2021-12-01 06:53:03 <RyuKurisu[m]> Yesterday I heard from Adriadne Conill that there are two other projects that want to create desktop distributions based on Alpine Linux.  Do any of you how I might get in contact with them?
2021-12-01 07:17:33 <capra[m]> I found the issue with the bad signatures for vim and xxd... (full message at https://matrix.org/_matrix/media/r0/download/matrix.org/wglrtExTBYGdGESDNzMdIXOI)
2021-12-01 07:53:12 <dngray[m]1> alpine_dev=usbdisk:vfat 
2021-12-01 07:53:15 <dngray[m]1> is that still current?
2021-12-01 07:53:26 <dngray[m]1> i installed alpine on a usb stick and for some reason it thinks it is /dev/sdi
2021-12-01 07:53:35 <dngray[m]1> https://wiki.alpinelinux.org/wiki/Create_a_Bootable_Device#Troubleshooting
2021-12-01 08:12:19 <dngray[m]1> APPEND alpine_dev=usbdisk:vfat modules=loop,squashfs,sd-mod,usb-storage quiet
2021-12-01 08:12:43 <dngray[m]1> i thought that would hajve worked in syslinux.conf
2021-12-01 08:21:28 <dngray[m]1> weird like my server ignores that option
2021-12-01 08:21:39 <dngray[m]1> and just picks whatever device label it feels
2021-12-01 09:56:52 <jhjacobs81> hmz
2021-12-01 16:28:59 <remexre> is one of gdb-multiarch or qemu built weirdly wrt the names it uses for architectures in the gdb server protocol? it seems that the QEMU gdb server calls it aarch64 where alpine gdb-multiarch calls it armv8-a
2021-12-01 16:31:39 <remexre> (it looks like it might be gdb-multiarch; on gentoo, it works, with architecture=aarch64)
2021-12-01 17:06:25 <newbie61> Hello all. I am having difficulties to install alpine linux to my old macbook pro. can I have help here? if this is not the right channel, please excuse me. 
2021-12-01 17:07:32 <lopid> i think this is the correct channel
2021-12-01 17:09:15 <newbie61> Thank you. Then I can write the details. I downloaded the standard iso. burnt it to my USB flash using etcher. I plugged it to my computer and boot it up. I held the option key down during the boot up. But nothing happened. I tried 3 times. No go. System justs hangs on white screen. I didn't get any command line or GUI installer environment. 
2021-12-01 17:15:15 <lopid> i'm not a mac person so i don't know what the option key is supposed to do there. this channel is quite low in traffic so you might have to wait a little while for somebody who knows better to reply
2021-12-01 17:17:02 <newbie61> Thank very much lopid. I appreciate it. 
2021-12-01 17:18:32 <newbie61> Option key , pressed during the bootup on macbook computers allows user to choose among available start up places, like harddrive or USB stick etc.
2021-12-01 17:19:11 <psykose> if you get nothing from holding option you have some other problem pretty sure, as that should always output something
2021-12-01 17:19:46 <psykose> you also should partition the disk first from osx recovery
2021-12-01 17:22:19 <newbie61> Thanks much msykose. I have just installed popos on the same machine using the very same USB stick and the same method. The disk on the machine still has Popos. and if I remove the stick and boot up, popos still boots up properly. 
2021-12-01 17:23:30 <invoked> which macbook is this
2021-12-01 17:23:36 <newbie61> I have to add that I verified the ISO I downloaded from the alpine linux site. So, I think it should work out of the box and there should not be any tricks.
2021-12-01 17:24:23 <mtm84[m]> Macs need the usb stick to have a certain partition layout or it won’t show up in the boot manager.  Maybe alpines iso is setup differently then popos?
2021-12-01 17:24:25 <newbie61> If my recollection is right, it must be 2009 macbook pro 17"
2021-12-01 17:24:48 <invoked> 2009, wow that's an oldie
2021-12-01 17:25:00 <newbie61> mtm84[m] this could be it.
2021-12-01 17:25:28 <newbie61> I am not sure what I should do? Should I give up Alpine altogether and stick to popos ? 
2021-12-01 17:26:05 <invoked> i think the earliest macbook i've installed alpine on was a 15" 2011 mbp, but i don't remember doing anything special
2021-12-01 17:26:13 <mtm84[m]> Did popos install grub?
2021-12-01 17:26:16 <newbie61> invoked it runs popos quite nicely. :)  (I upgraded the HDD to SSD of course)
2021-12-01 17:26:59 <invoked> if it were me, i'd try doing the usual mac stuff (reset pram, et al)
2021-12-01 17:27:13 <invoked> not that it will help you, but i don't think it will hurt either
2021-12-01 17:27:22 <newbie61> mtm84[m] I don't think so. It used UEFI I think
2021-12-01 17:27:40 <invoked> yeah all intel macs are uefi only
2021-12-01 17:27:59 <invoked> and apple's sort of strange efi at that.
2021-12-01 17:28:24 <invoked> strange = slightly custom subset of full uefi
2021-12-01 17:28:44 <newbie61> invoked what do you mean by reset pram et al? 
2021-12-01 17:29:12 <invoked> hold cmd + option + p + r at boot (i think that's the keys)
2021-12-01 17:29:32 <invoked> keep holding it down and let it chime 5 times if you want a really completely pram reset
2021-12-01 17:29:38 <invoked> but usually 1 cycle will suffice
2021-12-01 17:29:43 <mtm84[m]> I just didn’t know if grub was a standard thing on Intel mac linux installs or not.  If it has grub you could probably use that to boot the usb.  I’m really but sure why etcher would fail on alpine but not popos
2021-12-01 17:30:36 <newbie61> mtm84[m] I am surprised as well. strange behavior indeed. 
2021-12-01 17:30:48 <newbie61> I will try invoked's idea. and let you know. 
2021-12-01 17:30:59 <invoked> 
ACTION is not saying it will work :)

2021-12-01 17:30:59 <newbie61> thank you a lot guys. I appreciate it. Thanks
2021-12-01 17:31:15 <invoked> i just don't have any notes on my macbook installs, but like i said, i don't remember doing anything special
2021-12-01 17:31:18 <newbie61> but it wont hurt either :)  1 > 0 :)
2021-12-01 17:31:30 <invoked> but then sometimes i take the things i know for granted and don't remember my own workarounds :-/
2021-12-01 17:31:44 <mtm84[m]> And unfortunately all my alpine install experience is VM related and my mac linux experience is ppc32 related lol
2021-12-01 17:34:34 <mtm84[m]> In the interest of completeness, you did download the amd64 iso right?  Cause I’ve definitely downloaded the wrong architecture before by accident.
2021-12-01 17:36:10 <newbie61> I downloaded alpine-extended-3.15.0-x86_64.iso and alpine-standard-3.15.0-x86_64.iso
2021-12-01 17:36:18 <newbie61> I tried both.
2021-12-01 17:37:33 <mtm84[m]> Ok, Just a “Is it plugged in” sanity check
2021-12-01 17:38:34 <mtm84[m]> I don’t know that an alpine layout would be so different  from a popos layout that etcher wouldn’t make a bootable disk
2021-12-01 17:39:37 <newbie61> invokedI tried Option-Command-P-R: Reset NVRAM or PRAM. then tried booting again. still the same :\
2021-12-01 17:40:25 <psykose> i guess you could also try to dd the disk instead
2021-12-01 17:41:19 <newbie61> psykose it is a great idea. :)  thanks!
2021-12-01 18:04:11 <newbie61> I tried dd as psykose suggested. it didn't make any difference. as a last resort, I used a usb flasher utility named popsicle.. still the same. 
2021-12-01 18:04:19 <psykose> unfortunate
2021-12-01 18:04:27 <newbie61> thanks for your efforts guys. I really appreciate it
2021-12-01 18:04:35 <psykose> can you flash the popos installer and still boot to installer, just to make sure it's alpine alone?
2021-12-01 18:06:14 <newbie61> that's my next move actually. I hacked popos as I tried to enable hibernate on it, so I want to clean install it. I wanted to give Alpine Linux a try as it looks more interesting to be honest. 
2021-12-01 18:15:47 <newbie61> I flashed popos iso onto the same usb stick. booted it up. and it worked as expected. 
2021-12-01 18:16:40 <newbie61> so I think alpine linux iso is not uefi friendly and since there is no BIOS on this mac bookpro, I can not use it. 
2021-12-01 18:17:05 <newbie61> Thank you guys for your time. 
2021-12-01 18:22:00 <Hello71> apple uefi does a lot of non-standard things
2021-12-01 18:22:24 <Hello71> there have been reports of alpine not being compatible with older apple uefi
2021-12-01 18:22:49 <Hello71> i think it is not the hybrid uefi problem but some other strange behavior
2021-12-01 18:23:04 <helby> newbie61: it is uefi friendly
2021-12-01 18:23:48 <newbie61> I really do not know. 
2021-12-01 18:24:07 <newbie61> so, I decided I will not use alpine on this specific machine. 
2021-12-01 18:24:12 <helby> I would try to make that usb with something else if dd does not work
2021-12-01 18:25:43 <newbie61> I tried 2 different machines (1 windows 1 popos) and three different utility programs: etcher, dd, popsicle . still no good
2021-12-01 18:29:05 <psykose> probably is the mac uefi behaviour then, unlucky
2021-12-01 18:29:58 <newbie61> looks like it. :\  I got to find another machine to try alpine linux  somehow. 
2021-12-01 18:36:15 <invoked> obviously if it's working on other linux this is an alpine bug
2021-12-01 18:37:01 <invoked> but not one i would expect to be fixed anytime soon
2021-12-01 18:37:16 <invoked> especially without useful context/debug
2021-12-01 18:37:46 <psykose> the context is pretty useful
2021-12-01 18:38:43 <newbie61> is there anyway I can provide any further information ?
2021-12-01 18:39:17 <invoked> get the specific model # of the macbook, for starters
2021-12-01 18:39:50 <invoked> that would go a long way. once we know that, probably, some other distro has notes on it.
2021-12-01 18:40:03 <Hello71> i doubt it
2021-12-01 18:40:58 <invoked> you can doubt it, but i'll point to https://wiki.archlinux.org/title/MacBook5,2_(early-mid_2009) as an example
2021-12-01 18:40:59 <Hello71> i think it probably affects all intel macs before 2007-2009 or thereabouts. probably it is related to the precise xorriso invocation
2021-12-01 18:41:38 <Hello71> what does that have to do with efi
2021-12-01 18:42:18 <invoked> who said anything about efi
2021-12-01 18:42:52 <Hello71> sigh
2021-12-01 18:42:55 <invoked> but the link does mention refit specifically
2021-12-01 19:02:13 <newbie61> OK. sorry for the delay.. 
2021-12-01 19:03:50 <newbie61> My machine is Mac BookPro release date is mid 2009
2021-12-01 19:05:09 <newbie61> macbookpro5,2
2021-12-01 19:05:55 <newbie61> invoked the URL you sent just hits the mark.
2021-12-01 19:06:37 <newbie61> Thanks for your time and effort to help me. I appreciate it. 
2021-12-01 19:08:49 <invoked> newbie61: no probs, good luck
2021-12-01 19:59:09 <Nulo> Hey, the options for mounting / in fstab aren't actually used, any clue?
2021-12-01 20:04:42 <Newbyte> Nulo: I think you have to enable the localmount service or something like that
2021-12-01 20:08:44 <Nulo> Newbyte, thanks, I'll try that
2021-12-01 20:12:32 <helby> you can check mountin with 'mount -a' it will try to mount from fstab 
2021-12-01 20:12:50 <helby> if you did not hacked your system I assume localmount service is running anyway
2021-12-01 22:40:17 <kxzaon> exit
2021-12-01 23:36:02 <freq> 5hi
2021-12-01 23:38:06 <psykose> what's up
2021-12-01 23:39:41 <freq> well
2021-12-01 23:39:56 <freq> i guess i have a lot of questions not regarding alpine
2021-12-01 23:40:05 <freq> because alpine is pretty cut and dry lol
2021-12-01 23:40:15 <freq> installer is spot-on
2021-12-01 23:40:22 <freq> and the rest is solid
2021-12-01 23:40:28 <freq> why do you guys even have a support chat
2021-12-01 23:41:36 <PJ[m]> is this a serious question?
2021-12-01 23:41:46 <freq> nah i mean that was the joke
2021-12-01 23:41:48 <freq> carry on
2021-12-02 00:00:41 <helby> it's not support chat
2021-12-02 00:01:19 <helby> bc. bc. it's 'pretty cut and dry' then no issues, so we need irc where we can spend time instead of fuckering with system ;/
2021-12-02 00:08:24 <mtm84[m]> IRC is where you hang out while test patches compile, right?  That sounds like a good cover to me.
2021-12-02 01:27:03 <bdju> does anyone know what I have to install or do to get /dev/snd/seq to appear? I have alsa, pulseaudio, fluidsynth...
2021-12-02 03:08:43 <Hello71> modprobe something-or-other
2021-12-02 03:17:28 <Hello71> snd-seq i think
2021-12-02 03:23:27 <bdju> I tried that already the other day
2021-12-02 03:24:30 <bdju> modprobe: FATAL: Module snd-seq not found in directory /lib/modules/5.14.9
2021-12-02 03:35:08 <Hello71> sounds like you're not using alpine
2021-12-02 03:35:14 <Hello71> or you're using docker or some nonsense
2021-12-02 03:37:55 <bdju> yeah, sorry, using postmarketOS. I went there and elsewhere first but haven't been having luck
2021-12-02 03:54:35 <Hello71> well the kernel obviously doesn't have snd-seq enabled?
2021-12-02 03:55:01 <bdju> I did consider that possibility but it didn't seem obvious to me. thanks
2021-12-02 03:58:14 <Hello71> well pmos is open source, so why don't you check
2021-12-02 03:59:37 <Hello71> it would have saved considerable time if you had started with "i'm using pmos on DEVICE and can't get /dev/snd/seq to show up. i tried modprobe snd-seq but it gave modprobe: FATAL: Module snd-seq not found in directory /lib/modules/5.14.9"
2021-12-02 04:00:29 <bdju> I would say my thought process was not wanting to give misleading information that could block off certain ideas that could be helpful
2021-12-02 04:00:40 <bdju> like not wanting to tell the doctor too much in case you misrepresent your symptoms and get the wrong meds
2021-12-02 04:01:19 <bdju> starting from scratch allows new perspectives
2021-12-02 04:03:37 <Hello71> i see it as 99% of the time wasting an hour needling out basic details
2021-12-02 04:04:14 <Hello71> or more likely me just giving up halfway through because it's a waste of time
2021-12-02 04:04:50 <bdju> this was nowhere near that long and I have zero regrets about my approach. I understand your perspective, though.
2021-12-02 04:05:03 <bdju> I could just as easily see someone responding to a list of things I tried with "that has nothing to do with x..."
2021-12-02 04:06:08 <bdju> I was also being a bit sneaky about not using alpine because I had the feeling I wouldn't get help if I started with that
2021-12-02 04:06:14 <Hello71> 03:08 <Hello71> modprobe something-or-other
2021-12-02 04:06:22 <Hello71> 03:54 <Hello71> well the kernel obviously doesn't have snd-seq enabled?
2021-12-02 04:06:30 <Hello71> is your clock slow or something
2021-12-02 04:06:49 <bdju> short answer: no
2021-12-02 04:06:51 <Hello71> that's just trolling and wasting people's time
2021-12-02 04:07:00 <Hello71> as you wasted mine
2021-12-02 04:07:05 <bdju> I may have missed my own first message when glancing at the timestamps just now
2021-12-02 04:07:43 <bdju> I'd still argue that an hour of time wasn't wasted. I guess I'm not sure what you were doing on your end, but this is all async in my mind and after I posted I was doing other things for a bit
2021-12-02 04:08:30 <bdju> I don't think it's trolling at all either
2021-12-02 04:08:47 <bdju> I strongly believe there are situations where you could get an answer you might not have gotten otherwise if you'd for some reason locked yourself out of help
2021-12-02 04:08:52 <Hello71> that's a pretty crappy excuse
2021-12-02 04:08:58 <bdju> okay
2021-12-02 04:12:04 <Hello71> even if what you're saying is true, which it absolutely isn't, it's shitty to knowingly withhold information and waste volunteers' time
2021-12-02 04:13:07 <Hello71> at least the doctor is paid to listen to your crap, and if you get a bad answer because you didn't tell the doctor everything, that's your own problem
2021-12-02 04:23:53 <bdju> it was just an analogy, not like I've had issues with doctors. I'm withholding info I'm unsure about. seems like a safe bet to me
2021-12-02 04:24:05 <bdju> better than risking misleading people, imo
2021-12-02 04:24:48 <bdju> also worth noting the channel wasn't busy and I wasn't stopping other users from getting help from you or anyone else by asking my question as far as I know
2021-12-02 04:48:52 <Hello71> there's 526 non-package-request issues open at aports to work on.
2021-12-02 04:51:28 <Hello71> deliberately implying information that you know to be false is called "lying by omission". if you join a channel called "alpine linux" and ask for linux help and intentionally don't tell people that you're not using alpine linux, that's lying by omission.
2021-12-02 04:52:37 <bdju> it's not like I was running ubuntu, pmOS is based on alpine, and someone suggested I come here when I was troubleshooting this elsewhere
2021-12-02 04:53:18 <bdju> I also didn't expect there to be major differences
2021-12-02 05:11:01 <Hello71> if you knew what was relevant, then you would be able to fix it
2021-12-02 09:53:11 <telmich> Which browsers are you using? I am asking for a secondary browser that I can permanently setup with a proxy configuration that I need for specific work environments
2021-12-02 09:53:32 <telmich> My main two browsers are firefox/chromium and I am looking for something else that I can hardwire to a socks port
2021-12-02 09:54:06 <lopid> for this i use firefox multi-account containers. i used to use foxyproxy per site, but they recently added proxy support to containers
2021-12-02 09:54:31 <lopid> you could also use a different firefox profile
2021-12-02 09:54:56 <telmich> multi account containers? I'll need to check that out - you can run them in parallel?
2021-12-02 09:55:17 <lopid> yes, many tabs in many containers in one browser
2021-12-02 09:56:17 <telmich> That sounds great
2021-12-02 09:59:29 <telmich> hmm, it seems DNS resolution of a socks proxy is not getting through with the container tab
2021-12-02 09:59:58 <lopid> oh. what about dns over https?
2021-12-02 10:01:01 <telmich> I don't see a setting on the per container to reconfigure dns
2021-12-02 10:01:12 <lopid> not per container, it's a global
2021-12-02 10:01:43 <lopid> wouldn't dns resolution have to be done beforehand anyway, regardless? doesn't socks not handle not?
2021-12-02 10:03:20 <lopid> yes, i think the socks proxy itself does the lookup
2021-12-02 10:03:30 <lopid> https://stackoverflow.com/questions/33099569/how-does-sock-5-proxy-ing-of-dns-work-in-browsers
2021-12-02 10:04:14 <lopid> i think that's how i've mine set up, but unfortunately my company has exposed its internal ips to public dns, so i probably wouldn't have needed to(!)
2021-12-02 10:05:24 <telmich> So aside from the containers, there is a check flag
2021-12-02 10:05:43 <telmich> If you enable a socks proxy globally in firefox, you can choose whether or not to forward DNS requests via the socks proxy
2021-12-02 10:05:55 <telmich> And that setting doesn't exist in the multi container proxy config
2021-12-02 10:06:46 <lopid> it's just released, you might have to experiment
2021-12-02 10:13:21 <lopid> foxyproxy does support dns via socks5
2021-12-02 10:21:28 <telmich> ha! others have been ahead of me: https://github.com/mozilla/multi-account-containers/issues/2213
2021-12-02 10:22:58 <lopid> "We can definitely implement the second issue in the next release.", nice
2021-12-02 15:13:36 <donoban> ouch, after rebooting I can't start sway
2021-12-02 15:14:36 <helby> thanks for the info ;/
2021-12-02 15:14:48 <donoban> :D
2021-12-02 15:15:03 <helby> u get some error probably ...
2021-12-02 15:15:13 <donoban> but I'm not sure if some update broke it or it is due to some package cleanup
2021-12-02 15:16:01 <donoban> I don't find any error about tinydm or sway, there was some complain about elogind but I think that it's just a warning since autologin tries to load it optionally
2021-12-02 15:16:27 <helby> oh u don't start it from tty
2021-12-02 15:16:38 <donoban> if try to manually start sway it complains about my keymap and then segfaults
2021-12-02 15:16:50 <donoban> but the segfault only appears if I try to manually start it
2021-12-02 15:17:07 <donoban> I'm gonna try to remember how to copy/paste with tmux
2021-12-02 15:18:22 <psykose> start sway with -dV &> log and upload it
2021-12-02 15:22:08 <donoban> ok thnaks 
2021-12-02 15:23:37 <donoban> http://ix.io/3GMV
2021-12-02 15:23:55 <donoban> and the dmesg segfault...
2021-12-02 15:24:25 <donoban> [  546.571507] sway[4234]: segfault at 8 ip 00007f9a2eb607b9 sp 00007fffbdf41e38 error 6 in libxkbcommon.so.0.0.0[7f9a2eb4d000+16000]
2021-12-02 15:24:26 <donoban> [  546.571528] Code: 24 08 48 83 c4 18 e9 c0 21 00 00 83 c8 ff 48 83 c4 18 c3 31 c0 39 77 18 77 0f 39 77 1c 72 0a 89 f6 48 6b c6 30 48 03 47 20 c3 <ff> 47 08 48 89 f8 c3 48 85 ff 0f 84 65 01 00 00 41 56 41 55 41 54
2021-12-02 15:24:36 <psykose> looks like an xkbcommon issue alone
2021-12-02 15:24:58 <donoban> when I just reboot it doesn't log any segfault
2021-12-02 15:25:40 <donoban> I tried to change es-cat to es-es
2021-12-02 15:26:04 <donoban> maybe should try with us
2021-12-02 15:26:57 <donoban> could I be missing some package?
2021-12-02 15:28:01 <psykose> xkb_layout should just be `es` for spanish
2021-12-02 15:28:17 <donoban> it's ok
2021-12-02 15:28:20 <donoban> es-cat is the VARIANT
2021-12-02 15:28:42 <psykose> yes, and cat is in xkb_variant
2021-12-02 15:30:03 <psykose> like this https://img.ayaya.dev/J1BvT1nyy2mg.webp
2021-12-02 15:30:27 <donoban> hehe I can't see images at this moment
2021-12-02 15:30:35 <psykose> :p
2021-12-02 15:31:27 <psykose> input * { xkb_layout es \n xkb_variant cat }
2021-12-02 15:32:22 <donoban> well, the keymap was running fine for months . Is not more likely that I broke it removing some package?
2021-12-02 15:32:37 <psykose> all of them are from libx11
2021-12-02 15:32:39 <donoban> or there is some bug on an update
2021-12-02 15:32:50 <psykose> and xkeyboard-config
2021-12-02 15:32:51 <psykose> i think
2021-12-02 15:33:06 <psykose> but those are dependencies
2021-12-02 15:33:22 <psykose> start with apk fix to see if everything is actually updated
2021-12-02 15:33:45 <donoban> it semes ok
2021-12-02 15:34:34 <psykose> i can't think of why it would randomly break sadly
2021-12-02 15:34:42 <donoban> I'm gonna try with other desktop
2021-12-02 15:34:59 <donoban> do you know another not very big with wayland?
2021-12-02 15:35:05 <helby> river
2021-12-02 15:35:33 <psykose> you could also post your sway config i guess
2021-12-02 15:35:34 <psykose> just in case
2021-12-02 15:36:07 <helby> and try default by removing your own
2021-12-02 15:36:33 <psykose> just sway -c /etc/sway/config is fine
2021-12-02 15:38:36 <donoban> ouch, river started a grey screen where I could move the cursor but any key or button did anything
2021-12-02 15:38:40 <donoban> so I had to poweroff
2021-12-02 15:38:50 <psykose> that sounds normal
2021-12-02 15:38:53 <psykose> so i guess it works
2021-12-02 15:39:18 <psykose> something-enter opens a terminal by default iirc
2021-12-02 15:39:26 <helby> foot
2021-12-02 15:39:30 <donoban> let's try the plain sway
2021-12-02 15:40:02 <donoban> same error and segfault
2021-12-02 15:40:37 <psykose> are you on edge
2021-12-02 15:40:53 <donoban> yes
2021-12-02 15:41:06 <donoban> is river compatible with sway config?
2021-12-02 15:41:07 <psykose> no idea then sadly
2021-12-02 15:41:24 <helby> no, it's different, example config is 'river-doc'
2021-12-02 15:41:26 <psykose> you can `apk fix \*` to reinstall everything :p but i don't think that will fix it
2021-12-02 15:41:32 <donoban> well being able to open a terminal I'm ok for now
2021-12-02 15:41:49 <helby> so install foot
2021-12-02 15:41:53 <donoban> I have it
2021-12-02 15:42:36 <donoban> do you know what key combinatio opens it?
2021-12-02 15:43:00 <helby> or run setup-keymap and setup something it's there
2021-12-02 15:43:37 <helby> mod4 shifrt return by default
2021-12-02 15:43:46 <donoban> well I've alredy tried reconfiguring it
2021-12-02 15:44:03 <donoban> I would like to get a workable desktop, later I can investigate this deeper
2021-12-02 15:44:13 <donoban> I have root snapshots for everyday
2021-12-02 15:44:34 <donoban> wow, what a key combinatio :D
2021-12-02 15:44:45 <helby> same as in sway I assume
2021-12-02 15:45:00 <helby> most keybinding is the same
2021-12-02 15:45:09 <helby> maybe extra shift no idea
2021-12-02 15:50:40 <donoban> ouch nothing
2021-12-02 15:50:48 <donoban> I think that it's a better idea to just boot into an old snapshot
2021-12-02 16:15:41 <donoban> yeah!
2021-12-02 16:15:45 <donoban> sway working again
2021-12-02 16:16:16 <donoban> can I check how many updates I have pending without doing them?
2021-12-02 16:19:17 <artok> -s is for simulating
2021-12-02 16:19:51 <donoban> apk list -u 
2021-12-02 16:19:57 <donoban> also works
2021-12-02 16:20:22 <artok> aand that
2021-12-02 16:20:32 <donoban> I'm not missing too many updates but one of them is... sway-1.6.1-r2 x86_64 {sway} (MIT) [upgradable from: sway-1.6.1-r1]
2021-12-02 16:21:14 <donoban> well I will try to discover what happened later
2021-12-02 16:21:34 <donoban> thanks for the help guys
2021-12-02 16:59:15 <Jenkler> Anyone having issues with 1.15 install media? getting https://tpaste.us/lylX
2021-12-02 17:00:06 <Jenkler> i only have "http://alpine.mirror.far.fi/v3.15/main" in the /etc/apk/repositories
2021-12-02 17:02:04 <psykose> that mirror is weeks behind
2021-12-02 17:02:11 <psykose> and is generally always weeks behind, i wouldn't use it
2021-12-02 17:02:31 <Jenkler> psykose, ahh 
2021-12-02 17:03:26 <psykose> the last time they updated 3.15 was on 12 nov, which is almost 2 weeks before it was even formally released lol
2021-12-02 17:07:13 <psykose> if you're in europe i recommend the uk. mirror, it's pretty fast and never more than like 1-2 hours behind
2021-12-02 17:07:18 <Jenkler> psykose, is dl-cdn.alpinelinux.org the default mirror?
2021-12-02 17:07:21 <psykose> yes
2021-12-02 17:07:25 <psykose> that's where everything goes first
2021-12-02 17:16:23 <Jenkler> psykose, thanks man. Works as it should now ;)
2021-12-02 17:56:09 <Jenkler> does anyone have issues with the lxc 4.0.10 package here?
2021-12-02 17:56:22 <Jenkler> cgfsng - cgroups/cgfsng.c:__cgroup_tree_create:771 - File exists - Creating the final cgroup 13(lxc.monitor.ssh) failed
2021-12-02 18:02:30 <Jenkler> https://githubmemory.com/repo/lxc/lxc/issues/3964 ;) 
2021-12-02 18:35:55 <Nulo> Alpine is not applying the mount options I set in fstab to my root and home, any clue?
2021-12-02 18:36:13 <ikke> What mount options?
2021-12-02 18:37:59 <psykose> post your fstab and whether you have localmount enabled
2021-12-02 18:39:01 <psykose> though that gets started automatically i think anyway
2021-12-02 19:03:21 <wrongdong> wee woo, wee woo.
2021-12-02 19:03:36 <wrongdong> tried to chroot into my system which isn't booting because i use efistub
2021-12-02 19:03:41 <wrongdong> somehow efibootmgr wasn't around
2021-12-02 19:03:42 <wrongdong> so i install that
2021-12-02 19:03:47 <wrongdong> then i upgrade -U
2021-12-02 19:03:59 <wrongdong> initramfs: creating /boot/initramfs-lts
2021-12-02 19:04:18 <wrongdong> cpio: can't create symlink 'lib/libblkid.so.1' to 'libblkid.so.1.1.0': Operation not permitted
2021-12-02 19:04:20 <wrongdong> no shit it's not permitted
2021-12-02 19:04:30 <wrongdong> MY BOOT DIR IS VFAT BECAUSE THAT'S HOW EFI WORKS
2021-12-02 19:04:41 <wrongdong> so what do we do, local men?
2021-12-02 19:06:30 <wrongdong> apparently this don't register as an error to apk
2021-12-02 19:06:31 <wrongdong> wtf
2021-12-02 19:06:45 <wrongdong> what the fuck is going on
2021-12-02 19:07:35 <wrongdong> why is efivarfs not in my kernel
2021-12-02 19:07:37 <wrongdong> WHO DID THIS
2021-12-02 19:08:03 <helby> if you chroot you have to mount more crap than just / and boot
2021-12-02 19:08:11 <wrongdong> i did though
2021-12-02 19:08:20 <wrongdong> but also now it's looking in fuckin
2021-12-02 19:08:24 <wrongdong> i'm gay, literally just big gay
2021-12-02 19:08:38 <wrongdong> do i have to boot live media of the SAME kernel version
2021-12-02 19:08:39 <wrongdong> ?
2021-12-02 19:09:32 <wrongdong> it appears that i've booted 5.15.4 and edge is on 5.15.6
2021-12-02 19:09:42 <wrongdong> so if i try to modprobe, it's like "haha it's not in there stupid"
2021-12-02 19:10:21 <wrongdong> presumably i need to modprobe efivarfs before the chroot, then
2021-12-02 19:10:58 <wrongdong> aha.
2021-12-02 19:11:02 <wrongdong> sneaky
2021-12-02 19:15:08 <wrongdong> apparently efivars aren't carried over to the chroot either despite mounting /sys there
2021-12-02 19:23:35 <wrongdong> wtf my efivars don't have my old guy in there
2021-12-02 19:23:43 <wrongdong> mad
2021-12-02 19:24:33 <wrongdong> HELLA MAD
2021-12-02 19:26:16 <wrongdong> something tells me that this is partially my fault for using ventoy to boot instead of a raw uefi drive
2021-12-02 19:26:18 <wrongdong> let's see if it really is
2021-12-02 19:43:37 <wrongdong> nope
2021-12-02 19:43:46 <wrongdong> linux is mad and i assume something duplicated a variable somewhere
2021-12-02 19:44:01 <wrongdong> but all i need to do is read the motherfucking variable to know what to replace it with
2021-12-02 19:46:07 <hoel> Leave that variable’s sexual life alone.
2021-12-02 20:06:09 <minimal> wrongdong: "MY BOOT DIR IS VFAT BECAUSE THAT'S HOW EFI WORKS". Yes EFI requires its ESP partition to have a FAT filesystem but that is not necessarily /boot. General speaking many/most distros mount the ESP filesstem as /boot/EFI (i.e. /boot is ext4/xfs/btrfs/etc and /boot/EFI is FAT)
2021-12-02 20:06:54 <donoban> lol, my sway failure seems "fixed" enablig suid to it
2021-12-02 20:07:26 <donoban> I think that last last upgrade already broke it and i just reenabled it but I forgot about that 
2021-12-02 20:07:30 <wrongdong> what the hell would that even do, minimal
2021-12-02 20:07:54 <wrongdong> it's just files; why aren't they just copyin the goddamned files in there
2021-12-02 20:08:25 <wrongdong> i might just have to reread mkinitfs
2021-12-02 20:08:36 <wrongdong> for some reason i can see the boot option in my bios, but in linux it's not there
2021-12-02 20:08:48 <wrongdong> and i dunno in my bios how to read the MUFUCKEN VAR
2021-12-02 20:09:25 <ikke> wrongdong: please mind your language
2021-12-02 20:10:42 <wrongdong> how is it 2021 and people are still puritanical fuckboys
2021-12-02 20:10:56 <wrongdong> "mind your language," or fucking what? you gonna smite me, big skydaddy?
2021-12-02 20:11:07 <wrongdong> as if i got any fucking information on fixing the install from here anyway
2021-12-02 20:11:24 <wrongdong> if you're gonna treat me like a baby, i'm gonna treat you like the condescending prick you're being
2021-12-02 20:11:29 <wrongdong> mind this dick why don't you
2021-12-02 20:11:36 <psykose> i know exactly what your issue is probably, but you act like a child so i had no interest in saying anything
2021-12-02 20:11:38 <psykose> try being polite sometime
2021-12-02 20:11:45 <ikke> We just want this to be a welcoming respectful place, which you are not contributing to at the momenbt
2021-12-02 20:11:53 <psykose> donoban: do you have seatd set up properly?
2021-12-02 20:12:22 <wrongdong> me sitting around for forty five minutes isn't welcoming; maybe if you stopped kicking all the fucks who swore you wouldn't have such an empty channel
2021-12-02 20:12:34 <wrongdong> psykose: my issue is multivariate
2021-12-02 20:13:21 <wrongdong> the real issue was obviously me trying to fix a fucked system by just blindly upgrading it and it isn't doing even that because whatever mkinitfs is triggering is all sorts of fucked
2021-12-02 20:13:45 <wrongdong> and i assume my dickbag closed source bios has something to do with being mad that there are multiple efivars with the same name, despite those efivars not even showing up in the bios
2021-12-02 20:43:20 <invoked> jeffreybenzos i presume
2021-12-02 20:56:57 <ncopa> wrongdong: can you please follow the code of conduct, or leave. https://www.alpinelinux.org/community/code-of-conduct.html
2021-12-02 21:01:53 <Apachez> ncopa: dick is short for richard in englishspeaking countries
2021-12-02 21:02:59 <jn> only if used in a context where a name is clearly expected
2021-12-02 21:03:24 <wrongdong> i'm not violating shit
2021-12-02 21:03:39 <wrongdong> there are three tenets that describe behavior "in violation"
2021-12-02 21:03:48 <wrongdong> everything else is a "guideline" to "avoid"
2021-12-02 21:04:06 <wrongdong> how about you not make this personal, ncopa
2021-12-02 21:04:17 <wrongdong> and also not take it personally when i say fuck you
2021-12-02 21:04:35 <wrongdong> i found the efi shell by the way
2021-12-02 21:04:39 <wrongdong> and now it's a too-low number
2021-12-02 21:04:56 <hoel> wrongdong: I guess that “explicitly disallowed conduct” implies there is a “implicitly disallowed conduct”
2021-12-02 21:04:57 <wrongdong> good thing i joined two hours ago so i could enjoy having to cut this short because i've got real life to attend to
2021-12-02 21:05:11 <wrongdong> it doesn't, hoel, and you saying that sounds like you're a fuckboy from twitter
2021-12-02 21:05:29 <hoel> Haha, I’ve never used twitter in my life, but that’s not relevant. ^^
2021-12-02 21:05:32 <wrongdong> perhaps stop reading algorithmic dogshit smeared on your phone screen and start thinking with that noggin you almost killed your mother with
2021-12-02 21:05:50 <wrongdong> it entirely is, as that means that the internet's asshole is staining way more bowls than it ever had a right to
2021-12-02 21:06:07 <wrongdong> i recommend investigating your friendgroup
2021-12-02 21:06:16 <hoel> ô.ŏ
2021-12-02 21:06:32 <hoel> No need to get agressive.
2021-12-02 21:09:13 <ikke> wrongdong: this is your final warning. We expect you to be respectful to users (and calling people 'fuckboy' is not that
2021-12-02 21:09:24 <wrongdong> there's always a need, caused by the utter lack of anyone's personal inclination to match emotion in any conversation ever
2021-12-02 21:09:32 <wrongdong> you are all so devoid of empathy and justify it with passive aggression
2021-12-02 21:09:44 <wrongdong> respect my motherfucking dicknballs all over your drumset ikke
2021-12-02 21:09:53 <wrongdong> i'm doing you a service by outing the fuckboys
2021-12-02 21:10:01 <wrongdong> just because no one said it doesn't mean it didn't happen
2021-12-02 21:10:32 <jn> what a load of nonsense
2021-12-02 21:13:57 <Apachez> jn: you say you are not familiar with a richardbag? ;)
2021-12-02 21:14:27 <jn> ;D
2021-12-02 21:20:49 <Saijin_Naib[m]> Does anyone know of a way to do a full system repair/re-run of triggers under Alpine, sort of like the Repairing Packages/Re-run System-Wide Configuration Triggers that Solus Linux has?
2021-12-02 21:20:49 <Saijin_Naib[m]> https://getsol.us/articles/troubleshooting/boot-rescue/en/
2021-12-02 21:21:16 <Saijin_Naib[m]> My laptop bricked testing the latest kernels and all my fonts/icons are broken, despite apk fix of what I believe should cover them
2021-12-02 21:21:22 <lopid> `apk fix \*` might do it
2021-12-02 21:21:40 <Saijin_Naib[m]> Beautiful, let me try that. Thank you
2021-12-02 21:22:00 <Saijin_Naib[m]> I was kind of hoping there was an apk fix world target or something
2021-12-02 21:22:17 <lopid> i think that is :)
2021-12-02 21:22:33 <helby> you can apk fix * or apk fix package package ...
2021-12-02 21:22:40 <helby> * will reinstall all crap
2021-12-02 21:24:41 <lopid> hopefully there is no crap
2021-12-02 21:24:49 <Saijin_Naib[m]> lopid: It's running now, thank you. If this fails, is there an inplace upgrade mechanism, or should I just nuke/clean install?
2021-12-02 21:25:13 <Saijin_Naib[m]> The problem follows new profiles as well, so I think it must be system-wide
2021-12-02 21:25:24 <Saijin_Naib[m]> The lack of crap is a key draw to Alpine 🙂
2021-12-02 21:26:20 <psykose> inplace upgrade?
2021-12-02 21:27:02 <lopid> i don't know. i think it depends what's actually gone wrong
2021-12-02 21:29:18 <Saijin_Naib[m]> Inplace is more of a Windows Vista+ concept, but basically install the OS back over itself. More brute-force way to do DISM/SFC type work that also resets all file/folder permissions and re-runs any configuration/registry bits as well
2021-12-02 21:29:56 <Saijin_Naib[m]> I'm at a loss for what broke here, but I do see APK complaining about not being able to move/rename/create icons in usr/share/icons which sounds about right if everything is missing
2021-12-02 21:30:00 <lopid> maybe apk fix does most of that, but it shouldn't replace any customisations you've made
2021-12-02 21:30:20 <psykose> can you post the error
2021-12-02 21:30:27 <Saijin_Naib[m]> Yeah, APK fix does seem to be doing basically all that, but the permissions errors keep coming.
2021-12-02 21:30:58 <Saijin_Naib[m]> ERROR: failed to create usr/share/icons/hicolor/.... in numerous variants as each package tries to place their icons there
2021-12-02 21:31:00 <Saijin_Naib[m]> Busted permissions?
2021-12-02 21:31:08 <lopid> maybe check the destination
2021-12-02 21:31:25 <psykose> post ls -la of /usr/share/icons
2021-12-02 21:31:26 <ikke> --directory-permissions
2021-12-02 21:31:28 <psykose> or a tree of the whole thing
2021-12-02 21:31:36 <psykose> you probably have something very wrong :)
2021-12-02 21:32:11 <Saijin_Naib[m]> usr/share/icons seems to be drwxr-xr-x root like everything else in usr/share, so that's good at least
2021-12-02 21:32:34 <lopid> can you manually create a file there?
2021-12-02 21:32:48 <Saijin_Naib[m]> all the sub directories (Adwaita, ContrastHigh, HighContrast, hicolor, mate, meta) are all also drwxr-xr-x root
2021-12-02 21:33:04 <Saijin_Naib[m]> As user (wheel) or root?
2021-12-02 21:33:49 <lopid> root initially
2021-12-02 21:34:15 <Saijin_Naib[m]> yeah, sudo touch testfile works fine
2021-12-02 21:34:19 <lopid> since it seems you've shown no reason for the failure
2021-12-02 21:34:21 <psykose> anyway you can run apk fix * --directory-permissions as ikke says and that might fix it
2021-12-02 21:34:44 <lopid> \*
2021-12-02 21:34:44 <Saijin_Naib[m]> Oh I didn't know that flag. That'll help!
2021-12-02 21:37:02 <Saijin_Naib[m]> re-trying with that
2021-12-02 21:37:06 <Saijin_Naib[m]> Fonts are back at least
2021-12-02 21:37:10 <Saijin_Naib[m]> And some icons, so that's a huge help
2021-12-02 21:37:26 <Saijin_Naib[m]> Thanks, all
2021-12-02 21:38:24 <ikke> nice
2021-12-02 21:45:59 <Saijin_Naib[m]> Wow, sweet. Almost all good! Thanks so much!
2021-12-02 23:21:21 <Hello71> "in-place reinstall" is mostly a windows concept because of the sheer quantity of crap shoved in the registry and whatnot
2021-12-02 23:22:30 <Hello71> on linux the crap is mostly contained to $HOME and /var and to a lesser extent /etc, and it's a small enough quantity that it's feasible to manually scan through
2021-12-02 23:24:05 <helby> crap is everywhere! ;
2021-12-02 23:24:32 <mtm84[m]> There’s going to be a group of people who don’t want to deal with the hassle vs starting over clean, even among the linux crowd.  Of course that’s probably more a “nuke it from orbit” thing.
2021-12-02 23:25:29 <Hello71> true but at least with linux it's entirely fixable if you have some knowledge and put in a little bit of time
2021-12-02 23:26:13 <Naib> besides fs corruption and/or package db ... its quite fixable
2021-12-02 23:26:24 <Hello71> whereas with windows it's much more likely (although obviously not all the time) that there's some wrong value in millions of registry entries accessed by one of hundreds of processes
2021-12-02 23:26:44 <Hello71> and it's basically not possible to fix manually unless you have in-depth knowledge of the specific component
2021-12-02 23:29:54 <mtm84[m]> Yeah agree.  It’s way easier to, say, boot into a limited shell in linux and tinker with configs than in windows.
2021-12-02 23:30:22 <mtm84[m]> I’ve fixed many a misconfigured kernel that way
2021-12-02 23:30:38 <helby> alpine installer and apk is so fast, that with backup configs it's not so time consuming do even fresh reinstall
2021-12-02 23:32:27 <mtm84[m]> Yeah also true.  It’s  a lot easier to backup a home folder on a usb stick then to go through some random backup utility on windows.
2021-12-02 23:36:23 <helby> I use lbu and put that crap on usb
2021-12-02 23:36:39 <helby> bc. some /etc/crap is edited too
2021-12-02 23:37:54 <helby> and lbu is simple and fast, included so
2021-12-02 23:39:06 <mtm84[m]> Right.  I mostly use alpine in purpose run VMs so that’s not really an issue for me, but as a whole linux has a lot more options for “fixing” things easily.  Depending on how far you want to go. Verses windows were things might work ok a lot of time automatically but if something fails it will probably fail in a way that is harder or near impossible to correct. (Subjective of course)
2021-12-02 23:39:48 <helby> manually fix something in win .. pfuf
2021-12-02 23:40:12 <helby> you just hit a button and if it does not fix auto, it's over ;/
2021-12-02 23:40:48 <mtm84[m]> You would think.
2021-12-02 23:41:11 <mtm84[m]> I’ve spent way to long in windows repair rabbit holes
2021-12-02 23:44:33 <mtm84[m]> I had a windows update that would fail every time it tried to apply a certain update.  No amount of applying the common manual fixes would remove its desire to try and fail on that update.  Almost any decent linux distribution package manager could have fixed it or black listed it in a matter of minutes.
2021-12-02 23:49:12 <psykose> i've had that happen at least 5 separate times
2021-12-02 23:53:00 <helby> yes, that happens to me too
2021-12-02 23:53:54 <helby> also I have no idea what it can do during update, as it takes so long
2021-12-02 23:54:36 <helby> probably sending telemetry ;/
2021-12-02 23:55:20 <mtm84[m]> Twiddling it’s thumbs.  90 percent of the time it doesn’t even tell you any meaningful percentage completion.
2021-12-02 23:56:00 <helby> bc. it does not yet, how much data it will collect ;/
2021-12-03 00:05:07 <mtm84[m]> None because I block all that at my network router lol
2021-12-03 00:27:26 <hiro_> hahahahaha you guys have finally added the rc.local convention - but i already had a rc.local symlink to /etc/local.d/bla.start...
2021-12-03 00:28:12 <hiro_> and incidentally nothing works. guess how stupid i felt when i found out that everything was run twice????? 
2021-12-03 00:28:46 <hiro_> like what FUCKING KIND OF ERROR is that to debug, would anybody here even have guessed? :O
2021-12-03 00:29:03 <mtm84[m]> Whooops
2021-12-03 00:29:07 <helby> ;/
2021-12-03 00:29:53 <hiro_> it took me forever to find out that everything is run twice :D
2021-12-03 00:30:06 <hiro_> but you know my first symtpom?
2021-12-03 00:30:09 <helby> where did yo ufind rc.local or what's that?
2021-12-03 00:30:09 <hiro_> *symptom
2021-12-03 00:30:16 <hiro_> i run `pon` inside there
2021-12-03 00:30:24 <hiro_> so... NO INNANET
2021-12-03 00:30:33 <hiro_> but then i check and i see ppp1 device
2021-12-03 00:30:40 <hiro_> and i'm like, why the fuck didn't it pick ppp0
2021-12-03 00:30:44 <psykose> pon pon
2021-12-03 00:31:00 <hiro_> and you can only have one ppp connection with the provider
2021-12-03 00:31:15 <psykose> you should make that an actual service
2021-12-03 00:31:16 <hiro_> so actually it kept on toggling from ppp0 to ppp1
2021-12-03 00:31:26 <hiro_> so every few seconds a packet would get through!!!
2021-12-03 00:31:47 <hiro_> pon;pon-as-a-service
2021-12-03 00:32:03 <hiro_> i was like MULTI-HOMING YAAAAHHHH
2021-12-03 00:32:10 <hiro_> the FUTURE
2021-12-03 00:32:23 <hiro_> fucking absurd :D
2021-12-03 00:33:16 <hiro_> helby: in /etc
2021-12-03 00:33:33 <hiro_> it is run by /etc/local.d/something if it exists
2021-12-03 00:33:49 <hiro_> cat /etc/local.d/rc.local-compat.start 
2021-12-03 00:34:06 <hiro_> mega-troll
2021-12-03 00:34:37 <mtm84[m]> May I direct you all to xkcd 1172
2021-12-03 00:57:52 <hiro_> mtm84[m]: nah this was just for laughs, i'm not advocating for removal haha
2021-12-03 00:58:14 <mtm84[m]> Lol
2021-12-03 00:58:43 <mtm84[m]> Its more commentary on you never know what a persons use case might be, for any number of reasons
2021-12-03 01:03:45 <hiro_> so yeah. well played and good night.
2021-12-03 01:07:57 <psykose> sleep well
2021-12-03 01:17:41 <Saijin_Naib[m]> Hello71: had to remove 17 packages (and dependencies) to finally fully fix the errors, but it seems to be okay now. Still really confused how a hard lock caused all this
2021-12-03 01:18:01 <Hello71> are you using btrfs? :p
2021-12-03 01:18:26 <Saijin_Naib[m]> And for the windows repair thing, it's a more familiar concept for me since I've been working with OS/2 and windows personally and professionally for like, 20+ years so second nature
2021-12-03 01:18:48 <Saijin_Naib[m]> No! ext4 over LVM
2021-12-03 01:19:22 <Saijin_Naib[m]> Been stable prior. I also have tune2fs setup to check every week or 10 boots so the volume should have been consistent before the hang 
2021-12-03 01:24:00 <Saijin_Naib[m]> What does installing a package do that APK fix with xattr and repair permissions not do? Because APK fix \* strangely didn't resolve the errors. I don't understand the mechanics of what happened 😔
2021-12-03 01:38:35 <Hello71> perhaps there was some memory corruption and it wrote some garbage to your filesystem
2021-12-03 01:39:33 <Saijin_Naib[m]> Would fsck not catch and fix? 
2021-12-03 01:41:19 <Hello71> theoretically
2021-12-03 02:55:41 <execrate> Yo
2021-12-03 08:07:42 <dhruvin> hello #alpine-linux, i'm new to alpine packaging. i have packaged several software for guix in past. in guix, one must add all the dependencies to guix first, and then package the project in question. i saw some packages don't do that in alpine (in aports repository). do we just use pypi/npm/hackage directly?
2021-12-03 08:10:10 <lopid> are we to assume your package relies on something provided by pypi/npm/hackage?
2021-12-03 08:10:31 <dhruvin> yes, the package I'd like to contribute is hledger
2021-12-03 08:11:23 <dhruvin> i see pandoc for reference. it just uses hackage (via cabal install --only-dependencies)
2021-12-03 08:18:53 <dhruvin> (i read the APKBUILD_examples:Haskell wiki page before asking the question, the page suggests something similar too)
2021-12-03 08:26:23 <dhruvin> nevermind, i read APKBUILDs of several more packages, and they do the same as well. thanks!
2021-12-03 08:45:38 <ikke> dhruvin: note that these are exceptions for these particular languages / ecosystems (go, rust, haskell) 
2021-12-03 08:45:59 <ikke> And we do expect that dependencies are pinned
2021-12-03 09:37:21 <dhruvin> ikke: will keep that in mind.
2021-12-03 10:15:47 <Tackyone> Hi all - anyone here have any experience on getting OpenVAS to run under Alpine? - I've followed the "Setting up GVM11" Wiki guide, but run into a number of issues :(
2021-12-03 11:20:08 <Justin[m]1234> Is there a site where I can see what's in the build queue? I know the status at https://build.alpinelinux.org/ but I want to see what's coming up next and where a certain package is in the queue.
2021-12-03 11:26:51 <ikke> Justin[m]1234: We don't have that information public. It's something that's determined per builder, and can be kind of arbitrary (bar dependencies)
2021-12-03 11:27:16 <Justin[m]1234> Ahh OK thanks.
2021-12-03 11:27:33 <Justin[m]1234> Just that poppler-qt5 is currently broken looking for old libpoppler .so 
2021-12-03 11:27:51 <Justin[m]1234> No rush. I'll check in 12h or so.
2021-12-03 11:28:20 <ncopa> git log may give hints on whats in the queue
2021-12-03 11:42:30 <donoban> psykose: probably I don't, I just installed it, added it to default runlevel and my user to seat group
2021-12-03 14:18:35 <psykose> donoban: sure, that's all you need- just making sure
2021-12-03 15:53:13 <donoban> well, I suppose that I have something wrong
2021-12-03 15:53:34 <donoban> and I noticed something when upgrading from my old snapshot
2021-12-03 15:53:50 <donoban> [donoban@localhost][~]% apk list -u                                                                                                                                                                                                                                                                          [git][donoban/.][async-upgradeU]
2021-12-03 15:53:52 <donoban> libcrypto3-3.0.0-r4 x86_64 {openssl3} (Apache-2.0) [upgradable from: libcrypto3-3.0.0-r3]
2021-12-03 15:53:54 <donoban> libssl3-3.0.0-r4 x86_64 {openssl3} (Apache-2.0) [upgradable from: libssl3-3.0.0-r3]
2021-12-03 15:54:18 <donoban> this packages don't upgrade with apk upgrade (using --available)
2021-12-03 16:21:05 <xordspar0> dhruvin: Cool, I will definitely use your hledger package when it's ready. :D
2021-12-03 18:18:40 <frag> trying to upgrade system getting libffi errors https://x0.at/9snT.txt
2021-12-03 18:25:40 <psykose> what is you /etc/apk/world
2021-12-03 18:26:50 <psykose> it looks like something wants explicitly libffi-dev-3.3 and that breaks everything else
2021-12-03 18:28:23 <frag> psykose: https://x0.at/0RZR.txt
2021-12-03 18:29:33 <psykose> try remove some of the -dev packages
2021-12-03 18:37:57 <frag> hmm one dev package purged 25 packages :S are they usually ok to purge? https://x0.at/9Uw2.txt
2021-12-03 18:38:13 <psykose> yes
2021-12-03 18:38:44 <psykose> unless you want util-linux, in which case add it by itself
2021-12-03 18:39:08 <psykose> also like 3 things in your world don't exist on stable
2021-12-03 18:39:10 <frag> ok, nice :]
2021-12-03 18:39:26 <frag> ledger and?
2021-12-03 18:39:46 <psykose> catdoc epub2txt
2021-12-03 18:43:32 <frag> epub2txt very nice/small code
2021-12-03 18:43:47 <frag> psykose: i guess those dont cause the problem? 
2021-12-03 18:43:54 <frag> doesnt seem to depends on the error package
2021-12-03 18:44:05 <psykose> no, just mentioning
2021-12-03 18:44:36 <frag> okok :]
2021-12-03 18:44:47 <psykose> once you fix the dep issue, then it will fail on those instead
2021-12-03 18:45:05 <psykose> but it's unrelated
2021-12-03 18:45:12 <frag> hehe, damn
2021-12-03 18:45:28 <psykose> if you want to just go fast, delete every -dev line and those 3 and update
2021-12-03 18:45:48 <frag> delete=apk del?
2021-12-03 18:46:10 <psykose> you can just delete it in /etc/apk/world
2021-12-03 18:46:21 <frag> oh rly?  hm
2021-12-03 18:46:38 <psykose> apk add/del is just a frontend for modifying the world file
2021-12-03 18:46:53 <psykose> and then running a fix after, pretty much
2021-12-03 18:47:08 <frag> psykose: so if i remove lines from world, it will be deleted when i upgrade?
2021-12-03 18:47:12 <psykose> yes
2021-12-03 18:47:17 <frag> ah, nice :]
2021-12-03 18:47:32 <psykose> i mean don't let me confuse you, you can also just use del; just telling you the possibilities
2021-12-03 18:47:51 <frag> yes, i didnt know, very nice
2021-12-03 18:50:06 <frag> hmm, had to remove the 3 extra packages before that libffi errors disappeared
2021-12-03 18:51:49 <psykose> probably because they are installed and have dependencies, and you are trying to downgrade to stable where they don't exist, but they also do have the dep info on ffi somewhere
2021-12-03 18:51:54 <psykose> but yeah, no issues here
2021-12-03 18:52:35 <Guest7482> I've been searching how to increase the 'Filesystem' /tmpfs 'Mounted on' / above 50% of RAM, where can I change the setting? In /etc/fstab and use ',size=6G' ? The rpi4 I'm using has 8GB RAM now I only use 3.8G Size ...
2021-12-03 18:53:19 <psykose> yes, size=75% or 6G or something
2021-12-03 18:54:12 <psykose> the default is 50%, and the rpi4b has 7.62GiB, so you see 3.8
2021-12-03 18:54:44 <Guest7482> psykose: Currently there is no entry in my /etc/fstab for /tmpfs, but that is the right place to define it?
2021-12-03 18:55:16 <psykose> sure, `tmpfs /tmp tmpfs defaults,mode=1777,size=75% 0 0`
2021-12-03 19:00:12 <Guest7482> psykose: Thank you, just wanted to be sure to tweak the setting in the right place (although tmpfs / tmpfs defaults,mode=1777,size=75% 0 0) did the trick... :)
2021-12-03 19:00:34 <psykose> :)
2021-12-03 19:19:21 <helby> trying to setup wifi access point on laptop, everything looks okay, just phone can't get ip, it just 'obtaining IP address' any idea?
2021-12-03 19:19:55 <psykose> did you set up a dhcp server
2021-12-03 19:20:00 <helby> no
2021-12-03 19:20:05 <psykose> well how would it get an ip :)
2021-12-03 19:20:21 <helby> how laptop get one
2021-12-03 19:20:34 <psykose> with dhcp or static configuration
2021-12-03 19:20:54 <helby> so I have to install some crap or use udhcpc?
2021-12-03 19:21:08 <psykose> udhcpd can be a dhcp server sure
2021-12-03 19:21:17 <psykose> or use dnsmasq
2021-12-03 19:21:21 <helby> but funny few days ago I get it working, but probably from wlan not eth0
2021-12-03 19:21:30 <psykose> or, if you have a router already, it probably does dhcp
2021-12-03 19:21:33 <helby> without setting anything with dhpc
2021-12-03 19:22:10 <helby> well, this is network in work, so no idea how that shit work, so mobile using access point does not use same crap like laptop?
2021-12-03 19:22:45 <helby> should I change interface udhcp will use from eth0 to br0?
2021-12-03 19:23:46 <psykose> if you connect your phone to the same wifi your laptop connects to it would indeed work the same
2021-12-03 19:24:20 <psykose> but you are making your own access point, so you need to either run your own dhcp on your own subnet, or forward dhcp requests
2021-12-03 19:24:24 <helby> but bridge is from eth0 on laptop 
2021-12-03 19:24:30 <psykose> given that it's work you should probably find out 'how that shit work'
2021-12-03 19:25:26 <helby> but I did it with access point without tweaking udhpc or whatever related to dhcp
2021-12-03 19:25:35 <psykose> easiest way is to also start dnsmasq and set dhcp-range=ip-of-existing-dhcp,proxy
2021-12-03 19:25:39 <helby> maybe my router did that somehow
2021-12-03 19:45:58 <helby> there is no busybox udhcpd?
2021-12-03 19:50:17 <helby> why there is then udhcpd.conf
2021-12-03 19:57:12 <minimal> helby: install busybox-extras package
2021-12-03 19:59:58 <lopid> how to see what that provides? i see only only /bin/busybox-extras, and no indication of what extras is
2021-12-03 20:10:50 <xordspar0> It's tricky because they're all symlinks created in post install. You can sort of see by looking at the compile-time options: https://git.alpinelinux.org/aports/tree/main/busybox/busyboxconfig-extras
2021-12-03 20:11:27 <minimal> lopid: yeah, line 970 of that file
2021-12-03 20:11:48 <lopid> looks like alpine specific, which explains why there's nothing on the busybox site
2021-12-03 20:12:04 <minimal> whereas the busyboxconfig file has CONFIG_UDHCPD unset
2021-12-03 20:12:53 <minimal> lopid: as it says at the top of those files, "# Automatically generated make config: don't edit""
2021-12-03 20:13:09 <lopid> i won't edit
2021-12-03 20:14:03 <lopid> i'm wondering if i've installed a 'real' version of something when the busybox-extras package version might have sufficed
2021-12-03 20:14:15 <minimal> I meant its not Alpine specific, if you download Busybox source and use "make config" to configure it then it also creates a config file locally with your settings
2021-12-03 20:14:30 <lopid> oh
2021-12-03 20:15:15 <minimal> Alpine is just using a pre-prepared confogi file to automate the configuration, same way with Linux kernel you can copy across a .config file rather than going through "make menuconfig" each time
2021-12-03 20:15:30 <minimal> s/confogi/config/
2021-12-03 20:15:30 <alpine-meetbot> minimal meant to say: Alpine is just using a pre-prepared config file to automate the configuration, same way with Linux kernel you can copy across a .config file rather than going through "make menuconfig" each time
2021-12-03 20:19:07 <lopid> 
ACTION nods

2021-12-03 20:19:27 <lopid> still would be nice to see what it provided in a more appetising format
2021-12-03 20:20:08 <xordspar0> <busyboxconfig-extras grep 'CONFIG_.*=y' | grep -Ev '(^#|FEATURE)'
2021-12-03 20:20:58 <xordspar0> https://tpaste.us/xnlw
2021-12-03 20:21:37 <lopid> it provides pie? i'm sold
2021-12-03 20:21:47 <xordspar0> :D
2021-12-03 20:21:56 <xordspar0> It's still kind of a mess but it's at least a shorter list to sift through.
2021-12-03 20:22:14 <lopid> and you have to know to do that
2021-12-03 20:23:49 <aky> Hi
2021-12-03 20:23:52 <psykose> hi
2021-12-03 20:24:15 <aky> I'm having issues starting minidlna server
2021-12-03 20:24:48 <aky> I'm not sure what to do
2021-12-03 20:25:05 <lopid> start with describing the problem
2021-12-03 20:25:13 <aky> This is what i read from the logs: /etc/init.d/minidlna[3095]: start-stop-daemon: no matching processes found
2021-12-03 20:25:41 <minimal> lopid: "busybox --help" and "/bin/busybox-extras --help"
2021-12-03 20:25:47 <aky> I have just installed minidlna after update&&upgrade
2021-12-03 20:25:48 <lopid> it could say that when you try to stop a service that isn't running
2021-12-03 20:26:03 <aky> Actually.. it doesn't start
2021-12-03 20:26:09 <aky> It says "crashed"
2021-12-03 20:26:29 <aky> Also ps aux shows no process related to minidlna
2021-12-03 20:26:31 <lopid> you only see that help after you've installed it
2021-12-03 20:26:44 <aky> it is installed, yes
2021-12-03 20:26:49 <aky> it doesn't start
2021-12-03 20:27:13 <lopid> nothing in its own log, if it has one?
2021-12-03 20:27:19 <aky> # rc-service minidlna restart  * Stopping MiniDLNA ...  * start-stop-daemon: no matching processes found                                                                                                         [ ok ]  * Starting MiniDLNA ...                                                                                                                                  [ ok ]
2021-12-03 20:27:25 <aky> Sorry for the paste
2021-12-03 20:27:29 <aky> It's just 3 lines
2021-12-03 20:27:50 <psykose> can you post the logs from /var/log/minidlna
2021-12-03 20:28:16 <aky> I've had a look at /etc/init.d/minidlna but i'm lost, honestly
2021-12-03 20:28:57 <aky> psykose: there is no such log
2021-12-03 20:29:17 <aky> I suspect that the process doesn't start because there is something wrong with its initscript
2021-12-03 20:29:36 <lopid> are you running it as root?
2021-12-03 20:29:41 <aky> When i manually run: "minidlnad -d" it actually works
2021-12-03 20:29:47 <aky> Yes, as root
2021-12-03 20:31:01 <psykose> did you edit /etc/conf.d/minidlna at all
2021-12-03 20:31:17 <psykose> or /etc/minidlna.conf
2021-12-03 20:31:28 <aky> I am not familiar with openrc init scripts, but.. I suspect it's from there
2021-12-03 20:31:44 <aky> psykose: yep, edited it to point to my share location
2021-12-03 20:32:10 <aky> I mean /etc/minidlna.conf
2021-12-03 20:32:27 <lopid> i think it essentially runs command command_args
2021-12-03 20:32:55 <psykose> what did you change in the file
2021-12-03 20:33:04 <aky> media_dir points to a share location where all have access; a location from where samba successfully shares the files
2021-12-03 20:33:40 <psykose> who owns the folder
2021-12-03 20:34:44 <aky> You mean the share folder? nobody, and it has 777 recursively on all files and folders
2021-12-03 20:35:40 <psykose> are you completely sure there isn't a log under /var/log/minidlna/*
2021-12-03 20:36:02 <aky> Yes, I'm sure
2021-12-03 20:36:16 <aky> I'm very concerned about what logread reports: "/etc/init.d/minidlna[3095]: start-stop-daemon: no matching processes found"
2021-12-03 20:36:46 <aky> What's that "start-stop-daemon" supposed to mean? is it something openrc related?
