On 4 September 2013 10:39, Michael Biebl <bi...@debian.org> wrote: > Am 04.09.2013 01:45, schrieb Brian May: > > aquitard# umount > /var/lib/schroot/union/underlay/squeeze-f8ea98e7-1bac-43d1-b774-94c9c42fddc7 > > umount: > /var/lib/schroot/union/underlay/squeeze-f8ea98e7-1bac-43d1-b774-94c9c42fddc7: > device is busy. > > (In some cases useful info about processes that use > > the device is found by lsof(8) or fuser(1)) > > > > aquitard# fuser -vm > /var/lib/schroot/union/underlay/squeeze-f8ea98e7-1bac-43d1-b774-94c9c42fddc7 > > USER PID ACCESS COMMAND > > > /var/lib/schroot/union/underlay/squeeze-f8ea98e7-1bac-43d1-b774-94c9c42fddc7: > > root kernel mount > /var/lib/schroot/union/underlay/squeeze-f8ea98e7-1bac-43d1-b774-94c9c42fddc7 > > So I assume after the reboot it is the schroot init script which mounts > that file system or is that mount point in /etc/fstab? >
Yes. That is my understanding. > systemd only mounts a few internal API file systems and what it finds in > /etc/fstab. > pid says kernel, which is somewhat odd. Does systemd do anything to monitor mounted filesystems, e.g. to run actions if something changes? But if I read you correctly, the above means, that there is a running > mount process which was spawned to mount that lvm volume and that mount > process did not exit? > I don't see any evidence of any mount command still running. "ps auwx | grep mount" returns nothing. > And this behaviour you only get with systemd? > Yes. That is correct. Maybe I should try again now, however that would mean another reboot, and now is not a good time.
