On Fri, Nov 7, 2014 at 2:40 PM, Erik Logtenberg <erik@xxxxxxxxxxxxx> wrote: > Hi, > > My MDS is very slow, and it logs stuff like this: > > 2014-11-07 23:38:41.154939 7f8180a31700 0 log_channel(default) log > [WRN] : 2 slow requests, 1 included below; oldest blocked for > > 187.777061 secs > 2014-11-07 23:38:41.154956 7f8180a31700 0 log_channel(default) log > [WRN] : slow request 121.322570 seconds old, received at 2014-11-07 > 23:36:39.832336: client_request(client.7071:115 getattr pAsLsXsFs > #1000002bdbe 2014-11-07 23:36:39.000000) currently failed to rdlock, waiting > > Any idea what that rdlock is and what might cause it to fail? Each inode has a whole bunch of different locks associated with it for various things (file data, xattrs, ownership, linkage, etc) and different kinds of requests need different kinds locks. A getattr operation just needs read locks, which should generally not be a big deal unless you have another client currently using the file. In that case it will prompt the other client to flush everything and give up its locks (at least temporarily), and that is probably the slow step here. I'd look into why that might be the case. (You can explore a little bit more with the MDS' admin socket to get some more information out, or turn up the mds logging and search through it for the client requests reported as slow.) -Greg _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
 |