[Bug 107301] system hang during ext4 xattr operation

[bugzilla-daemon@bugzilla.kernel.org]

https://bugzilla.kernel.org/show_bug.cgi?id=107301

--- Comment #2 from Mehdi Abaakouk <sileht@sileht.net> ---
Yes ceph heavy uses xattr

I don't have a 'step by step to reproduce' list. But I have the issue ~4-5
times per day on a ceph cluster since the first incident.

On November 5, 2015 6:02:57 PM GMT+01:00, bugzilla-daemon@bugzilla.kernel.org
wrote:
>https://bugzilla.kernel.org/show_bug.cgi?id=107301
>
>Jan Kara <jack@suse.cz> changed:
>
>           What    |Removed                     |Added
>----------------------------------------------------------------------------
>             Status|NEW                         |NEEDINFO
>
>--- Comment #1 from Jan Kara <jack@suse.cz> ---
>Hum, from the stack traces in the log it seems mb_cache_entry_alloc()
>is racing
>with other operations on the LRU and restarting all the time. Another
>possibility is that there are lots of entries in the LRU and it takes a
>long
>time to scan (if I remember right ceph is a heavy user of xattrs).
>
>The first problem would be easily fixed by adding cond_resched() at
>appropriate
>place, the second problem would require more intrusive changes in how
>LRU is
>handled.
>
>Can you reproduce the issue?
>
>-- 
>You are receiving this mail because:
>You reported the bug.

-- 
You are receiving this mail because:
You are watching the assignee of the bug.