I installed kernel 3.3 from http://kernel.ubuntu.com/~kernel-
ppa/mainline/v3.3-precise/ on 1 of the hosts.

Next, i mounted my gfs2 volume on 1 node, created a new empty file in
it, unmounted the gfs volume, and then mounted it on all 3 nodes.

Then i simply rm'ed the file on each node ( starting with the normal
3.2-kernel based machines ).

On the 2 3.2 machines, rm triggered an oops, and the file wasn't gone. I
then rm'ed it on the 3.3 kernel machine, which went perfectly. gfs mount
still reachable afterwards, no oops. Both other nodes hung when trying
to ls into the mount.

I tried booting the kernel from http://kernel.ubuntu.com/~kernel-
ppa/mainline/v3.2.28-precise/ to try and narrow down the scope, but that
one doesnt seem to be coming up anymore, so i'll have remote hands look
at it asap.

But at least this kind of narrows down the scope?

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1020207

Title:
  gfs2 kernel oops when deleting file on other cluster node

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/gfs2-utils/+bug/1020207/+subscriptions

-- 
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Reply via email to