A Gluster NFS share hangs and the "glusterfsd" has high CPU utilization.

Issue

• A Gluster NFS share becomes unresponsive to NFS clients when doing an "rsync" of a huge file. This problem has also been observed when testing with some I/O bench marking tools, like "bonnie++" or "iozone". Below are some potential error messages in the "/var/log/glusterfs/nfs.log":

[2013-05-21 13:15:04.583293] E [rpc-clnt.c:208:call_bail] 0-mirror-client-0: bailing out frame type(GlusterFS 3.1) op(WRITE(13)) xid = 0x28841448x sent = 2013-05-21 12:45:01.137353. timeout = 1800