2023-09-01 11:49:12 <af> Hi, does apk keep track somewhere of whether a package was installed explicitly or as a dependency?
2023-09-01 11:49:47 <ikke> af: everything installed explicitly is listed in /etc/apk/world
2023-09-01 11:50:09 <af> perfect, thank you
2023-09-01 12:01:12 <blt[m]1> do we have raspi-config in the raspberrypi Alpine version ? I can't find it, I would like to disable the serial for console output?
2023-09-01 12:08:06 <blt[m]1> enable_uart=1 in config.txt seems to be close but not sure it disable the login shell over serial, I would like the serial port hardware to be enabled but no login shell over serial
2023-09-01 12:11:36 <ikke> blt[m]1: Isn't it a matter of modifying /etc/inittab then? 
2023-09-01 12:13:19 <blt[m]1> ikke: thanks, it seems that #ttyS0::respawn:/sbin/getty -L ttyS0 115200 vt100 is commented but I see console [ttyS0] enabled in my dmesg
2023-09-01 12:42:41 <dwfreed> blt[m]1: console != getty
2023-09-01 12:43:00 <dwfreed> console refers to where the kernel might right messages, and what device owns /dev/console
2023-09-01 12:43:04 <dwfreed> s/right/write/
2023-09-01 12:44:04 <dwfreed> kernel will printk to all consoles, and the last console= on the kernel command line is the one that becomes /dev/console
2023-09-01 12:44:40 <ikke> but getty is responsible for providing a login promt, right?
2023-09-01 12:44:45 <dwfreed> yeah
2023-09-01 12:45:01 <dwfreed> so if you're not starting a getty on the tty device, you're not going to get a login prompt
2023-09-01 13:39:13 <minimal> so it's a matter of what "console=" entries are present on cmdline (from cmdline.txt file in boot partition). Does it specify "console=ttyAMA0"? (ttyS0 is a a *pc* serial port, ttyAMA0 is an Arm serial port))
2023-09-01 15:39:50 <juan_> Hi. Wanted to check before filing an issue to see if I'm not missing something obvious
2023-09-01 15:40:38 <juan_> Been trying to use biber, but I keep getting a Perl error about mismatched versions
2023-09-01 15:41:23 <juan_> It seems to me that it means by perl binary was compiled with different settings than the dynamic libraries biber needs
2023-09-01 15:41:36 <juan_> I guess that is something that should be fixed on the distro level
2023-09-01 15:41:40 <juan_> Am I correct?
2023-09-01 15:46:38 <ikke> If you're not mixing different releases, then probably 
2023-09-01 15:49:17 <juan_> I don't think I am
2023-09-01 15:49:42 <juan_> Ok. Is the project where I'm supposed to create the issue the aports one?
2023-09-01 15:51:07 <ikke> yes
2023-09-01 15:51:45 <juan_> Ok. Thanks.
2023-09-01 17:39:41 <qaqland[m]> In the pkgs.alpinelinux.org, there is a build date
2023-09-01 17:39:41 <qaqland[m]> Does the time equal aport commit time?
2023-09-01 17:40:27 <ikke> No, to when the builder actually built the package
2023-09-01 17:40:35 <qaqland[m]> 🤔
2023-09-01 17:41:05 <qaqland[m]> Almost same time (maybe minutes later)?
2023-09-01 17:41:50 <ikke> can also be longer if the builder has a lot of packages to build or is stuck for some reason
2023-09-01 17:42:16 <qaqland[m]> No new commit == no new build ? (If success)
2023-09-01 17:43:05 <ikke> it checks if a package with the same pkgver and pkgrel already exists
2023-09-01 17:43:15 <ikke> (same for subpackages)
2023-09-01 17:44:34 <qaqland[m]> So we need change plgver or pkgrel +1 to make build
2023-09-01 17:44:36 <qaqland[m]> Got it, thank you 🥰
2023-09-01 17:44:42 <ikke> Yes
2023-09-01 17:44:45 <ikke> And I must correct myself
2023-09-01 17:44:59 <ikke> to facilitate reproducable builds, the build date has been set equal to the commit date
2023-09-01 17:45:30 <ikke> One clue is that a package has the same build dates for all architectures
2023-09-01 18:48:31 <whynothugo> Oh, that's neat.
2023-09-01 18:48:58 <whynothugo> Is there a database of package checksums? Can I build them locally in a chroot and compare which ones yield the same results?
2023-09-01 18:54:00 <Habbie> whynothugo, i think an APKINDEX is such a database
2023-09-01 18:56:45 <whynothugo> I like that I can just gunzip -c APKINDEX.00163840.tar.gz it and it's nice readable text.
2023-09-01 18:57:29 <whynothugo> Ah, the fields are right here: https://wiki.alpinelinux.org/wiki/Apk_spec#APKINDEX_Format
2023-09-01 19:18:16 <bosco> Is it possible to scroll up on the tty? S-PgUp isn't working
2023-09-01 19:19:56 <ovf> bosco: not since 5.9: https://www.phoronix.com/news/Linux-5.9-Drops-Soft-Scrollback
2023-09-01 19:21:32 <bosco> Thanks
2023-09-01 19:22:27 <bosco> My system seems to be kinda bricked and I can't get the first (out of many) error message, is there a way I can see it?
2023-09-02 05:30:58 <codebam> so I'm still having an issue where iwd starts after networking and I don't get a dhcp address. does anyone know how to fix that?
2023-09-02 05:31:19 <codebam> they're both in boot. should I move networking into default so that it starts later?
2023-09-02 05:32:29 <codebam> or should I add iwd to the depend after section of the networking init service?
2023-09-02 05:33:54 <codebam> iwd has before net but it doesn't seem to work
2023-09-02 05:37:49 <p1fk6ohrq9> how to check if ntpd is working properly?
2023-09-02 06:02:27 <zcrayfish> p1fk6ohrq9: which ntpd? 
2023-09-02 06:04:13 <zcrayfish> codebam: the default rc service for iwd should start before networking... Unless... rc_parallel="YES" in /etc/rc.conf 
2023-09-02 06:32:07 <p1fk6ohrq9> zcrayfish: busybox ntpd
2023-09-02 06:33:47 <zcrayfish> No great ways to do it; but it's not impossible
2023-09-02 06:33:51 <zcrayfish> https://serverfault.com/questions/845814/is-it-possible-to-check-ntpd-status-on-a-busybox-system <-- for instance
2023-09-02 06:34:27 <p1fk6ohrq9> thanks. will look into it.
2023-09-02 06:37:41 <blt[m]1> maybe I missed some news but I don't see influxdb anymore on Alpine 3.18 ?
2023-09-02 06:40:10 <blt[m]1> <minimal> "so it's a matter of what "..." <- "root=/dev/vg0/lv_root modules=sd-mod,usb-storage,ext4 quiet rootfstype=ext4" in my cmdline.txt
2023-09-02 06:41:55 <zcrayfish> blt[m]1: I guess it hasn't had a maintainer since 3.12 or so, so it was dropped in this release. :/
2023-09-02 06:54:48 <codebam> it's set to no, is there anything else I can check?
2023-09-02 06:56:42 <codebam> maybe it's starting but not connected before networking comes up to do dhcp?
2023-09-02 06:58:29 <zcrayfish> dunno
2023-09-02 06:59:28 <codebam> can I put a delay on networking starting?
2023-09-02 07:00:11 <codebam> how would I do that. do I just do 'sleep &&' in the service?
2023-09-02 07:01:11 <codebam> I guess I could also just switch back to wpa_supplicant
2023-09-02 07:03:42 <zcrayfish> are you using NetworkManager also?
2023-09-02 07:03:47 <zcrayfish> just curious
2023-09-02 07:05:30 <codebam2> I enabled the builtin dhcp client in iwd and that seems to work. it got an ip after restarting iwd
2023-09-02 07:07:56 <zcrayfish> sweet
2023-09-02 07:36:19 <ahmadraniri[m]> Hello. Is it just me or some lxqt tools (lximage-qt, pcmanfm-qt) refuse to launch? I'm using stable alpine right now. Anyone willing to confirm ?
2023-09-02 07:42:57 <zcrayfish> I use edge, but those all ran fine for me on 3.18.2
2023-09-02 07:45:12 <zcrayfish> If one of your terminals still opens, lximage-qt stays in the foreground and prints warnings and errors; could be useful.
2023-09-02 07:47:10 <ikke> ahmadraniri[m]: do you  get an error message?
2023-09-02 07:50:16 <ahmadraniri[m]> ikke: No at all, unfortunately. When I run it from terminal, nothing show up. I also check it using gdb but I got nothing. Both on Wayland and xorg.
2023-09-02 07:50:48 <ahmadraniri[m]> It's strange, indeed.
2023-09-02 07:52:07 <ikke> how about strace?
2023-09-02 07:53:37 <ahmadraniri[m]> Strace from gdb : No default breakpoint address now.
2023-09-02 08:34:06 <ahmadraniri[m]> It seems fixed by using dbus-run-session. Thanks.
2023-09-02 08:49:21 <zcrayfish> speaking of lxqt... latest labwc's got lxqt-panel not drawing on top of full-screen programs any more. Makes games a lot more enjoyable
2023-09-02 12:42:45 <ascensionwires> why if i do lsblk it prints to me sdb 0B? what the hell is this disk?
2023-09-02 12:43:56 <ascensionwires> or its bot disk
2023-09-02 12:44:10 <ascensionwires> i have no idea why it displaying
2023-09-02 13:19:31 <qaqland[m]> i'm looking into apkindex "p:", i know cmd is exec but what is py3.11
2023-09-02 13:19:41 <qaqland[m]> example: cmd:apparmor_status=3.1.6-r1 py3.11:apparmor=3.1.6-r1
2023-09-02 13:19:53 <qaqland[m]> from here https://paste.opensuse.org/pastes/820382b2b7c0
2023-09-02 13:20:57 <ikke> qaqland[m]: the idea is that you can depend / pull in a python project by upstream name rather than by package name
2023-09-02 13:21:25 <ikke> qaqland[m]: https://gitlab.alpinelinux.org/alpine/abuild/-/commit/ac3ee42458ebb6204c75135cbeb30201777e4116
2023-09-02 16:16:38 <oleovit> hi. i enabled: net.core.default_qdisc=fq and net.ipv4.tcp_congestion_control=bbr . ss -tin shows bbr is in use, but qdisc for wlan0 and lo = noqueue, not fq (eth0 is not in use on my laptop) i can force fq with a tc-command, but why are these kernel parameters not enough? i had no idea about this topic before yesterday, so sorry.. . this is my last question, after that, i am done with experiments.
2023-09-02 16:54:26 <oleovit> ah, i guess it's intel: https://bbs.archlinux.org/viewtopic.php?pid=1852605#p1852605 . 
2023-09-02 17:25:52 <qaqland[m]> ikke: thx~
2023-09-02 17:27:41 <qaqland[m]> eg. I can depend on py3.11:torch not pytorch😄
2023-09-02 17:28:37 <ikke> Yes, but when python 3.12 is pushed, the dependencies would need to be updated as well
2023-09-02 17:29:14 <frag> anyone got qmk working? cant flash my nyquist, "no bootloader found"
2023-09-02 17:30:07 <whynothugo> frag: I have it working, but might be fuzzy in the details
2023-09-02 17:31:21 <frag> whynothugo: fuzzy? i got a "minor problem": ⚠ We do not recommend avr-gcc newer than 8. Downgrading to 8.x is recommended.
2023-09-02 17:31:53 <frag> when doing 'qmk setup'
2023-09-02 17:33:43 <whynothugo> frag: What exactly command fails with "no bootloader found"
2023-09-02 17:34:17 <whynothugo> I'm using `qmk_flash` and provide a json generated with qmk-configurator
2023-09-02 17:34:56 <frag> whynothugo:  qmk flash -kb keebio/nyquist/rev3 -km default
2023-09-02 17:35:13 <frag> but it keeps trying to find it forever..
2023-09-02 17:35:36 <frag> and im in RESET mode with the nyquist
2023-09-02 17:36:13 <frag> i dont even have qmk_flash... did you not install it from repo (testing) whynothugo ?
2023-09-02 17:55:16 <frag> how do you restart udev? with rc-service?
2023-09-02 17:56:31 <ikke> Yes, there are multiple udev related services
2023-09-02 17:57:17 <ikke> https://tpaste.us/z5n5
2023-09-02 17:58:39 <frag> ikke: oh, need to reload /etc/udev/rules.d/
2023-09-02 17:59:04 <ikke> afaik, udev automatically does that
2023-09-02 17:59:10 <ikke> otherwise there is udevadm
2023-09-02 17:59:36 <ikke> udev control -R
2023-09-02 18:03:23 <frag> ikke: dont have the udev command, but tried udevadm, didnt work... i copyied something there after 'qmk setup' complained you see.. and it stopped complaining (so thought maube i needed to restart udev)
2023-09-02 18:03:39 <frag> but still cant flash so i guess its not that ..
2023-09-02 18:17:12 <caskd[m]> instead of restarting udev try restarting udev-trigger
2023-09-02 18:17:33 <caskd[m]> it does essentially the equivalent of a restart
2023-09-02 18:17:44 <caskd[m]> s/udev/udevd/
2023-09-02 18:22:06 <frag> caskd[m]: hmm get the same result
2023-09-02 18:44:20 <linearcannon> has anyone used firefox recently on ppc64le
2023-09-02 18:46:43 <linearcannon> i get a message about "The bookmarks and history system will not be functional because one of Firefox's files is in use by another application." it links to a page that gives advice on recreating those files or using a new profile, none of it helps, i can completely nuke all firefox's data folders and it doesn't change anything
2023-09-02 19:05:15 <codebam> so I set up bluetooth and installed the bluetooth firmware but when I scan I can't find any devices?
2023-09-02 19:06:04 <codebam> dmesg says this when I'm scanning https://pastebin.seanbehan.ca/0BKjQefcBO.txt
2023-09-02 19:06:29 <ascensionwires> why setup-devd not purging other (unused) device managers?
2023-09-02 19:07:15 <elly> why would it? setup scripts generally don't remove other packages like that
2023-09-02 19:07:27 <elly> you can remove them yourself if you'd like
2023-09-02 19:09:32 <ascensionwires> like as i understand i can use setup-devd when i want switch to other device manager?
2023-09-02 19:09:50 <ascensionwires> or its wrong usage?
2023-09-02 19:10:59 <elly> I would not do that, those setup scripts are intended for use during installation
2023-09-02 19:11:17 <elly> if I wanted to switch device managers, I would use apk to install the new one and configure it normally
2023-09-02 19:11:39 <caskd[m]> setup-devd does exactly what it's called, it only sets up a device manager, it won't migrate or reconfigure other ones
2023-09-02 19:12:26 <caskd[m]> it would be called manage-devd otherwise
2023-09-02 19:12:32 <caskd[m]> or similarly
2023-09-02 19:13:45 <codebam> I used rfkill to block and then unblock it and it worked
2023-09-02 19:14:09 <ascensionwires> but why is it on alpine's sway guide?
2023-09-02 19:14:38 <elly> those guides are generally written assuming you are doing initial system setup
2023-09-02 19:16:00 <ascensionwires> i underatand but why they dont do it as you said before?
2023-09-02 19:16:34 <elly> you'd have to ask the author, I don't know
2023-09-02 19:16:49 <ascensionwires> also eudev is dead?
2023-09-02 19:17:08 <ascensionwires> or alpine dont use eudev?
2023-09-02 19:17:11 <caskd> no, eudev is still the main implementation 
2023-09-02 19:18:45 <caskd> it's just that the e prefix is left out as there's no other implementation compatible with alpine at the moment, so it's implicitly assumed that udev = eudev in such context
2023-09-02 19:20:28 <elly> I think the default is still mdev but the only real options on alpine are mdev, mdevd, and eudev
2023-09-02 19:20:38 <elly> (installed as /sbin/udevd as caskd says)
2023-09-02 19:21:22 <caskd> correct
2023-09-02 19:22:10 <ascensionwires> why gentoo say eudev is dead, and use systemd udev even in openrc, who knows?
2023-09-02 19:22:52 <caskd> i am not sure if udev works without systemd but maybe they just patch the upstream instead of maintaining a fork nowadays
2023-09-02 19:23:34 <caskd> so in that case the project itself is "dead" because there's no fork being maintained, just a set of patches
2023-09-02 19:23:58 <ascensionwires> on alpine its not "dead"?
2023-09-02 19:24:22 <ascensionwires> or better using mdev+libudev-zero
2023-09-02 19:24:47 <elly> I don't know what you are talking about?
2023-09-02 19:24:51 <elly> https://github.com/eudev-project/eudev eudev is actively developed
2023-09-02 19:24:57 <elly> it is no longer *part of gentoo*
2023-09-02 19:25:35 <elly> eudev works just fine
2023-09-02 19:25:42 <caskd> libudev-zero is sadly still missing a lot of features, so if you have things relying on udev internals then i'd recommend sticking with udev or using mdev(d) otherwise, it can only fill a niche right now
2023-09-02 19:25:59 <elly> you could try using mdev + libudev-zero if you want to, but in the words of the libudev-zero maintainers, "Many functions and interfaces still aren't implemented, which may lead to breakage in some programs."
2023-09-02 19:26:06 <elly> so I'd just use eudev
2023-09-02 19:26:22 <caskd> a good example i met myself is multipathd
2023-09-02 19:26:35 <caskd> not gonna work with -zero due to missing properties
2023-09-02 19:33:38 <ascensionwires> somebody say now udev in gentoo faster then eudev (idk how they do udev with openrc "independently of systemd" as they say) 
2023-09-02 19:35:04 <zcrayfish> We have #alpine-offtopic to talk about Gentoo. 
2023-09-02 19:35:31 <ascensionwires> ok 
2023-09-02 20:23:20 <elly> (but also I would just ask the gentoo folks directly)
2023-09-02 22:00:29 <tree3> Morning.
2023-09-02 22:01:21 <tree3> I recently found out that Solaris has direct binding, and now I'm shocked, _shocked_ I tell you, that Alpine doesn't. I mean I know it needs musl support but why doesn't musl support it? It just seems like the right thing.
2023-09-02 22:25:47 <Tach> Hi all, why is alpine building packages using r1 and r2 for the same release ?
2023-09-02 23:01:19 <frag> is there a way to insert unicodes on alpine? (like ctrl+shift+u+code on some systems)
2023-09-02 23:04:15 <ptrc> i'm pretty sure that depends on whatever context you mean
2023-09-02 23:04:29 <ptrc> as in, terminal ( which? ), desktop environment, tty?
2023-09-02 23:06:04 <frag> oh .. st, ratpoison
2023-09-02 23:06:21 <frag> qmk supports inserting unicode now... guess it wont work great tho
2023-09-02 23:06:30 <frag> too bad hehe
2023-09-03 00:59:33 <quinq> frag, st supports ISO 14755
2023-09-03 06:09:29 <frag> quinq: what does that mean? ctrl+shift+code should work?
2023-09-03 06:45:04 <quinq> frag, actually it's been moved out to being a patch, so you'll have to add it back yourself
2023-09-03 06:45:53 <quinq> But the default key binding is Ctrl+Shift+i
2023-09-03 06:46:40 <frag> ah, seems it will not work everywhere anyway ...
2023-09-03 06:47:55 <quinq> how not?
2023-09-03 06:48:51 <quinq> Ah, you mean you want a global way of doing that
2023-09-03 06:49:28 <quinq> That's easy to do, you can write a couple-line script with dmenu and printf
2023-09-03 06:50:50 <quinq> Well, the only “complex” part is to output the codepoint to somewhere, could be the selection/clipboard
2023-09-03 06:53:22 <quinq> Though there must be tools that prints to the currently focused window
2023-09-03 06:55:26 <frag> was hoping you could program the keymap with qmk so you dont need any xkb config
2023-09-03 06:56:58 <quinq> You wouldn't
2023-09-03 08:35:05 <ikke> linearcannon: maybe you can open an issue about it?
2023-09-03 10:11:37 <ascensionwires> yesterday there said me the setup-devd bad way to switching between device managers
2023-09-03 10:24:14 <ascensionwires> is some way to choose device manager during setup-alpine?
2023-09-03 10:24:40 <ascensionwires> or better do it manually after installing
2023-09-03 10:26:13 <ikke> DEVDOPTS="-C <manager>" setup-alpine
2023-09-03 10:27:04 <ascensionwires> if i do installation wierd
2023-09-03 10:27:18 <ascensionwires> like do "diskless" install
2023-09-03 10:27:33 <Apachez> "ramless" is fun
2023-09-03 10:27:50 <ascensionwires> after partitioning and mount to /mnt
2023-09-03 10:28:27 <ascensionwires> then `BOOTLOADER=grub USE_EFI=1 setup-disk /mnt`
2023-09-03 10:31:16 <ascensionwires> i need specify this DEVOPTS="-C <manager> when setup-disk right?