2021-12-03 20:37:17 <lopid> yes
2021-12-03 20:37:32 <lopid> https://github.com/OpenRC/openrc/blob/master/service-script-guide.md
2021-12-03 20:37:57 <psykose> that just means when it gets stopped it has already crashed
2021-12-03 20:38:03 <lopid> run it manually with the same command and command_args, maybe it outputs something to terminal
2021-12-03 20:38:16 <aky> So.. when I run "minidlnad -d" it works, when I run "rc-service minidlna restart" it doesn't
2021-12-03 20:38:38 <lopid> stop confusing yourself with a restart, concentrate on the start
2021-12-03 20:38:55 <aky> I use "re"-start just to be sure; it's a habbit (if it was started to stop it before starting again)
2021-12-03 20:39:19 <psykose> `minidlnad -f /etc/minidlna.conf --user minidlna --group minidlna -d`
2021-12-03 20:40:37 <aky> Thanks, psykose ; I guess I see the problem now: "Unknown option: --user"
2021-12-03 20:41:02 <aky> # minidlnad -f /etc/minidlna.conf --user minidlna --group minidlna -d [2021/12/03 20:40:01] minidlna.c:964: error: Unknown option: --user [2021/12/03 20:40:01] minidlna.c:831: fatal: Unknown option: minidln
2021-12-03 20:41:19 <lopid> i wonder if they are options to  start-stop-daemon instead…
2021-12-03 20:41:23 <aky> So.. I suppose it needs a "minidlna" user to be created in order to work
2021-12-03 20:41:27 <psykose> lopid: they are
2021-12-03 20:41:35 <psykose> just remove those two
2021-12-03 20:41:38 <psykose> my mistake
2021-12-03 20:41:42 <aky> Am I wrong?
2021-12-03 20:42:00 <psykose> replace --user and --group with -u and -g
2021-12-03 20:42:13 <lopid> but you might want to run it as that user and group, in case running it as root creates some file that can't be removed by that user next time
2021-12-03 20:42:16 <Hello71> probably you have set minidlna to fork
2021-12-03 20:42:35 <minimal> aky: the pre-install file takes care of creating user/group
2021-12-03 20:42:41 <aky> Yep, psykose , replaced with -u -g and it starts
2021-12-03 20:42:57 <aky> Only it is not daemonized
2021-12-03 20:42:57 <psykose> then it runs fine
2021-12-03 20:43:15 <aky> It fills the screen with the files it indexes (movies, music, etc.)
2021-12-03 20:43:22 <psykose> the only thing that can be wrong is that you changed something in the config file other than media_dir
2021-12-03 20:43:41 <aky> My issue is that "rc-service start" doesn't actually start the service
2021-12-03 20:44:13 <aky> psykose: nope, just media_dir; I haven't even touched the "server name"
2021-12-03 20:44:24 <aky> Or anything else
2021-12-03 20:44:42 <psykose> well when i install it starts fine
2021-12-03 20:44:43 <aky> Just wanted to give it a quick start
2021-12-03 20:44:48 <psykose> all that can be different is you changed something
2021-12-03 20:44:59 <aky> Ok, I'm checking again
2021-12-03 20:45:31 <psykose> you can delete /etc/init.d/minidlna /etc/conf.d/minidlna /etc/minidlna.conf and run apk fix minidlna and start it with a fresh config
2021-12-03 20:45:31 <aky> Before changing configs I always make a backup copy of the original
2021-12-03 20:45:43 <aky> Ok, will do
2021-12-03 20:49:43 <aky> minidlna                                                                                                                                          [  crashed  ]
2021-12-03 20:49:50 <aky> It's the same
2021-12-03 20:50:29 <aky> Deleted all the configs and.. fixed/reinstalled the package but rc-service still doesn;t start it properly
2021-12-03 20:53:01 <psykose> post `rc-service -d minidlna start`
2021-12-03 20:55:31 <aky> https://pastebin.com/gNkfDK4y
2021-12-03 20:56:45 <aky> I see this at line 156 (in my paste): "eend 0 'Failed to start MiniDLNA'"
2021-12-03 20:56:56 <aky> But then I don't know where to go with that
2021-12-03 20:59:17 <aky> I've also checked that it created the pidfile
2021-12-03 20:59:19 <aky> It did
2021-12-03 21:00:22 <aky> But the pid in the pidfile doesn't exist in /proc
2021-12-03 21:00:50 <psykose> sorry, no idea
2021-12-03 21:00:52 <aky> # cat /run/minidlna/minidlna.pid  3962
2021-12-03 21:01:11 <aky> # stat /proc/3962 stat: can't stat '/proc/3962': No such file or directory
2021-12-03 21:01:34 <aky> Thanks anyway, psykose 
2021-12-03 21:24:35 <Hello71> aky: what happens if you run start-stop-daemon --start --exec /usr/sbin/minidlnad --pidfile /run/minidlna/minidlna.pid --user minidlna --group minidlna -- -f /etc/minidlna.conf -P /run/minidlna/minidlna.pid
2021-12-03 21:26:06 <aky> minidlna didn't start
2021-12-03 21:27:11 <aky> ps aux for it but there is nothing; rc-status reports that the service has the status "crashed"
2021-12-03 21:29:54 <aky> I even removed minidlna package and its config files, reinstalled it again and left the config untouched (with media_dir pointing to /opt) and it still doesn't start as a service
2021-12-03 21:31:11 <aky> 16 GB of RAM are free, plenty of disk space..
2021-12-03 21:31:36 <aky> I don't understand what happens
2021-12-03 21:33:22 <Hello71> what exactly is the output
2021-12-03 21:33:35 <aky> There was no output
2021-12-03 21:33:44 <Hello71> hm.
2021-12-03 21:33:55 <aky> ualp [~]# start-stop-daemon --start --exec /usr/sbin/minidlnad --pidfile /run/minidlna/minidlna.pid -u minidlna -g minidlna -- -f /etc/minidlna.conf -P /run/minidlna/minidlna.pid ualp [~]# echo $? 0 ualp [~]#
2021-12-03 21:34:14 <aky> And it returns 0
2021-12-03 21:34:27 <Hello71> does /run/minidlna exist and is writable by minidlna?
2021-12-03 21:34:32 <aky> Yep
2021-12-03 21:34:47 <Hello71> post strace -f start-stop-daemon etc
2021-12-03 21:34:56 <Hello71> (install strace first)
2021-12-03 21:35:28 <aky> Oh, so you suspect it might crash? - the binary, I mean
2021-12-03 21:35:49 <Hello71> depends what you mean by crash, but no
2021-12-03 21:35:56 <aky> Ok.. I could do that, but the "minidlnad -d" works each and every time
2021-12-03 21:36:22 <aky> I just don't want to run it manually
2021-12-03 21:43:55 <aky> Hello71: https://pastebin.com/EDT4JGgR
2021-12-03 21:44:41 <aky> Aside from a lot of "EBADF (Bad file descriptor)" I don't see anything suspicious
2021-12-03 21:46:04 <tobtobxx> Hey there, sorry to interrupt. I have a couple of problems with my fresh Alpine install. I'll just ask the first: gnome-terminal fails to start because it seems to expect some kind of locale or language. All of the guides I found online are based on gnu libc, but theroetically I should be able to fix this just with an env variable, right? The question is now which env variable do I have to set to which value...
2021-12-03 21:46:14 <tobtobxx> The error I get is this: Error constructing proxy for org.gnome.Terminal:/org/gnome/Terminal/Factory0: Error calling StartServiceByName for org.gnome.Terminal: Process org.gnome.Terminal exited with status 
2021-12-03 21:46:24 <Hello71> does it work with minidlnad -d -f /etc/minidlna.conf -P /run/minidlna/minidlna.pid?
2021-12-03 21:47:12 <aky> Yes, it works
2021-12-03 21:47:23 <Hello71> hm.
2021-12-03 21:47:41 <Hello71> tobtobxx: what about the status
2021-12-03 21:48:29 <tobtobxx> The status_ 
2021-12-03 21:48:31 <tobtobxx> ?
2021-12-03 21:48:46 <tobtobxx> Ah, it isn't a service, it's a binary
2021-12-03 21:51:07 <tobtobxx> Ah, wait no, now I get your question. I missed a character while copy pasting:
2021-12-03 21:51:33 <tobtobxx> Error constructing proxy for org.gnome.Terminal:/org/gnome/Terminal/Factory0: Error calling StartServiceByName for org.gnome.Terminal: Process org.gnome.Terminal exited with status 8
2021-12-03 21:52:07 <psykose> tobtobxx: what is your $LANG
2021-12-03 21:52:22 <tobtobxx> LANG=C
2021-12-03 21:53:00 <psykose> set it to an actual lang like en_US.UTF-8 then
2021-12-03 21:53:39 <tobtobxx> Just exporting the var?
2021-12-03 21:54:03 <psykose> and install musl-locales for it i think
2021-12-03 21:54:04 <psykose> yes
2021-12-03 21:55:10 <tobtobxx> btw, it's C.UTF-8 in the tty and C in the eshell (emacs). I can't tell what it generally is within the graphical session, because, yk, no terminal.
2021-12-03 21:55:59 <aky> Hello71: I'm still thinking that it could be something fishy with the rc file for the service (minidlna)
2021-12-03 21:56:17 <aky> But I don't know much about openrc
2021-12-03 21:56:25 <aky> So, I'm in the dark here
2021-12-03 21:57:40 <jhjacobs81> does alpine run mysql or anything? :O
2021-12-03 21:58:38 <jhjacobs81> ohh it does :)
2021-12-03 21:58:41 <jhjacobs81> thanfully :)
2021-12-03 21:58:44 <psykose> you can install mariadb
2021-12-03 21:58:53 <psykose> no idea why you would want to torture yourself with it, but it's there
2021-12-03 21:58:57 <psykose> :)
2021-12-03 21:59:13 <jhjacobs81> well... me neither, but unfortunatly we arent always free to choose
2021-12-03 21:59:17 <psykose> of course
2021-12-03 21:59:29 <jhjacobs81> i prefer postgresql
2021-12-03 22:10:21 <tobtobxx> psykose: That did work kindof... I installed musl-locales and set LANG=en_US.UTF-8 but turns out gnome overwrites all that with C again. However there's the language setting in the gnome settings, which I can only set to British English. I did that and it resolved my issue, though I'm now stuck with en_GB. en_US or en_EN or just en would've been nice. Is that a limitation of musl?
2021-12-03 22:11:15 <psykose> https://gitlab.gnome.org/GNOME/gnome-session/-/issues/80
2021-12-03 22:37:08 <jhjacobs81> *sigh* 
2021-12-03 22:37:31 <psykose> what's wrong
2021-12-03 22:37:34 <jhjacobs81> i have a program in /opt/program/program
2021-12-03 22:37:47 <jhjacobs81> when i do ./program ash complaints that the file cant be found
2021-12-03 22:37:50 <jhjacobs81> and rc does the same
2021-12-03 22:38:01 <jhjacobs81> yet i KNOW it is there.. i chmod +x'd it
2021-12-03 22:38:34 <jhjacobs81> its green on the commandline as well
2021-12-03 22:39:02 <jhjacobs81> its a golang program i wonder if maybe i need to build it from source?
2021-12-03 22:39:25 <jhjacobs81> i downloaded the default 64 bit program..
2021-12-03 22:40:25 <jhjacobs81> maybe i need the 32bits version?
2021-12-03 22:43:39 <xordspar0> jhjacobs81: I bet you it links to glibc. The file that it can't find is /lib/ld-linux.so. It's a confusing error message.
2021-12-03 22:45:12 <jhjacobs81> hmz
2021-12-03 22:45:50 <xordspar0> How you could have found that out: `strace ./program`, or `file program`. If you suspect it's a linking issue you might want to try `ldd program`, but that won't work either because ldd uses the loader configured in the binary.
2021-12-03 22:46:10 <jhjacobs81> ohh.. lemme try
2021-12-03 22:47:04 <xordspar0> Go binaries are *mostly* static linked, but by default they link to libc for DNS lookups. You can either compile on Alpine, or you can explicitly make it 100% statically linked.
2021-12-03 22:47:47 <jhjacobs81> well screw me.. but i get: -ash: strace: not found
2021-12-03 22:47:55 <jhjacobs81> and -ash: file: not found
2021-12-03 22:47:58 <helby> holy crap
2021-12-03 22:48:17 <xordspar0> Yeah, those are not installed by default. :) `apk add strace file`
2021-12-03 22:48:25 <xordspar0> Are you in Docker?
2021-12-03 22:48:28 <helby> installed unudhpcd but run it foreground, so it blocked boot and was not able to stop it or do whatever with machine ;/
2021-12-03 22:48:42 <jhjacobs81> ofcourse not.. i installed Alpine in a VM :)
2021-12-03 22:48:57 <helby> lucky that someone with usb was around
2021-12-03 22:49:26 <xordspar0> Ok, this happens a lot to Go + Docker users. They build on their host machine and then copy it into an Alpine container.
2021-12-03 22:49:58 <jhjacobs81> well.. im gonna try to install go and build it from source :)
2021-12-03 22:50:21 <helby> at least wifi access point now works
2021-12-03 22:51:31 <xordspar0> Good luck. It's usually pretty easy to compile Go programs from source: go install github.com/my/program@latest
2021-12-03 22:52:22 <jhjacobs81> i need to get go going first haha (pun..)
2021-12-03 22:52:54 <jhjacobs81> oh
2021-12-03 22:52:59 <jhjacobs81> thats rather easy i see :)
2021-12-03 22:56:43 <psykose> you can install patchelf and do patchelf --set-interpreter to the musl libc, or remove the needed of the existing one to play a bit with it
2021-12-03 22:56:52 <psykose> installing gcompat can help with some glibc things but not all
2021-12-03 22:56:55 <psykose> that's about it
2021-12-03 22:59:07 <xordspar0> Good point. I haven't tried, but I bet gcompat works well for Go binaries that only use Glibc for dns.
2021-12-03 23:00:00 <jhjacobs81> well.. installing gcompat worked :)
2021-12-03 23:02:25 <jhjacobs81> that throws out new errors :)
2021-12-03 23:02:52 <jhjacobs81> anyone know what this means: Usage: /lib/ld-musl-x86_64.so.1 [options] [--] pathname [args]
2021-12-03 23:14:16 <jhjacobs81> oh well, i'll give up on Alpine for that project foer the time beeing :)
2021-12-03 23:14:24 <jhjacobs81> i first need to learn some more i think
2021-12-03 23:14:37 <helby> shit was, that while 'service' in default runlevel blocked boot, I was not able even ctrl alt f2 to move to another tty
2021-12-03 23:14:55 <helby> did not know that default level can block everything
2021-12-04 00:41:46 <Justin[m]1234> A minor issue for whoever manages https://build.alpinelinux.org/ - the table is just a bit to wide for it's container, it has a scrollbar.
2021-12-04 00:46:32 <psykose> if you have the talent, you can propose something on https://gitlab.alpinelinux.org/alpine/infra/build-server-status i think
2021-12-04 01:46:52 <Justin[m]1234> <psykose> "if you have the talent, you..." <- I don't know much about css but I fixed it 😁 MR done.
2021-12-04 01:50:02 <psykose> :) now you just have to hope someone sees it
2021-12-04 01:50:53 <Justin[m]1234> Hopefully whoever is connected to that repo hasn't turned off emails about MRs.
2021-12-04 01:51:49 <Justin[m]1234> 
ACTION writes a script to announce the MR every 7 days via Matrix until status !=open

2021-12-04 01:52:17 <Justin[m]1234> There is a MR in there that's over 1 year old.. 😥
2021-12-04 02:09:29 <psykose> sounds about right
2021-12-04 02:10:07 <Justin[m]1234> clandmeter Might be able to merge it.
2021-12-04 09:42:41 <vyivel> what package should i install to use asan? right now gcc complains it can't find libasan_preinit.o and -lasan, and clang can't find libclang_rt.asan-x86_64.a
2021-12-04 09:42:50 <vyivel> couldn't find anything useful on pkgs.alpinelinux.org
2021-12-04 09:51:21 <vyivel> ...ah, apparently asan doesn't work with musl :\
2021-12-04 09:52:38 <vyivel> although gcc's APKBUILD mentions libasan.a when stripping, i'm confused
2021-12-04 11:20:52 <frag> whats the most basic tool that can show two txt files side by side, so its easy to compare visually? (paste doesnt seem to work too well for this)
2021-12-04 11:26:04 <frag> maybe a simple script could do it ...
2021-12-04 11:31:26 <consus[m]> vim
2021-12-04 11:31:43 <consus[m]> * vi
2021-12-04 11:37:33 <frag> busybox vi can do it?
2021-12-04 12:00:41 <tobtobxx> Hey, I'm getting this error when running apk add and other apk commands: ERROR: Unable to lock database: temporary error (try again later)
2021-12-04 12:01:02 <tobtobxx> My guess is that I maybe aborted an apk transaction and this might've corrupted a db.
2021-12-04 12:01:06 <consus[m]> Yep
2021-12-04 12:01:10 <tobtobxx> Is there a way to fix this?
2021-12-04 12:01:33 <tobtobxx> Like with pacman I'd just remove a specific lockfile.
2021-12-04 12:03:30 <lopid> maybe try `apk fix apk\*`
2021-12-04 12:03:42 <lopid> apk-tooks
2021-12-04 12:05:25 <tobtobxx> lopid: apk fix commands fail too.
2021-12-04 12:05:49 <consus[m]> frag: yap
2021-12-04 12:05:53 <tobtobxx> lopid: And I can't install any package (eg. apk-tools) either.
2021-12-04 12:06:02 <tobtobxx> Where is that lockfile?
2021-12-04 12:06:22 <ikke> tobtobxx: do you execute apk as root?
2021-12-04 12:06:31 <tobtobxx> ikke: Through doas, yes.
2021-12-04 12:07:21 <tobtobxx> without doas it would be "ERROR: Unable to lock database: Permission denied", but with doas it's "ERROR: Unable to lock database: temporary error (try again later)"
2021-12-04 12:07:30 <ikke> right
2021-12-04 12:07:43 <ikke> is there a process running that still has a lock on that file/
2021-12-04 12:07:45 <ikke> ?
2021-12-04 12:08:00 <tobtobxx> Yeah, WHICH FILE, I still don't know which.
2021-12-04 12:08:10 <ikke> are there any apk processes running?
2021-12-04 12:08:19 <ikke> just removing that file is not a real solution
2021-12-04 12:08:26 <ikke> /lib/apk/db/lock
2021-12-04 12:08:33 <tobtobxx> There's /usr/libexec/apk-polkit/apk-polkit-server running...
2021-12-04 12:09:31 <ikke> lsof | grep /lib/apk/db/lock
2021-12-04 12:09:58 <tobtobxx> Nope, none.
2021-12-04 12:10:48 <lopid> i've that lock file, dated back in may, and no problems here
2021-12-04 12:12:30 <ikke> yes, the file exists always
2021-12-04 12:12:40 <ikke> apk tries to open it exclusively to obtain the lock
2021-12-04 12:12:44 <ikke> it remains after the process exists
2021-12-04 12:12:54 <tobtobxx> And it should be empty?
2021-12-04 12:13:07 <lopid> tobtobxx, can you touch a new file in that dir?
2021-12-04 12:13:17 <ikke> tobtobxx: yes
2021-12-04 12:13:21 <tobtobxx> sure, as root.
2021-12-04 12:13:37 <ikke> can you execute as root (not with doas): strace -e openat apk update
2021-12-04 12:13:55 <tobtobxx> I don't have strace installed (I tried that before)
2021-12-04 12:14:02 <ikke> hmm, right
2021-12-04 12:14:17 <lopid> install it heh
2021-12-04 12:14:24 <tobtobxx> heh
2021-12-04 12:14:31 <tobtobxx> What could happen if I delete it_
2021-12-04 12:15:06 <tobtobxx> Worst case, apk screws up its db and looses which packages I've installed, but I could always just backup /etc/apk/world.
2021-12-04 12:16:03 <lopid> might as well back up all of /etc and /lib/apk for the size of them
2021-12-04 12:17:16 <tobtobxx> Good news: deleting the lockfile just fixes it.
2021-12-04 12:17:35 <tobtobxx> (The same experience I've had with pacman.)
2021-12-04 13:18:23 <vlp[m]> Hello ! I am trying to install grub on USB key. However, when I launch `grub-install --force --boot-directory /boot/ --efi-directory /boot/efi` I have the error: `cannot open /boot/efi/EFI/apine/grubx64.efi: No error information'.  The `no error information` make me sad :(
2021-12-04 13:27:17 <minimal> vlp: first of all for a USB key I'd expect you to pass grub-install the "--removable" option and possibly also the "--no-nvram" option. Also "--target=x86_64-efi" to signify its a UEFI install, not BIOS (assuming this is on Intel, not Arm)
2021-12-04 13:28:00 <minimal> vlp: have you the grub-efi package installed?
2021-12-04 13:28:29 <vlp[m]> yes it is installed 
2021-12-04 13:28:47 <vlp[m]> I am following the wiki tutorial: https://wiki.alpinelinux.org/wiki/Setting_up_a_laptop
2021-12-04 13:29:11 <vlp[m]> I'll try w/ this additionnal options !
2021-12-04 13:33:24 <minimal> vlp: have you installed grub-efi on the machine itself, distinct from the inside the /mnt/ directory? those instructions show running the "grub-install" from *outside* the chroot directory but installing the grub-efi package inside the chroot
2021-12-04 13:33:37 <minimal> s/from the/from/
2021-12-04 13:33:37 <alpine-meetbot> minimal meant to say: vlp: have you installed grub-efi on the machine itself, distinct from inside the /mnt/ directory? those instructions show running the "grub-install" from *outside* the chroot directory but installing the grub-efi package inside the chroot
2021-12-04 13:39:12 <minimal> vlp: also that Wiki article is almost 2 years old and may be inaccurate for current Alpine (e.g. it mentions the linux-hardened kernel file which no longer exists for several Alpine releases)
2021-12-04 13:44:21 <vlp[m]> <minimal> "vlp: also that Wiki article is..." <- Yes I noticeed some commands with missing option I just edit some lines
2021-12-04 13:45:57 <minimal> vlp: and my question about whether grub-efi is installed on the machine (in addition to inside /mnt directory)?
2021-12-04 13:45:57 <vlp[m]> It is 2y old but it seems to be a good start to learn install/boot process
2021-12-04 13:46:21 <vlp[m]> 
ACTION preparing fresh coffee :)

2021-12-04 13:46:32 <vlp[m]> yes you are right, there is a mistake here
2021-12-04 13:46:39 <vlp[m]> apk add grub-efi is OK
2021-12-04 13:47:16 <minimal> so the earlier "cannot open /boot/efi/EFI/apine/grubx64.efi" was probably because grub-efi was not installed on the machine
2021-12-04 13:47:32 <vlp[m]> but apk add grub-efi --root=/mnt/root/ --keys-dir ... -repo-....  is KO
2021-12-04 13:51:03 <minimal> vlp: this guide seems "dangerous", for example: "You need to manually edit the initramfs-init.in in mkinitfs to support two factor authentication using cryptsetup" - making changes to this file risks having future problems
2021-12-04 13:53:13 <vlp[m]> I am using a spare laptop juste for testing
2021-12-04 13:54:09 <vlp[m]> There is no data
2021-12-04 13:55:11 <minimal> vip: I mean that once you start modifying parts of mkinitfs then you are diverging from the Alpine mkinitfs package (and any future changes made to it)
2021-12-04 13:55:53 <minimal> in the future if you update the mkinitfs package things could then potentially break
2021-12-04 13:56:58 <vlp[m]> My goal (for fun & leran) is to set a full E2E encryption laptop w/ gpg, there is other and more standard way ?
2021-12-04 13:57:10 <vlp[m]> s/leran/learn/
2021-12-04 13:57:10 <alpine-meetbot> vlp[m] meant to say: My goal (for fun & learn) is to set a full E2E encryption laptop w/ gpg, there is other and more standard way ?
2021-12-04 13:59:54 <minimal> vlp: I don't believe standard mkinitfs supports GPG, so no there is no (current) standard way for that. The problem is not finding an unofficial way to get that working, the problem is doing so in a way that will not potentially suddenly break for you in the future due to changes in the standard Alpine packaged
2021-12-04 14:10:11 <vlp[m]> Thanks minimal , for a more standard E2E encryption I imagine that I can use: https://wiki.alpinelinux.org/wiki/LVM_on_LUKS#Preparing_the_Temporary_Installation_Environment
2021-12-04 14:11:21 <ikke> the install script now supports setting up luks
2021-12-04 14:11:51 <vlp[m]> Ah yes, I saw this in the 3.15 changelog
2021-12-04 14:11:54 <ikke> So I don't think you need to do it manually like that anymore
2021-12-04 14:12:07 <vlp[m]>  /boot included? 
2021-12-04 14:12:18 <ikke> I'm not sure
2021-12-04 14:13:11 <minimal> vlp: it depends on what you are trying to achieve. The original Wiki article you referenced seemed to be doing "2FA" for LUKS which the standard install script in Alpine 3.15 doesn't provide
2021-12-04 14:14:38 <vlp[m]> minimal: Yes, I'd like to go into 2FA conf 
2021-12-04 14:15:45 <minimal> vlp: ok, and as I explained that sort of stuff is not supported by "standard" Alpine and once you start doing things like editing the mkinitfs' initramfs-init script you risk having problems in the future with package update/upgrades
2021-12-04 14:16:18 <vlp[m]> But as you said, I'd like to stay as standard as possible to avoid any future failures due to custom conf
2021-12-04 14:16:53 <minimal> vlp: ok, but you can't have it both ways - you cannot stay standard and also do things that standard does not support
2021-12-04 14:18:50 <minimal> vlp: for example I developed changes to mkinitfs to support the remote unlocking of LUKS via SSH. I submitted a MR/patch to mkinitfs to get that incorporated into "standard" mkinitfs and this has not happened. I have not make use of this functionality myself since as I feel that if I did so there is too great a risk that it could break in the future whenever I do "apk upgrade"
2021-12-04 14:20:31 <vlp[m]> That would have been a great feature! 
2021-12-04 14:23:03 <minimal> vlp: If I did have a machine running Alpine 3.14 with that change manually applied (to initramfs-init) and then upgraded that machine to 3.15 then things would have problems as Alpine 3.15's mkinitfs package included changes to initramfs-init that are required on 3.15 but my manual modification to that file on 3.14 would have prevented the 3.15 changes being applied
2021-12-04 14:23:37 <minimal> and so would have broken such a machine, preventing it booting correctly
2021-12-04 15:45:00 <blt[m]> minimal: I second vlp this would have been a good feature specifically for server use case...
2021-12-04 15:48:34 <blt[m]> vlp: we may need to propose a MR to include GPG in initramfs as I like the concept of using GPG / smartkey to decrypt LUKS. For smartkey we also need to see what other package have to be including on top of GPG to make it work (scdaemon...etc)
2021-12-04 15:49:46 <minimal> blt: the MR still exists in the mkinitfs repo, it might at some point in the future get incorporated
2021-12-04 15:50:01 <minimal> blt[m]: you were asking about cloud-init the other day?
2021-12-04 15:51:38 <minimal> blt[m]: FYI: https://gitlab.alpinelinux.org/alpine/mkinitfs/-/merge_requests/86
2021-12-04 15:53:37 <blt[m]> minimal: yes ! Indeed I have prepared my current infrastructure (based on Hyperbola today) I would like to switch everything to Alpine as I really the concept and philosophy.. So I prepared to VM with nested emulation to simulate two hosts and tests Alpine with LXD and cloud-init but I was not finding a ton of documentation
2021-12-04 15:55:18 <minimal> blt[m]: what's Hyperbola? I'm the Alpine maintainer for cloud-init
2021-12-04 15:56:22 <minimal> blt[m]: depends what you mean by docs, if its about how to use cloud-init then that would be the general c-i docs: https://cloudinit.readthedocs.io/en/latest/
2021-12-04 15:56:53 <blt[m]> My main question is around the sequence of events during boot between init/lbu/openrc & cloud-init as my goal is to only to diskless Alpine VM and ideally I would prefer to host my lbu in a remote server through NFS for example but I am wondering if cloud-init would help to push all the connectivity and packages needed and launch lbu to save everything for the next reboot to have a full working VM
2021-12-04 15:57:22 <blt[m]> minimal: Hyperbola is my current Linux version I will replace with Alpine
2021-12-04 15:57:35 <frag> does browsers work better on 3.15? (or am i just lucky so far :p)
2021-12-04 15:57:57 <blt[m]> So I guess cloud-init is like other openrc init managed in default ?
2021-12-04 15:58:01 <caskd> frag: qt5-qtwebengine got bumped so likely
2021-12-04 15:58:03 <minimal> blt[m]: if "docs" is about how to build an Alpine image with cloud-init then, yes, the only docs as such as the README.Alpine in the cloud-init-docs package. I do have my own script for building c-i enabled Alpine images which could act as a "form" of documentation for how to do so :-)
2021-12-04 15:58:35 <frag> caskd: bumped? yes ive mostly tested qutebrowser
2021-12-04 15:58:48 <caskd> where does APK store it's database? i've setup a dev env and screwed up the db and i don't really wanna delete everything
2021-12-04 15:59:07 <caskd> frag: yes, the engine qutebrowser uses was updated, so it might be more stable
2021-12-04 15:59:48 <minimal> blt[m]: Right, have not yet found the time to setup c-i with Alpine diskless, so far have being doing so in "sys" (i.e. normal install-to-disk) mode. It is on my list of things to add to my own script
2021-12-04 16:00:22 <blt[m]> minimal: would be very nice!
2021-12-04 16:00:54 <blt[m]> For me the combination of c-i and lbu could be very powerful
2021-12-04 16:01:25 <blt[m]> One for kind of -initialization and the last for saving configuration and restoring it after reboot
2021-12-04 16:01:55 <caskd> nvm found the db
2021-12-04 16:02:45 <minimal> blt[m]: yes, I've thought the same - "forgetful" VM/Cloud server images where every boot is a "1st boot" from a c-i perspective
2021-12-04 16:03:26 <blt[m]> I think the main issus with diskless and remote lbu directory is how lbu would be able to connect and maybe c-i can modify the grub / kernel parameters to provide very early boot access to retrieve lbu cache and conf
2021-12-04 16:03:49 <blt[m]> I like the concept of "forgetful VM"
2021-12-04 16:05:51 <blt[m]> minimal: would have a minimal (word game on me ;) c-i working config you could share for me to start doing some tests ?
2021-12-04 16:06:06 <frag> caskd: nice!
2021-12-04 16:16:28 <minimal> blt[m]: some examples as part of my script: https://github.com/dermotbradley/create-alpine-disk-image/tree/main/Virtual/config-examples
2021-12-04 17:20:46 <leah2> fcolista: do you realize https://git.alpinelinux.org/aports/tree/community/aircrack-ng/make-check-fix.patch just papers over the bug on musl?
2021-12-04 17:23:06 <helby> gzip to kernel did 100MB saving, a lot
2021-12-04 17:37:31 <leah2> fcolista: i think this patch is better: https://github.com/void-linux/void-packages/blob/master/srcpkgs/aircrack-ng/patches/getopt.patch
2021-12-04 17:47:22 <blt[m]> <minimal> "blt: some examples as part of my..." <- Many thanks ! I will try that hopefuly next week
2021-12-04 17:48:22 <minimal> blt[m]: which c-i Data Source are you intending to use?
2021-12-04 17:48:59 <minimal> blt[m]: if you look at the script in that repo, that's what I use to build cloud-init Alpine images "from scratch"
2021-12-04 17:50:58 <Nulo> Hey, I'm getting "warning: the debug information found in "/usr/lib/debug//usr/lib/libwayland-server.so.0.1.0.debug" does not match "/usr/lib/libwayland-server.so.0" (CRC mismatch)." when trying to run gdb to analyze a core dump of a program that uses libwayland. Same thing happens with wlroots. Any clue?
2021-12-04 17:55:12 <Hello71> i also got that on musl, not sure why
2021-12-04 17:56:05 <Hello71> you can try apk fix wayland wayland-dbg but i think it won't help
2021-12-04 17:58:29 <Nulo> Nope, doesn't help sadly
2021-12-04 18:20:53 <Hello71> you're using dl-cdn?
2021-12-04 18:21:12 <Hello71> also i think it's wayland-libs-server wayland-dbg
2021-12-04 19:56:26 <blt[m]> <minimal> "blt: if you look at the script..." <- The create-seed I guess?
2021-12-04 20:09:27 <minimal> blt[m]: I was referring to the create-alpine-disk-image script that installs Alpine in a chroot and then sets up cloud-init etc
2021-12-04 20:10:11 <minimal> the create-seed script just creates an ISO image containing the cloud-init YAML files to be used by the cloud-init NoCloud DataSource when a VM boots
2021-12-04 20:13:37 <blt[m]> got it!
2021-12-04 20:16:11 <blt[m]> <minimal> "blt: I was referring to the..." <- this script is hugeee!
2021-12-04 20:17:10 <minimal> blt[m]: yupe, there's a lot of functionality in it - however all it does is "spits out" a far smaller script tailored to do exactly what you want
2021-12-04 20:18:01 <blt[m]> I am curious to know if you already tried Terraforms with cloud-init plugin with Alpine?
2021-12-04 20:19:59 <minimal> blt[m]: there should be nothing Alpine-specific really about Terraform, use it to create a VM for example, and supply the user-data as part of that (assuming you're using a cloud provider where you have the Alpine image uploaded)
2021-12-04 20:20:38 <minimal> if you're using something like libvirtd then you could use Terraform to create the seed iso to provide to the VM
2021-12-04 20:22:45 <minimal> haven't actually sat down and written a Terraform plan for that yet but will once I create a "local" (i.e. account-specific) AMI for my Alpine image in AWS
2021-12-04 20:23:07 <blt[m]> yep I am now (with Hyperbola Linux) but I am still thinkning about the next infra I want to create and I think to use LXD instead but never tried it so I think I still need to play to see the basics and how I can use c-i with it maybe Terraforms would be overkilled for me as my usecase is just to creae some VMs fast and having c-i to put everything I need
2021-12-04 20:24:05 <minimal> blt[m]: cloud-init added a DataSource for LXD in their latest release but have never tried LXD in general
2021-12-04 20:24:40 <minimal> blt[m]: are you looking to create infra "in the cloud" or on a local hypervisor?
2021-12-04 20:24:40 <ikke> I think ncopa tested someting with terraform, but I'm not sure if it integrated with cloud-init
2021-12-04 20:26:25 <minimal> ikke: terraform's had support for cloud-init in general for quite some time (certainly for its various cloud providers). I'm not sure if the likes of terraform's libvirt provider is "stable" these days for local use
2021-12-04 20:27:56 <ikke> ncopa did manage to setup a cluster of vms
2021-12-04 20:28:02 <blt[m]> minimal: local hypervisors (my 2 servers in two locations), my idea is to have them on Alpine in my new install and using LXD with c-i for creation and automation and idealy lbu for saving the conf
2021-12-04 20:28:23 <blt[m]> everything with diskless Alpine VMs
2021-12-04 20:28:41 <blt[m]> this is ultimately my goal 
2021-12-04 20:29:28 <blt[m]> ideally creating a wiki page for the community to reproduce the same kind of infra
2021-12-04 20:30:05 <minimal> blt[m]: generally with cloud-init for non-cloud use (i.e. where there is no metadata server) you use the NoCloud DS which expects a suitable ISO with the YAML config. However you could instead use the "seed" feature of NoCloud where it retries these files via http/https
2021-12-04 20:31:10 <minimal> blt[m]: what you're talking about is basically what I've been working on for some time - my script is just a step on the way there (to create the disk images) :-)
2021-12-04 20:31:16 <blt[m]> thanks that exactly what I was trying to understand as I though I could provide the yaml directly to LXD parameters
2021-12-04 20:32:29 <minimal> blt[m]: LXD for containers or for VMs?
2021-12-04 20:33:30 <ikke> https://tpaste.us/Qrkl 
2021-12-04 20:33:40 <ikke> It does use cloud-init
2021-12-04 20:33:46 <ikke> minimal: ^
2021-12-04 20:34:00 <blt[m]> good question  idont know if Alpine can run as pure "containers"
2021-12-04 20:34:31 <ikke> We do run alpine as lxc containers, not sure how much different that is from lxd
2021-12-04 20:35:16 <blt[m]> I think it is the same as far as I understand
2021-12-04 20:36:12 <minimal> ikke: yeah he doesn't seem to be actually using cloud-init, rather he's using some (shell?) code to read the c-i yaml as a simpler/smaller alternative to c-i. He is also using the terraform libvirt provider - haven't looked at that in several years when it was still in development
2021-12-04 20:36:28 <ikke> ok
2021-12-04 20:37:00 <minimal> ikke: never used it but AFAIK LXD is common infra for both (LXC) containers and KVM-based VMs
2021-12-04 20:37:12 <minimal> so you can run either/both with LXD
2021-12-04 20:39:28 <blt[m]> minimal: interesting I didn't know that LXD is not really using c-i but "basic" script instead? but how it interact with the c-i service inside the Alpine VM?
2021-12-04 20:40:18 <minimal> blt[m]: sorry, I was saying that to ikke about what ncopa seemed to be doing, it wasn't about LXD
2021-12-04 20:42:12 <minimal> regarding LXD with VMs, you can see its using the same cli as "lxc", just with a "--vm" flag to create VMs rather than containers: https://discuss.linuxcontainers.org/t/running-virtual-machines-with-lxd-4-0/7519
2021-12-04 20:42:59 <minimal> you'll also see in that like it says (regarding Ubuntu, by applies to other distros as well) "For official Ubuntu images, cloud-init must be used along with a config drive to seed a default user into the VM and allow console access."
2021-12-04 20:45:30 <minimal> as I mentioned, cloud-init added a LXD DataSource recently: https://cloudinit.readthedocs.io/en/latest/topics/datasources/lxd.html
2021-12-04 20:49:35 <idkrn> does anyone here use dmenu or rofi?
2021-12-04 20:49:53 <idkrn> I'm trying to get them to find the start tor browser script
2021-12-04 20:53:06 <bdju> is the script in your PATH?
2021-12-04 20:53:57 <bdju> those can also take arbitrary text for selection in case that would work with your workflow
2021-12-04 20:54:34 <psykose> if it's not in PATH but a .desktop file you need something that reads desktop file paths
2021-12-04 20:54:54 <psykose> for dmenu there is j4-dmenu-desktop that executes a dmenu after parsing some extra paths
2021-12-04 20:55:25 <psykose> if it's not a desktop file, just make sure it's in path i guess
2021-12-04 20:55:35 <bdju> rofi can read desktop files but may need an extra mode in the launch options
2021-12-04 20:57:04 <psykose> should just work in rofi i think
2021-12-04 20:57:22 <psykose> in any case even for desktop files the paths can be wrong
2021-12-04 20:57:51 <psykose> has to be under xdg_data_dirs or something like that
2021-12-04 20:58:30 <bdju> here's what I used to bind as my rofi launch command: rofi -kb-primary-paste Control+v,Insert -kb-secondary-paste Control+Shift+v,Shift+Insert -sidebar-mode -combi-modi window,drun,run -modi run,drun,window,ssh,combi -show run
2021-12-04 20:58:50 <bdju> iirc run is standard binaries in PATH and drun is desktop files
2021-12-04 20:59:42 <psykose> ah right, that makes sense
2021-12-04 21:04:40 <idkrn> bdju: ~/.local/share/applications
2021-12-04 21:05:30 <idkrn> so rofi would be better for .desktop files?
2021-12-04 21:06:08 <bdju> probably, yeah. I'm sure dmenu can handle them if patched, but it might not out of the box
2021-12-04 21:07:21 <bdju> apparently "i3-menu-desktop" which I think is a script that uses dmenu can handle desktop files. not sure what environment you're using it in
2021-12-04 21:08:43 <bdju> https://github.com/enkore/j4-dmenu-desktop found this as well
2021-12-04 21:09:34 <idkrn> well sway is my preference, so I'm assuming anything 13 related would help
2021-12-04 21:09:52 <idkrn> i3*
2021-12-04 21:10:42 <bdju> I'm also on Sway and I'm using a mix of bemenu and wofi, similarly because wofi easily handles desktop files and other stuff like rofi did, and bemenu because it's a lot faster than wofi and handles large amounts of text better.
2021-12-04 21:11:00 <idkrn> I use bemenu as well but I dont understand it
2021-12-04 21:11:09 <bdju> I switched off rofi because there wasn't native wayland support and I was having some weird issues with multiple monitors where it would appear on the opposite monitor or in the middle of them
2021-12-04 21:11:11 <idkrn> theres no patching like there is with dmenu so i dont understand
2021-12-04 21:11:16 <bdju> ah
2021-12-04 21:11:38 <idkrn> options are configured through environment variables or something
2021-12-04 21:11:47 <bdju> you can add options for bemenu that you always want on as env vars, yeah. I do some in my .profile
2021-12-04 21:11:57 <bdju> export BEMENU_OPTS="--no-overlap --monitor -1 --ignorecase"
2021-12-04 21:11:57 <idkrn> wofi no longer recieves updates fyi
2021-12-04 21:12:23 <bdju> Well, the maintainer is busy, yeah. There are commits from under a month ago, though.
2021-12-04 21:12:26 <idkrn> yea just a bit harder to find documentation for bemenu compared to dmenu
2021-12-04 21:12:41 <bdju> bemenu has a man page that's pretty helpful
2021-12-04 21:12:48 <idkrn> oh I had no idea
2021-12-04 21:12:54 <idkrn> I guess I should've checked
2021-12-04 21:13:49 <bdju> It didn't used to have one when I first started using it, or at least it wasn't in any releases.
2021-12-04 21:13:57 <idkrn> oh ok
2021-12-04 21:14:30 <idkrn> If there was an official patch that could get dmenu to launch tor's start script or any .desktop file I would just use that
2021-12-04 21:14:40 <idkrn> cause I rarely need to configure a run launcher
2021-12-04 21:15:18 <idkrn> it's also in the default sway config which is nice
2021-12-04 21:17:03 <helby> dmenu is in sway? does it even support wayland?
2021-12-04 21:17:54 <idkrn> sway can use xwayland
2021-12-04 21:17:56 <bdju> I think it assumes you have xwayland, and it kinda/mostly works. I remember having issues with it not appearing, or possibly appearing under my panel. I haven't tried it lately.
2021-12-04 21:18:02 <idkrn> but sway pipes dmenu through swaymsg I think
2021-12-04 21:18:31 <helby> interesting choice
2021-12-04 21:18:55 <helby> as bemenu works well
2021-12-04 21:19:03 <idkrn> not as popular
2021-12-04 21:19:21 <bdju> bemenu didn't used to work well. there was a horrible bug for a while where it'd eat your keypresses
2021-12-04 21:19:25 <idkrn> If I could I would rewrite dmenu in rust lol
2021-12-04 21:20:21 <helby> still if you have just few application it's simple to just set up keybinding for that
2021-12-04 21:20:25 <idkrn> I don't understand how bemenu is like dmenu when you don't just compile from source
2021-12-04 21:20:29 <helby> and avoid any menu at all
2021-12-04 21:20:41 <idkrn> well I can't launch tor browser like that
2021-12-04 21:20:50 <idkrn> its a .desktop file or what ever
2021-12-04 21:20:52 <bdju> bemenu has a bunch of built-on options, similar to rofi. it just looks/acts like dmenu in that it can launch things or be a menu used in scripts
2021-12-04 21:21:01 <idkrn> oh I see
2021-12-04 21:21:17 <bdju> you don't need to set any env vars, you can also just add arguments to the command every time
2021-12-04 21:21:28 <bdju> I used it over a year before figuring out the env var thing
2021-12-04 21:21:54 <idkrn> which way do you prefer?
2021-12-04 21:22:23 <bdju> for the options I was always using like setting the font, size, which monitor(s) to appear on, etc. I put them in an env var since it just saves typing
2021-12-04 21:22:34 <bdju> but there may be some options you only want for some cases, like case-insensitive search
2021-12-04 21:22:39 <idkrn> you didn't just change the keybinding?
2021-12-04 21:22:54 <idkrn> wait that makes no sense
2021-12-04 21:22:56 <bdju> well, I have the bemenu-run keybinding, but it's also usable in scripts / one-liners
2021-12-04 21:23:19 <bdju> you can pipe ls to bemenu and then your selection is printed to stdout
2021-12-04 21:23:37 <bdju> so you can use it for selecting things from a text file, like a DIY clipboard manager for example
2021-12-04 21:23:46 <idkrn> I found that suckless has an irc on oftc I'm gonna go ask there lol
2021-12-04 21:23:51 <bdju> okay
2021-12-04 21:24:50 <bdju> I'm guessing he only uses dmenu to launch programs
2021-12-04 21:27:27 <psykose> export the things you want global as env and set the rest locally? doesn't seem very hard
2021-12-04 21:27:47 <bdju> yeah maybe I said something weird but not sure which thing
2021-12-04 21:39:18 <Nulo> Hello71, yes, dl-cdn
2021-12-05 04:46:12 <Porkyofthepine[m]1> I'm trying to switch to networkmanager for wireless networks. I've got iwd installed and set as the backend per the wiki, I've got the device enabled in networkmanager, but when I try to connect, I get `Error: Insufficient privileges` when I run as a normal user and `Error: Connection activation failed: (7) Secrets were required but not provided.` when I run as root
2021-12-05 04:53:07 <psykose> no idea about the root error, but for user you want to be in `network` group
2021-12-05 04:54:36 <psykose> or was it netdev
2021-12-05 04:55:06 <psykose> ah, plugdev in alpine
2021-12-05 04:56:32 <zcrayfish> Are you running nm-applet? if so you might want to launch it from a terminal emulator in order to get more detailed error information
2021-12-05 04:56:59 <Porkyofthepine[m]1> my user is already in plugdev
2021-12-05 04:57:25 <Porkyofthepine[m]1> I'm using nmcli in a terminal emulator and getting those outputs
2021-12-05 04:59:00 <psykose> is the networking service enabled and also with stuff in /etc/network/interfaces for wlan0
2021-12-05 04:59:10 <psykose> afaik you are supposed to disable networking and enable networkmanager
2021-12-05 04:59:44 <Porkyofthepine[m]1> i've stopped wpa_supplicant and started iwd and networkmanager
2021-12-05 04:59:59 <psykose> did you stop networking
2021-12-05 05:00:21 <Porkyofthepine[m]1> oh is that it's own thing? let me try that
2021-12-05 05:01:15 <zcrayfish> I would recommend not doing that.
2021-12-05 05:01:28 <Porkyofthepine[m]1> now `nmcli d wifi list` gives nothing
2021-12-05 05:01:33 <zcrayfish> Instead make sure that /etc/network/interfaces is exclusively populated by the loopback interface.
2021-12-05 05:02:28 <Porkyofthepine[m]1> so don't have wlan0 in there?
2021-12-05 05:02:37 <zcrayfish> No, just lo
2021-12-05 05:04:42 <Porkyofthepine[m]1> getting the same output
2021-12-05 05:05:22 <Porkyofthepine[m]1> and I restarted the nm service after editing it to make sure
2021-12-05 05:11:13 <zcrayfish> can you pastebin at least the [main] and [device] sections from /etc/NetworkManager/NetworkManager.conf ??
2021-12-05 05:13:57 <Porkyofthepine[m]1> https://pastebin.com/3na2rnJR
2021-12-05 05:14:41 <Porkyofthepine[m]1> i'm not sure if I need to put keyfile in there because afaik I'm not using it. That was just copying others that i found online
2021-12-05 05:15:56 <Porkyofthepine[m]1> removing keyfile from it didn't change anything
2021-12-05 05:17:10 <zcrayfish> NetworkManager loads it by default anyway, so no need to declare it
2021-12-05 05:18:42 <zcrayfish> This is what I am using; no issues managing bluetooth, wifi, or ethernet networkin with it:
2021-12-05 05:18:44 <zcrayfish> https://pastebin.com/2Fib9EPg
2021-12-05 05:20:01 <zcrayfish> the auth-polkit directive looks like it could be helpful
2021-12-05 05:20:12 <Porkyofthepine[m]1> i take it auth-polkit=false removes the need for polkit? before I installed it, I was getting the secrets error for both user and root so I'm guessing it has something to do with my problems
2021-12-05 05:22:59 <Porkyofthepine[m]1> so now I'm getting the secrets required but not provided error still
2021-12-05 05:23:06 <Porkyofthepine[m]1> for both user and root
2021-12-05 05:24:18 <Porkyofthepine[m]1> and when I tried looking up the error, all i got were things relating to vpns. which i'm not using a vpn with this
2021-12-05 05:25:06 <zcrayfish> it can display that error when it has trouble reading the [wifi-security] section or the psk directive (among others) in the .nmconnection files also.
2021-12-05 05:25:36 <zcrayfish> They should be stored in /etc/NetworkManager/system-connections
2021-12-05 05:26:04 <Porkyofthepine[m]1> yeah I actually just found a stackexchange page talking about that or something similar
2021-12-05 05:26:43 <Porkyofthepine[m]1> `nmcli con delete <SSID>` fixed it
2021-12-05 05:26:51 <Porkyofthepine[m]1> thanks for your help
2021-12-05 05:27:12 <zcrayfish> No problem. 
2021-12-05 06:03:53 <Guest7603> anyone run a full desktop using alpinelinux?
2021-12-05 06:09:55 <psykose> me
2021-12-05 06:16:31 <zcrayfish> jwm & pcmanfm -d --desktop --one-screen count as a full desktop, right?
2021-12-05 06:20:56 <Guest7603> zcrayfish, yea
2021-12-05 06:21:06 <Guest7603> i like your setup
2021-12-05 06:21:22 <Guest7603> do we need to install additional driver for wifi ??
2021-12-05 06:23:01 <zcrayfish> that depends on your hardware. 
2021-12-05 06:23:27 <psykose> linux-firmware pulls in all the firmware for everything, aside from that it depends what card you have
2021-12-05 06:34:18 <Guest7603> zcrayfish, your setup is nice. i bet it is around 100mb memory on start up?
2021-12-05 06:34:53 <psykose> probably less
2021-12-05 06:35:45 <ktprograms> zcrayfish: If you use pcmanfm --desktop, how do you have access to the right click menu?
2021-12-05 06:39:21 <bdju> jwm was mentioned too
2021-12-05 06:39:49 <ktprograms> bdju: What I mean is that the pcmanfm right click menu is activated instead of the jwm right click menu
2021-12-05 06:40:04 <bdju> ah, gotcha
2021-12-05 06:40:47 <bdju> haven't used jwm, but if it's anything like openbox, there may also be a keyboard shortcut to get to the same menu for the wm
2021-12-05 06:41:52 <ktprograms> BTW is there any openbox-like wm for wayland?
2021-12-05 06:42:48 <bdju> there are one or two but I'm not sure if they're in a usable state
2021-12-05 06:43:01 <bdju> https://github.com/swaywm/wlroots/wiki/Projects-which-use-wlroots
2021-12-05 06:43:12 <ktprograms> bdju: How similar is weston to openbox?
2021-12-05 06:43:18 <bdju> tinybox and waybox
2021-12-05 06:43:29 <psykose> labwc as well
2021-12-05 06:43:31 <bdju> uh I'm not sure what weston is like to use. I thought it was mainly made to be an example
2021-12-05 06:43:41 <bdju> I use sway
2021-12-05 06:44:50 <bdju> labwc has been actively committed to more recently than the others, so it's good to know about that one
2021-12-05 06:45:32 <psykose> personally i have only used sway, and opened wayfire and gnome once
2021-12-05 06:45:44 <ktprograms> Thanks, I guess I'll try either weston or labwc when I get around to experimenting with wayland.
2021-12-05 06:45:46 <psykose> the wayfire.. swoopy effect thing when you drag windows is very.... satisfying
2021-12-05 06:46:37 <ktprograms> psykose: Is wayfire a compositor or a wm?
2021-12-05 06:46:49 <psykose> same thing
2021-12-05 06:47:02 <psykose> wayland wm's are called compositors since they do everything in the stack
2021-12-05 06:47:09 <psykose> or something
2021-12-05 06:47:11 <bdju> yeah
2021-12-05 06:47:14 <psykose>  i am too tired to think
2021-12-05 06:47:18 <ktprograms> psykose: Ah right
2021-12-05 06:47:40 <ktprograms> I just checked, labwc isn't packaged for alpine yet?
2021-12-05 06:47:51 <psykose> indeed not
2021-12-05 06:47:56 <psykose> do you want me to package it
2021-12-05 06:48:22 <ktprograms> psykose: I'll do it when I'm ready to experiment with it. Hopefully it shouldn't be too much trouble for me.
2021-12-05 06:49:00 <psykose> sure
2021-12-05 06:49:09 <psykose> it looks easy enough, there's been 300 commits since last tag
2021-12-05 06:49:16 <psykose> not sure how much has been fixed/improved in that time
2021-12-05 06:50:14 <psykose> looks like a bunch
2021-12-05 06:50:49 <ktprograms> psykose: So you're saying that when I package it I should probably do it from a recent git commit and not the last tag?
2021-12-05 06:51:04 <psykose> no, just talking out loud
2021-12-05 06:51:41 <ktprograms> psykose: Ok, I'll see how it is when I get around to it then.
2021-12-05 06:51:49 <psykose> hehe https://github.com/labwc/labwc/issues/144
2021-12-05 06:51:52 <psykose> but yeah it should be fine
2021-12-05 06:51:59 <psykose> standard wayland dependencies and all
2021-12-05 06:53:03 <ktprograms> psykose: Ok, thanks.
2021-12-05 07:25:43 <voltagex> Hi there, is there a way I can build an edge netboot ISO or virtual image?
2021-12-05 07:35:56 <psykose> there is https://github.com/alpinelinux/alpine-make-vm-image
2021-12-05 07:37:59 <voltagex> psykose: https://wiki.alpinelinux.org/wiki/How_to_make_a_custom_ISO_image_with_mkimage is what I was looking for in the end
2021-12-05 07:38:31 <psykose> yes, it is one of the two options one link above in https://wiki.alpinelinux.org/wiki/How_to_make_a_custom_ISO_image
2021-12-05 08:05:46 <frag> where does the vi man page come from? "POSIX Programmer's Manual" "VI(1P)"
2021-12-05 08:07:20 <psykose> man-pages?
2021-12-05 08:07:46 <MaximKarasev[m]> frag: apk info -W ...
2021-12-05 08:07:57 <psykose> https://man7.org/linux/man-pages/man1/vi.1p.html
2021-12-05 08:10:59 <frag> man-pages indeed
2021-12-05 08:15:11 <frag> 'apk info -all' seems useful :]
2021-12-05 08:17:08 <frag> do you use man-pages? i think i hardly ever have looked up those things, but prob usefull if you get in the habit..
2021-12-05 08:19:00 <mtm84[m]> In the age of the internet on your cell phone, probably not very useful to most people. But no reason not to have the option if you want local reference
2021-12-05 08:19:22 <nero> there is nothing better than the original manpages
2021-12-05 08:19:29 <lopid> all the time
2021-12-05 08:19:31 <nero> documents on the internet often refer to the wrong version of the software
2021-12-05 08:19:44 <lopid> …and are written by idiots
2021-12-05 08:19:58 <nero> or worse, people read the arch linux wiki and try it on alpine linux
2021-12-05 08:20:34 <frag> :]
2021-12-05 08:20:50 <nero> there are whole stacks of programming articles about less-known ssh and ssh-keygen functions
2021-12-05 08:21:01 <mtm84[m]> Many actual man pages are replicated verbatim on the net >.> 
2021-12-05 08:21:03 <nero> several days worth of read, just to save 10 minutes reading the damn manpage
2021-12-05 08:22:16 <nero> if you do apk add doc man-pages you get alot of them
2021-12-05 08:22:26 <mtm84[m]> To be fair some man pages don’t always go into high detail on what a command might do.   Most major or well known stuff will of course
2021-12-05 08:22:33 <lopid> a lot
2021-12-05 08:23:14 <mtm84[m]> One of these days a lot will officially merge into one word, like goodbye did
2021-12-05 08:23:24 <mtm84[m]> But it is not this day
2021-12-05 08:23:39 <lopid> alot already is a word, which means something completely different
2021-12-05 08:23:52 <lopid> i am thinking of allot
2021-12-05 08:24:11 <nero> in a world where people cant tell the difference between a virtual machine and an emulator, i dont feel like not saying "alot" makes alot of difference
2021-12-05 08:24:28 <mtm84[m]> Lead and lead
2021-12-05 08:24:39 <mtm84[m]> Lol
2021-12-05 08:25:11 <nero> like, people always say wine is not an emulator, but by the same standards, terminal emulators wouldnt be, either
2021-12-05 08:28:56 <mtm84[m]> To mean an emulator is something that implements hardware.  But that’s a poor rendering of my thoughts on the matter.  Ultimately they all work as some kind of translation layer.
2021-12-05 08:42:25 <nero> mtm84[m]: wrong
2021-12-05 08:42:45 <nero> just do a lookup on the web... its actually a pretty generic term
2021-12-05 08:46:13 <mtm84[m]> software implementing hardware is a pretty generic term
2021-12-05 08:49:34 <nero> emulation is basically just another flavor of "imitation"
2021-12-05 08:49:51 <nero> and also used in non-computing contexts
2021-12-05 08:50:01 <nero> someone or something tring to be/act like someone or something other
2021-12-05 09:25:45 <Apachez> emulator is something that mimics the hardware
2021-12-05 09:26:02 <Apachez> either by software (wine, qemu) or by hardware (fpga)
2021-12-05 09:31:51 <gwyef> D i d    u s a   t r a i n    &   s u p pl y      i s i s     wi th   w e a p o n s     l i k e     i t     d i d   w i t h       a l    q a e d a   to   j u st i f y  c r e a t i n g  w a r  s        > >       C R E A T  I V E     C H O A S
2021-12-05 09:31:55 <gwyef> d i d     c i a     d i d     9 /11    or    i t    j u s t     l e t       it      h a p p en
2021-12-05 09:31:57 <gwyef> if     a l   q a e d a      di d    it     w h y      t o    k i l l   9  m i l l i  on     i  r a q i s 
2021-12-05 09:32:00 <gwyef> ab o u t    9   m il i o n        i r a q i   l o s t  th e i r  li ves      e i th e r  be ca uz  o f   dir e ct  viol ence  o f   u sa   and   becauz of  si ege      u sa    ma de        ag a in s t     i r a q  t h at   pr eve nt ed    i raqi   f or m es se nt ial         l i v i n   g       ma te rial s , food ,m e d i c i n e   and  even     p e n c i l s
2021-12-05 09:32:05 <gwyef> i s i s   a n d a l q a e da    o n l y     a p p e a r e d  i n       i r  a  q    a f t e r  u s a  i n v a si on  , D i d    u s a     p a v e   th e w ay  f or  i s i s   to  i r a q ?
2021-12-05 09:32:10 <gwyef>  sa d  da m     hu ss i n    wh o    lo st   m o s t   of  h i s    p o w er  i n 1st , 2 nd  gu lf  w a    rs  an d   du ri n g    10   y ea rs   of    sie ge    ,  d i d   no t  a l l o w    i s is    o r  a l q a e dea   t o   e n  t e r    i  r   a   q  , e ve n to h e l p h i m    a g a i n s t    u s a
2021-12-05 09:32:14 <gwyef> p l e a s e    s h a r e  m y  qs to l e s s e n  u s a     a g g r e s s i o n    a g a i n s t    o t h e rs
2021-12-05 09:40:25 <Justin[m]1234> You able to remove the spam as well @ikke ?
2021-12-05 09:40:35 <ikke> no
2021-12-05 09:40:52 <Justin[m]1234> Is there anyone on the Matrix side that can?
2021-12-05 09:45:36 <vlp[m]> It seems that there is no matrix admin into the room 
2021-12-05 10:40:49 <ahmadraniri1994[m]> Hi, Is there a tutorial to install alpine linux with bootstrap like debootstrap or some kind like that? Last week I practiced via qemu (use bootstrap and chroot), it booted successfully but there was an error (sysroot not found) and I was prompted to busybox shell. 
2021-12-05 10:40:49 <ahmadraniri1994[m]> Assuming I use other distro and want to bootstrap alpine to a partition then chroot and configure it until I can reboot it. Thanks.
2021-12-05 10:45:57 <ikke> ahmadraniri1994[m]: easiest is to just extract the minirootfs: https://dl-cdn.alpinelinux.org/alpine/latest-stable/releases/x86_64/alpine-minirootfs-3.15.0-x86_64.tar.gz
2021-12-05 10:51:36 <ahmadraniri1994[m]> <ikke> "ahmadraniri1994: easiest is to..." <- Thanks,  I'll check it, one more question, can we use one partition (only root, no swap, boot isn't separated from root) instead of 3 (root, boot, swap) ?
2021-12-05 10:51:39 <ikke> You could also grab a static version of apk-tools and run `apk.static add -X https://dl-cdn.alpinelinux.org/alpine/latest-stable/main --allow-untrusted --initdb --root path/tho/chroot alpine-base
2021-12-05 10:53:48 <ahmadraniri1994[m]> ikke: Yeah, I Done that, but for advance configurations, I still feel confuse.
2021-12-05 12:19:04 <telmich> Something is wrong with the 3.15.0 x86_64 image
2021-12-05 12:19:10 <telmich> It does not detect the network devices anymore
2021-12-05 12:19:14 <ikke> hmm
2021-12-05 12:19:31 <telmich> Testing on an APU2 (pcengines)
2021-12-05 12:19:34 <telmich> Only lo exists
2021-12-05 12:20:06 <telmich> oh... or there is an error with the writing
2021-12-05 12:20:13 <telmich> Just seen: [   16.056331] SQUASHFS error: Unable to read metadata cache entry [74055fa]
2021-12-05 12:21:15 <telmich> Testing with a different usb stick, even though it was written with sync
2021-12-05 12:21:31 <telmich> that works - so disregard my comment :-)
2021-12-05 12:22:17 <ikke> good :-)
2021-12-05 12:30:57 <telmich> It would be great if the installer inherited the console=ttyS0,115200 command line after installation
2021-12-05 12:31:35 <nero> ... im pretty sure it does
2021-12-05 12:32:08 <telmich> ... well, I don't see any output after rebooting after the install
2021-12-05 12:32:30 <telmich> ...double checking, maybe it was just the quiet parameter...
2021-12-05 12:32:51 <telmich> No, negative.
