[systemd-devel] Infinite loop at startup on var fsck failure

[Vito Caputo]

Hello list,

Today I experienced an unclean shutdown due to battery dying unexpectedly,
and it left my /var in a state requiring a manual fsck to repair errors.

The normal startup process failed and dropped me to a rescue shell after
asking for my root password.  But I was unable to immediately run fsck
manually, because systemd was endlessly trying to fsck /var.

Stopping, disabling, masking, none of those obvious options to prevent
'systemd-fsck at dev-mapper-ssd\x2var.service' from starting again in
this loop worked, and I don't recall seeing any guidance in the journal on
what was the appropriate course of action.

Eventually I resorted to `systemctl emergency` which seemed to get things
quieted down enough for me to run the fsck manually.

All's well that ends well, but what an *awful* user experience.  Is this
really how things are supposed to play out when a fsck on something like
/var fails?  I was very much left in the dark at a root shell with systemd
pointlessly spinning its wheels hopelessly running the same fsck
repeatedly.

It's possible this is already better in a newer systemd release, but I just
wanted to document this experience in case it's an area that still needs
improvement.

This is on an old release (v232) in Debian 9.11 amd64.

Regards,
Vito Caputo