2023-09-03 10:31:30 <ikke> no, setup-alpine
2023-09-03 10:31:58 <ikke> It's setup-alpine that reads that environment variable, not setup-disk
2023-09-03 10:33:24 <ascensionwires> after all if i want change device manager after install, what is my steps?
2023-09-03 10:35:03 <ikke> If I look at setup-devd, it allows you to switch device managers, it just does not uninstall the other ones
2023-09-03 10:35:29 <ikke> https://gitlab.alpinelinux.org/alpine/alpine-conf/-/blob/master/setup-devd.in#L122-L124
2023-09-03 10:36:56 <ikke> so run setup-devd to switch to your desired manager, and afterwards, uninstall the other one you don't want anymore (although mdev is part of busybox, so you you can't uninstall that)
2023-09-03 10:42:15 <ascensionwires> why no guide how do it on wiki?
2023-09-03 10:44:33 <ikke> Because no one wrote it
2023-09-03 10:48:41 <Apachez> are we playing mindgames now? ;)
2023-09-03 11:00:51 <ikke> No, someone would need to take it up to write it
2023-09-03 11:09:58 <ascensionwires>  why in setup-devd its `rc-update delete --quiet --quiet hwdrivers sysinit || :` 
2023-09-03 11:10:23 <ascensionwires> then rc-update delete --quiet --quiet mdev sysinit || :
2023-09-03 11:10:53 <ikke> remove them from the sysinit runlevel and don't fail if they were never in there
2023-09-03 11:26:54 <dwfreed> scripts often run with set -e, which means if any command fails, terminate immediately; || : is one of the fastest (but more obtuse) ways to ignore a command's exit status when running with set -e
2023-09-03 11:27:19 <dwfreed> : is a do-nothing command, akin to true
2023-09-03 11:27:48 <ascensionwires> what i need if switching from mdev to eudev? https://pastebin.com/sjJn09mV
2023-09-03 11:28:23 <ascensionwires> setup-devd do sonething coldplug, i have no idea what is this
2023-09-03 11:28:35 <ascensionwires> setup-devd do sonething coldplug, i have no idea what is this
2023-09-03 11:28:43 <ikke> applying the policies to alreayd existing devices
2023-09-03 11:29:04 <ikke> hotplug is about detecting new devices  
2023-09-03 11:34:47 <ascensionwires> as i understand all i need to switch device managers is just running `setup-devd <manager>`
2023-09-03 11:36:03 <ikke> yes
2023-09-03 11:37:17 <ahmadraniri[m]> Is ash really underrated as daily drive shell unlike bash or zsh ?
2023-09-03 11:37:40 <ikke> The only thing I personally miss in ash is alt+<dot>
2023-09-03 11:37:45 <ikke> but your mileage may vary
2023-09-03 11:37:52 <zcrayfish> busybox ash is awesome. 
2023-09-03 11:39:06 <winny> tfw you forget you're in ash and brace expansion doesn't work
2023-09-03 11:39:18 <winny> and no M-y
2023-09-03 11:40:30 <ahmadraniri[m]> The only things I miss are cut and yank some parts of line (Ctrl+ u/k and Ctrl+y).
2023-09-03 12:05:00 <ascensionwires> how install and set bash as shekl 
2023-09-03 12:05:12 <ascensionwires> *shell on alpibe?
2023-09-03 12:05:20 <ikke> ascensionwires: Take one guess on how it install bash
2023-09-03 12:05:38 <ascensionwires> how set it as shell?
2023-09-03 12:06:30 <ikke> without installing any extra tools, change it in /etc/passwd
2023-09-03 12:06:47 <ascensionwires> ok
2023-09-03 12:06:52 <ikke> If you install `shadow`, there is chsh
2023-09-03 12:07:49 <ascensionwires> no i dont want shadow in my system
2023-09-03 12:08:39 <ikke> Just gave you your options
2023-09-03 12:12:56 <winny> you don't even get vipw, so it's easy to clobber a non shadow install
2023-09-03 12:34:50 <Tach> why is alpine building packages using r1 and r2 for the same release ?
2023-09-03 12:35:16 <Tach> it seems some repackaging after software releases, alpine builds r1 and like 2 weeks later r2
2023-09-03 12:36:13 <ikke> Tach: different reasons
2023-09-03 12:36:29 <ikke> one reason would be to rebuild after a library it depends on has been upgraded
2023-09-03 12:36:43 <Tach> ikke because of that I'm thinking about building my own
2023-09-03 12:36:51 <ikke> Why?
2023-09-03 12:36:56 <Tach> control
2023-09-03 12:37:04 <Tach> know what is running
2023-09-03 12:37:11 <lopid> your own what? all distros do it
2023-09-03 12:37:31 <Tach> I don't care what other distro's do
2023-09-03 12:37:57 <lopid> yours would have to, too, unless you build everything statically
2023-09-03 12:38:03 <ikke> Tach: 1) you _have_ to rebuild if a soname of a library you depend on changes
2023-09-03 12:38:17 <ikke> 2) the pkgrel bump is necessary for apk to automatically upgrade the package
2023-09-03 12:38:36 <ikke> (and for our builders to rebuild it)
2023-09-03 12:39:07 <ikke> I mean, you are free to build all packages for yourself, but the reason seems suspicious
2023-09-03 12:40:32 <Tach> ikke I automate such stuff :) but you can rely on them, if your software fails in a test you can check dependencies
2023-09-03 12:41:08 <Tach> ikke why suspicious ? It's just good to have it managed
2023-09-03 12:41:19 <ikke> Have what managed?
2023-09-03 12:41:28 <Tach> your own shit
2023-09-03 12:41:45 <lopid> sounds like a grand vision based upon a fundamental misunderstanding
2023-09-03 12:41:51 <vkrishn> i am guessing, 2 above probably the main reason, as any change src,apkbuild or otherwise would not get picked by builders
2023-09-03 12:42:12 <ikke> both reasons are valid
2023-09-03 12:42:20 <Tach> lopid you might think so, large companies do and need to as well
2023-09-03 12:43:08 <ikke> Tach: there are valid reasons, but stating package release versions changing is not a good one
2023-09-03 12:43:48 <Tach> ikke developing your own software is not as well ??
2023-09-03 12:43:49 <Tach> ;)
2023-09-03 12:44:08 <Tach> it's just beyond a lot of people their goals, with me it's not
2023-09-03 12:44:45 <lopid> you think a large company is going to go drop enterprise software and go with whatever you concoct?
2023-09-03 12:47:32 <Tach> lopid we are enterprise and people love our services because we control things right... enterprise is developed and supported as well; by people.
2023-09-03 12:47:57 <ikke> enterprises do a lot of things which not always make a lot of sense
2023-09-03 12:48:04 <Tach> lopid I always see jalous people reacting like you do; don't... take advantage of willing to do different.
2023-09-03 12:48:12 <ikke> cargo-culting is a thing
2023-09-03 12:48:25 <Tach> ikke no theirs engineers do, don't blame the company itself
2023-09-03 12:48:29 <lopid> jealous of what?
2023-09-03 12:48:38 <Tach> lopid I don't know and I don't care
2023-09-03 12:48:44 <lopid> ok
2023-09-03 16:28:23 <ascensionwires> how switch from lts to edge?
2023-09-03 16:29:01 <ikke> apk add linux-edge # make sure the community repo is enabled
2023-09-03 16:29:04 <ikke> apk del linux-lts
2023-09-03 16:30:55 <ascensionwires> ok
2023-09-03 16:31:22 <ascensionwires> also im on thinkpad how make fn keys work?
2023-09-03 16:31:42 <ikke> You man like media keys?]
2023-09-03 16:31:56 <ascensionwires> really 
2023-09-03 16:54:51 <linearcannon> ikke, there's a patch in alpine for it now, and bugs in firefox, and i'm working to fix the issue with the relevant upstream (sqlite)
2023-09-03 16:55:11 <ikke> linearcannon: ok good
2023-09-03 16:55:13 <linearcannon> turns out sqlite assumes __ppc__ being defined, means that you are big endian
2023-09-03 16:55:18 <ikke> ah, fun
2023-09-03 16:55:28 <linearcannon> gcc doesn't actually define that but clang does
2023-09-03 16:55:29 <ikke> and no one has noticed ebfore
2023-09-03 16:55:33 <ikke> before*
2023-09-03 16:55:42 <linearcannon> no, people have noticed, but it's never gotten fixed properly
2023-09-03 16:55:58 <linearcannon> since i'm not the first person to debug this to the end, i've decided i need to be the last, so i'm going to continue pushing until this bug does not exist anymore
2023-09-03 16:56:35 <linearcannon> meanwhile my firefox works now
2023-09-03 16:56:46 <ikke> glad to hear, and thank you!
2023-09-03 20:28:44 <winny> Tach: hosting your own package repos is somewhat straight forward. Then you can make your modified package conflict with the official one. Of course that means you'll have to keep on top of stuff like ABI rebuilds, other routine breakages, and other maintenance tasks
2023-09-03 20:29:41 <winny> If you haven't messed with APKBUILD much it's worth looking into... It's not too complicated compared to some other packaging systems 
2023-09-03 20:37:43 <Tach> wikky yeah running my own repo's for decades!
2023-09-03 20:37:59 <wikky> :)
2023-09-03 20:40:26 <Tach> but fancy proxying is what I like more these days
2023-09-03 21:47:06 <armin> it's a very very good thing if hosting your own package repositories is straight forward, but i'm happy if i don't need to, likewise.
2023-09-03 21:47:26 <armin> and alpine does a very good job here.
2023-09-03 21:47:38 <ikke> creating a custom repo is quite easy, keeping it up-to-date is a lot of work :)
2023-09-03 21:47:47 <armin> yup sure.
2023-09-03 21:47:49 <armin> hi ikke.
2023-09-03 21:48:02 <ikke> hi
2023-09-03 21:48:52 <armin> i must say i cannot believe what happened to alpine, when it started i was like "this will NEVER turn into a usable operating system.". well it did.
2023-09-03 21:49:36 <ikke> How long ago was that?
2023-09-03 21:49:36 <armin> it's totally usable nowadays, i love it for laptops, everything.
2023-09-03 21:49:44 <armin> 5 years maybe? 10?
2023-09-03 21:50:04 <armin> when docker came around i started to read "FROM: alpine-latest"
2023-09-03 21:50:39 <armin> i quickly realized it's way more than that. it's a 100% usable desktop operating system and something i love to have nowadays.
2023-09-03 21:52:11 <armin> i love systemd-less distro's anyway, but wow, i can just express my biggest of respect for what happened with alpine.
2023-09-03 21:52:49 <armin> you guys did your homework over the last years, hands down, maximum respect.
2023-09-03 21:53:28 <ikke> It all comes down to the great community
2023-09-03 21:54:36 <armin> you obviously manage to maintain a good one, and i really hope the best for you people and that you will continue to be such a strong community, using alpine over the last 10 years was heart-warming.
2023-09-03 21:55:12 <armin> always a good experience, great great great support from the community any time.
2023-09-03 21:56:30 <armin> and i love how modest and respectful you people are.
2023-09-03 21:57:29 <armin> honestly, alpine is one of the greatest linux projects of all time, i feel like it's so under-rated.
2023-09-03 21:57:49 <armin> you can build anything out of it.
2023-09-03 21:58:28 <armin> it helped me so tremendously, even in my job.
2023-09-03 22:00:10 <armin> anyways, thx and good night. :)
2023-09-03 22:05:37 <ikke> You too
2023-09-04 00:29:12 <qaqland[m]> 🥰
2023-09-04 04:29:31 <codebam> how do I run corectrl as root? it doesn't find a display if I use `doas corectrl`
2023-09-04 04:29:49 <qaqland[m]> su？
2023-09-04 04:30:35 <codebam> oh yeah su works
2023-09-04 04:30:37 <codebam> thanks
2023-09-04 10:07:53 <donoban> hi guys, anyone using foot? I noticed URL extraction not working since some days (weeks?), I guessed that it will work again after some update/reboot but nothing
2023-09-04 10:34:59 <ptrc> donoban: the keybind changed
2023-09-04 10:35:16 <donoban> oh, do you know what is now?
2023-09-04 10:35:18 <ptrc> Ctrl-Shift-U now enters unicode
2023-09-04 10:35:21 <ptrc> Ctrl-Shift-O is the new one
2023-09-04 10:35:25 <ptrc> for opening URLs
2023-09-04 10:35:41 <donoban> ohh, great!
2023-09-04 10:35:44 <donoban> thanks ptrc :D
2023-09-04 10:36:09 <p1fk6ohrq9> what's foot? i walked to office everyday then on foot
2023-09-04 10:36:27 <donoban> what is supposed to do with unicode? enter some unicode char code?
2023-09-04 10:36:34 <donoban> p1fk6ohrq9: it's a terminal emulator for wayland
2023-09-04 10:36:50 <p1fk6ohrq9> heh. got it. thanks
2023-09-04 10:41:47 <tree3> Yo. Anyways Solaris-style direct binding. I mentioned it a while back but never saw a reply (might have just missed it, bad client) -- would it be nice? Would Alpine like to have it if it were an option? musl isn't interested but maybe a fully-managed system has a different perspective?
2023-09-04 10:42:43 <p1fk6ohrq9> using alpine edge in my personal vps with podman and wireguard. works great sofar.
2023-09-04 10:47:35 <donoban> edge is near rock solid :)
2023-09-04 10:49:40 <p1fk6ohrq9> well, it failed to boot a few times. always keep backups!!
2023-09-04 11:15:30 <pj> >ask something
2023-09-04 11:15:30 <pj> >leave before anyone can reply
2023-09-04 11:27:48 <donoban> he has a bad client :\
2023-09-04 11:32:03 <quinq> and is also shocked, _shocked_ they tell you
2023-09-04 11:56:11 <lopid> how do you back up and restore the previous/current kernel for that situation? or do you refer to a whole system backup?
2023-09-04 11:56:29 <lopid> using syslinux
2023-09-04 11:59:22 <qaqland[m]> You can use matrix and join bridge room to get always online
2023-09-04 12:43:39 <quinq> qaqland[m], you make a good Darth Vador impersonation
2023-09-04 14:21:46 <Tach> can anyone help me out with this package that is needed for rspamd (building) ? It's not in later versions anymore. https://pkgs.alpinelinux.org/package/v3.13/community/x86_64/hyperscan-tools 
2023-09-04 14:43:31 <Tach> OK building with the old package
2023-09-04 14:55:52 <ascensionwires> there is no difference in setup-disk -k edge /mnt and after already installation with default (lts) doing: apk add linux-edge && apk del linux-lts
2023-09-04 14:59:43 <ascensionwires> ?
2023-09-04 15:04:31 <elly> ascensionwires: those are very different - read the help for setup-disk
2023-09-04 15:04:51 <elly> (setup-disk --help)
2023-09-04 15:18:24 <ascensionwires> both install edge kernel as i understand
2023-09-04 15:40:17 <elly> sure, but setup-disk also does a LOT of other stuff
2023-09-04 16:40:29 <cockliuser[m]> Would it be futile to try and create an Alpine image without an initramfs (as an experiment)
2023-09-04 17:21:35 <julie_harshaw> hey everyone, i have a very noob question: I would like to set up PAM on my desktop to use yubikey to authenticate, and the wikipage says I can replace busybox login either by shadow-login or util-linux-login
2023-09-04 17:21:39 <julie_harshaw> what's the difference?
2023-09-04 17:22:46 <minimal> busybox login does not support PAM
2023-09-04 17:23:02 <julie_harshaw> yup, but what's the difference between shadow-login and util-linux-login?
2023-09-04 17:23:04 <minimal> well I mean in Alpine it has PAM support disabled
2023-09-04 17:25:28 <julie_harshaw> i already have shadow because busybox doesn't have chsh
2023-09-04 17:25:39 <julie_harshaw> so i would default to shadow-login
2023-09-04 17:25:52 <minimal> they're 2 different implementations of login, you'd have to look through the code to see what functionality differences there are
2023-09-04 17:25:56 <julie_harshaw> but then again, shadow[-login] is from community, and util-linux-login is main
2023-09-04 17:26:17 <julie_harshaw> yeah I was hoping to avoid reading and comparing the code :D
2023-09-04 17:52:25 <PureTryOut> why are you posting every message twice?
2023-09-04 17:56:10 <minimal> PureTryOut: who? I don't see doubled messages
2023-09-04 17:56:57 <PureTryOut> you and julie_harshaw both
2023-09-04 17:57:09 <julie_harshaw> must be on your side, I don't see anything doubled either
2023-09-04 17:57:48 <PureTryOut> I only see you 2 doing it in this channel, not anywhere else. And not every message as julie_harshaw's last message is posted just once
2023-09-04 17:59:18 <minimal> we're not doing "it" though, that's the point
2023-09-04 17:59:21 <julie_harshaw> https://irclogs.alpinelinux.org/%23alpine-linux-2023-09.log seems okay too
2023-09-04 17:59:24 <PureTryOut> hmm it's more people actually, but only in this channel. Strange, a bridge thing I suppose
2023-09-04 18:33:27 <kwadroke[m]> It's doubling the non-Matrix users from the Matrix side of things as Matrix users aren't double posting
2023-09-04 18:33:59 <elly> from the IRC side I see no doubling at all
2023-09-04 18:35:02 <kwadroke[m]> it's got to be the Matrix bridge
2023-09-04 18:49:14 <zcrayfish> geeeze. :(
2023-09-04 19:10:39 <quinq> Just use regular IRC
2023-09-04 19:15:06 <p1fk6ohrq9> yeah. for chatting, pure irc is way better.
2023-09-04 19:15:19 <minimal> was a glitch in the matrix I guess lol
2023-09-04 19:42:25 <micahrl[m]> Is there a good way to build alpine packages for a custom APK repo (not main/community/etc from the aports git repository)? It looks like it just uses the name of the ../../ of the APKBUILD directory based on this https://gitlab.alpinelinux.org/alpine/abuild/-/blob/master/abuild.in#L2937 ... is there a good way to override that for packages developed outside of aports?
2023-09-04 19:44:32 <ikke> I don't see a way to override that
2023-09-04 19:45:42 <Ermine> ../../ doesn't have to be called 'master', 'community' or 'testing'
2023-09-04 19:47:12 <micahrl[m]> Right, my question is, is there a way to avoid changing the name of ../../ to be what I want my APK repo name to be, but it seems like there isn't.
2023-09-04 19:47:38 <micahrl[m]> Would adding a way to customize the repo name be a useful MR?
2023-09-04 19:48:07 <Ermine> So you want to e.g. have your ../.. to be named 'a' and the repo called 'b'?
2023-09-04 19:48:29 <micahrl[m]> yes
2023-09-04 19:52:04 <micahrl[m]> One use case is maintaining some project somewhere and  being able to provide an APKBUILD in the repo. Right now you need a build script that creates a temp dir named what you want the APK repository to be, then copies the source code over there, and builds the .apk package, which is kind of weird.
2023-09-04 20:08:30 <OdaNobunaga[m]> In case you didn't know yet:... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/FkpdAmJCqIJJkNAZNCpwjktW>)
2023-09-04 20:09:02 <OdaNobunaga[m]> Is it known when Kubernetes and CRI-O will be available to Alpine-LTS users?
2023-09-04 20:09:09 <ikke> "In case you didn't know yet:... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/FkpdAmJCqIJJkNAZNCpwjktW>)"
2023-09-04 20:09:17 <ikke> that link is hardly usable
2023-09-04 20:14:29 <ascensio1wires> then i meant - is there a difference between 1 and 2? https://hastebin.com/share/arateyebiy.bash
2023-09-04 20:15:25 <quinq> ascensio1wires, that text is broken
2023-09-04 20:16:01 <quinq> Oh, it's html+css+js
2023-09-04 20:16:36 <quinq> Please use text/plain
2023-09-04 20:16:37 <ascensio1wires> why broken?
2023-09-04 20:17:07 <quinq> ascensio1wires, I misread, it's not broken, it's just difficult to access
2023-09-04 20:18:28 <ascensio1wires> i just before asked about this in this chat
2023-09-04 20:18:48 <ascensio1wires> is there difference of result of this ways?
2023-09-04 20:19:41 <ikke> I don't expect any differences
2023-09-04 20:20:50 <ascensio1wires> ok thanks
2023-09-04 20:22:42 <ascensio1wires> just 2) variant faster and shorter)
2023-09-04 20:27:28 <ascensio1wires> i hope its same, also i hope my method of installation not shit
2023-09-04 20:32:01 <BlueCrayon> hi there
2023-09-04 20:32:19 <BlueCrayon> Am I allowed to ask questions about Linux in here?