2021-12-05 12:33:06 <nero> ah, you mean, the resulting system should also have that in its own cmdline?
2021-12-05 12:33:11 <nero> so you get kernel output there?
2021-12-05 12:33:41 <telmich> So I have to stop the installer and add console=ttyS0,115200 to the kernel before starting the installer
2021-12-05 12:33:51 <telmich> after the installation is done, the kernel command line is:
2021-12-05 12:33:59 <telmich> .linux vmlinuz-lts root=UUID=492d94ca-7acc-47b9-a00b-cf20f587acfa modules=sd-mod,usb-storage,ext4 quiet rootfstype=ext4 initrd=initramfs-lts
2021-12-05 12:37:47 <nero> setting the kernel cmdline is normally part of the setup-disk step
2021-12-05 12:39:42 <telmich> I was using setup-alpine today, i.e. almost completely automatic
2021-12-05 12:43:49 <nero> setup-disk cant just read it from /proc/cmdline, since thats the host system, which can be any alpine system, not the installer
2021-12-05 12:44:36 <telmich> Let me rephrase:
2021-12-05 12:44:52 <telmich> it would be great if there was a way to automate the serial console based installations even further
2021-12-05 12:45:52 <nero> use KERNELOPTS as listed in the setup-disk help, or use an pre-generated apkovl
2021-12-05 12:47:17 <telmich> nero: I was never successful with an apkovl, do you have a pointer on what/how I'd easiest take an up-to-date installer and apply the changes?
2021-12-05 12:48:44 <telmich> also: I guess everyone is aware that the installer is doing a great job, my comments are just about further improving - installation takes even manually less than 10m
2021-12-05 12:50:29 <nero> i have a set of makefiles here that i used to generate apkpvl's from scratch
2021-12-05 13:14:46 <ktprograms> ahmadraniri1994[m]: You can use `lbu package - | tar -C /mnt -xz` (assuming /mnt is the root partition), then `apk add --root=/mnt --initdb apk-tools alpine-baselayout alpine-conf alpine-base openssh openssl e2fsprogs`. You can just use the single partition if your platform has a way to boot a kernel without any bootloader (for example with qemu -kernel option). I found that it's simpler to use a 
2021-12-05 13:14:52 <ktprograms> kernel config that just disables the initramfs and you pass root=/dev/vda to the kernel command line (assuming no partition so it's just the full vda).
2021-12-05 13:46:00 <donoban> Hi, I still getting an upgradable package from 'apk list -u' -> "libcrypto3-3.0.0-r4 x86_64 {openssl3} (Apache-2.0) [upgradable from: libcrypto3-3.0.0-r3]" but doing 'apk upgrade -Ua' doesn't upgrade it
2021-12-05 14:01:53 <ikke> donoban: that means there is a package that is holding it back
2021-12-05 14:02:40 <ikke> donoban: try apk add libcrypto3=3.0.0-r4
2021-12-05 14:15:45 <lopid> when updating from 3.14 to 3.15 (php7 to php8), the symlink from /usr/bin/php disappeared
2021-12-05 14:16:35 <donoban> (1/2) Upgrading libcrypto3 (3.0.0-r3 -> 3.0.0-r4)
2021-12-05 14:16:38 <donoban> (2/2) Replacing libssh (0.9.6-r1 -> 0.9.6-r1)
2021-12-05 14:18:24 <donoban> Now I removed it from world and 'apk fix' purged it
2021-12-05 14:20:54 <donoban> oh!
2021-12-05 14:21:19 <donoban> ikke: the problem is a custom packaga that was build before last ssl3 change
2021-12-05 14:21:41 <donoban> s/packaga/package
2021-12-05 14:21:41 <alpine-meetbot> donoban meant to say: ikke: the problem is a custom package that was build before last ssl3 change
2021-12-05 14:21:50 <donoban> all fine thx
2021-12-05 14:23:15 <ikke> donoban: alright
2021-12-05 14:45:55 <amk> is there a good way to also automaticaly install -doc packages when installing a package?
2021-12-05 14:46:25 <ikke> apk add docs
2021-12-05 15:32:22 <lopid> "Your configuration in /etc/doas.conf was migrated to…" misleading—i didn't have a /etc/doas.conf
2021-12-05 15:33:40 <lopid> s/mis… oh no, i'm not triggering that thing. i meant it's confusing more than misleading
2021-12-05 15:36:45 <minimal> lopid: yeah it displays that on a new install, not just on update, I've been intended to raise a MR to fix it but never go around to it :-)
2021-12-05 15:39:51 <lopid> might have been prudent to use /etc/doas.d/01-doas.conf or 99-doas.conf, depending on precedence 
2021-12-05 15:40:06 <lopid> ah, 01
2021-12-05 15:40:50 <ikke> The doas.conf config is so simple that I don't think it should matter too much
2021-12-05 15:47:54 <lopid> how long is "some time" for the persist option?
2021-12-05 15:49:00 <minimal> at least twice as long as "a short time" ;-)
2021-12-05 15:49:20 <ikke> I have a feeling it's not working for some reason
2021-12-05 15:50:00 <lopid> persist is working for me (for now)
2021-12-05 15:50:37 <minimal> I'm failrly sure it was working about 1 month ago when I was adding (and testing) doas support to cloud-init
2021-12-05 15:53:51 <lopid> hope you don't introduce a syntax error in that file if you don't have sudo already installed
2021-12-05 15:54:32 <ikke> Hope you have a root password and don't solely rely on doas for privilege escalation
2021-12-05 15:55:20 <lopid> hope you have access to console in that case
2021-12-05 15:58:06 <minimal> hope Santa brings me a new PC as I've been a good boy this year :-)
2021-12-05 15:58:49 <minimal> "Alpine episode 3.15, A New Hope" :-)
2021-12-05 15:59:06 <ikke> :D
2021-12-05 16:00:36 <lopid> is there no equivalent protection that visudo offered?
2021-12-05 16:01:00 <lopid> 
ACTION writes a vim autocmd for doas.conf

2021-12-05 16:01:07 <ikke> There is no equivalent, no
2021-12-05 16:03:54 <minimal> conversely doas is smaller (and less functional) than sudo and so theoretically more secure (in the sense of less code to go wrong)
2021-12-05 16:04:36 <lopid> human typo
2021-12-05 16:05:08 <minimal> personally I'd get my stunt-double to write doas config files for me :-)
2021-12-05 16:05:18 <minimal> lopid: typo?
2021-12-05 16:05:59 <lopid> the code for doas can be as small and secure as possible but all it takes is for someone to typo its config
2021-12-05 16:06:35 <ikke> That's a better failure mode than a security vulnerability that gives everyone access :-)
2021-12-05 16:07:04 <minimal> in that scenario is would be very secure with a typo as then no-one could get in :-)
2021-12-05 16:07:16 <lopid> indeed
2021-12-05 16:09:20 <minimal> as ikke pointed out that's where something like a root password usable only via console, or from certain IP addresses, etc or some other out-of-band mechanism comes in to fix the typo
2021-12-05 16:10:24 <lopid> i am merely lamenting the loss of a feature that would have prevented that situation in the first place
2021-12-05 16:17:14 <Habbie> there is doas -C, so vidoas could be a 10 line shell script
2021-12-05 16:17:51 <lopid> why 10?
2021-12-05 16:20:00 <lopid> persist has timed out :(
2021-12-05 16:20:23 <lopid> long live sudo! heh
2021-12-05 16:50:03 <Habbie> lopid, because it's a nice and round number, i don't think 3 would cut it, and if it turns out to be 15, i wasn't off by too much ;)
2021-12-05 16:59:32 <psykose> 15 is an even number
2021-12-05 17:02:01 <vlp[m]> Hello! 
2021-12-05 17:02:39 <vlp[m]> Quick question: does the Grub2 support luks2 in the last Iso of Alpine 
2021-12-05 17:04:18 <telmich> I've got another funky issue today: trying to install / build python-ldap via pip fails with an ld/gcc error:   /usr/lib/gcc/x86_64-alpine-linux-musl/11.2.1/../../../../x86_64-alpine-linux-musl/bin/ld: cannot find -lldap_r
2021-12-05 17:04:30 <telmich> I wonder if anyone else has seen this before? To me this is new
2021-12-05 17:04:45 <MaximKarasev[m]> vlp[m]: It should.
2021-12-05 17:05:07 <telmich> Ah, seems to be known for 3.15: https://github.com/python-ldap/python-ldap/issues/445
2021-12-05 17:05:38 <vlp[m]> MaximKarasev[m]: Thanks!
2021-12-05 17:24:46 <blt[m]> <MaximKarasev[m]> "It should." <- As well for /boot ? It is a nice enhancement and is part of recent Grub2 release 
2021-12-05 17:25:40 <frag> getting some error when playing sound after upgrade/reboot : alpine linux [ao/alsa] pcm write error: Broken pipe[ao/alsa] Error writing audio to device. .... sounds plays but kinda like a bad record player
2021-12-05 17:26:42 <minimal> vlp[m]: Grub 2.06, which is in Alpine 3.15 and Edge does support LUKS2. However I have found issues with its LUKS2 support - grub-mkconfig does not create correct grub.cfg ("insmod lvm2" and some other stuff is miss).
2021-12-05 17:26:54 <minimal> s/miss/missing/
2021-12-05 17:26:54 <alpine-meetbot> minimal meant to say: vlp[m]: Grub 2.06, which is in Alpine 3.15 and Edge does support LUKS2. However I have found issues with its LUKS2 support - grub-mkconfig does not create correct grub.cfg ("insmod lvm2" and some other stuff is missing).
2021-12-05 17:29:34 <blt[m]> minimal: does it work with /boot on the same portion as / or it have to be on a separate partition? In the past because of Grub limitation to work with Luks it was required to have separate partition it would be much better to avoid that but not sure the 3.15 script support that
2021-12-05 17:29:42 <frag> in qutebrowser i get "[4502:4654:1205/182838.201214:ERROR:bus.cc(393)] Failed to connect to the bus: /usr/bin/dbus-launch terminated abnormally without any error message"
2021-12-05 17:31:57 <minimal> blt[m]: the whole point of LUKS support in Grub (whether v1 or v2) is to support encrypted /boot (whether separate partition or part of root partition)
2021-12-05 17:32:30 <minimal> however as I said it seems that Grub 2.06's LUKS2 support is buggy as grub-mkconfig does not create a valid grub.cfg in that situation
2021-12-05 17:32:56 <blt[m]> minimal Even if you do it manually?
2021-12-05 17:33:03 <minimal> I started to look into it but didn't have the time to figure out what patch (and upstream fix) would be needed for it to work as expected
2021-12-05 17:33:36 <blt[m]> Got it , so it may break the script I guess
2021-12-05 17:33:42 <minimal> blt[m]: manually? you mean after running grub-install? (as, from memory, that calls grub-mkconfig)
2021-12-05 17:34:50 <blt[m]> minimal: was more thinking about the 3.15 scripts where I assume it is calling the same commands (but I am writing without testing)
2021-12-05 17:35:51 <minimal> blt[m]L If you call grub-install with --install-modules="xxxxx" and *add* "gcry_rijndael gcry_sha256 luks2" to the list of modules that may *possibly* fix it
2021-12-05 17:36:14 <minimal> but not 100% sure, would have to look at it again to refresh my memory
2021-12-05 17:38:30 <minimal> blt[m]: I did some testing when Grub 2.06 as added to Alpine Edge and it wasn't working for LUKS2 then. I could retest very quickly (as have it scripted) but not expecting it to have magically fixed itself since
2021-12-05 18:08:04 <minimal> blt[m]: just quickly build a LUKS2 machine and the generated grub.cfg file is missing insmod entries for cryptodisk, gcry_rijndael, gcry_sha256, and luks2 as well as missing a line for "cryptomount -u "
2021-12-05 18:08:43 <blt[m]> Thanks to having test that
2021-12-05 18:09:01 <blt[m]> So it confirms what you mentioned
2021-12-05 18:09:26 <minimal> that's the issue, whereas if I build a LUKS1 machine then the equivalent entries are added as expected.
2021-12-05 18:10:16 <psykose> that sounds like the issue behing 'grub doesnt support luks2' ive been hearing for like 3 years
2021-12-05 18:10:21 <psykose> i'm surprised nobody has fixed that yet
2021-12-05 18:10:24 <minimal> I believe there's a "bug" in one of Grub's shellscripts and maybe else in Grub also, I just have not had enough time to track it down
2021-12-05 18:11:38 <minimal> psykose: not quite, the support was only added to Grub 2.06 (which only came out 1? 2? months ago) but it seems the scripts bundled aren't magically setting it up right
2021-12-05 18:11:50 <psykose> ahh
2021-12-05 18:11:54 <psykose> that makes more sense
2021-12-05 18:11:59 <minimal> if you were to create your own grub.cfg manually then I guess it would work as expected
2021-12-05 18:12:43 <minimal> I *think* I noticed someone raised it as an issue on the Grub mailing list shortly after 2.06 came out
2021-12-05 18:14:03 <psykose> that's good
2021-12-05 18:17:31 <minimal> psykose: trying to find evidence of it lol
2021-12-05 18:17:56 <minimal> also with LUKS2 Grub doesn't support Argon2i so you've have to use pbkdf2 as well
2021-12-05 18:20:15 <minimal> I've used LUKS2 fine on a none boot partition (i.e. encrypted root, unencrypted /boot), its the encrypted partition with /boot that's not right
2021-12-05 18:28:31 <elibrokeit> LUKS2 support is only added to the cryptomount module etc. -- it's not added to the detection routines for grub-install, so you need to assemble your own .img to make sure it is actually added into the stub loader
2021-12-05 18:29:12 <elibrokeit> it may also not be added to the grotty shellscripts powering grub-mkconfig, but I shun those on principle so idk
2021-12-05 19:09:43 <frag> anyone else with intel nuc? does alsa work after upgrade/reboot?
2021-12-05 19:28:21 <frag> can i list/install older versions of alsa on 3.15 repo?
2021-12-05 19:29:19 <minimal> elibrokeit: I'd expect grub-install/grub-mkconfig to behave the same for both LUKS1 and LUKS2, but that would be too sensible
2021-12-05 19:37:37 <elibrokeit> well, the good news is the lack of sense is only a matter of "we half implemented this, only the plumbing actually works"
2021-12-05 20:11:40 <frag> downgrading the system to 3.14 worked :{
2021-12-05 20:13:43 <lopid> is that process just the same as updating to a new version?
2021-12-05 20:13:56 <ikke> yes, make sure to include --available
2021-12-05 20:14:36 <frag> doas apk update && doas apk upgrade --update-cache --available
2021-12-05 20:14:54 <frag> the times ive done it, it works suprisingly well 
2021-12-05 20:42:57 <vlp[m]> minimal: I have started an E2E encrypted install with luks1 (I'll retry with luk2 after). And whent I launch `setup-disk -m sys /mnt` There is a message: "failed to find sysfs mount point" and at the end "you might need fix the MNR to be able to boot"
2021-12-05 21:45:41 <frag> upgraded again and rebooted and it still works :S
2021-12-05 21:45:44 <frag> jaja...
2021-12-05 22:12:25 <tobtobxx> vlp[m]: The "you need to fix the MBR" is normal (the wiki tells you), but the sysfs mount point... do you have sysfs mounted?
2021-12-05 22:13:16 <vlp[m]> I have mounted root into /mnt 
2021-12-05 22:13:42 <vlp[m]> (I'll check the wiki for the MBR point! Thanks) 
2021-12-05 22:14:37 <tobtobxx> About the MBR: You don't need to do anything, just continue on with the wiki guide.
2021-12-05 22:15:11 <tobtobxx> Do you have BIOS or UEFI?
2021-12-05 22:15:41 <vlp[m]> UEFI 
2021-12-05 22:15:51 <vlp[m]> tobtobxx: OK great 
2021-12-05 22:16:15 <tobtobxx> Yeah, then you shuold've mounted the sysfs als slave from /sys to /mnt/sys before chrooting: https://wiki.alpinelinux.org/wiki/LVM_on_LUKS#Grub_with_UEFI
2021-12-05 22:17:54 <vlp[m]> Yes but the setup-disk comes before into the wiki . So, I, in deed, didn't mount it 
2021-12-05 22:19:06 <tobtobxx> You're right, I got confused. I'd probably just continue then.
2021-12-05 22:19:28 <vlp[m]> I'll continue and see :) 
2021-12-05 22:19:31 <tobtobxx> I can't remember if I've had this message too.
2021-12-05 22:19:57 <vlp[m]> Thanks! 
2021-12-06 01:21:22 <joseaAlvarez> Hi everyone, good afternoon. I want to place alpine in azure, using this guide: https://wiki.alpinelinux.org/wiki/Replacing_non-Alpine_Linux_with_Alpine_remotely all perfect, only after choosing the disk as sys I get the following error: How would you like to use it? ('sys', 'data', 'crypt', 'lvm' or '?' for help) [?] sys ERROR: Failed to create lib / libblkid.so.1.1.0: Connection aborted ERROR: libblkid-2.37.2-r1: IO ERROR
2021-12-06 01:24:05 <Hello71> check dmesg
2021-12-06 01:25:36 <minimal> psykose: this seems to be the most complete summary of the GRUB LUKS2 workarounds: https://aur.archlinux.org/packages/grub-git/?O=10&PP=10#comment-769218
2021-12-06 01:26:40 <ahmadraniri1994[m]> <ktprograms> "ahmadraniri1994: You can use `..." <- Noted, I'll try it thanks.
2021-12-06 01:27:18 <ktprograms> ahmadraniri1994[m]: Are you trying to do it on hardware or in a VM?
2021-12-06 01:27:43 <ktprograms> Because I think the single partition/boot a kernel without a bootloader would only be possible in a VM.
2021-12-06 01:28:56 <psykose> minimal: i see, good to place in the back of my head for reference, thanks!
2021-12-06 01:29:14 <joseaAlvarez> on a VM, initially with ubuntu and I want to convert it to alpine. with the guide mentioned before.
2021-12-06 01:30:07 <minimal> psykose: I'd still like to workout patches for Alpine grub package to fix it, that'll be after the work I'm doing currently with dhcpcd :-)
2021-12-06 01:30:22 <psykose> having fun i see :p
2021-12-06 01:30:57 <minimal> psykose: "fun" with changing pid filename depending on number of interfaces present, yes lol
2021-12-06 01:31:10 <ahmadraniri1994[m]> ktprograms: For now I'll practice on vm.
2021-12-06 01:31:23 <mtm84[m]> How about a rage128 KMS driver for “fun” XD
2021-12-06 01:32:41 <minimal> ahmadraniri1994[m]: single partition works fine in general if you are using BIOS. For UEFI typically you need at least 2 partitions
2021-12-06 01:33:01 <ahmadraniri1994[m]> <ktprograms> "Because I think the single..." <- How about I configure the bootloader? I found a section in alpline wiki which explain it. I tried use syslinux before boot unable to boot, but grub did. I'll dig it more.
2021-12-06 01:33:39 <ahmadraniri1994[m]> minimal: Yeah, we need EFI partition, right? 
2021-12-06 01:33:49 <minimal> yes
2021-12-06 01:34:33 <ktprograms> ahmadraniri1994[m]: I was thinking use QEMU with -kernel but that can't be replicated on hardware, so you will need at least the boot partitions+bootloader.
2021-12-06 01:36:34 <minimal> ktprograms: for BIOS a single partition can hold everything, for UEFI you need a FAT-formatted ESP partition and then can have a 2nd ext4/xfs/brtfs/etc formatted partition for everything else
2021-12-06 01:37:39 <ktprograms> minimal: I see.
2021-12-06 01:44:16 <minimal> psykose: also more discussion here: https://www.reddit.com/r/archlinux/comments/o1opyj/psa_grub_fails_to_detect_the_fact_that_and_boot/
2021-12-06 01:44:17 <alpine-meetbot> [REDDIT] PSA: GRUB fails to detect the fact that / and /boot are LUKS encrypted (self.archlinux) | 18 points (0385.0%) | 13 comments | Posted by aperture_synce | Created at 2021-06-17 - 04:44:58UTC
2021-12-06 01:47:27 <joseaAlvarez> Help! 
2021-12-06 01:47:29 <joseaAlvarez>  How would you like to use it? ('sys', 'data', 'crypt', 'lvm' or '?' for help) [?] sys
2021-12-06 01:47:33 <joseaAlvarez>  ERROR: Failed to create lib / libblkid.so.1.1.0: Connection aborted 
2021-12-06 01:47:39 <joseaAlvarez>  ERROR: libblkid-2.37.2-r1: IO ERROR
2021-12-06 01:48:13 <joseaAlvarez> :'(
2021-12-06 01:58:17 <ktprograms> joseaAlvarez: Did you run `dmesg` like Hello71 told you to? (just type dmesg at the sh prompt)
2021-12-06 02:36:43 <joseaAlvarez> Yes, but what part of dmesg am I looking for?
2021-12-06 02:38:40 <ktprograms> joseaAlvarez: Anything that looks like an error, should be within the last few lines, so use `dmesg | tail`
2021-12-06 02:40:00 <dick> hi
2021-12-06 02:40:08 <joseaAlvarez> no, nothing related, but I try again and check.
2021-12-06 02:41:01 <dick> hi guys, i'm trying to use the new crypt feature on a sys installation of alpine but it failed last time. has anyone got any tips? using an nvme btw.
2021-12-06 02:46:16 <dick> getting the old mount: mounting UUID on /sysroot failed: no such file or directory. then i am dropped into the emergency shell
2021-12-06 02:47:58 <joseaAlvarez> here the result:  https://pastebin.com/YnHc3TyL
2021-12-06 02:53:19 <minimal> joseaAlvarez: what happens if you run "blkid" manually?
2021-12-06 02:54:58 <joseaAlvarez> here:  https://pastebin.com/MksyN1d8
2021-12-06 02:55:15 <joseaAlvarez> Thank you! I appreciate your support
2021-12-06 02:56:19 <ktprograms> joseaAlvarez: Did you try apk add libblkid?
2021-12-06 02:57:20 <joseaAlvarez> Hi, no, but I'll try it right away, one moment please.
2021-12-06 02:59:52 <minimal> joseaAlvarez: so is libblkid package installed? if it is installed then maybe try "apk fix libblkid"?
2021-12-06 03:06:05 <joseaAlvarez> https://pastebin.com/qcjRBFVr
2021-12-06 03:06:09 <joseaAlvarez> none :(
2021-12-06 03:06:57 <ktprograms> joseaAlvarez: Can you try installing any other package (for example apk add htop)?
2021-12-06 03:07:29 <minimal> joseaAlvarez: is your network connection actually working? you cannot "apk add" if it is not
2021-12-06 03:07:30 <joseaAlvarez> no, same result
2021-12-06 03:07:54 <joseaAlvarez> yes is working
2021-12-06 03:08:05 <joseaAlvarez> i have a meet yf you like to see direct
2021-12-06 03:08:06 <joseaAlvarez> https://meet.google.com/sew-mboq-bhd
2021-12-06 03:08:17 <joseaAlvarez> i speak spanish.
2021-12-06 03:08:31 <ktprograms> joseaAlvarez: same result as in IO error?
2021-12-06 03:08:52 <joseaAlvarez> yes IO Error, but ping works
2021-12-06 03:08:58 <minimal> "ping dl-cdn.alpinelinux.org" ? "curl dl-cdn.alpinelinux.org"? (or wget instead of curl)
2021-12-06 03:09:33 <joseaAlvarez> 64 bytes from 151.101.250.133: seq=0 ttl=55 time=0.703 ms
2021-12-06 03:09:55 <minimal> and wget/curl?
2021-12-06 03:10:02 <joseaAlvarez> yes
2021-12-06 03:10:05 <psykose> it looks like you just can't write to the disk
2021-12-06 03:10:20 <ktprograms> joseaAlvarez: Try running 'curl -o test https://google.com'
2021-12-06 03:10:34 <joseaAlvarez> localhost:~# wget dl-cdn.alpinelinux.org Connecting to dl-cdn.alpinelinux.org (151.101.250.133:80) saving to 'index.html' index.html           100% |********************************|   247  0:00:00 ETA 'index.html' saved
2021-12-06 03:11:02 <joseaAlvarez> dont have curl, only wget
2021-12-06 03:11:14 <ktprograms> joseaAlvarez: Nevermind, it shows that it can write to disk
2021-12-06 03:12:00 <joseaAlvarez> yes i can create a file with vi.
2021-12-06 03:13:03 <minimal> is libblk package already installed? then try "apk fix libblk"
2021-12-06 03:13:10 <minimal> oops libblkid
2021-12-06 03:13:10 <joseaAlvarez> but i am running from ram, 
2021-12-06 03:13:22 <joseaAlvarez> as mentioned in the guide:
2021-12-06 03:13:33 <joseaAlvarez> Reboot, and login again from vnc with root. Alpine is now running from ram, and should have mounted /dev/sda on /media/sda, paste it into ram.
2021-12-06 03:13:39 <joseaAlvarez> https://wiki.alpinelinux.org/wiki/Replacing_non-Alpine_Linux_with_Alpine_remotely
2021-12-06 03:13:43 <ktprograms> joseaAlvarez: How much RAM do you have in the VM?
2021-12-06 03:13:52 <joseaAlvarez> 512Mb
2021-12-06 03:13:52 <ktprograms> Is it running out of ram disk space?
2021-12-06 03:14:10 <minimal> ok, so then when you first created your "run from ram" setup you needed to ensure that libblkid was installed
2021-12-06 03:15:04 <ktprograms> minimal: But I thought joseaAlvarez was trying to use sys in setup-disk?
2021-12-06 03:15:24 <joseaAlvarez> mmmm im running the command  in ubuntu dd if=alpine-virt-3.5.2-x86_64.iso of=/dev/sda
2021-12-06 03:15:39 <joseaAlvarez> to flash the iso in hdd
2021-12-06 03:16:34 <joseaAlvarez> yes, I want to use sys.
2021-12-06 03:16:58 <joseaAlvarez> y want to: Replacing non-Alpine Linux with Alpine remotely
2021-12-06 03:17:11 <joseaAlvarez> im trying on azure.
2021-12-06 03:23:23 <joseaAlvarez> or some other way to be able to use alipne on azure? :'(
2021-12-06 03:48:11 <joseaAlvarez> blkid :
2021-12-06 03:48:53 <joseaAlvarez> -  /dev/sdb1: UUID="cd181cd0-ddce-459e-a0aa-e93d872628c3" TYPE="ext4"
2021-12-06 03:49:07 <joseaAlvarez> - /dev/sda2: UUID="4CEA-718F" TYPE="vfat"
2021-12-06 03:49:16 <joseaAlvarez> - /dev/sda2: UUID="4CEA-718F" TYPE="vfat"
2021-12-06 03:49:30 <joseaAlvarez> - /dev/sda2: UUID="4CEA-718F" TYPE="vfat"
2021-12-06 06:39:27 <ktprograms> Hi, how do I run rootless wayland? There doesn't seem to be much documentation around wayland so I'm not even sure if I should use dbus-run-session or the wm specific command.
2021-12-06 07:37:54 <consus[m]> You just run the binary
2021-12-06 07:38:13 <consus[m]> Like sway or labwc or <wmname>
2021-12-06 07:40:55 <ktprograms> consus[m]: Do I need to prefix it with doas?
2021-12-06 07:59:34 <consus[m]> Nope
2021-12-06 08:01:23 <ktprograms> consus[m]: What packages do I need to install to get weston working?
2021-12-06 08:01:29 <consus[m]> Hmm
2021-12-06 08:01:37 <consus[m]> Actually don't know, never used weston
2021-12-06 08:15:58 <ktprograms> consus[m]: With running sway, I get an error 'Could not get primary session for user: No data available' and 'No backend was able to open a seat' and 'Unable to create seat: Function not implemented'
2021-12-06 08:16:43 <consus[m]> I believe you need some sort of session manager running
2021-12-06 08:17:01 <consus[m]> I run sway on archlinux so I have systemd doing that
2021-12-06 08:17:40 <consus[m]> AFAIR elogind should do the trick here
2021-12-06 08:18:09 <ktprograms> consus[m]: How would I use it?
2021-12-06 08:18:24 <consus[m]> However I've switched away from OpenRC a couple of years ago so my knowledge is kinda dated
2021-12-06 08:18:43 <consus[m]> Try Gentoo wiki
2021-12-06 08:36:07 <ktprograms> consus[m]: Turns out I needed to install mesa-dri-gallium (I'm using a VM)
2021-12-06 08:38:51 <consus[m]> Great!
2021-12-06 09:14:33 <gijsboer> hello everyone, hoping to find some help here regarding installation of npm/nodejs on Alpine 3.15. Right now, when I install npm v8.1.3-r0, I get nodejs-current-17.2.0-r0 for 'free'. I'd like to stick with Node v16 for now, since that's the LTS version. For that, I need to first run 'apk -U add nodejs=16.13.1-r0' and then follow by 'apk -U add npm=8.1.3-r0'. I can't do this in one command, because nodejs-curre
2021-12-06 09:14:33 <gijsboer> nt apparently conflicts with the nodejs package, since they are on a different version. Is this intended behavior? It seems kind of odd to me, that I need two different commands and I can't combine the two.
2021-12-06 10:48:30 <Insmonia> Hey there
2021-12-06 10:49:05 <Insmonia> Is there any support for ksmbd in alpine linux yet? the ksmbd-tools are available in the community repo, but the kernel module itself seems to be missing. thanks for your help! 😊 
2021-12-06 10:56:51 <ktprograms> Insmonia: Install either linux-edge or linux-edge4virt
2021-12-06 10:59:20 <frag> anyone using the surf browser in repo? _can_ it donwload files from net-banks/places you have to log in?
2021-12-06 11:36:38 <ikke> gijsboer: `apk add npm nodejs` works for me
2021-12-06 11:56:17 <gijsboer> hmm yeah.. weird. For some reason it wasn't working for me. I'll have to see why :S
2021-12-06 13:30:01 <Insmonia> thx @ktprograms
2021-12-06 13:48:43 <ktprograms> I'm trying to run waydroid, and it seems to have started properly (doas waydroid shell works) but when running waydroid show-full-ui, it has the error message 'Failed to get service waydroidplatform, trying again...'. I looked at the relevant code, and it seems like it's using a gbinder.ServiceManager initialized with /dev/binder and calling the get_service_sync('waydroidplatform') function, but 
2021-12-06 13:48:49 <ktprograms> the remote object doesn't seem to be returned.
2021-12-06 14:22:18 <hugorodrigues> hello all, I've a quick question regarding a package name. I'm packaging crossplane that has both a Python3 module and application. Should I just call it crossplane or py3-crossplane?
2021-12-06 14:43:25 <mercenary> scapy is just called 'scapy'
2021-12-06 14:59:11 <hugorodrigues> will use crossplane then, thank you
2021-12-06 19:01:52 <jhjacobs81> hmz
2021-12-06 19:26:14 <helby> welcome matrix users ;/
2021-12-06 19:26:35 <helby> it's like raid
2021-12-06 19:27:53 <bdju> yeah
2021-12-06 19:28:36 <helby> still no idea what's the point of bridge
2021-12-06 19:30:40 <jhjacobs81> matrix.. *scratches head* im too old for all these modern whattamathingy's
2021-12-06 19:31:35 <helby> I think matrix was probably the worst thing for communication I ever tried
2021-12-06 19:33:17 <jhjacobs81> :)
2021-12-06 19:33:45 <jhjacobs81> once i know enough Rust i still want to make a XMPP client
2021-12-06 19:34:01 <helby> yes, I know one person, he use xmpp ;/
2021-12-06 19:34:09 <nero> helby: have you tried Atlassian HipChat
2021-12-06 19:34:17 <helby> nero: no ;/
2021-12-06 19:34:36 <hoel> jhjacobs81: isn’t Rust a modern whattamathingy? XD
2021-12-06 19:35:00 <nero> helby: i think microsoft lynx is also worse
2021-12-06 19:35:19 <helby> nero: well, who will use that ;/
2021-12-06 19:35:22 <helby> voluntarily
2021-12-06 19:35:43 <helby> it's like these google servie, 10 things for communicating, not one works well
2021-12-06 19:35:54 <helby> all kidn of meeting, duo, hangouts and whatever they hve now
2021-12-06 19:36:52 <jhjacobs81> hoel true, true! but every now and then i try some new whattamathingy's, just to prove that im too old for this shit :D
2021-12-06 19:39:18 <bdju> I like xmpp and matrix, but I love irc. the other things just don't have clients as good as irssi
2021-12-06 19:39:56 <bdju> (I know of the weechat matrix thing, but it was a bit buggy last time I used it)
2021-12-06 19:41:20 <jhjacobs81> i used to be using FOSS clients, then when i moved to macos a decade ago i went with.. err.. that one famous thingamabob.. and now im on TheLounge.. just because im too curious to log off
2021-12-06 19:42:05 <helby> u selfhost lounge?
2021-12-06 19:42:21 <jhjacobs81> yupyup
2021-12-06 19:42:49 <helby> btw. modern irc servers provide history/always-on mode
2021-12-06 19:43:19 <helby> jhjacobs81: not better to run just client on that server? some kind of terminal client I mean ;/
2021-12-06 19:44:48 <jhjacobs81> wwweeeeeelllll... i like web apps :) and it supports multi user :)
2021-12-06 19:45:09 <jhjacobs81> terminals on phones suck.. thelounge.. it works okay'ish
2021-12-06 19:45:25 <helby> is there a app on phone or just web?
2021-12-06 19:46:25 <jhjacobs81> for thelounge? its just web
2021-12-06 19:46:49 <jhjacobs81> foe my phone there's a terminal app that i sometimes use in emergency situations.. but i prefer not to :)
2021-12-06 19:50:41 <helby> u said, terminals on pohones sucks
2021-12-06 19:50:45 <helby> phones
2021-12-06 19:50:59 <jhjacobs81> they do
2021-12-06 19:51:21 <jhjacobs81> and i know this, because i do ;-)
2021-12-06 19:53:22 <bdju> terminals on phones are not too bad, imo. they are worse of course than at a real computer, but I run irssi in tmux and I often pick up a chat from my phone
2021-12-06 19:57:54 <helby> what is not worse on phone than on computer anyway
2021-12-06 20:01:12 <jhjacobs81> fair enough
2021-12-06 20:01:28 <clandmeter> making phone calls
2021-12-06 20:01:35 <jhjacobs81> on my ipad i actually enjoy using a terminal :)
2021-12-06 20:04:31 <helby> clandmeter: nobody asked you! ;/ continue to work in council, tsc and whatever other system structures exists ;/ I bet there is ton of work ;/
2021-12-06 20:05:55 <clandmeter> and dont forget shooting photos :p
2021-12-06 20:14:48 <jhjacobs81> photos? of what?
2021-12-06 20:46:50 <helby> he has probably wife ;/
2021-12-06 20:46:58 <helby> maybe will share, will see ;/
2021-12-06 20:47:13 <jvoisin> wtf.
2021-12-06 20:47:47 <psykose> are you ok
2021-12-06 20:48:44 <helby> some crime or what?
2021-12-06 21:23:37 <telmich> I think I have asked before, but do we already have a "standard way" to execute `mount --make-shared /` and/or to mark it as shared on boot?
2021-12-06 21:24:16 <telmich> Background: quite some kubernetes workloads like node_exporter from the prometheus-operator or kubevirt require the path to be shared
2021-12-06 21:24:43 <telmich> I can obviously create my own script and solve it, but eventually more and more people will run into this
2021-12-07 00:38:03 <muirrum> howdy, apologies if this is the wrong channel, but I'm trying to put a draft under my user page on the wiki and it got blocked for spam, saying that I needed to talk to an administrator. Screenshot of error:  https://webchat.devcara.com/uploads/e24f7c3489f64ceb/image.png 
2021-12-07 00:40:04 <helby> not so long ago I tried to register on wiki, but tried like 10 times and did not get thru captcha ;/
2021-12-07 00:42:20 <helby> https://wiki.alpinelinux.org/wiki/Special:ListUsers/sysop
2021-12-07 00:42:33 <helby> clandmeter has nothing to do, he will fix it
2021-12-07 02:41:53 <ktprograms> Hi, any idea why adding elogind to the default runlevel causes plasma to have its display messed up (it's like squished to half the display and when I set it back with xrandr then everything is very laggy)? I need to run elogind so that sddm/plasma can have a way to shut down/reboot (They don't seem to use /sbin/poweroff and /sbin/reboot)
2021-12-07 04:03:21 <nickname> hello
2021-12-07 04:03:32 <Guest7836> hllo
2021-12-07 04:03:59 <Guest7836> a question regard using musl to compile static binary
2021-12-07 04:04:19 <Guest7836> i run into this problem when check for dependacy of the finished build executable
2021-12-07 04:04:59 <Guest7836> ldd /path/binary 	/lib/ld-musl-x86_64.so.1 (0x7fb5eec57000) 	libstdc++.so.6 => /usr/lib/libstdc++.so.6 (0x7fb5ee918000) 	libgcc_s.so.1 => /usr/lib/libgcc_s.so.1 (0x7fb5ee8fe000) 	libc.musl-x86_64.so.1 => /lib/ld-musl-x86_64.so.1 (0x7fb5eec57000)
2021-12-07 04:05:40 <Guest7836> https://paste.linux.chat/?c88f4bd07118bd14#6nUjWXwfcAd4ZsAtmMUTEj6EzhaXRMQDqnUav86Kh29N
2021-12-07 04:06:54 <Guest7836> i did the compilation as such: ./configure prefix="/path/to/sf" CC="gcc -static --static"
2021-12-07 04:20:09 <psykose> putting the CC= in the configure options probably doesn't do anything
2021-12-07 04:21:03 <psykose> what to do depends on the configure script, usually you can just export CFLAGS="-Os -static"
2021-12-07 04:56:29 <Guest7836> psykose , i will give it a try
2021-12-07 05:38:10 <Guest7836> it is still giving the same dependency message
2021-12-07 09:30:42 <ugifarukh> new to alpine linux, just installed it on a raspberry pi and working fine so far. But want to change the default ssh ports for the dropbear ssh server that I selected during setup-alpine. Also want to add only ssh key authentication. But cannot seem to find any documentation regarding how to do this. Can anyone please help?
2021-12-07 09:33:40 <lopid> https://github.com/mkj/dropbear#readme
2021-12-07 09:34:45 <ugifarukh> there's nothing regarding changing the ssh port in the dropbear readme
2021-12-07 09:37:01 <lopid> "see the manpage"
2021-12-07 09:37:59 <lopid> startup scripts are in /etc/init.d
2021-12-07 09:38:42 <ugifarukh> So, what should I do with the startup scripts?
2021-12-07 09:39:09 <lopid> did you read the man page?
2021-12-07 09:39:53 <ugifarukh> I read this -> https://linux.die.net/man/8/dropbear
2021-12-07 09:40:09 <ugifarukh> but the command provided here to change the port is not working on alpine linux
2021-12-07 09:53:55 <frag> trying to remove python3 i get : World updated, but the following packages are not removed due to: python3: qutebrowser youtube-dl zsh gcc rsync imagemagick git ... but i dont think all those depends on python eh?
2021-12-07 09:54:25 <frag> also is there a way to list packages that a package depends on? apk info -R seems to list modules?
2021-12-07 10:03:48 <lopid> apk search -r package
2021-12-07 10:04:35 <lopid> might want -er
2021-12-07 10:09:29 <frag> hmm so theres no way to list dependencies as package names?
2021-12-07 10:09:47 <lopid> aren't they?
2021-12-07 10:11:14 <frag> revers dependencies yes
2021-12-07 10:12:04 <frag> apk info -R zsh, doesnt list package names 
2021-12-07 10:12:26 <lopid> `apk search -er zsh` does
2021-12-07 10:12:50 <ugifarukh> any suggestions on the port change lopid?
2021-12-07 10:13:39 <lopid> what happens when you specify the port?
2021-12-07 10:14:20 <frag> lopid: isnt that still reverse? different from dependencies listed here https://pkgs.alpinelinux.org/package/edge/main/x86_64/zsh
2021-12-07 10:14:34 <ugifarukh> nothing. command executes. but the port doesnt change. it is still the default port 22
2021-12-07 10:14:59 <lopid> how can it be *any* port if nothing executed?
2021-12-07 10:15:28 <ugifarukh> by default ssh is already running on port 22 after setup-alpine
2021-12-07 10:15:42 <lopid> dropbear is already running?
2021-12-07 10:15:46 <ugifarukh> yes
2021-12-07 10:15:59 <ugifarukh> i can access ssh using putty on port 22
2021-12-07 10:16:03 <lopid> how are you changing the port and running it anew?
2021-12-07 10:16:24 <ugifarukh> i gave the command: dropbear -p 500
2021-12-07 10:16:29 <ugifarukh> according to the man page
2021-12-07 10:16:32 <ugifarukh> and then rebooted the pi
2021-12-07 10:16:33 <lopid> as root?
2021-12-07 10:16:41 <ugifarukh> yeah as root user
2021-12-07 10:16:46 <lopid> you ran it then rebooted?
2021-12-07 10:16:51 <ugifarukh> yes
2021-12-07 10:16:55 <ugifarukh> ran the command and rebooted
2021-12-07 10:17:08 <ugifarukh> then tried ssh, same port 22 it allows me to ssh in
2021-12-07 10:17:10 <lopid> rebooting stops all processes
2021-12-07 10:17:36 <lopid> stop dropbear, edit /etc/init.d/dropbear if it's there, start it
2021-12-07 10:17:42 <ugifarukh> why the configuration for port change isn't saved?
2021-12-07 10:18:05 <lopid> because you didn't save the port
2021-12-07 10:19:11 <ugifarukh> there's nothing related to port in /etc/init.d/dropbear
2021-12-07 10:20:39 <lopid> in it is DROPBEAR_OPTS, which is read from /etc/conf.d/dropbear
2021-12-07 10:21:26 <ugifarukh> so I modify the conf.d/dropbear?
2021-12-07 10:21:31 <lopid> yes
2021-12-07 10:21:43 <ugifarukh> ah I see. lemme try. will get back to you soon
2021-12-07 10:22:37 <lopid> dependencies are things that a package depends on. reverse deps are things that rely on the package
2021-12-07 10:24:32 <ugifarukh> worked like a charm lopid.
2021-12-07 10:24:44 <ugifarukh> thanks a lot man
2021-12-07 10:24:58 <ugifarukh> maybe you could please help me with pub key authenntication
2021-12-07 10:25:27 <lopid> unfortunately i'm too busy right now and i don't use dropbear, so i'd have to look it up just like you
2021-12-07 10:25:36 <ugifarukh> when I selected dropbear as the ssh option in setup-alpine, it created dsa, rsa and ecdsa keys automatically
2021-12-07 10:25:48 <ugifarukh> its ok, just help when you are free
2021-12-07 10:27:31 <ugifarukh> I just cannot seem to copy the public equivalent keys from the pi. now, I am left with no way to get the private keys and convert them to check the public keys. any way to do that?
2021-12-07 10:38:34 <ugifarukh> made it work @ lopid
2021-12-07 10:38:38 <ugifarukh> no worries man
2021-12-07 10:38:45 <ugifarukh> thanks for the port change help though
2021-12-07 16:25:14 <palliyil> Hey
2021-12-07 16:25:18 <palliyil> Is anyone online
2021-12-07 16:25:24 <palliyil> I need a bit of help
2021-12-07 16:25:50 <jvoisin> ask your question
2021-12-07 16:26:18 <palliyil> I am trying to install i3 on a fresh install of alpine on my laptop
2021-12-07 16:26:30 <palliyil> I finished the xorg-setup
2021-12-07 16:26:38 <palliyil> How do I boot into i3
2021-12-07 16:27:32 <vlp[m]> startx? 
2021-12-07 16:28:08 <palliyil> Xinitrc: server error
2021-12-07 16:28:31 <bdju> put `exec i3` into .xinitrc
2021-12-07 16:29:26 <palliyil> Nope didn't work
2021-12-07 16:29:29 <palliyil> Server error
2021-12-07 16:29:41 <palliyil> Server terminated with error 1
2021-12-07 16:55:22 <mdt> hi, i am looking for a explanation how to create a headless but not diskless installation on a raspberry zero - is there a method to prepare a sd to directly boot into such an installation?
2021-12-07 17:13:24 <artok> mdt usually adding wpa_supplicant and ssh enabling to boot is enough (if there is network connectivity), or then you'll need serial cable to be able to login on serial console
2021-12-07 17:18:42 <artok> on alpine it is https://wiki.alpinelinux.org/wiki/Raspberry_Pi_-_Headless_Installation
2021-12-07 17:53:28 <mdt> artok: yes, i read that but it is not headless - i want to prepare a sd on a linux box ready to go but i dont find the root fs, only the /boot one for alpine which is as i understand the diskless installation
2021-12-07 17:53:51 <mdt> artok: so you need to boot, login and issue commands which i dont want
2021-12-07 17:55:59 <mdt> artok: it reads as well that is supports either wifi or cable but doesnt tell if 'cable' means usb-ether 
2021-12-07 17:56:04 <durrendal> I don't think anyone has written a guide on how to do that, but you should be able to just do a chroot install and setup the sd card that way
2021-12-07 18:17:56 <mdt> so i try that wiki above (a chroot need that arch which i dont want to install either), but that does not boot, the led blinks 7 times... is that an error code?
2021-12-07 18:18:17 <clandmeter> mdt: you can also try the ssh_key cmdline option
2021-12-07 18:18:33 <clandmeter> you will probably need to combine it with net=dhcp or similar.
2021-12-07 18:18:54 <clandmeter> ip=dhcp 
2021-12-07 18:20:11 <mdt> clandmeter: this doesnt seem to boot at all, after pluggin in in starts blinking nearly immediatly - is a zero a possible candidate for a diskless boot?
2021-12-07 18:20:29 <clandmeter> which release did you use?
2021-12-07 18:21:00 <mdt> alpine-rpi-3.15.0-armhf.tar.gz
2021-12-07 18:21:33 <clandmeter> and you copied it to a sd card?
2021-12-07 18:22:13 <mdt> downloaded today, i understood its the current version. i use armhf (while the doc says aarch64) because i use that in other configurations too
2021-12-07 18:22:39 <clandmeter> zero is not aarch64
2021-12-07 18:22:45 <mdt> yes, i did what the wiki tells: vfat partition 256mb, tar x... from that file, copy headless..
2021-12-07 18:22:55 <mdt> ok, than that's fine
2021-12-07 18:24:04 <mdt> sync, unmount, boot - should be it, right? i have a linux box (but not alpine) where i write the sd. i have other SDs which boot fine, so the zero should be ok too
2021-12-07 18:24:29 <mdt> does the blinking tell something? like an error code?
2021-12-07 18:24:42 <clandmeter> probably
2021-12-07 18:24:48 <clandmeter> you need to look it up on the official docs
2021-12-07 18:24:55 <clandmeter> from rpi
2021-12-07 18:25:15 <clandmeter> from what i understand is that the sd card needs to be fat32 formatted
2021-12-07 18:25:26 <clandmeter> and you copy the contents to this partition 
2021-12-07 18:25:43 <mdt> so alpine just repacks /boot somehow? (i didnt see a kernel in there but in boot/boot)
2021-12-07 18:26:01 <mdt> sorry, yes, i followed the doc: mkdosfs -F 32 /dev/mmcblk1p1
2021-12-07 18:26:17 <clandmeter> i dont know the doc you are refering to, never tried it.
2021-12-07 18:26:35 <mdt> the one artok mentioned above
2021-12-07 18:26:45 <mdt> Raspberry_Pi_-_Headless_Installation
2021-12-07 18:27:28 <clandmeter> you also do not need 2 partitions, but you can ofc.
2021-12-07 18:27:39 <clandmeter> by default alpine will boot from ram
2021-12-07 18:28:21 <Hello71> #12368
2021-12-07 18:28:28 <Hello71> https://gitlab.alpinelinux.org/alpine/aports/-/issues/12368
2021-12-07 18:32:11 <clandmeter> Hello71: ah nice, didnt know about this.
2021-12-07 18:33:52 <Hello71> probably we should rename that directory
2021-12-07 18:34:00 <Hello71> i don't think rpi foundation will fix it any time soon
2021-12-07 18:35:42 <mdt> ok, blinking is gone, thank you Hello71 
2021-12-07 18:35:55 <Hello71> or just mount whole drive to /boot, i think that's what raspbian^Wraspberry pi os does
2021-12-07 18:36:11 <mdt> yes
2021-12-07 18:36:19 <aru> hi, anyone running alpine on odroid c2?
2021-12-07 18:36:22 <mdt> you have a boot in that partition
2021-12-07 18:40:19 <mdt> so it boots but doesnt show in wifi lan. how can i find out what happened - anything but serial? logfiles are gone i assume, anything else?
2021-12-07 18:41:10 <ikke> aru: maybe ask the postmarketos project. They are more involved with mobile devices
2021-12-07 19:22:47 <Hello71> ikke: odroid c2 is not particularly mobile
2021-12-07 19:23:09 <ikke> oh, my bad
2021-12-07 19:39:57 <aru> yeah, it is a sbc, not really a mobile thingy. I mean it should be supported by mainline kernel, but if anyone had a deployment script so I wouldn't have to fiddle with things then that would be great
2021-12-07 19:41:46 <durrendal> https://lord.re/en/posts/173-odroid-n2/ <- aru looks like this person managed to get it running on an odroid2
2021-12-07 19:41:56 <durrendal> err n2, well maybe not quite the same
2021-12-07 19:42:28 <durrendal> might point you in a direction ish
2021-12-07 19:43:54 <aru> oh, nice
2021-12-07 19:43:55 <aru> thanks
2021-12-07 19:44:01 <psykose> "it wears the mark of the evil" lmfao
2021-12-07 19:59:23 <aru> oh well, hopefully I'll get around to messing with it over the holidays
2021-12-07 20:02:32 <Lord> durrendal : hey that's me !
2021-12-07 20:03:15 <durrendal> Lord: I didn't know you were in here haha, thank you for the wonderful blog content :)
2021-12-07 20:03:24 <Lord> ;-)
2021-12-07 20:05:21 <aru_> Lord: now that I have you on the line, any additional pointers which weren't mentioned in the blog post?
2021-12-07 20:31:19 <lopid> be good unto others, my child
2021-12-07 21:19:49 <helby> true true ;/
2021-12-07 21:20:05 <helby> is there a way to disable useless histfile in busybox ash?
2021-12-07 21:23:44 <psykose> HISTFILE=/dev/null probably
2021-12-07 21:25:26 <helby> if used for whatever /dev/null write is still there no?
2021-12-07 21:26:23 <lopid> `unset HISTFILE` might be better, if ash recognises unset
2021-12-07 21:26:59 <helby> no
2021-12-07 21:27:08 <psykose> ash has unset, but i'm not sure how that helps
2021-12-07 21:27:11 <helby> probably just point it to crap like /dev/null
2021-12-07 21:27:26 <helby> it does not help
2021-12-07 21:27:39 <helby> it just point to default $/.ash_history
2021-12-07 21:27:40 <psykose> pointing it to /dev/null makes it not write a history file
2021-12-07 21:27:52 <helby> yes, that I Understand
2021-12-07 21:28:00 <psykose> yes, because the logic is if HISTFILE is unset, it uses ~/.ash_history
2021-12-07 21:28:16 <helby> I was asking if I write anything to /dev/null is there writting to nowhere?
2021-12-07 21:28:23 <lopid> fing cointreau
2021-12-07 21:28:36 <psykose> yes? what do you think writing to dev null does
2021-12-07 21:28:36 <helby> but still writting?
2021-12-07 21:28:37 <lopid> /dev/null is a void
2021-12-07 21:28:48 <psykose> what do you mean still writing?
2021-12-07 21:28:56 <psykose> it doesn't do any disk io
2021-12-07 21:29:02 <lopid> it goes through the motions but nothing results. like a phantom shit
2021-12-07 21:29:02 <helby> oh that I meant
2021-12-07 21:29:08 <helby> thanks
2021-12-07 22:04:30 <helby> ajaj
2021-12-07 22:04:55 <helby> reinstall openrc somehow overwrite all /etc/crap to default ;/
2021-12-07 22:05:54 <helby> no one apk-new
2021-12-07 22:08:00 <helby> in edge 
2021-12-07 22:09:29 <lopid> are you, helby, sure?
2021-12-07 22:10:03 <helby> no, just joking obviously
2021-12-07 22:10:19 <lopid> ok
2021-12-07 22:10:22 <helby> I am sure ;/
2021-12-07 22:10:51 <psykose> are you ok
2021-12-07 22:11:20 <helby> what?
2021-12-07 22:11:46 <helby> why u point ot me? I was just describing issue I experienced right now
2021-12-07 22:12:11 <helby> if I apk fix openrc it overwrite all files it handle in /etc/
2021-12-07 22:12:20 <helby> so yes I am ok and I am sure about it
2021-12-07 22:12:49 <helby> bc. did it few times now and I m not brave to even reboot
2021-12-07 22:13:07 <bdju> fyi saying you were joking was confusing, I guess it was sarcasm but it was hard to tell
2021-12-07 22:13:58 <helby> okay, thanks, that helps ;/
2021-12-07 22:14:08 <psykose> apk fix openrc doesn't overwrite anything with changes in it for me
2021-12-07 22:14:32 <helby> in edge?
2021-12-07 22:14:39 <psykose> yes
2021-12-07 22:14:47 <helby> so what can cause that? 
2021-12-07 22:14:48 <lopid> psykose, are you joking?
2021-12-07 22:14:58 <Lord> aru_ : sorry it was pizza time. It still should work i think.
2021-12-07 22:15:08 <Lord> My sbc is still running alpine since this blog post
2021-12-07 22:15:35 <psykose> lopid: no
2021-12-07 22:16:10 <lopid> helby, restore from backup
2021-12-07 22:16:38 <helby> I did some files
2021-12-07 22:16:43 <helby> but then it happen again 
2021-12-07 22:17:25 <helby> I also tried to reinstall all and these packages does not crate groups/users
2021-12-07 22:17:30 <helby> create
2021-12-07 22:18:16 <psykose> maybe my system is magic, but afaik this was always the default behaviour https://img.ayaya.dev/JNHNbpnX3383
2021-12-07 22:18:37 <psykose> and there is an mtab.apk-new placed next to it
2021-12-07 22:19:01 <helby> yes that is how it works
2021-12-07 22:19:21 <helby> but few minutes again it somehow overwrited everything fo rme
2021-12-07 22:23:21 <helby> oh shit
2021-12-07 22:23:28 <helby> not openrc ;/
2021-12-07 22:23:36 <helby> alpine-baselayout
2021-12-07 22:23:44 <helby> don't try at home
2021-12-07 22:25:07 <helby> not sure what can cause that
2021-12-07 22:25:39 <helby> not sure if it was updated in this edge 3.15 ever, probably not
2021-12-07 22:51:41 <helby> I did not believe this installation from the beginning, it was done in hurry and it was weird
2021-12-07 22:53:13 <helby> the issue is, that anything I reinstall overwrite these files for own package in /etc
2021-12-07 22:54:44 <helby> like alpine-conf too
2021-12-08 09:50:41 <aru_> Lord: cool, thanks
2021-12-08 09:53:11 <Lord> aru_: the odroid n2 works with a mainline kernel now. I think that a some people were working on the same thing for the odroid c2 so maybe it's done now
2021-12-08 09:56:08 <aru_> would you recall who the people were so I could get in touch with them>
2021-12-08 09:56:12 <aru_> s/>/?
2021-12-08 09:56:12 <alpine-meetbot> aru_ meant to say: would you recall who the people were so I could get in touch with them?
2021-12-08 10:01:54 <vlp[m]> Hello all, I have set up a alpine laptop installation with Grub2/Luks2 in order to have a FDE. It "almost" works. Indeed Grub refuse to automatically open the encrypted disk and I fall back into grub-rescue. However, with grub-rescue, I can manually mount the encrypted disk and continue the boot process normally. 
2021-12-08 10:03:46 <vlp[m]> I have read that I may have to creak my own grub image w/ grub-mkinstall 
2021-12-08 10:54:01 <minimal> vlp[m]: this was talked about a couple of days ago - LUKS2 FDE with Grub2 does *not* work currently as grub does not handle it correctly
2021-12-08 11:16:38 <Carrot4123> I just upgrade from 3.14 to 3.15 and now my Laptop no longer boots.
2021-12-08 11:16:39 <Carrot4123> The initial kernel log looks visually broken and I see the message "drm_plane_enable_fb_damage_clips() not  called". It looks broken because after that I get a few more messages that are wrongly indented and overlap with eachother or are cutoff. Then the output seems to have frozen.
2021-12-08 11:16:39 <Carrot4123> I wonder if this is related to this item in the changelog: Framebuffer drivers have been disabled in kernel and replaced by simpledrm..
2021-12-08 11:16:39 <Carrot4123> My setup is somewhat unconventional because I boot through efistub and decrypt my luks root partition during boot. Not sure if that matters.
2021-12-08 11:16:40 <Carrot4123> Does anyone have any ideas how I can fix this?
2021-12-08 11:16:41 <Carrot4123> If not, is there a way to downgarde back to 3.14 or to the previous kernel?
2021-12-08 11:18:11 <Carrot4123> I have an alpine 3.14 live cd (usb stick) from which I can boot chroot into the system.
2021-12-08 11:19:30 <psykose> you can downgrade by updating the repositories to the old ones in /etc/apk/repositories and just running and upgrade -Ua
2021-12-08 11:19:39 <psykose> aside from that, yeah probably the simpledrm change
2021-12-08 11:19:59 <psykose> you did upgrade with -a/--available right?
2021-12-08 11:21:35 <vlp[m]> <minimal> "vlp: this was talked about a..." <- Yes, and thanks again for your explanations! When we talk about it, I have had successfully setup my laptop w/ grub2/luks1. And, I wanted to try with luks2 as well. And I have exactly the issue that you have mentioned 
2021-12-08 11:21:50 <Carrot4123> I forgot that the first time but after I saw it failing I ran `apk upgrade --available` from the chroot and mkinitfs.
2021-12-08 11:22:31 <minimal> vlp[m]: ok, so what are you asking now?
2021-12-08 11:23:35 <psykose> Carrot4123: sure, if it fails with a full upgrade then it's most likely the simpledrm change and you should use 3.14 for a bit
2021-12-08 11:23:42 <psykose> i am not sure what the path forward is though yet
2021-12-08 11:24:14 <vlp[m]> minimal: Any tips for a workaround 
2021-12-08 11:24:35 <vlp[m]> Sorry I was really clear! 
2021-12-08 11:25:25 <vlp[m]> * I was not really clear!
2021-12-08 11:25:57 <minimal> psykose, Carrot4123: in some cases of people upgrading from 3.14 to 3.15 mkinitfs was not updated (I'm assuming as they did not specify "-aU" to the apk upgrade), I opened a MR to address this, !28023, a week ago but its not yet been merged
2021-12-08 11:26:11 <psykose> in this case it's upgraded
2021-12-08 11:27:09 <minimal> vlp[m]: I don't believe there is a complete workaround yet. I don't think changing grub.cfg manually fixes all aspects of the problem
2021-12-08 11:28:16 <Carrot4123> I made sure to run upgrade --available and rebuild the initfs from the chroot afterwards so it should be fine. Attempting the downgrade right now. Thanks for your help. Should I report my problem and hardware in an issue somewhere maybe?
2021-12-08 11:28:44 <minimal> psykose: if its run-from-ram then does the initramfs file also include any required graphics driver firmware files?
2021-12-08 11:30:36 <Carrot4123> Downgrade worked. I could try upgrading again this time from proper boot and with --available.
2021-12-08 11:31:03 <Carrot4123> I'll try it I guess. Doesn't take that long to fix.
2021-12-08 11:31:08 <minimal> Carrot4123: ah you're using EFISTUB - does you "embedded" initramfs include the kernel module for your graphics card and any firmware this card needs?
2021-12-08 11:32:50 <Carrot4123> I... don't know. Since I don't know maybe not? All I know is this used to work. Does the SimpleDrm change now need drivers when it didn't before?
2021-12-08 11:34:17 <psykose> what does efistub have to do with embedded initramfs? it still uses the normal one
2021-12-08 11:34:36 <minimal> Carrot4123: there were several issues recently - in Alpine 3.15 kernel modules are compressed but only the newer versions of mkinitfs handle this - older versions do not place kernel modules in the initramfs as the filenames changed (due to compression)
2021-12-08 11:36:18 <minimal> Carrot4123: in addition for run-from-ram setup there was a 2nd issue where code that checked the kernel codes in the initramfs to see what (if any) firmware they also required to be placed in the initramfs also did not handle compressed kernel modules correctly and so did not add relevant firmware files to the initramfs
2021-12-08 11:36:36 <minimal> s/codes/modules/
2021-12-08 11:36:36 <alpine-meetbot> minimal meant to say: Carrot4123: in addition for run-from-ram setup there was a 2nd issue where code that checked the kernel modules in the initramfs to see what (if any) firmware they also required to be placed in the initramfs also did not handle compressed kernel modules correctly and so did not add relevant firmware files to the initramfs
2021-12-08 11:38:22 <minimal> psykose: was thinking of the situation with EFISTUB where kernel and initramfs are "merged" together. Anyway the underlying point was whether the initramfs contained both the kernel modules and any required firmware
2021-12-08 11:38:46 <psykose> ah, that exists, but no alpine tooling does that yet to my knowledge
2021-12-08 11:39:19 <minimal> psykose: was looking at it recently and so that's why it was fresh in my mind :-)
2021-12-08 11:39:41 <psykose> :^)
2021-12-08 11:39:56 <minimal> Carrot4123: so in summary, have you checked your initramfs to see what's in it?
2021-12-08 11:40:03 <Carrot4123> (got disconnected). Thanks for the explanation. For reference this laptop has an Intel Atom processor and Intel HD graphics.
2021-12-08 11:40:05 <Carrot4123> How do I check it?
2021-12-08 11:40:22 <psykose> mkinitfs -l will tell you what gets generated into it
2021-12-08 11:40:24 <psykose> list of files
2021-12-08 11:40:35 <minimal> Carrot4123: "mkinitfs -l" will show you what *would* be placed in it if it was regenerated
2021-12-08 11:40:43 <psykose> then you can see if /lib/firmware contains firmware for your gpu, as simpledrm needs it
2021-12-08 11:40:54 <minimal> so not strictly speaking the current contents
2021-12-08 11:40:58 <psykose> but of course it's also possible simpledrm just doesn't work for your gpu, even with the firmware there
2021-12-08 11:41:43 <minimal> I'd assume you should have "i915" as your graphics driver and some files in /lib/firmware/i915/
2021-12-08 11:42:21 <minimal> assuming i915 is the correct driver for the Intel HD chipset you have
2021-12-08 11:43:47 <minimal> if you do a "modinfo 915" and look for firmware fields that shows the file(s) needed by the kernel module
2021-12-08 11:49:12 <Carrot4123> It is this laptop https://www.notebookcheck.net/Review-Lenovo-IdeaPad-Yoga-2-11-Convertible.116051.0.html . I do have the files in /lib/firmware/i915 however I don't see them in mkinitfs -l  (running from my working again downgarde 3.14). I do notice a mkinitfs.conf.apk-new file now that I didn't before. Maybe the issue that I'm missing some
2021-12-08 11:49:13 <Carrot4123> "features" it has changed.
