Re: [gentoo-user] Mysteriously dismounting partition

[Peter Humphrey <peter@prh.myzen.co.uk>]

On Tuesday 27 October 2015 18:18:18 J. Roeleveld wrote:
> On 27 October 2015 17:36:00 CET, Peter Humphrey <peter@prh.myzen.co.uk> 
wrote:
> >On Tuesday 27 October 2015 15:16:26 J. Roeleveld wrote:
> >> If a disk is umounted/removed, something needs to be logged
> >>somewhere.
> >
> >> Might even be a comment from the scsi-subsystem or the SATA driver.
> >> 
> >> I usually only grep the log to try to find specific messages.
> >> If I know the time-period something weird happened in, I tend to go
> >> through the unfiltered log for that period.
> >
> > I have been scanning dmesg and /var/log/messages by eye and not noticed
> > anything. I'll keep doing it though.
> 
> What does your fstab look like?

I gave the two relevant lines in my first message, but here's the whole thing:

LABEL=RescueSys     /                               ext4 relatime        1 1
LABEL=RescUsrBits   /usr-bits                       ext4 relatime        1 2
/dev/md1            /boot                           ext2 relatime,noauto 1 2
/dev/md5            /mnt/main             ext4   relatime,noauto,dev,exec 0 2
/dev/vg7/local      /mnt/main/usr/local             ext4 relatime,noauto 0 3
/dev/vg7/home       /mnt/main/home                  ext4 relatime,noauto 0 3
/dev/vg7/common     /mnt/main/home/prh/common       ext4 relatime,noauto 0 4
/dev/vg7/virt       /mnt/main/home/prh/.VirtualBox  ext4 relatime,noauto 0 4
/dev/vg7/boinc      /mnt/main/home/prh/boinc        ext4 relatime,noauto 0 4
/dev/vg7/var        /mnt/main/var                   ext4 relatime,noauto 0 2
/dev/vg7/portage    /mnt/main/usr/portage           ext4 relatime,noauto 0 2
/dev/vg7/packages   /mnt/main/usr/portage/packages  ext4 relatime,noauto 0 3
/dev/vg7/distfiles  /mnt/main/usr/portage/distfiles ext4 relatime,noauto 0 3
/dev/vg7/opt        /mnt/main/opt                   ext4 relatime,noauto 0 3
/dev/vg7/atom       /mnt/main/mnt/atom              ext4 relatime,noauto 0 3
/dev/vg7/atomresc   /mnt/main/mnt/atomresc          ext4 relatime,noauto 0 3
/dev/vg7/tpad       /mnt/main/mnt/tpad              ext4 relatime,noauto 0 3
/dev/vg7/vartmp     /mnt/main/var/tmp               ext4 relatime,noauto 0 3
/dev/sdc5           /mnt/sdc              ext4    relatime,noauto,user 0 0
/dev/sdd5           /mnt/sdd              ext4   relatime,noauto,user 0 0
/dev/sr0            /mnt/dvd                      iso9660 noauto,user    0 0
/dev/sda2           none                            swap sw              0 0
/dev/sdb2           none                            swap sw              0 0
tmpfs               /tmp                  tmpfs  nodev,nosuid,size=6G 0 0
proc                /proc                           proc defaults        0 0
shm                 /dev/shm              tmpfs  nodev,nosuid,noexec 0 0
/dev/md8            /mnt/qt5                        ext4 noauto,relatime 0 0
/dev/vg9/home       /mnt/qt5/home                   ext4 noauto,relatime 0 0
/dev/vg9/var        /mnt/qt5/var                    ext4 noauto,relatime 0 0
/dev/vg9/vartmp     /mnt/qt5/var/tmp   ext4  noauto,relatime,nosuid,nodev 0 0
/dev/vg9/local      /mnt/qt5/usr/local              ext4 noauto,relatime 0 0
/dev/vg9/portage    /mnt/qt5/usr/portage            ext4 noauto,relatime 0 0
/dev/vg9/packages   /mnt/qt5/usr/portage/packages   ext4 noauto,relatime 0 0
/dev/vg9/distfiles  /mnt/qt5/usr/portage/distfiles  ext4 noauto,relatime 0 0

> And maybe some more info, like which kernel version. Mount version.
> And maybe check for some weird crontab entry somewhere?

No cron installed. Kernel was 4.0.5 at the time of writing, upgraded to 4.0.9 
yesterday. Mount is in sys-apps/util-linux-2.26.2 - was 2.25.2-r2 until 27 
September but the problem occurred with both versions.

> You could also rule out the use of umount by replacing it with a wrapper
> script that logs every call with as much info as is possible?

Hm. That may be above my bash grade.

I'm inclined to suspect the kernel. No real evidence, just that I've booted 
the rescue system twice since installing the new kernel and everything worked 
as it should.

-- 
Rgds
Peter