2023-09-04 20:33:29 <lopid> alpine linux?
2023-09-04 20:34:40 <BlueCrayon> I have it downloaded, but I couldn't get it to boot. Will Alpine work in VM?>
2023-09-04 20:36:02 <lopid> it should do
2023-09-04 20:36:15 <BlueCrayon> So just an ISO?
2023-09-04 20:36:24 <BlueCrayon> I also want to try Chimera Linux
2023-09-04 20:45:38 <zcrayfish> The Alpine and Chimera ISOs should both work fine in most VMs.
2023-09-04 23:31:02 <M-u-m-p-i-t-z> Hi! Anybody confim with awall here ⸮
2023-09-04 23:34:15 <M-u-m-p-i-t-z> I need an example for a base zones and policies file, if alpine is a chat server in DMZ on a raspy with one networkcard.
2023-09-05 03:22:37 <dpward> mirror.math.princeton.edu has not synced in 7 days. Admin wrote back that rsync.alpinelinux.org seems to be blocking all connections from the Princeton network? (He can access off-campus fine)
2023-09-05 10:47:39 <WhyNotHugo> dhcpcd overwrites its cmdline, so /proc/4257/cmdline just reads "dhcpcd: [manager] [ip4] [ip6]".
2023-09-05 10:47:44 <WhyNotHugo> How can I determine which flags were passed to it?
2023-09-05 12:32:00 <zcrayfish> ugh. openrc-run manpage seems to be wrong about the RC_REBOOT variable
2023-09-05 12:32:21 <zcrayfish> I never see it set... so how is an init script that only needs to run on reboot (but not run on shutdown) supposed to accomplish that.... heh
2023-09-05 12:36:45 <SmthSweethinher[m]> framebuffer graphical environments on alpine, like fbpad fbterm and more, are something anyone has experience or wanna drop some anecdotes about?
2023-09-05 12:37:33 <zcrayfish> That's pretty old school. :)
2023-09-05 12:38:11 <SmthSweethinher[m]> heh yeah, just wanna get away from the security holes in X11 now- and Wayland maybe can't run on a (g)old T60
2023-09-05 12:40:34 <SmthSweethinher[m]> also looked into [TWIN](https://github.com/cosmos72/twin) and [CLIFM](https://github.com/leo-arch/clifm) because both can be run in (virtual)consoles
2023-09-05 12:48:59 <SmthSweethinher[m]> however TWIN can lack mouse support then, and dunno how to solve that really
2023-09-05 12:58:32 <zcrayfish> GPM not working there?
2023-09-05 12:59:06 <SmthSweethinher[m]> ohh, GPM is what to use?
2023-09-05 12:59:50 <zcrayfish> Yes definitely
2023-09-05 12:59:50 <SmthSweethinher[m]> [archwiki](https://wiki.archlinux.org/title/General_purpose_mouse) says it all
2023-09-05 13:00:09 <zcrayfish> It's near the bottom of TWIN's github page as well.
2023-09-05 13:00:28 <zcrayfish> morning minimal 
2023-09-05 13:00:47 <SmthSweethinher[m]> morning minimal
2023-09-05 13:01:01 <minimal> afternoon
2023-09-05 13:04:53 <SmthSweethinher[m]> ah, afternoon minimal
2023-09-05 13:07:51 <minimal> SmthSweethinher[m]: do those support DRM or only fbdev devices?
2023-09-05 13:08:07 <minimal> for several releases now Alpine uses DRM drivers rather than fbdev drivers
2023-09-05 13:08:15 <SmthSweethinher[m]> hmmm
2023-09-05 13:08:33 <SmthSweethinher[m]> good sanity check, yes TWIN relies on x11 or fbdev drivers
2023-09-05 13:08:56 <SmthSweethinher[m]> meanwhile CLIFM uses nothing- not even ncurses
2023-09-05 13:09:06 <zcrayfish> Some drivers (i915) have some kind of fbdev emulation crap. Dunno if it makes any difference here.
2023-09-05 13:09:11 <SmthSweethinher[m]> * good sanity check, yes TWIN relies on x11 or fbdev drivers
2023-09-05 13:09:11 <SmthSweethinher[m]> edit: but it *does* work in an ncurses mode
2023-09-05 13:10:06 <SmthSweethinher[m]> maybe it's best to forsake all of fbdev X11 (and wayland is out the window as too big)- and ascetically embrace the console, ncurses, gpm
2023-09-05 13:10:46 <SmthSweethinher[m]> oh but there was one that relies on KMS- probably more- but came across one just now
2023-09-05 13:11:15 <SmthSweethinher[m]> [KMSCON](https://wiki.archlinux.org/title/KMSCON) if KMS matters in the context of DRM=
2023-09-05 13:11:22 <SmthSweethinher[m]> s/=/?/
2023-09-05 13:11:42 <minimal> yes I should have said DRM/KMS
2023-09-05 13:12:08 <minimal> Alpine was one of the first distros to move from fbdev to DRM/KMS
2023-09-05 13:12:47 <SmthSweethinher[m]> neat
2023-09-05 13:13:20 <minimal> https://elinux.org/images/6/6d/Trading_Fbdev_for_DRM_no_returns_accepted_Handouts.pdf
2023-09-05 13:13:56 <SmthSweethinher[m]> appreciated reading, and kmscon exist in alpine package repo!
2023-09-05 13:14:11 <SmthSweethinher[m]> along with gpm and clifm for that matter- for interest
2023-09-05 13:14:32 <SmthSweethinher[m]> however twin isn't included- and for reference can be run in console solely
2023-09-05 13:15:06 <SmthSweethinher[m]> endearing to see all architectures supported too :')
2023-09-05 13:23:27 <ascensionwires> as i mentiozed 
2023-09-05 13:23:48 <ascensionwires> before, fn keys dont worj
2023-09-05 13:24:03 <ascensionwires> *dont work on my thinkpad
2023-09-05 13:26:24 <zcrayfish> ascensionwires: Please provide the folks who could potentially help you with more information... such as the specific model of thinkpad, and which environment you are having trouble in (e.g. vt, Xorg, or Wayland).
2023-09-05 13:28:38 <ascensionwires> t480 i have, mb keys is working, but i cant regulate backlight (or how it named) with f5-f6
2023-09-05 13:29:10 <ascensionwires> tty
2023-09-05 14:34:52 <vkrishn> is it ok that, hostapd installs files in /etc/hostapd/ with perms -rwxr-xr-x, or something else happening ?
2023-09-05 14:39:35 <minimal> vkrishn: what do you mean "is it ok"? what is the problem you are having?
2023-09-05 14:40:34 <vkrishn> minimal: does it have same perm if you install ?
2023-09-05 14:41:28 <minimal> the hostapd package sets the files it places in /etc/hostapd/ as rwx,rx,rx
2023-09-05 14:41:33 <vkrishn> then is it ok ? coz I don't know, I keep changing them to -rw------- in new install
2023-09-05 14:42:36 <minimal> then if you think the permissions are wrong perhaps you should raise either an Issue or a MR in gitlab to address this
2023-09-05 14:44:45 <vkrishn> i don't know, hostapd.conf can have open text wpa_passphrase= but admins use wpa_psk=
2023-09-05 14:47:52 <minimal> then perhaps raise a Gitlab issue suggesting a change
2023-09-05 14:57:35 <em6> hey, I am starting to code c++ for raspberry pi pico w. I cant fint the apk pack for the cross compiler. anyone know if there is one?
2023-09-05 15:00:41 <em6> is it gcc-arm-none-eabi?
2023-09-05 15:42:04 <p00f[m]> hello, does alpine linux have an equivalent of systemd-coredump?
2023-09-05 15:53:27 <ikke> There is nothing included
2023-09-05 15:55:55 <p00f[m]> Are there 3rd party openrc services which do the same
2023-09-05 15:56:15 <ikke> I cannot find any alternative to it
2023-09-05 15:56:21 <p00f[m]> Like I tried searching "openrc coredump"
2023-09-05 15:56:47 <ikke> p00f[m]: do you just want to get coredumps?
2023-09-05 15:58:58 <WhyNotHugo> py3-pylsp has a strict dependency on jedi<0.19.0, and fails at runtime if this version doesn't match.
2023-09-05 15:59:19 <WhyNotHugo> However, we have jedi 0.19.0 in repos, and that's what gets pulled as a dependency.
2023-09-05 16:00:13 <WhyNotHugo> Maybe I can just patch py3-pylsp with this? https://github.com/python-lsp/python-lsp-server/pull/416.patch?
2023-09-05 16:00:21 <WhyNotHugo> The upstream PR says that there are no breaking changes.
2023-09-05 16:00:41 <ikke> Yeah, we do that more often
2023-09-05 16:00:57 <ikke> mostly these upper limits are not based on anything other than that was what'a available at the time
2023-09-05 16:01:13 <ikke> ofcourse, it means upstream hasn't tested it against newer versions
2023-09-05 16:01:49 <WhyNotHugo> Yeah, I've done the same a few times (pinning dependencies like this).
2023-09-05 16:01:58 <WhyNotHugo> I wish this would come up at install time and not at runtime.
2023-09-05 16:02:46 <ikke> Yeah, that is anoying
2023-09-05 16:25:28 <p00f[m]> ikke: it does more than just get coredumps - it runs gdb and stops it at the segfault or whatever
2023-09-05 16:25:41 <p00f[m]> ideally something like this would be nice
2023-09-05 19:31:24 <customkernel> hey everyone
2023-09-05 19:31:54 <ikke> Let me guess, you want to build a custom kernel? :P
2023-09-05 19:31:54 <customkernel> i am trying to build a custom kernel in docker using abuild and failing miserably...
2023-09-05 19:32:11 <customkernel> ikke: lol, I think you might be spot on :D
2023-09-05 19:32:39 <ikke> So how is it failing?
2023-09-05 19:32:40 <customkernel> so to summarize, I spun up a docker container, added an underprivileged user called abuild
2023-09-05 19:32:54 <customkernel> then added the abuild user to have doas privileges
2023-09-05 19:33:05 <customkernel> generated signing keys using abuild -a -i
2023-09-05 19:33:21 <customkernel> cloned the aports repo, then cd into aports/main/linux-lts
2023-09-05 19:33:49 <customkernel> and abuild deps fails, I can't even get further... :( not sure what I am missing, followed official docs, akthough I know that is not for docker
2023-09-05 19:34:02 <ikke> fails how?
2023-09-05 19:34:02 <customkernel> WARNING: opening /home/abuild/packages//main: No such file or directory
2023-09-05 19:34:09 <ikke> That's just a warning
2023-09-05 19:34:22 <customkernel> this is the error I get, the double slash after packages is strange, like there should be something coming from a variable
2023-09-05 19:34:53 <customkernel> ERROR: unable to select packages: gcc-12.2.1_git20220924-r10: breaks: .makedepends-linux-lts-20230905.192628[gcc>=13.1.1_git20230624]
2023-09-05 19:35:40 <ikke> So you are trying to build an older kernel on an edge image?
2023-09-05 19:35:57 <ikke> or the opposite I guess
2023-09-05 19:36:00 <customkernel> docker image is 3.18.3
2023-09-05 19:36:21 <ikke> Right, you need edge for the latest kernel
2023-09-05 19:36:29 <customkernel> and I cloned the aports repo and i am at HEAD, did not check out any branch
2023-09-05 19:36:40 <customkernel> ohh OK
2023-09-05 19:36:41 <ikke> the APKBUILD has: "gcc>=13.1.1_git20230624"
2023-09-05 19:36:51 <ikke> and alpine 3.18 only has gcc 12
2023-09-05 19:37:01 <ikke> fyi, you can also use alpinelinux/build-base as image
2023-09-05 19:37:02 <customkernel> so try to rerun the same, juse use alpine:edge container image instead of 3.18.3?
2023-09-05 19:37:23 <ikke> only thing you need to do there is abuild-keygen -ain and clone the repo
2023-09-05 19:37:26 <ikke> the rest is all setup
2023-09-05 19:37:32 <customkernel> ohh this will have the abuild stuff preconfigured I guess: "alpinelinux/build-base" thanks for that tip!
2023-09-05 19:37:47 <customkernel> perrrrfect thank you so much ikke!
2023-09-05 19:37:54 <customkernel> will try rn
2023-09-05 19:38:07 <ikke> That image is used in our CI (technically another image that uses it with automation on top)
2023-09-05 19:38:32 <customkernel> do I pull alpinelinux/build-base:edge? or :latest?
2023-09-05 19:39:04 <ikke> latest
2023-09-05 19:40:13 <ikke> the double // is a red-herring, it's innocent
2023-09-05 19:40:40 <customkernel> thanks for confirming that
2023-09-05 19:40:46 <valerius> it's better than the \\ you have to use in Windows to quote out the stupid backslash they went with
2023-09-05 19:40:48 <customkernel> will report back how it goes
2023-09-05 19:45:13 <customkernel> OK "abuild deps" ran through just fine
2023-09-05 19:45:20 <customkernel> I suppose the rest will be OK too
2023-09-05 19:45:22 <customkernel> but will confirm
2023-09-05 19:56:22 <customkernel> will come back later to ocnfirm for anyone interested, but everything looks right so far. thanks again ikke!
2023-09-05 19:57:02 <ikke> np
2023-09-06 06:21:54 <DanielBrenot[m]> Can anyone help me with a small issue with alpine login? I have a custom live iso i'm working on and i'm trying to make it so that when the alpine iso starts it will load a terminal as root with no login prompt. I've tried changing the flags to getty with no avail. I'm not super well versed when it comes to the login prompt and i've tried some of the things i've seen online, but none of them have worked so far. Thanks in advance
2023-09-06 06:21:54 <DanielBrenot[m]> for any help.
2023-09-06 06:30:16 <ikke> DanielBrenot[m]: you need to install agetty and switch to that 
2023-09-06 06:32:03 <DanielBrenot[m]> ikke: Thank you! I'll try that.
2023-09-06 06:35:10 <zcrayfish> look up the --autologin flag on the agetty manpage :)
2023-09-06 06:56:06 <DanielBrenot[m]> in the file /etc/inittab I have replaced the following line:... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/BpzANLIXtCyvAuwZyCVoNrFN>)
2023-09-06 07:41:46 <zcrayfish> I'm thinking on this one, but... 0 is the baud rate... is that allowed? it's not in the list of agetty --list-speeds
2023-09-06 07:43:30 <zcrayfish> doesn't make agetty crash though. hehe
2023-09-06 07:48:30 <zcrayfish> that's not the problem in any case
2023-09-06 07:50:18 <WhyNotHugo> Neochat won't start with "Error relocating /usr/bin/neochat: _ZN8Quotient4Room14userIdsAtEventERK7QString: symbol not found"
2023-09-06 07:50:20 <WhyNotHugo> ldd prints: /lib/ld-musl-x86_64.so.1: cannot load neochat: No such file or directory
2023-09-06 07:50:29 <WhyNotHugo> Could it need to be rebuilt due to upgraded depndencies?
2023-09-06 07:52:08 <dwfreed> did you provide the full path to neochat when running ldd ?
2023-09-06 07:52:14 <dwfreed> it doesn't PATH search
2023-09-06 07:52:54 <DanielBrenot[m]> <zcrayfish> "I'm thinking on this one, but......" <- I don't think it's supposed to be a baud rate. I think it's the argument to the -L flag which is supposed to control CLOCAL in getty
2023-09-06 07:54:41 <zcrayfish> It doesn't matter because that's not the issue.
2023-09-06 07:55:04 <zcrayfish> the only way I could replicate your inittab not working was by forgetting to install agetty :)
2023-09-06 07:55:11 <dwfreed> baud rate's after the port, not before, anyway :P
2023-09-06 07:58:59 <DanielBrenot[m]> zcrayfish: Oh? So with the exact command i have there it worked? Did you modify the inittab any other way?
2023-09-06 08:03:52 <DanielBrenot[m]> And did you remove the last re spawn command?
2023-09-06 08:05:01 <zcrayfish> actually looks like I didn't use the -L/--local-line flag like you did :/
2023-09-06 08:05:54 <zcrayfish> the only respawns I have are serial port 0 and the VTs
2023-09-06 08:06:15 <DanielBrenot[m]> can you post your inittab file?
2023-09-06 08:06:49 <zcrayfish> yeah give me a sec
2023-09-06 08:08:15 <zcrayfish> https://tpaste.us/KE9b
2023-09-06 08:09:08 <zcrayfish> Wish my old Esprit terminal wasn't packed up in a moving box. I'd test it on something other than QEMU. haha
2023-09-06 08:09:11 <DanielBrenot[m]> zcrayfish: I appreciate this so much, thank you
2023-09-06 08:09:20 <DanielBrenot[m]> I just need qemu
2023-09-06 08:09:32 <zcrayfish> oh cool
2023-09-06 08:09:35 <DanielBrenot[m]> That's the only place i'm going to be running this image lol
2023-09-06 08:09:37 <zcrayfish> no problem
2023-09-06 08:10:18 <DanielBrenot[m]> also can i just say, mkimage is a godsend
2023-09-06 08:10:29 <DanielBrenot[m]> anyways, goodnight all and thanks for the help
2023-09-06 08:10:59 <zcrayfish> nini. \o/
2023-09-06 08:12:46 <WhyNotHugo> How can I force the vfat filesystem module to be included in the initramfs?
2023-09-06 08:19:38 <ikke> Add a vfat.modules file in /etc/features.d with the path to the module(s)
2023-09-06 08:19:57 <ikke> Then enable that feature
2023-09-06 08:22:29 <quinq> (/etc/mkinitfs/features.d/, no?)
2023-09-06 08:22:58 <ikke> Sorry, yes
2023-09-06 08:26:17 <quinq> That's actually nice and pretty simple
2023-09-06 08:27:43 <zcrayfish> n.b.: the vfat module is often pulled in by default as part of the usb feature.
2023-09-06 08:28:44 <quinq> good point :D
2023-09-06 09:18:43 <WhyNotHugo> Odd, I do see kerne/fs/fat in /etc/mkinitfs/features.d/usb.modules.
2023-09-06 09:18:59 <WhyNotHugo> This makes the module get bakes into the initramfs, but it's possible that it doesn't get auto-loaded, right?
2023-09-06 09:19:43 <WhyNotHugo> Background: due to an oversight, my initramfs didn't get rebuilt, so I booted off a kernel older than the installed one. My main partitin no longer had the fat32 module, so I couldn't easily mount /boot to fix this.
2023-09-06 09:20:27 <WhyNotHugo> This has happened twice, so I'd like to make sure that the fat32 module is always loaded form the initfs and available.
2023-09-06 09:25:19 <zcrayfish> understandable, I definitely want fat+msdos+vfat on my UEFI systems in case I get thrown into a rescue shell
2023-09-06 09:25:39 <zcrayfish> I like to `zat /boot/initramfs-edge | cpio -t | grep -i fat' to make sure
2023-09-06 09:25:49 <zcrayfish> zat == cat
2023-09-06 09:25:51 <zcrayfish> lol. 
2023-09-06 09:25:57 <zcrayfish> damn typos.
2023-09-06 09:28:28 <WhyNotHugo> yea, it's there: lib/modules/6.5.1-0-edge/kernel/fs/fat (+3 .ko.gz files).
2023-09-06 09:29:09 <WhyNotHugo> I think that the problem is that the kernel does not load the modules early. It then switching the root to the "real" root partition, and that one has modules for an older kernel.
2023-09-06 09:29:18 <WhyNotHugo> If I try to mount at that point, the  module from the initramfs is no longer visible.
2023-09-06 09:29:31 <WhyNotHugo> SO I need to force the kernel to LOAD the module during initramfs, even if it's not using it yet.
2023-09-06 09:33:48 <zcrayfish> I wonder if mdev can be used to force load it
2023-09-06 09:35:59 <ikke> WhyNotHugo: add it to the modules kernel arg I suppose 
2023-09-06 09:52:49 <WhyNotHugo> Is stuff in modules.load.d loaded from the initfs or the "final" root?
2023-09-06 10:14:32 <ikke> Final
2023-09-06 10:17:18 <ikke> It's read by the modules openrc service
2023-09-06 17:43:10 <armin> ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIGOx6UAmAz7MMMXPA0dvlESPaCfh8AQJD8sUdrSPgvf7 armin@acid.my.domain
2023-09-06 17:47:39 <Ermine> 
ACTION logs into acid.my.domain

2023-09-06 17:48:11 <minimal> using *public* key?
