD_MODULES and GRUB_ENABLE_CRYPTODISK.
2023-05-31 21:47:35 <msiism> I'll have ti investigate this tomorrow.
2023-05-31 21:48:07 <minimal> WhyNotHugo: you'll only have any cgroup mounts if the init.d/cgroups is enabled. Also the type and number of such mounts also depends on what rc_cgroup_mode is set to.
2023-05-31 21:48:28 <minimal> msiism: did you look at the shellscript example I pointed you to?
2023-05-31 21:49:18 <msiism> You mean the one you pointed me to yesterday?
2023-05-31 21:51:26 <msiism> No, didn't have time yet. But I'll put it on my TODO.
2023-05-31 22:01:26 <msiism> Having trouble installing Git. I'm beginning to suspect either my RAM or my HDD are on their way out.
2023-05-31 22:07:32 <msiism> Actually, seems to be a netowkring issue. I'll deal with that tomorrow.
2023-05-31 22:11:56 <LAC-LapTop> finally time to install alpine on my main machine. here's hoping I can run the glibc stuff I want through flatpak and it'll all be fine.
2023-05-31 22:44:53 <LAC-LapTop> has setup alpine changed? I'm sure it used to ask you what kind of install you want to do
2023-05-31 22:44:59 <LAC-LapTop> ie a sys install
2023-05-31 22:46:47 <LAC-LapTop> nvm, I didn't select a disk so it didn't answer
2023-05-31 22:49:57 <zcrayfish> flatpak works fine on Alpine. For the win.
2023-05-31 22:58:11 <LAC-LapTop> so the install I'm overrwriting had a partition for /, and a partition for the home directory. Do I just mount the root directroy in alpine, then call setup-disk /mnt? I assume I can assign the home directory to my existing home partition later
2023-05-31 22:58:15 <LAC-LapTop> (and yes I've backed up)
2023-05-31 23:55:36 <psykose> WhyNotHugo: those are docker mounts iirc for random stuff
2023-05-31 23:55:41 <psykose> i don't have 14 mounts