[Linux-cluster] Re: Linux-cluster Digest, Vol 64, Issue 10

[Bob Peterson]

----- "Wendell Dingus" <wendell at bisonline.com> wrote:
| Well, here's the entire list of blocks it ignored and the entire
| message section.
| Perhaps I'm just overlooking it but I'm not seeing anything in the
| messages
| that appears to be a block number. Maybe 1633350398 but if so it is
| not a match.

Your assumption is correct.  The block number was 1633350398, which
is labeled "bh = " for some reason.

| Anyway, since you didn't specifically say a new/fixed version of fsck
| was 
| imminent and that it would likely fix this we began plan B today. We

Yesterday I pushed a newer gfs_fsck and fsck.gfs2 to their appropriate
git source repositories.  So you can build that version from source if
you need it right away.  But it sounds like it wouldn't have helped
your problem anyway.  What would really be nice is if there is a way
to recreate the problem in our lab.  In theory, this error could be
caused by a hardware problem too.

| plugged
| in another drive, placed a GFS2 filesystem on it and am actively
| copying files
| off to it now. Fingers crossed that nothing will hit a disk block that
| causes
| this again but I could be so lucky probably...

It's hard to say whether you'll hit it again.

Regards,

Bob Peterson
Red Hat File Systems