💾 Archived View for gemini.ctrl-c.club › ~phoebos › logs › kisslinux-2023-06-05.txt captured on 2024-05-26 at 16:00:33.
⬅️ Previous capture (2023-06-14)
-=-=-=-=-=-=-
[2023-06-05T04:33:17Z] <sewn> https://github.com/kisslinux/website/issues/52
[2023-06-05T04:33:22Z] <sewn> can we redo this
[2023-06-05T04:36:50Z] <illiliti> dilyn: hi. yes
[2023-06-05T06:22:58Z] <testuser[m]> Hi
[2023-06-05T06:23:07Z] <testuser[m]> sewn: redo what
[2023-06-05T06:23:56Z] <sewn> the context of the issue I linked, which is hosting screenshots of kiss linux user's desktops and such
[2023-06-05T13:52:46Z] <sad_plan> hi
[2023-06-05T14:13:26Z] <dilyn> illiliti: I finally have the capacity to actually rigorously use ZFS but I'm hitting an issue. What's your /etc/tinyramfs/config and kernel command line? mine:
[2023-06-05T14:13:27Z] <dilyn> compress=zstd
[2023-06-05T14:13:27Z] <dilyn> root_type=zfs
[2023-06-05T14:13:28Z] <dilyn> zfs_root=UUID=15497905189281842687
[2023-06-05T14:13:28Z] <dilyn> hooks=systemd-udev,zfs
[2023-06-05T14:13:29Z] <dilyn> commandline: root=rpool/ubuntu rw initrd=\efi\uwu\initrd.zst
[2023-06-05T14:13:29Z] <dilyn> pool structure is like:
[2023-06-05T14:13:30Z] <dilyn> rpool/ubuntu{/some extra datasets} <- location of /, mountpoint=legacy
[2023-06-05T14:13:44Z] <dilyn> I believe the error at boot is "cannot import pool ' ' "
[2023-06-05T15:03:02Z] <sewn> why do you use an initrd
[2023-06-05T15:03:23Z] <dilyn> because I didn't want to swap bootloaders yet and as a result it's the only way to boot from zfs on root
[2023-06-05T15:04:30Z] <sewn> oh yeah zfs is out of tree and is needed because root partition
[2023-06-05T15:04:31Z] <sewn> oops
[2023-06-05T15:07:28Z] <dilyn> yeah; I actually boot my ubuntu install without an initrd but in this case I can't (trivially) avoid it
[2023-06-05T16:05:05Z] <illiliti> my config: https://illiliti.pastes.sh/1685981014272276512
[2023-06-05T16:06:19Z] <illiliti> cmdline: vmlinuz initrd=\initramfs quiet i915.fastboot=1 lsm=landlock
[2023-06-05T16:06:40Z] <dilyn> gotcha gotcha! thank you
[2023-06-05T16:06:41Z] <illiliti> i boot from uefi directly
[2023-06-05T16:06:48Z] <dilyn> also I believe that fastboot option is default on 5.12+
[2023-06-05T16:07:07Z] <illiliti> maybe i don't know
[2023-06-05T16:07:28Z] <illiliti> they break it from time to time
[2023-06-05T16:07:43Z] <dilyn> I'll test later today with an updated config, seems like it should work then. thank you so much <3
[2023-06-05T16:07:57Z] <dilyn> yeah I saw a lot of bug reports regarding fastboot breaking for i915 when debugging something haha
[2023-06-05T16:09:12Z] <illiliti> how did you build zfs module?
[2023-06-05T16:09:18Z] <illiliti> as builtin?
[2023-06-05T16:09:22Z] <illiliti> or external
[2023-06-05T16:09:22Z] <dilyn> yessir
[2023-06-05T16:09:30Z] <illiliti> builtin?
[2023-06-05T16:09:33Z] <dilyn> yeah builtin
[2023-06-05T16:09:40Z] <illiliti> me too
[2023-06-05T16:09:43Z] <dilyn> I think I'm JUST screwing up the config tbh
[2023-06-05T16:09:57Z] <dilyn> builtin is the way to go! keep it simple
[2023-06-05T16:10:13Z] <illiliti> maybe because you use systemd-udev which i don't test regularly
[2023-06-05T16:10:55Z] <dilyn> it's possible. that was my next point to investigate
[2023-06-05T16:11:28Z] <illiliti> cannot import pool sounds sus
[2023-06-05T16:11:58Z] <dilyn> well I think it can't import the pool because it doesn't know the name of the pool
[2023-06-05T16:12:08Z] <dilyn> because I'm passing it to zfs.init incorrectly
[2023-06-05T16:12:50Z] <illiliti> ah wait
[2023-06-05T16:12:57Z] <illiliti> you use uuid right
[2023-06-05T16:13:11Z] <illiliti> that may be a problem
[2023-06-05T16:13:58Z] <illiliti> like, your blkid may not support zfs and discovery by uuid will not work in that case
[2023-06-05T16:15:52Z] <dilyn> it should, it's ubuntu :clown-face:
[2023-06-05T16:16:10Z] <illiliti> double check
[2023-06-05T16:16:27Z] <dilyn> I have run it myself (albeit in userspace) and it works as intended, but I'll play around with it
[2023-06-05T16:20:43Z] <illiliti> hmm are you sure rpool/ubuntu is correct root?
