Forgot to mention gluster version and O/S... Both client and server use:
Scientific Linux 6.4 (Kernel 2.6.32-431.11.2.el6.x86_64)
[root@gfs01a ~]# rpm -qa | grep gluster
glusterfs-libs-3.5.0-2.el6.x86_64
glusterfs-server-3.5.0-2.el6.x86_64
glusterfs-3.5.0-2.el6.x86_64
glusterfs-fuse-3.5.0-2.el6.x86_64
glusterfs-cli-3.5.0-2.el6.x86_64
glusterfs-rdma-3.5.0-2.el6.x86_64
------ Original Message ------
From: "David F. Robinson" <[email protected]>
To: [email protected]
Sent: 5/19/2014 10:58:57 AM
Subject: rsync + stale file handle
When I do an rsync to backup my workstations onto a gluster mounted
file system, I end up with thousands of healing problems. The heal
status repeatedly shows the same number of healed/failed during a
"gluster volume heal homegfs info statistics" check. There are over
9,000 files healed and roughly 1,000 that fail the heal. Each time
that heal runs, it reports the exact same number of healed and
heal-failed... The log file shows thousands of "W
[client-rpc-fops.c:1529:client3_3_inodelk_cbk] 0-homegfs-client-1:
remote operation failed: Stale file handle" errors messages and the log
file has grown to over 6GB.
These problems show up when I backup a workstation onto the
gluster-mounted file system using rsync. The file system is mounted as
"gfsib01a:/homegfs on /homegfs type fuse.glusterfs
(rw,allow_other,max_read=131072)" and I do the backup using "rsync -avh
/ /homegfs/wks_backup".
Note, that if I delete all of the backups from the gluster file system
(rm -rf /homegfs/wks_backup), the heal statistics return to normal and
show zero healed/split-brain/heal-failed...
Any suggestions for what I can do to resolve these issues and use my
gluster file system as a storage location for my workstations backups?
Thanks in advance for any help you can provide... Greatly
appreciated...
David
_______________________________________________
Gluster-users mailing list
[email protected]
http://supercolony.gluster.org/mailman/listinfo/gluster-users
