#hi all i did something stupid with my
blissful pollen
I tried single on the kernel cmdline which drops me into a root shell, but / is a squashfs so i'm unsure how to edit network settings.
also noticed that the service that fails just before it reboots is systemd-time-wait-sync, so i tried booting with systemd.mask=systemd-time-wait-sync.service which skips that, but it fails on something else (it reboots too quickly for me to see what)
what's the best way to resolve this? i can't easily get a local shell because something in hassos startup scripts reboots it automatically before the getty launches for me to log in
ok i figured based on https://github.com/home-assistant/operating-system/blob/e6147184c094711e5a783de996960ac27419b9a2/buildroot-external/rootfs-overlay/etc/systemd/system/docker.service.d/failure.conf#L2 that docker is failing to start and that's what causes the reboot. booting with systemd.mask=docker.service has got me eventually to a console
bleak creek
Was about to link you the git issue, but I think you found it already
bleak creek
That's been the reason I fear updating/rebooting my instance
blissful pollen
the debugging experience was bad, but thankfully i know systemd quite well so followed a hunch
and now that a few ways to solve it are documented in that issue, no need to fear!
also i think a clean shutdown should be OK in general
someone was fixing my A/C and turned off the wrong breaker
i'm so used to no corruption after unclean shutdowns that my first thought was recent config changes that I've made (DNS) rather than a corrupted file
bleak creek
Yeah, once you know what's the error it's not that bad, tho I can tell you that it happened to me 4 times already (before figuring it out). It happened 3 times on clean shutdown and once on os upgrade