2021-12-08 11:49:52 <Carrot4123> I'm going to update that conf with the new values and upgrade again.
2021-12-08 11:56:26 <minimal> Carrot4123: if you "diff" the old and new files you'll likely see that the different is that in any features *.apk-new files how reference ".ko*" rather than ".ko" so that they match the compressed kernel module files
2021-12-08 11:58:04 <minimal> actually thinking about it I guess mkinitfs' should use a pre-install file to rewrite existing features.d/*.modules files to change the ".ko" suffix automatically
2021-12-08 12:04:23 <Carrot4123> The only apk-new file was mkinnitfs.conf (not features.d/*) which only contains one line features="<stuff>". Sadly still broken. I do see some i915 references in mkinitfs -l now so maybe that's not the issue. Will stay downgraded for a bit I guess. Thanks for your help again.
2021-12-08 12:09:55 <minimal> Carrot4123: if you see i915 entries now that previously were not showing then, yes, that may have been the issue - note "mkinitfs -l" does not actually update the initramfs file, it only shows up what the result would be if you ran it without the "-l" option
2021-12-08 12:11:48 <Carrot4123> I made it regenerate. Even checked the timestamp on disk.
2021-12-08 12:15:23 <minimal> Carrot4123: and i915 is definately the correct driver for your HW? have a look at "lsmod" output on a working kernel to see if it or another GPU module is loaded
2021-12-08 12:15:40 <minimal> or look at "dmesg" output
2021-12-08 16:36:58 <Carrot4123> (came back from being afk). I ran `lshw -c video` which says `configuration: driver=i915 latency=0`.
2021-12-08 16:52:00 <Nulo> How do I make Alpine generate /run/user/*?
2021-12-08 16:52:40 <ikke> You need a seat / session manager
2021-12-08 16:55:44 <ikke> It's something systemd specific
2021-12-08 16:55:51 <Nulo> I use seatd
2021-12-08 16:56:24 <Nulo> Yay for podman assuming systemd things (it's Red Hat after all)
2021-12-08 16:57:03 <ikke> Maybe setting XDG_RUNTIME_DIR to something that exists would help
2021-12-08 16:58:16 <Nulo> I'm already doing that
2021-12-08 17:01:14 <ikke> but that does not work?
2021-12-08 17:10:38 <bdju> Nulo: I'm using elogind on a non-alpine distro (but one without systemd). I have that directory. I also have dbus installed. Can't remember for sure which of the two were needed or if both were. It was to get Sway to work.
2021-12-08 17:10:54 <ikke> bdju: do you have pam_systemd?
2021-12-08 17:11:14 <bdju> ikke: Is that a package or something else? How would I verify?
2021-12-08 17:11:34 <bdju> I'm thinking I most likely do not have it, but not sure.
2021-12-08 17:12:26 <ikke> ls -l /usr/lib/security/pam_systemd.so
2021-12-08 17:14:13 <bdju> ah, I don't even have /usr/lib (this machine is running Guix System which has nonstandard dirs a bit like NixOS)
2021-12-08 17:14:42 <ikke> ah ok
2021-12-08 17:14:50 <bdju> no results in /gnu/store either for it, so yeah, I don't think I have it
2021-12-08 17:14:52 <ikke> well, look for pam_systemd.so somewhere I guess
2021-12-08 17:14:55 <ikke> ok
2021-12-08 17:15:14 <bdju> unless it's not at the top level... I just did an ls | grep and not a find. I'll do one more search
2021-12-08 17:15:41 <ikke> find / -type f -name pam_systemd.so
2021-12-08 17:17:31 <Nulo> ikke: No, it doesn't work as the directory is under /tmp.
2021-12-08 17:18:35 <bdju> the search is taking a while with your suggested method. I have some sshfs dirs that are several TB large that may be slowing it down. no results yet anyway.
2021-12-08 17:23:33 <bdju> I ran a second find on /gnu/store specifically and there were no results there. I'll leave the one on / going and check back on it later
2021-12-08 17:52:45 <Nulo> apk fix has no --no-chown but apk add does
2021-12-08 18:11:33 <Carrot4123> Re previous discussion on my boot problem on 3.15 . I decided to do a clean installation and now I see that `mkinitfs -l` again is missing i915 . How do I make sure mkinitfs includes them? I couldn't find any documentation on the alpine specific mkinitfs.conf .
2021-12-08 18:16:21 <ikke> Carrot4123: create a file /etc/mkinitfs/features.d/i915.modules
2021-12-08 18:16:31 <ikke> and add kernel/drivers/gpu/drm/i915/i915*
2021-12-08 18:23:10 <ikke> Then you can add i915 to features in mkinitfs.conf
2021-12-08 18:23:34 <Carrot4123> Thank you. It worked.
2021-12-08 18:24:42 <ikke> nice
2021-12-08 18:27:34 <Carrot4123> To clarify. After a fresh installation AND force adding the i915 modules my previous problem went away and system boots properly on 3.15.
2021-12-08 18:33:16 <MAD_AS_HELL_> I'M MAD AS HELL
2021-12-08 18:33:58 <hoel> Wouldn’t have guessed it.
2021-12-08 18:34:24 <lopid> let's see if they're going to take it any more
2021-12-08 18:34:40 <lopid> 
ACTION looks out the window to check

2021-12-08 18:36:17 <MAD_AS_HELL_> That movie's so goddamned good.
2021-12-08 18:36:28 <MAD_AS_HELL_> I've narrowed the situation back down to what I originally thought it was: the initramfs.
2021-12-08 18:36:39 <MAD_AS_HELL_> I've unfortunately tried remaking it multiple times to no avail.
2021-12-08 18:37:00 <MAD_AS_HELL_> Makes me think I should try foregoing an initramfs at all, but at that point, maybe I'll just switch to gentoo.
2021-12-08 18:37:18 <MAD_AS_HELL_> All of this so I can use btrfs for literally no reason and uefi because fuck bios.
2021-12-08 18:37:49 <MAD_AS_HELL_> Why do linux distros not give any fucks about uefi, ten years after widespread adoption?
2021-12-08 18:37:57 <MAD_AS_HELL_> "But you CAN do uefi, you just need to use grub!"
2021-12-08 18:38:01 <MAD_AS_HELL_> Who gives a fuck about grub?
2021-12-08 18:38:11 <MAD_AS_HELL_> Fucking no one. They don't even know it fucking exists.
2021-12-08 18:38:47 <MAD_AS_HELL_> I'll say I did away with grub on a laptop a while ago and never had issues. No clue what the difference has been here.
2021-12-08 18:43:03 <minimal> Carrot4123: did you have "kms" listed in the FEATURES line of your /etc/mkinitfs/features.conf file? that's the standard way to ensure graphics modules are added to initramfs
2021-12-08 18:43:30 <ikke> minimal: ah, good to know
2021-12-08 18:44:20 <minimal> ikke: I'd assume kms to be in there as standard, will go take a look in aports now to check...
2021-12-08 18:46:41 <minimal> ikke: "kms" is in the stock file already: https://gitlab.alpinelinux.org/alpine/mkinitfs/-/blob/master/Makefile#L84
2021-12-08 18:46:52 <ikke> ok, so it should work already
2021-12-08 18:48:15 <MAD_AS_HELL_> kms obviously stands for kill myself
2021-12-08 18:48:29 <minimal> ikke: yes unless its what I originally thought, related to the compressed kernel modules (i.e. out of date mkinitfs and/or alpine-conf)
2021-12-08 18:48:33 <MAD_AS_HELL_> which i want to do any time video is involved
2021-12-08 18:49:00 <Carrot4123> kms isn't there. I might have made a mistake recreating that file.
2021-12-08 18:49:12 <MAD_AS_HELL_> you done fucked up
2021-12-08 18:49:59 <MAD_AS_HELL_> what the fuck
2021-12-08 18:50:36 <MAD_AS_HELL_> oh nevermind, not necessarily what the fuck
2021-12-08 19:14:18 <MAD_AS_HELL_> the desire to give in is too great
2021-12-08 19:14:28 <MAD_AS_HELL_> i must persist
2021-12-08 19:29:04 <MY-R> hmm, do you guys got full "dmesg" output or the first ling start from "[    0.927418] simple-framebuffer ......" ?
2021-12-08 19:30:10 <MY-R> weird because in log file in /var/log/dmesg I have little more and start from "[    0.746404] usbcore: registered ...." but still lot of log missing
2021-12-08 19:30:16 <minimal> MY-R: dmesg output will only show a (certain) maximum number of like due to kernel log buffer size
2021-12-08 19:30:45 <minimal> MY-R: if you look at /var/log/dmesg it should show the full dmesg output from boot
2021-12-08 19:30:56 <MY-R> minimal: read above
2021-12-08 19:31:12 <minimal> MY-R: was typing my reply as you sent that :-)
2021-12-08 19:31:17 <MY-R> :)
2021-12-08 19:31:43 <MY-R> on rpi where simpledrm doesnt exist I have all, on x86_64 like above...
2021-12-08 19:32:38 <MY-R> I think it was too early to switch to simpledrm, even gentoo removed from their modificated fedora kernel configs few releases ago...
2021-12-08 19:33:41 <minimal> MY-R: the 1st thing I see in /var/log/dmesg on a x86_64 VM is at 0.006749
2021-12-08 19:34:09 <minimal> I think basically any output before simpledrm is initilised is "lost"
2021-12-08 19:34:42 <MY-R> that is bad
2021-12-08 19:36:19 <MY-R> I have no idea what and when basic and most important stuff are loaded :/
2021-12-08 19:36:33 <minimal> MY-R: I think its the nature of simpledrm - in the "old" way I think basic VGA text console (on BIOS) was always there for the kernel whereas now I think there's nothing available for output until simpledrm initialises
2021-12-08 19:37:55 <minimal> MY-R: actually thinking about, its primarily likely to simpledrm being loaded as a module by the initfs' init scripr rather than being compiled into the kernel
2021-12-08 19:37:56 <MY-R> I dont like simpledrm stuff already, not to mention some random artifacts when switching resolution in console :\
2021-12-08 19:38:10 <minimal> if it was compiled in then it would init earlier
2021-12-08 19:38:50 <MY-R> minimal: any tip which modules I could blacklist
2021-12-08 19:39:04 <minimal> MY-R: for what specific purposes/reason?
2021-12-08 19:39:17 <MY-R> to see full dmesg output? :P
2021-12-08 19:42:43 <minimal> actually thing about it, that doesn't make sense - simpledrm is loaded by initram which, on this VM, is not unpacked by kernel until 0.214284 (which is in the dmesg output) and so the kmsg buffer for output prior to simpledrm init must obviously still be accessible to simpledrm
2021-12-08 19:42:51 <minimal> s/thing/thinking/
2021-12-08 19:42:51 <alpine-meetbot> minimal meant to say: actually thinking about it, that doesn't make sense - simpledrm is loaded by initram which, on this VM, is not unpacked by kernel until 0.214284 (which is in the dmesg output) and so the kmsg buffer for output prior to simpledrm init must obviously still be accessible to simpledrm
2021-12-08 19:43:47 <minimal> could it perhaps be related to CONFIG_EARLY_PRINTK? not sure if that was set in previous Alpine kernels
2021-12-08 19:45:08 <MY-R> hmm maybe
2021-12-08 19:46:33 <minimal> anyway I agree it would be "good" to have all the kernel output in dmesg/logfile
2021-12-08 19:47:31 <free_mind> Is aarch64  version 3.15.0 able to run on 'Raspberry Pi Zero 2 W' (quad-core 64-bit ARM Cortex-A53) ?
2021-12-08 19:48:12 <free_mind> I tried but it just hangs at the rainbow image...
2021-12-08 19:48:24 <minimal> free_mind: not sure if all the DTB-related stuff is in place currently
2021-12-08 19:49:09 <minimal> free_mind: plus I believe the move in general of linux-rpi to 3.15.x was premature
2021-12-08 19:49:51 <farukh> clock skew detected and start-stop daemon failed problem in raspberry pi alpine boot process. Any suggestions?
2021-12-08 19:50:03 <free_mind> Trying to figure out if I'm missing a config item in usercfg.txt
2021-12-08 19:50:19 <free_mind> any suggestions minimal?
2021-12-08 19:50:21 <minimal> farukh: sell the RPI and buy another SBC with a RTC on board? or buy a RTC for your RTC?
2021-12-08 19:50:22 <MY-R> I see in "boot" bcm2710-rpi-zero-2.dtb 
2021-12-08 19:50:54 <free_mind> Yes that's why I figuring out what I'm doing wrong. :) MY-R 
2021-12-08 19:51:17 <minimal> farukh: the "clock skew" message is because the RPI does not have an RTC (battery-based clock), so each time you turn it on it thinks its 1970 :-)
2021-12-08 19:52:57 <minimal> MY-R: even with that DTB file present there may still be other problems booting the Zero 2 - I don't have one to try.
2021-12-08 19:53:05 <MY-R> maybe "config.txt" point to "wrong" images
2021-12-08 19:53:58 <minimal> Basically Alpine 3.15 and Edge are using RPI Foundation 5.15.x kernels (whereas Alpine 3.14 uses 5.10.x Foundation kernels)
2021-12-08 19:56:22 <minimal> my understanding is that RPI Foundation 3.15.x kernel is still under development - it is not stable, Foundation expect people to use 5.10.x kernel
2021-12-08 19:57:53 <free_mind> Thanks for sharing; I know understand your statement  (... "move in general of linux-rpi to 3.15.x was premature"). BTW: would I make a diff choosing armv7/armhf minimal ?
2021-12-08 19:59:06 <minimal> free_mind: 3.15.x does work fine for me using aarch64 on a RPI3, but I'm using it as a server so have never tested if Wifi/Bluetooth/etc functionality works or not
2021-12-08 20:02:37 <MAD_AS_HELL_> i bet
2021-12-08 20:02:39 <MAD_AS_HELL_> i could reinstall
2021-12-08 20:02:43 <MAD_AS_HELL_> without changing my /home at all
2021-12-08 20:03:04 <MAD_AS_HELL_> that'd be so fucking cool
2021-12-08 20:15:40 <MAD_AS_HELL_> i've realized
2021-12-08 20:15:54 <MAD_AS_HELL_> i'm the best
2021-12-08 20:19:07 <artok> minimal: my 2x rpi3 and 1x rpi4 are running 3.15 aarch64 with wifit and bt so no problemos
2021-12-08 20:19:24 <artok> just reporting =)
2021-12-08 20:21:52 <free_mind> artok: the image boots fine here on a rpi4 (prepared wifi on it and that works and wanted to move the sdcard to the Zero 2W) Didn't  test bt so far, but thanks for sharing though.
2021-12-08 20:22:03 <minimal> artok: ok, I was just pointing out that a working machine does not mean that 100% of the RPI functionality (I used Wifi and BT as examples) is ok
2021-12-08 20:22:25 <farukh> is there any way to stop the clock skew messages? and make ialpine only use software clock and never look for any hwclocks?
2021-12-08 20:22:26 <minimal> bottom line: I still think Foundation only support/recommend 5.10.x currently
2021-12-08 20:23:12 <free_mind> so changes for Zero 2W are better with 3.14, minimal?
2021-12-08 20:23:12 <minimal> farukh: no, at least some clock skew messages are "normal" on a RPI without RTC
2021-12-08 20:23:18 <artok> farukh: get rtc
2021-12-08 20:24:27 <farukh> why am I getting start-stop daemon : no processes red error while rebooting pie everytime?
2021-12-08 20:24:33 <minimal> farukh: it is caused by OpenRC noticing that a specific file is newer than the "current" date/time, where the "current" date/time at boot is 1970
2021-12-08 20:25:13 <artok> kernel needs to ask rtc before running init to get "clean" boot dates and times
2021-12-08 20:25:24 <minimal> free_mind: Zero 2W is relatively new hardware, don't think the kernel in Alpine 3.14 supports it but not sure
2021-12-08 20:27:07 <artok> ncopa runs diffs to rpi kernels from foundation patches so if they're implemented on what ever kernel version you're running, they will work
2021-12-08 20:28:01 <minimal> artok: he was talking about Alpine 3.14 which has 5.10.78 kernel
2021-12-08 20:28:18 <artok> hence I did generalization =)
2021-12-08 20:28:50 <artok> not that I know about bootloader on 2W
2021-12-08 20:29:08 <minimal> artok: I don't think "new hardware" patches are applied to old Alpine releases
2021-12-08 20:29:11 <artok> nor the dtb
2021-12-08 20:29:32 <artok> if the kernel has sub update ?
2021-12-08 20:29:38 <minimal> normally back patches are to fix security problems only
2021-12-08 20:30:59 <artok> just checking aport from raspberry kernel, it does diff creation from foundation repo against linux repo, and pushes diff to alpine server
2021-12-08 20:31:45 <minimal> artok: yes, but in Edge and 3.15 its using the 5.15.x branch which as far as I know is not supported/recommended yet by Foundation
2021-12-08 20:32:21 <minimal> perhaps that has changed, RPI Foundation are not very clear in their communications regarding such things
2021-12-08 20:32:48 <artok> sure, but they do also some backporting to older kernels for newer chips, and if they're on repo, aport will include them
2021-12-08 20:33:02 <artok> just matter of maintainer who runs the diff...
2021-12-08 20:34:51 <minimal> artok: but if linux-rpi on Alpine 3.14 is not updated for some time due to a policy of only updating in old Alpine releases to bring in security fixes then it could be some time before linux-rpi in Alpine 3.14 moves on from 5.10.59-r0
2021-12-08 20:35:17 <minimal> s/5.10.59/5.10.78/
2021-12-08 20:35:18 <alpine-meetbot> minimal meant to say: artok: but if linux-rpi on Alpine 3.14 is not updated for some time due to a policy of only updating in old Alpine releases to bring in security fixes then it could be some time before linux-rpi in Alpine 3.14 moves on from 5.10.78-r0
2021-12-08 21:04:31 <kevint> So I am getting some strange "symbol not found" errors on two of my packages after upgrading them: https://dpaste.com/79HJAUA8S Both of these packages are built with boost and I noticed that boost seemed to get upgraded at the same time as these two packages, so maybe it has something to do with that? I'm not really sure how to fix this.
2021-12-08 21:08:32 <Hello71> upgrading boost usually requires upgrading all packages that depend on boost and their recursive dependencies
2021-12-08 21:08:38 <Hello71> unfortunately this wasn't done for boost 1.77
2021-12-08 21:39:25 <lamineer> someone seems to be editing the wiki maliciously 
2021-12-08 21:39:45 <lamineer> is there anyone here that can address this?
2021-12-08 21:42:36 <ikke> lamineer: Examples?
2021-12-08 21:44:35 <lamineer> sure. I'll try not to make this a gish gallop, but here's the gist of it
2021-12-08 21:47:34 <lamineer> give me a minute...
2021-12-08 21:49:50 <ikke> Is it about pages being deleted/
2021-12-08 21:57:08 <lamineer> 1) user creates new account and instantly starts editing multiple pages in a short amount of time
2021-12-08 21:57:15 <lamineer>  Some edits appear helpful, yet some edits appear destructive, e.g.
2021-12-08 21:57:41 <lamineer> 2) a number of their edits delete entire pages examples of pages deleted include Nextcloud, Snort, Setting_up_a_VPN_with_tinc, SLiM e.g. with SLiM the user clearly didn't research the page  before deleting. The package is still active in alpine.
2021-12-08 21:58:07 <lamineer> If pages are going to be edited, as I said in a talk page, the user should run through the installand update the information if certain steps are needed, not blanket deletion of whole pages. It's poor editing.
2021-12-08 21:58:34 <lamineer> 3) user questionably edits certain pages e.g. the flatpak page links are deleted. such links include info to the official flatpak page for alpine, and external links showing usage of flatpak (e.g. how to run windows programs through winepak).
2021-12-08 21:59:03 <lamineer> So, at first I thought he was just being helpful, but he seems to be recklessly deleting items, and otherwise making significant changes to the wiki. Unless you know what is going on, it looks like someone with an agenda.
2021-12-08 22:00:39 <lamineer> There are some other small examples, but I tried to keep this succinct
2021-12-08 22:01:13 <ikke> Yes, if I look at recent changes, I see what you mean
2021-12-08 22:05:49 <MAD_AS_HELL_> lamineer: The wiki is in utter disrepair.
2021-12-08 22:06:49 <MAD_AS_HELL_> Almost every article I've needed to use came up as outdated when given a second look by someone in here. My activity alone has been the cause of multiple "DON'T USE THIS ARTICLE" additions to the site.
2021-12-08 22:07:12 <MAD_AS_HELL_> When I say multiple, I mean at least 4, and that's in half as many months.
2021-12-08 22:07:27 <MAD_AS_HELL_> With that impression, mass deletion isn't suspect.
2021-12-08 22:08:25 <lamineer> Deletion of active pages is not proper editing.
2021-12-08 22:09:14 <lamineer> Outdated pages are still useful resources. Where appropriate a notice saying "information may need to be updated" is better than deleting all information, which may be useful to those trying to install the programs
2021-12-08 22:23:36 <zcrayfish> In addition to the notice template; comments could be added to the talk pages of articles in question to help editors out.
2021-12-08 22:28:50 <zcrayfish> Also proposing an article for deletion does not mean actually deleting the article... Are they even reading the template?
2021-12-08 22:52:05 <MAD_AS_HELL_> lamineer: In the cases I had, it was detrimental in all cases to follow any of the documentation.
2021-12-08 22:52:11 <MAD_AS_HELL_> (On the pages mentioned)
2021-12-08 22:57:02 <zcrayfish> The user in question has even blanked up-to-date pages . . . .
2021-12-08 22:57:26 <zcrayfish> afk
2021-12-08 23:02:27 <MAD_AS_HELL_> in that case, it's only circumstantially convenient that some pages deserved to die
2021-12-08 23:05:22 <zcrayfish> Haha.
2021-12-08 23:05:43 <zcrayfish> Ok fair enough. Some of the pages should be kept, and some of them should be deleted indeed
2021-12-08 23:09:53 <lamineer> MAD_AS_HELL_: can you give examples?
2021-12-08 23:11:11 <MAD_AS_HELL_> lamineer: all the pipewire shit
2021-12-08 23:26:45 <bdju> ikke: never found that pam thing. only output was a bunch of permission denied on weird dirs like proc and suc
2021-12-08 23:26:50 <bdju> s/suc/such
2021-12-08 23:26:50 <alpine-meetbot> bdju meant to say: ikke: never found that pam thing. only output was a bunch of permission denied on weird dirs like proc and such
2021-12-08 23:43:36 <lamineer> MAD_AS_HELL_: Pipewire has only one page that I can see. It was created in Jan of 2021, so it's less than a year old. Any other examples of pages with bad documentation you followed? I might get inspired to go through them.
2021-12-08 23:44:02 <lamineer> I can understand Pipewire being rough documentation, but on the positive side, it looks like it is actively being edited.
2021-12-08 23:44:40 <MAD_AS_HELL_> actively after my bitching, that is
2021-12-08 23:44:51 <MAD_AS_HELL_> i'm trying to recall the next one in order when it was going insane
2021-12-08 23:45:13 <MAD_AS_HELL_> i'll post it as it comes
2021-12-08 23:46:31 <zcrayfish> I encourage complaining on the article talk pages. 👍👍
2021-12-08 23:47:21 <MAD_AS_HELL_> pretty sure my default "fuck this article, it sucks" gets autoremoved
2021-12-08 23:48:55 <zcrayfish> Probably, you could make your complaint more specific... Like... "Why do we have this article, this package is no longer a part of Alpine"... Or "Step 5 in the directions of this how-to no longer works on the current release"...
2021-12-09 00:06:07 <helby> there is nothing what can help alpine wiki
2021-12-09 00:07:57 <lamineer> I would propose having a category tag for outdated articles. Then someone such as myself can systematically go through the list and update where needed. I can search for the template string obsolete or expand but that only amounts to less than maybe 20 articles combined.
2021-12-09 00:11:03 <zcrayfish> lamineer:  that sounds doable.
2021-12-09 00:13:40 <ktprograms> Speaking of the wiki, in https://wiki.alpinelinux.org/wiki/Gitea, since Gitea is in 3.15 community, the 'Installation' section doesn't need the messy 'add most deps from community' then 'switch to edge and add gitea'
2021-12-09 00:13:59 <ktprograms> Should I fix it or is there some other problem I don't know about?
2021-12-09 00:16:18 <ktprograms> On the wiki, is there a way to propose an edit without changing the user facing page? I think the Gitea page should be re-written to be more step-by-step (for example, currently the database configuration section is right at the bottom and it doesn't have any info about how to set up the users etc)
2021-12-09 00:18:55 <helby> sooner or later one smart person will delete whole wiki
2021-12-09 00:20:31 <zcrayfish> ktprograms: if you don't feel comfortable editing the article itself, hit the talk button and add a comment on the talk page explaining your concerns
2021-12-09 00:21:29 <ktprograms> zcrayfish: It's the 'Discussion' button, right? And is there any format for the discussion page?
2021-12-09 00:22:08 <zcrayfish> Yes. Not really, just sign your edit/addition of the talk page with ~~~~
2021-12-09 00:22:32 <ktprograms> zcrayfish: Ok
2021-12-09 00:39:11 <ktprograms> zcrayfish: https://wiki.alpinelinux.org/wiki/Talk:Gitea
2021-12-09 00:40:57 <zcrayfish> That's perfect. When I get back on my computer (I'm on mobile right now), I'll see if I can implement some of your proposals
2021-12-09 00:41:39 <ktprograms> zcrayfish: If you're ok with it, I can edit the wiki myself. I just wanted some feedback first.
2021-12-09 00:44:06 <Cheddoleum> Running Alpine 5.15 x86_64, ethtool coalesce features for igb driver are unavailable.  With another Linux like OpenWRT on a nearly identical platform it works fine.
2021-12-09 00:45:41 <Cheddoleum> Wondering if there's something about the way alpine builds the igb driver that limits the available features.
2021-12-09 00:45:47 <zcrayfish> I agree that a history of gitea is not needed. A very brief blurb at the beginning explaining what gitea is is ok... Points 2-4 are accurate
2021-12-09 00:45:58 <Hello71> openwrt uses old kernel, and on some devices, *very* old kernel
2021-12-09 00:48:34 <Cheddoleum> Not sure that matters here, ethtool -c has been around a very long time. the openwrt kernel in this case is 5.4
2021-12-09 00:51:05 <zcrayfish> Is there an option on the kernel driver itself for that? (Usually at/near the bottom of modinfo for the driver)
2021-12-09 00:51:36 <Cheddoleum> Checking the options...
2021-12-09 00:55:26 <Cheddoleum> There's something about dma coalescing in the docs, might be relevant.
2021-12-09 00:57:14 <ktprograms> zcrayfish: What do you mean by 'Points 2-4 are accurate'? As in you agree to remove it?
2021-12-09 00:58:00 <zcrayfish> Yes. For point 2 I'd link to the documentation, as you mentioned
2021-12-09 00:58:53 <ktprograms> zcrayfish: Ok. Would you like me to do it or are you going to do it?
2021-12-09 00:59:50 <zcrayfish> You can give it a shot and I'll review it in a little while 👍👍
2021-12-09 01:02:06 <ktprograms> zcrayfish: How does reviewing work?
2021-12-09 01:03:13 <zcrayfish> Informally. I'll just run the diff against your edit(s) and see if there is anything that could be improved upon. That's it really
2021-12-09 01:04:16 <ktprograms> zcrayfish: Meaning I just edit it and if you think changes should be made you'll just change it after that? So there's no 'staging' area for edits?
2021-12-09 01:07:47 <zcrayfish> That's right
2021-12-09 01:08:16 <ktprograms> zcrayfish: ok, thanks.
2021-12-09 01:08:22 <zcrayfish> np
2021-12-09 01:08:44 <minimal> Cheddoleum: could be a issue with kernel config for netlink
2021-12-09 01:09:23 <minimal> although CONFIG_ETHTOOL_NETLINK does appear to be "=y"
2021-12-09 01:35:08 <ktprograms> zcrayfish: Do you have experience with PostgreSQL?
2021-12-09 01:35:38 <zcrayfish> I'm afraid not
2021-12-09 01:37:44 <ktprograms> Ok, in that case I'll just add instructions for Mariadb to the Gitea page, and the postgresql instructions can be done later?
2021-12-09 01:41:29 <psykose> https://docs.gitea.io/en-us/database-prep/ lists how to use every database
2021-12-09 01:41:55 <zcrayfish> Yes
2021-12-09 01:42:10 <zcrayfish> you could include the URL psykose mentioned as well.
2021-12-09 01:42:17 <MAD_AS_HELL_> psykose, tell me what you thought my fucked up system's problem was
2021-12-09 01:42:20 <MAD_AS_HELL_> i guarantee you were WRONG
2021-12-09 01:42:47 <psykose> although it is a bit outdated, as scram-256-sha is now the default :)
2021-12-09 01:42:56 <ktprograms> zcrayfish: psykose: Ok, I'll add detailed instructions for mariadb since I know that works (I've tested), and like to the docs for other databases.
2021-12-09 01:43:00 <ktprograms> s/like/link
2021-12-09 01:43:00 <alpine-meetbot> ktprograms meant to say: zcrayfish: psykose: Ok, I'll add detailed instructions for mariadb since I know that works (I've tested), and link to the docs for other databases.
2021-12-09 01:43:04 <psykose> aye
2021-12-09 01:43:15 <zcrayfish> MAD_AS_HELL_: I would have to guess that the user neglected to take into account that *nix based systems have case-sensitive passwords and filenames. ;) ;) 
2021-12-09 01:43:16 <psykose> ah, only in 14
2021-12-09 01:43:42 <MAD_AS_HELL_> zcrayfish: unfortunately not
2021-12-09 01:44:09 <zcrayfish> Bummer. I didn't scroll up far enough to even attempt to diagnose your problem TBH.
2021-12-09 01:46:50 <MAD_AS_HELL_> it'd take years
2021-12-09 01:47:45 <zcrayfish> so did you figure it out?
2021-12-09 01:53:22 <Cheddoleum> minimal: thanks (at dinner)
2021-12-09 02:03:53 <MAD_AS_HELL_> zcrayfish: nah, i'm considering installing gentoo, admitting full defeat
2021-12-09 03:37:24 <ktprograms> zcrayfish: When you've taken a look at the changes and you think it's ok, please remove the Expand notice. Or do you want me to just do it first?
2021-12-09 05:23:01 <dhruvin> where should i ask questions about packaging software for alpine? here or in #alpine-devel?
2021-12-09 05:26:38 <psykose> either, here is fine
2021-12-09 05:32:35 <dhruvin> i'm stating what i've learned so far from reading other package definitions. correct me if im wrong. when packaging pythong packages: the source should be from pypi, the url should point to whatever is in Homepage section on pypi (not the url of pypi). skip tests that fail (do i have to comment why they fail?). specify !check in options, with a comment "no tests" if there are no
2021-12-09 05:32:37 <dhruvin> tests.
2021-12-09 05:33:53 <dhruvin> if i update some package, contributed by someone else, should i also update the Maintainer: ... comment on top with my information (leaving the Contributor: ... as is)?
2021-12-09 05:35:06 <ikke> dhruvin: source doesn't have to be pypi necessarily. We'd rather have tests fixed than skipped (or at least reported upstream).
2021-12-09 05:35:27 <ikke> dhruvin: the contributor field is not that important, I never add it
2021-12-09 05:45:08 <dhruvin> thanks ikke, guix has a command that helps with generating package definitions from various package repositories. the package definition file is pre-filled with data from said repositories. do we have something similar in alpine? i know about newapkbuild that generates APKBUILD from the information i give. is there something i'm missing?
2021-12-09 05:45:39 <ikke> We don't have something that contacts the source to fill in the template atm
2021-12-09 05:46:03 <dhruvin> okay
2021-12-09 05:47:17 <psykose> apkbuild-pypi can get you started with a pypi link, but it doesn't take super long to fill out like.. one source link and some fields anyway
2021-12-09 05:47:57 <psykose> apkbuilds are quite small and fast to write from just an empty newapkbuild, such is minimalism
2021-12-09 05:48:14 <ikke> most tricky thing is run-time deps
2021-12-09 05:48:17 <psykose> yep
2021-12-09 05:48:34 <psykose> especially because non-compiled script packages can't have the deps traced
2021-12-09 05:48:56 <psykose> so you either need to test in a minimal chroot, or read the python files to see what gets import'd, etc
2021-12-09 05:49:16 <psykose> some amount of cases of things missing a py3- dep after getting merged on install
2021-12-09 05:49:56 <dhruvin> re: apkbuild-pypi: i'll look into it. thanks.
2021-12-09 05:50:35 <dhruvin> yes, minimalism is what i've recently started appreciating a lot
2021-12-09 05:51:47 <psykose> if you ever want to see what a 'big' apkbuild looks like look at something like chromium, firefox, some of your bigger favorite server software, etc
2021-12-09 05:52:03 <psykose> even there it's fairly easy to understand, unless you look at that nginx templated module stuff
2021-12-09 05:54:00 <dhruvin> i was looking for a canonical way of defining things in APKBUILD. For example, there are some packages that use ${_pyname:0:1} and some that use ${_pyname%${_pyname#?}}. i'm relatively new to shell scripting and hence to linux in general. i'm not certain which to use when.
2021-12-09 05:54:50 <dhruvin> psykose: will see firefox one :)
2021-12-09 05:55:36 <psykose> _pyname seems to just be the name of the upstream minus py3-
2021-12-09 05:55:47 <psykose> you can just write it out, no point in using shell tricks for a static name
2021-12-09 05:56:06 <ikke> yeah, I'm with psykose 
2021-12-09 05:56:33 <dhruvin> re: canonical way: i believe APKBUILD file authors have more autonomy as to how they see it should be, than other distros. right?
2021-12-09 05:56:40 <psykose> you will come across some amount of inconsistency as things have been updated over time, old things didn't, etc, but just do what seems simplest
2021-12-09 05:56:47 <psykose> dhruvin: pretty much, and you do get it reviewed in the mr
2021-12-09 05:56:54 <psykose> obvious stuff should get caught by that
2021-12-09 05:58:22 <ikke> https://gitlab.alpinelinux.org/alpine/aports/-/blob/master/CODINGSTYLE.md
2021-12-09 05:59:16 <dhruvin> okay. seems good. i'll soon submit patches for todoman (and dependencies) and hledger.
2021-12-09 06:00:25 <psykose> and the COMMITSTYLE.md to tell you how to lay out the commits
2021-12-09 06:00:29 <psykose> it's fairly simple
2021-12-09 06:02:43 <dhruvin> should i submit a patchset with one commit per package? for example to update todoman, i need added py3-sphinx-click, py3-sphinx-theme-rtd, py3-sphinx-autorun, etc. the patchset should contain one commit per said dependency and the last commit will update todoman. is that right?
2021-12-09 06:03:18 <dhruvin> don't have to answer if that's written in COMMITSTYLE
2021-12-09 06:03:52 <dhruvin> (got it from COMMITSTYLE. ignore the question above)
2021-12-09 06:04:04 <dhruvin> thanks ikke and psykose for your help :)
2021-12-09 06:04:10 <psykose> glad to be of help :)
2021-12-09 07:44:16 <kmmndr> Hi all :-)
2021-12-09 07:44:49 <psykose> hey
2021-12-09 07:45:01 <kmmndr> Is there something to do in order to trigger snapshot function on gitlab runners ?
2021-12-09 09:11:33 <ikke> No
2021-12-09 09:11:45 <ikke> kmmndr: what do you want to achieve?
2021-12-09 09:38:33 <kmmndr> ikke: I'd like my MR (https://gitlab.alpinelinux.org/alpine/aports/-/merge_requests/18552/diffs#88e47fe095d1ef998a33d542a080d384239005bc) to pass CI :-)
2021-12-09 09:40:37 <ikke> Is there a reason you cannot use archives from upstream?
2021-12-09 09:40:47 <ikke> https://repology.org/project/tuxpaint/information#Downloads lists a few
2021-12-09 09:41:46 <kmmndr> I haven't found any archive of the SDL2 version
2021-12-09 09:43:19 <kmmndr> Tuxpaint authors agreed to create a tag for each release, but they haven't completly decided to switch to sdl2
2021-12-09 09:44:06 <psykose> how do you get the -sdl2 version yourself
2021-12-09 09:44:19 <ikke> git archive from a sdl2 suffixed tag
2021-12-09 09:44:42 <psykose> the default archive i got off their site has a makefile with a bunch of sdl references in it, i'm not sure what is different in this sdl version
2021-12-09 09:46:33 <psykose> in any case something from here works fine i think https://sourceforge.net/p/tuxpaint/tuxpaint/ci/tuxpaint-0_9_27-2021-11-25-sdl2/tarball?path=
2021-12-09 09:51:17 <kmmndr> psykose: The link does not work properly
2021-12-09 09:51:30 <frag> anyone else experiencing screen-freeze in firefox? if i toggle workspace back and fort it updates, i can experience the same in telegram-desktop
2021-12-09 09:52:00 <psykose> kmmndr: it loads for me, it's not a tarball itself- something on the page should be, the download link on it packages a specific commit i guess
2021-12-09 09:52:36 <kmmndr> Sourceforge has a redirection to a page where download begins when the archive has been generated, but they often clean the cache
2021-12-09 09:53:01 <kmmndr> This means that the download link won't work very long
2021-12-09 09:57:08 <psykose> hm
2021-12-09 09:57:46 <psykose> no clue then, i guess upstream should properly tag things, but aside from that no idea
2021-12-09 10:19:17 <ikke> kmmndr: they do offer a direct link?
2021-12-09 10:19:22 <ikke> https://sourceforge.net/code-snapshots/git/t/tu/tuxpaint/tuxpaint.git/tuxpaint-tuxpaint-fa071f85b841e7d07b065ea66e613d79124d1af7.zip
2021-12-09 10:23:52 <psykose> that was the one on the page, so i assume that's the one they mean expires
2021-12-09 11:51:36 <frag> 'apk search -e cmd:mount' doesnt list busybox
2021-12-09 11:51:50 <ikke> Because it's not part of the package
2021-12-09 11:52:10 <ikke> The commands are created as symlink to the bb binary in a post-install step
2021-12-09 11:52:18 <ikke> (or rather a trigger)
2021-12-09 11:54:08 <frag> yeah not sure how its intended to work, was just a bit surprised :]
2021-12-09 12:28:04 <frag> is there a column alternative (in busybox)? seems the only command i want from util-linux-misc
2021-12-09 12:52:38 <jhjacobs81> anyone around who got Wireguard working on Alpine?
2021-12-09 13:18:13 <aru_> jhjacobs81: I had it working some time ago. What seems to be the issue?
2021-12-09 16:55:38 <Saijin_Naib[m]> dhruvin: thanks for updating Sphinx dependencies! 
2021-12-09 17:10:22 <user1> Hello, I'm trying out Alpine (specifically, Plasma on Alpine) with an Intel iGPU (Ivy Bridge to be precise) and I've got it to run, but for some reason it seems to use software rendering instead of hardware acceleration. Anyone know of a similar situation?
2021-12-09 17:11:19 <user1> As far as I understand Alpine's Wiki suggests that I shouldn't need to tweak anything with regards to e.g. Xorg settings and boot parameters so I'm not sure what's gone wron here.
2021-12-09 17:12:28 <user1> Admittedly I've only logged into the Wayland session so far (but that's what I would want to use anyway) and it's still alpine 3.14 (I started playing with it before 3.15 was released)
2021-12-09 17:12:39 <Hello71> how did you determine that it is using software rendering
2021-12-09 17:16:10 <user1> I don't have the strongest of all evidence.
2021-12-09 17:16:15 <user1> Dragging Windows feels subjectively way worse than it does on the other OS on the computer (running GNOME, which I must admit is a different OS) and when I switch from software to irc OpenGL on the renderer setting app the session crashes at the next login and the app shows software rendering again. Plus changing the OpenGL2/3/Xrender setting to any of the OpenGL options makes the system hang.
2021-12-09 17:17:28 <user1> I mean gnome is a different shell (and compositor when running wayland) of course
2021-12-09 17:18:24 <Hello71> post glxinfo
2021-12-09 17:18:59 <Hello71> and also /etc/apk/world
2021-12-09 17:20:04 <user1> okay, I'll get to that at some point (hopefully soon)
2021-12-09 17:25:23 <user1> Actually I can just load the /etc/apk/world now
2021-12-09 17:26:13 <user1> alpine
2021-12-09 17:26:13 <user1> alpine-base
2021-12-09 17:26:13 <user1> !bash
2021-12-09 17:26:13 <user1> iwd
2021-12-09 17:26:13 <user1> eudev
2021-12-09 17:26:14 <user1> !linux-firmware
2021-12-09 17:26:16 <user1> linux-firmware-none
2021-12-09 17:26:18 <user1> linux-lts
2021-12-09 17:26:20 <user1> mate-terminal
2021-12-09 17:26:22 <user1> mesa
2021-12-09 17:26:24 <user1> networkmanager
2021-12-09 17:26:26 <user1> pipewire-pulse
2021-12-09 17:26:28 <user1> plasma
2021-12-09 17:26:31 <hoel> You shouldn’t paste that in the channel.
2021-12-09 17:26:32 <user1> sudo
2021-12-09 17:26:34 <user1> wireless-tools
2021-12-09 17:26:36 <user1> wpa_supplicant
2021-12-09 17:26:38 <user1> xf86-input-libinput
2021-12-09 17:26:40 <user1> xorg-server
2021-12-09 17:26:42 <user1> (I did experiment with forbidding some packages there)
2021-12-09 17:26:44 <user1> Oops sorry
2021-12-09 17:26:59 <hoel> Put it on a pastebin and send the link here.
2021-12-09 17:29:25 <user1> https://pastebin.com/A6xuwhW3
2021-12-09 17:36:02 <minimal> user1: I think you're missing linux-firmware-i915 (IvyBridge uses i915 graphics driver AFAIK)
2021-12-09 17:39:26 <minimal> user1: also x86-video-modesetting or xf86-video-intel?
2021-12-09 17:40:55 <minimal> doh! ignore xf86-video-modesetting, its a dummy package
2021-12-09 17:42:50 <Hello71> iirc i915 firmware is only needed for power saving and hw decoding
2021-12-09 17:43:02 <Hello71> it should technically work without it but not as good
2021-12-09 17:43:47 <Hello71> main problem is that mesa is (nearly) empty package
2021-12-09 17:43:59 <Hello71> you need mesa-dri-classic
2021-12-09 17:46:14 <user1> OK I'll try mesa-dri-classic the next time I boot to Alpine. Thanks for now!
2021-12-09 17:47:25 <minimal> Hello71: what about mesa-dri-gallium instead? https://airlied.blogspot.com/2021/04/crocus-gallium-for-gen4-7-generation.html
2021-12-09 17:47:39 <Hello71> still experimental
2021-12-09 17:47:44 <Hello71> i guess to be safe you can install mesa-dri
2021-12-09 17:48:06 <Hello71> then it will switch to crocus whenever that becomes default upstream
2021-12-09 17:51:20 <user1> mesa-dri? I don't find anything under that name in pkgs.alpinelinux.org. But I guess I'll just try adding that to see if it works first.
2021-12-09 17:57:07 <Hello71> hm, possibly i am misremembering
2021-12-09 17:57:13 <Hello71> i can't find it either
2021-12-09 19:56:55 <mipmeb>   c /  e  r t  a i n l y     ,    z	 a m e r i c a n  s 	n e e d e d       s o m e t h i n  g     l i  k e   9  /  11    t  o    	 J u s t i f y      	i n v a  d i n g  -  	i r -  a  q	   w h i c h    h  a s   b e e n   		a l r e a d y   	p la n n e d 		a s 	  a  p a r t  	o f   	c r e  a t i v e    	c -h a  o s    D    i   d     u  s  a    t r a i n   &      s u p p l y     i  s  i  s     w i t h  
2021-12-09 19:57:02 <mipmeb>    w e a p  o  n s     l i k e   i t    d i d     w i  t h    a  l  - q  a- e  d a    t o 		j u s  t i f y 		c r e a  t i n g       	w  a  r .s      C  R  E - A T  - I V E	   	C  H  A - O  S   	   	d  i  d     	c   i   a    -   d i d   	 9     /   11   	 o r  	i t  	j u s t    	l e  t    i t   	h a  p p e n   	 	p l   a n  	 f o r  	  m i d d l e	     -e a s t   	i f      a l q a e d a 	d i d 		i t 
2021-12-09 19:57:06 <mipmeb>       	W  H  Y   t o     	 k - i - l - l     9	 m i l l i o n   		i r -a q i - s 	a b o u t    	9   m i l l i  o n        	i r a q  i -   d i -e  d   	  E i t h e r    -	b e c a u s e  - 	o f  	d i r e c t 		a m e r ic a n  	v i o  l e n c e 	a n d  b e c a u s e  	o f  	s i e g e   	u s a  	m a d e    a g a i n s t    	i r - a q   	t h a t  		p r e v  e n t e d    i -r a -q i    	f r o m  		e s s  
2021-12-09 19:57:11 <skintimidation> oh hell yeah
2021-12-09 19:57:11 <mipmeb> e n t i a l 	l i v i n g 	m  a t  e r i a l s  , 	f o o d  , 		m e d i c i n e	 	a n d  	e v e n  	p e n c i l s 	f o r	  z  c h i l d e r n  - 		i s i s 	  a n d	   a l q-  a e  d  a -   o n l y     a p p e a r e d    i n    i r - a q       a f  t e r   u s a  	i n v a s i o n  ,  	D i d    	u  s  a     	p  a  v e   	t h e  w  a y     for    	i  s- i  s    t o   		i   r   a   -q  ?  		s a  d  -d  
2021-12-09 19:57:17 <mipmeb> -a  m  	h u s- s i  n		 who   	 l o s t    	m o s t   of  his    	p o -w  e r   	in  	1  st ,  2 nd  g u l f  	w a r s	  	a n d 	d u r i n g      	 1 0   	 	y e a r s  	  	of    	s i e g e   ,  	d i d   	n o t  	a l l o w       	i  s  ` i  s        or    		a l , q a , e  d e a   	t o   	e n t e r      	i  r  -a  q   ,   	e v e n    	t o `    		h e  l p   	h i  m    	a g a  i n 	s	t    	u s  a
2021-12-09 19:57:21 <skintimidation> -a  m  	h u s- s i  n		 who   	 l o s t    	m o s t   of  his    	p o -w  e r   	in  	1  st ,  2 nd  g u l f  	w a r s	  	a n d 	d u r i n g      	 1 0   	 	y e a r s  	  	of    	s i e g e   ,  	d i d   	n o t  	a l l o w       	i  s  ` i  s        or    		a l , q a , e  d e a   	t o   	e n t e r      	i  r  -a  q   ,   	e v e n    	t o `    		h e  l p   	h i  m    	a g a  i n 	s	t    	u s  a
2021-12-09 19:57:28 <skintimidation> fuck, it doesn't work for me as well
2021-12-09 19:57:33 <skintimidation> c / e r t a i n l y , z a m e r i c a n s n e e d e d s o m e t h i n g l i k e 9 / 11 t o J u s t i f y i n v a d i n g - i r - a q w h i c h h a s b e e n a l r e a d y p la n n e d a s a p a r t o f c r e a t i v e c -h a o s D i d u s a t r a i n & s u p p l y i s i s w i t h 
2021-12-09 19:57:44 <psykose> it looks the same
2021-12-09 19:57:48 <psykose> also, don't join the spam
2021-12-09 19:57:58 <skintimidation> don't tell me what to do with my dead channel psykose
2021-12-09 19:58:01 <skintimidation> >:(
2021-12-09 19:58:09 <skintimidation> (he's frowning because he's sad)
2021-12-09 19:58:11 <skintimidation> (it's a he)
2021-12-09 19:58:23 <skintimidation> i'll do you one better psykose
2021-12-09 19:58:31 <skintimidation> 1v1 me on halo infinite snipers only
2021-12-09 19:58:32 <Tulala> Good evening!
2021-12-09 19:58:44 <psykose> don't have it
2021-12-09 19:58:48 <skintimidation> evening, tulala
2021-12-09 19:58:52 <skintimidation> what about xonotic, psykose
2021-12-09 19:59:04 <psykose> never even heard of that
2021-12-09 19:59:10 <skintimidation> open source quakelike
2021-12-09 19:59:24 <Tulala> Quick question: Is the package "linux-edge" part of the edge repositories and thus unsafe to use or is it simply the very latest kernel, but production ready? Thanks in advance
2021-12-09 19:59:39 <MAD_AS_HELL_> the very latest kernel is not production ready.
2021-12-09 19:59:53 <psykose> both -lts and -edge are the same version currently in edge
2021-12-09 19:59:57 <MAD_AS_HELL_> but nothing is production ready when your production isn't a useful definition.
2021-12-09 20:00:05 <psykose> -edge is also technically better for desktop use
2021-12-09 20:00:19 <MAD_AS_HELL_> psykose: but -edge isn't the actual latest kernel available to the world, right?
2021-12-09 20:00:30 <psykose> it's latest release branch
2021-12-09 20:00:30 <MAD_AS_HELL_> i thought alpine didn't package it.
2021-12-09 20:00:32 <psykose> non-rc
2021-12-09 20:00:36 <MAD_AS_HELL_> ah.
2021-12-09 20:00:42 <MAD_AS_HELL_> Tulala: thoughts?
2021-12-09 20:00:56 <psykose> current release branch is also stable, so lts is also there
2021-12-09 20:01:06 <psykose> but when 5.16 becomes stable only edge goes forth
2021-12-09 20:01:10 <MAD_AS_HELL_> why do people insist on treating irc like email in the modern day
2021-12-09 20:01:26 <MAD_AS_HELL_> tulala literally JUST joined
2021-12-09 20:01:30 <MAD_AS_HELL_> and then left terminal
2021-12-09 20:01:40 <Tulala> Okay it does need to be a bleeding egde kernel, but just stable enough to use it safely as an internet facing router which serves internally a ksmbd server
2021-12-09 20:01:52 <MAD_AS_HELL_> see? could have started with that
2021-12-09 20:02:05 <psykose> is there a reason it has to be bleeding edge? if it has the features on 5.15 it will be fine in the future too
2021-12-09 20:02:12 <psykose> unless you are implementing things on the latest stable release perpetually
2021-12-09 20:02:33 <psykose> but yes, it's not -rc, just latest stable, and it should be fine for that
2021-12-09 20:02:47 <Tulala> Oh sorry, I have misspelled: It does NOT need to be a bleeding edge kernel 😅 
2021-12-09 20:02:52 <psykose> haha
2021-12-09 20:03:05 <psykose> well if 5.15 is what you need you are fine with 3.15-alpine 5.15 lts as well
2021-12-09 20:03:14 <MAD_AS_HELL_> is it possible for people to build initrds for others? are they transposable? do they depend on a specific kernel version?
2021-12-09 20:03:40 <psykose> they include some modules that are signed to the current kernel to my knowledge
2021-12-09 20:03:46 <Tulala> Jep basically 5.15 is enough, but the currently used kernel does not have the ksmbd module included. the linux-edge kernel does
2021-12-09 20:03:53 <psykose> so if an alpine system user packaged it, it should be fine- same sigs for same kernel version
2021-12-09 20:04:25 <psykose> Tulala: ah yeah, it was forgot to be enabled, you are right
2021-12-09 20:04:55 <psykose> but yeah edge 'should' be okay, just know the system can be broken randomly
2021-12-09 20:05:02 <Tulala> Oh okay. So if I wait a little it will be enabled like in alpine linux 3.15.1?
2021-12-09 20:05:20 <MAD_AS_HELL_> psykose: ah damnit, big damnit
2021-12-09 20:05:25 <psykose> you would have to ask ncopa
2021-12-09 20:05:32 <Tulala> @pykose okay a random breakdown is not a good thing, thus I will wait
2021-12-09 20:05:48 <minimal> 192.145
2021-12-09 20:05:56 <psykose> i think it would be fine, i'll ask him if i remember when he's online
2021-12-09 20:06:38 <Tulala> Thanks 👍 
2021-12-09 20:07:13 <psykose> you can also make a gitlab issue
2021-12-09 20:07:17 <psykose> i think that is better
2021-12-09 20:07:31 <Tulala> Okay I have to check that.
2021-12-09 20:07:33 <psykose> just ping him in there and he should see it tomorrowish
2021-12-09 20:07:47 <ikke> (or rather a trigger)
2021-12-09 20:08:55 <Tulala> Which repository is the right one to create the very issue in?
2021-12-09 20:08:59 <minimal> Tulala: thought there was an Issue to track adding ksmbd to kernel but don't see it
2021-12-09 20:09:36 <psykose> alpine/aports
2021-12-09 20:10:00 <psykose> also CONFIG_SMB_SERVER_SMBDIRECT should be enabled too i think, disabled on edge as well it seems
2021-12-09 20:14:35 <Saijin_Naib[m]> Does anyone run fwupd under Edge? Mine seems to be misbehaving and I'm not sure if that's a local issue or not
2021-12-09 20:14:46 <Saijin_Naib[m]> 
ACTION uploaded an image: (12KiB) < https://matrix.org/_matrix/media/r0/download/matrix.org/vXvjuDOOfEzezRlYjHjnwoTL/image.png >

2021-12-09 20:19:18 <Tulala> Another quick question: On an older machine I have been using alpine 3.12. Later on if have upgraded it to alpine 3.14 by editing the /etc/apk/repositories file (using the main repository). Altough the system has been updated without any issues I am still stuck with an older version of the linux kernel. How can i upgrade the kernel itself as well, as apk add linux-lts does not apply the update?
2021-12-09 20:19:38 <Saijin_Naib[m]> Have you issued an apk update -al?
2021-12-09 20:19:55 <Saijin_Naib[m]> That should reset all packages to track the current repository and fetch their latest versions
2021-12-09 20:20:02 <Saijin_Naib[m]> sorry
2021-12-09 20:20:07 <Saijin_Naib[m]> apk upgrade -al
2021-12-09 20:20:12 <ikke> Tulala: make sure you have linux-lts, not linux-vanilla or linux-hardened
2021-12-09 20:23:47 <Tulala> Yes, it the linux-lts package is installed. I will try the apk upgrade -al command shorty
2021-12-09 20:23:54 <vlp[m]> hey, I have added to the wiki the process for grub2/luks2 FDE. If anyone want to try, I would be happy to have your feedback. (thanks minimal  for your help. I have added a warning to explain that it coud be reasky): https://wiki.alpinelinux.org/wiki/LVM_on_LUKS
2021-12-09 20:31:12 <skintimidation> vlp: Have you covered installation where /boot is a completely different disk?
2021-12-09 20:31:37 <Tulala> Okay, apk upgrade -al did not work. I am still on the older 5.4 kernel and not on the available 5.10 kernel. Any suggestions?
2021-12-09 20:32:35 <vlp[m]> skintimidation: at that time I have just supposed that /boot is into / or into an other dedicated partition but not in a complete different disk
2021-12-09 20:33:04 <vlp[m]> but it won't change a lot
2021-12-09 20:37:27 <psykose> Tulala: can you post /etc/apk/repositories and world
2021-12-09 20:37:31 <Saijin_Naib[m]> yep
2021-12-09 20:37:49 <Saijin_Naib[m]> Are you tracking latest-stable as your release in your repositories file?
2021-12-09 20:39:03 <minimal> Tulala: try adding "-U" to the upgrade command
2021-12-09 20:40:13 <minimal> vpl[m]: a suggestion would be to replace haveged with rng-tools
2021-12-09 20:40:23 <minimal> s/vpl/vlp/
2021-12-09 20:40:23 <alpine-meetbot> minimal meant to say: vlp[m]: a suggestion would be to replace haveged with rng-tools
2021-12-09 20:40:52 <Tulala> The repositories files contains the http://dl-cdn.alpinelinux.org/alpine/v3.14/main as well as the 3.14 community repository (using the @com tag)
2021-12-09 20:41:22 <Tulala> Unfortunately apk upgrade -U also does not do any upgrade 😬 
2021-12-09 20:41:56 <ikke> if it has a pin, it's not used by default
2021-12-09 20:41:58 <minimal> vlp[m]: also even if you don't overwrite with random data it is important to have either haveged or ideally rngd (from rng-tools) running before you generate the crypto_keyfile.bin file
2021-12-09 20:42:47 <psykose> Tulala: you can remove the @repo from all the world entries by hand, then upgrade
2021-12-09 20:42:54 <psykose> and remove the @ in repositories too
2021-12-09 20:45:01 <minimal> skintimidation: as long as /boot is mounted when grub-install is run then it should figure things out ok in a multi-disk setup
2021-12-09 20:45:08 <Tulala> There is no @repo anywhere and I have removed the @com tag as well. Now apk upgrade -U does not let me continue, as I have installed tagged packages like docker@com before
2021-12-09 20:46:12 <Tulala> I am going to delete all those packages now. The server is going to be fully reinstalled anyway, I was just curious why the kernel hold back occured in the first place
2021-12-09 20:47:13 <vlp[m]> minimal: haveged  is mentioned at the beginning, but I'll look at rng-tools ! 
2021-12-09 20:48:06 <Tulala> Interessting. By deleting all @com tagged packages the new kernel has automatically been installed. I guess there was a dependency left, which hold the kernel back? 🤔 
2021-12-09 20:48:54 <minimal> vlp[m]: haveged was mentioned only in the context of if you wanted to securely wipe the disk
2021-12-09 20:49:16 <vlp[m]> minimal: yes, you are right
2021-12-09 20:49:43 <minimal> vlp[m]: rngd (from rng-tools) can use both software "jitter" (like haveged) and also any hardware RNG if present as entropy sources
2021-12-09 20:51:33 <skintimidation> minimal: will efibootmgr write the correct disk over to it? will that disk be interchangeable with other boot media?
2021-12-09 20:52:58 <minimal> minimal: not sure, never used efibootmgr. grub-install should write a grub.cfg file that references the UUID of the boot partition regardless of which disk its on
2021-12-09 20:54:26 <skintimidation> aye, though how does the boot process get involved? does grub-install handle efi vars?
2021-12-09 20:56:30 <psykose> Tulala: with the @ the world is invalid i guess, you can just remove the @ part of them :) happy upgrade
2021-12-09 20:58:30 <minimal> skintimidation: yes, unless you pass grub-install the "--no-nvram" parameter it will manipulate EFI variables
2021-12-09 20:59:04 <Tulala> Thanks @psykose. The upgrade to the latest kernel went fine afterwards.
2021-12-09 21:00:02 <minimal> skintimidation: I tend to build single OS boot disk images so don't need to set EFI vars but rather use the "fallback" path location instead
2021-12-09 21:04:12 <minimal> skintimidation: grub.cfg is confusing as the "--set=root" UUID shown for the "search" command is the *boot* UUID, not the root UUID whereas the root UUID does appear in the "linux" command (which is the full /proc/cmdline passed to kernel)
2021-12-09 21:10:33 <Tulala> Regarding the installation of alpine: Is there any reason for installing chronyd (which is the default) over the busybox for handling ntp stuff? In my experience chronyd delays the boot by a view seconds, but other than that I did not see any difference?
2021-12-09 21:12:42 <psykose> it supports more ntp settings and is more.. power efficient maybe? i'm not sure if it's meaningful
2021-12-09 21:13:10 <psykose> can also run ntp server with it
2021-12-09 21:14:03 <Saijin_Naib[m]> Oh, that's surprising. I would have thought that the BusyBox stuff would be the most power efficient
2021-12-09 21:14:31 <Tulala> Running an ntp server is with the busybox version possible as well
2021-12-09 21:16:17 <psykose> ah
2021-12-09 21:19:10 <jason123onirc> I was wanting to get rid of chronyd and replace it with busybox ntp. It delays boot by a few secconds and all I want is for the time to be set correctly. 
2021-12-09 21:20:20 <jason123onirc> How would I replace chronyd with busybox ntp?
2021-12-09 21:20:40 <Tulala> For using as a server or just as client?
2021-12-09 21:21:58 <psykose> rc-service chronyd stop, apk del chrony, rc-update add ntpd default, rc-service ntpd start
2021-12-09 21:22:01 <psykose> or something along those lines
2021-12-09 21:22:46 <jason123onirc> I can try that
2021-12-09 21:22:55 <jason123onirc> I just want the client to sync the time 
2021-12-09 21:28:11 <jason123onirc> Yeah that did work fine. Now my boot time is even faster thanks 
2021-12-09 21:35:16 <Hello71> minimal: or --removable
2021-12-09 21:35:48 <Hello71> jason123onirc: it would be better to turn off waiting for chrony timesync
2021-12-09 21:36:01 <Hello71> otherwise it's not faster, it's just "faster"
2021-12-09 21:36:16 <ikke> remove initstepslew
2021-12-09 21:39:53 <minimal> Hello71: not entirely sure of the differences between "--removable" and "--no-nvram", I use both
2021-12-09 21:40:09 <Hello71> --removable puts it at bootx64.efi
2021-12-09 21:40:43 <minimal> right, whereas "--no-nvram" does not write EFI vars - which is what was asked about
2021-12-09 21:42:53 <Hello71> --no-nvram is not usually useful on its own because the firmware will be unable to locate the file
2021-12-09 21:43:39 <minimal> like I said I use both (and other options) when installing grub inside a chroot
2021-12-09 21:44:15 <minimal> been meaning to experiment with startup.nsh
2021-12-09 21:51:10 <jason123onirc> Hello71: I am now just using busybox ntp and it works fine  
2021-12-09 21:52:07 <Hello71> sure, a lot of things will work fine. ubuntu "works fine" for a lot of people, but we use alpine for a reason (or many reasons)
2021-12-09 21:52:36 <Hello71> if busybox ntp didn't work at all then we would delete it
2021-12-09 21:53:34 <jason123onirc> that is true 
2021-12-09 21:55:42 <Hello71> busybox ntp isn't *awful* but using it because the init script doesn't wait for sync isn't a good reason
2021-12-09 21:56:47 <minimal> Hello71: agreed, the whole reason chronyd pauses it to ensure any necessary time change is made before other processes/daemons are started by the rest of the boot sequence
2021-12-09 21:57:13 <Hello71> yes
2021-12-09 21:57:22 <Hello71> if you want it to not wait then just turn that off
2021-12-09 22:01:43 <minimal> in a similar view someone a while ago wanted to change udhcpc (or was it dhcpcd) so that it didn't pause boot for a while (wanted it to background immediately), which would mean that other stuff later in the boot sequence that expected network (like ntpd/chronyd for example) might be affected if network was not up
2021-12-09 22:05:01 <jason123onirc> What is bad about replacing crony with busybox ntp? 
