[Ocfs2-users] Read Only FileSystem on both Cluster nodes due to internal FileSystem Error in OCFS2

[Harsha A Patankar]

Hi,
     We observed that due to below error "ocfs2_commit_truncate: Inode
387098 has empty extent block at 387098".
Due to this The /dev/drbd0 (/dev/sdb2) filesystem on both Nodes we set to
read only even though the actual error was on disk of one Node.
So I wanted to know how to fence the Node when DISK I/O Error is seen in
OCFS2 filesystem.
Is this the recommended solution.
echo 1 > /proc/fs/ocfs2_nodemanager/fence_method


Feb 26 15:32:27 MC0 kernel: : [7823386.409471] OCFS2: ERROR (device drbd0):
ocfs2_commit_truncate: Inode 387098 has empty extent block at 387098
Feb 26 15:32:27 MC0 kernel: : [7823386.409474]
Feb 26 15:32:27 MC0 kernel: : [7823386.421715]
(ocfs2_wq,2850,0):ocfs2_truncate_for_delete:597 ERROR: status = -30
Feb 26 15:32:27 MC0 kernel: : [7823386.421719]
(ocfs2_wq,2850,0):ocfs2_wipe_inode:773 ERROR: status = -30
Feb 26 15:32:27 MC0 kernel: : [7823386.421722]
(ocfs2_wq,2850,0):ocfs2_delete_inode:1071 ERROR: status = -30

regards,
Harsha
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://oss.oracle.com/pipermail/ocfs2-users/attachments/20140331/037422c2/attachment.html