Disabled Filesystem

[Regina Mitchell]

I am getting the following errors and my syslog file and /u05 is disabled. Any ideas, could this be a kernel parameter that may need to be adjusted?
I have a n4000 server running 11.0.
un 2 00:52:04 usahs009 vmunix: vxfs: mesg 019: vx_log_add - /dev/vg05/lvol1 fi
le system log overflow
Jun 2 00:52:04 usahs009 vmunix: vxfs: mesg 031: vx_disable - /dev/vg05/lvol1 fi
le system disabled
Jun 2 00:52:09 usahs009 vmunix: vxfs: mesg 049: vx_esum_bad - /dev/vg05/lvol1 f
ile system extent allocation unit summary number 338 marked bad
Jun 2 00:52:09 usahs009 vmunix: vxfs: mesg 049: vx_esum_bad - /dev/vg05/lvol1 f
ile system extent allocation unit summary number 452 marked bad
Jun 2 00:52:09 usahs009 vmunix: vxfs: mesg 019: vx_log_add - /dev/vg05/lvol1 fi
le system log overflow
Jun 2 01:03:23 usahs009 sshd[6060]: Accepted keyboard-interactive/pam for root
from 206.122.126.25 port 43568 ssh2
Jun 2 01:08:28 usahs009 vmunix: vxfs: mesg 019: vx_log_add - /dev/vg05/lvol1 fi
le system log overflow
Jun 2 01:08:30 usahs009 above message repeats 244 times
Jun 2
Thanks!

[Patrick Wallek]

From TKB document "JFS 3.3 Kernel messages and solutions"

Submitted Date: 10/19/04
Title: JFS 3.3 Kernel messages and solutions
Document ID: UFSKBAN00001139
Last Modified Date: 10/26/04
http://www1.itrc.hp.com/service/cki/docDisplay.do?docLocale=en_US&docId=200000076281156


WARNING: msgcnt x: vxfs: mesg 019:
vx_log_add - mount_point file system log overflow

"Edited to comply with ITRC Guidelines"

[A. Clay Stephenson]

Unmount the filesystem and run a full fsck.

[Regina Mitchell]

This has happened approx 2 weeks ago and we did what you suggested here. It has happended again. I'm trying to determine the root cause. The DBA wants me to TC the server to get a crash dump for HP to analyze? Do you think a crash dump will capture the root cause?

[A. Clay Stephenson]

I don't think an analysis is going to help
but this message is very suspicious:

extent allocation unit summary number 452 marked bad

This strongly suggest that you are seeing i/o problems. Is syslog reporting any disk errors?

It never hurts to make sure that you are up to date on the latest vxfs and SCSI patches but if this is recurring after a -o full,nolog fsck then disk problems (or poor termination) are certainly something to be considered.

The only thing that really comes to mind is to make a script that uses adb to probe the kernel for some of the vx_ variables but I am really leaning towards a hardware problem.