2021-12-09 22:08:23 <psykose> i mean i did basically that, by backgrounding networking
2021-12-09 22:08:33 <psykose> nothing in my setup depends on the network though, so works for me
2021-12-09 22:20:33 <minimal> jason123onirc: we were not saying its wrong to replace chrony with busybox ntp, rather that whichever is used should pause boto until the time is synced - which is what you initially indicated was what you wanted to avoid
2021-12-09 22:20:46 <minimal> s/boto/boot/
2021-12-09 22:20:46 <alpine-meetbot> minimal meant to say: jason123onirc: we were not saying its wrong to replace chrony with busybox ntp, rather that whichever is used should pause boot until the time is synced - which is what you initially indicated was what you wanted to avoid
2021-12-09 22:23:40 <jason123onirc> Yeah I could of paused it till boot but its easier to just replace it 
2021-12-10 05:30:20 <anjan> hi all, I want to use alpine but I rely on glibc programs - specifically matlab. Can I run it on alpine?
2021-12-10 05:30:26 <anjan> Anyone got it running on alpine?
2021-12-10 05:39:48 <deckard> anjan: you might be able to get by with octave, which is in the community repo
2021-12-10 05:39:58 <deckard> in many ways it's a clone of matlab
2021-12-10 05:40:04 <anjan> deckard: I cant get by with octave
2021-12-10 05:40:07 <anjan> I would if I could
2021-12-10 05:40:08 <deckard> ah okay
2021-12-10 05:40:25 <anjan> have you ran graphical programs in chroot? de
2021-12-10 05:40:27 <anjan> deckard:
2021-12-10 05:41:36 <deckard> i havent, sorry. my install is a headless server
2021-12-10 05:41:44 <anjan> okay....
2021-12-10 05:41:53 <anjan> I guess I can install it in a virtual machine lol
2021-12-10 05:43:24 <anjan> Im really frustrated that debian doesnt run the latest browser
2021-12-10 05:43:37 <anjan> and I dont like any of the other mainstream distros.
2021-12-10 06:31:09 <pomwv> //  …p l e a s e      .s h `a r e    m  y .`    q s  2  ` .l e s s e n   ..  u s a .     a g g r` e s s i o n    a g a` i n s t    o t h e r s.  D i d     u` s a   t  r a i n   ,  &   s u p `pl y     i  s ` i  s         wi th   w e. a  .p o n s   l i k e     i t     d i d    w i t h       a. l    q. a e d a  to    j. u  st i f y  c r  e a t i n g    w. a r  s         C. R E A T  I V E     C .H AO S-  d i d       c i` a      d i d  
2021-12-10 06:31:09 <pomwv>   9. /.11  	.  or,    i t    j u s t     l e t    `    it      h a p p en   *   c er t a i n l y     ,the  .  .. a m e  r i c` a n s   	  ne ed e d     something l   i k e      9. /  1 1   to   j /u s t i f y  i n v `a  d i n g   .i r  /a q   `  w h i c h/  has   b e e n ...alrea dy     p l a n  n e d  as   a  p art    of   c r e/ a t i v e    c h  a` o s   p l a n  for .    m i d d l e   e a s t
2021-12-10 06:31:10 <pomwv> if       .a l `  q a` e d a       / ` di d    ..	it     w h y  ,  `   t o    k `i l l     9   . m i l` l i  on    .  i  r` a q i s 
2021-12-10 06:31:13 <pomwv> a b o u t      9   m `il i o n     -   r a `q i    l o s t  `  th e i r  li` ves      e i th e r   be ca uz .  o f   dir e ct ameri`can   viol e`nce ` and ..  .	beca uz /of  si e`ge   .    u  s a    ma de        . a`g a in s t     i r a q  t h at   pr eve nt ed    i r a  qi   f or m    es se nt ial    `  l i v i n   g    `  ma t e   ri a l s ,  food , m e d i c i n e   
2021-12-10 06:31:18 <pomwv> i  s . i  s      /a n d     a l q  a ` e da      o n l y   a p p e  a  r e d     . `i n.       i r `  a  q    a f t e r    u s a    i n v a` si on  , D i d    u s a    . p a v e.   th e  /w ay   ..?f or     i s` i s   to  i r  `a q ?
2021-12-10 06:31:22 <pomwv>  sa d  `d a m        h u `s  si n    .w h o  l o s t ` .  m o s t   . of   h i s    p o w er  i n 1st , 2 nd  gu `lf  w a    `rs  an d     .d u ri n g. / 10    y ea rs    of   . ` s  i `e g e    ,.   d i d    n o t   a l l o w     i s i s    o r    a l q a e d e a      t o   e n  t e r  /   i  r  /  a  q .  e ve n   to    h e l p    h i m        a g a i n s t       u` s a
2021-12-10 06:31:27 <pomwv> `p l e a s e      .s h `a r e    m  y .`    q s  2  ` .l e s s e n   ..  u s a .     a g g r` e s s i o n    a g a` i n s t    o t h e r s
2021-12-10 08:42:10 <xsteadfastx> i tried to upgrade my raspberry pi from 3.14 to 3.15 and get this: fetch http://dl-cdn.alpinelinux.org/alpine/v3.15/main/aarch64/APKINDEX.tar.gz
2021-12-10 08:42:12 <xsteadfastx> ERROR: http://dl-cdn.alpinelinux.org/alpine/v3.15/main: UNTRUSTED signature
2021-12-10 08:42:18 <xsteadfastx> any ideas?
2021-12-10 08:43:10 <ikke> https://alpinelinux.org/posts/Alpine-3.15.0-released.html#upgrade_notes
2021-12-10 08:43:32 <ikke> https://wiki.alpinelinux.org/wiki/Release_Notes_for_Alpine_3.15.0#New_package_signing_keys
2021-12-10 08:43:39 <xsteadfastx> yeah thanks :)
2021-12-10 13:03:55 <dhruvin> i think creating MR from ~alpine/aports@lists.sr.ht doesn't work with patchsets. is that right?
2021-12-10 13:04:09 <dhruvin> https://lists.alpinelinux.org/~alpine/aports/patches/3795
2021-12-10 13:05:21 <dhruvin> (i see many patches sent to the mailing list, but not patchsets)
2021-12-10 13:06:15 <ikke> ddevault: ^
2021-12-10 13:06:37 <ddevault> it should work
2021-12-10 13:06:42 <ddevault> send me an email and I will investigate when I have time
2021-12-10 13:07:43 <dhruvin> im looking for MR here: https://gitlab.alpinelinux.org/alpine/aports/-/merge_requests, is that the correct location to look for?
2021-12-10 13:09:20 <dhruvin> ddevault: will do
2021-12-10 16:50:00 <skintimidation> ddevault: Go fucking kill yourself you useless swine
2021-12-10 16:50:18 <skintimidation> Can't believe I finally got the opportunity to say it to your stupid fucking face
2021-12-10 16:50:37 <MAD_AS_HELL_> ^
2021-12-10 17:03:34 <Hello71> ikke: you missed one
2021-12-10 17:05:43 <MAD_AS_HELL_> yeah, ddevault's still in here
2021-12-10 18:53:52 <cybb38> Hi all, does anybody here know how to get apache2 webserver running on Alpine w/ mod_dav_svn? 
2021-12-10 18:56:22 <nero> whats the problem?
2021-12-10 20:06:44 <asdflkj> anybody know if there's a bug in the imake package? might be fixed in 3.15, I'm still on 3.14/pmOS stable, but this worked fine on debian buster: https://termbin.com/yga4
2021-12-11 05:51:45 <dhruvin> when building hledger, should I enable static executable?
2021-12-11 05:52:43 <dhruvin> i built one without and it worked fine too (as in it was able to detect library dependencies). i wonder what's the right way to go now.
2021-12-11 05:54:26 <dhruvin> fwiw, haskell example APKBUILD doesn't mention --enable-static-executable, but the document seems outdated. so i asked it here.
2021-12-11 07:01:52 <Gh0stW0lf[m]> apk install people
2021-12-11 07:13:48 <ikke> dhruvin: we prefer not, in case of security vulnerabilities in of the libraries.
2021-12-11 07:14:40 <dhruvin> okay
2021-12-11 09:01:52 <dhruvin> newbie question: why is it recommended to use "$pkgdir"/some/dir over "$pkgdir/some/dir" in wiki?
2021-12-11 09:02:15 <psykose> just code style
2021-12-11 09:02:27 <psykose> there is an 80% chance you put the latter and nobody says a thing
2021-12-11 09:02:49 <dhruvin> functionally they're the same right?
2021-12-11 09:02:52 <psykose> yes
2021-12-11 09:02:56 <dhruvin> thanks :)
2021-12-11 09:04:59 <dhruvin> i just packaged hledger, web, ui, their docs, and their bash-completions. upstream updated it from 1.24 to 1.24.99 (pre-release of 1.24.1). i'll wait for them to update to 1.24.1.
2021-12-11 10:46:29 <ktprograms> zcrayfish: Did you look at my edits on the Gitea wiki page?
2021-12-11 11:01:07 <lopid> `apk -U version` doesn't actually seem to do an update for me
2021-12-11 11:28:57 <lamineer> For the conference, I only see two 5-6 hour recordings on https://bbb.dereferenced.org/b/adm-ec4-bx7-ypm is that right?
2021-12-11 11:29:02 <lamineer> Is there any way to download them?
2021-12-11 12:42:36 <zcrayfish> ktprograms: Not yet; been away :x
2021-12-11 13:10:55 <markpash> anyone here use alpine for a router with PPPoE? How do you handle drop-outs? I find I need to re-apply nft rules and restart my dhcpv6 client when that happens, what do you do to handle it?
2021-12-11 13:15:29 <kitzman> hmm, what could be the reason for an usb storage device (ssd adapter) not being detected on boot? maybe i have to add some udev rules?
2021-12-11 13:16:55 <nero> check dmesg
2021-12-11 13:21:24 <kitzman> the device does not appear there. first i get the hubs going online, and an usb device getting recognized (it's probably the sd card), but not the external ssd
2021-12-11 13:22:34 <nero> if plugging and unplugging the ssd does not result in dmesg output, its some electrical problem
2021-12-11 13:22:54 <nero> it might be that the hub does not give enough current for the ssd
2021-12-11 13:23:03 <kitzman> that's the thing : on boot it does not get recognized, but after boot, if i replug, there's no issue
2021-12-11 13:23:30 <kitzman> i don't know who sends first the message: the adapter or the sbc. but my guess at this point is that the external ssd simply refuses to answer/announce itself after the first time it does that
2021-12-11 13:26:18 <ktprograms> zcrayfish: Ok, np
2021-12-11 13:28:42 <kitzman> i am now looking if i can do the 'replug' using udev. but would it work if it doesn't even detect the device?
2021-12-11 14:01:24 <zcrayfish> ktprograms: I think your edits look awesome. Good job
2021-12-11 14:02:08 <ktprograms> zcrayfish: Thanks!
2021-12-11 14:02:27 <zcrayfish> np
2021-12-11 14:02:37 <ktprograms> zcrayfish: Should I remove {{Expand}} in the Gitea wiki page?
2021-12-11 14:03:54 <zcrayfish> I'm not sure if we have an expand template for a section-only; but I'd put it in the Post installation section
2021-12-11 14:05:17 <ktprograms> zcrayfish: Putting the expand template in the Post installation section works.
2021-12-11 14:05:53 <zcrayfish> awesome
2021-12-11 14:07:48 <Hello71> lamineer: afaik bbb doesn't provide built-in download, as it is "dynamically composited" in your browser. there are various softwares to download bbb recordings, some of which simply automate opening the browser and recording the screen
2021-12-11 15:03:27 <robertos> Hi I am trying to install Alpine on a virtual machine (QEMU/KVM with virt-manager) but I am getting the following error: "unsupported cpio format. You may need to fix the MBR to be able to boot"
2021-12-11 15:03:52 <ikke> What are you booting?
2021-12-11 15:03:57 <robertos> Couldn't find info on the manual
2021-12-11 15:04:52 <robertos> I have just partitioned the virtual disk and was running "setup-disk /mnt" to install the system
2021-12-11 15:06:37 <robertos> @ikke what do you mean?
2021-12-11 15:06:49 <ikke> How are you booting / running alpine?
2021-12-11 15:09:08 <robertos> I was folllowing the instructions for BIOS+MSDOS in the manual. Installation is not finished
2021-12-11 15:42:48 <lemonsh> Hi, I've got a question about OpenRC. The service script guide in the openrc repo (https://github.com/OpenRC/openrc/blob/master/service-script-guide.md) has a section about the depend() function, which mentions services 'net' and 'dns'. There are service scripts on my system (like sshd) that do use these services, but I can't find the services themselves. Running 'service net'
2021-12-11 15:42:50 <lemonsh> tells me '* service: service `net' does not exist', so I imagine these are some kind of 'pseudoservices' built into OpenRC. Are they documented somewhere? Or, at least, is there a list of them?
2021-12-11 15:45:40 <ikke> lemonsh: grep -r provide /etc/init.d
2021-12-11 15:46:06 <minimal> lemonsh: ikke just beat me to it, networking has "provide net"
2021-12-11 15:46:26 <minimal> as does dhcpcd (if installed)
2021-12-11 15:47:23 <ikke> staticroute as well
2021-12-11 15:49:11 <lemonsh> oh thanks! now it makes sense
2021-12-11 15:49:45 <lemonsh> that's much cleaner than what I had thought it to be
2021-12-11 15:50:03 <lemonsh> tysm
2021-12-11 15:52:59 <minimal> ikke: its somewhat confusing as boh dhcpcd and staticroute both have "use network" but nothing indicates that it provides that
2021-12-11 17:40:14 <goldeneye> Did usa train & suppl y isis with  weapon s  like  it did with  al qaeda   to   justify  creating  war s
2021-12-11 17:40:14 <goldeneye> about  9   million iraqi   lost  their  lives either becauz of direct  violence  of  usa and   becauz of  siege  usa  made against iraq  that  prevented iraqi  form essential living  materials  food, medicine and  even pencils 
2021-12-11 17:40:14 <goldeneye> did cia did 9 /11 or it  just let it happen. if alqaeda did it why to invade and kill 9 million iraqi
2021-12-11 17:40:14 <goldeneye> certainly ,the american administarin needed something like 9/11 to justify invading iraq which has been  already  planned as a part of creative chaos plan for middle east
2021-12-11 17:40:14 <goldeneye> isis  and alqaeda only appeared  in iraq after  usa invasion, Did  usa  pave the way for  isis to iraq ?
2021-12-11 17:40:16 <goldeneye>  saddam hussin who lost   most   of  his power  i n 1st , 2nd  gulf  wars  and   during 10 years   of  siege ,did   not  allow    isis    or  al-qaedea   to  enter    iraq  , even to help him against usa
2021-12-11 17:40:31 <goldeneye> Did usa train & suppl y isis with  weapon s  like  it did with  al qaeda   to   justify  creating  war s
2021-12-11 17:40:31 <goldeneye> certainly ,the american administarin needed something like 9/11 to justify invading iraq which has been  already  planned as a part of creative chaos plan for middle east
2021-12-11 17:40:31 <goldeneye> did cia did 9 /11 or it  just let it happen. if alqaeda did it why to invade and kill 9 million iraqi
2021-12-11 17:40:31 <goldeneye> about  9   million iraqi   lost  their  lives either becauz of direct  violence  of  usa and   becauz of  siege  usa  made against iraq  that  prevented iraqi  form essential living  materials  food, medicine and  even pencils 
2021-12-11 17:40:31 <goldeneye> isis  and alqaeda only appeared  in iraq after  usa invasion, Did  usa  pave the way for  isis to iraq ?
2021-12-11 17:40:32 <goldeneye>  saddam hussin who lost   most   of  his power  i n 1st , 2nd  gulf  wars  and   during 10 years   of  siege ,did   not  allow    isis    or  al-qaedea   to  enter    iraq  , even to help him against usa
2021-12-11 17:42:35 <goldeneye>                                       dnc jvb asdj wl;akf awfelk                                                                   
2021-12-11 17:42:41 <goldeneye> Did usa train & suppl y isis with  weapon s  like  it did with  al qaeda   to   justify  creating  war s
2021-12-11 17:42:41 <goldeneye> isis  and alqaeda only appeared  in iraq after  usa invasion, Did  usa  pave the way for  isis to iraq ?
2021-12-11 17:42:41 <goldeneye> about  9   million iraqi   lost  their  lives either becauz of direct  violence  of  usa and   becauz of  siege  usa  made against iraq  that  prevented iraqi  form essential living  materials  food, medicine and  even pencils 
2021-12-11 17:42:41 <goldeneye> certainly ,the american administarin needed something like 9/11 to justify invading iraq which has been  already  planned as a part of creative chaos plan for middle east
2021-12-11 17:42:41 <goldeneye> did cia did 9 /11 or it  just let it happen. if alqaeda did it why to invade and kill 9 million iraqi
2021-12-11 17:42:42 <goldeneye>  
2021-12-11 17:42:42 <goldeneye>  saddam hussin who lost   most   of  his power  i n 1st , 2nd  gulf  wars  and   during 10 years   of  siege ,did   not  allow    isis    or  al-qaedea   to  enter    iraq  , even to help him against usa
2021-12-11 17:42:54 <lopid> ikke?
2021-12-11 17:44:33 <ikke> oof
2021-12-11 17:44:52 <zcrayfish> thanks
2021-12-11 20:18:28 <anjan> hi all, I made this change; https://wiki.alpinelinux.org/w/index.php?title=Include%3ASetup_your_system_and_account_for_building_packages&type=revision&diff=20767&oldid=20384
2021-12-11 20:18:38 <anjan> but it seems it's not rendering as I thought it would
2021-12-11 20:18:40 <anjan> any help?
2021-12-11 20:18:48 <anjan> How is the command {{{1}}}?
2021-12-11 20:21:53 <ikke> It's apparently triggered by the SUDO=doas part
2021-12-11 20:22:22 <anjan> ikke: I tried doing export SUDO=doas and the keygen as a seperate line
2021-12-11 20:22:24 <anjan> no luck...
2021-12-11 20:23:01 <anjan> well, im going for a run. Feel free to make the change. I would really appreciate it. That has been tripping me up for so long lol
2021-12-11 20:26:56 <Hello71> put 1=SUDO=doas ...
2021-12-11 20:27:18 <Hello71> but probably abuild-keygen should automatically detect it
2021-12-11 20:29:05 <ikke> Hello71: ah, thanks
2021-12-11 20:30:37 <anjan> Hello71: there is a merge request for it  https://gitlab.alpinelinux.org/alpine/abuild/-/merge_requests/121
2021-12-11 20:30:41 <anjan> hasnt been merged
2021-12-11 20:30:41 <Hello71> or SUDO{{=}}doas ..., or SUDO&#61;doas
2021-12-11 20:30:48 <anjan> but the code looks good to me
2021-12-11 20:31:19 <ikke> Hello71: I tried &equals;, but that didn't work
2021-12-11 20:32:12 <Hello71> ah, yes, it doesn't work in this context
2021-12-11 20:32:43 <Hello71> so right, you need either 1= or {{=}} in this case
2021-12-11 20:32:53 <ikke> ok, thanks for the info
2021-12-11 20:34:05 <Hello71> anjan: doesn't it have the issue that ncopa said
2021-12-11 20:34:15 <Misthios> yeah, never fixed it 
2021-12-11 20:34:46 <ikke> ncopa did make some changes regarding this
2021-12-11 20:35:20 <anjan> Hello71: ugh, I guess it does break the test suite
2021-12-11 20:35:31 <anjan> Im not familiar with it but I would love to fix
2021-12-11 20:35:58 <anjan> ok, let me know if someone else is working on a fix
2021-12-11 20:36:21 <anjan> so I can build off it
2021-12-11 20:36:56 <ikke> and I think the SUDO env variable should remain supported for backwards compattibility
2021-12-11 20:40:23 <ikke> If you have questions about the test-suite, feel free to ask
2021-12-11 20:43:27 <Misthios> anjan i kinda forgat about that mr so if u want to can fix the issue
2021-12-11 20:44:05 <ikke> And we've added CI for the test suite after the MR
2021-12-11 20:44:32 <Misthios> but should the fix just be renaming the auth to sudo?
2021-12-11 20:44:37 <Misthios> or am i missing something now
2021-12-11 20:44:46 <ikke> I think that would fix it, yes
2021-12-11 20:45:16 <Misthios> hmm i might just get my laptop and try it
2021-12-11 20:45:16 <Hello71> it also needs to check if it is already specified
2021-12-11 20:45:24 <ikke> right
2021-12-11 20:59:10 <Misthios> anjan the failing tests with that mr are 15 and 16 (both related to the key install with permission denied). if u want to fix the mr please tell me. if not i or someone else can fix it
2021-12-11 22:17:12 <anjan> Misthios: please fix it
2021-12-11 22:17:17 <anjan> sorry I went for a run
2021-12-11 23:18:26 <markpash> hello, is there a way to see debug output from the boot of the alpine installer?
2021-12-11 23:19:01 <markpash> I've tried to boot up the installer on this old laptop and it gets stuck on loading hardware drivers
2021-12-11 23:21:40 <Hello71> what version
2021-12-11 23:21:45 <Hello71> and what installer
2021-12-11 23:21:56 <markpash> standard, downloaded it today
2021-12-11 23:22:04 <markpash> so I think 3.15
2021-12-11 23:22:59 <Hello71> what happens if you add nomodeset to kernel parameters
2021-12-11 23:25:27 <markpash> worked
2021-12-11 23:26:58 <markpash> thanks, is there anything I can do to help make this not happen in the future?
2021-12-11 23:27:05 <markpash> like a log output or something?
2021-12-11 23:27:55 <markpash> Hello71: Also, what's happening? Why does that kernel param make it work?
2021-12-11 23:28:11 <Hello71> probably related to simpledrm
2021-12-11 23:28:28 <Hello71> but i thought it was reverted for 3.15, hm
2021-12-11 23:28:44 <Hello71> what graphics card
2021-12-11 23:29:05 <markpash> intel graphics, 7th gen laptop CPU
2021-12-11 23:29:32 <Hello71> i think the current theory is that there is some problem passing off simpledrm to i915
2021-12-11 23:30:00 <markpash> I see, so this disables modesetting?
2021-12-11 23:30:23 <markpash> And, if this is known, I guess I don't need to report this anywhere :)
2021-12-11 23:30:35 <Hello71> well the precise cause is not known
2021-12-11 23:30:57 <Hello71> can you try modprobe.blacklist=simpledrm
2021-12-11 23:31:07 <markpash> sure
2021-12-11 23:31:16 <Hello71> (without nomodeset)
2021-12-11 23:31:42 <Hello71> also if you have time, modprobe.blacklist=i915 (without nomodeset or modprobe.blacklist=simpledrm)
2021-12-11 23:32:42 <markpash> I have time :) Trying the first thing now
2021-12-11 23:33:17 <Hello71> thanks
2021-12-11 23:33:50 <Hello71> some people reported this with 3.15 rc, but it was supposed to be fixed. possibly it is still broken on some hardware
2021-12-11 23:37:18 <markpash> I've tried both blacklisting, neither worked
2021-12-11 23:47:25 <Hello71> hm.
2021-12-11 23:49:44 <markpash> yeah if you have other suggestions, I have a few hours I can help with this :P
2021-12-11 23:50:19 <Hello71> if you install with nomodeset then you can't use x or wayland
2021-12-11 23:50:38 <markpash> I don't need those, so it's fine
2021-12-11 23:50:41 <atka> hey alpine, possible to run dockerd rootless on alpine?
2021-12-11 23:50:41 <Hello71> also the resolution will be stuck to vga but if you don't want x or wayland then you probably don't care about that either
2021-12-11 23:50:52 <Hello71> atka: possibly but have you considered podman
2021-12-11 23:51:40 <atka> Hello71: I have not
2021-12-11 23:51:47 <tobtobxx> atka: https://wiki.alpinelinux.org/wiki/Podman for rootless podman setup
2021-12-11 23:52:22 <Hello71> podman is still bloat but at least it's less bloat than docker
2021-12-11 23:52:38 <atka> ok, so does it replace docker altogether?
2021-12-11 23:52:44 <tobtobxx> atka: you can also install podman-docker to make the ui almost like docker.
2021-12-11 23:52:55 <tobtobxx> Yeah, it's an alternative.
2021-12-11 23:53:31 <tobtobxx> The only difference you have to know, is that you have to explicitely prefix docker.io/postgres IIRC.
2021-12-11 23:53:52 <atka> ok, thanks I'll look into getting that set up asap
2021-12-11 23:55:02 <Hello71> tobtobxx: you can configure unqualified-search-registries in /etc/containers/registries.conf or somesuch
2021-12-11 23:55:37 <tobtobxx> ah, good to know. I prefer fully qualified names anyways though.
2021-12-11 23:56:00 <markpash> Hello71: I'll go ahead and install, but I'll be here for a little bit, and I can boot up the iso again to test anything you like
2021-12-11 23:56:16 <tobtobxx> Can someone who understands a bit more take a quick glance at the luarocks package and tell mi if I'm wrong or if this is really just a dummy package: https://git.alpinelinux.org/aports/tree/community/luarocks/APKBUILD
2021-12-11 23:56:39 <Hello71> markpash: i don't really have any better ideas at the moment
2021-12-11 23:56:54 <tobtobxx> The package() is empty and the subpackage would do something, but there are none specified in subpackages="".
2021-12-11 23:56:55 <markpash> Ok, no problem. Thanks for the help
2021-12-11 23:57:25 <Hello71> tobtobxx: but subpackages is modified at line 22
2021-12-11 23:57:44 <Hello71> https://pkgs.alpinelinux.org/contents?file=&path=&name=luarocks*&branch=edge
2021-12-11 23:58:07 <tobtobxx> Ah, didn't catch that. So I shouldn't even bother installing `luarocks`, but install `luarocks5.1`?
2021-12-11 23:58:37 <tobtobxx> Or does the `luarocks` package serve purpose?
2021-12-12 00:02:18 <Hello71> it does appear that luarocks doesn't do anything by itself
2021-12-12 00:03:01 <Hello71> this is somewhat inconsistent between alpine packages. sometimes installing the unqualified package will install the latest version, sometimes it will install all of them, and sometimes it will install none
2021-12-12 00:04:03 <psykose> for me it installs 5.1 and 5.3 but not 5.2
2021-12-12 00:04:20 <psykose> as i have 5.1 and 5.3 on system i guess
2021-12-12 00:04:21 <Hello71> ah, right, it uses install_if
2021-12-12 00:05:25 <tobtobxx> Ah, so I can install `luarocks5.3` first and then abuild `luarocks` so that it has `luarocks5.3` as a dependency?
2021-12-12 00:05:38 <tobtobxx> *could...
2021-12-12 00:05:53 <psykose> it has nothing to do with abuild
2021-12-12 00:06:04 <psykose> if you install `luarocks` then install lua5.2 it will pull in luarocks5.2 as well
2021-12-12 00:06:58 <Hello71> markpash: it works fine on my intel... kaby lake? laptop, so probably it is related to the exact hardware (and maybe firmware)
2021-12-12 00:07:39 <Hello71> oh, another downside of nomodeset is that it will possibly slightly increase the power consumption
2021-12-12 00:08:46 <Hello71> do other linux distros work?
2021-12-12 00:09:44 <markpash> this laptop previously had ubuntu, so I guess
2021-12-12 00:10:32 <markpash> I'm gonna run this as a router, so I'm good with it being a little more power hungry, as long as it's not running full tilt :P
2021-12-12 00:12:09 <Hello71> at a guess, it will use between 0 and 6 watts more
2021-12-12 00:12:36 <markpash> oof, 6 is significant
2021-12-12 00:12:42 <Hello71> if you could try booting arch linux, that would help
2021-12-12 00:13:09 <markpash> I have a fresh iso of that on my flash drive too
2021-12-12 00:13:11 <markpash> give me a min
2021-12-12 00:13:26 <Hello71> iso? you're not using standard dd?
2021-12-12 00:13:49 <markpash> ah, if it makes a difference, I'm using something called ventoy
2021-12-12 00:14:06 <Hello71> ventoy screws with the system significantly
2021-12-12 00:14:15 <Hello71> it probably doesn't cause this particular issue though
2021-12-12 00:14:45 <markpash> I can test it with a different flash drive... but I'll boot arch first
2021-12-12 00:17:40 <markpash> huh, the arch iso didn't boot properly, dropped me into some limited shell
2021-12-12 00:18:08 <markpash> ubuntu iso on the other hand, boots to gui fine
2021-12-12 00:19:56 <Hello71> the initramfs shell is almost certainly ventoy's fault
2021-12-12 00:20:20 <markpash> ok, I have plenty of flash drives, gonna dd some isos
2021-12-12 00:24:28 <Hello71> considering how much it tries to do, ventoy doesn't do a terrible job, but it's fundamentally a brittle solution. because most linux distros aren't designed to run this way, it goes and rummages around and pokes at stuff
2021-12-12 00:30:44 <markpash> yeah I don't really have any opinion on it, usually if it works it works, but I guess it could be problematic
2021-12-12 00:32:02 <markpash> I took a different drive and dd'd alpine, booted it but it didn't change anything
2021-12-12 00:32:08 <markpash> gonna try arch next
2021-12-12 00:37:51 <markpash> ok, arch is up
2021-12-12 00:38:41 <markpash> booted straight to a shell no problem
2021-12-12 00:40:44 <markpash> Hello71: happy to run anything you like in the archiso
2021-12-12 00:40:59 <Hello71> hm.
2021-12-12 00:41:52 <Hello71> maybe lspci?
2021-12-12 00:42:00 <markpash> interesting, while sitting in the shell, I have an error from drm
2021-12-12 00:42:39 <markpash> from i915 : *ERROR* CPU pipe A FIFO underrun
2021-12-12 00:45:25 <markpash> do you know of a paste site which can work with a basic curl? not gonna trouble you with a picture from my phone
2021-12-12 00:46:14 <Hello71> ix.io, paste.rs, termbin.com
2021-12-12 00:46:45 <Hello71> lspci | curl -F 'f:1=<-' ix.io
2021-12-12 00:46:54 <markpash> weird, I see weird graphical glitches as it sits in the shell
2021-12-12 00:47:05 <Hello71> could be fbc
2021-12-12 00:47:05 <markpash> like every few mins
2021-12-12 00:47:12 <Hello71> or psr
2021-12-12 00:47:14 <markpash> thanks for the snippet
2021-12-12 00:48:06 <Hello71> hm, seems like archiso bundles i915/amdgpu/nouveau in the initramfs
2021-12-12 00:48:33 <markpash> http://ix.io/3HLI
2021-12-12 00:49:00 <Hello71> nothing particularly weird there
2021-12-12 00:49:06 <markpash> dmesg http://ix.io/3HLJ
2021-12-12 00:49:23 <Hello71> [    5.022956] fb0: switching to i915 from EFI VGA
2021-12-12 00:49:47 <Hello71> so it's switching directly from efifb to i915, skipping simpledrm
2021-12-12 00:50:03 <Hello71> hm... can you try booting arch with modprobe.blacklist=i915
2021-12-12 00:50:34 <markpash> sure, let's hope the thing has a way to change the kernel boot params
2021-12-12 00:51:21 <Hello71> yes
2021-12-12 00:53:41 <markpash> http://ix.io/3HLK
2021-12-12 00:53:57 <markpash> I hope I did the boot param correctly
2021-12-12 00:55:00 <markpash> I don't think I did
2021-12-12 00:55:06 <markpash> but I don't know what I should do
2021-12-12 00:55:42 <psykose> it's on the cmdline at the start correctly
2021-12-12 00:55:59 <Hello71> i think systemd might ignore it
2021-12-12 00:56:08 <markpash> sure, but a quick CTRL+F shows lines from i915
2021-12-12 00:56:11 <markpash> yeah
2021-12-12 00:56:12 <Hello71> eh, it probably won't be that helpful anyways
2021-12-12 00:56:23 <Hello71> i'm still mostly lost. thanks for the help anyways
2021-12-12 00:57:19 <markpash> hey no problem at all, at least we tried. Besides my alpine install worked with nomodeset, and that's enough
2021-12-12 01:08:18 <Hello71> you can probably fix it by using early kms (loading i915 in the initramfs)
2021-12-12 01:09:04 <markpash> you mean, once I've installed the system, I go and tweak the initramfs generation?
2021-12-12 01:09:08 <Hello71> mmhmm
2021-12-12 01:09:16 <Hello71> assuming you're using sys install anyways
2021-12-12 01:09:21 <markpash> I see, I can try that
2021-12-12 01:10:06 <Hello71> the procedure is roughly to put kernel/drivers/gpu/drm/i915 in /etc/mkinitfs/features.d/i915.modules, then add i915 to features in /etc/mkinitfs.conf or something like that
2021-12-12 01:11:56 <markpash> would doing this help with solving the problem upstream? or just help loading i915 in my specific case?
2021-12-12 01:12:48 <Hello71> i guess it would help debug the issue. if it works then the issue is almost certainly simpledrm-related
2021-12-12 01:13:05 <Hello71> if it doesn't work then the issue is almost certainly not simpledrm-related
2021-12-12 01:13:12 <Hello71> if you don't test it then it's probably simpledrm-related
2021-12-12 01:13:13 <psykose> isn't kernel/drivers/gpu in it's entirety part of the kms feature
2021-12-12 01:13:38 <markpash> ok cool, I'm down to do help debug this
2021-12-12 01:14:00 <Hello71> hm, good point
2021-12-12 01:14:01 <markpash> but I'm gonna need a little more guidance
2021-12-12 06:40:35 <atka> ok, I have rootless podman and some containers running. anyone know how to get containers to autorestart after reboot with open-rc? lots of guides for using systemd for this, anyone using alpine for this?
2021-12-12 07:20:48 <Asmadeus> atka: I've just slapped the podman run commands in my own service
2021-12-12 07:21:48 <Asmadeus> (this is pretty much what the systemd variant does, just that there are helpers to generate service files..)
2021-12-12 07:52:45 <atka> Asmadeus: thanks, I'll do that
2021-12-12 09:21:50 <frag> do you have alsa service running? what does that actually do?
2021-12-12 09:30:06 <lopid> no. provides sound
2021-12-12 09:46:21 <frag> sound seems to work ok without it tho
2021-12-12 09:46:28 <frag> without it running as a service
2021-12-12 10:37:07 <lopid> the service probably just sets some things, doesn't need a daemon
2021-12-12 10:38:42 <lopid> https://git.alpinelinux.org/aports/tree/main/alsa-utils/alsa.initd?h=3.15-stable#n6
2021-12-12 10:41:11 <lopid> ie. rc-service alsa describe
2021-12-12 10:43:47 <frag> oh didnt know about that option :]
2021-12-12 10:45:34 <frag> trying to remove alsa-utils-openrc fails with error: World updated, but the following packages are not removed due to: alsa-utils-openrc: alpine-base alsa-utils ...
2021-12-12 10:45:43 <frag> i cant uninstall because of the sub-packages? https://pkgs.alpinelinux.org/package/v3.15/main/x86_64/alsa-utils-openrc
2021-12-12 10:46:33 <lopid> hrm. i have alpine-base with no alsa packages
2021-12-12 10:47:16 <frag> yeah, doesnt seem to make any sense
2021-12-12 10:48:09 <lopid> maybe it means you have to uninstall a sub-package in order to remove this package
2021-12-12 10:48:13 <frag> it doesnt even depend on alsa-utils afaict
2021-12-12 10:49:40 <frag> i think i wont try to uninstall alpine-base :p
2021-12-12 10:49:52 <lopid> ;)
2021-12-12 10:50:59 <frag> hmm actually it doesnt look so dangerous, buuut ...
2021-12-12 10:51:00 <frag> jaja
2021-12-12 10:51:36 <lopid> i don't see anything about sub-packages on the wiki
2021-12-12 10:52:02 <frag> often sub-packages is just the -doc package i think
2021-12-12 10:52:19 <frag> but in this case its harder to understand
2021-12-12 10:52:31 <lopid> but how do they relate? why does it stop the uninstallation?
2021-12-12 10:52:46 <frag> i dunno if its a bug or what
2021-12-12 10:53:49 <lopid> does it look like it would allow the removal if you tried to remove all the sub-packages at the same time? use -i so it prompts
2021-12-12 10:54:03 <frag> if anything i would think alsa-utils-openrc would be a sub-package to alsa-utils...
2021-12-12 10:54:13 <lopid> yeah
2021-12-12 10:56:30 <frag> yes, it seems so
2021-12-12 10:57:01 <lopid> a bit laborious
2021-12-12 11:01:34 <frag> ok, did it and reinstalled alsa-utils, and alsa-utils-openrc also was intalled :p
2021-12-12 11:01:39 <frag> jaja...
2021-12-12 11:02:00 <frag> wonder if alsa is easy to replace with tinyalsa
2021-12-12 12:30:22 <frag> do one usually need xf86-input-libinput and xf86-video-fbdev (i use alpine as desktop and have xf86-video-intel installed)?
2021-12-12 12:31:29 <lopid> wouldn't surprise me
2021-12-12 12:32:01 <lopid> are the deps of something?
2021-12-12 12:33:13 <frag> nope :]
2021-12-12 12:33:20 <frag> but i have them in my world file
2021-12-12 12:40:28 <lopid> on another distro, they're deps of xorg-input-drivers and xorg-video-drivers, respectively
2021-12-12 12:50:06 <PureTryOut> you almost definitely need xf86-input-libinput, but I believe that depends on the DE you're using
2021-12-12 12:50:07 <frag> oh, good idea :]
2021-12-12 12:50:29 <frag> i dont use DE, i have dwm
2021-12-12 12:50:36 <frag> so tiny :]
2021-12-12 13:29:15 <sliwkr> Hey, attempting to install the vm-optimized Alpine in qemu. I'm partitioning the disk space to have a single encrypted sys partition. After rebooting I'm getting a blank screen in qemu's -nographic mode. When I boot a with graphics interface, it seems to work (i'm able to enter the encryption password and log in), but there's another issue with the screen not scrolling down (apparently that happens in qemu, haven't found a solution yet). Th
2021-12-12 17:09:46 <graywolf> Hi, quick question: Is there a quick way to get what packages are dependencies of another package? apk info -R seems to give list of .so, not packages. What would be a simplest way to get list of actual packages?
2021-12-12 17:10:51 <ikke> apk add lua-aports; cd aports/<repo>; ap revdep <pkg>
2021-12-12 17:15:35 <graywolf> That requires cloning the aports etc, isn't there way using just apk? Best options I found would be apk dot firefox | awk SOME_MAGIC or apk add --simulate firefox | awk SOME_MAGIC; ; Thanks for the lua-aports, that is probably "correct" way, so I will use it but was hoping for something slightly more lightweight
2021-12-12 17:18:41 <ikke> apk has as far as I know no direct way to list them
2021-12-12 17:45:21 <minimal> graywolf: not cli-based but if you look at the package's details on pkgs.alpinelinux.org the "Required by" dropdown will show this
2021-12-12 20:09:36 <tobtobxx> Is there a way to tell abuild to skip the checksums? Like the SKIP on AUR?
2021-12-12 20:09:45 <tobtobxx> (it's for a private nightly package)
2021-12-12 20:12:19 <ikke> Only way would be to override the verify() function
2021-12-12 20:12:31 <ikke> but that would skip it completetly
2021-12-12 20:22:32 <Hello71> better use a custom unpack
2021-12-12 20:29:43 <tobtobxx> Works perfectly, thanks!
2021-12-12 21:11:11 <Srket> I have a small laptop i had Alpine installed, and i was running some Docker Containers and file transfer using SFTP. It was running for about 30 days until today, my SFTP transfers decided to not work, so i SSH'd into the machine and rebooted it. Upon booting, It gave an error : "mount: mounting /dev/sda3 on /sysroot failed: No error information"
2021-12-12 21:11:11 <Srket> and it went into recovery shell. I have no idea what to do, and the machine didn't have anything changed, no updates or anything. How can i fix it, or atleast recover my data?
2021-12-12 21:11:47 <Hello71> dmesg
2021-12-12 21:12:27 <MaximKarasev[m]> Maybe hard drive failure?
2021-12-12 21:12:41 <MaximKarasev[m]> Can you mount sda3 by hand from recovery shell?
2021-12-12 21:12:48 <MaximKarasev[m]> fsck?
2021-12-12 21:13:15 <Srket> I see some checksum errors, and "failed to initialize system zone", drive is near brand-new so should be no problem on that
2021-12-12 21:13:28 <Srket> can upload pictures of the outputs
2021-12-12 21:14:28 <Hello71> sure, send a picture
2021-12-12 21:14:35 <Hello71> also i guess try mount /dev/sda3 /sysroot manually
2021-12-12 21:14:49 <Hello71> "No error information" is a bit odd, but maybe a busybox bug
2021-12-12 21:15:07 <Srket> keyboard layout is defaulted, hard to execute commands, uploading the pictures
2021-12-12 21:15:45 <Hello71> print out a qwerty and then tape it to your keyboard :p
2021-12-12 21:15:55 <Srket> mount gives error : "Can't find /dev/sda in /etc/fstab"
2021-12-12 21:16:06 <Hello71> did you try putting the 3
2021-12-12 21:16:09 <Hello71> or the /sysroot
2021-12-12 21:16:47 <Srket> i didn't put the /sysroot, when i did, it says the same error message as it first did
2021-12-12 21:19:00 <Srket> running "fsck /dev/sda3" gives "fsck.auto no such file or directory"
2021-12-12 21:20:37 <Srket> https://transfer.sh/ncm1YY/20211213_001736.jpg
2021-12-12 21:20:47 <Srket> sorry for crappy picture, but here it is
2021-12-12 21:21:27 <Srket> when i exit, it just kernel panic and don't run
2021-12-12 21:21:51 <Hello71> fsck.ext4 /dev/sda3
2021-12-12 21:22:20 <Srket> fsck.ext4 not found
2021-12-12 21:22:29 <Hello71> uh... e2fsck /dev/sda3
2021-12-12 21:22:44 <Srket> e2fsck not found
2021-12-12 21:22:53 <Hello71> find a usb drive
2021-12-12 21:22:57 <Hello71> :p
2021-12-12 21:23:02 <Srket> write alpine in it?
2021-12-12 21:23:07 <Hello71> mmhmm.
2021-12-12 21:23:20 <Hello71> (i mean it doesn't technically have to be new, or a usb, or alpine)
2021-12-12 21:24:51 <Srket> downloading rn
2021-12-12 21:26:07 <Hello71> probably your drive is failing (bathtub curve), but possibly something else is wrong
2021-12-12 21:26:56 <Srket> holds all my backups, would be horrible if it failed
2021-12-12 21:27:31 <Srket> if i can access the data, i have spare space to hold it tho
2021-12-12 21:27:43 <lopid> you also have the masters
2021-12-12 21:27:58 <Hello71> one is none, two is one
2021-12-12 21:29:07 <Srket> thanks, i don't have a lot alpine experience
2021-12-12 21:31:55 <Srket> USB is done, booting it up now
2021-12-12 21:34:13 <Srket> on the system, logged in as root,
2021-12-12 21:39:52 <Srket> i try doing "fsck /dev/sda3" and "fsck.ext4 /dev/sda3" but they seem to not work, giving "fsck.auto no suck file or directory" or "command not found"
2021-12-12 21:42:33 <Hello71> apk add e2fsprogs
2021-12-12 21:43:01 <Srket> says OK: 12MiB
2021-12-12 21:43:32 <Srket> ok, command worked now
2021-12-12 21:44:26 <Srket> by the output of it, it looks bad. It asks me if i wanna clean inode garbage
2021-12-12 21:45:27 <Srket> is there a way that i can check SMART data of the disk in the live system?
2021-12-12 21:47:24 <Hello71> yes, use google
2021-12-12 21:48:19 <Srket> is it normal the command spamming stuff on the screen?
2021-12-12 21:49:00 <Srket> its bunch of numbers and im concerned about what caused this to happen in the first place
2021-12-12 21:49:02 <lopid> smartctl -H /dev/sda3, but you might want to run a self test
2021-12-12 21:49:13 <lopid> that output is expected
2021-12-12 21:59:05 <rdavidson> i'm trying to install Alpine on a Raspberry Pi Zero 2 W. The config.txt in the released rpi tarballs does not include a setting for the Zero 2 W.
2021-12-12 21:59:16 <rdavidson> Has anyone succeeded in installing Alpine on this device?
2021-12-12 22:08:30 <Srket> finally, the fsck finished, gonna run the self tests now
2021-12-12 22:14:02 <Srket> SMART info shows, PASSED, gonna run selfcheck just in case, thanks for the support and good night
2021-12-12 23:11:08 <rdavidson> ended up opening an issue for it: https://gitlab.alpinelinux.org/alpine/aports/-/issues/13302
2021-12-13 06:24:08 <jhjacobs81> hmz
2021-12-13 06:46:36 <dhruvin> hello, CODINGSTYLE.md allows the use of ${var:offset:length}, whereas shellcheck running while lint step in gitlab pipeline doesn't.
2021-12-13 06:47:36 <dhruvin> https://gitlab.alpinelinux.org/alpine/aports/-/blob/master/CODINGSTYLE.md, https://gitlab.alpinelinux.org/alpine/aports/-/jobs/561753#L41
2021-12-13 06:48:16 <dhruvin> Should I ignore the shellcheck warning?
2021-12-13 08:11:41 <ikke> dhruvin: Yes, I still need to update it to ignore the dash warnings
2021-12-13 09:49:27 <xsteadfastx> i set up a md raid with the alpine installer. its a raid 5. now if i reboot i get a blinking cursor on black screen. i have to turn it off completly and start to get alpine to boot. on every reboot i get this blinking cursor
2021-12-13 09:49:51 <xsteadfastx> i checked if the bootloader is installed on every device... and i think it is
2021-12-13 09:51:35 <xsteadfastx> here some more infos https://gist.github.com/xsteadfastx/1abb299dd2e622f72c7a58ccb3a5b7f5
2021-12-13 09:52:20 <xsteadfastx> the only think that looks strange to me is the same label-id on /dev/sda and /dev/sdb
2021-12-13 12:24:55 <steinbach> good day =)
2021-12-13 12:26:03 <steinbach> I am using Alpine Linux as KVM Host. I noticed resource graphs not being displayed when connecting to libvirt running on the alpine instance.
2021-12-13 12:26:13 <steinbach> And it seems like an issue with cgroups.
2021-12-13 12:26:54 <steinbach> So I created the dir /dev/cgroup and mounted it with "mount -t cgroup -o devices,cpu,memory,blkio,cpuset none /dev/cgroup"
2021-12-13 12:27:21 <steinbach> but then, when running a VM I get:     raise libvirtError('virDomainCreate() failed')
2021-12-13 12:27:21 <steinbach> libvirt.libvirtError: Unable to write to '/dev/cgroup/machine/qemu-13-ubuntu20web1.libvirt-qemu/devices.deny': No such file or directory
2021-12-13 12:27:47 <steinbach> when i restart, it works again, just without the resource meters
2021-12-13 12:28:20 <steinbach> after creating the cgroup dir in dev, it starts with this behaveiour requiring that file
2021-12-13 12:32:43 <psykose> cgroups are normally mounted on /sys/fs/cgroup by the cgroup service and set to unified/hybrid/legacy per /etc/rc.conf
2021-12-13 12:33:48 <psykose> if the default hybrid doesn't work you can try change it to unified and see if that works
2021-12-13 12:50:32 <steinbach> ah! thanks for the hint
2021-12-13 12:50:50 <steinbach> I'll update the dir in the qemu config to point to that directoray
2021-12-13 12:53:10 <steinbach> I reset everything to default settings and set cgroups to legacy, ill reboot the machine now and if not ill check the other variables
2021-12-13 13:03:08 <steinbach> mm
2021-12-13 13:03:11 <steinbach> virsh # cpu-stats 1
2021-12-13 13:03:11 <steinbach> error: Failed to retrieve CPU statistics for domain 'ubuntu20_web1'
2021-12-13 13:03:11 <steinbach> error: Requested operation is not valid: cgroup CPUACCT controller is not mounted
2021-12-13 13:06:07 <steinbach> In the qemu.conf file it say "does not matter where the controllers are mounted, qemu will locate the mount path automatically"
2021-12-13 13:08:24 <steinbach> damn.... i cant believe it.......
2021-12-13 13:08:25 <psykose> you could also try unified 
2021-12-13 13:08:31 <steinbach> please dont kill me...
2021-12-13 13:08:54 <steinbach> 
ACTION din't had the cgroup service running

2021-12-13 13:09:06 <steinbach> 
ACTION hides

2021-12-13 13:14:26 <steinbach> thx =)
2021-12-13 14:58:21 <earendel>     
2021-12-13 15:48:03 <earendel>   
2021-12-13 15:48:33 <earendel> pardon. screensaver 
2021-12-13 16:05:05 <nero> earendel: i have the feeling that you were involved with the fall of gondolin
2021-12-13 16:54:47 <earendel> people say it was you who put rome on fire - with gasoline
2021-12-13 17:02:41 <nero> heh
2021-12-13 17:20:45 <frag> is it less/unsecure to put internet straight into computer (not using router)?
2021-12-13 17:25:50 <nero> alpine base doesnt bind to many ports per default, so there isnt much an attacker could do
2021-12-13 17:26:19 <bdju> your modem might be a combo modem/router already
2021-12-13 17:26:35 <nero> its different for distros like ubuntu (think zeroconf) or windows (think upnp)
2021-12-13 17:28:11 <frag> nice :)
2021-12-13 17:31:19 <bovis> I'd like to upgrade my system to edge, but I get this error https://bpa.st/26BQ . I'm happy to delete streamlink, and looks like I can delete wine* for the time being to upgrade libldap. But when I try apk del, I get several errors https://bpa.st/VPEA . How do I get around this?
2021-12-13 17:32:51 <MY-R> bovis: use --available upgrade
2021-12-13 17:33:35 <MY-R> with --interactive to see what it will do
2021-12-13 17:33:49 <MY-R> so... apk -Uia upgrade
2021-12-13 17:35:50 <bovis> MY-R: I get the initial error https://bpa.st/26BQ but without streamlink listed, so it's just stuck on libldap
2021-12-13 17:37:47 <MY-R> bovis: what is "streamlink" package? it doesnt exist in Alpine
2021-12-13 17:38:12 <MY-R> ah it is in testing
2021-12-13 17:38:25 <bovis> https://pkgs.alpinelinux.org/package/edge/testing/x86_64/streamlink
2021-12-13 17:39:10 <MY-R> bovis: try remove streamlink, upgrade and install it again
2021-12-13 17:39:33 <psykose> you still didn't run upgrade -Uia
2021-12-13 17:40:45 <psykose> though wine-libs is gone pretty sure
2021-12-13 17:40:48 <bovis> MY-R: In trying to remove streamlink, I get all the errors I originally posted here https://bpa.st/VPEA
2021-12-13 17:41:15 <bovis> psykose: upgrade -Uia gives me the first set of errors, but only with libldap as the conflict.
2021-12-13 17:41:34 <bovis> I'm fine with deleting streamlink permanently. I just can't.
2021-12-13 17:41:39 <MY-R> bovis: when using edge then be sure you always run with -a
2021-12-13 17:41:52 <psykose> could you post your world file
2021-12-13 17:43:30 <bovis> psykose: https://bpa.st/LFHA
2021-12-13 17:43:54 <MY-R> ye post /etc/apk/word file, some app depends on old libldap version, so should find it and remove and after all can install again
2021-12-13 17:44:51 <bovis> MY-R: Looks like the only thing libldap 2.4 conflicts with is the 2.6 version
2021-12-13 17:44:51 <MY-R> psykose: wth are those hashes? Q12AZebngyujWxmPYhjlx2YibAKU0 etc? virtual?
2021-12-13 17:45:06 <psykose> i don't remember
2021-12-13 17:45:24 <psykose> just delete the hash off every line and i guess it will be fine
2021-12-13 17:45:25 <MY-R> bovis: ye conflict is obvious but something using old libldap and that is why apk cant remove it
2021-12-13 17:46:02 <bovis> The hashed ones all look like apks I've built/installed locally
2021-12-13 17:46:47 <psykose> the streamlink you built still depends on old ldap
2021-12-13 17:46:57 <MY-R> bovis: so dont expect that apk will upgrade whole system since older libs are used with those local apps
2021-12-13 17:49:13 <MY-R> bovis: remove those hashes from localy build apps in word file, let apk do -a upgrade and then rebuild them if you want again
2021-12-13 17:49:25 <bovis> MY-R psykose: Removing the hashes didn't change anything. Same error with libldap only
2021-12-13 17:50:23 <MY-R> bovis: because there is still some app which was probably build with old libldap, maybe some which isnt even in repo?
2021-12-13 17:50:40 <psykose> could you paste a -Ua
2021-12-13 17:51:14 <psykose> and /etc/apk/repositories
2021-12-13 17:52:10 <bovis> https://bpa.st/ZLUA
2021-12-13 17:53:12 <psykose> delete wine-libs from the world file
2021-12-13 17:53:20 <bovis> https://bpa.st/OG5A
2021-12-13 17:54:54 <bovis> psykose: Deleting wine-libs worked and allowed the upgrade to continue
2021-12-13 17:58:03 <MY-R> would be nice if apk could return more readable informations in such a situations, not to mention that dealing with removed packages/sub-packages/downgraded packages is still a nightmare
2021-12-13 17:59:24 <bovis> Is there a log for apk?
2021-12-13 18:10:42 <asdflkj> is there a way to search uninstalled packages by files included in them?
2021-12-13 18:11:13 <ikke> not with apk
2021-12-13 18:11:27 <ikke> there is apk-files which scrapes pkgs.alpinelinux.org
2021-12-13 18:13:09 <asdflkj> ikke: how do I use it?
2021-12-13 18:14:35 <lopid> https://pkgs.alpinelinux.org/contents
2021-12-13 18:15:52 <asdflkj> thanks!
2021-12-14 05:15:42 <ktprograms> Has anyone had 'Renderer process crashed (status 11)' when opening youtube.com with Qutebrowser?
2021-12-14 05:17:38 <zcrayfish> Can't say I have, but the first thing I would do (because it worked for all other browsers I tried on Alpine) is run as quickly as possible to the settings and disable all GPU acceleration options :x
2021-12-14 05:23:19 <ktprograms> zcrayfish: That was recommended to me by someone on #qutebrowser (on LiberaChat), but it didn't change anything.
2021-12-14 05:43:04 <dhruvin> !28411 (incorrect commit message) and !28381 (superseded by !28415) can be closed.
2021-12-14 05:44:50 <ikke> algitbot: !ping
2021-12-14 05:45:10 <ikke> w00t
2021-12-14 05:46:10 <ikke> w00t
2021-12-14 05:46:26 <ikke> !28411
2021-12-14 05:46:33 <ikke> !28415
2021-12-14 05:46:56 <ikke> !28318
2021-12-14 05:48:26 <dhruvin> ikke: that's !28381
2021-12-14 05:48:31 <ikke> yes, noticed it
2021-12-14 05:49:16 <dhruvin> I replied to the mailing list, but the messages weren't relayed to MR.
2021-12-14 05:50:32 <dhruvin> I'll create an account on GitLab instance and start contributing using that.
2021-12-14 05:54:45 <ikke> dhruvin: One thing I'm wondering, I don't see that hledger has any lockfiles for dependencies, is that correct?
2021-12-14 05:57:47 <dhruvin> yes, should I cabal freeze, and add that file to sources?
2021-12-14 05:57:54 <ikke> yes
2021-12-14 05:59:03 <dhruvin> is there a plan for having a prefix for haskell packages like we have for python, i.e. py3-?
2021-12-14 06:00:32 <dhruvin> maybe a canonical home for haskell libraries, /var/lib/ghc-$ghcver/$pkgname/? I'm not sure about the path, but something like it/
2021-12-14 06:00:50 <dhruvin> s|/|.|
2021-12-14 06:01:10 <ikke> not at the moment, but that's because so far we choose not to ship separate haskell libraries
2021-12-14 06:02:23 <dhruvin> so for future haskell packages, one needs to cabal freeze the dependencies, add them to sources, and fetch and build the dependencies from hackage. right?
2021-12-14 06:03:14 <ikke> That's what we have done so far
2021-12-14 06:03:19 <ikke> ie in shellcheck
2021-12-14 06:03:58 <ikke> What I understand from archlinux is that packaging invidiual haskell libraries is a lot of work keeping it up-to-date
2021-12-14 06:06:18 <ikke> It's the same we do for go and rust, but their projects ship the lock files, we do not have to generate them ourselves
2021-12-14 06:08:12 <dhruvin> It indeed is. This came up in #hledger. They mentioned that haskell packages in arch are a bit problematic. For example reverse dependencies are not updated. And since not all dependencies are there in arch repo yet, the builds also fetch some packages from hackage.
2021-12-14 06:09:08 <dhruvin> (relaying this from the creator and maintainer of hledger-iadd-bin)
2021-12-14 06:14:47 <ikke> it doesn't help that cabal v2-* does not make this process easier
2021-12-14 11:14:56 <dhruvin> py3-humanize-v3.3.0 should have py3-setuptools in depends. What should I do? Submit a patch incrementing release number?
2021-12-14 11:15:34 <dhruvin> this breaks todoman currently
2021-12-14 11:28:07 <lopid> looks like the alpine.mirror.wearetriple.com mirror disabled http and now uses https, but after updating /etc/apk/repositories accordingly, `apk update` shows the new https…APKINDEX.tar.gz urls giving a "network error", even though i can download the tarball separately just fine
2021-12-14 11:29:08 <lopid> oh, but not from that host
2021-12-14 11:32:44 <lopid> 
ACTION switches mirror

2021-12-14 11:38:57 <ncopa> dhruvin: should it be a runtime dependency?
2021-12-14 11:39:15 <ncopa> py3-setuptools is normally only needed during install time
2021-12-14 11:40:53 <dhruvin> yes, it seems it is runtime dependency of py3-humanize. I was surprized too: https://github.com/jmoiron/humanize/blob/3.3.0/setup.cfg#L36
2021-12-14 11:46:07 <dhruvin> ncopa: ^
2021-12-14 11:55:10 <ncopa> looks so indeed. maybe create an issue
2021-12-14 11:57:13 <ncopa> hum. i don't think its correct
2021-12-14 11:59:54 <dhruvin> it imports pkg_resources to set __version__ here: https://github.com/jmoiron/humanize/blob/3.3.0/src/humanize/__init__.py#L23
2021-12-14 12:00:09 <mgax> Hello! I've received a security advisory (https://github.com/advisories/GHSA-55x5-fj6c-h6m8) for LXML (`py3-lxml` in Alpine, it's a "community" package). Alpine 3.15 comes with lxml v4.6.4 (https://pkgs.alpinelinux.org/packages?name=py3-lxml&branch=v3.15&arch=x86_64) and the fix should be in v4.6.5. Edge has v4.6.5 but it's "flagged" (don't really know what that means?).
2021-12-14 12:00:09 <mgax> What is the right approach here? Should I install this single package from edge? Can I expect the package to be upgraded sometime in Alpine 3.15, and is there a way to subscribe to a notification when that occurs?
2021-12-14 12:00:09 <mgax> Thanks!
2021-12-14 12:03:16 <ncopa> we should update the package in alpine 3.15
2021-12-14 12:04:00 <lopid> i think alpine's repo has an rss feature
2021-12-14 12:04:40 <ncopa> our security tracker does not find it https://security.alpinelinux.org/vuln/CVE-2021-43818
2021-12-14 12:11:25 <mgax> Yep, found the RSS feed for that folder: https://git.alpinelinux.org/aports/atom/community/py3-lxml?h=master thanks @lopid!
2021-12-14 12:11:25 <mgax> It looks like there was a commit yesterday about `CVE-2021-43818` adding it to the `secfixes` list: https://git.alpinelinux.org/aports/commit/community/py3-lxml?id=e54a8acab6327ee1a21a457f587b2a02da272e57
2021-12-14 12:11:55 <ncopa> mgax: I did that a few mins ago :)
2021-12-14 12:12:12 <ncopa> it is still December 14 here :)
2021-12-14 12:12:38 <mgax> 😆 fair
2021-12-14 12:13:09 <ncopa> and I just pushed the fix to alpine 3.15. should be available from cdn within 15 mins, and the other mirrors within an hour or so
2021-12-14 12:13:39 <mgax> oh wow, totally did not expect such a quick fix, amazing! thanks so much!
2021-12-14 12:14:10 <ncopa> mgax: you were lucky. it was a trivial fix and i had a few mins at hand
2021-12-14 12:39:20 <ncopa> dhruvin: lovley... they need setuptools only to figure out the version. I guess we can patch or sed it to use $pkgver and get rid of the runtime dep
2021-12-14 12:40:52 <ktprograms> ncopa: dhruvin: Maybe this diff (for Qtile) would work: https://github.com/qtile/qtile/pull/2811/files ?
2021-12-14 12:41:21 <ktprograms> Do you want me to do it?
2021-12-14 12:44:10 <ncopa> i was thinking something like sed -i -e "s/VERSION =.*/VERSION = '$pkgver'/" ....
2021-12-14 12:44:33 <ktprograms> ncopa: Ah I suppose that could work too.
2021-12-14 12:44:46 <ncopa> oh, we would need to remove the import as well
2021-12-14 12:46:58 <ktprograms> ncopa: Where is the code that uses setuptools?
2021-12-14 12:52:56 <ncopa> ktprograms: testing/py3-humanize
2021-12-14 12:55:07 <ktprograms> ncopa: I meant the actual source code.
2021-12-14 12:55:38 <ncopa> <dhruvin> it imports pkg_resources to set __version__ here: https://github.com/jmoiron/humanize/blob/3.3.0/src/humanize/__init__.py#L23
2021-12-14 12:57:35 <ktprograms> ncopa: Oh that's the problem, the dep on pkg_resources seems to have been removed in master
2021-12-14 12:59:41 <ktprograms> (Actually since 3.12.0: https://github.com/jmoiron/humanize/releases/tag/3.12.0)
2021-12-14 13:00:36 <dhruvin> ktprograms: https://gitlab.alpinelinux.org/alpine/aports/-/issues/13316
2021-12-14 13:00:53 <dhruvin> the latest humanize code does what qtile did in that PR
2021-12-14 13:01:14 <ktprograms> dhruvin: Yes
2021-12-14 13:01:34 <dhruvin> so we can upgrade it to laatest as well. I'm not sure what other packages will have to be updated though. New to alpine packaging.
2021-12-14 13:01:42 <dhruvin> s/laatest/latest/
2021-12-14 13:01:42 <alpine-meetbot> dhruvin meant to say: so we can upgrade it to latest as well. I'm not sure what other packages will have to be updated though. New to alpine packaging.
2021-12-14 13:06:18 <ktprograms> dhruvin: Seems to be required by 4 other packages: https://pkgs.alpinelinux.org/package/edge/testing/x86_64/py3-humanize (see the 'Required by' dropdown)
2021-12-14 13:15:04 <dhruvin> 
ACTION checking for humanize version constrains in those packages

2021-12-14 13:15:58 <dhruvin> I wonder how twtxt is working with humanize<1 in requirements, but we provide it humanize==3.3 https://github.com/buckket/twtxt/blob/v1.2.3/setup.py#L34
2021-12-14 13:22:11 <dhruvin> pgcli-3.2.0 isn't requiring/using humanize, it uses pendulum instead: https://github.com/dbcli/pgcli/blob/v3.2.0/setup.py#L8
2021-12-14 17:06:26 <vlp[m]> Hello ! On debian I use xinput to manage the touchpad, is there any equivalent on Alpine ?