2023-09-06 17:48:53 <Ermine> Ah, yes
2023-09-06 17:49:10 <lopid> Ermine is that good
2023-09-06 17:49:21 <armin> no idea how this got pasted though, weird
2023-09-06 17:49:35 <armin> especially so since my irssi has some paste protection that hits on 1 line already
2023-09-06 17:49:48 <Ermine> But given enough computational power, I could find out private key
2023-09-06 17:49:51 <armin> but, sure, feel free to add my pubkey everywhere :)
2023-09-06 17:50:59 <Ermine> adduser -s /sbin/nologin armin
2023-09-06 17:51:58 <Ermine> echo "$pubkey" > ~armin/.ssh/authorized_keys
2023-09-06 17:52:07 <Ermine> you're welcome
2023-09-06 18:14:43 <frag> using ERE or BRE (not -P), trying to grep/higlight from ' m ' to ' m ' but not the longest match in case there are multiple pairs of ' m ' ... eg   'foo m test m bar m lala m bar' would match/higlight ' m test m ' and ' m lala m ' .. is it possible?
2023-09-06 18:26:39 <quinq> frag, just ' m [^ ]\{1,\} m ', no?
2023-09-06 18:27:19 <quinq> Or do you have things like ' m foo bar m '?
2023-09-06 18:30:00 <frag> quinq: could be ' m longe sentence m ' too yeah
2023-09-06 18:31:03 <frag> i want to use it to mark/yellow text ...
2023-09-06 18:52:26 <quinq> frag, a bit of brute-forcing ^^
2023-09-06 18:52:28 <quinq> ' m \([^ ]\| [^m]\| m[^ ]\)\{1,\} m '
2023-09-06 19:00:57 <frag> quinq: interesting!
2023-09-06 19:13:27 <frag> seems to work and i learnt some new regfu :]
2023-09-06 19:18:27 <quinq> nice :D
2023-09-06 19:20:47 <frag> '\bm \([^ ]\| [^m]\| m[^ ]\)\{1,\} m\b' seems to work too, why not '\bm\b\([^ ]\| [^m]\| m[^ ]\)\{1,\}\bm\b'
2023-09-06 19:24:02 <mfny> Hello
2023-09-06 19:44:43 <quinq> Wrong door?
2023-09-06 20:28:31 <codebam> what provides the manpath command?
2023-09-06 20:29:12 <lopid> apk search cmd:manpath
2023-09-06 20:29:29 <codebam> okay thank you
2023-09-06 20:52:25 <DanielBrenot[m]> I'm trying to set the PS1 environment variable to empty but not having much luck. I set it in /root/.profile using an apkovl file but it seems to not exist when i boot into the image. Any advice?
2023-09-06 20:53:10 <p1fk6ohrq9> 
ACTION empty PS1

2023-09-06 20:56:50 <OdaNobunaga[m]> Someone has taken inspiration from Alpine: Alpaquita Linux from Bell SW https://bell-sw.com/alpaquita-linux/
2023-09-06 20:58:51 <DanielBrenot[m]> <p1fk6ohrq9> "empty PS1" <- > * <@_oftc_p1fk6ohrq9:matrix.org> empty PS1
2023-09-06 20:58:51 <DanielBrenot[m]> ?
2023-09-06 21:13:56 <minimal> OdaNobunaga[m]: yeah I had dealings with them several months back when I discovered they'd removed my copyright messages from 1 of my packages and put their own instead
2023-09-06 21:17:01 <minimal> (or put their copyright on stuff I'd written but hadn't put a copyright message on, can't remember which)
2023-09-06 21:31:31 <invoked> sounds like a grift
2023-09-06 22:08:12 <minimal> here's their fork of aports: https://github.com/bell-sw/alpaquita-aports
2023-09-06 22:56:23 <invoked> "As you have probably noted Alpaquita is very similar to Alpine"  haha, wat.
2023-09-06 22:58:59 <invoked> i'm afraid to dig any further.  i'd expect to read stuff like "our exclusive 'apkay' packaging technology ensures the smoothest installation possible"
2023-09-06 22:59:06 <invoked> along with the "your choice of 3 mallocs"
2023-09-07 01:39:03 <zcrayfish> oh boy mckaygerhard on their issue tracker. Blast from the past there. xD
2023-09-07 10:13:10 <Ermine> Is kernel supposed to dump panic output to tty1?
2023-09-07 10:16:46 <Ermine> I've got myself comfast cf-953ax wifi adapter (driver mt7921au). It doesn't appear in ip link, and kernel apparently panics when I unplug it
2023-09-07 10:18:05 <lopid> they usually did that in the past
2023-09-07 10:18:11 <lopid> well, hopefully not usually!
2023-09-07 10:18:33 <Ermine> Ah, actually, there's some disk activity, but the screen got frozen
2023-09-07 10:26:58 <Ermine> Well, fetched required firmware package with the old adapter, no more panics
2023-09-07 16:44:31 <ascensionwires> i asked before but quited, what differense between `setup-disk /mnt` then `apk add linux-edge && apk del linux-lts` and `setup-disk -k edge /mnt`?
2023-09-07 16:49:20 <ikke> and I have answered it before
2023-09-07 16:57:05 <ascensionwires> ikke, what you answered?
2023-09-07 16:57:21 <ikke> That it should not matter 
2023-09-07 17:02:52 <ascensionwires> ok thanks (and i also asked the wrong question then i didnt specify that i ran setup-disk in both cases)
2023-09-07 17:15:34 <Saijin_Naib[m]> Anyone missing menulibre for editing desktop menu entries? 
2023-09-07 17:16:03 <Saijin_Naib[m]> It's on my list to try and package, and can get bumped up if anyone else also wants it 
2023-09-07 17:16:34 <Saijin_Naib[m]> Otherwise, owntone has priority 🤣
2023-09-07 18:28:20 <ascensionwires> what i need to do if while running setup-disk connection abborted was
2023-09-07 18:28:30 <ascensionwires> like unstable network was
2023-09-07 18:28:45 <ascensionwires> i just need rerun command?
2023-09-08 01:32:11 <qaqland[m]> When will alpine have a new initd instead of openrc?
2023-09-08 01:32:50 <humm> the current default init is busybox init
2023-09-08 01:33:16 <humm> you can change init and service manager to whatever you want
2023-09-08 01:34:04 <humm> it’s just that the Alpine packages don’t give you service definitions for other service managers
2023-09-08 01:35:25 <qaqland[m]> Oh, so openrc is not initd
2023-09-08 01:35:47 <qaqland[m]> it is the service manager?
2023-09-08 01:46:35 <humm> qaqland[m], yes
2023-09-08 01:46:47 <humm> look at /etc/inittab
2023-09-08 01:48:02 <humm> sometimes, busybox init (pid 1) spawns an openrc process which might start or stop services
2023-09-08 01:48:49 <humm> otherwise, it’s just init and the services (and their start-stop-daemon or the like) running
2023-09-08 01:50:29 <qaqland[m]> thanks
2023-09-08 02:02:31 <Quillith> hey, I think I might've borked my flatpak install. How do I tell my flatpak install not to need the --user anymore, and how do I re-create the repo?
2023-09-08 02:03:08 <Quillith> my problem is when I try to run apps, even with the --user flag, I get this "ERROR: while opening repository /var/lib/flatpak/repo: opening repo: opendir(/var/lib/flatpak/repo): No such file or directory
2023-09-08 02:03:54 <Quillith> and then it crashes. So I tried to uninstall flatpak and reinstall, but it's just picking up where we left off, and I'm not sure what to erase to start with a cleaner slate (preferably with my individual app data still intact)
2023-09-08 02:19:09 <Quillith> I deleted the 'flatpak' folder in my ~/.local folder as well as /var/lib/flatpak , then I uninstalled/reinstalled flatpak, but it did not help. Now when I try and add the flathub repo it says "authorization error: the name org.freedesktop.PolicyKit1 was not provided by any .service files
2023-09-08 14:16:03 <ashie[m]> I don't have the time to open a proper issue right now but "zathura-pdf-mupdf" is compiled against outdated version of mupdf and complains about incompatible version, recompiling from aports with no changes to APKBUILD fixed it
2023-09-08 14:50:35 <tuzu_> I am trying to setup flatpak and if I try to add flathub it gives me this error: error: Flatpak system operation ConfigureRemote not allowed for user
2023-09-08 14:52:37 <tuzu_> I know that I could run flatpak as root but I would prefere to run them as 
2023-09-08 14:52:50 <tuzu_> normal user
2023-09-08 14:53:11 <zcrayfish> try passing the --user flag to whatever commands you are using.
2023-09-08 14:53:18 <zcrayfish> oh wait.
2023-09-08 14:53:26 <zcrayfish> not sure what's going on there.
2023-09-08 14:53:49 <tuzu_> thanks that worked
2023-09-08 14:54:05 <zcrayfish> awesome, I didn't think it was gonna work xD
2023-09-09 11:54:33 <qaqland[m]> If i can install one package by apk and pip, which method is recommended 
2023-09-09 11:55:15 <qaqland[m]> e.g.  jupyter notebook
2023-09-09 11:55:18 <lopid> apk
2023-09-09 11:55:55 <qaqland[m]> 🤔
2023-09-09 12:52:46 <ashie[m]> qaqland[m]: if installing to root always use apk, pip is fine if you contain it to its own virtualenv
2023-09-09 13:18:53 <qaqland[m]> There is no conda on alpine, is there any alternative recommended 
2023-09-09 15:19:30 <ashie[m]> sorry but no idea
2023-09-09 16:33:26 <nu[m]1> does anyone have a small alpine based container which i could use to automate netbooting? (auto start dhcp/tftpd, cache main repo for offline envs)
2023-09-09 16:35:23 <p1fk6ohrq9> even smaller than the image on docker?
2023-09-09 16:36:44 <nu[m]1> sry, i didnt mean the base image itself, but an already functional container with netbooting configured
2023-09-09 20:10:30 <wk> Hey guys 
2023-09-09 20:10:57 <wk> I came here to thank you all developers of Alpine linux 
2023-09-09 20:11:19 <wk> your distro could make me revive my old pentium-M with only 900mb of ram 
2023-09-09 20:11:51 <wk> and with i3wm it opens firefox more faster than my Ryzen Desktop with xfce 
2023-09-09 20:25:25 <lopid> nice
2023-09-09 20:48:27 <mothership> hello
2023-09-09 20:48:40 <mothership> i have some problems(
2023-09-09 20:52:54 <armin> mothership: you could at least *try* to increase the chance of someone helping you by actually mentioning what these could be.
2023-09-09 23:32:36 <NepNepdmsalwaysopen[m]> Is there anything blocking synapse 1.91.2 right now?
2023-09-10 00:28:57 <ptrc> NepNepdmsalwaysopen[m]: just the maintainer's approval
2023-09-10 00:31:13 <ptrc> as in, there's !51180 open
2023-09-10 00:34:58 <NepNepdmsalwaysopen[m]> ahh
2023-09-10 02:26:29 <Saijin_Naib[m]> qaqland: If you use pipx, that apparently makes things a bit nicer to manage, and as a bonus, TopGrade can manage/update that easily, so that's neat
2023-09-10 13:02:37 <witcher> is there an elegant way to start multiple instances of a service with openrc with different arguments? something similar in function to systemd's unit@arg.service where arg can be accessed with %i
2023-09-10 13:03:31 <ikke> witcher: you can symlink services, and refer to RC_SVCNAME to get the name of the service
2023-09-10 13:04:00 <ikke> witcher: a common practice is to call the service service.instance, and then chop of the "service." part to get the instance name
2023-09-10 13:04:28 <witcher> that sounds like a good idea, i'll try that. thanks, ikke 
2023-09-10 13:05:00 <ikke> https://www.mankier.com/8/openrc-run#Files
2023-09-10 13:05:34 <caskd> stunnel and postfix are examples which use this
2023-09-10 15:51:39 <fancsali[m]> Hey peeps, my LXC started to act up recently. Anyone else seen this:... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/OEKUXPXCYiLqrftoloDVvgzh>)
2023-09-10 15:51:56 <fancsali[m]>  * Hey peeps, my LXC started to act up recently. Anyone else seen this:... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/tWfeBcUDMKqfNzIqIsaNvvcE>)
2023-09-10 15:52:17 <fancsali[m]> Any pointers appreciated...
2023-09-10 15:54:29 <qaqland[m]> openlinux21 polls in matirx😕
2023-09-10 15:57:20 <caskd> fancsali[m]: have you upgraded your kernel before executing this without rebooting? a old kernel cannot load the new kernel modules, so you will need a reboot to execute in the new kernel version
2023-09-10 15:58:40 <fancsali[m]> That was my hunch too, but I did reboot after the last update. 
2023-09-10 15:59:22 <caskd> what happens if you try executing 'modprobe nf_tables'
2023-09-10 15:59:29 <fancsali[m]> Let me check my kernel version vs /lib/modules again...
2023-09-10 16:00:44 <caskd> try the same with nf_masq
2023-09-10 16:01:15 <caskd> there seems to be some modules that either got corrupted or are straight up missing for some reason
2023-09-10 16:03:36 <fancsali[m]> Well, it's already loaded, that's the funny thing. At least lsmod says so...
2023-09-10 16:04:43 <fancsali[m]> But nf_masq cannot be found in /lib/modules/6.1.27... 🤔
2023-09-10 16:04:55 <caskd> aha, okay so those are missing then
2023-09-10 16:05:02 <fancsali[m]> Btw uname says, that's the correct version...
2023-09-10 16:05:16 <fancsali[m]> Well, nf_tables is there, and loaded, as said before...
2023-09-10 16:05:23 <caskd> one second
2023-09-10 16:06:32 <caskd> what flavour are you running?
2023-09-10 16:07:36 <caskd> i cannot find that version anywhere
2023-09-10 16:08:05 <fancsali[m]> To be fair, this is a Google Veyron Minnie with pmOS.
2023-09-10 16:08:14 <caskd> oh
2023-09-10 16:08:30 <fancsali[m]> I asked here first, because dnsmasq is from aports...
2023-09-10 16:09:02 <caskd> apk info -L linux-FLAVOUR | grep nft_masq
2023-09-10 16:09:04 <caskd> try this
2023-09-10 16:09:21 <fancsali[m]> Good idea, let me have a butchers...
2023-09-10 16:10:05 <caskd> if it returns nothing then it means pmOS devs cut down their config and removed the feature you rely on, so you should go have a chatter with them
2023-09-10 16:10:40 <fancsali[m]> Yeah, got your thinking. However:... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/DAwntRcyEftCOOkqQKDmiZkP>)
2023-09-10 16:11:20 <caskd> try apk fix linux-google-veyron then
2023-09-10 16:11:50 <caskd> something must've went wrong during the package upgrade/install
2023-09-10 16:12:13 <caskd> possibly ran out of storage space?
2023-09-10 16:15:16 <fancsali[m]> Hm. Funnily it doesn't emit any error message, runs fine - but in the end it says "1 error". 🤔
2023-09-10 16:15:43 <caskd> what does `apk fix -s` show?
2023-09-10 16:21:02 <fancsali[m]> You mean apk fix -s linux-google-veyron?
2023-09-10 16:21:19 <ikke> just apk fix
2023-09-10 16:22:21 <fancsali[m]> Just plain `apk fix -s` complains about `obexd-enhanced`; I think two `obex-*` packages are trying to claim the same udev rule...
2023-09-10 16:22:21 <fancsali[m]> ... I've spotted that a while ago, but I do hope this is not caused by that...
2023-09-10 16:23:31 <fancsali[m]> Doing `apk fix`, would try to reinstall `obexd-enhanced` and complains about this:... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/IYyJKPhSOcXyufWwijznvHVx>)
2023-09-10 16:23:48 <fancsali[m]> Is that perhaps the single error I am seeing?
2023-09-10 16:25:22 <ikke> yes
2023-09-10 16:25:37 <ikke> but that should not cause issues with linux-google-veyron
2023-09-10 16:26:52 <caskd> yeah
2023-09-10 16:27:36 <caskd> so does modprobe nft_masq work now in that case?
2023-09-10 16:30:45 <fancsali[m]> Yapp; I think we were trying nf_masq before... ☝️
2023-09-10 16:31:43 <caskd> okay, so it was most likely just that something must've went wrong and those files were either never extracted(installed) or just vanished
2023-09-10 16:31:54 <caskd> it should hopefully work now
2023-09-10 16:32:29 <caskd> oh wait
2023-09-10 16:32:30 <fancsali[m]> Well, as I said, I was trying nf_masq and not nft_masq before...
2023-09-10 16:32:37 <caskd> that was my bad
2023-09-10 16:33:17 <fancsali[m]> Never mind, I should've spotted it; but the point is, the modules are now there, for sure, and the issue is still there...
2023-09-10 16:38:45 <caskd> this is beyond me but maybe see if the ipt_tcp module is present too
2023-09-10 16:39:09 <caskd> i am not sure if the nftables iptables shim translates everything 1:1 
2023-09-10 16:51:23 <fancsali[m]> Hm, interesting proposal...
2023-09-10 16:51:23 <fancsali[m]> ... that might be a contributing factor; but then again, it seemed to have been working before...
2023-09-10 16:52:16 <caskd> well yeah, but since probably the shim and kernel have updated
2023-09-10 16:52:33 <caskd> so now it's important to figure out if the shim misbehaves or if the kernel is missing what the shim needs
2023-09-10 16:56:22 <fancsali[m]> Makes sense...
2023-09-10 16:56:22 <fancsali[m]> ... I was just hoping, someone has seen this before, and knows which one is the case. ;)
2023-09-10 16:58:33 <fancsali[m]> Or it could be the lxc-bridge package, that owns /etc/init.d/dnsmasq.lxcbr0
2023-09-10 16:58:49 <caskd> the problem isn't really that, it's the iptables call
2023-09-10 16:58:54 <caskd> which adds the firewall rules
2023-09-10 16:59:14 <caskd> i guess you could disable those but i cannot guarantee stuff will work as before if you do it
2023-09-10 17:00:22 <fancsali[m]> My bad, the error messages seem to be emitted by the iptables calls, not modprobe...
2023-09-10 17:01:25 <fancsali[m]> Where does the shim reside? Kenel module, or is it just a plain binary?
2023-09-10 17:01:31 <caskd> it's a binary
2023-09-10 17:01:34 <fancsali[m]> * just a "plain, * plain" binary?
2023-09-10 17:01:39 <caskd> yeah
2023-09-10 17:02:51 <fancsali[m]> Let me guess, it's the iptables package... ;)
2023-09-10 17:02:56 <caskd> yep
2023-09-10 17:03:26 <fancsali[m]> I shall play the lottery... :D
2023-09-10 17:03:48 <caskd> iptables-legacy is still the old iptables
2023-09-10 17:04:00 <caskd> iptables by default is now iptables-nft
2023-09-10 17:04:03 <caskd> which is the shim
2023-09-10 17:05:18 <fancsali[m]> Got ya!
2023-09-10 17:16:39 <fancsali[m]> That's odd: linux-google-veyron doesn't seem to have changed recently; at least on my end. In the repo there seems to be a 6.1.39 available. But apk -U upgrade won't consider it...
2023-09-10 17:22:12 <ikke> what does `apk policy linux-google-veyron` return (and place do not use the matrix multi-line feature)
2023-09-10 17:38:29 <fancsali[m]> Hm, 6.1.27-r0: lib/apk/db/installed and my local repo; and then 6.1.39-r0: http://mirror.postmarketos.org...
2023-09-10 17:39:06 <fancsali[m]> Odd, 6.1.27-r1 should (also) be present, that's something I've contributed a while ago... 
2023-09-10 17:41:36 <fancsali[m]> Would a locally present and installed package always be higher priority?
2023-09-10 17:41:53 <ikke> no
2023-09-10 17:42:02 <ikke> otherwise apk would never upgrade a package
2023-09-10 17:42:23 <ikke> Check the repos if that package / version is present
2023-09-10 17:45:00 <fancsali[m]> Yes, it was there. I removed it from my local dev repo, and rebuilt the indexes. Now I only have 6.1.27-r0: lib/apk/db/installed and below that 6.1.39: http://...
2023-09-10 17:45:22 <fancsali[m]> Running apk -U upgrade still won't try to actually upgrade...
2023-09-10 17:45:30 <fancsali[m]> What am I missing?
2023-09-10 17:45:33 <ikke> You can use a paste service like https://tpaste.us or https://ix.io
2023-09-10 17:45:51 <ikke> if apk policy doesn't return it, it means that there is no repo configured with that version
2023-09-10 17:46:37 <fancsali[m]> ikke: it does, that's the funny thing.
2023-09-10 17:47:00 <ikke> Where is the APKBUILD for that package?
2023-09-10 17:47:46 <fancsali[m]> The policy says this: https://pastebin.com/83EYZj92.
2023-09-10 17:48:08 <fancsali[m]> ikke: The APKBUILD is in the pmOS repos, why?
2023-09-10 17:48:17 <fancsali[m]> Also in my local mirror, which I just removed.
