[Ocfs2-users] OCFS2 for shared Oracle home

[Roeschley, Dale]

I am new to this list so I hope this is the correct list to post this
question.

 

We are using OCFS2 for our shared Oracle home for CRS, ASM, and 2
database homes and the agent home.  However, this shared filesystem
locked up on us on Friday.  The linux ls command would hang sometimes;
CRS began evicting nodes.  We had to stop everything, reboot the nodes
with the OCFS2 mount point commented out, run fsck.ocfs2 in repair mode
which found a bunch of stuff and repaired it.  When we ran fsck.ocfs2
while the filesystem was mounted, it came up with 3 Inode messages.
When we ran it without the filesystem mounted, it came up with a number
of Inode  zero length directory messages, many, many global cluster
bitmap messages, and then a few Inode orphan directory messages.  When
we ran it a second time, it came up with 2 Inode link count messages.
The third time we ran it, it same up clean.

 

My questions are:

1)       Is OCFS2 stable enough to support shared homes for all the
Oracle homes?

2)       Is there any way to find out if an OCFS2 mount point is having
a problem before it gets to a critical point?

3)       Would moving all the db dump directories to local mount point
alleviate any of the issues with OCFS2?  The reason I ask this is
because of all the trace files and cdmp junk that Oracle creates, in
addition everything CRS is logging.

 

Thanks in advance for any suggestions or comments.

 

 

Dale Roeschley

Lead Database Administrator

Chicago Stock Exchange, Inc.

312-663-2328

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://oss.oracle.com/pipermail/ocfs2-users/attachments/20080225/f7739f79/attachment.html