2021-12-14 17:07:02 <vlp[m]> (to precise: alpine + sway - no X)
2021-12-14 17:22:32 <Misthios> sway-input?
2021-12-14 17:22:54 <ikke> Any knows why nullmailer conflicts on x86? (assuming that it passed on other arches)
2021-12-14 17:25:56 <vlp[m]> Misthios: thanks ! Perfect
2021-12-14 17:26:06 <vlp[m]> I have put it into the sway config file
2021-12-14 19:11:07 <Hello71> ikke: world[coreutils]?
2021-12-14 19:30:43 <ikke> aha
2021-12-14 19:51:56 <zcrayfish> ikke: can you update the Template:AlpineLatest on the wiki ( https://wiki.alpinelinux.org/wiki/Template:AlpineLatest ) to 3.15.0 ?
2021-12-14 19:52:26 <ikke> Done
2021-12-14 19:52:33 <zcrayfish> thanks \o/
2021-12-14 20:39:20 <prohacke_> hello
2021-12-14 20:39:48 <prohacke_> could anyone held me out with init.d scripts?
2021-12-14 20:40:10 <lopid> let's see
2021-12-14 20:42:19 <zcrayfish> what in particular is the issue?
2021-12-14 20:42:30 <prohacke_> i am trying to run docker-compose services via OpenRC services, which works flawlessly. now i extended the services to use environment variables in the docker-compose files, which gets taken from the shell they are started in... my problem, is that i can't get the openrc-run scripts to pass an environment variable to the started script in start(). i even tried starting them with bash -c '<ENV>= s
2021-12-14 20:42:30 <prohacke_> omething docker-compose ....'
2021-12-14 20:43:05 <prohacke_> the equivalent in systemd would be using Environment=">ENV:"
2021-12-14 20:43:44 <prohacke_> i am not used to a non-multiline text input, sorry for the message formatting
2021-12-14 20:50:03 <lopid> did you export the vars?
2021-12-14 20:50:53 <prohackerbro> i just placed them before the call to docker-compose. i'll try exporting them in the same function (fstart())
2021-12-15 06:07:24 <anjan> I sometimes need some programs in testing/ while running stable. Obv making a franken alpine is unsupported but what about compiling from testing/ using abuild?
2021-12-15 06:07:39 <anjan> and just installing the apk that it spits out
2021-12-15 06:08:26 <anjan> I usually send a MR to move to community if the program works.
2021-12-15 06:08:43 <ikke> should work if you build it yourself
2021-12-15 06:09:24 <anjan> okay. I dont want to break my system with upgrades and such.
2021-12-15 06:09:30 <anjan> so thats good to know it's fine to do that
2021-12-15 06:10:23 <ikke> You do need to keep it up-to-date / rebuild it yourself when libraries are bumped
2021-12-15 06:10:58 <anjan> ikke: I see, but the program itself would just fail to launch.
2021-12-15 06:11:07 <anjan> and stable doesnt upgrade libraries that often anyways
2021-12-15 06:11:27 <anjan> Im more worried about breaking the stuff I installed from official repos
2021-12-15 06:11:55 <anjan> which I guess is unlikely since I only install small extras like maim and monero from aports
2021-12-15 07:24:05 <jbg> what conditions exactly need to be met for a repository to be considered trusted?
2021-12-15 07:24:34 <jbg> i'm building a package via `abump`, and it creates a local repository at ~/packages. i've set $PACKAGER_PRIVKEY to the path to an RSA private key in PEM format
2021-12-15 07:24:54 <jbg> the build ends with "Signing the index..." without errors, so I think that part works
2021-12-15 07:25:06 <jbg> the corresponding pubkey has been written to a file ending in .rsa.pub in /etc/apk/keys/
2021-12-15 07:25:11 <jbg> also in PEM format
2021-12-15 07:25:22 <jbg> yet trying to install the package results in "WARNING: Ignoring /home/alpine/packages//main: UNTRUSTED signature"
2021-12-15 07:30:26 <psykose> is the key named the same in both places
2021-12-15 07:31:01 <jbg> ah no
2021-12-15 07:31:08 <jbg> thanks, didn't realise that was important
2021-12-15 07:31:16 <psykose> ran into that earlier myself
2021-12-15 07:31:59 <psykose> the .SIGN file inside the index has the name as part of it
2021-12-15 07:32:10 <psykose> i guess the .RSA implies algo and the rest implies the name of the key
2021-12-15 07:32:50 <jbg> got it
2021-12-15 07:33:23 <jbg> i guess it's for looking up which pubkey in /etc/apk/keys since there could be a lot of them so testing them all would be wasteful
2021-12-15 07:35:08 <psykose> yep
2021-12-15 08:02:25 <jbg> psykose: that fixed it, thanks so much
2021-12-15 08:02:36 <psykose> glad it works :)
2021-12-15 08:03:24 <fcolista> itin
2021-12-15 08:03:38 <fcolista> sorry...wrong window...
2021-12-15 15:05:55 <markand> do we need to setup something to get list of connected users?
2021-12-15 15:06:03 <markand> coreutils's who returns empty list
2021-12-15 15:06:56 <minimal> markand: utmp support is currently not enabled in Alpine for most tools
2021-12-15 15:07:47 <minimal> I have patches for several packages to enable this and have it on my list to submit these in the near future (have other stuff to do first)
2021-12-15 15:08:13 <markand> okay
2021-12-15 15:08:52 <minimal> markand: utmp/wtmp are sort of the "databases" to store the details of logged in users
2021-12-15 15:31:28 <Grabunhold> is there anybody who could do something about https://gitlab.alpinelinux.org/alpine/aports/-/issues/13029 ? 
2021-12-15 16:42:40 <frojnd> Jejks it's been a long time since I've diagnosed a broken hardware. I tried multiple distros including Microsoft Windows so I know it's hardware. At one point compute just becomes unresponsive as in picture freezes. No matter what's behind: Linux/Windoze... any ideas what should how I diagnose? I know hard drive isn't issue because that's what I've checked already. I think I can
2021-12-15 16:42:42 <frojnd> also check ram with memtest or smth
2021-12-15 16:43:25 <frojnd> But besides that if ram is OK, then it might be motherboard some capacitor or smth..
2021-12-15 16:51:23 <koollman> frojnd: gpu overheating
2021-12-15 17:16:58 <Ariadne> does anybody know why firefox screen capture does not work in GNOME wayland? :)
2021-12-15 17:19:09 <Habbie> is your firefox running in X or in Wayland?
2021-12-15 17:19:19 <Ariadne> wayland
2021-12-15 17:19:30 <Habbie> ok, then I do not know :)
2021-12-15 17:19:56 <psykose> do you have pipewire running
2021-12-15 17:20:05 <Ariadne> yes
2021-12-15 17:20:08 <Habbie> oh, do you have pipewire and xdg-desktop-portal-gtk?
2021-12-15 17:20:15 <Ariadne> yes
2021-12-15 17:20:24 <Habbie> these are the hints i recently received for getting screen captures -into X apps-
2021-12-15 17:20:27 <Habbie> (I did not try it yet)
2021-12-15 17:20:28 <psykose> sadly no clue then
2021-12-15 17:20:32 <Hello71> does it work in chromium
2021-12-15 17:20:56 <psykose> chromium just now got the pipewire integration added and needs a flag to enable, but i guess that can be tried
2021-12-15 17:21:20 <Ariadne> Hello71: have not tried
2021-12-15 17:21:29 <Hello71> hm, i thought it needed a flag for firefox too
2021-12-15 17:21:33 <Hello71> maybe it is enabled by default now
2021-12-15 17:21:38 <Hello71> (upstream)
2021-12-15 17:46:34 <frojnd> koollman: this happens after a few minutes of boot. Live usb. Will try to stress test gpu with ultimate boot cd
2021-12-15 18:10:43 <frojnd> Well after I dd it to usb it's not recognized by system. Shame
2021-12-15 18:39:17 <frojnd> Well at this point I would be better to make my own alpine live usb iso with specific packages to stress test gpu
2021-12-15 21:47:52 <anjan> Do you have to configure xdg-desktop-portal-gtk like you have to wlroots?
2021-12-15 21:47:58 <anjan> that might be the issue
2021-12-15 22:02:28 <zcrayfish> Hi there ikke, when you are around again, can you edit another protected page on the wiki... The dead link to the forum on MediaWiki:Sidebar 
2021-12-15 22:02:30 <zcrayfish> https://wiki.alpinelinux.org/w/index.php?title=MediaWiki:Sidebar&action=edit
2021-12-15 22:57:28 <Saijin_Naib[m]> Is anyone using eudev-netifnames with busybox networking, or am I doing yet another Bad Idea (TM)? It changes wlan0 to wlp1s0, which I believe I adjust for properly in /etc/networking/interfaces, but I can not get a link with this package installed. Reboot, change /etc/networking/interfaces back to wlan0, restart and good to go again
2021-12-15 22:58:50 <psykose> did you change wlan0 to wlp1s0 in all the places it's used
2021-12-15 23:00:06 <Saijin_Naib[m]> Within that file? Yes.... (full message at https://matrix.org/_matrix/media/r0/download/matrix.org/McytvuQPhtVFVMzpbJCqfSKU)
2021-12-15 23:00:30 <Saijin_Naib[m]> No errors on boot about bad interface or anything, it just fails to get a DHCP lease
2021-12-15 23:00:56 <Saijin_Naib[m]> Oh, combined with wpa_supplicant. IWD and I are not getting along
2021-12-15 23:00:59 <Saijin_Naib[m]> If that matters
2021-12-15 23:01:26 <psykose> well you could ps the wpa_supplicant line and see if it's correct there too
2021-12-15 23:01:32 <psykose> i forget if networking passes -i for you
2021-12-15 23:01:53 <psykose> if not, it's part of the wpa config
2021-12-15 23:02:03 <Saijin_Naib[m]> I'm also not sure if this is even worth debugging. I don't know if the eudev-netifnames interfaces are more correct or standards compliant or what. Or if I need this for IWD
2021-12-15 23:02:14 <Saijin_Naib[m]> Hmm, let me poke at WPA then, thank you
2021-12-15 23:02:21 <Saijin_Naib[m]> Had not even thought about that half of it
2021-12-15 23:02:53 <psykose> you don't need it for iwd
2021-12-15 23:04:01 <Saijin_Naib[m]> Ahh, okay.
2021-12-15 23:05:28 <Saijin_Naib[m]> If networking is passing the -i option, I should be seeing something like this, right?:... (full message at https://matrix.org/_matrix/media/r0/download/matrix.org/lhlZsDSykjpFxEMbmjcOyJmI)
2021-12-15 23:06:13 <psykose> probably
2021-12-15 23:06:27 <Saijin_Naib[m]> Alright, so it appears that is all working as it should be
2021-12-15 23:06:49 <Saijin_Naib[m]> Is there any objective benefit to using the eudev-netifnames package to get these systemd-style interface names?
2021-12-15 23:13:06 <Saijin_Naib[m]> Don't mind my random leavning -_- DE crashes if I use multitouch gestures too much
2021-12-16 03:22:53 <greatwielder[m]> how do you add the non-free repository? its not on the cdn
2021-12-16 03:27:48 <zcrayfish> I'm not aware of one. main, community, and testing are the only ones AFAIK.
2021-12-16 03:28:16 <zcrayfish> but if I am wrong, I hope someone speaks up and embarrasses me.
2021-12-16 03:28:46 <greatwielder[m]> apparently there is non-free in aports. and they moved unrar to it. widevine is also there.
2021-12-16 03:30:28 <bdju> give p7zip a shot, it should be able to extract rars. `7z e filename.rar`
2021-12-16 03:31:00 <zcrayfish> Indeed. I have not used WinRAR in years thanks to 7zip
2021-12-16 03:31:09 <greatwielder[m]> i know, but not all rar files.
2021-12-16 03:31:11 <greatwielder[m]> https://git.alpinelinux.org/aports/tree/non-free
2021-12-16 03:32:59 <zcrayfish> Looks really easy to build at least judging by the APKBUILD and single patch
2021-12-16 03:33:52 <PJ[m]> https://wiki.alpinelinux.org/wiki/Creating_an_Alpine_package
2021-12-16 03:34:11 <PJ[m]> setup your system and account explains how to prepare your machine for building packages locally
2021-12-16 03:35:12 <PJ[m]> then section 5. Build the package
2021-12-16 04:10:50 <Hello71> use unar or bsdtar (although neither is very secure)
2021-12-16 05:40:12 <greatwielder[m]> is there a difference between rc-update  and  ln -fs /etc/init.d/$svc /etc/runlevels/$level ?
2021-12-16 06:29:43 <PureTryOut> so something is holding back the python3.10 upgrade on my system, but the "conflicts and satisfies" output of apk is so big, I have no clue what is actually blocking it...
2021-12-16 06:31:22 <PureTryOut> oh I'm probably just too quickly, seems the builders haven't finished rebuilding all packages that got bumped yet
2021-12-16 08:37:56 <ikke> Only main
2021-12-16 08:38:06 <ikke> Community not yet
2021-12-16 10:58:01 <Alphare> Hi all. Is there a way of getting a list of all packages in the different categories? (main, community, etc.) I'm trying to get an estimate on how many there are for the Software Heritage initiative
2021-12-16 11:03:45 <ktprograms> Alphare: By listing the number of folders in each repo's directory in https://gitlab.alpinelinux.org/alpine/aports (clone it locally)?
2021-12-16 11:05:20 <Alphare> seems like it's about 8544 packages
2021-12-16 11:06:12 <Alphare> thanks ktprograms I thought that would be the solution as well
2021-12-16 11:21:48 <frag> can apk list world, or only cat /etc/apk/world?
2021-12-16 11:25:27 <ktprograms> frag: Alphare: wants to be able to list the total number of available packages, not number of installed packages.
2021-12-16 11:26:50 <Alphare> correct
2021-12-16 11:28:11 <frag> ktprograms: i was just asking a question for myself
2021-12-16 11:28:21 <frag> can 'apk' list world?
2021-12-16 11:28:50 <ikke> No
2021-12-16 11:29:21 <frag> okok
2021-12-16 11:30:10 <ktprograms> frag: Sorry about the mistake
2021-12-16 11:31:13 <frag> ktprograms: np buddy
2021-12-16 13:52:47 <Hello71> Alphare: you may have better results asking repology
2021-12-16 14:11:39 <greatwielder[m]> are the devfs, procfs, sysfs services needed for a functioning system? does standard alpine some with them enabled? using minirootfs as bootstrap.
2021-12-16 14:13:08 <Alphare> Hello71: good point!
2021-12-16 17:08:08 <wyk72> hi
2021-12-16 17:08:28 <wyk72> I have a big problem running a samba server from RAM, diskless mode
2021-12-16 17:08:38 <wyk72> the passwords are not saved in the overlay file
2021-12-16 17:08:52 <wyk72> and I have no clue to where those passwords are
2021-12-16 17:09:13 <wyk72> triev to add /var/lib/samba to lbu (lbu include /var/lib/samba)
2021-12-16 17:09:18 <wyk72> but is not working
2021-12-16 17:10:03 <wyk72> I can't find anywhere the location of the smbpasswd file or the binary archive or something else
2021-12-16 17:10:25 <wyk72> any pointers??
2021-12-16 17:10:55 <wyk72> latest Alpine, 3.15, Samba 4.15.2
2021-12-16 17:11:47 <wyk72> I mean ,the passwords of the commands "smbpasswd -a"
2021-12-16 17:12:12 <wyk72> I can write a shell script to local startup, I think, but it's plain ugly
2021-12-16 17:12:26 <wyk72> and samba/smb is ugly enough
2021-12-16 17:12:26 <ikke> wyk72: did you look in $HOME?
2021-12-16 17:12:36 <wyk72> yup
2021-12-16 17:12:45 <wyk72> nothing there
2021-12-16 17:17:58 <psykose> smbpasswd is in like /usr/local/samba by default
2021-12-16 17:18:34 <psykose> but you can change the location in the samba conf
2021-12-16 17:24:57 <wyk72> samba 4.15 does not use any smapasswd
2021-12-16 17:25:04 <wyk72> and it's not in /usr/local/samba
2021-12-16 17:25:11 <psykose> it is wherever the private dir is
2021-12-16 17:25:20 <wyk72> nope
2021-12-16 17:25:27 <wyk72> it's not there
2021-12-16 17:25:38 <wyk72> maybe I've found a solution
2021-12-16 17:26:00 <wyk72> had to add the line
2021-12-16 17:26:04 <wyk72> " passdb backend = tdbsam:/etc/samba/passdb.tdb"
2021-12-16 17:26:10 <wyk72> to the smb.conf file
2021-12-16 17:26:15 <nickersonm> ls /var/cache/samba
2021-12-16 17:26:23 <wyk72> so that the tdb passwd database is stored there
2021-12-16 17:26:58 <wyk72> ok
2021-12-16 17:27:02 <psykose> yes, the default for that is the same just in $prefix/private/passdb.tdb or something
2021-12-16 17:27:09 <wyk72> so I have to add /var/cache to lbu ?
2021-12-16 17:27:10 <psykose> use find or whatever for .tdb
2021-12-16 17:27:15 <wyk72> it seems is in /var/cache
2021-12-16 17:27:19 <wyk72> looks ugly to me
2021-12-16 17:27:21 <psykose> ah, good to move it then
2021-12-16 17:28:53 <ikke> yeah, adding /var/cache to lbu does not sound like a good strategy
2021-12-16 17:33:54 <wyk72> I can confirm that adding "passdb backend = tdbsam:/etc/samba/passdb.tdb"
2021-12-16 17:33:57 <wyk72> solves the problem
2021-12-16 17:34:13 <wyk72> and the users are properly loaded after a cold reboot
2021-12-16 17:34:33 <wyk72> this thing took me 4 hours
2021-12-16 17:34:58 <wyk72> it seems this samba password file/backed is a moving target
2021-12-16 17:35:59 <wyk72> it should be written somewhere, I think
2021-12-16 17:36:05 <wyk72> in the alpine docs
2021-12-16 17:36:17 <wyk72> or wiki
2021-12-16 17:40:12 <nickersonm> looks like the default is specified at compile time via --with-lockdir : https://git.alpinelinux.org/aports/tree/main/samba/APKBUILD
2021-12-16 17:44:33 <nickersonm> But you can also find it via testparm -vs | grep "lock directory"
2021-12-16 17:47:33 <wyk72> lock directory = /var/cache/samba
2021-12-16 17:47:39 <wyk72> thanks
2021-12-16 17:49:25 <wyk72> it's a not a good option for a ram/diskless/lbu Alpine samba server to have stuff in /var/cache, I think
2021-12-16 17:50:18 <wyk72> anyway adding that line to smb.conf solves the problem nicely and more "cleanly" if I can say so
2021-12-16 17:52:56 <psykose> what other files are in there
2021-12-16 17:53:13 <psykose> based on the name it is temporary things, i guess it's more weird that lockdir is used for private dir
2021-12-16 17:53:21 <psykose> but the default smb.conf should perhaps be changedd
2021-12-16 18:04:10 <nickersonm> wyk72: there's also some persistent files in /var/lib/samba by default ('state directory')
2021-12-16 18:04:58 <nickersonm> wyk72: I believe that's actually where the default passdb files are
2021-12-16 18:10:57 <nickersonm> psykose: looks like /var/cache/samba is actually for temporary files; /var/lib/samba is for the state files
2021-12-16 18:15:48 <psykose> yes, and for some reason by default it writes the password file to cache
2021-12-16 18:17:11 <nickersonm> psykose: the default passdb is actually in the state directory/private (/var/lib/samba/private), I was incorrect when I directed wyk72 to the lock dir before.
2021-12-16 18:17:24 <psykose> ah
2021-12-16 18:17:38 <psykose> seems like everything is fine there
2021-12-16 19:45:27 <blurt> anyone know how to get net.* dependencies working for a service?
2021-12-16 19:46:08 <blurt> for example i'd like dnsmasq to require eth1, but adding rc_need="net.eth1" or rc_need="!net net.eth1" just results in this output when starting dnsmasq:  Service `dnsmasq' needs non existent service `net.eth1'
2021-12-16 20:01:47 <Hello71> i think that's gentoo-specific
2021-12-16 20:02:00 <Hello71> for dnsmasq you should use freebind anyways
2021-12-16 20:02:31 <Hello71> hm, although i'm not sure if that works for dhcp-range constructor
2021-12-16 20:05:45 <blurt> i set the bind interface in dnsmasq.conf so i'm not too worried about that service.  i wanted to turn up a VPN on an external interface and apply some IPTables when both interfaces are up using two different custom services though so i was trying to tie those to their respective interfaces
2021-12-16 20:06:28 <blurt> both interfaces meaning the internal network and external one
2021-12-16 20:07:24 <blurt> i might just write my own net.eth1 service too then.  from what i can see in /etc/init.d/networking it doesn't seem too difficult 
2021-12-16 20:42:31 <bl4ckb0ne> is there an ETA for the python 3.10 package community/testing rebuild?
2021-12-16 20:51:21 <psykose> when it's done on the builders https://build.alpinelinux.org/
2021-12-16 20:56:06 <bl4ckb0ne> ty
2021-12-16 20:57:01 <greatwielder[m]> is there a way to move an interface from a named netns into an unnamed netns? ip -n named link set iface0 netns NAME   actually expects a name and not  /proc/$$/ns/net     and do it without actually creating a named interface.   or perhaps there is a way to get a named netns to auto delete itself when nothing is running in it?
2021-12-16 21:14:18 <psykose> what do you mean by unnamed netns? why would you want to make one without a name
2021-12-16 21:18:58 <greatwielder[m]> psykose:  once execution stops in an unnamed netns it is gone and interfaces in it are deleted. named netns may become orhpans.
2021-12-16 21:19:49 <psykose> how do you make one
2021-12-16 21:23:24 <greatwielder[m]> unshare -n
2021-12-16 21:23:39 <greatwielder[m]>  * unshare -n bash
2021-12-16 21:33:33 <anjan> someone flagged one of my packages out of date
2021-12-16 21:33:37 <anjan> but it's not
2021-12-16 21:33:45 <anjan> https://pkgs.alpinelinux.org/package/edge/community/armhf/ufw#
2021-12-16 21:33:52 <anjan> they said it doesnt work with 3.10.2
2021-12-16 21:33:57 <psykose> i don't think there is a way to address that netns privately
2021-12-16 21:34:02 <anjan> is that version of alpine even supported anymore?
2021-12-16 21:34:13 <psykose> no
2021-12-16 21:34:21 <psykose> only 3.15 community is supported
2021-12-16 21:34:30 <anjan> ok, how do I remove this flag
2021-12-16 21:34:36 <anjan> ugh, why do people do this
2021-12-16 21:35:28 <ikke> anjan: we do not have any interface for that atm
2021-12-16 21:35:54 <anjan> they said 3.16 is out even though it's not
2021-12-16 21:36:00 <ikke> lol
2021-12-16 21:36:09 <anjan> so that they could say "why arent you supporting 3.10.2
2021-12-16 21:36:18 <anjan> Imagine annoying people giving something you for free
2021-12-16 21:36:40 <anjan> ok, no worries
2021-12-16 21:58:15 <Habbie> greatwielder[m], openwrt netifd can move netnses into unnamed netns given a PID that is in them
2021-12-16 22:13:52 <Hello71> according to ip-link(8), set netns takes either PID or NETNSNAME
2021-12-16 22:23:11 <greatwielder[m]> Hello71: where did you find PID in the manpage? There is ip netns attach NETNSNAME PID  but nothing for set.
2021-12-16 22:24:27 <Hello71> what man page are you looking at
2021-12-16 22:26:42 <greatwielder[m]> Hello71: apparently the wrong one, thanks.
2021-12-16 22:40:16 <Habbie> Hello71, ah yes!
2021-12-17 03:20:00 <ktprograms> Hi, I'm getting the error 'Error relocating /mnt/mcmtools/host/bin/ld: qsort_r: symbol not found' during the build of musl when using https://git.zv.io/toolchains/bootstrap/-/blob/master/bootstrap. The full build log (ignore the lines after the collect2 error) is here: https://tpaste.us/w1B9. Thanks!
2021-12-17 03:29:05 <Hello71> apk upgrade -a; rm -fr "/mnt/mcmtools"
2021-12-17 03:30:02 <ktprograms> Hello71: How will that help?
2021-12-17 03:30:19 <ktprograms> FYI I just apk upgradeed before starting the build.
2021-12-17 03:30:55 <Hello71> hm, wait, it should be backward compatible, not forward
2021-12-17 09:20:11 <ph88> hello
2021-12-17 09:20:18 <ph88> how can i install the psql cli tool on alpine ?
2021-12-17 09:24:52 <clandmeter> ph88: apk search cmd:psql
2021-12-17 09:26:03 <Misthios> i think its in postgresql-client
2021-12-17 09:27:49 <ph88> clandmeter, i get this when i run that command https://bpa.st/4AUA
2021-12-17 09:28:06 <clandmeter> ph88: apk update first
2021-12-17 09:28:11 <clandmeter> with first :)
2021-12-17 09:28:14 <clandmeter> without*
2021-12-17 09:28:45 <ph88> yes works thanks clandmeter 
2021-12-17 09:28:53 <ph88> Misthios, you were right about the package name
2021-12-17 09:29:30 <ph88> wow apk is so fast, ive never seen that before
2021-12-17 09:29:51 <clandmeter> :)
2021-12-17 09:30:49 <clandmeter> any command can be searched for with cmd:xxx
2021-12-17 09:30:54 <ph88> ok
2021-12-17 09:31:16 <clandmeter> as long as you have indexes updated of course
2021-12-17 09:31:46 <clandmeter> for a oneliner you could add, --update-cache or --no-cache
2021-12-17 09:39:42 <Misthios> cmd: doesnt work for everything toh
2021-12-17 09:39:56 <Misthios> cmd:firefox does not return anything for example
2021-12-17 09:40:20 <clandmeter> it depends where the command lives
2021-12-17 09:40:29 <clandmeter> if its in general path it should be included
2021-12-17 09:40:48 <Misthios> does it show fr u?
2021-12-17 09:43:15 <clandmeter> hmm, not sure why its not included
2021-12-17 09:45:23 <ph88> would be cool if alpine had a website like ubuntu where you can search for packages after some text in a filename
2021-12-17 09:45:40 <ph88> i also use that for library and header files
2021-12-17 09:49:20 <clandmeter> ph88: something like https://pkgs.alpinelinux.org/packages ?
2021-12-17 09:51:33 <ph88> ye something like that but then with added functionality like here https://packages.ubuntu.com/ where you can search into package contents
2021-12-17 09:51:51 <clandmeter> check the menu
2021-12-17 10:05:35 <psykose> contents has contents
2021-12-17 10:05:42 <psykose> in fact i have never had it fail me, it's quite good
2021-12-17 10:31:10 <ph88> oh i see cool
2021-12-17 10:50:08 <Misthios> any chance of it being added to apk itself like apk search content:x
2021-12-17 10:50:16 <Misthios> or is there something like that arleady
2021-12-17 10:50:24 <psykose> it needs a database to search that is not part of the index
2021-12-17 10:50:41 <psykose> there is a tool i forgot the name of that generates it, same as on the site
2021-12-17 10:50:45 <psykose> i forgot what the package was
2021-12-17 10:51:04 <Misthios> apk-ttols-lua?
2021-12-17 10:51:28 <Misthios> or lua- something *
2021-12-17 10:52:15 <psykose> maybe an ap command from lua-aports, but i really don't remember
2021-12-17 10:53:47 <psykose> Misthios: are you by any chance the same misthios from ao3
2021-12-17 10:53:58 <Misthios> ao3?
2021-12-17 10:54:22 <psykose> don't worry about it :)
2021-12-17 10:57:47 <clandmeter> it has nothing to do with lua-aports
2021-12-17 10:58:06 <clandmeter> there is a cli tool that queries pkgs.a.o 
2021-12-17 10:58:14 <clandmeter> apkfile or something like that
2021-12-17 10:58:37 <clandmeter> adding that info to the database has been disucussed a few times
2021-12-17 10:58:43 <psykose> apk-file
2021-12-17 10:58:55 <clandmeter> it would increase the apkindex by 1000x
2021-12-17 10:59:06 <psykose> yeah, it should not be in there
2021-12-17 10:59:50 <clandmeter> it could be added in a seperate file, but this would also be huge
2021-12-17 11:00:03 <psykose> it's certainly takes forever to give results
2021-12-17 11:00:11 <psykose> and the output is broken
2021-12-17 12:00:20 <minimal> anyone know of a package to HOTP/TOTP (aka Google Authenticator) 2FA-enable both sshd and console logins *without* using PAM?
2021-12-17 12:08:08 <psykose> map backwards (no)
2021-12-17 12:08:47 <psykose> i don't think one exists
2021-12-17 12:11:12 <minimal> psykose: yes my searches so far haven't turned up anything
2021-12-17 12:11:33 <psykose> the most i've seen without pam is ssh-sk for security key, which is obviously not the same thing and wildly different
2021-12-17 12:13:28 <minimal> psykose: yes openssh recently added builtin support for U2F hardware keys. I guess something could be kludged using "ForceCommand" with OpenSSH
2021-12-17 23:55:33 <Thermi> Is the libldap.so that is part of openldap the threaded one (actually called libldap_r.so), or the normal, unthreaded one?
2021-12-18 00:46:48 <ktprograms> Hello71: I just realised, I've never used the bootstrap script to bootstrap musl-cross-make on Alpine, I've only done it on Debian, so I don't think it's any recent update that's causing the symbol not found.
2021-12-18 00:47:00 <Hello71> maybe
2021-12-18 00:47:46 <ktprograms> Although I think it was working at some point because the repo has a docker based build using Alpine
2021-12-18 02:40:41 <ktprograms> Hello71: Hmm, it seems to be building fine on aarch64, not sure if it's an x86_64 specific problem.
2021-12-18 04:13:25 <amberlee[m]> something is wrong with alpine kernels lately. extract_vmlinux seems to be failing. readelf -n on result ends with  readelf: Warning: note with invalid namesz and/or descsz found at offset 0x80    and  readelf: Warning:  type: 0xffffffff, namesize: 0x006e6558, descsize: 0x80000000, alignment: 8
2021-12-18 04:13:34 <amberlee[m]> linux-virt sppecifically
2021-12-18 04:14:06 <amberlee[m]> tried main and edge
2021-12-18 04:15:48 <amberlee[m]> extraction succeeds on  other kernels  I tried.
2021-12-18 04:17:09 <amberlee[m]> noticed problem when i tried using linux-virt as kernels for a VMM: VmBoot(KernelMissingPvhHeader)   but after checking  alpine does have CONFIG_PVH=y.
2021-12-18 14:46:47 <Hello71> main and edge?
2021-12-18 15:14:34 <amberlee[m]> 5.15.5 and and 5.15.10
2021-12-18 15:19:00 <ikke> Anyone else having issues with firefox complaining it from already running when exucuting it more then one or opening a link through xdg-open?
2021-12-18 15:25:09 <psykose> you probably need to enable MOZ_DBUS_REMOTE=1 if you are on wayland and have dbus
2021-12-18 15:27:31 <psykose> maybe not even wayland related though, and for the whole environment
2021-12-18 15:27:51 <ikke> Not using wayland
2021-12-18 15:28:02 <ikke> but let me check if that helps
2021-12-18 15:34:21 <ikke> psykose: does not help
2021-12-18 15:34:49 <psykose> weird
2021-12-18 16:00:01 <ikke> https://momi.ca/posts/2021-10-11-startxdebian.html
2021-12-18 16:04:27 <psykose> ah, you can also do `startx dbus-run-session -- wm` or similar
2021-12-18 16:04:46 <ikke> ok, let me try that
2021-12-18 16:05:33 <psykose> or `startx dbus-launch --exit-with-x11 wm -- :0` which might be better
2021-12-18 16:05:37 <psykose> some examples in here https://lists.freedesktop.org/archives/dbus/2018-November/017640.html
2021-12-18 16:09:38 <ikke> So I typically just run startx, and in my .xinitrc I run exec awesomewm
2021-12-18 16:13:06 <ikke> So I wonder if I should change it to startx dbus-launch .. -- awesomewm, or change it in my .xinitrc
2021-12-18 16:13:53 <hoel> I guess you can set exec dbus-launch … -- awesomwn in your .xinitrc.
2021-12-18 16:14:15 <ikke> right
2021-12-18 16:16:12 <ikke> Nice, that works
2021-12-18 16:19:55 <ikke> Hmm, it even works without it :D
2021-12-18 16:35:26 <psykose> grats :)
2021-12-19 04:08:45 <ubq323> is there anything like w or who that doesn't require installing all of coreutils
2021-12-19 04:12:07 <bdju> `users` maybe
2021-12-19 04:17:54 <ubq323> what package is that in?
2021-12-19 04:22:24 <bdju> oh, damn. I tihnk i
2021-12-19 04:22:28 <bdju> it's also coreutils
2021-12-19 04:23:06 <bdju> not at an alpine machine atm but was just looking up similar commands
2021-12-19 04:30:57 <bigboi> im finally in. hi folks, i came in about a week ago but i am still looking for a little bit of help. to put it short i am trying to use alpine on my nvme uefi setup - and the new cyryptdisk feature in the installer kept failing me. i forgot the specific details but i was dropped into emergency shell on boot. is there anything specific i should try
2021-12-19 04:30:57 <bigboi> here in these next few minutes?
2021-12-19 04:39:32 <bigboi> i'm seeing an update to setup-disks on 11-16 which adds the cryptdisk kernel parameter. going to try updating before install and report my findings here
2021-12-19 04:46:27 <bigboi> apk update, apk upgrade and apk fix alpine-conf should be sufficient to ensure the installer is upgraded correct?
2021-12-19 04:48:02 <ktprograms> bigboi: It should
2021-12-19 04:55:28 <bigboi> same issue, but i did see the installer mention there was no uuid for the swap partition. i will try to chroot and see if disabling swap would help
2021-12-19 04:55:36 <bigboi> or rather invalid uuid
2021-12-19 05:19:28 <bigboi> lol finally chrooted in, commented it out of the fstab and did swapoff -a. gonna check grub and see if the uuid is right next.
2021-12-19 05:47:06 <bigboi> :(  tried rebuilding mkinitfs with the flags it was missing but still at the same spot. now im in no mans land and the actual error im getting isn't even fully in view. im guessing from some of the letters i can make it out it is the  "can not mount /sysroot" or similar error
2021-12-19 06:19:48 <amberlee[m]> when trying to open firefox or chromium I get: Error relocating /usr/bin/firefox: _ZSt28__throw_bad_array_new_lengthv: symbol not found     I installed alpine from a minirootfs bootstrap, am I missing something? used   apk add alpine-base util-linux   and  for svc in devfs procfs sysfs; do ln -fs /etc/init.d/$svc /etc/runlevels/boot; done
2021-12-19 12:42:24 <ikke> telmich: cdist fails to build because https://code.ungleich.ch/ungleich-public/cdist/uploads/d342a7e52f1dab308e9ae155c7204a61/cdist-6.9.8.tar.gz returns 404
2021-12-19 12:45:28 <psykose> https://code.ungleich.ch/ungleich-public/cdist/archive/6.9.8.tar.gz
2021-12-19 13:56:35 <Hello71> amberlee[m]: try apk upgrade -a
2021-12-19 13:56:55 <Hello71> also that's not enough services for a full installation and openrc doesn't run in chroot
2021-12-19 14:47:41 <amberlee[m]> Hello71: I'm running it in a VM and apk upgrade -a didn't seem to do anything, problem persists.   while firefox has just a single error chromium has 4. all are Error relocating /usr/lib/chromium/chrome  followed by _ZSt28__throw_bad_array_new_lengthv: symbol not found    _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE7reserveEv: symbol not found   _ZNSt7__cxx1112basic_stringIwSt11char_traitsIwESaIwEE7reserveEv: symbol not found
2021-12-19 14:47:41 <amberlee[m]> FT_Get_Color_Glyph_ClipBox: symbol not found
2021-12-19 14:47:56 <amberlee[m]> not all software is like this, a terminal emulator runs fine
2021-12-19 14:47:57 <Hello71> try upgrading to edge
2021-12-19 14:52:21 <amberlee[m]> that seems to have fixed the problem, but why is 3.15 broken??
2021-12-19 14:55:06 <Hello71> probably boost
2021-12-19 14:55:46 <Hello71> although it's odd that those aren't boost symbols
2021-12-19 14:55:59 <Hello71> can you file an issue on gitlab
2021-12-19 15:00:03 <amberlee[m]> I checked the setup scripts carefully and firefox and chromium are installed from edge while the system is 3.15. that used to never be a problem but maybe now there are compatibility issues?
2021-12-19 15:00:21 <Hello71> oh, yes, that's never supported
2021-12-19 15:00:31 <Hello71> it will sometimes work but when it doesn't you get to pick up the pieces
2021-12-19 15:01:08 <psykose> which setup scripts? are these some scripts you wrote yourself
2021-12-19 15:01:42 <amberlee[m]> other people in the org, yes. does edge break often?
2021-12-19 15:02:47 <psykose> it's not to do with edge breaking, but simply dependencies get updated and so the 3.15 dependencies wont work anymore. this happens fairly often i would assume for something with 100 dependencies like a web browser, it's not supported for a reason
2021-12-19 15:02:53 <psykose> as far as edge itself breaking, ehhh
2021-12-19 15:03:11 <psykose> depends what you consider broken, usually it's random small things, and mostly it's testing
2021-12-19 15:04:33 <psykose> this specific break is probably because edge got gcc11 and so libstdc++ from it has different symbols
2021-12-19 15:04:49 <psykose> but it doesn't really matter what specifically caused it
2021-12-19 15:05:42 <amberlee[m]> sorry i meant that as an unrelated question, does edge break as often or more often than rolling release distros such as arch and void (which is basically never)? and therefore suitable for normal use.
2021-12-19 15:06:34 <psykose> more often, but i use it daily without real issue- it depends how much you value something randomly being broken in the middle of work that you don't expect, and depends what you use i guess
2021-12-19 15:07:39 <psykose> if you stay away from testing/ you will dodge.. most of the random breaks that i can think of, but no promises :)
2021-12-19 17:26:38 <MY-R> I was really so happy to install Alpine on another machine, until next reboot when only what I see is "Loading initrd...." and nothing else, pc completly hang
2021-12-19 17:26:48 <MY-R> is really all about that crap simpledrm?
2021-12-19 17:27:57 <donoban> do you have fde MY-R ?
2021-12-19 17:29:04 <MY-R> is fresh install, after first reboot all was fine, installed eudev then setup-udev and few small packages, happy disconnected monitor, so could use ssh and boom....
2021-12-19 17:29:49 <MY-R> ah ye and rebuild initramfs with "mkinitfs" so that probably mess up or added some gpu kms modules what conflict with that damn simpledrm stuff
2021-12-19 17:30:20 <donoban> can you ping or ssh it?
2021-12-19 17:30:21 <MY-R> so most probably after next kernel upgrade I would stuck with it
2021-12-19 17:30:24 <MY-R> no 
2021-12-19 17:30:44 <MY-R> zero ping, it doesnt boot more at all, hang, nothing work
2021-12-19 17:31:29 <MY-R> normal grub sys install without anything extra, so no encryption
2021-12-19 17:32:00 <donoban> if it was a problem with simpldrm it should boot fine without screen
2021-12-19 17:32:15 <donoban> I had problem with it and I could blindly enter my passphrase
2021-12-19 17:33:32 <MY-R> I have got that problem before too on gentoo with simpledrm too and same, even capslock didnt work, rebuilt kernel without simpledrm extra stuff resolved the problem
2021-12-19 17:34:39 <donoban> can do anything with a kernel argument?
2021-12-19 17:34:54 <MY-R> ye ofc, tried nomodeset but didnt change anything
2021-12-19 17:35:04 <MY-R> grub edit mode working
2021-12-19 17:35:25 <MY-R> there was some simpledrm thing which could pass to kernel but already forgot what was it
2021-12-19 17:35:45 <MY-R> and it isnt fun to replugin monitor and doing hard resets over and over :\
2021-12-19 17:35:49 <MY-R> donoban: any sugestions?
2021-12-19 17:36:35 <donoban> it's a real or virtual machine?
2021-12-19 17:37:07 <MY-R> real older x86_64 pc, amd cpu and old radeon gpu
2021-12-19 17:37:57 <donoban> have you tried to blacklist simpledrm?
2021-12-19 17:38:00 <MY-R> pretty sure if wouldnt do mkinitfs then all would work fine but then in next kernel update would have same
2021-12-19 17:38:56 <MY-R> donoban: module.blacklist=simpledrm.ko or how it can look?
2021-12-19 17:39:50 <donoban> I'm not sure, maybe 'modprobe.blacklist=drivername'
2021-12-19 17:51:50 <amberlee[m]> any ideas why libxkbcommon.so segfaults and application reports and xkbcommon: ERROR: Couldn't read rules file "/usr/share/X11/xkb/rules/evdev": No such device ?   /usr/share/X11/xkb/rules/evdev exists
2021-12-19 17:54:14 <MY-R> donoban: eh, nothing work, it suposed to be headless server and Im fighting with some gpu crap
2021-12-19 17:54:40 <psykose> amberlee[m]: do you have eudev or mdevd running
2021-12-19 17:54:57 <donoban> paradoxical
2021-12-19 17:55:33 <psykose> lmarz: did you get my replies
2021-12-19 17:55:44 <donoban> if do you chroot and try to regenrate initramfs?
2021-12-19 17:56:48 <MY-R> donoban: faster will reinstall it and remove any kms from features.d if any there, maybe that will help
2021-12-19 17:58:13 <psykose> you can chroot, remove it, and apk fix mkinitfs
2021-12-19 18:00:51 <amberlee[m]> psykose: that may be it, thanks. is there a list of which services should be running on a fresh on-hardware install?
2021-12-19 18:01:57 <psykose> aside from the simplest of apk add udev, setup-udev, i can't think of any
2021-12-19 18:02:11 <psykose> seatd if you use sway i guess, 
2021-12-19 18:02:18 <psykose> cgroups if you need those
2021-12-19 18:02:23 <amberlee[m]> or a guide on how to manually install alpine on hardware that would include these details? such as a chroot install with apk from a different linux distro (independant of alpine isos)
2021-12-19 18:03:08 <psykose> setup-udev is part of the whatever setup-xorg stuff randomly scattered on the wiki, but i forget if there is a decent guide
2021-12-19 18:03:40 <psykose> i'm also not sure how it is on-hardware if it's on a chroot
2021-12-19 18:04:41 <amberlee[m]> ex usecase: chroot install on partition from another distro running then reboot to alpine.
2021-12-19 18:05:36 <psykose> probably easier to just boot an iso and install to that partition just like you would any multiple-os setup
2021-12-19 18:05:38 <amberlee[m]> but mainly the rpoblem appears to be with a VMM that direct boots a kernel. the isos are useless and bootstrap from minirootfs results in these problems.
2021-12-19 18:11:07 <amberlee[m]> psykose: mdev is now running but the problem persists, should i use udev instead?  Couldn't read rules file "/usr/share/X11/xkb/rules/evdev": No such device
2021-12-19 18:12:42 <psykose> maybe linux-firmware
2021-12-19 18:13:18 <psykose> but if this isn't hardware and is some virtual machine booting thing then i don't know what issues you could run into
2021-12-19 18:14:45 <psykose> you can also post rc-update -a
2021-12-19 18:16:44 <amberlee[m]> acpid |      default    agetty.ttyS0 |      default   crond |      default   devfs | boot    loadkmap | boot   local |      default  mdev |                                 sysinit    networking | boot    procfs | boot     sysfs | boot     urandom | boot
2021-12-19 18:22:08 <psykose> for reference mine is https://img.ayaya.dev/Z7Nq6cE8ESQV (pretend async is default)
2021-12-19 18:22:22 <psykose> idk if this is what the issue is but you are missing some of the ones normally there from a normal iso install
2021-12-19 18:24:58 <minimal> MY-R: Alpine 3.15 or Edge?
2021-12-19 18:40:10 <MY-R> minimal: 3.15
2021-12-19 18:41:03 <MY-R> but ye, I didnt have any "kms" in initfs config, added, done fix mkinitfs in chroot, reboot and is same
2021-12-19 18:41:35 <MY-R> Im glad it came out now and later when kernel upgrade enforce mkinitfs and leave me with "brick" router
2021-12-19 18:41:44 <MY-R> and not later*
2021-12-19 18:42:54 <minimal> MY-R: ok, so are you staying on 3.15 or upgrading to Edge?
2021-12-19 18:43:47 <MY-R> minimal: I wanted make router from this machine so 3.15
2021-12-19 18:44:48 <minimal> MY-R: and when you installed did you try running "apk -Ula upgrade" to ensure latest version of packages are installed?
2021-12-19 18:45:39 <MY-R> minimal: I used official 3.15 iso and setup-alpine with cdn mirror
2021-12-19 18:46:28 <MY-R> minimal: you think that some package could stuck and didnt download correctly durin installation?
2021-12-19 18:47:05 <minimal> MY-R: that's not what I asked. there were issues with mkinitfs and alpine-conf after 3.15 release
2021-12-19 18:47:33 <minimal> "apk -Ula upgrade" ensure latest versions from mirror are installed
2021-12-19 18:48:15 <MY-R> ye but same doing installer, ok I will chroot once again and run it but doubt it will be anything to upgrade and check with apk version if nothing left
2021-12-19 18:48:17 <psykose> the 3.15 iso doesn't have those issues
2021-12-19 18:48:52 <minimal> then assuming you do have latest versions the next point is mkinitfs config - you already found you didn't have "kms" in the features list and so module for your graphics card was missing from initramfs
2021-12-19 18:49:40 <minimal> the next point is whether you had the firmware package for your graphics card installed *prior* to running running "mkinitfs" command so that any relevany firmware is also included in the new initramfs that mkinitfs generates
2021-12-19 18:50:33 <minimal> so typically linux-firmware-i915, or linux-firmware-amdgpu, or linux-firmware-radeon
2021-12-19 18:52:30 <minimal> psykose: was not sure if the 3.15 iso contained the fixed versions of mkinitfs and alpine-conf (for compressed kernel modules and correctly determining firmware needed for compressed GPU kernel modules)
2021-12-19 18:53:42 <minimal> MY-R: did you have firmware installed? ^^^
2021-12-19 18:54:04 <psykose> the 3.15 iso contains 3.6.0 which was released 2 days before 3.15
2021-12-19 18:55:44 <minimal> psykose: ok, there's still the matter of "kms" and of graphics firmware...
2021-12-19 18:56:09 <jhjacobs81> hmz
2021-12-19 18:56:49 <MY-R> since when kms is default enabled in alpine?
2021-12-19 18:57:23 <MY-R> minimal: all is up to date, every firmware is installed since installer doing it
2021-12-19 18:59:51 <minimal> MY-R: you said you ran mkinitfs after initial installation - that will *re-generate* initramfs so if the /etc/mkinitfs/mkinitfs.conf config does not match what it was when the installer's initramfs was built then you will end up with different contents, which might explain why it fails to boot again after you ran mkinitfs
2021-12-19 19:00:56 <MY-R> minimal: but why id doesnt match, so how default mkinitfs.conf should look?
2021-12-19 19:02:06 <amberlee[m]> psykose: adopting your services fixed the problem, not sure which one did it. will try eliminating some later. thanks.
2021-12-19 19:04:10 <MY-R> minimal: after installation and reboot I have got just "ata base ide scsi usb virtio ext4" 
2021-12-19 19:06:12 <MY-R> all was upgraded so "upgrade" return nothing, all firmwares were installed already too, dunno, will reinstall and check again, if same happen then dunno
2021-12-19 19:10:43 <minimal> MY-R: are you running from RAM or is this a SYS (disk) install?
2021-12-19 19:14:58 <minimal> MY-R: if you did a disk install then setup-disk (whether called directly or via setup-alpine) modified the mkinitfs.conf contents which would explain the contents you saw (missing "kms")
2021-12-19 19:22:28 <feuerteufel> https://www.ebay.de/itm/294205415537
2021-12-19 19:22:54 <feuerteufel> Ups sorry, wrong channel
2021-12-19 19:29:51 <MY-R> minimal: ok, I reinstall alpine maybe 4 times, repeated same few commands after reboot with different order and everytime all was working fine, couldnt repeat that hang anymore
2021-12-19 19:32:04 <minimal> MY-R: ok, good that its gone. Re: mkinitfs.conf, simply put the contents of that file on the ISO installer may or may not differ from the contents of that file (and therefore the initramfs contents) on a disk-based machine you create using setup-alpine/setup-disk.
2021-12-19 19:32:05 <MY-R> minimal: BUT I saw only one little difference once, when done first reinstall then noticed that during setup-alpine didnt discover any ext4 signature, it discover vfat signature, swap signature but NOT ext4 what done in next other reinstalls
2021-12-19 19:33:24 <MY-R> so once again on my first installation something went wrong with partitioning so after when regenerated initramfs in already running system it catch some error
2021-12-19 19:34:50 <MY-R> disk was "pure" when was running setup-alpine for the first time
2021-12-19 19:35:08 <MY-R> was fully "dd" from begining to the end (hdd)
2021-12-19 19:36:22 <minimal> "some error"? without details there's no way to debug
2021-12-19 19:37:21 <MY-R> I call it error but it isnt first time when Alpine acting so strange during partitioning, but is hard to debug it if it happens so randomly and now again, on different disk and different machine
2021-12-19 19:38:02 <MY-R> like said few times when using few parition tools in live iso alpine then at some point some partitions start dissapear in /dev/sd*
2021-12-19 19:39:45 <MY-R> but ye whatever, done so many reinstalls and cant repeat, only I could earse whole disk again and check if Im abble to reproduce it but not now, eh
2021-12-19 20:11:56 <spiralw> hey, the raspberry pi armhf image doesn't seem to work correctly with my Pi Zero. is there a list of compatible / tested Pi versions?
2021-12-19 20:20:16 <caskd> aren't the mdev scripts provided by alpine supposed to make by-id symlinks using /lib/mdev/persistent-storage?
2021-12-19 20:21:56 <minimal> caskd: I've seen them make by-uuid links but not by-id links
2021-12-19 20:22:08 <caskd> they do neither in my case
2021-12-19 20:22:13 <caskd> not sure if i gotta enable something
2021-12-19 20:22:43 <minimal> caskd: ah, I checked on a box with eudev installed...
2021-12-19 20:26:28 <caskd> nvm i just tested again, i was looking in the wrong place
2021-12-19 20:26:30 <caskd> my bad
2021-12-19 20:26:35 <caskd> works as expected
2021-12-19 20:31:49 <minimal> caskd: checked another *eudev* box and everything except by-id is populated (label, partuuid, path, uuid). I should have a VM without eudev somewhere here to have a look at...
2021-12-19 20:32:45 <caskd> minimal: i did confirm it works as expected by deleting /dev/disk and repopulating via mdev -s
2021-12-19 20:39:35 <minimal> caskd: not seeing it working here, only by-uuid is populated though there is an empty by-id directory. If I run "mdev -s" it then creates and populates by-partuuid but by-id is still empty
2021-12-19 20:39:45 <caskd> hm
2021-12-19 20:42:00 <minimal> caskd: I could have a misconfig - this is a VM that had eudev where I simply did a "apk del eudev", "rc-update add mdev boot", and rebooted
2021-12-19 20:48:03 <minimal> caskd: think I see why I might not be getting by-id entries, persistent-storage has an entry for dev/vd* (virtio-blk) to create /dev/disk/by-id entries but I'm using virtio-scsi (which creates /dev/sd* devices)
2021-12-19 20:48:22 <minimal> it doesn't appear to create by-id entries for /dev/sd* devices
2021-12-19 23:14:05 <caskd> Is someone also getting missing setuptools module during builds?
2021-12-19 23:14:14 <caskd> with the latest py3-setuptools pkg?
2021-12-19 23:16:34 <psykose> could you post an error message
2021-12-19 23:16:52 <caskd> Seems like it has to do with me mixing repos, looking into it
2021-12-19 23:16:54 <caskd> nothing upstream
2021-12-19 23:17:00 <caskd> my bad
2021-12-20 00:57:05 <ahills> I just upgraded a usbdisk-based install and on reboot it's mounted at /media/sdc1 instead of /media/usb, and /dev/usbdisk doesn't exist
2021-12-20 00:57:10 <ahills> can I get the old behavior back?
2021-12-20 01:06:14 <minimal> ahills: this was talked about last week - a couple of mdev scripts were replaced by a single persistent-storage script which does not include setting up /media/usb
2021-12-20 01:07:09 <ahills> minimal: thanks, which package/s has/had the scripts?
2021-12-20 01:08:21 <minimal> busybox-initscripts - however the old script are gone - replaced by persistent-storage which DOES NOT create the usbdisk softlink
2021-12-20 01:09:03 <ahills> presumably I could use the old ones as a hint for how to add the old behavior back myself?
2021-12-20 01:16:35 <minimal> ahills: I can't find the discussion in the irc logs (sometimes the logging "wedges" and misses logs for a time period)
2021-12-20 01:16:50 <ahills> minimal: do you have a time range? I can search mine
2021-12-20 01:17:58 <minimal> ahills: from memory ncopa didn't seem to want to reintroduce the /media/usbdisk software (I could be wrong in my recollection)
2021-12-20 01:18:06 <minimal> s/software/softlink/
2021-12-20 01:18:06 <alpine-meetbot> minimal meant to say: ahills: from memory ncopa didn't seem to want to reintroduce the /media/usbdisk softlink (I could be wrong in my recollection)
2021-12-20 01:18:32 <ahills> I'm willing to maintain my own solution if it's as trivial as adding an mdev script
2021-12-20 01:18:36 <minimal> ahills: I think it was either last week or the week before - not sure in in alpine-devel or alpine-linux.
2021-12-20 01:18:45 <ahills> ok, will search both, thanks for the tip
2021-12-20 01:24:40 <minimal> ahills: this is the commit that replaced usbdev_link with persistent-storage: https://git.alpinelinux.org/aports/commit/?id=7141b493a223fdb910ed90439cc044a9d5fe8c46
2021-12-20 01:24:54 <ahills> minimal: do you happen to know if the alpine_dev kernel command-line parameter is still used? I don't see it referenced anywhere in the mkinitfs package anymore except in a list of options
2021-12-20 01:25:00 <ahills> yep found it, thank you
2021-12-20 01:38:34 <ahills> wait... if /media/usb is not automatically mounted, then how does lbu work? only "floppy" and "usb" are allowed for LBU_MEDIA
2021-12-20 01:40:36 <ahills> I need something better than "sdXY" for the specification, can I give it a uuid instead maybe?
2021-12-20 01:51:23 <minimal> ahills: I see that setup-apkcache mentions /dev/usbdisk and /media/usb, setup-bootable mentioned /media/usb, setup-disk mentions /dev/usbdisk and /media/usb, and setup-lbu mentions /media/usbdisk, /media/usb and /dev/usbdisk
2021-12-20 01:51:43 <ahills> setup-lbu has special handling for it
2021-12-20 01:53:25 <ahills> I guess the CLI help is just wrong, more values than just "usb" and "floppy" are accepted
2021-12-20 01:53:39 <ahills> in my case I can't guarantee that usb will always be "sdc1" but that seems to be the only option
2021-12-20 02:25:26 <ahills> where can I find the mkinitfs config used for the initramfs included on the ISOs?
2021-12-20 02:39:41 <minimal> ahills: in aports/scripts/ in the mkimg.base.sh, mking.standard.sh etc
2021-12-20 02:40:20 <minimal> look for initfs_features
2021-12-20 02:40:25 <ahills> thanks
2021-12-20 02:41:20 <minimal> ahills: but for what's used when you install to disk ("sys" mode), setup-disk defines the resultant features list
2021-12-20 02:41:53 <ahills> yeah, this isn't a problem for my "sys" installs, just the "data"/usb-only ones
2021-12-20 02:42:12 <ahills> where the number of local sd* disks can change between boots
2021-12-20 02:42:27 <minimal> ahills: you should raise an Issue for it
2021-12-20 02:42:29 <ahills> but only one of them will be usb
2021-12-20 02:46:45 <ahills> minimal: maybe the missing irc discussion was on this issue: https://gitlab.alpinelinux.org/alpine/aports/-/issues/13254
2021-12-20 02:46:57 <ahills> > This was sort of expected. I am not sure that we want to keep the /dev/usbdisk link.
2021-12-20 02:48:25 <minimal> ahills: ah you're right, it was that issue, not on IRC. I did search the Issue list but obviously didn't use the right "magic" search string :-)
2021-12-20 02:48:33 <ahills> yeah, it took me a few tries
2021-12-20 02:48:52 <ahills> thanks for all the help, I'll discuss on the issue (if needed)
2021-12-20 02:49:49 <minimal> ahills: you should put a comment on that issue about the "data" mode type issue - i.e. where a HDD/SSD is /dev/sda and the bootable USB stick is /dev/sdb (but, as you said, change change name depending on the discovered order of devices, which can change depending on which usb port you plug it into)
2021-12-20 02:50:04 <ahills> yes, I will
2021-12-20 02:50:33 <minimal> you see someone else has mentioned the setup-apkcache-related naming issue
2021-12-20 02:51:36 <minimal> that was me in the issue flagging the relevant commit that caused the impact :-)
2021-12-20 12:09:18 <frag> i remember there was some discussion about this here earlier, was there some solution to firefox ads/monitoring? https://reclaimthenet.org/firefox-ads-address-bar/
2021-12-20 12:10:05 <psykose> a lot of discussion yes
2021-12-20 12:11:36 <frag> hmm..
2021-12-20 13:11:25 <frag> someone recommended palemoon on another chan, anyone using that?
2021-12-20 13:12:23 <frag> probably easy to package too..
2021-12-20 13:13:09 <psykose> is that the one that is 10% browser and 90% licencing department
2021-12-20 13:13:42 <frag> hehe, no idea, i used it many years ago ... http://linux.palemoon.org/download/mainline/
2021-12-20 13:13:56 <frag> seems it should work even without installing (i cant test right now)
2021-12-20 13:14:07 <psykose> regardless i don't think anyone is going to be packaging any web browsers as it's too much maintenance
2021-12-20 13:15:58 <frag> alpine needs a working browser eh
2021-12-20 13:16:08 <frag> are ppl happy with firefox?
2021-12-20 13:16:11 <psykose> it works
2021-12-20 14:26:50 <Hello71> it remains the least worst option
2021-12-20 14:35:16 <lopid> i am happy with it
2021-12-20 18:09:17 <q[m]12> Hi all! I'm trying to set up alpine according to this guide[0], but I had a problem mounting the root volume on boot. It would drop to a recovery shell, from where I could manually activate and mount the root LV, but apparently it failed at automatically doing that. It's been a few days, I don't remember exactly how it went. Anyway, I know it isn't required anymore just to get an encrypted rootfs, but I also want it to be BTRFS, but still
2021-12-20 18:09:17 <q[m]12> if anyone can recommend another way of achieving that, I will look into it. Otherwise I will try again and possibly come back with a more specific question/problem description. I'd much appreciate any pointers, thanks!
2021-12-20 18:09:17 <q[m]12> [0] https://wiki.alpinelinux.org/wiki/LVM_on_LUKS
2021-12-20 18:16:34 <minimal> q[m]12: how is it failing? does it ask for the passphrase to unlock and fail after that or fail before then?
2021-12-20 18:18:13 <q[m]12> Yes, it would ask for the password and after unlocking fail to find the root volume
2021-12-20 18:28:43 <minimal> q[m]12: using grub? on BIOS or UEFI? if Grub then what is the contents of your /etc/default/grub file?
2021-12-20 18:30:52 <q[m]12> I get:... (full message at https://matrix.org/_matrix/media/r0/download/matrix.org/oMqWFISeSOOgcuWhVoznOqXs)
2021-12-20 18:33:56 <minimal> q[m]12: ok, but what cmdline options is your bootloader (Grub?) passing to the kernel/initramfs?
2021-12-20 18:34:01 <q[m]12> it is using GRUB and UEFI, the contents of /etc/default/grub are as follows... (full message at https://matrix.org/_matrix/media/r0/download/matrix.org/XrSVWcVMnVeaahdLIvdRDVVk)
2021-12-20 18:35:22 <minimal> q[m]12: there are things missing from your /etc/default/grub file, that guide clear mentions setting GRUB_CMDLINE_LINUX_DEFAULT
2021-12-20 18:37:09 <q[m]12> * it is using GRUB and UEFI, the contents of /etc/default/grub are as follows... (full message at https://matrix.org/_matrix/media/r0/download/matrix.org/cDpaeBuhfZYvDEtlYxaONtPX)
2021-12-20 18:37:34 <q[m]12> yes, just edited, I accidentally sent to early
2021-12-20 18:40:33 <q[m]12> I'm doing this inside a VM, not sure if that matters
2021-12-20 18:53:32 <minimal> q[m]12: are you using LUKS v1 or v2?
2021-12-20 18:56:04 <minimal> q[m]12: also in /etc/default/grub does the GRUB_CMDLINE_LINUX_DEFAULT value *really* contain "cryptroot=UUID=$UUID"? the bit "$UUID" should actually be a UUID value
2021-12-20 19:00:51 <q[m]12> I am using LUKS v1, yes the actual UUID is in there, I just didn't want to type as much
2021-12-20 19:03:36 <minimal> q[m]12: ok, so that will create /dev/mapper/lvmcrypt for LUKS once unlocked. What's in your /etc/fstab?
2021-12-20 19:11:42 <q[m]12> #cat /etc/fstab... (full message at https://matrix.org/_matrix/media/r0/download/matrix.org/vtmvIBFHYhllfDysHFjRzmsY)
2021-12-20 19:13:40 <q[m]12> I just noticed the UUID in fstab is different from the one in /etc/default/grub, but that is to be expected, because one refers to the physical disk, the other one to the logical volume, right?
2021-12-20 19:19:05 <minimal> q[m]12: yes the cryptroot UUID is for luks, whereas the fstab one is for btrfs inside LVM inside LUKS
2021-12-20 19:19:38 <minimal> blkid will show you these
2021-12-20 19:20:48 <minimal> for fstab I prefer to refer to the LVs using "/dev/mapper/vg0-root" type names as they are clearer than UUID refs
2021-12-20 19:25:46 <q[m]12> Yes, those are actually more helpful
2021-12-20 19:26:41 <minimal> q[m]12: and you have checked that the /boot/grub/grub.cfg file reflects/uses the settings from /etc/default/grub?
2021-12-20 19:43:33 <q[m]12> hmm I am not quite sure; the kernel cmdline is
2021-12-20 19:43:33 <q[m]12> linux /vmlinuz-lts root=/dev/mapper/vg0-root ro modules=sd-mod,usb-storage,btrfs quiet rootfstype=btrfs
2021-12-20 19:43:33 <q[m]12> but all the volumes should be active by the time GRUB is running right?
2021-12-20 19:46:11 <minimal> q[m]12: wrong, GRUB is the bootloader, its loads the kernel and the initramfs (which unlocks LUKS and handles LVM, and mounts root)
2021-12-20 19:46:35 <minimal> so Grub happens BEFORE kernel runs and all that other stuff
2021-12-20 19:47:31 <minimal> I'd expect to see the cryptdm, cryptkey entries etc in your cmdline. If they are missing then was grub-mkconfig run after /etc/default/grub was edited?
