[Pkg-systemd-maintainers] Bug#733232: Bug#733232: systemd: Systemd fails to boot after fscking filesystem if plymouth not installed

[Michael Stapelberg]

Hi Philip,

Philip Armstrong <phil at kantaka.co.uk> writes:
> If systemd decides to fsck one or more of my filesystems, then the
> boot consistently fails, dropping me to a root password prompt. If I
> login & run journalctl -xb than I find the following error in the
> journal:
>
> Dec 27 12:39:53 xanthus systemd[1]: Started LSB: option to manually
> manipulate the IPsec SA/SP database.
> Dec 27 12:39:53 xanthus systemd[1]: Startup finished in 4.659s
> (kernel) + 1min 30.662s (userspace) = 1min 35.321s.
> Dec 27 12:39:53 xanthus systemd[685]: Failed at step EXEC spawning
> /bin/plymouth: No such file or directory
>
> (I don't have plymouth installed & there is no /bin/plymouth)
>
> The system always boots successfully if no fsck is required.
I think you are misinterpreting the situation.

In the systemd source, there are only two occurrences of /bin/plymouth:

$ grep -r '/bin/plymouth' .
./units/rescue.service.m4.in:ExecStartPre=-/bin/plymouth quit
./units/emergency.service.in:ExecStartPre=-/bin/plymouth quit

Both are prefixed with a minus, meaning it’s not a problem if they
cannot be started.

Furthermore, both occur in rescue.service and emergency.service,
respectively, so I think something else is failing before that. Can you
reproduce this with the kernel parameter systemd.log_level=debug and
then attach the entire journalctl -xb output please?

-- 
Best regards,
Michael