http://oss.sgi.com/bugzilla/show_bug.cgi?id=767
------- Additional Comments From gabriel@xxxxxxxx 2008-12-27 11:43 CST -------
(In reply to comment #1)
> Do you still see this?
It happened at least 2-3 times in 2008, last time being around september (not
related to the server load).
The 2 volumes shutting down have some things in common: they contain a large
number of files (I don't know if 2-5 million files - not in the same dir - is
"large" or "huge" to handle for that kind of FS), and they have directories with
20000+ files and long filenames (40 characters). On the other hand, I have a
mail archive containing very big Maildir directories (XFS, 10k+ files, long
filenames) without a single problem on it.
BTW, all the XFS volumes are all created with a big log size (-l size=128m) and
mounted with big log buffers (logbufs=8).
Is there some changes in the XFS current kernel code since 2.6.21.5 that could
be worth trying? Note this is a production file server that cannot be easily
upgraded.
--
Configure bugmail: http://oss.sgi.com/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.
|