Bug#788050: systemd-fsck : Check disks at each reboot

[Sander Smeenk]

Hi,

Discovered this bug while researching trouble i'm having with systemd
boot and (forced) fscks. All that has been said in this bug, i'm seeing
the same. Since this bug was last updated december 20th 2015, i'm
wondering what the state of this is atm.

I'm running Ubuntu 16.04 though, systemd 229-4ubuntu6.

I have a huge filesystem (20T, 12T used) that experiences errors from
time to time. I've set the max-mount-count for this fs to 1, it has pass
flag '2' in fstab, so when i reboot it always checks the FS. I've set
'fsck.repair=yes' on the kernel cmdline so it should fix errors found.

All that was said in this bug-thread applies to me. During root-fs check
i see fsck output, during non-root-fs check i see an obfuscation message
showing a percentage. After a while it switches to 'A startup task for
... is running', a few moments later it stops and mounts the filesystem
with errors.

TimeoutSec is already set to 0 on relevant services on my system:

| # grep Timeout *fsck*
| systemd-fsck-root.service:TimeoutSec=0
| systemd-fsck at .service:TimeoutSec=0

I can also reproduce this issue at will.

Any progress on this?

-Sndr.
-- 
| Hey.. I'm done talkin'. Now check out my pretty!
| 4096R/20CC6CD2 - 6D40 1A20 B9AA 87D4 84C7  FBD6 F3A9 9442 20CC 6CD2