It was indeed an error, thank you, Jens.
So it seems that I can't find the ceph-mds process at all. Similarly, my
ceph.log file is littered with these messages:
2013-05-06 13:14:31.750391 mon.0 10.81.2.100:6789/0 6280 : [INF] mdsmap
e25013: 1/1/1 up {0=a=up:active(laggy or crashed)}
Clearly this indicates a pretty deep problem, how can I go about seeing
what's wrong with ceph-mds?
On Sun, May 5, 2013 at 1:03 PM, Jens Kristian Søgaard <
j...@mermaidconsulting.dk> wrote:
> Hi Wyatt,
>
> [You sent me this email off list, I'm assuming that was an error, so I'm
> bringing it back on the list]
>
>
> ceph -s
>> health HEALTH_WARN mds a is laggy
>>
>
> What does this mean, and what can be done to fix it? Thanks, Jens!
>>
>
> It means that the mds hasn't responded to the monitor for too long.
>
> This can happen for example when the mds has crashed, if it is too busy or
> if the network has problems.
>
> Check that the server running the mds is still alive. Use for example
> "top" to determine that the server isn't terribly loaded or swapping, and
> that the mds process is still running.
>
> Then check the mds logfile to see if it has logged errors or a crash
> message.
>
> If you still haven't found any problems, try checking the network to see
> if you connectivity between the servers. Ensure that you haven't firewalled
> the mds so that it cannot talk to the monitors.
>
>
> --
> Jens Kristian Søgaard, Mermaid Consulting ApS,
> j...@mermaidconsulting.dk,
> http://www.mermaidconsulting.**com/ <http://www.mermaidconsulting.com/>
>
_______________________________________________
ceph-users mailing list
ceph-users@lists.ceph.com
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
