[Gluster-users] GlusterFS on XFS + ext4 runs out of memory

Vladimir Yakovlev mrquesty at gmail.com
Thu Jan 31 04:26:26 UTC 2013


Hi Everyone,

I was running GlusterFS 3.3.1 on ext4 with CentOS 6.3, but it ran out of
memory quickly because of known bug in 64/32 bits operations with ext4.

I built another machine with XFS and added it to volume, after full heal,
switched to the new machine.

The configuration is the following: 1st machine with CentOS 6.3 with XFS,
2nd machine with CentOS 6.3 with ext4. They are in replica (factor 2).

But now I have memory leaks on 1st machine with XFS as well. Glusterfs
process ate 8GB RAM for 3 days.

Can it be because of 2nd machine with ext4 or is it anything else?

E.g. if I migrate 2nd machine to XFS, will memory leaks disappear?

Thanks a lot in advance,
BR,
Vladimir.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://supercolony.gluster.org/pipermail/gluster-users/attachments/20130131/7b10d3ed/attachment.html>


More information about the Gluster-users mailing list