!PSmBFWNKoXmlQBzUQf:helsinki-systems.de

Stage 1 systemd

82 Members
systemd in NixOs's stage 1, replacing the current bash tooling https://github.com/NixOS/nixpkgs/projects/5126 Servers

Load older messages


SenderMessageTime
7 Dec 2022
@hexa:lossy.network@hexa:lossy.networkthat gets mounted via impermanence 😲21:08:56
@hexa:lossy.network@hexa:lossy.networkwell bind-mounted21:09:20
@hexa:lossy.network@hexa:lossy.network
In reply to @hexa:lossy.network
... does not look like a NixOS installation - not activating
this begins with /nix/store/....-sytem
21:10:01
@hexa:lossy.network@hexa:lossy.networkso I'm not sure how this would be about /etc/nixos21:10:12
@elvishjerricco:matrix.org@elvishjerricco:matrix.org hexa: Oh wait, I misremembered what generates that error. It's this: https://github.com/NixOS/nixpkgs/blob/master/nixos/modules/system/boot/systemd/initrd.nix#L487 21:11:42
@elvishjerricco:matrix.org@elvishjerricco:matrix.org I guess that makes sense if you said /sysroot is empty? 21:12:22
@hexa:lossy.network@hexa:lossy.networki guess21:12:39
@elvishjerricco:matrix.org@elvishjerricco:matrix.orgI can't imagine why sysroot is empty though21:12:50
@elvishjerricco:matrix.org@elvishjerricco:matrix.org Oh, I wonder if zfs rollback is erasing the mountpoints 21:13:36
@elvishjerricco:matrix.org@elvishjerricco:matrix.orgSo maybe I was wrong when I suggested it doesn't matter much when the rollback happens :P21:13:53
@elvishjerricco:matrix.org@elvishjerricco:matrix.orgIf the snapshot being rolled back to doesn't have those mountpoints, I'm not sure what happens...21:14:16
@hexa:lossy.network@hexa:lossy.networkmy point exactly 😄21:14:40
@elvishjerricco:matrix.org@elvishjerricco:matrix.org So I guess try adding before = ["sysroot.mount"]; 21:14:45
@hexa:lossy.network@hexa:lossy.networkthough I would expect them to get created during activation21:14:48
@elvishjerricco:matrix.org@elvishjerricco:matrix.orgThat wouldn't help. If a file systems mountpoints simply vanish, then making new directories at the same path wouldn't be reviving the old mount points21:15:19
@elvishjerricco:matrix.org@elvishjerricco:matrix.org * That wouldn't help. If a file systems mountpoints simply vanish, then making new directories at the same path wouldn't be reviving the old mountpoints21:15:21
@elvishjerricco:matrix.org@elvishjerricco:matrix.orgI'm just... surprised it's possible to make mountpoints vanish21:15:31
@hexa:lossy.network@hexa:lossy.networkwantedby initrd.target, before sysroot.mounts => odering cycle found, skipping Local File Systems21:16:23
@elvishjerricco:matrix.org@elvishjerricco:matrix.orgoh21:16:35
@elvishjerricco:matrix.org@elvishjerricco:matrix.org If you're going to come before local-fs.target, you need DefaultDependencies=no, which also means you'll need to order after the import service 21:17:10
@hexa:lossy.network@hexa:lossy.networkwhere do I set DefaultDeps=no?21:17:55
@elvishjerricco:matrix.org@elvishjerricco:matrix.orgunitConfig21:18:12
@hexa:lossy.network@hexa:lossy.networkok21:18:15
@elvishjerricco:matrix.org@elvishjerricco:matrix.orgthe ordering cycle is because the default dependencies of a unit include After=local-fs.target, and sysroot.mount has Before=local-fs.target21:18:41
@elvishjerricco:matrix.org@elvishjerricco:matrix.org * the ordering cycle is because the default dependencies of a service include After=local-fs.target, and sysroot.mount has Before=local-fs.target21:18:49
@elvishjerricco:matrix.org@elvishjerricco:matrix.org(more reasons we need to use initrd-fs.target instead of local-fs.target...)21:19:49
@hexa:lossy.network@hexa:lossy.networkcool 😄 now it starts quite early and fails21:19:50
@elvishjerricco:matrix.org@elvishjerricco:matrix.orgoof21:19:57
@hexa:lossy.network@hexa:lossy.networkeven before luks21:20:00
@hexa:lossy.network@hexa:lossy.networkso after zfs-import.target21:20:10

Show newer messages


Back to Room ListRoom Version: 6