2023-09-10 17:48:29 <ikke> So that I can check what's inside it (and don't have to figure out where it's located)
2023-09-10 17:49:01 <ikke> fancsali[m]: can you try: `apk add 'linux-google-veyron>6.1.39'`
2023-09-10 17:51:18 <fancsali[m]> Well, nothing happens...
2023-09-10 17:52:26 <ikke> you did include the single quotes, right?
2023-09-10 17:53:25 <fancsali[m]> 🤦
2023-09-10 17:53:59 <ikke> :D
2023-09-10 17:54:35 <fancsali[m]> Let me just hide in a corner, roll over and start crying in embarrassment...
2023-09-10 17:54:40 <fancsali[m]> Will be back soon... :D
2023-09-10 18:05:38 <fancsali[m]> Got it; no change in the error message, though. Others claim, it would need iptables-legacy.
2023-09-10 18:05:48 <fancsali[m]> Otoh, I am still confused, why I had to specify the version...
2023-09-10 19:04:13 <frag> anyone got switching tab in sakura to work?
2023-09-10 19:22:31 <kwadroke[m]> frag: works on my Fedora system, think it's fighting with JWM on my Alpine system
2023-09-10 19:23:17 <kwadroke[m]> ALT-LeftArrow & ALT-Right arrow work on sakura on Fedora with IceWM
2023-09-10 19:23:34 <kwadroke[m]> JWM on Alpine wants to change workspaces
2023-09-10 19:26:53 <frag> im on ratpoison, so would think it doesnt interfere .. also C+1-9 doesnt work ..
2023-09-10 19:28:14 <kwadroke[m]> C+1-9 doesn't work for either of my systems, ALT+1-9 does on Fedora/IceWM
2023-09-10 19:28:35 <kwadroke[m]> JWM also wants to change workspaces with those keys
2023-09-10 19:29:31 <frag> very strange ...
2023-09-10 19:30:20 <kwadroke[m]> don't believe I'm doing any custom keys for sakura
2023-09-10 19:30:28 <kwadroke[m]> I know I'm not on Alpine/JWM
2023-09-10 19:33:00 <kwadroke[m]> do you have "disable numbered tabswitch" checked in the right-click options?
2023-09-10 19:33:22 <frag> kwadroke[m]: aha! changed switch_tab_accelerator=4 from 8, and ctrl+1-9 seems to work
2023-09-10 19:33:34 <kwadroke[m]> cool
2023-09-10 19:34:33 <kwadroke[m]> hmm, my fedora system has that set and ALT+1-9 is the keys
2023-09-10 19:34:41 <kwadroke[m]> set to 8 that is
2023-09-10 19:35:13 <kwadroke[m]> think I'm using the defaults that was packaged with sakura
2023-09-10 19:35:50 <kwadroke[m]> only thing I changed was the colors
2023-09-10 19:36:21 <kwadroke[m]> but good that you found the issue
2023-09-10 19:36:46 <frag> kwadroke[m]: yeah, seems switch_tab_accelerator sets it for both 1-9 and the arrow keys, even thought the defaults are supposed to be ctrl+1-9 and alt+left/right
2023-09-11 14:20:07 <eletrotupi> does anybody else in here uses slack via flatpak or some other method? I've been using it via firefox, but folks at work are using huddle more and more and it refuses to work on firefox and on chromium is just too painfully slow to use it
2023-09-11 15:26:21 <Saijin_Naib[m]> Yeah, I do 
2023-09-11 15:26:44 <Saijin_Naib[m]> It can be a nightmare to activate your account with the magic token 
2023-09-11 15:30:37 <Saijin_Naib[m]> https://www.reddit.com/r/Slack/comments/q42gk2/comment/hfwh2vv/
2023-09-11 18:05:07 <frag> does alpine has categories in repo, so one category for terminals? and what terminal are you using?
2023-09-11 18:06:18 <ikke> no, there are not categories
2023-09-11 18:06:20 <ikke> no*
2023-09-11 18:06:45 <frag> ah okok
2023-09-11 19:45:50 <frag> hmm trying to run gropdf from the groff package, getting : /bin/ash: /usr/bin/gropdf: not found (even thought the file is there)
2023-09-11 19:46:05 <frag> its 84K too..
2023-09-11 19:50:30 <quinq> frag, try apk fix
2023-09-11 19:51:13 <ikke> ldd /usr/bin/gropdf
2023-09-11 19:53:53 <frag> /lib/ld-musl-aarch64.so.1: /usr/bin/gropdf: Not a valid dynamic program
2023-09-11 19:54:12 <ikke> file /usr/bin/gropdf
2023-09-11 19:54:46 <frag> /usr/bin/gropdf: Perl script text executable
2023-09-11 19:54:53 <frag> hm, i dont have perl installed
2023-09-11 19:55:15 <ikke> That would explain
2023-09-11 19:55:26 <frag> mhmhm
2023-09-12 14:37:55 <ovf> i see there's a criu package, but the linux-edge config seems to be missing some required options, e.g. CONFIG_PROC_CHILDREN. should these be enabled, or is there a different kernel (why?) for use with criu?
2023-09-12 14:41:36 <minimal> ovf: linux-lts?
2023-09-12 14:41:47 <zcrayfish> The linux-edge maintainer doesn't maintain parity with -lts (I don't think it has debugfs either), so your mileage will vary
2023-09-12 14:48:01 <ovf> thanks. i wonder why the disparity.
2023-09-12 14:48:49 <zcrayfish> idk, I never got around to asking
2023-09-12 14:49:17 <minimal> ovf: because linux-edge is maintained by a different person that linux-lts/linux-virt and does their own thing regarding the linux-edge configuration
2023-09-12 14:57:18 <zcrayfish> I thought it was cool when he enabled the config option to fully support my bizarro laptop ( https://gitlab.alpinelinux.org/alpine/aports/-/issues/15161 ), so I guess it doesn't hurt to file an issue 
2023-09-13 11:09:50 <armin> hm, when using "startx" from a tty, i don't have mouse and keyboard input under Xorg, however, when i use "lightdm" to log in instead, i do. any help?
2023-09-13 11:12:32 <quinq> quite strange
2023-09-13 11:16:20 <armin> oh and when doing "sudo /etc/init.d/lightdm stop" Xorg seems to hang in a restart-loop.
2023-09-13 11:16:26 <armin> eww.
2023-09-13 11:21:02 <ikke> armin: is your user member of the input group?
2023-09-13 11:21:54 <armin> ikke: now it is, let me retry.
2023-09-13 11:22:54 <armin> also the default font was extremely tiny, it was actually hard to install. :(
2023-09-13 11:23:43 <armin> ikke: that was it, thank you.
2023-09-13 11:23:48 <armin> :)
2023-09-13 12:13:34 <armin> so how can i find out what package contains the "pulseaudio" binary? the pulseaudio package doesn't seem to.
2023-09-13 12:16:50 <pj> apk search cmd:pulseaudio
2023-09-13 12:18:27 <armin> pj: says pulseaudio and pulseaudio-ctl, and i have both installed.
2023-09-13 12:19:35 <armin> hmmm.
2023-09-13 12:20:10 <pj> I installed pulseaudio package and I have pulseaudio
2023-09-13 12:20:15 <pj> CLI
2023-09-13 12:20:26 <armin> what does "which pulseaudio" say?
2023-09-13 12:21:04 <pj> I don't have "which"
2023-09-13 12:21:18 <pj> but it's /usr/bin/pulseaudio
2023-09-13 12:21:51 <armin> armin@skull:~ sudo apk add pulseaudio
2023-09-13 12:21:51 <armin> OK: 2490 MiB in 582 packages
2023-09-13 12:21:51 <armin> ls: /usr/bin/pulseaudio: No such file or directory
2023-09-13 12:21:51 <armin> armin@skull:~ ls -alrt /usr/bin/pulseaudio
2023-09-13 12:21:51 <armin> armin@skull:~
2023-09-13 12:22:15 <pj> do: apk fix --reinstall pulseaudio
2023-09-13 12:22:29 <pj> which alpine version do you use?
2023-09-13 12:22:56 <armin> 3.18.3
2023-09-13 12:23:08 <armin> same thing, still no /usr/bin/pulseaudio:
2023-09-13 12:23:14 <minimal> pj: isn't "which" provided by busybox?
2023-09-13 12:23:48 <pj> probably, I remember not having it
2023-09-13 12:24:14 <minimal> you don't have busybox?
2023-09-13 12:24:16 <pj> you should always recommend `command -v <name>` or `type [-a] <name>` though
2023-09-13 12:24:24 <pj> minimal: which
2023-09-13 12:24:45 <minimal> pj: /usr/bin/which is just a softlink to /bin/busybox
2023-09-13 12:24:51 <pj> great
2023-09-13 12:25:10 <minimal> so it is strange if you don't have that softlink
2023-09-13 12:25:11 <pj> that doesn't change the fact that I had a situation where I didn't have "which"
2023-09-13 12:26:35 <minimal> as such softlinks are generally only removed when you install a package containing a "full fat" version of an app, and are reinstated whenever you uninstall the "full fat" package
2023-09-13 12:27:05 <pj> I'm aware of how the package works
2023-09-13 12:27:26 <armin> sudo apk del pipewire-pulse pulseaudio pulseaudio-bluez pavucontrol; sudo apk add pulseaudio <- this here worked.
2023-09-13 12:27:34 <minimal> and I'm just pointing out that the non-existence of that softlink is strange/unusual
2023-09-13 12:27:55 <pj> And I'm telling you it's not about the softlink but about not existing which command
2023-09-13 12:28:20 <pj> this also could have happened on system that wasn't alpine
2023-09-13 12:28:29 <minimal> eh? if that softlink exists then running "which <whichever prog>" works
2023-09-13 12:28:45 <minimal> I'm not talking about which vs command -v etc
2023-09-13 12:29:16 <pj> I know
2023-09-13 12:29:26 <pj> You keep talking about specific thing about busybox and alpine
2023-09-13 12:29:52 <pj> I don't even remember if it was alpine thing or not 
2023-09-13 12:30:07 <minimal> so I don't understand it when you say "it's not about the softlink but about not existing which command" as if /usr/bin/which exists as a softlink to /bin/busybox then "which" command does exist
2023-09-13 12:30:15 <pj> I don't use which for years, because I learned it's not available on every system
2023-09-13 12:30:17 <pj> and I recommend something that exists on every system
2023-09-13 12:30:49 <minimal> that's a different matter however
2023-09-13 12:31:31 <pj> minimal: I typed "which", I got "which: not found", does that answer the problem?
2023-09-13 12:32:05 <minimal> well you already indicated you didn't have the /usr/bin/which softlink so that's hardly a surprise
2023-09-13 12:32:28 <minimal> as I said already, it is just strange that the softlink doesn't exist
2023-09-13 12:32:56 <pj> It could have been not even Alpine
2023-09-13 12:33:02 <pj> so the softlink doesn't matter here
2023-09-13 12:33:43 <minimal> pj: we're talking in an Alpine channel about Alpine however
2023-09-13 12:34:09 <pj> and we're on earth so we can talk about earth only
2023-09-13 12:34:15 <pj> please
2023-09-13 12:34:44 <minimal> you were helping someone with an Alpine problem, he asked you to run which, you said you didn't have it...all in the course of helping with an Alpine problem
2023-09-13 12:35:48 <pj> I can not even use Alpine and still be here and help people
2023-09-13 12:36:11 <pj> One can just lookup package content on pkgs.a.o
2023-09-13 13:18:25 <armin> i have absolutely no clue how to build this here, though (yet): https://github.com/Oncorporation/obs-shaderfilter.git - looked at the PKGBUILD for the arch package, but that didn't actually help much...
2023-09-13 13:18:57 <armin> *should* be just cmake/make, but i'm getting lost somewhere.
2023-09-13 14:03:12 <j_v> armin: looks like a ms windows project. and the CMakeLists.txt seems to confirm it
2023-09-13 14:06:09 <armin> j_v: i just succeeded in compiling that, though.
2023-09-13 14:06:47 <armin> (cmake is a bit weird!)
2023-09-13 14:07:31 <j_v> ok, i just noticed that the CMakeLists.txt only added the sources in a if win32 block
2023-09-13 16:46:13 <armin> damnit, i'm trying to build obs with browser source plugin added since an hour or so, but i have no success. :)
2023-09-13 16:47:10 <armin> got the aports repo with the APKBUILD file and whatnot, thought that adding "BUILD_BROWSER" and "ENABLE_BROWSER" would at least show some effect but it doesn't.
2023-09-13 16:47:39 <armin> so basically running "abuild -r" every time to see if that did it, but whatever i try is obviously wrong.
2023-09-13 16:47:50 <armin> <- newb.
2023-09-13 16:48:12 <armin> any pointer super appreciated. <3
2023-09-13 16:48:47 <ikke> armin: can you show the APKBUILD?
2023-09-13 16:49:01 <armin> i think? 1sec.
2023-09-13 16:49:38 <armin> https://paste.debian.net/plain/1291847
2023-09-13 16:51:21 <kattjevfel> How would I go about overriding something in a /etc/conf.d file? like the user:group for a service
2023-09-13 16:51:23 <armin> ikke: i could well be overlooking something completely obvious, i'm by no means experienced or something.
2023-09-13 16:52:02 <ikke> armin: seems like you need CEF_ROOT_DIR as well
2023-09-13 16:52:14 <ikke> cmake/Modules/FindCEF.cmake
2023-09-13 16:52:17 <armin> ikke: ok!? will read up on that i guess?
2023-09-13 16:52:27 <minimal> kattjevfel: you'd edit the conf.d file to change the value used
2023-09-13 16:52:31 <ikke> Just grepping for BROWSER
2023-09-13 16:52:35 <armin> ikke: ok you just gave me some homework to do :)
2023-09-13 16:52:48 <kattjevfel> minimal: wouldn't that be overwritten by the package being updated?
2023-09-13 16:53:06 <armin> ikke: thank you so much for looking into this. <3
2023-09-13 16:54:03 <minimal> nope, if newer version of package has different contents of that file then a new file with a different extension (forget which) would be placed in /etc/conf.d/ directory
2023-09-13 16:54:14 <minimal> so it wouldn't overwrite
2023-09-13 16:54:32 <ikke> if it's directly set in the init.d file, then setting it in conf.d would not override it though
2023-09-13 16:55:05 <kattjevfel> ikke: it's set in the conf.d file https://git.alpinelinux.org/aports/tree/community/qbittorrent/qbittorrent-nox.confd?h=3.18-stable
2023-09-13 16:55:12 <kattjevfel> anyway thanks minimal, that does make sense :)
2023-09-13 16:55:18 <ikke> then what minimal says indeed
2023-09-13 16:55:24 <minimal> ikke: if it's in the conf.d file then I'm assuming the init.d would source the conf.d and only set a value of the conf.d didn't
2023-09-13 16:55:26 <kattjevfel> thanks both of you :)
2023-09-13 16:55:47 <ikke> the conf.d is automatically sourced by openrc-run
2023-09-13 16:56:50 <ikke> You can set options mentioned in https://www.mankier.com/8/openrc-run without explicit init.d support
2023-09-13 16:57:06 <ikke> but if the init.d sets those options itself, then it would overwrite it
2023-09-13 16:57:14 <minimal> ikke: ok, what I mean is that I'd expect the init.d to only set a value if one was not already set
2023-09-13 16:57:48 <minimal> which is what many/most init.d files do
2023-09-13 16:58:43 <ikke> many directly set things like command_user
2023-09-13 16:58:51 <ikke> not like https://git.alpinelinux.org/aports/tree/community/qbittorrent/qbittorrent-nox.initd?h=3.18-stable#n8
2023-09-14 03:51:38 <Naruni> hello, i have a fresh install in sys mode. trying to add package xauth which exists here: https://pkgs.alpinelinux.org/package/edge/community/x86_64/xauth but 'apk search xauth' returns "strongswan-5.9.11-r1"
2023-09-14 03:52:25 <Naruni> i changed mirrors and did a few 'apk update' in case there was a pointing issue but so far everything is the same
2023-09-14 03:52:54 <Naruni> so i tried 'apk add strongswan' but xauth was not installed and followed it up with 'which xauth' with no returns
2023-09-14 03:52:57 <zcrayfish> can you pastebin the contents of /etc/apk/repositories ?
2023-09-14 03:53:23 <Naruni> aaaaand i just realized community repo is commented out
2023-09-14 03:53:39 <Naruni> sorry about that, and thanks zcrayfish
2023-09-14 03:53:59 <zcrayfish> \o/
2023-09-14 03:54:08 <zcrayfish> all good!
2023-09-14 03:54:16 <Naruni> i'm new to alpine so that's my one dumb question used :)
2023-09-14 03:55:00 <Naruni> cya
2023-09-14 06:27:51 <Wumbo> hi
2023-09-14 06:38:34 <sicelo> hi, i would like to get my networkmanager patch into alpine, https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/merge_requests/1731/
2023-09-14 06:39:48 <sicelo> i need it for my postmarketOS device (Nokia N900). what is the best (and quickest) way to get it done? open an issue in alpine gitlab? or submit a merge request there?
2023-09-14 06:44:34 <sicelo> or wait for upstream (i wonder if this won't take long)
2023-09-14 06:51:49 <ikke> In preferred order: wait for upstream, open an MR, create an issue
2023-09-14 07:09:31 <sicelo> ok. let me wait to see any comments on the upstream MR. if they are OK with it, then i'll make an MR backporting it to alpine's NM 1.42.8, since i would like to get it in pmOS asap. even if accepted upstream, it'd obviously only be available in some future tagged version
2023-09-14 09:06:40 <pki> Hi there
2023-09-14 09:07:50 <pki> I see the officiall kernel is switched to i686 and requires PAE, what is the last version which uses i586 without PAE as i need it for ALIX boards with AMD Geode CPU? Anyone remember? :)
2023-09-14 09:20:44 <jophar> Hello, it's been a while since I last used an IRC channel. I am currently using Alpine on an RPI Zero W v1.1 board (armv6) which is by far the best distro in matters of speed and responsiveness. Happy with that. That said, I'm having problems with bcm2835 initialization so I can read values from sensors. I've tried to run as root but even then it says "bcm2835_init: gpio mmap failed: 
2023-09-14 09:20:44 <jophar> Operation not permitted". The same code works in the DietPi distro (which is Raspbian-based and SUPER slow). I am using nodejs to create an endpoint to send sensor data. Is there any flag on the kernel that needs to be activated or something? Basic "Led on, led off" works with the onoff package but the node-dht-sensor gives this error. Any thoughts?
2023-09-14 09:35:12 <Tach> Hi guys! Is the amavis package for alpine the  amavisd-new one by default ?
2023-09-14 10:22:32 <ikke> Tach: https://gitlab.alpinelinux.org/alpine/aports/-/blob/master/main/amavis/APKBUILD#L21
2023-09-14 10:22:36 <ikke> does that answer your question?
2023-09-14 10:50:44 <Tach> ikke yes and now... I found out on gitlab amavis is 200% amavis-new now
2023-09-14 10:50:51 <Tach> alreayd for some time but still
2023-09-14 10:50:54 <Tach> thanks!
2023-09-14 13:07:58 <WhyNotHugo> Is there any way to connect expose a vsock via a unix socket?
2023-09-14 13:08:09 <WhyNotHugo> E.g.: given a vsock, create a unix socket which transparently forwards traffic to the vsock.
2023-09-14 13:10:15 <ikke> perhaps socat?
2023-09-14 13:10:30 <ikke> though not sure it supports vosck
2023-09-14 13:10:32 <ikke> vsock
2023-09-14 13:11:08 <ikke> https://man.archlinux.org/man/socat.1 mentions vosck
2023-09-14 13:18:27 <WhyNotHugo> I was rather looking for some sort of syscall where one is plugged into the other (rather than read/writing each message)
2023-09-14 15:37:51 <xordspar0> WhyNotHugo: Hm, neither sendfile(2) nor splice(2) work between two sockets, do they?
2023-09-14 15:45:46 <xordspar0> pki: PAE was reverted for the default kernel: https://gitlab.alpinelinux.org/alpine/aports/-/commit/e81a0ff4246ffa3cc57dca06225c83fecafcfc5a
2023-09-14 16:51:49 <ilyaigpetrov> Hello. Is it a bad idea to use "data" installation type on a machine to be used as a server?
2023-09-14 16:54:35 <ilyaigpetrov> If disk is used as data only then, I guess, I will have to keep all installed packages on a usb. And maybe usb will be actively used by a server so it may wear off.
2023-09-14 17:11:53 <ilyaigpetrov> I also have another question. 2) I'm installing alpine linux from a flash drive, at the end of setup-alpine I input "usb" to setup-lbu, it succeeds but on the next step setup-apkcache fails with "Resource busy" while trying to mount /dev/sdbdisk (points to sdb1) on /media/usb. I guess, sdb1 is used by the booted installer so it's busy.
