[Ocfs2-users] ocfs2_delete_inode kernel bug

[Andre Nathan]

On Tue, 2010-10-26 at 09:44 -0700, Sunil Mushran wrote:
> You'll have to run fsck.ocfs2 -fy to clear the older orphaned inodes.
> The upgrade will prevent future occurrences of this issue.

I actually did that right after rebooting into the new kernel. The
system ran fine for a day, but the kernel bug showed up after the backup
script ran.

My setup is the following: I have two servers sharing OCFS2 filesystems
through one dedicated 10Gbps interface.

There is also a backup server which exports OCFS2-formatted devices via
ATA-over-ethernet. This machine is connected to each server through
standard gigabit ethernet interfaces. The two servers mount their
respective volumes and run the backup script.

Could there be an issue with the fact that the backup storage network
interface is not the same as the interface used by the cluster? I'm
guessing not, since the AoE devices act like local disks for its
clients.

Thanks,
Andre