2021-12-20 19:47:37 <q[m]12> Ok, but the grub config is inside vg0-boot inside lvm inside luks
2021-12-20 19:49:32 <q[m]12> yes, that confused me too, but I am sure I ran grub-mkconfig; anyway I should be able to chroot and re-run it from the recovery shell I guess
2021-12-20 19:50:39 <minimal> right, but you set GRUB_ENABLE_CRYPTODISK=y which means that Grub unlocks LUKS and deals with LVM itself *just* to get to the grub.cfg file, once the kernel and initramfs then load and run the initramfs seperately has to unlock LUKS, deal with LVM etc - that's why if you do *not* specify "cryptkey" (and set up a keyfile) you would be prompted for the LUKS passphrase twice - once by Grub and then by initramfs's init
2021-12-20 20:09:09 <minimal> q[m]12: I have scripted all this sort of stuff to make things easier :-)
2021-12-20 20:11:27 <q[m]12> <minimal> "right, but you set GRUB_ENABLE_C..." <- Ah ok, my understanding of GRUB and the whole boot process is still rather limited, but that clarifies it. I chrooted and ran grub-mkconfig. It boots now, although it prompts for the password twice, I did create the keyfile (with the proper name), well I'll see, I can surely figure it out, thanks so much for your help!
2021-12-20 20:11:27 <q[m]12> My ultimate goal is to automate this whole process (and then use ansible to automate the configuration too)  so I can setup alpine on my laptop within minutes
2021-12-20 20:13:29 <q[m]12> minimal: Ah yes... nice my script does not work properly yet, for example for some reason my sed commands wouldn't actually do anything
2021-12-20 20:13:40 <minimal> q[m]12: sort-of been there, done that - script for creating Alpine server disk images for VMs, physical, cloud etc
2021-12-20 20:14:50 <minimal> q[m]12: https://github.com/dermotbradley/create-alpine-disk-image
2021-12-21 03:21:11 <Saijin_Naib[m]> Hello71: Would you mind sharing your ownership/permissions for /tmp/.wine-1000/ ? My wine will not run anything because it "must not be accessible by other users"
2021-12-21 04:05:59 <Hello71> you need to create the directory first
2021-12-21 04:06:27 <Hello71> or just put it in your home directory like you're supposed to
2021-12-21 05:19:43 <Saijin_Naib[m]> Directory exists in tmp, and I have the .wine in my local user profile
2021-12-21 07:33:57 <jhjacobs81> alpine would be rather good to run a puppet master on..
2021-12-21 08:56:51 <blt[m]> wondering if it is expected to do a `service networking restart` after a setup-interface ? I noticed that it is not done at the end of the script even if the parameters are passed when I do a service restart, not a big deal just a curiosity
2021-12-21 08:59:45 <ktprograms> blt[m]: Pass -r to setup-interfaces: https://gitlab.alpinelinux.org/alpine/alpine-conf/-/blob/master/setup-interfaces.in#L422 (that's what setup-alpine does)
2021-12-21 09:00:25 <blt[m]> ktprograms: ha thanks ! it is as simple, I missed this parameter
2021-12-21 10:14:35 <circ-user-fDpxv> Hola. Can someone help with segfault:  httpd[24876]: segfault at 1fe0 ip 00007f61ff8b0e7e sp 00007ffeac5971e8 error 4 in ld-musl-x86_64.so.1 ?
2021-12-21 11:18:04 <ikke> circ-user-fDpxv: hello
2021-12-21 11:18:46 <ikke> can you add rc_ulimit='-c unlimited' to /etc/conf.d/<servicename>?
2021-12-21 11:21:33 <circ-user-fDpxv> ikke: I think yes. But the problem is floating. I thought maybe there be a known issue and a fix in newer versions.
2021-12-21 11:24:05 <ikke> Not aware of any existing issues
2021-12-21 13:06:21 <blt[m]> is there a reason why lbu is not keeping owner on some files/directories ? I tried to play with mariadb which is failing to start and I quickly saw that mysql owner diseapeard from my db datadir (the group was style good after a reboot) . manualy chown fix the issue, but it was "60" after the reboot, wondering if there is a way to keep user/uid with lbu?
2021-12-21 14:20:19 <ikke> There's an issue open about it
2021-12-21 14:21:18 <Hello71> Saijin_Naib[m]: maybe you have lost sticky bit on /tmp
2021-12-21 14:25:49 <Saijin_Naib[m]> Hello71: thanks! I'll look into what this is/means. 
2021-12-21 14:41:21 <Saijin_Naib[m]> Awesome! You saved me with that. Had to delete and recreate my tmp with the proper permissions (1777), and now it runs WINE
2021-12-21 15:10:54 <blt[m]> ikke: thanks I didn't know that do you have the issue I can maybe add a comment?
2021-12-21 16:52:59 <ikke> blt[m]: https://gitlab.alpinelinux.org/alpine/apk-tools/-/issues/10743
2021-12-21 16:53:26 <ikke> Not sure if that's the same issue
2021-12-21 18:48:36 <blt[m]> ikke: thanks, it is close but no exactly the same but I can jump into it as it may be related
2021-12-21 19:39:12 <asdflkj> apk -U upgrade gave an error about no such package as libtls, required by sacc. does that mean it won't update my other packages?
2021-12-21 19:41:00 <asdflkj> (sacc and 2 other packages in world are @testing, everything else's still on 3.14/pmOS 21.06 sp4)
2021-12-21 19:45:56 <ikke> asdflkj: likely. What is the exact error message?
2021-12-21 19:51:10 <asdflkj> ikke:  https://termbin.com/wy8gw
2021-12-21 19:57:11 <ikke> sacc needs libraries that are not preset in 3.14
2021-12-21 19:57:23 <ikke> mixing edge with a stable release like this is not supported
2021-12-21 20:00:20 <asdflkj> so is uninstalling sacc the only way to update anything else?
2021-12-21 20:00:58 <ikke> yes
2021-12-21 20:09:47 <asdflkj> hmm, wonder how I got sacc installed in the first place
2021-12-21 20:10:14 <asdflkj> maybe it didn't depend on libtls before, it shouldn't actually need it
2021-12-22 02:42:55 <Levi_> Hi All, just trying to install virtualbox guest additions. Wikiw says to install virtualbox-guest-modules-virt. repo shows that most recent version is for branch v3.11. is there any way to install this in a v3.15 instance ?
2021-12-22 02:46:15 <raven523> it looks like the linux-virt package now includes the vboxvideo, vboxguest, and vboxsf modules
2021-12-22 02:46:45 <raven523> also linux-lts and I assume the other kernel packages
2021-12-22 02:48:21 <Levi_> ok great, thanks very much.
2021-12-22 02:55:46 <minimal> Levi_: there are packages virtualbox-guest-additions and virtualbox-guest-additions-x11
2021-12-22 02:57:05 <minimal> these provide tools like VBoxService
2021-12-22 03:34:46 <zcrayfish> Levi_, raven523, minimal: any objections to me copying a log of y'alls conversation above verbatim to the wiki talk page for later use in fixing the article?
2021-12-22 03:35:47 <raven523> I'm fine with that but you don't need to copy iot verbatim
2021-12-22 03:36:09 <minimal> zcrayfish: no problem. I use Virtualbox frequently for running Alpine server (i.e. non-graphical) VMs so can help with any clarifications
2021-12-22 03:36:46 <zcrayfish> raven523: it wouldn't be verbatim in the final article, only on the talk page.
2021-12-22 04:11:00 <bleb> is there a nice and easy way to add a package from a previous version of alpine?
2021-12-22 04:11:08 <bleb> I want to use nvi
2021-12-22 04:15:13 <zcrayfish> what's the last version it was in?
2021-12-22 04:16:02 <psykose> it was in testing
2021-12-22 04:16:16 <psykose> it was removed in da085140b8
2021-12-22 04:16:42 <psykose> because it depends on oracle berkeleydb
2021-12-22 04:16:43 <psykose> or something
2021-12-22 04:18:21 <zcrayfish> I thought it shipped with its own db. 
2021-12-22 04:18:31 <zcrayfish> but anyway
2021-12-22 04:23:13 <psykose> funnily berkeleydb is still in the repos
2021-12-22 04:24:11 <raven523> that removal happened in february
2021-12-22 04:35:22 <bleb> it was a rash move which ultimately puts more power into the hands of oracle, allowing them to dictate the security status of versions of db which they don't control
2021-12-22 04:35:56 <bleb> but I mostly just want to use nvi, haven't had the time to set up my own repo
2021-12-22 04:36:29 <bleb> no easy way?
2021-12-22 04:38:15 <zcrayfish> Not that I can think of for that one.
2021-12-22 04:38:35 <bleb> I figured the precompiled packages would be sitting on a server somewhere
2021-12-22 04:38:52 <bleb> maybe alpine doesn't keep them though
2021-12-22 04:39:46 <zcrayfish> If I'd had seen one, I would have suggested repository pinning. :/
2021-12-22 13:50:46 <Srket> I had a problem some weeks ago where my system would crash, and then i would be greeted by "mounting /dev/sda3 on /sysroot failed : No error information".
2021-12-22 13:50:46 <Srket> People here helped me to get it working by running fsck on the disk, and then i could use it again. But today, when i came home i saw my server in a crashed state again, and this same error is presented on boot. I'll run fsck again, but it would be better if the system wouldn't crash like this.
2021-12-22 13:52:04 <nero> crashes are normally kernel issues
2021-12-22 13:52:47 <lopid> perhaps your drive is bad
2021-12-22 13:52:59 <Srket> Drive extensive test, no errors reported.
2021-12-22 13:53:02 <nero> then that should be in dmesg
2021-12-22 13:53:15 <nero> Srket: what kind of crash?
2021-12-22 13:53:43 <Srket> System becomes non-bootable, /dev/sda3 breaks and i need to run fsck by hand
2021-12-22 13:53:59 <Srket> Throws me to recovery shell.
2021-12-22 13:54:29 <psykose> what is in the logs
2021-12-22 13:54:40 <nero> Srket: that is not a crash, that is failing to boot
2021-12-22 13:55:03 <Srket> System was running, no manual reboots or shotdowns, it just restarts all of a sudden and this happens.
2021-12-22 13:56:18 <Srket> "mount : Mounting /dev/sda3 on /sysroot failed: No error information.
2021-12-22 13:56:19 <Srket> Mounting root failed.
2021-12-22 13:56:19 <Srket> initramfs emergency recovery shell launched, Type exit to continue boot" is what i get
2021-12-22 13:56:32 <psykose> the logs from when it restarts
2021-12-22 13:56:36 <nero> yeah because the FS was not unmounted properly while in use
2021-12-22 13:56:52 <nero> the ext issues are a symptom of what is probablya kernel panic
2021-12-22 13:57:11 <nero> you need to get hold of what is causing the kernel panic (and the reboot is probably a result from that)
2021-12-22 13:57:30 <qman> Right, the system crashed for some reason and left the filesystem in an inconsistent state
2021-12-22 13:57:44 <Srket> Like i said, no electricity loss or nothing, machine was running for a long time. I don't know what could crash in it.
2021-12-22 13:57:53 <psykose> it would perhaps be in the logs
2021-12-22 13:57:56 <qman> Which could still be a disk issue, but could be a number of other things too
2021-12-22 13:58:04 <nero> disable auto-reboot in case of kernel panics, wait until it crashes again, then check the console output
2021-12-22 13:58:11 <Srket> Disk is fine, checked and there is no problem.
2021-12-22 13:58:17 <Srket> How do i disable that?
2021-12-22 13:58:53 <qman> SMART tests passing are not a guarantee the disk is good
2021-12-22 13:59:08 <Srket> I did tests on it, not enough?
2021-12-22 13:59:50 <nero> stop talking about the disk, a sudden crash somewhat guaranteres that a FS is in a state that requires fsck'ing
2021-12-22 13:59:51 <qman> It's enough to not directly suspect the disk right now, but it doesn't completely rule it out
2021-12-22 14:00:20 <qman> You need more data on the actual crash condition to point at the issue
2021-12-22 14:00:58 <nero> Srket: /etc/sysctl.d/00-alpine.conf has panic=120
2021-12-22 14:02:32 <nero> `sysctl kernel.panic=0` as root should disable it for the current boot
2021-12-22 14:03:27 <Srket> Will do after the fsck ends.
2021-12-22 14:04:40 <qman> Yep, if it's a kernel panic, that will give you some error data, if it still reboots then it's either hardware resetting it or an external cause
2021-12-22 14:04:46 <nero> if the server has a serial console, you can also have some screen/com/logger running on there and record the panic output
2021-12-22 14:20:18 <Nulo> Hi, any clue on why most packages don't have a -dbg counterpart? For example: pipewire and derivates, iwd...
2021-12-22 14:22:50 <psykose> they take a lot of space, and are added on demand
2021-12-22 16:54:23 <armin> Checking for stray cats under /usr/share/man...
2021-12-22 16:54:23 <armin> Checking for stray cats under /var/cache/man...
2021-12-22 16:54:34 <armin> checking for STRAY CATS!? :)
2021-12-22 16:54:40 <armin> that's a funny one.
2021-12-22 16:54:54 <armin> :)
2021-12-22 16:56:38 <psykose> a cat is a specific format like man, a stray cat is a cat without a corresponding man
2021-12-22 23:49:39 <villechae> Hello everyone.
2021-12-23 06:34:39 <villechae> Hello Everyone!
2021-12-23 11:17:24 <jhjacobs81> hello ;-)
2021-12-23 12:33:28 <gjabell> has anyone gotten mpris-proxy to work with pipewire audio devices?
2021-12-23 12:35:36 <psykose> hm, it used to work for me
2021-12-23 12:35:42 <psykose> now it seems to kind of not
2021-12-23 12:41:08 <psykose> only volume up/down does
2021-12-23 12:41:15 <gjabell> yeah same
2021-12-23 12:41:25 <psykose> but i remember it working maybe 3 pipewire versions ago
2021-12-23 12:41:33 <gjabell> hmm odd
2021-12-23 12:41:49 <psykose> let me try a downgraded one
2021-12-23 12:41:51 <gjabell> maybe one of the options changed and needs to be reenabled
2021-12-23 12:50:05 <psykose> 0.0.36 works
2021-12-23 12:50:15 <psykose> 0.3.36*
2021-12-23 12:50:21 <psykose> .39 doesnt as well
2021-12-23 12:50:59 <psykose> i guess i have to do some bisecting, but effooooort
2021-12-23 12:51:18 <gjabell> yeah that doesn't seem worth it haha
2021-12-23 12:51:30 <gjabell> thanks for checking, glad to know it's not just a misconfig on my part
2021-12-23 12:55:32 <psykose> still gonna try
2021-12-23 12:56:27 <psykose> thankfully it takes like 2 minutes to build
2021-12-23 13:06:11 <MY-R> dont you have have fram drops in youtube videos in firefox when using pipewire?
2021-12-23 13:16:42 <psykose> gjabell: 4b831021fbfa31b51d43d292582b00994143543e is the bad commit
2021-12-23 13:16:53 <psykose> and the commit itself seems to actually be related to bluetooth
2021-12-23 13:17:07 <psykose> bluez5: add a dummy AVRCP player as a workaround :p
2021-12-23 13:19:04 <gjabell> oh lol
2021-12-23 13:19:05 <psykose> it does add a config option to disable it
2021-12-23 13:19:07 <gjabell> nice find!
2021-12-23 13:19:16 <psykose> bluez5.dummy-avrcp-player = true
2021-12-23 13:19:26 <psykose> inside media-session.d, i think wireplumber has an equivalent
2021-12-23 13:19:41 <psykose> let me test that...
2021-12-23 13:20:54 <gjabell> oh yeah there's even a comment in the conf file for that version...
2021-12-23 13:21:05 <gjabell> I feel like that should really be disabled by default though
2021-12-23 13:22:26 <gjabell> seems to work for me on 0.3.40
2021-12-23 13:23:22 <psykose> yep, setting it to false seems to fix it for me
2021-12-23 13:23:57 <psykose> thanks, now i have working mpris too
2021-12-23 13:24:21 <gjabell> 🎉
2021-12-23 13:24:25 <gjabell> thanks for finding the fix
2021-12-23 13:26:02 <psykose> the commit itself links to an issue on pipewire, i guess they have a good reason to have it as on given all the devices broken otherwise
2021-12-23 13:26:03 <psykose> or something
2021-12-23 15:14:11 <jvoisin> it's such a pleasure to contribute to packages in alpine linux compared to debian
2021-12-23 15:44:33 <minimal> I have rootfs on LVM. During shutdown/reboot the /etc/init.d/lvm script attempts to deactive LVM VGs - but of course fails as the VG that rootfs is on is in use. Trying to figure out how this is supposed to be handled (without any errors)
2021-12-23 15:45:38 <minimal> later in the shutdown sequence /etc/init.d/remount-ro runs and correctly sets rootfs ro, however the VG would still be active at this point
2021-12-23 15:49:05 <Hello71> isn't the kernel supposed to clean it up
2021-12-23 15:49:32 <Hello71> remount read-only, sync, reboot
2021-12-23 15:51:17 <minimal> Hello71: does the kernel have support for deactivating VGs though? I didn't think so
2021-12-23 15:51:33 <Nulo> Is there a way to make apk _not_ delete the current kernel's /lib/modules?
2021-12-23 15:51:44 <Hello71> well you have to think about what "deactivating VGs" actually means
2021-12-23 15:52:05 <minimal> even if it does the /etc/init.d/lvm script should not generate an error - it should handle "errors" with trying to deactive the VG for the rootfs
2021-12-23 15:52:51 <Hello71> really it means deleting some device mapper targets
2021-12-23 15:53:29 <minimal> Hello71: yes I have not yet dug into the vgchange code but assume there may be some VG metadata that needs to be written to deactivate "cleanly"
2021-12-23 15:55:49 <Hello71> that sounds unnecessary. the point of dirty markers on filesystems is to note the potential need to run fsck; normally lvm doesn't write to metadata, so no fsck should be needed in case of unclean shutdown
2021-12-23 15:56:37 <Hello71> filesystems need to write to their metadata in the course of normal operation, but with the exception of thin provisioning, i think lvm blocks are fixed in place
2021-12-23 15:57:07 <minimal> Hello71: I'm not talking about fsck - the init.d mount-ro correctly switches the ext4 fs to ro, I'm talking about VG metadata/attributes
2021-12-23 15:57:58 <Hello71> yes
2021-12-23 15:58:10 <Hello71> but if there is no need for fsck, there is no need for dirty marker
2021-12-23 15:59:40 <minimal> dirty marker for ext4 you mean. I'm talking about PV/VG/LV, not filesystems.
2021-12-23 16:02:09 <Hello71> 
ACTION bails

2021-12-23 17:43:47 <jhjacobs81> oh.. well.. heh.. see!
2021-12-23 17:58:35 <jhjacobs81> question: i would assume there's a docker package available for Alpine, but would docker-compose also work?
2021-12-23 17:59:01 <jhjacobs81> ls -a
2021-12-23 17:59:09 <jhjacobs81> whoops
2021-12-23 18:14:58 <jhjacobs81> oh.. well.. no docker for Alpine :|
2021-12-23 18:20:48 <jhjacobs81> ohh.. thankfully for community repo's
2021-12-23 18:22:30 <jhjacobs81> time to kill everything! bbl!
2021-12-23 18:57:23 <jhjacobs81> hmz
2021-12-24 01:04:52 <Naib> mmm the community apk have bumped the need of py310 
2021-12-24 03:28:43 <amberlee[m]> is there a way to get the kernel to disable or minimize the usage of page cache (guest vm setting, guest is touching memory and hypervisor process grows in memory usage) vm.dirty* seems to not be working.
2021-12-24 07:07:52 <ktprograms> Alpine doesn't package android udev rules, right? (I'm talking about something like https://github.com/M0Rf30/android-udev-rules) So how are users supposed to use (for example) adb and scrcpy?
2021-12-24 13:39:46 <Hello71> amberlee[m]: that's not page cache, that's, like, virtual memory
2021-12-24 13:40:06 <Hello71> if you want qemu to use less ram, then set less ram for the vm
2021-12-24 13:42:29 <Hello71> ktprograms: i guess you can run it as root, or chown device manually, or write mdev rule, or install udev and rules manually
2021-12-24 13:44:20 <ktprograms> Hello71: I think what amberlee[m] means is that the standard disk caching method of the kernel doesn't work well with VMs because the VM will use as much memory as it is allocated for caching, which will use up a lot of host memory. For example, if I make a VM with 8GB of memory (which I might need for memory-heavy compiles), if I run a disk-heavy operation it will use a lot of RAM and I have to 
2021-12-24 13:44:26 <ktprograms> manually drop_caches.
2021-12-24 13:44:54 <ktprograms> Hello71: I see. It already works with me manually adding the rules, but I was just surprised it wasn't packaged yet.
2021-12-24 13:45:21 <Hello71> well it will use 8GB of ram, i don't see the issue
2021-12-24 13:46:20 <ktprograms> Hello71: It's probably a problem also because QEMU can't seem to free host memory even after the guest memory usage has reduced.
2021-12-24 13:46:37 <Hello71> i think possibly mdev should have plugdev rule
2021-12-24 13:46:55 <Hello71> that is the traditional, not-very-secure solution
2021-12-24 13:48:29 <Hello71> well for one, the guest is probably still using it as cache. two, because qemu doesn't actually allocate a bit of ram at a time; it allocates the full 8 GB, but through the magic of virtual memory, the memory is not used until the guest touches it
2021-12-24 13:49:04 <Hello71> the "solution" for some time has been memory ballooning
2021-12-24 14:18:41 <ktprograms> Hello71: I see, thanks
2021-12-24 17:37:04 <amberlee[m]> Once the guest touches any ram the hypervisor can no longer reclaim it for the host without active measures such as virtio-balloon. When using virtiofs, guest fs access is done directly on shared memory with the host making guest caching useless. But it seems impossible to disable it. /sys/ drop_caches every 0.5 seconds seems to work but it's janky.
2021-12-24 17:52:20 <jhjacobs81> has anyone succesfully configured Caddy webserver on Alpine?
2021-12-24 18:01:10 <argsh> hi
2021-12-24 18:02:23 <zico[m]1> jhjacobs81: I've configured it a couple times, what issue are you running into?
2021-12-24 18:12:19 <jhjacobs81> i cant get it to run on port 443
2021-12-24 18:12:49 <jhjacobs81> http	server is listening only on the HTTPS port but has no TLS connection policies; adding one to enable TLS	{"server_name": "srv0", "https_port": 443}
2021-12-24 18:13:09 <jhjacobs81> i havw never seen that error before.. nd the same Caddyfile works on Debian and Ubuntu..
2021-12-24 18:21:49 <zico[m]1> How are you starting/running Caddy and what error do you get?
2021-12-24 18:22:05 <jhjacobs81> the above error :) its started automaticly
2021-12-24 18:23:38 <jhjacobs81> the Caddyfile has a simple <url> { reverse_proxy <internal url> }
2021-12-24 18:23:45 <bdju> sounds more like a warning and like it is running to me
2021-12-24 18:24:30 <zico[m]1> yeah is it showing as bound to the port with netstat?
2021-12-24 18:24:38 <jhjacobs81> i thought so too :)
2021-12-24 18:24:45 <jhjacobs81> netstat only shows port 80
2021-12-24 18:25:49 <jhjacobs81> if i replace <url> with http://<url> then i can open the webapp
2021-12-24 18:26:07 <jhjacobs81> i can confirm that its no firewall or router issue
2021-12-24 18:26:24 <jhjacobs81> so it has to do with caddy.. somehow..
2021-12-24 18:26:52 <jhjacobs81> googling that message doesnt help either :(
2021-12-24 18:28:16 <jhjacobs81> it even has the certificates for the url..
2021-12-24 18:28:24 <jhjacobs81> and yet, its not running on port 443
2021-12-24 18:28:37 <zico[m]1> weird
2021-12-24 18:29:38 <ikke> https://caddy.community/t/port-and-certificate-errors-with-https-file-server-on-localhost/8260
2021-12-24 18:30:50 <ikke> https://github.com/caddyserver/caddy/issues/3635
2021-12-24 18:30:55 <jhjacobs81> thanks, but i already found that link, and it doesnt solve my problem :)
2021-12-24 18:36:42 <zico[m]1> when you say caddy is started automatically, do you mean with the caddy initd script? (openrc-run)
2021-12-24 18:36:52 <jhjacobs81> yupyup
2021-12-24 18:37:10 <jhjacobs81> never mind... i think the problem is in the router somehow
2021-12-24 18:37:30 <jhjacobs81> when i add the alpine host to my /etc/hosts, i can get https to work
2021-12-24 18:38:53 <bdju> nice
2021-12-24 18:39:04 <jhjacobs81> not really :D
2021-12-24 18:39:21 <jhjacobs81> but at least its not a problem with caddy on alpine :D
2021-12-24 18:52:37 <jhjacobs81> now im even more compelled to build a router based on alpine
2021-12-24 18:52:44 <jhjacobs81> it just needs a good web gui
2021-12-24 23:46:52 <tobtobxx> Hey there, I have a question. I'm trying to setup a RbPi as a IPv6-only router and with diskless mode. Thus on boot, it only connects to an IPv6 network. However ntpd fails somewhat. I've found out that _some_ domains of the NTP pool project support IPv6. So I've put them into the /etc/ntp.conf file. Running ntp -N -d (debug) works just fine, but the rc-service unit does not set the time on boot. This is due to the unit specifying its peer
2021-12-24 23:48:43 <tobtobxx> * I meant running ntpd -n -d (debug). ie. don't fork; print debug. The default unit file has the parameters ntpd -N -p <some-peer>. -N means high priority, -p sets the peer explicitly. I removed the -p flag in the init.d/ntpd file.
2021-12-24 23:50:44 <tobtobxx> Oh and btw, ntpd service is added to default.
2021-12-24 23:50:54 <tobtobxx> ... not boot.
2021-12-24 23:53:45 <Nulo> warning: the debug information found in "/usr/lib/debug//lib/ld-musl-x86_64.so.1.debug" does not match "/lib/ld-musl-x86_64.so.1" (CRC mismatch).
2021-12-24 23:53:47 <Nulo> Any clue?
2021-12-24 23:58:52 <tobtobxx> Nulo: Context?
2021-12-24 23:59:31 <Nulo> tobtobxx: trying to get a backtrace of a coredump using "gdb <program> <coredump>" and then running "bt full". Some of the backtrace comes from musl.
2021-12-25 00:00:07 <tobtobxx> On the same system the dump originated or on another?
2021-12-25 00:00:58 <Nulo> same system.
2021-12-25 00:04:36 <tobtobxx> apk update && apk upgrade?
2021-12-25 00:04:52 <tobtobxx> Maybe musl-dbg and musl have different versions...
2021-12-25 00:05:00 <tobtobxx> Running edge?
2021-12-25 00:05:18 <minimal> tobtobxx: I meant to raise a MR a while to change ntpd to use a ntp.conf inside of the hardcoded "-p"
2021-12-25 00:06:42 <tobtobxx> minimal: How does ntpd act if there's no /etc/ntp.conf? Just query {0,1,2,3}.pool.ntp.org?
2021-12-25 00:07:34 <tobtobxx> minimal: but changing a init.d file doesn't have any effect when running in diskless mode, does it?
2021-12-25 00:07:57 <minimal> tobtobxx: assuming this is the Busybox ntp, then the "-p pool.ntp.org" in the init.d is used. Only if no "-p" is specified does it fallback to read ntp.conf
2021-12-25 00:08:33 <minimal> tobtobxx: yeah for diskless you'd have to use lbu to write the modified ntp.conf file
2021-12-25 00:09:24 <tobtobxx> I did, but ps aux | grep ntpd show that ntpd was run with -p pool.ntp.org, even though I edited /etc/init.d/ntpd...
2021-12-25 00:12:12 <minimal> ah, yeah the init.d file is unpacked from the busybox-initscripts package at boot. I was talking about intending to push a code change (aka MR) to the busybox-initscripts package to remote "-p" from the supplied /etc/init.d/ntpd file
2021-12-25 00:14:48 <tobtobxx> Ah, ok. Thank you for filing the MR, whenever you come around to it.
2021-12-25 00:15:25 <tobtobxx> But is there a way to run ntpd automatically after startup but not through the init.d scripts in diskless mode?
2021-12-25 00:19:47 <minimal> you could use rc.local I guess - but if the ntpd init.d is enabled by default it would still run via that (also)
2021-12-25 00:20:21 <minimal> I'm doing a bit of general IPv6-only testing currently myself.
2021-12-25 00:20:39 <minimal> though not with diskless mode
2021-12-25 00:26:09 <tobtobxx> I think running ntpd twice won't hurt terribly.
2021-12-25 00:26:25 <tobtobxx> btw (you may have figured that out already): only the 2.*.pool.ntp.org servers support IPv6. Not the 0.*, 1.* or 3.*
2021-12-25 00:37:50 <tobtobxx> ... and the local.d trick worked indeed. Thanks!
2021-12-25 00:38:59 <minimal> tobtobxx: not a good idea to run 2 ntpd processes - they might interfere with each other
2021-12-25 00:46:00 <tobtobxx> Ok, I'll run rc-service ntpd stop && ntpd -N  ;)
2021-12-25 00:48:11 <tobtobxx> Nevermind, rc-update del works across lbu commit && reboot.
2021-12-25 00:51:40 <minimal> tobtobxx: what are you using for IPv6 address configuration and which which "client"? DHCPv6? SLAAC? static?
2021-12-25 00:51:51 <minimal> s/which w/via w/
2021-12-25 00:51:51 <alpine-meetbot> minimal meant to say: tobtobxx: what are you using for IPv6 address configuration and via which "client"? DHCPv6? SLAAC? static?
2021-12-25 00:52:20 <tobtobxx> static
2021-12-25 00:53:12 <tobtobxx> I have very few static devices  in the network, so I thought slaac would be overkill.
2021-12-25 00:53:47 <tobtobxx> and who uses dhcp when you already have ipv6??
2021-12-25 00:55:32 <Hello71> wat
2021-12-25 00:57:40 <Hello71> amberlee[m]: what does "directly on shared memory" mean
2021-12-25 00:57:46 <Hello71> and have you considered directio
2021-12-25 00:58:51 <amberlee[m]> Hello71: DAX. "File contents can be mapped into a memory window on the host, allowing the guest to directly access data from the host page cache."
2021-12-25 01:00:34 <Hello71> you have pmem?
2021-12-25 01:00:41 <amberlee[m]> afaik its not possible to force all IO be direct io. and its errant io by a multitude of applications that is the problem. but it looks like even patching the kernel to that effect would be a big undertaking. the kernel was written to touch all the memory it can.
2021-12-25 01:00:46 <amberlee[m]> virtiofs
2021-12-25 01:01:42 <Hello71> pretty sure dax only works with pmem (or emulated pmem)
2021-12-25 01:02:06 <amberlee[m]> virtios is a virtio-user, uses FUSE.
2021-12-25 01:02:23 <amberlee[m]> s/virtios/virtiofs
2021-12-25 01:02:23 <alpine-meetbot> amberlee[m] meant to say: virtiofs is a virtio-user, uses FUSE.
2021-12-25 01:02:25 <Hello71> you don't actually want the guest to use direct io. the vmexit overhead will be huge
2021-12-25 01:02:42 <Hello71> set qemu to use direct io so it doesn't occupy the host ram
2021-12-25 01:03:04 <amberlee[m]> im not sure it applies here? because direct io will be done on a virtiofs layer which doesnt actually do direct io just pretends to.
2021-12-25 01:03:21 <amberlee[m]> direct io does however bypass guest cache which is good.
2021-12-25 01:03:31 <Hello71> and also just because it's wrong to assume that cached ram is all used, it's also wrong to assume that cached ram is all unused
2021-12-25 04:41:47 <Nulo> tobtobxx: on v3.15, musl and musl-dbg do not have different versions
2021-12-25 04:47:09 <franklinyu> Hello, I have a question about Alpine Linux. Is there any way to know whether I need to restart after upgrade? For example, in Debian: https://serverfault.com/questions/667076/how-can-i-know-if-reboot-is-required-after-update
2021-12-25 04:47:24 <raven523> you may want to reboot after kernel upgrades
2021-12-25 04:47:46 <franklinyu> Thanks. How about others? Like maybe libc upgrades?
2021-12-25 04:48:44 <raven523> you may want to do that if you need to restart init. I don't remember if there's a way to reexec openrc but I haven't had a need to do it since I reboot my alpine VMs frequently
2021-12-25 04:49:57 <raven523> there is a restart-services package for determining which opewnrc services need to be restarted: https://dev.gentoo.org/~mschiff/restart-services/
2021-12-25 04:51:18 <raven523> that's only in the testing repo for edge, so it's probably not ready to use in production
2021-12-25 05:02:57 <franklinyu> I see. I guess I’ll try it on my personal server (nobody is losing money if it is down). Having an apk package is so much easier than `make install` myself.
2021-12-25 05:05:41 <franklinyu> Follow-up question: I see that the package has some dependencies, like Bash and ncurses. It doesn’t seem to specify the version of those dependencies; would `apk` pick the version from my current distribution (v3.15) or would it pull all dependencies from Edge?
2021-12-25 05:08:17 <raven523> it will choose from the repositories that you enabled
2021-12-25 05:15:07 <franklinyu> What if I don’t have the edge repository in `/etc/apk/repositories`, but install the restart-services package with the `--repository` flag of `apk add`?
2021-12-25 05:16:24 <raven523> I think it needs to be defined in /etc/apk/repositories, but mixing edge with v3.15 is probably not a good idea
2021-12-25 05:19:22 <franklinyu> I see; I thought that cherry-picking a package from edge/testing is a common practise. I’ll look elsewhere.
2021-12-25 05:33:44 <franklinyu> raven523: By the way do you use Stack Exchange? If so, I can give you some credit for this answer. `restart-services` isn’t super easy to know from Google.
2021-12-25 05:34:11 <raven523> I don't, but I just knew of that from it being availble for openrc on gentoo
2021-12-25 06:39:40 <argsh> hi
2021-12-25 08:34:47 <argsh> wich is the package to install obconf for openbox in alpine?
2021-12-25 08:37:13 <bdju> argsh: there's obconf-qt
2021-12-25 08:37:17 <bdju> not sure if there is another version also
2021-12-25 12:39:22 <argsh> i try to install obconf-qt
2021-12-25 12:39:32 <argsh> apk says no such package
2021-12-25 12:39:53 <ikke> It's in edge/testing
2021-12-25 12:40:02 <argsh> ah ok thanks ikke
2021-12-25 12:40:08 <ikke> https://pkgs.alpinelinux.org/packages?name=obconf-qt&branch=edge
2021-12-25 12:44:56 <argsh> then i must run apk with --repository option?
2021-12-25 12:45:13 <ikke> You can add it to /etc/apk/repositories
2021-12-25 12:46:51 <argsh> but if want to install one package from testing, i can do it with a single order?
2021-12-25 12:46:59 <ikke> yes
2021-12-25 12:47:17 <argsh> --repository option
2021-12-25 12:47:25 <ikke> correct
2021-12-25 12:47:30 <argsh> ok thanks again :)
2021-12-25 14:08:07 <q[m]12> Does someone run xmonad on alpine? I just failed to compile it, I guess it depends on glibc?
2021-12-25 14:09:17 <ktprograms> q[m]12: I got it working (self compiled GHC on aarch64). How did you install it?
2021-12-25 14:15:03 <q[m]12> I installed ghc and cabal using apk and the required headers according to xmonad's installation guide, and did cabal install --package-env=... --lib xmonad xmonad-contrib; it complains about a missing header file (inttypes.h), I just searched the repos for that file, apparently multiple packages provide it
2021-12-25 14:17:54 <q[m]12> * I installed ghc and cabal using apk and the required headers according to xmonad's installation guide, and did cabal install --package-env=... --lib xmonad xmonad-contrib; it complains about a missing header file (inttypes.h), I just searched the repos for that file, apparently multiple packages provide it
2021-12-25 14:20:24 <q[m]12> * I installed ghc and cabal using apk and the required headers according to xmonad's installation guide, and did cabal install --package-env=... --lib xmonad xmonad-contrib; it complains about a missing header file (inttypes.h), I just searched the repos for that file, apparently multiple packages provide it
2021-12-25 14:20:24 <q[m]12> EDIT: I installed musl-dev, now it fails during linking, I guess I should start by setting up a proper environment, then try again. thank's for the heads up
2021-12-25 14:35:20 <ktprograms> q[m]12: I installed it by following the instructions on the xmonad website: https://xmonad.org/INSTALL.html (using the cabal section)
2021-12-25 14:35:51 <ikke> q[m]12: please do not use the edit function in matrix, it's really noisy on IRC
2021-12-25 14:36:44 <ktprograms> q[m]12: You can also take a look at https://gist.github.com/ktprograms/011000bda16fbcdd5e00002b42d051d5#compiling-xmonad (a unpolished guide written by me to document the steps I did)
2021-12-25 14:45:27 <q[m]12> ikke: ok... (full message at https://matrix.org/_matrix/media/r0/download/matrix.org/qxDmNtBQhHYQbQpQNgXxPcvX)
2021-12-25 14:46:01 <q[m]12> s/thise/those/
2021-12-25 14:47:41 <ktprograms> q[m]12: Also don't send multiline messages, it requires clicking on a link for those using IRC
2021-12-25 14:48:00 <ktprograms> I think you'll want to install libffi-dev
2021-12-25 15:03:49 <q[m]12> Thanks again ktprograms, that fixed it
2021-12-25 20:46:45 <bmmcginty> Should have asked during my upgrade last night. I was going from 3.14 to 3.15. Changed the version number in my apk/repositories file, and I got an error about invalid signing keys. I'm also using community and edge as repos. Is that normal?
2021-12-25 20:47:16 <bmmcginty> Sorry, error came up when doing apk upgrade --available.
2021-12-25 20:53:28 <ikke> bmmcginty: https://wiki.alpinelinux.org/wiki/Release_Notes_for_Alpine_3.15.0#New_package_signing_keys
2021-12-25 21:09:06 <bmmcginty> ikke: Thank you. And shame on me for not reading that page.
2021-12-25 21:46:42 <neneko> Is it normal to not have a /boot when running diskless?
2021-12-25 21:48:27 <ikke> yes
2021-12-25 21:48:51 <ikke> The bootloader is part of the bootmedia
2021-12-25 21:48:57 <neneko> Syslinux broke after trying to update my boot disk, and now I'm trying to fix it. But update-extlinux complains that there's no /boot
2021-12-25 21:49:20 <neneko> And I see no way to specify a different path.
2021-12-25 21:49:34 <ikke> How did you update your bootdisk?
2021-12-25 21:50:10 <neneko> With setup-bootable
2021-12-25 21:50:20 <ikke> ok, sounds reasonable
2021-12-25 21:50:29 <ikke> but I wonder how syslinux broke then
2021-12-25 21:51:45 <neneko> Setup-bootable gave me a warning, don't remember exactly what it said but now it gets stuck on "boot:" with an error message saying a config is missing
2021-12-25 21:52:35 <neneko> It told me to run setup-bootable again with -f and -k, which I did
2021-12-25 21:56:46 <ikke> In anycase, update-extlinux is not suitable for boot media
2021-12-25 21:57:04 <ikke> You could manually run extlinux 
2021-12-25 21:57:57 <neneko> I ran `extlinux --install /media/sda1/boot` but that didn't seem to help
2021-12-25 22:05:03 <neneko> There is no /media/sda1/boot/extlinux.conf. I'm guessing that's the problem
2021-12-25 23:27:26 <franklinyu> I noticed that the aports repository doesn’t have a license. Would we appreciate work on this? https://gitlab.alpinelinux.org/alpine/aports/-/issues/9074
2021-12-25 23:29:16 <tobtobxx> Hmm... openrc unit scripts sometimes have options in them, that only get set if there's no exisiting value, for example NTPD_OPTS in https://git.alpinelinux.org/aports/tree/main/busybox-initscripts/ntpd.initd. Is there a recommended way of setting them? Maybe in rc.conf or something?
2021-12-25 23:46:42 <franklinyu> tobtobxx: My understand is setting them in the corresponding file in `conf.d`.
2021-12-26 00:07:05 <tobtobxx> minimal: What franklinyu said worked for me too. I just set NTPD_OPTS in /etc/conf.d/ntpd  to "-N" and it seems to override the default in /etc/init.d/ntpd. Still in favor of changing it in busybox-initscripts, but if you file the MR, be sure to adjust [ntpd.confd](https://git.alpinelinux.org/aports/tree/main/busybox-initscripts/ntpd.confd) too. In the meantime, adjusting conf.d/ntpd works for diskless systems.
2021-12-26 00:17:30 <minimal> tobtobxx: I'm still intending to change it to use the separate conf file as that is something that cloud-init creates
2021-12-26 00:18:52 <tobtobxx> Yeah, no, you're absolutely right. I was referring to removing the -p <explicit-peer> parameter. I removed this parameter in conf.d/ntpd and then it defaults to /etc/ntpd.conf.
2021-12-26 00:23:35 <minimal> actually the init.d and conf.d ntp files in busybox-initscripts appear to be shared by both the busybox and ntpsec packages I'll have have to come up with something that works with both of those
2021-12-26 03:35:18 <bmmcginty> I've got an issue showing up in znc-1.8.2r7, not a problem in znc-1.8.2r5. "throw_bad_array_new_length: symbol not found". I'm trying to track down the commit that might have changed things, but the change from r5 to r7 only touches the version number. I suppose I shoudl report this to the r7 commit author?
2021-12-26 03:46:17 <bmmcginty> And...the commiter is at the top of the file. I'll email them.
2021-12-26 06:54:33 <ikke> bmmcginty: and you are not mixing different releases?\
2021-12-26 12:44:21 <atagen> hello, i'm wondering if there is an equivalent to glibc's /etc/ld.so.preload we can use
2021-12-26 12:44:50 <atagen> i'd like to drop in a new allocator as my default
2021-12-26 12:45:40 <atagen> strace doesn't show musl/busybox stuff looking in anything like this, only ld-musl-$arch.path but that seems only for searching, not overriding
2021-12-26 12:47:16 <ikke> atagen: probably better to ask in #musl
2021-12-26 12:47:28 <ikke> on libera.chat
2021-12-26 12:48:59 <atagen> cool, thanks :)
2021-12-26 13:00:13 <ove> Hi! I'm having some trouble to get wifi up and running on raspberry pi, I haven't gotten to the install yet but I ran "apk add wireless-tools wpa_supplicant". "lsmod" lists "brcmfmac" but "ip a" doesn't show "wlan0". Do I need to add something to the usb?
2021-12-26 13:08:54 <ove> raspberry pi 400*
2021-12-26 13:15:26 <ktprograms>  ove: Did you try using setup-interfaces?
2021-12-26 13:16:10 <ktprograms> ove: Oh, sorry, if it doesn't show up in 'ip a' it probably means the firmware isn't loaded.
2021-12-26 13:20:58 <ktprograms> You could try installing the linux-firmware-brcm (I'm guessing that's the correct one because it has bcm43xx-0.fw, if that isn't the right chip model you'll need to check dmesg to see which chip the pi has)
2021-12-26 13:36:43 <ove> dmesg gives "brcmfmac mmc1:0001:1: Direct firmware load for brcm/brcmfmac43456-sdio.raspberrypi,400.bin failed with error -2"
2021-12-26 13:44:59 <ktprograms> ove: Is this before or after installing linux-firmware-brcm?
2021-12-26 13:47:12 <ove> "apk add linux-firmware-brcm" gives "no such package"
2021-12-26 13:49:06 <ove> I downloaded it from "https://dl-cdn.alpinelinux.org/alpine/v3.15/main/aarch64/linux-firmware-brcm-20211027-r0.apk" but not sure where I should put it
2021-12-26 13:56:18 <ktprograms> ove: What do you have in /etc/apk/repositories? (BTW I don't think it's a good idea to manually install a .apk file)
2021-12-26 13:59:50 <psykose> manually installing apks is fine, it's not any different than installing them from the repo, you get the same version checks
2021-12-26 14:00:26 <ktprograms> psykose: I mean not manually with tar
2021-12-26 14:00:29 <psykose> ah, sure
2021-12-26 14:03:04 <ove> It's just the things that came with the tarball and "linux-firmware-brcm-20211027-r0.apk" that I added
2021-12-26 14:04:12 <ove> but the brcm driver isn't in there if you just extract the tar and adding it doesn't seem to fix the problem neither
2021-12-26 14:06:01 <ktprograms> ove: Unfortunately I'm not sure what the problem is so I'm not sure how to fix it.
2021-12-26 14:08:15 <bmmcginty> ikke: re znc, I did have edge and latest-stable repos enabled at the same time, but libstdc++, which that array function seems to come from, doesn't have an updated version in edge far as I can see.
2021-12-26 14:15:49 <psykose> no, the issue is that you are mixing repos
2021-12-26 14:16:28 <psykose> the rebuild znc relies on gcc11 libstdc++, and you cannot upgrade to that without upgrading 90% of your system to edge to use it as well
2021-12-26 14:16:41 <psykose> so just move to edge
2021-12-26 14:17:22 <psykose> or don't, and remove the edge repos
2021-12-26 18:27:37 <ove> ktprogram: maybe the drivers are just missing in the tarball? I will try setting it up later with ethernet cable tho, thx for the help
2021-12-27 02:56:29 <bmmcginty> psykose: Thank you.
2021-12-27 03:42:02 <Saijin_Naib[m]> Hello71: Thanks for the WINE upgrade. Massive functionality boost for my limited usage here.
2021-12-27 10:55:09 <nico[m]> Do we have an `apk search` or a wiki entry for all browsers included in alpine? I find myself searching for various browsers so that I can pre-configure them from time to time, but not all typical browsers are there
2021-12-27 10:59:33 <lopid> maybe look for "web browser" in the info, somehow
2021-12-27 11:44:48 <feuerteufel> Hi, I yust upgraded my AL edge and there was the following error:
2021-12-27 11:44:54 <feuerteufel> stat: can't stat 'usr/bin/who': No such file or directory
2021-12-27 11:44:54 <feuerteufel> Executing busybox-1.34.1-r5.post-upgrade
2021-12-27 11:45:07 <psykose> that's fine, no real error occured
2021-12-27 11:45:49 <feuerteufel> It looks like ther is a "/" missing somewhere in a script ...
2021-12-27 11:48:17 <feuerteufel> Ok, so nothing to do?
2021-12-27 11:49:07 <psykose> no, it's just erroneous error log
2021-12-27 11:49:11 <psykose> nothing actually happened
2021-12-27 11:49:43 <feuerteufel> Ok, thank you!
2021-12-27 12:50:09 <minimal> feuerteufel: you can ignore that message. I did raise a MR, !28733, several days ago to fix this, it just has not been merged yet
2021-12-27 15:38:59 <Tenkawa> I was curious if there is a non-x86 specific channel? I didn't see one listed on the wiki.
2021-12-27 15:39:50 <nero> im somewhat sure this channel is not x86 specific
2021-12-27 15:43:29 <Tenkawa> Good. I'm going to be starting to look at porting some os builds from Debian on Odroid (and possibly RPI hardware depending on how much the hardware fights back) to Alpine and just am trying to make note of where to look for individual components I might have to rewrite that might already have equivalents already done.
2021-12-27 15:49:35 <earendel> what does that [m] stand for?
2021-12-27 15:49:43 <nero> matrix bridge
2021-12-27 15:49:56 <nero> its another messenger linked with irc
2021-12-27 15:50:55 <earendel> heard of. is it a convention, or are all nicks joined thru matrix marked?
2021-12-27 15:54:31 <bdju> I know you used to be able to change your nick after connecting to remove the [m], not sure if you still can
2021-12-27 15:54:59 <bdju> so they should all be marked by default anyway
2021-12-27 16:00:24 <nero> mh, im looking for a way to transfer files via serial port
2021-12-27 16:00:43 <nero> traditionally there has been the xmodem protocol, but it seems the tooling for it has disappeared
2021-12-27 16:16:09 <minimal> nero: ppp?
2021-12-27 16:17:20 <nero> the remote is a microcomputer
2021-12-27 16:17:39 <nero> i can write a program for there, but ppp has ip stack on it
2021-12-27 16:24:31 <minimal> ok, about zmodem?
2021-12-27 16:24:54 <nero> i dont see it in alpine
2021-12-27 16:25:30 <nero> i remember there used to be a sx and sz command, for send xmodem and send zmodem
2021-12-27 16:25:33 <fcolista> ckermit?
2021-12-27 16:25:37 <nero> but i dont know where i've seen that
2021-12-27 16:26:54 <fcolista> I think that lrzsz + ckermit are enough nero 
2021-12-27 16:27:18 <fcolista> lrzsz is x/z/y-modem 
2021-12-27 16:27:36 <nero> ahh, usr/bin/lsx is what i was looking for
2021-12-27 16:29:06 <Tenkawa> fcolista: I wondered when someone was going to mention kermit
2021-12-27 16:30:50 <minimal> kermit brings back decades-old memories
2021-12-27 16:30:51 <fcolista> iirc kermit allows by his own to transfer file... or ofc you can use lrzsz for [xyz]modem 
2021-12-27 16:30:53 <nero> kermit is like a terminal for modem use, ofc it has file transfer functionality
2021-12-27 16:31:07 <nero> i need something scriptable
2021-12-27 16:31:40 <fcolista> kermit is scriptable..
2021-12-27 16:32:00 <Tenkawa> minimal: reminds me of my flipswitch modems
2021-12-27 16:32:08 <nero> fcolista: sz/sx wins in terms of manpage length
2021-12-27 16:32:19 <fcolista> but, yeah...we are talking about eons ago...
2021-12-27 16:32:28 <Tenkawa> (early 80's
2021-12-27 16:32:28 <Tenkawa> )
2021-12-27 16:32:40 <fcolista> nero, good that you have found what you were looking for :)
2021-12-27 16:38:40 <nitsky> Hi, is it possible to build an alpine package and index using `abuild` from another distro? I tried and observe that `apk` errors stating that it cannot open the apk database, which makes sense. Is it possible to work around this issue?
2021-12-27 16:43:07 <minimal> nero: there's alway "cat" combined with uuencode/uudecode - used that decades ago to transfer files over a terminal connection :-)
2021-12-27 16:54:50 <Tenkawa> minimal: haahaa wow thats ancient
2021-12-27 16:57:19 <nero> background: i got an IBM PC compatible, im writing bare metal code in assembly for it
2021-12-27 16:57:31 <nero> idea was having a boot sector that fetches the data via xmodem or such
2021-12-27 16:58:26 <nero> ideally its in such a way that i only need to run a makefile client-side, and the pc picks up the start and end of transmission automatically
2021-12-27 17:03:38 <minimal> Tenkawa: did that back in the day when I was on a X.25 network using a PAD-to-telnet gateway to get to a TCP/IP network - as it was not a 8-bit clean connection I used uuencode/uudecode to transfer files over it :-)
2021-12-27 17:04:25 <Tenkawa> minimal: yeah I was doing that in late 80's early 90's as well
2021-12-27 17:04:54 <Tenkawa> I did my X.25 time in mid 90's though for a living in telecom.
2021-12-27 17:05:12 <Tenkawa> I actually enjoyed it
2021-12-27 17:06:42 <minimal> Tenkawa: I remember the X.25 software stack taking up a large percentage of PCs RAM before anything else was even run :-(
2021-12-27 17:07:09 <vlp[m]> Hello, I have a 404 on rofi-pass package. Is this a bug or the package has been decommissioned? 
2021-12-27 17:08:51 <Tenkawa> minimal: ahh yeah I was on SVR4 machines so different hardware resources to deal with 
2021-12-27 17:09:00 <Hello71> https://gitlab.alpinelinux.org/alpine/aports/-/merge_requests/28901
2021-12-27 17:16:44 <minimal> Tenkawa: UK universities in the last 80s, early 90s used X.25 rather than TCP/IP for the country-wide network
2021-12-27 17:17:34 <Tenkawa> minimal: we barely used dialup where I was for "anything" much less anything that advanced.
2021-12-27 17:19:23 <minimal> Tenkawa: 64Kbps leased lines (often via Telco microwave towers), back when that was really advanced.
2021-12-27 17:20:08 <Tenkawa> haahaa
2021-12-27 17:35:22 <earendel> electro-magnetic 
2021-12-27 17:35:25 <earendel> universe
2021-12-27 17:37:57 <lopid> intergalactic
2021-12-27 17:37:59 <lopid> radiostation
2021-12-27 17:38:47 <vlp[m]> <Hello71> "https://gitlab.alpinelinux.org/..." <- Tks
2021-12-27 17:39:09 <earendel> https://www.youtube.com/watch?v=VSygqiOyD9M
2021-12-27 21:08:56 <notaname> hey
2021-12-27 21:10:04 <notaname> i've done some research on installing alpine on my asus chromebook c201, and I came across these scripts: https://git.tilde.town/mio/alpine-c201. however, it seems they do not work anymore, and even with tinkering, they refuse to work. are there any other installation guides for the c201?
2021-12-27 21:11:06 <notaname> i also looked at the page https://wiki.alpinelinux.org/wiki/Alpine_on_ARM on the wiki, but it seems my chipset isn't there
2021-12-27 21:38:53 <oneinsect> you can port it
2021-12-27 21:39:35 <oneinsect> With Rockchip 3288-C Quad-Core its easy
2021-12-27 21:41:49 <notaname> what firmware is that on the wiki page>
2021-12-27 21:42:30 <notaname> searched for 3288 and rockchip and got no firmware
2021-12-27 21:45:27 <notaname> oneinsect: ^^?
2021-12-27 22:09:49 <notaname> again, does anyone have any advise on this?
2021-12-28 02:07:00 <bogen85> https://nopaste.ml/#XQAAAQCCAQAAAAAAAAARiAjm+lNpdB2PYGC8fkgz5b4IKH1PfEBk42G2vYQmP7pVjyQuO0TQ6DXjQwHR6ijH68llbWKe2UN/WKlaG8mvrWODb1zpWvJ+YPTuuBpzUNHoceSxHQZ/1a1AhHpVg45lX3UJoXdDm/WyDgVSDoptbYbFPHeIJZY3oCTAo2h/L/XzaLKIgjpHaH+XauatXvmbHnrp/P3c3ZSgCYTcBBePUBNyM+0dS/aYxO0m8x5TKImzV5HwUQvySNFMFHUwxfFoyB9fQh2FxAldEqLA6GrIUXu7rKyyKWomanGcX5hFIAlMXZpXgJo3DR/f8BCtLK7J+cpsUSll800+XCLH8Av5k7GdGBTqH8hvmO2G/WYVwoaF+x/8/NoOV67Z6VGbLf/hXiyA 
2021-12-28 02:07:03 <bogen85> how do I add a module using mkinitfs?
2021-12-28 02:08:05 <bogen85> I tried what is at the above paste, but when I get to the emergency shell I can't load or find that module
2021-12-28 02:10:19 <bogen85> I am using /boot/initramfs-virt which is the result of  mkinitfs 5.15.11-0-virt
2021-12-28 02:10:22 <minimal> bogen85: can't read that with Javascript disabled.....use one of the paste sites that doesn't require javascript?
2021-12-28 02:10:33 <bogen85> minimal: sure, hold on
2021-12-28 02:11:10 <bogen85> minimal, do you have one that you would recommend? (that does not require js?)
2021-12-28 02:11:48 <Hello71> bpaste, ix, debian, ubuntu, anything but pastebin.com and the security theater javascripts
2021-12-28 02:12:34 <minimal> I think tpaste is the most common one on here, there's even an Alpine package for it
2021-12-28 02:14:09 <bogen85> https://tpaste.us/gB7l
2021-12-28 02:15:51 <minimal> bogen85: ok, you need to add "9p" to the features list in /etc/mkinitfs/mkinitfs.conf
2021-12-28 02:16:04 <minimal> then run "mkinitfs" to rebuild the initramfs file
2021-12-28 02:16:21 <bogen85> minimal, ok, I will try that, rebuild, and try booting it
2021-12-28 02:16:38 <minimal> I'm not sure what "root=root9p" is supposed to be on your command line
2021-12-28 02:16:58 <minimal> this is assuming you are using a disk *sys" install of Alpine rather than run-from-ram
2021-12-28 02:17:23 <bogen85> this is 9p, a pass through filesystem
2021-12-28 02:17:34 <minimal> did you change that "root=root9p" value from something else?
2021-12-28 02:17:55 <bogen85> root9p is the label of the 9p device
2021-12-28 02:17:57 <Hello71> apart from alpine, there is no way for the system to detect that "root9p" means 9p type. you obviously must use some fs type option, e.g. rootfstype
2021-12-28 02:18:12 <Hello71> for e.g. cifs this is detected by // or \\ prefix
2021-12-28 02:18:15 <minimal> so then if would need to be "root=LABEL=root9p"
2021-12-28 02:18:34 <Hello71> as a separate issue, the initramfs doesn't come with 9p modules by default
2021-12-28 02:18:59 <minimal> Hello71: I talked him through adding 9p to his initramfs
2021-12-28 02:19:02 <Hello71> theoretically i guess if you enable 9p feature it could try 9p type but that's too magic for alpine
2021-12-28 02:19:17 <Hello71> yes i did read that already
2021-12-28 02:19:22 <Hello71> and it's not that kind of label
2021-12-28 02:19:38 <Hello71> also virtiofs is better in most cases despite its somewhat dubious security
2021-12-28 02:19:50 <minimal> right, label is a different thing. bogen71: which bootloader are you using? syslinux or grub?
2021-12-28 02:19:56 <bogen85> [    0.843759] 9p: Installing v9fs 9p2000 file system support <-- I got further
2021-12-28 02:20:10 <bogen85> no bootloader, kvm command line
2021-12-28 02:20:18 <Hello71> it's also not related to bootloader if you got to initramfs already
2021-12-28 02:20:23 <minimal> ah ok
2021-12-28 02:20:49 <minimal> Hello71: my question to him about bootloader was to ensure any cmdline options were added to the appropriate bootloader config file to make them persistent
2021-12-28 02:21:25 <Hello71> i guess, but wouldn't it be easier to ask about /proc/cmdline
2021-12-28 02:21:47 <minimal> that was in his tpaste output
2021-12-28 02:22:19 <minimal> bogen85: so how far has it got now?
2021-12-28 02:22:50 <minimal> you will see more if you drop "quiet" for the cmdline
2021-12-28 02:22:51 <Hello71> well it still can't work if you don't specify any rootfstype
2021-12-28 02:22:53 <bogen85> $ sudo qemu-system-x86_64 -enable-kvm -kernel $(pwd)/miniroot/boot/vmlinuz-virt -initrd $(pwd)/miniroot/boot/initramfs-virt -fsdev local,id=root9p,path=$(pwd)/miniroot,security_model=mapped -device virtio-9p-pci,fsdev=root9p,mount_tag=root9p -nographic -append 'root=root9p rw modules="9p" rootfstype=9p rootflags=trans=virtio console=ttyS0'
2021-12-28 02:22:54 <minimal> s/for/from/
2021-12-28 02:22:54 <alpine-meetbot> minimal meant to say: you will see more if you drop "quiet" from the cmdline
2021-12-28 02:23:42 <bogen85> [    6.990711] Mounting root: ok.
2021-12-28 02:23:42 <bogen85> qemu-system-x86_64: warning: 9p: Multiple devices detected in same VirtFS export, which might lead to file ID collisions and s!
2021-12-28 02:23:42 <bogen85> ok.
2021-12-28 02:23:42 <bogen85> switch_root: can't execute '/sbin/init': Symbolic link loop
2021-12-28 02:24:08 <bogen85> and then I get a kernel panic
2021-12-28 02:24:12 <Hello71> sure
2021-12-28 02:24:16 <bogen85> so, it is getting closer
2021-12-28 02:25:41 <minimal> normally /sbin/init is a softlink to /bin/busybox
2021-12-28 02:25:50 <minimal> no idea how you're getting a link loop
2021-12-28 02:26:11 <bogen85> yeah: /sbin/init -> /bin/busybox
2021-12-28 02:30:14 <bogen85> https://tpaste.us/eVBb <- that is with init=/sbin/openrc-init  # ok... I may need to fix the root filesystem I'm using
2021-12-28 02:31:54 <bogen85> I added openrc and udev-init-scripts-openrc to the mini root fs download from the front page (chrooted into it) https://dl-cdn.alpinelinux.org/alpine/v3.15/releases/x86_64/alpine-minirootfs-3.15.0-x86_64.tar.gz
2021-12-28 02:32:22 <bogen85> added linux-virt and mkinitfs
2021-12-28 02:32:38 <minimal> bogen85: Alpine does not (normally) use openrc-init, it uses Busybox init
2021-12-28 02:33:05 <bogen85> hmm, ok, I added that earlier because I was running into issues
2021-12-28 02:33:19 <minimal> openrc is used, but not as init
2021-12-28 02:33:47 <bogen85> I'll start clean and not add it. (just linux-virt mkinitfs and the 9p configuration)
2021-12-28 02:34:02 <minimal> also udev/eudev is not normally used, rather mdev
2021-12-28 02:39:31 <eletrotupi> hello
2021-12-28 02:40:46 <eletrotupi> while plugging a SD card (my laptop has a SD card slot) I can't seem to be able to mount them 
2021-12-28 02:40:52 <eletrotupi> they don't appear on lsblk or fdisk
2021-12-28 02:41:02 <eletrotupi> any tips on how should I proceed to find it?
2021-12-28 02:45:34 <bogen85> ok: "switch_root: can't execute '/sbin/init': Symbolic link loop" again... I'll install alpine in a regular fashion, grab the rootfs from that, and try agin. Rather than starting with the miniroot from the download page
2021-12-28 02:45:51 <minimal> electrotupi: does any device show up when the SD card is plugged in, like /dev/mmcblk0?
2021-12-28 03:14:09 <bogen85> ok, I extracting the rootfs/bootfs tree from a working install's image... Same problem: qemu-system-x86_64: warning: 9p: Multiple devices detected in same VirtFS export, which might lead to file ID collisions and severe misbehaviours on guest! You !
2021-12-28 03:14:09 <bogen85> switch_root: can't execute '/sbin/init': Symbolic link loop
2021-12-28 03:14:30 <bogen85> so I guess I'll need to track that down (the multiple devices detected)
2021-12-28 03:14:52 <bogen85> but... it is still seeing /sbin/init, so that might not be the case
2021-12-28 03:19:13 <bdju> eletrotupi: I've had weird SD card mounting issues before on a different distro. I think a reboot fixed it.
2021-12-28 03:30:43 <bogen85> ok, I believe it is related to how the symlinks are being resolved in the 9p mount... I tried pointing init to other stuff, and keep getting the symlink loop. I installed bash and pointed to it for init, and got this
2021-12-28 03:30:44 <bogen85> Error loading shared library libreadline.so.8: Symbolic link loop (needed by /bin/bash)
2021-12-28 03:31:36 <bogen85> so 9p seems to have issues when used as an alpine root... (due to the way symlinks are resolved...)
