LinuxMCE Forums

General => Users => Topic started by: Viking on August 03, 2009, 09:15:10 am

Title: Why does LinuxMCE 0710/0810 mount/umount the Storage partition every 6 seconds?
Post by: Viking on August 03, 2009, 09:15:10 am

Hi,

in the sys log i see that my video partition (/dev/md1 three disks RADI5) is mounten and unmounted all the time. Is that normal ?
I have seen this on 0710 and now also on 0810 alpha.

Syslog :

Code: [Select]

Aug  3 09:10:13 dcerouter kernel: [29179.315607] kjournald starting.  Commit interval 5 seconds
Aug  3 09:10:13 dcerouter kernel: [29179.316853] EXT3-fs: mounted filesystem with ordered data mode.
Aug  3 09:10:14 dcerouter kernel: [29180.306370] kjournald starting.  Commit interval 5 seconds
Aug  3 09:10:14 dcerouter kernel: [29180.307593] EXT3-fs: mounted filesystem with ordered data mode.
Aug  3 09:10:15 dcerouter kernel: [29181.082995] kjournald starting.  Commit interval 5 seconds
Aug  3 09:10:15 dcerouter kernel: [29181.084233] EXT3-fs: mounted filesystem with ordered data mode.
...

From /var/log/pluto/umount-wrapper.log :

Code: [Select]

9937 CALLED TO 'mount /dev/md1 /tmp/tmp.mDrhne9936'
9937 Mon Aug  3 09:07:59 CEST 2009 WaitLock 'mtab' (mount-wrapper)
9937 Mon Aug  3 09:07:59 CEST 2009 WaitLock 'mtab' (mount-wrapper) success
9937 Mon Aug  3 09:07:59 CEST 2009 Unlock 'mtab' (mount-wrapper)
9937 Mon Aug  3 09:07:59 CEST 2009 Unlock 'mtab' (mount-wrapper) success
9937 EXIT WITH STATUS 0


9952 CALLED TO 'umount -lf /tmp/tmp.mDrhne9936'
9952 Mon Aug  3 09:07:59 CEST 2009 WaitLock 'mtab' (umount-wrapper)
9952 Mon Aug  3 09:07:59 CEST 2009 WaitLock 'mtab' (umount-wrapper) success
9952 Mon Aug  3 09:07:59 CEST 2009 Unlock 'mtab' (umount-wrapper)
9952 Mon Aug  3 09:07:59 CEST 2009 Unlock 'mtab' (umount-wrapper) success
9952 EXIT WITH STATUS 0


10374 CALLED TO 'mount /dev/md1 /tmp/tmp.nswPx10372'
10374 Mon Aug  3 09:08:05 CEST 2009 WaitLock 'mtab' (mount-wrapper)
10374 Mon Aug  3 09:08:05 CEST 2009 WaitLock 'mtab' (mount-wrapper) success
10374 Mon Aug  3 09:08:05 CEST 2009 Unlock 'mtab' (mount-wrapper)
10374 Mon Aug  3 09:08:05 CEST 2009 Unlock 'mtab' (mount-wrapper) success
10374 EXIT WITH STATUS 0


10407 CALLED TO 'umount -lf /tmp/tmp.nswPx10372'
10407 Mon Aug  3 09:08:05 CEST 2009 WaitLock 'mtab' (umount-wrapper)
10407 Mon Aug  3 09:08:05 CEST 2009 WaitLock 'mtab' (umount-wrapper) success
10407 Mon Aug  3 09:08:05 CEST 2009 Unlock 'mtab' (umount-wrapper)
10407 Mon Aug  3 09:08:05 CEST 2009 Unlock 'mtab' (umount-wrapper) success
10407 EXIT WITH STATUS 0
Greetings
Viking

Title: Re: Why does LinuxMCE 0710/0810 mount/umount the Storage partition every 6 seconds?
Post by: pigdog on August 03, 2009, 12:50:55 pm

Hi,

This link looks the same.

http://forum.linuxmce.org/index.php?topic=6456.0

Title: Re: Why does LinuxMCE 0710/0810 mount/umount the Storage partition every 6 seconds?
Post by: Viking on August 03, 2009, 01:09:58 pm

Hi,

OK, thanks. Don't see the reason though.

Greetings
Viking

Title: Re: Why does LinuxMCE 0710/0810 mount/umount the Storage partition every 6 seconds?
Post by: merkur2k on August 03, 2009, 05:27:00 pm

yep, I just learned to live with it too. It's just how lmce works, most of it is caused by UpdateMedia since it is always running and looking for stuff to index.

Title: Re: Why does LinuxMCE 0710/0810 mount/umount the Storage partition every 6 secon
Post by: tschak909 on August 03, 2009, 06:17:08 pm

We have an intricate auto-mounter, with a very short timeout, and when updatemedia does its runs, it triggers the automount again.

-Thom

Title: Re: Why does LinuxMCE 0710/0810 mount/umount the Storage partition every 6 seconds?
Post by: merkur2k on August 03, 2009, 07:17:31 pm

I dont mind the constant mounting/unmounting, the annoying part for me is that it makes the kernel log completely useless because of the filesystem messages that get logged each time. I will try to find out if there is a way to suppress those...

Title: Re: Why does LinuxMCE 0710/0810 mount/umount the Storage partition every 6 seconds?
Post by: Viking on August 04, 2009, 12:21:41 pm

Quote from: merkur2k on August 03, 2009, 07:17:31 pm

I dont mind the constant mounting/unmounting, the annoying part for me is that it makes the kernel log completely useless because of the filesystem messages that get logged each time. I will try to find out if there is a way to suppress those...

Same here.

though I noticed on sunday that the server was really busy mounting/umounting. There was around 50 mount processes when I did an "ps ax|grep mount" and "mount" (without any options) took 10-20 sek. to complete.
So sometimes the system seems to get into trouble. I waited several hours, but it did not really get much better so I rebootet the core and it was gone.

Greetings
Viking