[2023-06-05T16:20:52Z] <dilyn> yessir
[2023-06-05T16:21:54Z] <illiliti> my zfs list: https://illiliti.pastes.sh/1685982091095676750
[2023-06-05T16:22:54Z] <illiliti> mount: https://illiliti.pastes.sh/1685982156177005752
[2023-06-05T16:23:05Z] <illiliti> if that helps
[2023-06-05T16:24:20Z] <dilyn> useful, thank you; I assume based on the way the zfs hooks work you *have* to mark pools that need to be mounted as legacy mount, otherwise they'll all mount at /mnt/root instead of /
[2023-06-05T16:25:40Z] <illiliti> yes
[2023-06-05T16:26:00Z] <illiliti> iirc there is a workaround to this issue but i didn't implement it
[2023-06-05T16:28:24Z] <illiliti> which kernel and zfs version you use?
[2023-06-05T16:45:09Z] <dilyn> 6.2.12+ (ubuntu 23.04 kernel), whatever latest zfs is
[2023-06-05T16:45:19Z] <dilyn> have to be bleeding edge on ubuntu because I got an arc GPU xD
[2023-06-05T16:45:40Z] <illiliti> intel arc?
[2023-06-05T16:46:08Z] <dilyn> yessir
[2023-06-05T16:46:10Z] <dilyn> a750
[2023-06-05T16:46:13Z] <illiliti> cool
[2023-06-05T16:46:44Z] <dilyn> got it for encoding/decoding once I build my NAS; i'll switch my main machine to like a 7900xtx or something when that happens. but that's a long-term strategy
[2023-06-05T16:52:42Z] <illiliti> well your only way to understand what's wrong is to boot into initramfs and type commands by hand until you get an error
[2023-06-05T16:53:30Z] <illiliti> there are some breakpoints you can use to drop into shell before specified stage is ran
[2023-06-05T16:54:04Z] <illiliti> grep init.sh for rdbreak
[2023-06-05T16:57:20Z] <dilyn> for sure, that was the last thing I was going to do (arguably maybe it should be the first thing I did but oh well)
[2023-06-05T16:58:30Z] <illiliti> it's worth to check if /dev/disk/by-uuid/<your uuid> gets created in initramfs at boot
[2023-06-05T16:58:43Z] <illiliti> if not, then it's device manager issue
[2023-06-05T16:59:00Z] <illiliti> in your case it's systemd obviously
[2023-06-05T17:03:09Z] <illiliti> if uuid does not exist in /dev/disk/by-uuid, you can try to specify path to devnode in zfs_root instead as a quick workaround
[2023-06-05T17:05:16Z] <dilyn> definitely
[2023-06-05T17:10:22Z] <illiliti> can you send me output of zfs list if possible?