2021-12-28 03:37:03 <bogen85> yeah... I booted into the rescue shell and mounted the root filesystem
2021-12-28 03:37:04 <bogen85> ls: /mnt/bin/ash: cannot read link: Symbolic link loop
2021-12-28 03:37:05 <bogen85> lrwxrwxrwx    1 root     root            12 Dec 28 02:54 ash
2021-12-28 03:38:51 <minimal> bogen85: perhaps as Hello71 suggested you should try virtiofs instead
2021-12-28 03:51:45 <eletrotupi> minimal: no
2021-12-28 03:52:12 <bogen85> the symlink issue is discussed here: https://lwn.net/Articles/788333/
2021-12-28 03:53:21 <minimal> eletrotupi: perhaps then the driver for it is not installed/built into your kernel
2021-12-28 03:53:53 <eletrotupi> hm, had some similar issues with wifi drivers a while ago 
2021-12-28 03:54:04 <eletrotupi> thanks minimal, will investigate further then
2021-12-28 03:56:45 <minimal> eletrotupi: for my laptop the SDcard slot is handled by the rtsx_pci and rtsx_pci_sdmmc kernel modules
2021-12-28 03:57:06 <minimal> that's for a builtin Realtek card reader
2021-12-28 03:58:44 <minimal> eletrotupi: try "lspci | grep -i card" to try and find the device type
2021-12-28 04:24:33 <Hello71> eletrotupi: you updated kernel without rebooting?
2021-12-28 05:23:16 <bogen85> Hello71, I got it working (no disks attached) with virtiofs (Based a rootfs copied from an alpine installed to a disk). I'll try it now the miniroofs (hopefully that will work, as it makes the bootstrapping process easier)
2021-12-28 06:23:10 <rwhalen> Hey all, I'm trying to set up my wifi following the instructions in the wiki but `ip link set wlan0 up` doesn't change the state of the card. `rfkill list` also shows that nothing is blocked... Any ideas?
2021-12-28 06:37:30 <bogen85> Hello71, minirootfs also requires openrc and busybox-initscripts to actually boot... so I got a minimal functioning boot with no errors (also vm "shutdown" does not trigger a shutdown). But I'll troubleshoot that later.
2021-12-28 06:43:16 <rwhalen> xcape -e 'Caps_Lock=Escape' &
2021-12-28 06:43:28 <rwhalen> woops.. disregard
2021-12-28 06:59:01 <rwhalen> bindsym --locked XF86MonBrightnessDown exec brightnessctl set 5%-
2021-12-28 06:59:01 <rwhalen> bindsym --locked XF86MonBrightnessUp exec brightnessctl set 5%+
2021-12-28 13:46:12 <blt[m]> I got `gpg: connecting dirmngr at '/root/.gnupg/S.dirmngr' failed: IPC connect call failed` when I try to fetch a key from gpg --card-edit, I got dirmngr installed and gpg-agent running, don't have S.dirmngr created automaticaly nor crls.d when I touch/mkdir with the right permission I got it work (sometimes) is there anything missing?
2021-12-28 14:21:35 <bogen85> Hello71, got all the services and acpi working, thanks for your help last night
2021-12-28 14:21:41 <Hello71> mmhmm
2021-12-28 14:23:00 <bogen85> well, I'll come up with a cleaner way of bootstrapping, but for now I'm learning a lot more of how alpine boots (what services packages needed by a base system) (which when you just do the normal install you don't know...)
2021-12-28 14:24:44 <Hello71> it is in initramfs-init
2021-12-28 14:24:57 <Hello71> which is a confusing place to put it
2021-12-28 14:28:59 <bogen85> Hello71, you mentioned virtiofs being a security risk... From the standpoint of the files not being in an opaque container, or something else?
2021-12-28 14:29:47 <Hello71> the daemon has been written by people not intimately familiar with C security
2021-12-28 14:30:08 <bogen85> hmmm.... ok
2021-12-28 14:30:24 <Hello71> i found CVE-2020-35517 within ten minutes of viewing the source code
2021-12-28 14:30:35 <bogen85> well, not "ok"... I mean "ack..."
2021-12-28 14:31:12 <Hello71> the good news is that it's probably not much worse than unsandboxed qemu with default devices
2021-12-28 14:31:32 <Hello71> the bad news is that's a really low bar
2021-12-28 14:32:56 <Tenkawa> Hello71: that is also very dependent on what "version" of virtiofs you are running if you read that CVE to be fair
2021-12-28 14:33:04 <bogen85> "written by people not intimately familiar with C security"... too much of that across the spectrum
2021-12-28 14:34:03 <Hello71> Tenkawa: what?
2021-12-28 14:34:18 <Tenkawa> virt:8.2
2021-12-28 14:34:18 <Tenkawa> March 8, 2021
2021-12-28 14:34:18 <Tenkawa> Fixed
2021-12-28 14:34:18 <Tenkawa> https://access.redhat.com/errata/RHSA-2021:0743
2021-12-28 14:34:46 <Tenkawa> thats directly from : https://access.redhat.com/security/cve/cve-2020-35517
2021-12-28 14:35:43 <Hello71> 8.2 isn't a qemu version
2021-12-28 14:35:50 <Tenkawa> so virtio "has" been fixed depending where you look
2021-12-28 14:36:16 <Hello71> yes, most cves have fixes available. what is your point
2021-12-28 14:36:19 <Tenkawa> neither is that CV
2021-12-28 14:36:22 <Tenkawa> CVE
2021-12-28 14:36:36 <Hello71> wat
2021-12-28 14:36:47 <Tenkawa> that CVE is in relationship to the virtio driver
2021-12-28 14:36:59 <Tenkawa> not qemu
2021-12-28 14:37:17 <Hello71> virtiofsd is bundled with qemu
2021-12-28 14:37:28 <Tenkawa> its the virtio driver "for" qemu
2021-12-28 14:37:35 <Hello71> the "virtio driver" is part of linux and is not relevant
2021-12-28 14:37:45 <Tenkawa> point being it "has" been fixed
2021-12-28 14:37:53 <Hello71> normally host side code is not called driver, it is called device emulator
2021-12-28 14:37:59 <Hello71> 14:36 <Hello71> yes, most cves have fixes available. what is your point
2021-12-28 14:39:05 <Tenkawa> my point is why are you being so negative about using it to people
2021-12-28 14:39:46 <Tenkawa> when they are trying to figure out ways to fix problems/implement solutions
2021-12-28 14:41:02 <Hello71> because this is extremely security-critical code, and the fact that i found a critical vulnerability in ten minutes suggests that the design is weak
2021-12-28 14:41:32 <Hello71> it completely breaks guest isolation for the vast majority of configurations
2021-12-28 14:43:29 <Hello71> and "virtio" is not virtiofs
2021-12-28 14:45:42 <Tenkawa> Hello71: I'm done. If you want to keep wasting your keystrokes feel free
2021-12-28 14:46:52 <Hello71> i'll be waiting for you to say a single thing of value, or really even basic comprehensibility
2021-12-28 15:22:08 <Tenkawa> Any Odroid Alpine users on per chance?
2021-12-28 15:22:17 <bogen85> minimal, I got everything working bootstrapped off the mini root fs download (booting, network, dropbear, acpi shutdown, etc, etc...) and booting from and using the virtiofs root. Thanks for your help last night!
2021-12-28 15:23:21 <minimal> bogen85: no problem. I haven't used p9 or virtiofs before so couldn't help with that aspect of things
2021-12-28 15:23:29 <Tenkawa> I've noticed the wiki doesn't list any current support for the N2. I wondered if anyone has had any luck with experimenting with it
2021-12-28 15:24:12 <Tenkawa> (the N2+ would be nice too.. I have both however I was going to start with the N2 first)
2021-12-28 15:24:52 <bogen85> minimal, yeah, the main thing was the module inclusion in the initramfs. Just need to get over the hump over alpine in general... (mkinitfs, explicity what services are needed in what levels for a basic system, and so on...)
2021-12-28 15:26:44 <psykose> Tenkawa: afaik people have gotten it to work on the n2
2021-12-28 15:27:05 <psykose> sadly i don't own one myself
2021-12-28 15:27:20 <Tenkawa> psykose: that's good news. I'll do some more digging through google/etc then
2021-12-28 15:27:35 <Tenkawa> psykose: thanks for the response
2021-12-28 15:36:57 <PureTryOut> Odroid HC2 is supported by pmOS, but I doubt it's usable for the N2
2021-12-28 15:44:12 <psykose> the worst you would have to do is find a way to boot downstream kernels and use some other version of uboot i think
2021-12-28 15:46:39 <Tenkawa> PureTryOut: yeah I thought about trying to pull the dtbs/u-boot I use in our custom built Debian64 images and integrating them into a generic arm64 image.. just need to figure out the one with the widest kernel build
2021-12-28 15:47:03 <Tenkawa> psykose: yeah.. that was my other thought was overlaying the kernel too
2021-12-28 15:48:02 <Tenkawa> I work on an imager project and have that all right here at my disposal.. Its just merging the output into an Alpine image that I need to do really I think
2021-12-28 16:07:56 <PureTryOut> Fun, the Mesa 21.3.2 upgrade (from 21.3.1) broke the Dolphin Emulator 😢
2021-12-28 16:08:46 <psykose> is there something broken post-startup?
2021-12-28 16:14:48 <PureTryOut> Not sure what you mean? The GUI of Dolphin launches but it failes to initialize any video backends when trying to start a game
2021-12-28 16:30:10 <jhjacobs81> :) good day!
2021-12-28 16:40:17 <psykose> it fails with wayland-egl but works if i set QT_QPA_PLATFORM=xcb for some reason
2021-12-28 16:43:31 <PureTryOut> Oh wow you are right
2021-12-28 17:01:55 <rwhalen> Hey all. When I try to `ip link set wlan0 up` it doesn't give me any errors but doesn't actually change the state. I have checked that `rfkill list` doesn't show the interface as blocked but I'm not sure what else to check. Any ideas?
2021-12-28 17:03:57 <psykose> i think i've had these wayland-egl Qt issues a bunch of times, but idk what even causes them, considering that is provided by qt5/6-qtwayland
2021-12-28 17:05:45 <psykose> rwhalen: what is the output of `ip a`, in a pastebin
2021-12-28 17:06:26 <Hello71> iproute2 shows two states: administrative and operational. you can't administratively change the operational state, you need to "plug in cable"
2021-12-28 17:06:56 <PureTryOut> psykose: it also only broke after upgrading Mesa somehow... The main UI of Dolphin launches fine too
2021-12-28 17:07:12 <psykose> yea, it's weird
2021-12-28 17:07:41 <psykose> maybe a rebuilt version will work, once it's compiled for me
2021-12-28 17:07:56 <PureTryOut> rebuild of Mesa? No I have tried that already
2021-12-28 17:08:05 <psykose> dolphin
2021-12-28 17:08:13 <PureTryOut> Tried that too, I updated it locally
2021-12-28 17:08:16 <psykose> hrm
2021-12-28 17:08:19 <psykose> very strange
2021-12-28 17:09:30 <rwhalen> psykose: https://pastebin.com/eSnt0ng4  ... now that I think about it... I think I'm looking at `state` instead of UP which comes earlier in that line.
2021-12-28 17:09:46 <psykose> this interface is up
2021-12-28 17:09:55 <psykose> no carrier means you are just not connected to anything
2021-12-28 17:10:19 <rwhalen> Yea... I see that now.. I was misreading it. Thanks for the help!
2021-12-28 17:41:02 <Saijin_Naib[m]> Can anyone else using xfce run xfce4-screensaver-configuration from the terminal and let me know if it bombs on gettext for you too?
2021-12-28 17:42:54 <nero> yeah it fails
2021-12-28 17:47:01 <Saijin_Naib[m]> Hmm, okay, thank you
2021-12-28 17:47:11 <Saijin_Naib[m]> I'll open an issue on our tracker then. I think we may need a patch
2021-12-28 18:39:46 <jhjacobs81> Does anyone know if Alpine runs well on pcengines boards?
2021-12-28 18:59:10 <bdju> they're regular amd64, right? I imagine they should be okay. the only people I know with them run OpenBSD on them. one of them uses it as a router
2021-12-28 20:21:56 <Apachez> jhjacobs81: alpine is pretty vanilla so it should run just fine on about any x86 or x86-64 based cpu out there including motherboards and other peripheral devices (gpu's etc)
2021-12-28 20:22:49 <Apachez> vanilla linux have way better driver covarage compared to windows these days
2021-12-28 20:22:56 <Apachez> coverage
2021-12-28 20:27:01 <Tenkawa> Is there a build of the alpine kernel for RPI4+USB instead of MicroSD by chance? I'm not finding any builds with USB kernel booting enabled and I will build it myself if I need to however just wanted to check in case if it is packaged already,
2021-12-28 20:27:12 <Tenkawa> (64 bit)
2021-12-28 20:27:24 <ikke> Tenkawa: I don't think so
2021-12-28 20:27:52 <Tenkawa> ok.. I didn't think so.. good thing I compile a few 100 kernels a month
2021-12-28 20:29:33 <Tenkawa> thanks though.
2021-12-28 20:38:45 <psykose> you can try rpi4 uefi
2021-12-28 20:39:30 <psykose> you just flash it to the sd card, then you boot normal images off usb- but you have to deal with the potential issues/incompleteness of it :)
2021-12-28 20:40:27 <Tenkawa> once I get it to boot I can deal with that… just getting it to run is the tricky part atm
2021-12-28 20:41:13 <psykose> in my case it was as simple as copying the stuff to an sd, then flashing the aarch64 alpine iso to a usb, and that was that
2021-12-28 20:42:12 <Tenkawa> I'm also playing the tricky game of headless/usb-ttl install so this is a very interesting install
2021-12-28 20:42:23 <psykose> that's how i did it
2021-12-28 20:42:48 <psykose> accidentally fried a usb-ttl cable in the process
2021-12-28 20:43:36 <Tenkawa> psykose: I'm on #3
2021-12-28 20:43:38 <Tenkawa> lol
2021-12-28 20:43:41 <psykose> :p
2021-12-28 20:43:51 <Tenkawa> I have 4 spares now lol
2021-12-28 20:43:52 <psykose> it's too easy to plug the leads into +5v on random shit
2021-12-28 20:45:46 <Tenkawa> I had 1 that came off and jumped on a pin and fried it… and the other one got me because I'm colorblind and I have to be "very" careful and the light caught just at the right angle and I didn't see the right one
2021-12-28 20:46:39 <Tenkawa> this one is now marked and the extra + pin I dont use is taped off
2021-12-28 20:47:16 <Tenkawa> I am being extra cautious now 
2021-12-28 20:47:23 <psykose> i would do the same on #3
2021-12-28 20:47:30 <psykose> but on #2.. ehh, i can still be a bit careless :p
2021-12-28 20:48:08 <psykose> why don't they just make one that has some circuits that don't break when you mix up the connection
2021-12-28 20:48:08 <psykose> smh
2021-12-28 20:48:40 <Tenkawa> heheh they have one thats an actual board but its next to impossible to find
2021-12-28 20:49:11 <Tenkawa> it is spaced out and fits in order 
2021-12-28 21:23:25 <Tenkawa> nice
2021-12-28 21:24:05 <Tenkawa> psykose: first part worked… I'll get the install finished and try the transfer
2021-12-28 21:25:14 <psykose> awesome
2021-12-28 21:57:00 <MY-R> Tenkawa: rpi4 with fresh enough EEPROM can boot without problem from usb sticks, usb disks etc or you asked about something else?
2021-12-28 21:57:52 <heat> hi, I got a tiny rpi zero 2 W yesterday and I wanted to get alpine going, do you support it?
2021-12-28 21:58:13 <heat> i've tried the raspberry pi instructions and it didn't work
2021-12-28 22:00:13 <jhjacobs81> Apachez thanks :-)
2021-12-28 22:01:02 <Tenkawa> MY-R: the first image was giving infamous boot flash error however I have a m.2 interface though that boots uas though so I'm going to  try booting with a usb quirk next. I agree it "should" boot under the right config and kernel compile however this chassis is peculiar
2021-12-28 22:02:52 <Tenkawa> MY-R: I have 7 PI4's *6 regular and 2 argon40 m.2 units* currently all running a custom Debian64 build against USB devices so I agree it should work in principle
2021-12-28 22:03:04 <Tenkawa> er 8 units
2021-12-28 22:04:52 <Tenkawa> it could be as simple as a storage quirk itself
2021-12-28 22:05:13 <Tenkawa> I don't remember if I had one in before I unplugged the old os image
2021-12-28 22:08:50 <MY-R> I only need unpack alpine-rpi-3.15.0-aarch64.tar.gz on any pendrive/usb disk on prepared fat partition and raspberry will boot it up, then "setup-alpine" without problem can use such a drive and install system on it ready to boot
2021-12-28 22:09:28 <Tenkawa> yes.. I'm not using a regular pi 4
2021-12-28 22:09:48 <MY-R> Tenkawa: then is matter of extra modules in kernel, right?
2021-12-28 22:10:00 <MY-R> or need some quirks?
2021-12-28 22:11:28 <Tenkawa> possibly however highly unlikely since I can see the hardware booted… the uas/storage quirks are usually a problem on socs
2021-12-28 22:11:38 <Tenkawa> thats what I said above
2021-12-28 22:12:12 <Tenkawa> Tenkawa
2021-12-28 22:12:12 <Tenkawa> 5:01
2021-12-28 22:12:12 <Tenkawa> MY-R: the first image was giving infamous boot flash error however I have a m.2 interface though that boots uas though so I'm going to  try booting with a usb quirk next. I agree it "should" boot under the right config and kernel compile however this chassis is peculiar
2021-12-28 22:13:03 <Tenkawa> ok.. time to put the quirk in
2021-12-28 22:16:12 <MY-R> Tenkawa: ahh so sounds like a fun :P
2021-12-28 22:16:30 <Tenkawa> this "is" what I call fun.. I actually enjoy this
2021-12-28 22:16:45 <Tenkawa> (this is what I did for most of my career though)
2021-12-28 22:18:02 <MY-R> usually Im trying to get as much compatible devices which "should" work :)
2021-12-28 22:20:12 <MY-R> and with pi4 checking https://github.com/raspberrypi/rpi-eeprom/blob/master/firmware/release-notes.md from time to time is good idea
2021-12-28 22:21:16 <Tenkawa> looks good now
2021-12-28 22:21:51 <Tenkawa> now I have it on a 8gb pi 4 w m.2 SSD :)
2021-12-28 22:22:53 <MY-R> congrats :)
2021-12-28 22:23:40 <Tenkawa> thanks.. now to start setting up and rebuilding packages for my experiments
2021-12-29 01:19:07 <rwhalen> Hey all, I'm running into an issue with qutebrowser crashing with a qwindow error: ```Wayland does not support QWindow::requestActivate()```. From searching around it seems this is a wayland issue, but does anyone know of a way around it via compatibility libraries or something?
2021-12-29 01:25:42 <psykose> that is a warning and not the source of the crash
2021-12-29 01:26:00 <psykose> almost any qt program prints that one
2021-12-29 01:26:04 <psykose> you sure there's no other output?
2021-12-29 01:28:57 <rwhalen> hmm... I only see "Renderer process crashed (status 11)" as the error and the wayland warning. Everything else is debug information in the logs. Maybe the folks in #qutebrowser might know what that status code means?
2021-12-29 01:29:35 <bdju> is this a full browser crash or just a tab crash? and have you updated recently? are you on edge? there was an issue with some sites like youtube in qutebrowser, but I thought it just got fixed
2021-12-29 01:29:58 <bdju> also, yeah, asking in #qutebrowser on libera is probably a good idea as well
2021-12-29 01:30:34 <bdju> https://gitlab.alpinelinux.org/alpine/aports/-/issues/13333
2021-12-29 01:32:13 <rwhalen> It is specifically on youtube, yea. I'm on the main branch.. figured since I was new to alpine that I shouldn't bork my system by going to edge yet :P
2021-12-29 01:32:21 <rwhalen> And yes, it's a tab crash only on the youtube tab.
2021-12-29 01:33:32 <bdju> I haven't tested yet, but I think it's probably fixed on edge. I was hitting that issue the other day myself on my pinebook pro running postmarketOS (alpine-based)
2021-12-29 01:33:41 <bdju> unsure when that sort of change would hit main
2021-12-29 01:34:24 <psykose> that is fixed on edge most likely
2021-12-29 01:34:32 <psykose> it's caused by the stack size issue that got fixed today
2021-12-29 01:34:40 <rwhalen> Is edge relatively stable? I was on arch before and am used to some small amount of fixing things after updates.
2021-12-29 01:34:58 <psykose> more or less
2021-12-29 01:36:50 <rwhalen> Cool... maybe I can upgrade then. Thanks for the help psykose and bdju 
2021-12-29 01:39:31 <psykose> i made a backport for the fix in !28964 , whenever that gets merged
2021-12-29 01:42:14 <rwhalen> awesome! thanks so much :)
2021-12-29 01:42:31 <rwhalen> I'll check in on it to see how it goes
2021-12-29 11:47:50 <frag> is busybox crond good/recommended?
2021-12-29 11:50:24 <lopid> good enough for me
2021-12-29 11:51:04 <frag> nice
2021-12-29 11:51:20 <lopid> thanks!
2021-12-29 12:16:19 <frag> i have "(export) DISPLAY=:0" in crontab but it crond doesnt seem to register it... is it possible in busybox crond?
2021-12-29 12:16:45 <lopid> i haven't seen it in parens befotre
2021-12-29 12:17:27 <frag> just meant i tried with and withoput export
2021-12-29 12:17:51 <lopid> all cron implementations support a limited number of variables (they're not shell vars)
2021-12-29 12:18:33 <lopid> so you probably want that in your command, not separately at the top
2021-12-29 12:20:22 <frag> ah yes
2021-12-29 13:57:34 <heat> I asked this yesterday but it might've gotten lost in all the messages, I have a rpi zero 2 w I just got and I wanted to know if alpine supports it
2021-12-29 13:58:09 <heat> I tried the regular rpi instructions but they didn't work and if the fix is relatively simple I don't mind getting my hands dirty
2021-12-29 13:58:48 <lopid> the wiki is usually old 
2021-12-29 14:02:20 <Tenkawa> heat: even on other distros right now there is a bit of "manual work" usually needing to be done for the 0w2
2021-12-29 14:02:42 <Tenkawa> mostly dealing with the dtb
2021-12-29 14:03:10 <Tenkawa> otherwise it is basicly identical to a 3a+ setup
2021-12-29 14:03:32 <heat> yea the rpi os installed fine, manjaro also installed fine, but I tried ubuntu server and it didn't work, same with alpine
2021-12-29 14:04:02 <heat> what exactly do I need to do with the dtb?
2021-12-29 14:04:18 <Tenkawa> yeah if you can get a copy of the dtb from one of those and drop it in /boot it in theory might work
2021-12-29 14:04:33 <Tenkawa> thats how the custom os I build is currently doing it
2021-12-29 14:05:06 <Tenkawa> I use a custom debian64 build on it
2021-12-29 14:05:28 <Tenkawa> soon I get use to alpine more I am going to start testing on it
2021-12-29 14:05:38 <Tenkawa> (just started learning it 2 days ago)
2021-12-29 14:07:26 <Tenkawa> my fun part of the learning process is the ram vs persistent part.. I need to keep things persistent and not use lbu and my first attempt didnt go quite as planned
2021-12-29 15:00:25 <heat> Tenkawa: alright I'll try it out then, thanks!
2021-12-29 15:00:57 <Tenkawa> good luck. I'm having cable fun myself this morning heheh
2021-12-29 15:02:08 <Tenkawa> (trying to rebuild 2 SoC's)
2021-12-29 15:26:09 <Hello71> iirc the wifi doesn't work
2021-12-29 15:30:54 <Tenkawa> Hello71: why wouldn't it? it uses the standard brcmfmac. He might need the  brcm/brcmfmac43436-sdio firmware
2021-12-29 15:31:04 <Tenkawa> I'm running it right now
2021-12-29 16:01:32 <Apachez> heat: select the proper cpu arch of alpine (preferly diskless mode for rpi's) and you should be fine
2021-12-29 16:02:22 <heat> Tenkawa, is that firmware already included in the standard linux-firmware stuff or do I need to hack it in?
2021-12-29 16:02:26 <Apachez> https://wiki.alpinelinux.org/wiki/Raspberry_Pi
2021-12-29 16:02:32 <Apachez> https://wiki.alpinelinux.org/wiki/Raspberry_Pi_-_Headless_Installation
2021-12-29 16:02:55 <Apachez> https://www.alpinelinux.org/downloads/
2021-12-29 16:03:39 <heat> Apachez, I've read that and followed that but it didn't boot, that's the issue (I was probably lacking the .dtb which I see isn't present in the alpine tarball)
2021-12-29 16:03:57 <Hello71> #13302
2021-12-29 16:04:05 <Apachez> zero 2 W is a 32-bit and 64-bit capable
2021-12-29 16:04:05 <Apachez> ARMv7 + ARMv8-A support
2021-12-29 16:04:40 <Apachez> did you use https://dl-cdn.alpinelinux.org/alpine/v3.15/releases/armv7/alpine-rpi-3.15.0-armv7.tar.gz ?
2021-12-29 16:04:47 <Apachez> note the cpu arch
2021-12-29 16:04:59 <Apachez> there are 3 flavours for rpi depending on which cpu arch your model is using
2021-12-29 16:05:16 <heat> no
2021-12-29 16:05:22 <heat> I use the aarch64 one
2021-12-29 16:05:28 <heat> used*
2021-12-29 16:05:53 <heat> Hello71: I still have an image of the old manjaro installation so I'll probably look into that
2021-12-29 16:06:05 <Apachez> dunno if aarch64 would work for that
2021-12-29 16:06:18 <Apachez> armv7 should work for sure
2021-12-29 16:06:40 <heat> why would it not
2021-12-29 16:06:50 <Apachez> and so is the armhf (which would work on anything rpi but doesnt utilize all cpu features as the others)
2021-12-29 16:07:16 <heat> I'm running manjaro aarch64 on it just fine
2021-12-29 16:07:38 <Tenkawa> firmware is OS/arch agnostic
2021-12-29 16:07:47 <Tenkawa> it works on all
2021-12-29 16:08:10 <Tenkawa> that's its purpose
2021-12-29 16:09:18 <Apachez> seems to be mixed results if aarch64 should work or not with rpi2zeroW
2021-12-29 16:11:13 <Tenkawa> Apachez:  na@zbox2:/lib/systemd$ uname -a ; dmesg | grep -i  Model
2021-12-29 16:11:13 <Tenkawa> Linux zbox2 5.15.11-zbox2 #1 SMP Thu Dec 23 12:27:01 EST 2021 aarch64 GNU/Linux
2021-12-29 16:11:13 <Tenkawa> [    0.000000] Machine model: Raspberry Pi Zero 2 W Rev 1.0
2021-12-29 16:11:24 <Apachez> the W models claims to support 64bit but all results are with only 32bit?  https://www.jeffgeerling.com/blog/2021/look-inside-raspberry-pi-zero-2-w-and-rp3a0-au
2021-12-29 16:11:28 <Tenkawa> Apachez: no question here
2021-12-29 16:11:34 <Apachez> well there ya go then :)
2021-12-29 16:16:33 <heat> Tenkawa, do you know where the firmware is kept inside the boot partition? I can't find an obvious linux-firmware package in the apks
2021-12-29 16:16:42 <heat> maybe I could just overlay it
2021-12-29 16:17:06 <Tenkawa> that I do not know yet (only being day 2 on alpine unfortunately)
2021-12-29 16:17:53 <Tenkawa> I'm a Debian, Devuan, Ubuntu, and Slackware person up until now
2021-12-29 16:37:31 <Apachez> slackware but no archlinux?
2021-12-29 16:40:43 <heat> is there a way to get persistent logs into the sd card?
2021-12-29 16:40:47 <Tenkawa> Apachez: nah.. I tried it out but wasn't that interested. I used Slackware in the mid 90's
2021-12-29 16:41:10 <heat> i have no serial so I'm relying on ssh for communication, and without wireless up that's kind of hard...
2021-12-29 16:42:04 <Tenkawa> heat: yeah I'm having that same fun.. no display here for me either… everything I do on all of my arm socs is by usb ttl or network
2021-12-29 16:43:42 <heat> Tenkawa, I don't have serial either :) so it's kind of tough to debug this
2021-12-29 16:43:58 <heat> the system looks like it's running but I have no way to know :D
2021-12-29 16:44:05 <heat> definitely no wireless though
2021-12-29 17:06:03 <minimal> heat: the linux-firmware-brcm package is where it typically would be found - but I don't see it there, looks like the upstream repo hasn't added it yet.
2021-12-29 17:14:50 <heat> is there a way to write to the boot partition or something, from an overlay script?
2021-12-29 17:15:53 <heat> i've tried editing the headless.start with a dmesg > ${ovlpath}/log but it doesn't seem work, I imagine this is running from ram
2021-12-29 17:35:39 <MY-R> heat: you need to first remount it to be abble to write to it, normally on diskless setup is enough to run: mount /media/mmcblk0p1 -o remount,rw
2021-12-29 18:28:31 <bogen85> alpine is making me disklike some software projects even more... (ones that I was already frustrated with...) (especially slow bloated and inefficient ones).. Install some package X on alpine... (And your total disk usage is increased 5x+...)
2021-12-29 18:29:01 <bogen85> (in my case python3 + ansible)
2021-12-29 18:32:27 <bogen85> looking at pyinfra... far less bloated than ansible, and works better with alpine for me. (pyinfra has no dependency on python on the target machine, and works with an alpine chroot for me)... (ansible does not work with an alpine chroot due to busybox symlinks).
2021-12-29 18:34:43 <bogen85> though "out of the box" pyinfra does have some issues with alpine... https://github.com/Fizzadar/pyinfra/issues/729
2021-12-29 18:35:52 <psykose> this sounds more like a dropbear issue since the PATH doesn't have sbin in it for whatever reason
2021-12-29 18:37:26 <bogen85> yeah, but also chroot connector for pyinfra does not work out of the box for alpine. (but neither does ansible's chroot connector, as ansible can't deal with the /bin/sh symlink)
2021-12-29 18:38:35 <bogen85> (works when the pyinfra code is patched to replace apk with /sbin/apk
2021-12-29 21:01:16 <kmmndr> Hi all :-)
2021-12-30 02:31:45 <bleb> anyone know of a guide for tunneling traffic from only a specific program through wireguard, on linux?
2021-12-30 02:32:08 <bleb> alpine
2021-12-30 02:32:44 <psykose> you can use network namespacing, though it takes some effort
2021-12-30 02:33:17 <psykose> you create a wireguard interface, move it into a netns, then netns exec the program in the netns, and it will go through wg only, everything else unaffected
2021-12-30 02:34:02 <bleb> I guess it takes effort because nobody has bothered to write down the commands?
2021-12-30 02:34:17 <psykose> they're written down in plenty of places
2021-12-30 02:34:34 <bleb> yeah that's probably true
2021-12-30 02:35:37 <bleb> publicly though?
2021-12-30 02:42:09 <durrendal> https://www.wireguard.com/netns/ looks like what psykose is referring to
2021-12-30 02:42:46 <psykose> yeah, that's basically all of it
2021-12-30 02:48:42 <bleb> this article has some iptables stuff https://www.aaflalo.me/2019/08/wireguard-and-torrent-on-linux/
2021-12-30 02:49:16 <bleb> says it's "to let use forward traffic into our virtual interfaces," is that not needed?
2021-12-30 02:59:13 <Hello71> only if you're doing it wrong
2021-12-30 02:59:38 <Hello71> it's easier to just configure the program to bind on the desired address/interface
2021-12-30 02:59:59 <Hello71> now that SO_BINDTODEVICE is unrestricted i should write a program to do that
2021-12-30 03:00:13 <Hello71> i guess bind would work too if the right ip rules are set up...
2021-12-30 03:04:25 <rwhalen> Hey again, I'm back with a new issue.. my sound doesn't seem to be working and I saw on the arch wiki that I'm probably missing some firmware. They mention the firmware is part of either sof-firmware or alsa-firmware. Is there any package on Alpine that might have this firmware?
2021-12-30 03:06:22 <rwhalen> I saw this package https://pkgs.alpinelinux.org/package/edge/testing/x86_64/sof-bin but I'm not on edge. I wasn't sure how to tell if the firmware is included either. If you know how I can tell that would be helpful for future reference too
2021-12-30 03:09:23 <Hello71> post dmesg
2021-12-30 03:09:52 <Hello71> the vague rule is that alsa-firmware is for very very old cards and sof-firmware is for very very new cards
2021-12-30 03:11:12 <rwhalen> dmesg: https://pastebin.com/gsmmVFpw
2021-12-30 03:12:46 <psykose> you do indeed want sof firmware it seems
2021-12-30 03:13:20 <psykose> it's in edge testing, sof-firmware
2021-12-30 03:13:37 <psykose> it doesn't depend on anything, i would say it's safe to prefix install it
2021-12-30 03:14:13 <rwhalen> oh okay.. I wasn't sure if I could or not. The wiki had a bunch of warnings. I can give it a shot.
2021-12-30 03:14:20 <Tenkawa> it has a very direct line in the message: 	1	[    2.752509] sof-audio-pci-intel-tgl 0000:00:1f.3: error: sof firmware file is missing, you might need to
2021-12-30 03:14:20 <Tenkawa> 	2	
2021-12-30 03:14:20 <Tenkawa> 	3	[    2.752509] sof-audio-pci-intel-tgl 0000:00:1f.3:        download it from https://github.com/thesofproject/sof-bin/
2021-12-30 03:14:23 <psykose> they are there for good reason
2021-12-30 03:14:31 <psykose> but this has no dependencies and is just firmware
2021-12-30 03:17:00 <rwhalen> Got it, I can give it a shot.
2021-12-30 03:17:34 <rwhalen> Also thanks again for helping with the qt patch yesterday. I checked today and I'm now browsing youtube crash-free :)
2021-12-30 03:17:36 <Hello71> well the binaries might not work
2021-12-30 03:17:49 <Hello71> er, the tools
2021-12-30 03:20:01 <rwhalen> I suppose that's okay as long as my card is detected. Right?
2021-12-30 03:20:53 <psykose> i would bet just the fw is fine
2021-12-30 03:21:00 <psykose> but i have not had such a .. fancy card :)
2021-12-30 03:21:36 <psykose> otherwise might as well go to edge i guess
2021-12-30 03:34:38 <bleb> Hello71: so the wireguard netns tutorial is how you'd do it?
2021-12-30 03:34:43 <bleb> or are you suggesting a third option
2021-12-30 03:35:10 <Hello71> there are a number of ways to do it, all of which are either annoying or insecure
2021-12-30 03:36:16 <Hello71> if your program supports binding to an interface, i recommend that. otherwise, the netns assignment (without veth) as documented on the wireguard website is probably the least worst way
2021-12-30 03:36:29 <bogen85> unsecure? not sure about insecure :D
2021-12-30 03:37:02 <Hello71> well one method is firejail --net=wg0
2021-12-30 03:37:23 <Hello71> but this method requires installing firejail which basically grants root to everyone on your box
2021-12-30 03:37:33 <Hello71> so i say it is not very secure
2021-12-30 03:38:45 <psykose> i think bwrap has a net option too
2021-12-30 03:44:47 <Hello71> yes, --unshare-net
2021-12-30 03:45:28 <Hello71> you would need to add slirp4netns, which would be secure but annoying and also slow
2021-12-30 03:46:04 <Hello71> if you just want to turn off the network, that's easy, just do unshare -cn
2021-12-30 04:07:31 <rwhalen> okay, so I installed sof-firmware and see the module under /usr/lib/firmware/intel/sof/sof-tgl.ri but I'm still seeing the dmesg errors. This is probably a rookie question but do I need to add a module for it to get detected? I tried a couple but I'm not sure how to tell which is right if that's what I'm missing.
2021-12-30 04:10:31 <psykose> it should get loaded correctly after you reboot
2021-12-30 04:11:41 <psykose> ah, it loads intel/sof/sof-tgl.ri
2021-12-30 04:11:54 <psykose> but the packaged ones are intel/sof-v1.92/
2021-12-30 04:12:23 <zcrayfish> my my, is it symlink time?
2021-12-30 04:13:09 <rwhalen> The symlink is there, yea. Not sure why it's not picked up
2021-12-30 04:19:20 <Hello71> did you reboot
2021-12-30 04:21:18 <rwhalen> yea... let me try it again as a sanity check though. maybe I've been staring at it for too long now. brb
2021-12-30 04:24:29 <rwhalen> okay.. rebooted. This is my new dmesg: https://pastebin.com/5xL8NNud
2021-12-30 04:26:39 <rwhalen> The error is the same as the old one and the firmware file is at /usr/lib/firmware/intel/sof/sof-tgl.ri
2021-12-30 04:30:41 <rwhalen> is the full search path configurable? it only shows the last part in dmesg so maybe I somehow misconfigured the prefix part
2021-12-30 04:36:45 <rwhalen> tried adding a link at /lib/firmware/intel/sof/sof-tgl.ri ... will reboot and see if that works. brb again
2021-12-30 04:47:32 <rwhalen> back.. that worked. I needed to link both /lib/firmware/intel/{sof,sof-tplg} and it finally picked it up. Is that a discrepancy with edge or should I try to send a patch to add those links?
2021-12-30 06:16:24 <lyubov> ./buffer 20
2021-12-30 08:39:06 <kmmndr> Hi all :-)
2021-12-30 08:39:20 <kmmndr> I'm having issues with qt5 application on gnome wayland
2021-12-30 08:39:43 <kmmndr> Whenever I start an app, the entire Gnome session crash
2021-12-30 08:40:14 <kmmndr> Has anyone similar issue ?
2021-12-30 08:43:42 <kmmndr> (It may be related to something between wayland and qt ? as I haven't had any problem with these application on i3wm and Xorg)
2021-12-30 09:13:47 <frag> does setup-keymap setup keymap for xorg too? can loadkmap work for both console/xorg?
2021-12-30 09:20:37 <markand> no it's only for console
2021-12-30 09:22:11 <markand> now it's even worse since all wayland compositors do have their own configuration to adjust keymap
2021-12-30 09:22:30 <markand> in the past you would call setxkbmap in your .xinitrc, now you have to find in individual wayland compositor on how to change the keyboard layout
2021-12-30 09:22:42 <markand> and same for video resolution, screen config, etc
2021-12-30 11:41:45 <frag> i uninstalled eudev :p managed to get net back manually and reinstall eudev but some stuff seem a bit off, is there an _easy_ way to fix it?
2021-12-30 11:42:18 <frag> eg some services dont work as they used to... i guess i have to manually fix em?
2021-12-30 11:42:26 <frag> prob not too much work tbh
2021-12-30 11:44:44 <psykose> setup-udev places them all in the correct place
2021-12-30 11:46:23 <frag> oh nice
2021-12-30 11:46:50 <frag> will reboot test
2021-12-30 11:48:58 <MY-R> damn, my msata small ssd is louder from old 2.5" HDD during read/write :P
2021-12-30 11:50:42 <frag> hmm didnt put everything back seems, eg have to do 'doas ip link set eth0 up;doas udhcpc' to get net, and power on bluetooth before it can connect to my bt keyboard
2021-12-30 11:55:15 <frag> maybe the bt thing was not there actually, i guess you set that in /etc/bluetooth/main.conf phaps
2021-12-30 11:57:20 <MY-R> frag: what do you have in "/etc/network/interfaces" ?
2021-12-30 11:59:57 <frag> MY-R: not sure if it was like this before i ran the two commands above| https://x0.at/FwCc.txt
2021-12-30 12:01:18 <MY-R> frag: looking good
2021-12-30 12:03:00 <MY-R> frag: anything other on screen when booting? something like "* Starting dhcpcd ... " ?
2021-12-30 12:05:18 <psykose> perhaps you disabled networking and bluetooth by accident
2021-12-30 12:05:48 <MY-R> rc-service networking status
2021-12-30 12:06:22 <frag> MY-R: oh.. no.. and when trying to start udhcpd i get(broken symlink to busybox) https://x0.at/hcKU.txt
2021-12-30 12:07:05 <frag> i think, ah i tried to make a new one to busybox, but doesnt seem to work
2021-12-30 12:07:07 <MY-R> frag: udhcpd is DHCP server and not client
2021-12-30 12:07:26 <psykose> you also don't start a udhcpc service
2021-12-30 12:07:28 <MY-R> remove udhcpd from services
2021-12-30 12:07:54 <frag> ok, i thought udhcpd was supposed to run
2021-12-30 12:07:57 <MY-R> networking will automaticly run udhcpc
2021-12-30 12:08:24 <frag> i dont think mine does now.. eth0 is not even 'up' at boot i think
2021-12-30 12:08:29 <MY-R> or anything else if installed different dhcp client
2021-12-30 12:11:40 <MY-R> frag: working?
2021-12-30 12:14:05 <frag> MY-R: well network/internet works now, since i ran the two commands above, not sure how to make them work at boot
2021-12-30 12:14:50 <MY-R> frag: what you have when run "rc-update"
2021-12-30 12:15:23 <frag> MY-R: https://x0.at/SF9N.txt
2021-12-30 12:15:54 <frag> love that overview btw :D
2021-12-30 12:16:08 <psykose> networking should be in default instead, but i don't think it makes much difference
2021-12-30 12:18:04 <MY-R> psykose: netowrking is by default in boot
2021-12-30 12:18:36 <MY-R> I would put bluetooth in default
2021-12-30 12:18:52 <frag> MY-R: will it get loaded before login?
2021-12-30 12:18:58 <frag> not a must but...
2021-12-30 12:19:06 <MY-R> frag: yes
2021-12-30 12:19:09 <frag> MY-R: okok
2021-12-30 12:19:26 <MY-R> login is at the end
2021-12-30 12:19:56 <MY-R> frag: so if you reboot now you wont have network?
2021-12-30 12:20:17 <MY-R> frag: enable rclog and send after /var/log/rc.log to paste service
2021-12-30 12:21:16 <MY-R> frag: in "/etc/rc.conf" change '#rc_logger="NO"' to 'rc_logger="YES"'
2021-12-30 12:22:06 <frag> MY-R: ok ill try
2021-12-30 12:26:41 <frag> MY-R: now it did work :S
2021-12-30 12:26:55 <frag> even bt, ah but i did edit that file to autoenable
2021-12-30 12:28:01 <MY-R> frag: so comment that line and enjoy, did you move bluetooth service to default or didnt touch it?
2021-12-30 12:28:32 <frag> MY-R: yes, moved to default :]
2021-12-30 12:28:44 <MY-R> frag: and isnt in boot, right?
2021-12-30 12:29:03 <frag> MY-R: mhmh
2021-12-30 12:29:19 <MY-R> rc-update to be sure, I sometimes forgeting add "boot" when removing stuff from there :P
2021-12-30 12:30:38 <frag> MY-R: yes, checks out ;)
2021-12-30 12:31:50 <MY-R> :)
2021-12-30 13:33:05 <frag> is there some special way to make rules for udev in alpine? i think something like 'ACTION=="add", KERNEL=="hidraw[0-9]*", RUN+="/usr/bin/xmodmap /1/s/xmodmap"' worked  on void
2021-12-30 13:33:30 <psykose> it's the same format with eudev running, as it's probably the same eudev
2021-12-30 13:33:44 <psykose> in the same directory for wherever people are meant to put .rules files
2021-12-30 13:35:21 <frag> im getting this in dmesg: [ 3897.893897] udevd[4831]: failed to execute '/lib/udev/xmodmap' 'xmodmap /1/s/xmodmap': No such file or directory
2021-12-30 13:35:55 <psykose> i dunnow what /1/s/xmodmap is, but you could try with something simple
2021-12-30 13:46:06 <frag> seems you have to restart udev, got rid of error, but did not implement change
2021-12-30 18:12:07 <jhjacobs81> is there any rust programmer inhere?
2021-12-30 18:12:35 <lopid> i don't know
2021-12-30 18:15:16 <jhjacobs81> im starting to learn rust, gonna write a web frontend for the alpine firewall, if i succeed :)
2021-12-30 18:44:21 <bleb> iptables?
2021-12-30 19:01:36 <jhjacobs81> no no
2021-12-30 19:02:02 <jhjacobs81> awall
2021-12-30 19:08:36 <incinia_> hi, I just installed Alpine (I'm a newbie) and everything went OK, no errors happened during installation, I chose "crypt" and "sys" during the partition part of the install. however after removing the installation media and booting, it isn't working properly.
2021-12-30 19:08:56 <incinia_> oops, meant to do a line break, wait for me to finish typing D:
2021-12-30 19:11:04 <incinia_> grub works and I boot into a mounting error screen (sorry I don't know the right terms), it says "mount: mounting UUID=(long string of characters here) on /sysroot failed: No such file or directory. Mounting root failed. initramfs emergency recovery shell launched. Type 'exit' to continue boot. sh: can't access tty: job control turned off
2021-12-30 19:11:40 <incinia_> on the wiki the install seems pretty simple so I don't understand what's going wrong, I didn't do anything out of the ordinary, just followed the guided installer
2021-12-30 19:12:27 <Hello71> i think it is broken with nvme
2021-12-30 19:12:52 <Hello71> what iso did you download
2021-12-30 19:12:59 <Hello71> and i guess also what disk type
2021-12-30 19:13:43 <incinia_> Standard x86_64, and I'm using nvme so that's probably the issue 
2021-12-30 19:20:59 <incinia_> are there currently any workarounds to get the installer to work with nvme drives?
2021-12-30 19:25:30 <Hello71> what happens if you enter modprobe nvme
2021-12-30 19:25:39 <Hello71> actually, lsmod | grep nvme, then modprobe nvme
2021-12-30 19:25:47 <Hello71> (in the emergency shell)
2021-12-30 19:29:45 <minimal> setup-disk does appear to handle nvme devices appropriately: https://gitlab.alpinelinux.org/alpine/alpine-conf/-/blob/master/setup-disk.in#L524
2021-12-30 19:35:44 <incinia_> "modeprobe: FATAL: Module nvme not found in directory /lib/modules/5.15.11-0-lts"
2021-12-30 19:35:55 <incinia_> modprobe*
2021-12-30 19:46:35 <Tenkawa> nvme worked fine on my hp omen  with no extra modules needed
2021-12-30 19:48:59 <Tenkawa> lsmod |  grep -i nvm
2021-12-30 19:48:59 <Tenkawa> nvme_core             131072  5 nvme
2021-12-30 19:48:59 <Tenkawa> nvme                   49152  3
2021-12-30 19:49:28 <Tenkawa> thats straight from the 3.15 install
2021-12-30 19:50:00 <Tenkawa>  /dev/nvme0n1p3          463820      9978    430209   2% /
2021-12-30 19:50:37 <Hello71> i mean https://lists.alpinelinux.org/~alpine/devel/%3CCGPE0WPWYYFK.1Z9XCTP3JYJWX%40nitro%3E
2021-12-30 19:50:56 <Hello71> ah, it's probably checking the mapped name
2021-12-30 19:52:14 <Tenkawa> likely yes.. 
2021-12-30 19:53:09 <Hello71> yeah
2021-12-30 19:53:17 <minimal> he mentioned installing with "crypt" so the rootfs reference will be for the LUKS /dev/mapper/ device
2021-12-30 20:16:57 <rwhalen> I had the same issue and ended up just removing encryption for now. Not a great option, but it's okay if you just want to get a working system
2021-12-30 20:34:53 <Hello71> all you need to do is add nvme to mkinitfs.conf
2021-12-30 20:35:57 <minimal> Hello71: it is added by setup-disk
2021-12-30 20:36:11 <Hello71> the point is that it isn't
2021-12-30 20:36:23 <Tenkawa> Hello71: staticly compiled in the kernel would do it too right?
2021-12-30 20:36:37 <Tenkawa> nvme and  nvme_core
2021-12-30 20:36:41 <minimal> https://gitlab.alpinelinux.org/alpine/alpine-conf/-/blob/master/setup-disk.in#L524
2021-12-30 20:36:44 <psykose> sure but it isn't in the alpine kernels
2021-12-30 20:38:30 <minimal> Hello71: or are you referring to the variable $rootdev being a "/dev/mapper/" reference and so not matching that "nvme" reference?
2021-12-30 20:40:15 <Tenkawa> psykose: will adding nvme in mkinitfs.conf implicitly pull in nvme_core too? 
2021-12-30 20:40:27 <psykose> yes
2021-12-30 20:40:32 <Tenkawa> (that was a q for all)
2021-12-30 20:40:36 <Tenkawa> ok good
2021-12-30 20:40:41 <psykose> you can see in mkinitfs/features.d/nvme.modules
2021-12-30 20:40:51 <psykose> and then see the things under that path in /lib/modules/kern/
2021-12-30 20:41:44 <psykose> that is all the features list does :)
2021-12-30 20:41:46 <Tenkawa> yeah it would need it (I'm much more use to statically defining things so I don't get to look at that code as much)
2021-12-30 20:42:19 <minimal> I assume the same problem must also occur with LVM on NVME
2021-12-30 20:43:12 <Tenkawa> minimal: I think its even been a legacy problem in debian too
2021-12-30 20:43:19 <Tenkawa> for all 3 scenarios
2021-12-30 20:44:14 <minimal> it looks more like the logic in setup-disk only adds nvme to the features when the root dev is /dev/nvme* but for both LVM and LUKS the root dev will be a /dev/mapper device
2021-12-30 20:48:28 <Hello71> yes, that's what i said an hour ago, except you're still wrong because 517 loops over $pvs
2021-12-30 20:50:32 <jhjacobs81> well.. that sucked..
2021-12-30 20:51:26 <minimal> Hello71: I'm happy to admit when I'm wrong - I didn't go through all the code with a toothcomb
2021-12-30 20:54:42 <jhjacobs81> anyone know why erlang-runtime-tools isnt found?
2021-12-30 20:56:23 <psykose> it's not packaged
2021-12-30 20:56:25 <psykose> in what context
2021-12-30 21:10:59 <bleb> recently shucked an old external hd, want to use it in a new zfs pool
2021-12-30 21:11:40 <bleb> but I thought it had some data on it and I can't seem to mount with zfs, ext2, ext3, ext4, exfat, vfat
2021-12-30 21:11:48 <bleb> so I'm worried that the drive is no good
2021-12-30 21:12:16 <bleb> any good tools to determine the type of filesystem or the health of the drive?
2021-12-30 21:12:30 <Tenkawa> bleb: does fdisk even show any partitions?
2021-12-30 21:12:48 <bleb> yes
2021-12-30 21:12:48 <bleb> /dev/sdd1   2048 7814035455 7814033408  3.6T FreeBSD ZFS
2021-12-30 21:12:59 <Tenkawa> oh a freebsd zfs 
2021-12-30 21:13:02 <Tenkawa> fun
2021-12-30 21:13:11 <bleb> is it fun
2021-12-30 21:13:25 <Tenkawa> there "might" be a fuse module… let me see
2021-12-30 21:13:35 <bleb> I thought zol was supposed to be cross compatible
2021-12-30 21:14:21 <bleb> thought I was being safe by creating the pool on freebsd where there are fewer zfs features supported
2021-12-30 21:19:37 <Tenkawa> did you also apk add zfs then modprobe zfs first?
2021-12-30 21:20:19 <Tenkawa> it "might" register that way
2021-12-30 21:20:49 <Tenkawa> I did get zfs to at least start up
2021-12-30 21:20:50 <Tenkawa> zfs list 
2021-12-30 21:20:50 <Tenkawa> no datasets available
2021-12-30 21:21:06 <bleb> well zpool import -d /dev/sdd1 is now taking a long time to say anything
2021-12-30 21:21:13 <Tenkawa> (I don't have any drives running it with any data atm)
2021-12-30 21:21:33 <bleb> ok yeah now it says tank is online
2021-12-30 21:21:53 <bleb> weird, I already had another pool mounted so the zfs module was already loaded
2021-12-30 21:22:08 <bleb> but modprobe zfs must have jiggled it?
2021-12-30 21:22:11 <Tenkawa> oh thats a hopefully good sign
2021-12-30 21:23:28 <bleb> yeah I was able to import it lol
2021-12-30 21:23:30 <bleb> so weird
2021-12-30 21:23:44 <Tenkawa> nice.. 
2021-12-30 21:24:16 <Tenkawa> bleb: hey I'll take weird and working over easy and broke...
2021-12-30 21:24:42 <Tenkawa> I've had too many "easy" procedures go wrong
2021-12-30 22:03:44 <bleb> regardless you've given me my life back
2021-12-30 22:03:46 <bleb> Katt Williams - Pimp Chronicles Part 1.avi
2021-12-30 22:04:45 <frag> why vim and similar use : instead of easier to punch ; ...
2021-12-30 22:05:58 <bleb> good question
2021-12-30 22:11:42 <lopid> because
2021-12-30 22:18:58 <earendel> probably because it's easier to punch.
2021-12-30 22:20:32 <earendel> how do you even write a : in vim? (:
2021-12-30 22:21:00 <earendel> ah insert mode.
2021-12-30 22:21:34 <earendel> hm. well. then forget what i said.
2021-12-30 22:22:33 <Tenkawa> bleb: np… I'm glad to help when I can. Going to be re-starting my Alpine journey tomorrow now that I've got my development machine ready to go :)
2021-12-31 00:38:29 <Nulo> What can I use to convert ePub to a MOBI file in Alpine
2021-12-31 00:40:44 <bdju> doesn't look like calibre or kindlegen are packaged
2021-12-31 00:52:41 <Nulo> kindlegen is apparently proprietary and deprecated
2021-12-31 00:53:08 <bdju> ah, I just did a quick search. didn't know it was proprietary
2021-12-31 00:53:31 <bdju> personally I prefer things to be epub so I have never converted anything to mobi
2021-12-31 00:54:08 <Nulo> I do, too, but my Kindle only supports mobi!
2021-12-31 00:54:26 <bdju> a shame :(
2021-12-31 01:04:40 <bleb> frag: looks like : may have been easier to type when vi was written: https://static.giffgaff.com/content/iocms/LSI_ADM_3_A_full_keyboard_e29a8bb405.jpg
2021-12-31 01:05:01 <bleb> I'm guessing you would type it the same way you would type - on a modern keyboard
2021-12-31 01:05:56 <bleb> still not as easy as ; but maybe they wanted to avoid accidents
2021-12-31 01:06:16 <bleb> (also note the position of the esc key)
2021-12-31 06:55:50 <cha2axa> hi, I'm trying to install Alpine via nvme with crypt (using guided installer) but when it installs and boots it puts me in an emergency shell. I asked about this a few hours ago but didn't have the time to try anything. I just went ahead and reinstalled, made sure NVME was in mkinitfs and rebooted, but it still put me in an emergency shell. I'm not sure how to get it working exactly, could anyone help walk me through what I'm meant to do? (
2021-12-31 06:56:30 <cha2axa> related: https://lists.alpinelinux.org/~alpine/devel/%3CCGPE0WPWYYFK.1Z9XCTP3JYJWX%40nitro%3E | https://gitlab.alpinelinux.org/alpine/mkinitfs/-/merge_requests/73
2021-12-31 06:57:37 <cha2axa> I also tried doing this: https://wiki.alpinelinux.org/wiki/NVME
2021-12-31 07:00:07 <cha2axa> when I ran the command in the link above, initramfs complained that a file didn't exist (or something like that) and aborted, but nvme was still listed in mkinitfs -L
2021-12-31 08:55:12 <frag> bleb: interesting  :D
2021-12-31 09:12:34 <Feuermagier> I'd like to install geckodriver as a requirement for my dockercontainer. However it only seems available in the edge/testing repository (https://pkgs.alpinelinux.org/packages?name=geckodriver). How would I install that, if my container is based on alpine-3.15?
2021-12-31 09:14:14 <ikke> You would have to build it for 3.15
2021-12-31 09:15:38 <Feuermagier> ikke, can I not somehow "hack-pull" it? - If I build it myself, I'll have to build it for every CPU-arch (and that comes with additional problems when it comes to multi-arch-containers)
2021-12-31 09:15:58 <ikke> You can, but it can break at any moment (if it's working in the first place)
2021-12-31 09:16:39 <Feuermagier> if a package is in the edge/testing, can I expect it to be in the next release version of alpine?
2021-12-31 09:16:50 <ikke> No, it would have to be moved to community at least
2021-12-31 09:17:14 <ikke> Nulo maintains it
2021-12-31 09:22:46 <Feuermagier> ah, interesting. As I'm building for docker and pushing the entire container after build I can simply try it out and see if it breaks. How would I install "only that package" from edge?
2021-12-31 09:23:51 <ikke> add '@edge_testing https://dl-cdn.alpinelinux.org/alpine/edge' to /etc/apk/repositories and then apk add geckodriver@edge_testing
2021-12-31 09:24:29 <Feuermagier> on the other hand I could also pull myself the edge-base-container and install python3 into that (instead of using python:3-alpine3.15)
2021-12-31 09:24:58 <Feuermagier> would you recommend this method over the other?
2021-12-31 09:31:46 <ikke> Though edge might be unstable at times, issues should eventually get fixed, so I think it has more chance to keep working on the long run, especially if you can keep using existing docker images
2021-12-31 09:36:02 <Feuermagier> ikke, WARNING: The repository tag for world dependency 'geckodriver@edge_testing' does not exist - should I add --force-broken-world, or is this a different kind of issue (running on alpine:edge)
2021-12-31 09:37:17 <ikke> Did you add the testing repository?
2021-12-31 09:37:32 <ikke> fyi, if you are using edge, you can leave out the pin (@edge_testing)
2021-12-31 09:38:01 <Feuermagier> Do I need to add the testing repo on edge?
2021-12-31 09:38:05 <ikke> yes
2021-12-31 09:38:12 <ikke> only main and community are enabled by default
2021-12-31 09:38:22 <Feuermagier> ah, alright
2021-12-31 09:39:01 <Feuermagier> "echo "@edge_testing https://dl-cdn.alpinelinux.org/alpine/edge" >> /etc/apk/repositories" this what you had in mind?
2021-12-31 09:39:18 <ikke> if you use edge, leave out @edge_testing
2021-12-31 09:39:58 <Feuermagier> excellent
2021-12-31 09:50:09 <Feuermagier> nothing like when your docker build aborts with "don't know how to compile Fortran code on platform 'posix'   Traceback (most recent call last):"...
2021-12-31 09:50:34 <ikke> heh
2021-12-31 10:24:12 <Feuermagier> ikke, ah, apparently it needs to build numpy completely from source, as there is no package for it in python3.10 - any way I can use python3.9 on edge?
2021-12-31 10:24:31 <ikke> No
2021-12-31 10:24:58 <ikke> py3-numpy should be available
2021-12-31 10:25:07 <ikke> https://pkgs.alpinelinux.org/contents?branch=edge&name=py3-numpy&arch=aarch64&repo=community
2021-12-31 10:26:43 <Feuermagier> ikke, ah, that is a *system* package! - I guess it tried to acquire it via pip
2021-12-31 10:27:16 <ikke> For packages with native extensions, it's often better to get them from apk
2021-12-31 10:27:48 <ikke> I think things are improving, but many packages don't have wheels for musl, so if you install them with pip, it will try to compile them
2021-12-31 10:28:30 <Feuermagier> what exactly is a "wheel" - it came up in the logs
2021-12-31 10:28:34 <ikke> That's what I do for netbox for exmaple, everything with native extensions I just install with apk, and update the dependencies to make sure it will use those
2021-12-31 10:28:49 <ikke> A wheel is the archive/package format for python
2021-12-31 11:27:07 <Feuermagier> ikke, apparently opencv just doesn't provide wheels for alpine :(
2021-12-31 12:50:53 <Nulo> Hi yes, I maintain geckodriver. I just recently packaged it because my coop. might use it Soon :tm: for browser testing and figured it was simple enough
2021-12-31 14:34:54 <frag> what process/app made this? its close to 5G /tmp/magick-SX8zS6Soz_mp2mdN8unlmh8dOKglaNjh
2021-12-31 14:35:02 <frag> image magick?
2021-12-31 14:47:57 <Tenkawa> frag: is it still running/growing?
2021-12-31 14:49:32 <Tenkawa> if so run a sudo fuser -cu /tmp/magick-SX8zS6Soz_mp2mdN8unlmh8dOKglaNjh
2021-12-31 14:49:34 <frag> Tenkawa: not growing
2021-12-31 14:49:54 <Tenkawa> it will show you the process that owns it
2021-12-31 14:49:56 <frag> Tenkawa: its doas now :p
2021-12-31 14:50:21 <Tenkawa> also lsof | grep magick-SX8zS6Soz_mp2mdN8unlmh8dOKglaNjh
2021-12-31 14:50:50 <Tenkawa> (if you have the utilities installed)
2021-12-31 14:51:20 <frag> 181 processees...
2021-12-31 14:52:01 <frag> xinit st ...
2021-12-31 14:53:09 <frag> firefox mpv busybox ...
2021-12-31 14:53:30 <frag> fuser -cu didnt work on busybox fuser, so used -m instead
2021-12-31 14:54:00 <Tenkawa> and yes its imagemagick
2021-12-31 14:54:14 <Tenkawa> do you use pimcore by chance?
2021-12-31 14:54:39 <Tenkawa> or a lot of php?
2021-12-31 14:54:56 <frag> Tenkawa: not that i know of
2021-12-31 14:55:53 <Tenkawa> wordpress and others apparently cause imagemagick to do this to
2021-12-31 14:56:04 <Tenkawa> I'm getting tons of hits on this problem
2021-12-31 14:56:14 <frag> Tenkawa: surfing wordpress sites? (many sites..)
2021-12-31 14:56:31 <frag> weird stuff..
2021-12-31 14:56:55 <frag> prob firefox eh..
2021-12-31 14:57:08 <Tenkawa> nah it was for plugins
2021-12-31 14:57:18 <Tenkawa> scripting stuff
2021-12-31 14:57:34 <Tenkawa> most people appear to be fixing with cron jobs
2021-12-31 14:57:39 <frag> i only have ublock
2021-12-31 14:57:42 <Tenkawa> just to clean up on timers
2021-12-31 14:59:08 <Tenkawa> here's one bug report for example. See if it helps you. : https://github.com/ImageMagick/ImageMagick/issues/395
2021-12-31 15:00:14 <frag> finding problem description dating back years hehe
2021-12-31 15:00:20 <frag> *that feeling*
2021-12-31 15:00:45 <Tenkawa> notice though the discussion continues for years lol
2021-12-31 15:04:21 <frag> i think ill just delte and see if it reappears
2021-12-31 15:04:35 <Tenkawa> heheh
2021-12-31 15:04:56 <Tenkawa> good luck.. hopefully it will behave
2021-12-31 18:50:11 <tiotags> happy new years to everybody !
2021-12-31 19:02:24 <Ermine> Happy new year!
2021-12-31 20:13:01 <frag> can one add an extra partition to the live usb? i dont understand "StartCHS    EndCHS"
2021-12-31 20:13:32 <psykose> afaik that is tricky for an .iso vs a .img
2021-12-31 20:14:44 <frag> okok, i thought i remembered something like that ... could be convenient tho
2021-12-31 23:27:44 <Apachez> happy new year!
2021-12-31 23:29:15 <Misthios> U2