[Ocfs2-users] FSCK may be failing and corrupting my disk???

[Eric Raskin]

Hi:

I am running a two-node Oracle VM Server 2.2.2 installation.   We were
having some strange problems creating new virtual machines, so I shut
down the systems and unmounted the OVS Repository (ocfs2 file system on
Equallogic equipment).

I ran a fsck -y first, which replayed the logs and said all was clean. 
But, I am pretty sure there are other issues, so I started an fsck -fy

One of the messages I got was:

Cluster 161213953 is claimed by the following inodes:
  <76289548>
  /running_pool/450_gebidb/System.img
[DUP_CLUSTERS_CLONE] Inode "(null)" may be cloned or deleted to break
the claim it has on its clusters. Clone inode "(null)" to break claims
on clusters it shares with other inodes? y

I then watched with an strace -p <fsck process> to see what was
happening, since it was taking a long time with no messages.  I see:

pwrite64(3,
"INODE01\0H\26O}\377\377\22\0\0\0\0\0\0$\0\0\0\0\0\0\0\0\0\0"..., 4096,
90112) = 4096
pwrite64(3,
"EXBLK01\0\0\0\0\0\0\0\0\0\0\0+\3H\26O}\306\374&\0\0\0\0\0"..., 4096,
10465599488) = 4096
pwrite64(3,
"GROUP01\0\300\17\0\4P\0\0\0H\26O}\0\0\0\0\0\0\0\0\0\0\0\0"..., 4096,
10462699520) = 4096
pwrite64(3,
"INODE01\0H\26O}\377\377\22\0\0\0\0\0\0$\0\0\0\0\0\0\0\0\0\0"..., 4096,
90112) = 4096
pwrite64(3,
"EXBLK01\0\0\0\0\0\0\0\0\0\0\0/\3H\26O}\302\374&\0\0\0\0\0"..., 4096,
10465583104) = 4096
pwrite64(3,
"GROUP01\0\300\17\0\4Q\0\0\0H\26O}\0\0\0\0\0\0\0\0\0\0\0\0"..., 4096,
10462699520) = 4096
pwrite64(3,
"INODE01\0H\26O}\377\377\22\0\0\0\0\0\0$\0\0\0\0\0\0\0\0\0\0"..., 4096,
90112) = 4096
pwrite64(3,
"EXBLK01\0\0\0\0\0\0\0\0\0\0\0003\3H\26O}\274\374&\0\0\0\0\0"..., 4096,
10465558528) = 4096
pwrite64(3,
"INODE01\0H\26O}\0\0L\0\0\0\0\0\24\346\17\0\0\0\0\0\0\0\0\0"..., 4096,
2686701568) = 4096
pwrite64(3,
"GROUP01\0\300\17\0~\3\0#\0H\26O}\0\0\0\0\0n\0\1\0\0\0\0"..., 4096,
100940120064) = 4096
pwrite64(3,
"INODE01\0H\26O}\377\377\7\0\0\0\0\0\0\6\0\30\0\0\0\0\0\0\0\0"..., 4096,
45056) = 4096
pwrite64(3,
"GROUP01\0\300\17\0\4P\0\0\0H\26O}\0\0\0\0\0\0\0\0\0\0\0\0"..., 4096,
10462699520) = 4096
pwrite64(3,
"INODE01\0H\26O}\377\377\22\0\0\0\0\0\0$\0\0\0\0\0\0\0\0\0\0"..., 4096,
90112) = 4096
pwrite64(3,
"EXBLK01\0\0\0\0\0\0\0\0\0\0\0\272\2H\26O}\274\374&\0\0\0\0\0"..., 4096,
10465558528) = 4096
pwrite64(3,
"EXBLK01\0\0\0\0\0\0\0\0\0\0\0003\3H\26O}\274\374&\0\0\0\0\0"..., 4096,
10465558528) = 4096
pwrite64(3,
"GROUP01\0\300\17\0\4O\0\0\0H\26O}\0\0\0\0\0\0\0\0\0\0\0\0"..., 4096,
10462699520) = 4096
pwrite64(3,
"INODE01\0H\26O}\377\377\22\0\0\0\0\0\0$\0\0\0\0\0\0\0\0\0\0"..., 4096,
90112) = 4096

This is going on and on.  It looks like it is writing lots of entries to
fix one duplicate inode???

At this point, I have aborted the fsck, as I am worried that it is
completely trashing our OVS repository disk.

Can anybody shed some light on this before I restart the fsck?  We need
to be back up and running ASAP!

Thanks in advance!
-- 
-----------------------------------------------------------------------------------------------------------------------------------------------
Eric H. Raskin 	914-765-0500 x120
Professional Advertising Systems Inc. 	914-765-0503 fax
200 Business Park Dr Suite 304 	eraskin at paslists.com
Armonk, NY 10504 	http://www.paslists.com

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://oss.oracle.com/pipermail/ocfs2-users/attachments/20140322/1ef20914/attachment.html