2023-09-14 17:12:32 <ilyaigpetrov> s/sdbdisk/usbdisk
2023-09-14 19:11:50 <frag> are there other minimalist wms like ratpoison that has notification/prompt/run/primary-copy/paste/wrapping the mouse/++ inbuilt?
2023-09-14 20:51:21 <lopid> that stuff is beyond mere window management
2023-09-14 22:28:39 <leah2> ion maybe
2023-09-15 00:25:50 <hurd> hello, i have a general question about linux, i'm currently using alpine so i'm asking here
2023-09-15 00:29:43 <hurd> I am writing a file that I'd eventually like to mount directly (i.e. mount /dev/loop0 or mount file.iso)
2023-09-15 00:32:07 <hurd> My process is: touch x.iso -> truncate -s 100M x.iso -> losetup -fP x.iso -> fdisk x.iso -> mkfs.ext/vfat on multiple partitions
2023-09-15 00:33:30 <hurd> mounting the partitions alone (i.e. mount /dev/loop0p1) works as expected. However mounting the device itself (i.e. /dev/loop0) causes a "cannot mount bad fs" error
2023-09-15 00:35:48 <hurd> I suspect that this is occurring because the first 2048 sectors aren't formatted properly. But I cannot find any guides or programs for identifying this
2023-09-15 00:38:16 <hurd> So I looked through Alpine's .iso and found that it mounts correctly via /dev/loop0 . I also adjusted the partition-scheme, format, and fs-type of Alpine's iso - verified the change through diff - and it was still mountable
2023-09-15 00:39:35 <hurd> All that to say, does anyone know what I am doing wrong that makes my file unmountable?
2023-09-15 00:50:41 <humm> a disk is generally not a file system
2023-09-15 00:51:03 <humm> you probably can’t mount /dev/sda
2023-09-15 01:01:36 <minimal> hurd: mount is for mouting filesystems, obviously on a partitioned device you can't mount the device itself as that is not a filesystem
2023-09-15 01:02:30 <minimal> now if you have a device which you did not partition and just created a single filesystem on that device then yes you could mount it
2023-09-15 01:30:09 <hurd> minimal: would that mean that iso9660 is holding partitions in its filesystem? Like a wrapper?
2023-09-15 01:31:26 <hurd> that would actually make the most sense and explain all of the properties. so that would mean I've been asking the wrong questions
2023-09-15 01:32:36 <minimal> hurd: no, ISO9660 *is* a filesystem
2023-09-15 01:33:08 <minimal> there is no partitioning
2023-09-15 01:33:46 <minimal> though you could partition a device and create a ISO9660 filesystem in one of its partitions
2023-09-15 01:35:16 <minimal> so it is not clear what you are trying to do
2023-09-15 01:41:45 <hurd> minimal: yeah, that makes sense. on the same note, the alpine.iso has a MBR with two partitions (EFI and Empty), it seems to be inside the iso9660 filesystem (as I can mount /dev/loop0 directly)
2023-09-15 01:43:57 <hurd> minimal: everything makes sense now though, thanks for the help - it was hard to find direct answers online
2023-09-15 01:44:16 <hurd> especially since I was thinking about it differently
2023-09-15 01:44:37 <minimal> hurn: the Alpine ISO is a "hybrid" image I believe
2023-09-15 01:45:01 <minimal> hurd: ^^^
2023-09-15 01:47:58 <hurd> minimal: yeah, for the boot process. The structure is " iso9660 (/dev/loop0) [ iso9660 (/dev/loop0p1) [ apk/  boot/ ] && vfat (/dev/loop0p2) [ efi/ ] ] "
2023-09-15 01:49:35 <hurd> This explains why when mounting /dev/loop0 all of apk/ boot/ efi/ were accessible. And why it was a continuous filesystem despite being on separate partitions. Also why it was able to mount
2023-09-15 01:50:33 <minimal> I assume it is a hybrid GPT/MBR, so it uses ISO9660 for BIOS booting, and Grub for UEFI booting
2023-09-15 01:50:58 <hurd> when booting 99% of BIOS and UEFI can mount iso9660, and the rest is history
2023-09-15 01:51:14 <hurd> minimal: yeah agreed
2023-09-15 01:51:43 <minimal> normally a device is either MBR or GPT partitioning
2023-09-15 01:52:16 <minimal> the hybrid is a funky way to create something that boots on BIOS-only and UEFI-only machines
2023-09-15 01:53:09 <minimal> otherwise you'd need 2 CDs/USB sticks - one for BIOS booting, the other for UEFI
2023-09-15 01:59:36 <hurd> minimal: I'm gonna head out, thanks again for the help
2023-09-15 06:34:36 <ilyaigpetrov> Is https://gitlab.alpinelinux.org/alpine/aports/-/issues/11589 (diskless mode with btrfs) still actual?
2023-09-15 06:41:44 <livy> Hello. I have trouble sharing a directory with virtual machine using virt-viewer WebDAV. My QEMU invocation is:
2023-09-15 06:41:53 <livy> -chardev spiceport,id="spiceport-webdav0",name=org.spice-space.webdav.0 \
2023-09-15 06:41:58 <livy> -device virtserialport,id="webdav0",name=org.spice-space.webdav.0,bus="virtio_serial0.0",nr=2,chardev="spiceport-webdav0" \
2023-09-15 06:42:36 <livy> It works on Ubuntu, but on Alpine it does not. virt-viewer grays out the "Share folder" checkbox.
2023-09-15 06:43:31 <livy> Did I miss some package or anything?
2023-09-15 06:45:56 <livy> My host is Alpine, and guest is Windoze (with Spice WebDAV driver installed).
2023-09-15 08:52:39 <ahmadraniri[m]> Hello, I got  kernel update on stable version and facing kernel panic. Are you guys also facing the same thing ?
2023-09-15 08:54:17 <lopid> and no way to revert to the previous kernel?
2023-09-15 10:36:19 <ahmadraniri[m]> My bad, dracut seems messed up... Load kernel via grub shell works fine.
2023-09-15 14:48:07 <frag> https://vid.pr0gramm.com/2023/09/15/15335fc0587359aa.mp4
2023-09-15 14:50:28 <lopid> no smut
2023-09-15 14:52:47 <frag> its clean
2023-09-15 14:53:09 <lopid> oh :(
2023-09-15 14:53:17 <frag> sorry ..
2023-09-15 15:02:56 <strajder> it's a naughty vid
2023-09-15 15:32:18 <openlinux21[m]> 
ACTION uploaded an image: (121KiB) < https://matrix.org/_matrix/media/v3/download/matrix.org/fttvxXxWtukMGTJNatpqFrHJ/ima_0bc5439.jpeg.jpg >

2023-09-15 15:37:47 <frag> nice setup
2023-09-15 15:46:24 <eddsalkield> does anyone know how to enable broadcast udp packets on alpine? i'm setting up a software-defined radio and it uses these packets for service discovery
2023-09-15 15:46:47 <eddsalkield> the discovery program is supposed to emit broadcast packets, but tcpdump can't find them
2023-09-15 15:58:53 <itwrx> hi, my php apps on one server quit being able to send email via starttls.
2023-09-15 15:59:06 <itwrx> i think maybe due to upgrade to 3.17 and openssl 3.0? 
2023-09-15 15:59:18 <itwrx> I notice php -i says openssl header is 3.0, but lib version is 1.1.
2023-09-15 15:59:38 <itwrx> openssl1.1-compat is installed. Is that the likely problem? Is it safe to remove openssl1.1-compat?
2023-09-15 16:38:37 <nu[m]1> /join #_oftc_#ceph:matrix.org
2023-09-15 16:43:18 <numono> thanks hydrogen ^^"
2023-09-15 17:54:36 <itwrx> loading the php-curl ext later made the header and lib versions match. still can't send email but i'll try to play with that some more...
2023-09-15 18:04:38 <itwrx> i meant to ask: how foolish would it be to change my apk repos to 3.16 to downgrade from 3.17? :) would it work or explode?
2023-09-15 18:07:48 <vova> your computer would explode
2023-09-15 18:08:04 <vova> joke aside I think it's fine 3.16 is still supported 
2023-09-15 18:08:35 <ikke> itwrx: most likely will work (don't forget to provide --available to the upgrade command)
2023-09-15 18:12:40 <itwrx> thansk people
2023-09-15 20:15:50 <frag> not to brag but only have 31 packages in world now: alpine-base aria2 cryptsetup curl doas e2fsprogs feh ffmpeg grep imagemagick irssi isync links-graphics linux-rpi4 mblaze mpv openssh qutebrowser raspberrypi raspberrypi-bootloader ratpoison rsync sfeed slock st telegram-desktop xf86-input-libinput xorg-server xsel yt-dlp zathura-pdf-mupdf
2023-09-15 20:18:30 <ikke> nice and slim
2023-09-15 20:37:48 <jacksonchen666> noticed a lot of packages that have log files but not logrotate config. i've already created 4 issues for each 4 packages. should i continue doing that for the rest or how else? 
2023-09-15 21:11:37 <lopid> is that desired?
2023-09-15 21:15:33 <jacksonchen666> i don't think hundreds of megabytes of logs left persisted on disk is desired
2023-09-15 21:21:22 <lopid> i might not want configs for unrelated packages. i'll handle any logs how i see fit
2023-09-15 21:24:09 <markand> o/
2023-09-15 21:24:59 <markand> I have a resume problem on my thinkpad x1 carbon (4gen), it suspends and resume fines but after 3-4 seconds it completely freezes and is no longer pingable
2023-09-15 21:25:12 <markand> the only last messages I can see are: http://markand.fr/static/alpine-crash.jpeg
2023-09-15 21:31:26 <dwfreed> lopid: the packages themselves can ship a corresponding logrotate config
2023-09-15 22:38:22 <omni> jacksonchen666: are you opening an issue for every aport that doesn't ship with a preconfigured logrotate file?
2023-09-15 22:40:12 <jacksonchen666> omni: yes (with packages i hve), and it doesn't seem like logrotate has a default fallback (checked --debug, doesn't contain all)
2023-09-15 22:40:58 <omni> maybe there should be a single issue with a list, idk
2023-09-15 22:41:50 <jacksonchen666> not sure either, but i've exhausted all log files that doesn't seem to be handled by logrotate
2023-09-15 22:46:34 <minimal> jacksonchen666: I think the underlying issue is whether logrotate is considered the Alpine "official" way to rotate logfiles or not
2023-09-15 22:46:41 <minimal> as there are alternatives
2023-09-15 22:48:17 <minimal> as if there is not officially-defined mechanism then it is hard to expect various packages to implement logrotate (and/or other utility) config files
2023-09-15 22:54:26 <omni> a quick search and it looks like 6 out of 7 aports that come with a .initd file doesn't come with a logrotate config
2023-09-15 22:55:06 <minimal> omni: it's not that simple - some of those apps could perhaps use syslog rather than writing directly to files
2023-09-15 22:55:27 <omni> right
2023-09-15 23:09:43 <omni> I just need to install all the packages and start all the services
2023-09-15 23:10:11 <minimal> omni: and enable all possible configs for all packages ;-)
2023-09-15 23:12:57 <omni> yes! what a glorious monster it'll be!
2023-09-16 10:38:21 <frag> running diskless, can i umount the boot medium (/media/mmcblk0p1/) ? need to run fsck...
2023-09-16 10:38:31 <frag> it says its "busy"
2023-09-16 10:53:30 <zcrayfish> dunno... lazy umount?
2023-09-16 10:53:53 <ikke> rc-service modloop stop
2023-09-16 10:54:16 <frag> what is modloop for?
2023-09-16 10:54:26 <ikke> It contains the kernel modules and firmware
2023-09-16 10:54:34 <ikke> which otherwise would need to live in memory
2023-09-16 10:54:51 <frag> its also 'busy'
2023-09-16 10:57:02 <frag> ill just boot into another media i gues ..
2023-09-16 13:17:40 <julie> hey, ansible seems broken in edge right now
2023-09-16 13:17:52 <julie> ❯ podman run --rm -it alpine:edge
2023-09-16 13:17:52 <julie> # apk --update-cache add ansible
2023-09-16 13:17:52 <julie> fetch https://dl-cdn.alpinelinux.org/alpine/edge/main/x86_64/APKINDEX.tar.gz
2023-09-16 13:17:55 <julie> fetch https://dl-cdn.alpinelinux.org/alpine/edge/community/x86_64/APKINDEX.tar.gz
2023-09-16 13:17:58 <julie> ERROR: unable to select packages:
2023-09-16 13:18:00 <julie>   ansible-core-2.15.4-r0:
2023-09-16 13:18:03 <julie>     breaks: ansible-8.3.0-r0[ansible-core~2.15.3]
2023-09-16 13:18:30 <julie> # there is 8.4.0 in git, but the CDN hasn't been updated yet?
2023-09-16 13:22:00 <julie> also, why does it break? the old package APKBUILD asks for ansible-core~=2.15.3
2023-09-16 13:22:05 <julie> what does the tilde mean?
2023-09-16 13:22:32 <julie> does that mean exact match?
2023-09-16 13:26:56 <ikke> no, it means any component that has not been specificed can very
2023-09-16 13:27:09 <ikke> in this case, 2.15.3-r*
2023-09-16 13:27:10 <julie> oh ok
2023-09-16 13:27:17 <ikke> s/very/vary
2023-09-16 13:27:24 <julie> that makes sense
2023-09-16 13:29:55 <ikke> so ansible 8.3.0 will only work wiith ansible-core 2.15.3, not 2.15.4
2023-09-16 13:30:48 <ikke> the upgrades have been pushed seperately, so one was uploaded before the other
2023-09-16 13:32:36 <julie> the 8.4.0 is shown on https://pkgs.alpinelinux.org/packages but the CDN doesn't have it yet
2023-09-16 13:32:56 <ikke> the builder was stuck, so the package hasn't been built yet / uploaded yet
2023-09-16 13:33:13 <ikke> I've fixed the issue, so should appear soon
2023-09-16 13:34:32 <julie> thank you! <3
2023-09-16 14:34:18 <julie> @ikke it's fixed!
2023-09-16 14:37:45 <ikke> julie: glad to hear
2023-09-16 15:02:45 <CornflakeDeath> hi, i'm trying to run alpine on an allwinner board
2023-09-16 15:03:34 <CornflakeDeath> i keep my rootfs on a usb stick and the default kernel is unable to mount it
2023-09-16 15:04:16 <CornflakeDeath> my custom built one has no problem with it and i think the main difference is that it has usb support built in instead of built as a module
2023-09-16 15:22:52 <frag> is it easy to disable modloop? would be nice to be able to install new system on boot medium from diskless 
2023-09-16 15:24:02 <frag>  /.modloop/ is _only_ 100MB also for me
2023-09-16 15:25:32 <ikke> I think the initramfs will always lood the modloop
2023-09-16 15:26:38 <frag> i see
2023-09-16 15:28:01 <minimal> frag: what exactly are you asking? Do you want to boot "diskless" Alpine from a device and then install Sys-mode Alpine onto that same device?
2023-09-16 15:28:45 <ikke> From what I understood, boot diskless but without modloop, so modules / firmware installed in memory
2023-09-16 15:30:27 <frag> what ikke said
2023-09-16 15:31:26 <minimal> ok, that's not what I understood from "to be able to install new system on boot medium from diskless"
2023-09-16 15:33:19 <minimal> if modloop were to be disabled then even more RAM would be taken up by Alpine and so more likely to cause problems if you ever tried to run update-kernel script
2023-09-16 20:41:35 <lopid> github
2023-09-16 20:41:57 <SmthSweethinher[m]> ai?
2023-09-16 20:42:36 <lopid> no
2023-09-16 20:42:41 <lopid> an
2023-09-17 00:08:05 <Armote[m]> do apparmor profiles apply to binaries which run in different mount namespaces as with containers or other pivot_root tools?
2023-09-17 09:43:41 <conffrey> dwfreed: sorry I messed up, closing the hexchat
2023-09-17 09:44:23 <conffrey> yes, I got the correct path in virtual console
2023-09-17 09:44:47 <conffrey> then to have the same result in X terminals, what have I to change?
2023-09-17 09:45:37 <dwfreed> you should put something like this in your ~/.bashrc (assuming you're using bash): if [ -d $HOME/.local/bin ]; then PATH="$HOME/.local/bin:$PATH"; fi
2023-09-17 09:47:09 <conffrey> dwfreed: thanks, just for learning, I ca understand the second part (then) but I miss the meaning of the 'if', what's about -d option?
2023-09-17 09:49:10 <dwfreed> checks the dir exists first
2023-09-17 09:49:39 <dwfreed> then you can just copy/paste that snippet everywhere, and it'll automatically kick in when you have a ~/.local/bin
2023-09-17 09:50:28 <conffrey> ok, thanks for the explanation, I'll do that
2023-09-17 13:23:29 <tyler[m]123> Why can't I run this program?... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/MqeJMqNFepGCnKrOXhNzhXBI>)
2023-09-17 13:24:10 <ikke> Would it be possible not to use the matrix multi-line feature, it's really anoying for IRC users, especially since the matrix.org url forces downloads
2023-09-17 13:24:15 <tyler[m]123> Btw, I am LD_PRELOADing because Alpine doesn't have libexecinfo anymore
2023-09-17 13:25:21 <Forza_> Hi. I managed to enable edge repo and did an update+upgrade. Is there a way to revert? I removed the edge from repositories config and tried update and upgrade but apk says all is OK
2023-09-17 13:25:36 <ikke> apk upgrade --available
2023-09-17 13:25:46 <Forza_> ikke: thanks 
2023-09-17 13:26:45 <tyler[m]123> tyler[m]123: > <@tyler:matrix.f1gh7.club> Why can't I run this program?... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/smLTrkJuyfbBgXdKJdaKQikP>)
2023-09-17 13:27:32 <ikke> "13:26 <tyler[m]123> tyler[m]123: > <@tyler:matrix.f1gh7.club> Why can't I run this program?... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/smLTrkJuyfbBgXdKJdaKQikP>)"
2023-09-17 13:27:40 <tyler[m]123>  * The code of `libexecinfo.so` (just to resolve the missing symbols - they are basically never used during my runtime anyways):... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/AHMmbFoKdXGkumIwWHFBwMlS>)
2023-09-17 13:27:41 <ikke> that's what I see
2023-09-17 13:27:45 <Forza> tyler[m]123: linux_64_client: not found 
2023-09-17 13:27:52 <Forza> Where is that file? 
2023-09-17 13:28:10 <tyler[m]123> In the current directory. This message generally happens when there is a linker error
2023-09-17 13:28:13 <tyler[m]123> But there is none
2023-09-17 13:28:25 <tyler[m]123> execinfo.c       libexecinfo.so   linux_64_client  linux_64_server
2023-09-17 13:28:25 <tyler[m]123> $ ls
2023-09-17 13:28:46 <Forza> Tyler try "./linux_64_client"
2023-09-17 13:28:56 <Forza> The dot slash is important 
2023-09-17 13:29:00 <tyler[m]123> E.g if you try to load mybinary and it's missing libtest.so, it will say "./mybinary: not found"
2023-09-17 13:29:25 <tyler[m]123> Forza: Same error
2023-09-17 13:29:53 <tyler[m]123> Funny thing is, this was working a couple of days ago
2023-09-17 13:29:56 <tyler[m]123> Idk what happened
2023-09-17 13:29:59 <Forza> And if you run without ld_preload? 
2023-09-17 13:30:18 <tyler[m]123> It will be missing backtrace and backtrace_symbols, which are unavailable on alpine linux
2023-09-17 13:30:44 <tyler[m]123> 
ACTION sent a code block: https://matrix.org/_matrix/media/v3/download/matrix.org/eMWTcXzDiwMTIuSZAbHcPeHo

2023-09-17 13:31:07 <tyler[m]123> That's why I LD_PRELOAD the custom library, for resolving these missing symbols.
2023-09-17 13:31:17 <Forza> Please use a pastebin service 
2023-09-17 13:31:46 <tyler[m]123> Ok
2023-09-17 13:31:55 <Forza> I mean without ldd. 
