On 25 October 2011 18:05, Burnash, James <jburnash at knight.com> wrote: > Top posting for visibility:**** > > ** ** > > Here is the pertinent line in the log:**** > > ** ** > > [2011-10-25 11:39:56.762720] E [glusterfsd-mgmt.c:628:mgmt_getspec_cbk] > 0-glusterfs: failed to get the 'volume file' from server > [2011-10-25 11:39:56.762775] E [glusterfsd-mgmt.c:695:mgmt_getspec_cbk] > 0-mgmt: failed to fetch volume file (key:/mnt) > > **** > > ** ** > > Did you move your volume file or in some other way make it inaccessible for > the client? The volume file tells the client essentially how to access files > and directories on the GlusterFS ? if the client can get that, the timeout > is expected (it figures that the server has gone away and that you will try > something different.**** > > ** ** > > I suggest trying this mount line instead (assuming that your servers are > named consistently:**** > > ** ** > > mount -t glusterfs gluster02:/volume01 /mnt/gluster**** > > ** ** > > and see if you get the same message**** > > ** ** > > James Burnash**** > > Unix Engineer**** > > Knight Capital Group**** > > I thank you for the advice. I've gave another look to the bricks, ando on gluster03 someone tried to mount the volume inside the /mnt, so when the client machine tried to mount /mn/gluster the gluster03 crashed all the servers. Problem solved. Thanks MV. -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://gluster.org/pipermail/gluster-users/attachments/20111025/ec2e5de8/attachment.htm>
