Hi,
I did try to restart the MDS server. The logs show the following error[187846.234448] init: ceph-mds (ceph/blade2-qq) main process (15077) killed by ABRT signal
[187846.234493] init: ceph-mds (ceph/blade2-qq) main process ended, respawning
[187846.687929] init: ceph-mds (ceph/blade2-qq) main process (15099) killed by ABRT signal
[187846.687977] init: ceph-mds (ceph/blade2-qq) respawning too fast, stopped
On Fri, May 17, 2013 at 3:33 PM, John Wilkins <john.wilkins@xxxxxxxxxxx> wrote:
Have you tried restarting your MDS server?
http://ceph.com/docs/master/rados/operations/operating/#operating-a-cluster
> _______________________________________________
On Fri, May 17, 2013 at 12:16 AM, Sridhar Mahadevan
<msridhar87@xxxxxxxxx> wrote:
> Hi,
>
> I have deployed the ceph object store using ceph-deploy.
> I tried to mount cephfs and I got struck with this error.
>
> sudo mount.ceph 192.168.35.82:/ /mnt/mycephfs -o
> name=admin,secret=AQDa5JJRqLxuOxAA77VljIjaAGWR6mGdL12NUQ==
>
> mount error 5 = Input/output error
>
> The output of the command
>
> # ceph -s
>
> health HEALTH_WARN 64 pgs degraded; mds cluster is degraded; mds
> blade2-qq is laggy
> monmap e1: 1 mons at {blade2-qq=192.168.35.82:6789/0}, election epoch 1,
> quorum 0 blade2-qq
> osdmap e56: 4 osds: 4 up, 4 in
> pgmap v834: 192 pgs: 128 active+clean, 64 active+clean+degraded; 0 bytes
> data, 50871 MB used, 437 GB / 513 GB avail
> mdsmap e6457: 1/1/1 up {0=blade2-qq=up:replay(laggy or crashed)}
>
> As it says the MDS has crashed. I dont see ceph-mds running in the MDS_Node.
> I executed ceph-deploy mds create <mds_node> and this starts the ceph-mds
> daemon in the mds_node, but I see that the ceph-mds daemon crashes after
> sometime.
>
> Kindly help me on this issue.
>
> --
> --sridhar
>
> ceph-users mailing list
> ceph-users@xxxxxxxxxxxxxx
> http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
>
--
John Wilkins
Senior Technical Writer
Intank
john.wilkins@xxxxxxxxxxx
(415) 425-9599
http://inktank.com
--
--sridhar
_______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