2023-09-17 13:32:21 <tyler[m]123> It will say the same message, but it makes sense now since it's missing the two symbols
2023-09-17 13:32:41 <tyler[m]123> It doesn't make sense with LD_PRELOAD because all symbols are resolved
2023-09-17 13:33:17 <Forza> And this gives the ash error of missing linux_64_client? `LD_PRELOAD=./libexecinfo.so ./linux_64_client`
2023-09-17 13:33:59 <tyler[m]123> Yes
2023-09-17 13:34:09 <tyler[m]123> $ LD_PRELOAD=./libexecinfo.so ./linux_64_client
2023-09-17 13:34:09 <tyler[m]123> /bin/ash: ./linux_64_client: not found
2023-09-17 13:34:37 <ikke> is the interpreter available in the fist place
2023-09-17 13:35:00 <ikke> afaik, when it says that the file it's not found, it means the interpreter is not there, rather then a needed so lib missing
2023-09-17 13:35:05 <Forza> Is it executable? `chmod +x linux_64_client`
2023-09-17 13:35:28 <Forza> Ikke good point 
2023-09-17 13:35:50 <tyler[m]123> ikke: Interpreter?
2023-09-17 13:35:56 <ikke> tyler[m]123: libc
2023-09-17 13:36:13 <tyler[m]123> I would hope so lol
2023-09-17 13:36:18 <Forza> I wrote a init script that sets up a dm cache device for existing block devices. Would that be something worth making a package for Alpine for? 
2023-09-17 13:36:40 <ikke> tyler[m]123: what does `ldd ./linux_64_client` show?
2023-09-17 13:36:42 <tyler[m]123>  $ file linux_64_client
2023-09-17 13:36:42 <tyler[m]123> linux_64_client: ELF 64-bit LSB executable, x86-64, version 1 (SYSV), dynamically linked, interpreter /lib64/ld-linux-x86-64.so.2, BuildID[sha1]=a9380296e2e744bbc8c7e42027a9985b4c85fd3b, for GNU/Linux 3.2.0, with debug_info, not stripped
2023-09-17 13:36:56 <tyler[m]123> <tyler[m]123> "$ ldd ./linux_64_client..." <- > <@tyler:matrix.f1gh7.club> ```... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/sVpqmuEFScqaacLTiTMnkXsr>)
2023-09-17 13:37:03 <ikke> tyler[m]123: ls -l /lib64/ld-linux-x86-64.so.2
2023-09-17 13:37:36 <Forza> 
ACTION is not happy with the matrix multiline stuff 

2023-09-17 13:37:48 <tyler[m]123> ikke: Now that's a start:... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/TslgWRtNKMIjRZSDSfPLpFKS>)
2023-09-17 13:38:07 <tyler[m]123> > <@_oftc_ikke:matrix.org> tyler: ls -l /lib64/ld-linux-x86-64.so.2... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/wsnRqwFFjpxWvZzZOcIXIMRD>)
2023-09-17 13:38:18 <tyler[m]123> I didn't think that could be possible
2023-09-17 13:38:25 <tyler[m]123> Interestinfg
2023-09-17 13:38:31 <tyler[m]123>  * Interesting
2023-09-17 13:38:54 <ikke> This is built against glibc, so you need something like gcompat to run it on musl systems
2023-09-17 13:39:05 <Forza> tyler[m]123: this is what we see on IRC side https://paste.tnonline.net/files/5SuAiPbOvSkC_Screenshot_20230917_153827_Quasseldroid.png 
2023-09-17 13:39:21 <tyler[m]123> 
ACTION sent a code block: https://matrix.org/_matrix/media/v3/download/matrix.org/LmSMVDlPfsiexVDARTNUNhsZ

2023-09-17 13:39:48 <tyler[m]123> Forza: Oh, sry. I thought he meant to use the pastebin because it was long
2023-09-17 13:40:31 <Forza> Yes. Matrix makes those links with your messages if you have multiple lines or too long messages. 
2023-09-17 13:40:35 <tyler[m]123> $ ls -l /lib64/ld-linux-x86-64.so.2... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/gLSgYDkrgWTCGstLigAGcHqu>)
2023-09-17 13:40:55 <Forza> Use https://gist.tnonline.net/ or pastebin.com for pasting code or command line output etc
2023-09-17 13:40:57 <tyler[m]123> I will try using gcompat, let's see
2023-09-17 13:42:25 <tyler[m]123> Spot on ikke , `gcompat` worked.
2023-09-17 13:42:28 <tyler[m]123> Thanks guys
2023-09-17 13:43:01 <ikke> tyler[m]123: please don't use the multi-line / code block feature, it's really anoying for us
2023-09-17 13:43:12 <tyler[m]123> Sorry
2023-09-17 13:50:44 <itwrx> i've been using "apk update" and "apk upgrade" in /etc/periodic weekly to automate upgrades for months without issue but i changed it to include "apk add --upgrade apk-tools" and added "--available" to the "apk upgrade" line.
2023-09-17 13:51:48 <itwrx> this seems to have enborkened some things... postgres14 is not in init.d anymore and is not even shown as installed but postgres-bdr is shown as installed.
2023-09-17 13:58:26 <itwrx> If i reinstall 14 will it try to overwrite any existing 14 files?
2023-09-17 14:00:29 <ikke> what does apk version postgres14 say?
2023-09-17 14:00:51 <ikke> postgresql14 I mean
2023-09-17 14:05:33 <itwrx> Installed:                                Available:
2023-09-17 14:09:34 <itwrx> prosody is also missing from init.d now but:
2023-09-17 14:09:52 <itwrx> apk version prosody
2023-09-17 14:10:52 <itwrx> shows: Installed: prosody-0.12.0-r2 Available: < 0.12.1-r2
2023-09-17 14:11:35 <itwrx> so it seems like it didn't upgrade prosody but removed the init.d file
2023-09-17 14:13:40 <itwrx> this server is running alpine 3.17.5, btw
2023-09-17 14:14:30 <itwrx> apk-tools-2.12.10-r1
2023-09-17 14:19:41 <itwrx> i did not install postgres-bdr myself/manually. i had to look up what it even was just now.
2023-09-17 14:36:57 <itwrx> reinstalling prosody init.d file from backup solved that problem.
2023-09-17 14:38:42 <itwrx> i notice prosody doesn't come with an init.d file for some reason. so maybe that's why it got removed, although i don't remember having to replace it after update before.
2023-09-17 14:45:30 <minimal> itwrx: it does have an init.d file in the prosdy-openrc package
2023-09-17 14:45:41 <minimal> prosody-openrc
2023-09-17 14:58:57 <itwrx> minimal: thanks. installed. i may have had that at one point, but it was not installed.
2023-09-17 14:59:47 <minimal> itwrx: it should be *automatically* installed when you install the prosody package (assuming, as is usual, that you have openrc installed)
2023-09-17 15:00:49 <minimal> so I don't understand how is was not installed if you had both prosody and openrc installed
2023-09-17 15:06:06 <itwrx> minimal: idk either.right now i'm trying to reinstall postgresql14 which was replaced by postgresql-bdr somehow, but there are package conflicts... maybe that's why bdr got installed? to satisfy a dep?
2023-09-17 15:29:10 <witcher> i'm unable to mount a samba share on alpine. dmesg prints "attempting to connect" but nothing ever happens. where else can i look to figure out what the problem could be?
2023-09-17 15:42:18 <Forza> witcher: does the dns name resolve, or are you connecting with ip
2023-09-17 15:43:46 <itwrx> i just ran "apk add postgresql14" and it installs 14 and 15 and sets 15 as the new default. is that expected behavior?
2023-09-17 15:43:49 <itwrx> https://pastebin.com/XERDTaHV
2023-09-17 15:49:46 <ikke> itwrx: I would not expect it
2023-09-17 15:49:52 <ikke> what's in your /etc/apk/world?
2023-09-17 15:51:26 <itwrx> postgresql
2023-09-17 15:51:26 <itwrx> postgresql-contrib
2023-09-17 15:51:26 <itwrx> postgresql14
2023-09-17 15:52:37 <itwrx> so that's why i guess, maybe postgresql was 14 in previous release so now that i upgraded that became 15. 
2023-09-17 16:01:32 <ikke> yes
2023-09-17 16:07:42 <itwrx> i haven't done any real packaging, so forgive my ignorance, but why in linux don't all package names have the version in the name?
2023-09-17 16:09:50 <minimal> normally a particular Alpine release only supports a single version of any package - postgresql is one of the few exceptions to that "rule"
2023-09-17 16:12:53 <itwrx> right, but why is this still the case for most distros? couldn't we avoid all this flatpak, snaps, appimage stuff if every upstream could just declare it's version and dep versions in a super simple manifest that all distro pkg managers could then support? 
2023-09-17 16:13:30 <itwrx> besides limited dev resources is there some technical reason?
2023-09-17 16:14:38 <quinq> That's not really the problem
2023-09-17 16:14:59 <quinq> The problem is why would you have to use an obsolete dependency as part of a regular workflow?
2023-09-17 16:15:02 <minimal> why would you want to have package *names* such as app-1.0, app-1.1, app-1.1.1, app-1.2 etc?
2023-09-17 16:18:03 <itwrx> because, just like some distros can't keep up with upstream, (thus all these not-so-universal packaging schemes) some upstreams (or custom apps) can't keep up with some distros. hosting web apps causes all kinds of fun dep issues.
2023-09-17 16:22:55 <itwrx> I'm guessing some desktop apps cause similar dep issues, thus flatpak, snaps, appimage as a symptom of an underlying, long standing issue.
2023-09-17 16:24:22 <quinq> software isn't dependent on some distro being too up to date
2023-09-17 16:24:33 <quinq> They depend on their chosen dependencies
2023-09-17 16:25:11 <quinq> If their chosen dependencies mark themselves as obsolete, then they should either adapt, or vendor it directly (ugly but that works)
2023-09-17 16:25:33 <quinq> Trying to solve problems at the wrong level just brings more problems
2023-09-17 16:26:08 <quinq> Of course, there's no solution to the web
2023-09-17 16:28:59 <itwrx> i'm not clear on the "vendor it directly" part.
2023-09-17 16:29:51 <quinq> Feel free to explain
2023-09-17 16:50:29 <itwrx> i, of course, agree that they should update their deps, but i don't love any of this direct vendoring i see with langs, web packages, and now desktop packages. 
2023-09-17 16:51:02 <itwrx> Ideally, I think i would want one totally universal packaging system (likely two-tiered) and everything to then go through the distros.
2023-09-17 16:51:14 <itwrx> I'm just wondering what technical hurdles would prevent that hypothetical approach, assuming many people were getting paid to work on it full time.
2023-09-17 16:54:40 <minimal> universal packaging system? defined by who? agree by "everyone" including those who may have various valid reasons to disagree with it? maybe if we were living in Utopia...
2023-09-17 16:55:47 <minimal> most packaging systems are defined by distros to meet their specific needs/wishes
2023-09-17 16:57:35 <minimal> no of which directly relates to the problems you've been having in your Alpine upgrade
2023-09-17 16:57:46 <minimal> s/no /none /
2023-09-17 17:00:44 <itwrx> the alpine upgrade issues just reminded me of this overall linux situation. I'm not complaining about alpine. 
2023-09-17 17:00:50 <quinq> itwrx, this is a dead end, diversification of build system is fine, as they're responsible for themselves
2023-09-17 17:01:23 <quinq> Look at all those that try this pseudo “universal packaging/distributions system” like plethora of npm / pip things
2023-09-17 17:02:55 <quinq> The actual issue here about postgresql14/15 is weird though
2023-09-17 17:06:34 <itwrx> i was thinking of a simple manifest for upstreams to use that distro package manager could then support.
2023-09-17 17:06:44 <itwrx> distros keep things how they want except i'm guessing multiple pkg versions would be required. 
2023-09-17 17:06:54 <itwrx> doing away with npm/pip and all the rest would be the goal. No need to link the XKCD. :)
2023-09-17 17:07:49 <minimal> who says the Python devs *want* to do away with pip, likewise the Node devs for npm?
2023-09-17 17:08:20 <minimal> remember some upstreams for some software do NOT want distros to package their software, they want people to get it directly from upstream
2023-09-17 17:12:08 <itwrx> I think they only did this due to the state of packaging in distros, as it relates to their needs.  Generally to get software to their users faster. If packaging for distros was more automated and flexible there would be less need for these work-arounds. 
2023-09-17 17:13:20 <minimal> all I can say is that you can continue dreaming about this Utopia but it's not going to happen...
2023-09-17 17:14:58 <itwrx> it's a feverish nightmare, not a dream. :)
2023-09-17 17:17:23 <user71> I suppose it all started with perl's cpan, it set a strong example for others
2023-09-17 17:18:41 <quinq> itwrx, that's called a README
2023-09-17 17:19:22 <quinq> itwrx, distros have no trouble packaging sane software
2023-09-17 17:20:03 <quinq> $ apk search py3
2023-09-17 17:20:22 <quinq> user71, and yet they reproduced it
2023-09-17 17:20:45 <quinq> That's one problem with those kind of projects, one doesn't know (willingly or not) history, and so is bound to repeat the same mistakes
2023-09-17 19:06:09 <witcher> Forza: i'm connecting with ip. it's working on other minimal machines but i can't reproduce it now
2023-09-17 19:42:59 <mla> q - so i have a usb drive that has a write-protect switch that i store a diskless alpine install on; when i have it set to ro mode boot fails
2023-09-17 19:43:19 <mla> what it says on syslinux is - "unable to remount cache read/write" and "failed to open apk database: read-only filesystem"
2023-09-17 19:43:43 <mla> does alpine require rw access on boot for diskless, i would have assumed booting a diskless install in ro mode should "just work", maybe theres some config somewhere i can change?
2023-09-17 20:45:51 <sloshy> i installed open-vm-tools but the guest autoresize isnt working
2023-09-17 21:27:02 <sloshy> i installed open-vm-tools but the guest autoresize isnt working
2023-09-17 21:27:38 <dwfreed> You don't need to ask the same question repeatedly
2023-09-17 21:28:03 <sloshy> what are you gonna do about it tough guy
2023-09-17 21:28:39 <dwfreed> I mean, if you'd prefer, I can ban you from the channel
2023-09-17 21:29:42 <dwfreed> Or I could go a step further and ban you from the entire network
2023-09-17 21:30:34 <sloshy> you won't do that you're too scared
2023-09-17 21:31:29 <dwfreed> the last thing I'm scared of is someone on IRC
2023-09-17 21:31:30 <quinq> I can ban you from my town because I'm the sheriff
2023-09-17 21:31:44 <sloshy> i can ban you from the entire internet because i'm elon musk
2023-09-17 21:31:53 <dwfreed> uh huh
2023-09-17 21:32:23 <quinq> Actually, only if your Ukranian on the war front using Starlink
2023-09-17 21:32:57 <sloshy> one push of a button and your router will light up like a christmas ham
2023-09-17 21:35:19 <quinq> Must be nice at christmas
2023-09-17 21:35:46 <txnintn> I installed Alpine Standard aarch64 with RPI4 UEFI firmware to create a Alpine Vanilla install, the microsd card reader and wireless nic stops working any ideas?
2023-09-18 01:04:57 <livy> Hello. I try to mount a NTFS filesystem and sometimes it is mounted as rw, other times as ro.
2023-09-18 01:05:05 <livy> What is the cause of the issue and how to solve it?
2023-09-18 01:05:33 <livy> doas mount -o remount,rw /mnt/bitlk -> mount: mounting /dev/dm-3 on /mnt/bitlk failed: Read-only file system
2023-09-18 01:07:46 <PopLamina> did it stop working after being mounted from windows?
2023-09-18 01:09:29 <livy> It is an external HDD, encrypted with BitLocker. I use it to copy data between Windows and Linux.
2023-09-18 01:09:34 <livy> It works normally under Windows.
2023-09-18 01:14:16 <PopLamina> does ntfsck say anything?
2023-09-18 01:15:02 <PopLamina> doas ntfsck /dev/dm-3 
2023-09-18 01:19:17 <PopLamina> *might be ntfsinfo
2023-09-18 04:37:15 <livy> I connected that drive into Windows, unlocked BitLocker, then performed a filesystem check.
2023-09-18 04:37:27 <livy> After that it is writable on Alpine again.
2023-09-18 04:38:17 <livy> Looks like Alpine (or the kernel) does not like an unclean disconnection of NTFS, and requires a filesystem check.
2023-09-18 04:38:26 <PopLamina> yeah the linux drivers avoid allowing writing to ntfs if any error occors
2023-09-18 04:38:40 <livy> Actually Windows scanned the drive and found no error.
2023-09-18 04:38:44 <PopLamina> like hotplugging or hibernation
2023-09-18 04:39:09 <PopLamina> yeah it's not considered an error on windows the windows end
2023-09-18 04:39:44 <livy> I thought since 5.15 we have ntfs3 driver from Paragon Softare, Linux should handle NTFS without any issues.
2023-09-18 04:40:04 <PopLamina> it's just a data safety thing idk
2023-09-18 04:41:03 <PopLamina> it's still unoffical so it defaults to read only to prevent damange if any little thing is off
2023-09-18 04:41:13 <PopLamina> I don't really know what im talking about lol
2023-09-18 04:41:47 <PopLamina> I have just dealt with ntfs drives not mounting right before
2023-09-18 12:45:06 <frag> seems some packages dont work without perl, but do not depend on perl? eg surfraw
2023-09-18 12:46:35 <ikke> Might be that a dependency which would've pulled in perl, no longer does 
2023-09-18 12:51:16 <frag> l3
2023-09-18 15:26:31 <CcxWrk> Is there a way to make apk install kernel images to a different directory (for purposes of dualboot)?
2023-09-18 15:27:22 <ikke> Only as a seperate root
2023-09-18 15:28:47 <CcxWrk> That is annoying. I guess I could do with a script that updates the real /boot but it feels slightly footgunny.
2023-09-18 15:33:04 <CcxWrk> Hm, given I probably won't need anything automatically updating the bootloader conf I guess I can just bind-mount a subdirectory to /boot
2023-09-18 15:34:27 <minimal> CcxWrk: dualboot?
2023-09-18 15:36:16 <CcxWrk> minimal: With another linux, yes. Specifically here I'm doing major upgrade by fresh install so both Alpine versions would be clobbering /boot/vmlinuz-lts for each other. But this isn't the first time I've ran into wanting this.
2023-09-18 15:36:58 <minimal> CcxWrk: with another Alpine installation you mean? so why not install them onto separate partitions?
2023-09-18 15:37:56 <CcxWrk> minimal: They are on separate LVM volumes. But /boot is obviously not, so extlinux can work with it. (this is MBR/BIOS boot)
2023-09-18 15:40:07 <minimal> ah, if you were using Grub then I believe you wouldn't need any separate /boot, you could have 2 rootfs' (including /boot) inside different LVM LVs and have Grub use the appropriate one
2023-09-18 15:45:03 <CcxWrk> I guess though this also assumes support for the rootfs (I'm not sure how is it's xfs support nowadays). I prefer to keep critical components simple. ;−)
2023-09-18 15:50:59 <minimal> it would still be kind of messy as you'd have 2 sets of grub config (and have to keep them in sync)
2023-09-18 15:59:29 <CcxWrk> Yeah. I'll go oldschool here. No automagic configs, one partition for bootloader and it's data, separate system roots. Party like LILO update just came out. :−)
2023-09-18 16:00:06 <minimal> "LILILILI" ;-)
2023-09-18 16:01:19 <CcxWrk> What is the Alpine's initrd generator called again?
2023-09-18 16:01:31 <minimal> default is mkinitfs
2023-09-18 16:01:43 <minimal> though there are also booster and dracut
2023-09-18 16:03:46 <CcxWrk> Thanks
2023-09-19 05:13:02 <qaqland[m]> When should I create an extra system user instead of root to start the service🙋
2023-09-19 05:14:09 <ikke> qaqland[m]: if the service does not require root to run
2023-09-19 05:14:17 <ikke> (or can drop privileges)
2023-09-19 05:15:18 <qaqland[m]> I see some init.d has this line: supervisor=supervise-daemon
2023-09-19 09:58:37 <frag> does ctrl-shift-pageup/down work for changeing font size for your st terminal (from repo)
2023-09-19 10:02:45 <lopid> i don't use st
2023-09-19 10:03:55 <frag> lopid: what do you use?
2023-09-19 10:06:50 <lopid> kitty in void
2023-09-19 10:07:39 <lopid> fonts usually change with the - and + keys, try those in your shortcuts
2023-09-19 10:13:10 <ikke> You probably have to change a header value and recompile 😋
2023-09-19 10:21:44 <frag> :o
2023-09-19 14:36:50 <frag> lopid: dunno if you remember this, why wont it highlight this when theres a double space : echo 'm lalal  m dfd'|grep --color '\bm \([^ ]\| [^m]\| m[^ ]\)\{1,\} m\b\|$'
2023-09-19 14:38:41 <lopid> i don't remember that
2023-09-19 14:39:13 <lopid> i don't think that grep will understand \b
2023-09-19 14:40:14 <frag> i added the \b :< but i dont think that should ruin it
2023-09-19 14:46:44 <lopid> play with it at https://regex101.com, after removing the escapes
2023-09-19 14:55:59 <frag> does actually makes it easier to read without the escaping..