[2023-06-05T17:15:17Z] <dilyn> you want the whole thing?? http://ix.io/4xxO
[2023-06-05T17:16:02Z] <dilyn> things like rpool/ubuntu/{usr,var} are set to can't be mounted btw. the mountpoint is a red herring
[2023-06-05T17:16:19Z] <dilyn> need to set rpool/ubuntu/usr/local etc to legacy tho
[2023-06-05T17:17:12Z] <dilyn> (though I might end up using https://openzfs.github.io/openzfs-docs/man/8/zfs-mount-generator.8.html to get around this, haven't decided yet)
[2023-06-05T17:22:28Z] <illiliti> fuck
[2023-06-05T17:22:39Z] <illiliti> i know what's wrong
[2023-06-05T17:23:02Z] <illiliti> you specified root= on cmdline
[2023-06-05T17:23:08Z] <illiliti> that's unsupported
[2023-06-05T17:23:22Z] <illiliti> specify it in config
[2023-06-05T17:23:36Z] <dilyn> yeah that's the change I made and why I asked for yours lmfao
[2023-06-05T17:23:43Z] <dilyn> but *technically* it SHOULD be supported:P
[2023-06-05T17:23:50Z] <illiliti> yeah
[2023-06-05T17:23:53Z] <dilyn> I'm about to test the changes i'll lyk
[2023-06-05T17:24:20Z] <illiliti> the problem is that it requires major changes to tinyramfs
[2023-06-05T17:24:33Z] <illiliti> like event-based system where hooks= order won't matter
[2023-06-05T17:24:40Z] <dilyn> aha aha
[2023-06-05T17:24:49Z] <illiliti> that's something i want to implement
[2023-06-05T17:24:51Z] <dilyn> I never used tinyramfs enough to notice when I was messing around with it last time
[2023-06-05T17:25:02Z] <illiliti> but have no time for this yet
[2023-06-05T17:25:22Z] <illiliti> i don't use tinyramfs often too
[2023-06-05T17:25:40Z] <illiliti> i just generated it once and forgot
[2023-06-05T17:28:14Z] <illiliti> btw i don't know how zfs mounts pools with nested pools
[2023-06-05T17:28:27Z] <illiliti> i never tested such configuration
[2023-06-05T17:28:29Z] <dilyn> wdym?
[2023-06-05T17:28:44Z] <dilyn> it's just a single pool
[2023-06-05T17:29:17Z] <dilyn> how mounting datasets is handled is a tricky bitch tho
[2023-06-05T17:29:41Z] <illiliti> i mean if you mount rpool/ubuntu, does it mean that rpool/ubuntu/usr will get mounted too?
[2023-06-05T17:30:10Z] <illiliti> i don't know how mount works in this case
[2023-06-05T17:30:24Z] <illiliti> mine configuration is simple as you see
[2023-06-05T17:30:24Z] <dilyn> rpool/ubuntu/usr won't get mounted because it's canmount=off. it only exists to create rpool/ubuntu/usr/local
[2023-06-05T17:30:53Z] <dilyn> you can't have rpool/ubuntu/usr mount to, say, /usr, because it won't exist for initrd to handoff to (because if you set it to automount it will get mounted to /mnt/root instead of /)
[2023-06-05T17:31:15Z] <dilyn> you'd have to extend tinyramfs' zfs hook to mount datasets before switch_root
[2023-06-05T17:33:46Z] <dilyn> I think the way would just be a zfs.init.late that checks if a var is set in config and mounts datasets - trivial with a zfs mount -a BUT if you're doing a dual boot off a single pool and have datasets setup like I do you'd need to set a different passphrase on the root datasets so that you don't get a collision -- because rpool/kiss was made after
[2023-06-05T17:33:47Z] <dilyn> rpool/ubuntu, rpool/kiss would get mounted over rpool/ubuntu
[2023-06-05T17:34:21Z] <dilyn> if you provide separate keys you just fail on the key for the dataset you don't care about and I think zfs mount -a only mounts unlocked rpools
[2023-06-05T17:34:32Z] <illiliti> i think that's something host system should do, e.g via fstab
[2023-06-05T17:34:39Z] <dilyn> I think I agree
[2023-06-05T17:34:45Z] <illiliti> initramfs task is to mount root and switch root to it
[2023-06-05T17:34:50Z] <dilyn> 100%
[2023-06-05T17:35:36Z] <dilyn> you can use the zfs mount generator to create a bunch of systemd services to properly map out the mount order and do the mounting during init after you've switch rooted but... why would I let systemd do that when I can just use fstab do the one thing it was designed for lmfao
[2023-06-05T17:36:55Z] <illiliti> yes
[2023-06-05T18:01:04Z] <illiliti> if you manage to get it to work, lmk. i'm going to sleep
[2023-06-05T18:27:48Z] <sewn> what's it with zig devs trying to blame kiss linux for zig builds being segfault
[2023-06-05T18:50:08Z] <sad_plan> maybe theyre drunk?
[2023-06-05T19:09:41Z] <dilyn> illiliti: figured out *one* thing: was using the wrong ID for the pool
[2023-06-05T19:10:04Z] <dilyn> zpool wants the ID of the pool itself, not the disk. I guess -- hard to debug. My keyboard doesn't work in initrd
[2023-06-05T19:13:14Z] <dilyn> the keyboard is seen by the kernel, the module is loaded. so why it doesn't function, I'm not certain of
[2023-06-05T21:10:36Z] <dilyn> setting hostonly in config results in success. Will investigate why soon enough
[2023-06-05T21:51:54Z] <dilyn> illiliti: http://ix.io/4xyE
[2023-06-05T21:51:55Z] <dilyn> Because I'm missing HID