| 24 Apr 2023 |
 @elvishjerricco:matrix.org | wait wut | 15:34:41 |
| @elvishjerricco:matrix.org | oh | 15:34:49 |
| @elvishjerricco:matrix.org | are you using zfs? | 15:34:56 |
 oddlama | yep | 15:35:10 |
| @elvishjerricco:matrix.org | https://github.com/NixOS/nixpkgs/pull/227208 | 15:35:18 |
| @elvishjerricco:matrix.org | zfs-import-${pool}.service is in local-fs.target, which is incorrect in initrd | 15:35:46 |
| @elvishjerricco:matrix.org | but that incorrectness does result in it being pulled in by rescue.target | 15:36:04 |
| oddlama | I see. It's probably not correct but in my case it shouldn't be a problem either | 15:36:47 |
| @elvishjerricco:matrix.org | that doesn't explain why sshd is starting though | 15:36:50 |
| oddlama | Wait I have to add some utils to the initrd before continuing, I can't even ping :P | 15:37:23 |
| @elvishjerricco:matrix.org | Can you systemctl list-dependencies --reverse sshd.service? | 15:39:05 |
| oddlama | Will do, but I have to find the proper option to add packages first | 15:40:29 |
| @elvishjerricco:matrix.org | boot.initrd.systemd.initrdBin is kinda like enviroment.systemPackages in that it just puts all their /bins on PATH | 15:41:04 |
| @elvishjerricco:matrix.org | boot.initrd.systemd.extraBin.$BINNAME = pathToBinary will add just the one binary | 15:41:28 |
| @elvishjerricco:matrix.org | (not great names, I know... Kinda want to change them) | 15:41:43 |
| oddlama | Thanks, that'll do | 15:42:19 |
| oddlama | In reply to @elvishjerricco:matrix.org
Can you systemctl list-dependencies --reverse sshd.service? Seems like I was talking bullshit. The sshd service is not pulled in except by initrd.target, so I must have started it accidentally when debugging other stuff. My bad. | 15:48:39 |
 @aktaboot:tchncs.de | the systemd initrd seem way bigger though ? I can't fit 8 of these :c | 21:13:37 |
| @elvishjerricco:matrix.org | aktaboot: How big, and how big was the traditional one? | 21:14:05 |
| @elvishjerricco:matrix.org | initrd size is something we would like to be careful about, but sometimes we make sacrifices | 21:14:23 |
| @aktaboot:tchncs.de | 25Mi | 21:14:58 |
| @elvishjerricco:matrix.org | IIRC networking does blow up the size quite a bit... | 21:15:22 |
| @aktaboot:tchncs.de | * 25.4 MiB on new, 5.1 MiB on old | 21:15:29 |
| @elvishjerricco:matrix.org | not sure with the regular initrd though | 21:15:35 |
| @elvishjerricco:matrix.org | 5.1MiB O_o I've never seen it that small, scripted or systemd | 21:15:52 |
| @aktaboot:tchncs.de | I'm not sure if that's just the one that wasn't written fully because there was no space | 21:16:56 |
| @aktaboot:tchncs.de | and I can't really know because they all have the same date ? | 21:18:18 |
| @aktaboot:tchncs.de | hmm, on another laptop the older ones are 40MiB | 21:21:10 |
| @aktaboot:tchncs.de | or not these are not the initrds (?) | 21:22:08 |
| @elvishjerricco:matrix.org | If you're using plymouth, that makes up the majority of the initrd's space regardless of systemd vs scripted | 21:23:38 |