2023-09-19 16:12:52 <dewdude> ERROR: linux-virt-6.1.47-r0: package mentioned in index not found (try 'apk update')
2023-09-19 16:12:56 <dewdude> apk update does not help.
2023-09-19 16:13:10 <ikke> dewdude: what mirror?
2023-09-19 16:13:16 <dewdude> ette.biz
2023-09-19 16:13:55 <ikke> What release?
2023-09-19 16:14:14 <dewdude> 3.18.2
2023-09-19 16:14:33 <dewdude> i'll just grab 3.18.3 iso
2023-09-19 16:14:39 <dewdude> i had two systems fail and i gotta rebuild
2023-09-19 16:15:00 <ikke> seems like that mirror is missing packages
2023-09-19 16:16:13 <dewdude> which is sad because it's practically right next door and just saturates my link
2023-09-19 17:14:06 <em6> Hello, I am trying to install armhf on a fat32 card but setup-bootable say "Could not find any syslinux.cfg on new Iso?". any ideas??
2023-09-19 17:33:35 <frag> 'aria2c magnet-link' to download torrents, so easy, no setup o,O
2023-09-19 17:42:50 <em6> now its working after making it fat 32 in cfdisk, noob alarm I guess. but I did not flag it as bootable in cfdisk but its still working...
2023-09-19 17:45:58 <ikke> I don't think bootable matters for these SBCs
2023-09-19 17:49:59 <em6> ikke: thanks!!
2023-09-19 17:51:13 <Habbie> pi definitely doesn't care about the bootable flag
2023-09-19 17:52:19 <minimal> yeah I think its more of a BIOS thing
2023-09-19 17:52:56 <minimal> other bootloaders may or may not use/require it
2023-09-19 17:53:50 <Habbie> well, not BIOS, but the boot loader code on sector 0
2023-09-19 17:54:01 <Habbie> a BIOS (pre-EFI) in theory doen't even know what a partition is
2023-09-19 18:11:24 <strajder> is there a reason openrc automatically mounts debugfs if /sys/kernel/debug exists and debugfs is in /proc/filesystems?
2023-09-19 18:13:01 <strajder> found this by searching the cause of "df: '/sys/kernel/debug/tracing’: Permission denied"
2023-09-20 06:30:48 <Dizzy611> Hi, I'm trying out alpine linux as a desktop distro on a low-powered old laptop. I'm trying to follow the handbook, specifically the post-installation steps here: https://docs.alpinelinux.org/user-handbook/0.1a/Working/post-install.html but I'm running into a problem, in that apk says the alpine-desktop package mentioned in the docs does not exist
2023-09-20 06:49:48 <quinq> Hi Dizzy611, I guess that's outdated info
2023-09-20 08:15:23 <qaqland[m]> Dizzy611: it is setup-desktop now
2023-09-20 10:11:44 <smollkitten> There is a strange thing on version 3.18.3 (fresh installation). During shutdown everything stops at the stage: Stopping busybox mdev.... namely when stopping sysfs.
2023-09-20 10:11:48 <smollkitten> Pressing the "i" key (enter interactive openrc mode) continues without errors and the system reboots normally. Nothing like this happened on 3.17.
2023-09-20 10:12:19 <smollkitten> Did I break something?
2023-09-20 11:04:18 <strajder> about the df reporting "permission denied" problem I mentioned yesterday, with linux-lts-6.1.51 it doesn't happen, with at least linux-lts-6.1.53 it does
2023-09-20 11:04:50 <strajder> also I tried it with linux-lts-6.1.54, still happens
2023-09-20 11:09:50 <strajder> correction, just upgraded to 6.1.54 on that other system and there it doesn't happen; busybox seems to be the same version, 1.36.1-r2, hmm
2023-09-20 11:32:53 <strajder> I seem to have found what was behind the error: hwdrivers service
2023-09-20 11:33:57 <strajder> on the second machine where df gave no error, hwdrivers was not added to any runlevel; when I added it to sysinit and started it, df started to give "permission denied" on /sys/kernel/debug/tracing
2023-09-20 11:34:25 <strajder> so the solution seems to not start hwdrivers
2023-09-20 11:34:35 <strajder> *to be
2023-09-20 11:48:31 <txnintn> All, I'm attempting to create a mirror of an existing logical volume (root partition) sda2 on a Raspberry PI 4. When I reboot after creating the mirror it can not find sba2.  What step(s)am I missing? 
2023-09-20 11:50:11 <txnintn> I'm thinking I need to copy the uuid of the logical volume and replace the sda2 uuid.
2023-09-20 11:57:04 <txnintn> When I run blkid, there is no sda2 and uuid, but there is for /dev/mapper/vg0-lv_root:  UUID=
2023-09-20 13:01:19 <Sheila> I get a bunch of missing symbols attempting to load apache2's mod_fcgid on alpine 3.18, anyone know what's up with that?
2023-09-20 13:01:35 <Sheila> apr and apr-util are both installed
2023-09-20 13:06:00 <Sheila> (also this is after `apk upgrade -al` to ensure I'm on latest available packages)
2023-09-20 16:03:25 <svmhdvn> Hi, where can I find the scripts to generate an 'alpine-uboot-<custom>-aarch64.tar.gz' tarball with my own '.config' file for the kernel? I need to generate an initramfs and modloop file along with the kernel image as provided in the release tarball
2023-09-20 17:01:01 <svmhdvn> alternatively, where can I find the '/etc/mkinitfs/features.d' configuration used for the official alpinelinux release tarballs listed on the website? The mkinitfs package from aports seems to have a much slimmer config than the official releases
2023-09-20 17:21:50 <minimal> svmhdvn: https://gitlab.alpinelinux.org/alpine/aports/-/blob/master/scripts/mkimg.arm.sh#L147
2023-09-20 17:58:33 <svmhdvn> minimal: perfect, thanks!
2023-09-20 18:36:46 <HelloNeedHelp123> Hello there, I recently installed alpine and have multiple issues: When I boot, udhcpc usually can't get a lease and I have to restart newtorking to make it work
2023-09-20 18:37:08 <ikke> ethernet or wifi?
2023-09-20 18:37:13 <HelloNeedHelp123> It's as if the wlan0 interface wasn't initialized
2023-09-20 18:37:17 <HelloNeedHelp123> Both honestly
2023-09-20 18:37:37 <HelloNeedHelp123> When I plug an ethernet cable in I get the same issue
2023-09-20 18:37:44 <HelloNeedHelp123> But with eth0
2023-09-20 18:37:55 <ikke> What network interface, do you see anything relevant in dmesg?
2023-09-20 18:38:30 <HelloNeedHelp123> I send the exact things I gte when booting
2023-09-20 18:38:34 <HelloNeedHelp123> Give me a second
2023-09-20 18:39:24 <HelloNeedHelp123> ip: ioctl 0x8913 failed: No such device
2023-09-20 18:39:45 <HelloNeedHelp123> udhcpc: ioctl 0x8933 failed: No such device
2023-09-20 18:40:09 <HelloNeedHelp123> ifup: failed to change interface wlan0 state to 'up'
2023-09-20 18:40:41 <HelloNeedHelp123> Then if I log in, wait a bit and run: rc-service networking restart
2023-09-20 18:40:59 <HelloNeedHelp123> It all works as expected
2023-09-20 18:43:45 <HelloNeedHelp123> Also another one of my questions/issues is that I installed libreoffice and it looks really old
2023-09-20 18:44:09 <HelloNeedHelp123> I dug around a bit and installed libreoffice-gnome instead, but that doesn't work at all
2023-09-20 18:46:13 <HelloNeedHelp123> Here's the error: terminate called after throwing an instance of 'com::sun::star::uno::RuntimeException
2023-09-20 18:53:14 <ikke> What version of Alpine Linux are you running?
2023-09-20 18:54:22 <ikke> Edge has libreoffice 7.5, which is not the latest, but also not that old
2023-09-20 18:59:07 <HelloNeedHelp123> I'm not such which version but I installed it just a couple days ago
2023-09-20 18:59:24 <ikke> What does cat /etc/alpine-release return?
2023-09-20 18:59:28 <HelloNeedHelp123> Also, here's exactly what libreoffice looks like: https://askubuntu.com/questions/24213/why-does-libreoffice-look-old
2023-09-20 19:00:03 <HelloNeedHelp123> 3.18.3
2023-09-20 19:10:16 <ikke> okay, so you are referring to the theme
2023-09-20 19:13:22 <strajder> rc-update | grep -E '(net|wpa)'
2023-09-20 19:17:48 <HelloNeedHelp123> strajder: networking | boot
2023-09-20 19:21:30 <strajder> you should have either wpa_supplicant or iwd service in boot runlevel
2023-09-20 19:23:36 <quinq> Speaking about busybox and networking, it seems that the last update broke the iproute2 implementation, my openvpn network would fail adding *some* given routes
2023-09-20 19:24:03 <quinq> Replacing it with the original iproute2 implementation fixed the issue
2023-09-20 19:24:25 <quinq> (was working fine before with the busybox one, and no change on the network side)
2023-09-20 19:24:33 <quinq> Just an unuseful bug report
2023-09-20 21:19:51 <iota[m]1> Is there a way to control the timeout and retry attempt count for the networking service, preferably on a per interface basis?
2023-09-20 21:56:22 <invoked> iota[m]1: retry for what?  dhcp?
2023-09-20 23:42:00 <minimal> quinq: which busybox update broken iproute2?
2023-09-21 00:25:40 <Sheila> …I think apache-mod-fcgid might actually be underlinked? because `objdump` is reporting it only needs libc, which is obviously wrong.
2023-09-21 04:50:48 <ikke> Sheila: that can happen with dynloaded objects 
2023-09-21 04:51:03 <ikke> So maybe it just needs to be rebuilt 
2023-09-21 07:22:31 <Lingo_> Folks, is anyone running Alpine on macmini? I do not see SSD within /dev/ on my macmini 2012 late :( but see related faults and errors with dmesg (relevant copypaste: https://pastebin.com/raw/jxHiHM8J)
2023-09-21 07:31:28 <iota[m]1> invoked: yes, is there a way to pass flags to udhcpc whenever you start the networking service?
2023-09-21 12:26:17 <HelloNeedHelp123> Hello there, is there a way to list explicitly installed packages on alpine? (Something like `pacman -Qe` on arch)
2023-09-21 12:26:27 <ikke> cat /etc/apk/world
2023-09-21 12:26:46 <HelloNeedHelp123> I love you
2023-09-21 12:27:41 <HelloNeedHelp123> Thank you so much
2023-09-21 12:28:38 <ikke> fun fact, that's exactly what apk add / del do, they add and remove entries from that list
2023-09-21 12:30:10 <HelloNeedHelp123> Ilove how minimal alpine is but I hate the issue with networking I have
2023-09-21 12:31:03 <HelloNeedHelp123> Yesterday someone suggested to add iwd/wpa_supplicant to the boot runlevel
2023-09-21 12:31:16 <HelloNeedHelp123> The issue: I use iwd and it already was added
2023-09-21 12:32:07 <HelloNeedHelp123> Also, I'm still having the same issue with libreoffice
2023-09-21 12:32:16 <HelloNeedHelp123> It looks like it's from 2008
2023-09-21 12:32:41 <HelloNeedHelp123> And the other version (libreoffice-gnome, libreoffice-gtk) don't work
2023-09-21 12:32:55 <HelloNeedHelp123> They throw the following error: terminate called after throwing an instance of 'com::sun::star::uno::RuntimeException'
2023-09-21 12:36:34 <ikke> HelloNeedHelp123: I have libreoffice-gnome installed and libreoffice runs without issue for me (but I'm on edge)
2023-09-21 12:37:34 <HelloNeedHelp123> Does libreoffice alone look old for you too?
2023-09-21 12:38:15 <HelloNeedHelp123> It's not because the package is old either, it's new
2023-09-21 12:38:21 <HelloNeedHelp123> Recent version
2023-09-21 12:38:35 <HelloNeedHelp123> It's just the UI that looks like it's from windows XP
2023-09-21 12:50:39 <HelloNeedHelp123> Sorry I'm back
2023-09-21 12:56:54 <Ermine> Can I make abuild grab sources from git master? I guess I need to use giturl= ?
2023-09-21 13:00:00 <ptrc> Ermine: nope, the checksums have to match
2023-09-21 13:04:06 <Ermine> ptrc: I see, thank you
2023-09-21 13:21:11 <invoked> iota[m]1: yes. you need to add "udhcpc_opts=foo" to the relevant interface in /etc/network/interfaces, indented under the 'iface' line
2023-09-21 13:21:29 <invoked> options are in the busybox manpage
2023-09-21 13:21:55 <Apachez> any of you who have tested xanmod linux kernel and what was your experience from that?
2023-09-21 13:22:09 <invoked> sorry, not "udhcpc_opts=foo" but "udhcpc_opts foo bar baz"
2023-09-21 13:22:09 <ikke> Never heard of it, let alone tested it
2023-09-21 13:34:41 <iota[m]1> invoked: thanks, it works like a charm. Feel like it should be better documented though
2023-09-21 13:42:51 <invoked> well this is kind of normal for embedded space. i mean, if you want to know more you have to dig into the source.
2023-09-21 13:43:26 <invoked> tradeoffs
2023-09-21 16:03:31 <Sheila> ikke: hmm. something more deeply fucky is going on here, it seems; rebuilding locally, including by hand, has the same results.
2023-09-21 16:05:57 <ikke> Not sure if it's the case hear, but I recall one project where the order of loading mattered
2023-09-21 16:06:02 <ikke> s/hear/here
2023-09-21 16:25:01 <Sheila> it's possible, but I'm using mod_fcgid on Ubuntu with the same configuration; it works there and ldd doesn't have missing symbols there (objdump there doesn't report any dep on apr/apr-util, either). so idk.
2023-09-21 16:30:16 <ikke> possibly different build options?
2023-09-21 16:30:18 <ikke> but dunno
2023-09-21 16:30:20 <ikke> just guessing here
2023-09-21 16:32:08 <Sheila> I thought about that so I grabbed the rules file from ubuntu, and they don't do anything differently over there.
2023-09-21 16:41:27 <Sheila> it may be an apache configuration thing, but that's getting in to 'above my paygrade' territory, and I'd prefer to not switch to another httpd because then I'll have to figure out another solution to the problem that mod_md solves for me (which is managing LE certs).
2023-09-21 16:43:00 <ikke> more then 10 years ago since a last touched apache
2023-09-21 16:44:40 <XFaCiEer> what is a last?
2023-09-21 16:45:02 <Habbie> s/ a / i /, i bet
2023-09-21 16:45:30 <ikke> ahuh
2023-09-21 16:45:56 <XFaCiEer> Habbie: You better you better you bet.
2023-09-21 16:46:02 <Habbie> :)
2023-09-21 19:18:30 <Ermine> minimal: is cloud config schema in cloud-init something that provider gives to cloud-init
2023-09-21 19:30:01 <minimal> Ermine: when yuo create a Cloud instance you can optionally provide "user data" for that instance via the Cloud Provider's console/API. That user data may be 1 or more shell scripts or a cloud-config YAML document
2023-09-21 19:30:25 <minimal> so it is provided by you to the provider who then provides it to the instance via their metadata server
2023-09-21 19:33:12 <minimal> Ermine: more info here: https://cloudinit.readthedocs.io/en/latest/explanation/format.html
2023-09-21 19:39:45 <Ermine> minimal: the thing is that I'm getting this warning: https://paste.sr.ht/~ermine/2d363ebb901099975a7e76c976e27e739a29492b
2023-09-21 19:40:53 <minimal> Ermine: what's the actual user-data you're providing?
2023-09-21 19:42:01 <Ermine> my provider shows this https://paste.sr.ht/~ermine/6a08e3e8101fc3f0b2e1f069c3e694c923885f5e
2023-09-21 19:42:13 <Ermine> I didn't type this though
2023-09-21 19:43:21 <Ermine> I have 2 guesses: 1) cloud-init didn't like sudo config, or 2) Provider messed up and I should contact them
2023-09-21 19:47:35 <minimal> hard to know without seeing cloud-init.log file. The ssh key is specified in a similar fashion to the examples here? https://cloudinit.readthedocs.io/en/latest/reference/examples.html#yaml-examples
2023-09-21 19:49:31 <Ermine> Yes
2023-09-21 19:51:06 <minimal> try running "cloud-init schema --system"
2023-09-21 19:52:15 <minimal> have you checked /var/log/cloud-init.log to see if anything in there shows what's going on?
2023-09-21 19:55:06 <Ermine> in cloud-init.log, there's one warning: schema.py[WARNING]: Invalid cloud-config provided: Please run 'sudo cloud-init schema --system' to see the schema errors.
2023-09-21 19:55:27 <Ermine> and output of this command is in the first link I've sent to you
2023-09-21 20:00:57 <minimal> Ermine: ah, for ssh-authorized-keys replace all the "-" with "_" and try again
2023-09-21 20:01:05 <minimal> for ssh_authorized_keys
2023-09-21 20:01:09 <minimal> s/for/so/
2023-09-21 20:02:12 <minimal> that parsed fine for me (tested using "cloud-init schema -c test.yml")
2023-09-21 20:07:15 <Ermine> minimal: that worked, thank you!
2023-09-21 20:08:31 <minimal> Ermine: I don't see ssh-authorized-keys being defined as a valid config, I've checked as far back as cloud-init 18.1
2023-09-21 20:10:30 <Ermine> I guess this is something my provider generated, so...
2023-09-21 20:11:25 <minimal> Ermine: how would they know what to generate? who told them to create a "peter" user? who told them which public SSH to use for it?
2023-09-21 20:12:04 <Ermine> I did while creating VM
2023-09-21 20:12:41 <Ermine> they allow you to specify which users are to be created and their SSH keys
2023-09-21 20:13:41 <minimal> creating via some web interface rather than via API or CLI?
2023-09-21 20:13:46 <Ermine> Yes
2023-09-21 20:14:45 <minimal> ok, sound a little unusual as I'd expect their console to have a field where you'd paste in a complete user-data entry
2023-09-21 20:15:45 <minimal> but rather they seem to be creating a user-data YAML file for you using provided users data.....in which case then they handle a mistake in their templating
2023-09-21 20:16:38 <minimal> recent releases (at least for the past couple of years) of cloud-init do strict validation of cloud-config files whereas older versions did not
2023-09-21 20:19:54 <Ermine> Actually this warning was shown all this time, but I didn't notice it...
2023-09-21 20:23:19 <minimal> Ermine: I've never heard of a cloud provider creating a cloud-config doc on your behalf. What if you want to use some other cloud-config functionality like runcmd: or ntp: etc ?
2023-09-21 20:25:38 <Ermine> This web interface has 'metadata' section which allows you to specify key-value pairs. Or, I guess, cli may have more functionality
2023-09-21 20:29:36 <minimal> Ermine: you should raise a bug with them regarding "ssh-authorized-keys"
2023-09-21 20:31:56 <Ermine> Yeah
2023-09-22 06:40:47 <sewn_> hi, is there a way to list orphaned (packages with no recursive dependencies, or no packages depend on them) in apk?
2023-09-22 06:41:14 <sewn_> im currently getting around this by using a apk info pipe to apk info -r but its incredibly slow. i tried to use xargs -P0 but apk got a segfault.
2023-09-22 06:44:14 <sewn_> that script also still has inaccuracies as other packages still depend on it
2023-09-22 06:54:46 <hd9hd9e2> hello, does anyone know their way around with enterprise wifi? I manually added an entry to wpa_supplicant.conf and starting it works, but I have to manually rerun udhcpc because it doesn't get the default gateway alone. any ideas?
2023-09-22 07:17:31 <hd9hd9e2> hmm the issue seems to have fixed itself after a second reboot
2023-09-22 12:56:58 <dok> hi, i am new to alpine-linux, is there a way to install -doc packages by default ?
2023-09-22 12:57:04 <ikke> apk add docs
2023-09-22 12:58:22 <dok> oh thanks
2023-09-22 13:00:26 <raspbeguy> Hello, I upgraded my laptop (edge) and Gnome doesn't respect my dark theme setting anymore. It worked before, and I'm using default theme (adwaita).
2023-09-22 13:06:28 <ikke> raspbeguy: was gtk updated?
2023-09-22 13:07:08 <raspbeguy> sadly I don't have the list of what was upgraded
2023-09-22 13:07:09 <ikke> gsettings set org.gnome.desktop.interface color-scheme prefer-dark
2023-09-22 13:07:13 <ikke> that's for gtk4
2023-09-22 13:07:19 <raspbeguy> yep, did that already
2023-09-22 13:07:44 <raspbeguy> doesn't work for any GTK version