up vote 0 down vote favorite
share [g+] share [fb]

I have 5 OCFS2 volumes on an iSCSI SAN that a handful (5) of servers mount. There is one volume in particular that never mounts cleanly after reboot. The error you get when trying to mount is

hamguy:/dev/disk/by-label # /etc/init.d/ocfs2 start
Starting Oracle Cluster File System (OCFS2) mount.ocfs2: Bad magic number in inode while
trying to determine heartbeat information

The only way I've found to fix it is by

fsck.ocfs2 /dev/sde1

fsck reports an error trying to lock the cluster down, and then exits. The volume is now able to be mounted.

OCFS2 information

kendal@hamguy:~> rpm -qa |grep ocfs2
ocfs2console-1.4.1-25.6.x86_64
ocfs2-tools-o2cb-1.4.1-25.6.x86_64
ocfs2-tools-1.4.1-25.6.x86_64

Operating system is OpenSuSE 11.2 x86.

Has anyone some insight into this?

Thanks,

Kendall

link|improve this question
72% accept rate
feedback

1 Answer

up vote 0 down vote accepted 
dump2fs /dev/sde1 | grep 'superblock'
fsck.ocfs2 -b ALTERNATE_SB # is one of the backup superblocks.

i would have a backup before ...

link|improve this answer
Thanks. Any idea what the root cause is? – Kendall Apr 27 '11 at 21:38
maybe had a dirty umount ... as in crashed server - if is what you need it please accept the answer - thanks – silviud Apr 28 '11 at 1:03
I guess the crash server is possible...AFAIK, this has been going on for a few years, and as been around since the volumes were created. – Kendall Apr 28 '11 at 18:20
feedback

Your Answer

 
or
required, but never shown
discard

Not the answer you're looking for? Browse other questions tagged or ask your own question.