On 08/09/2013 10:58 AM, Jeff Moskow wrote:
Hi,
I have a 5 node ceph cluster that is running well (no problems using any of the
rbd images and that's really all we use).
I have replication set to 3 on all three pools (data, metadata and rbd).
"ceph -s" reports:
health HEALTH_WARN 3 pgs degraded; 114 pgs stuck unclean; recovery 5746/384795 degraded (1.493%)
I have tried everything I could think of to clear/fix those errors and they persist.
Did you restart the primary OSD for that PGs?
Wido
Most of them appear to be a problem with not having 3 copies....
0.2a0 0 0 0 0 0 0 0 active+remapped 2013-08-06 05:40:07.874427 0'0 21920'388 [4,7] [4,7,8] 0'0 2013-08-04 08:59:34.035198 0'0 2013-07-29 01:49:40.018625
4.1d9 260 0 238 0 1021055488 0 0 active+remapped 2013-08-06 05:56:20.447612 21920'12710 21920'53408 [6,13] [6,13,4] 0'0 2013-08-05 06:59:44.717555 0'0 2013-08-05 06:59:44.717555
1.1dc 0 0 0 0 0 0 0 active+remapped 2013-08-06 05:55:44.687830 0'0 21920'3003 [6,13] [6,13,4] 0'0 2013-08-04 10:56:51.226012 0'0 2013-07-28 23:47:13.404512
0.1dd 0 0 0 0 0 0 0 active+remapped 2013-08-06 05:55:44.687525 0'0 21920'3003 [6,13] [6,13,4] 0'0 2013-08-04 10:56:45.258459 0'0 2013-08-01 05:58:17.141625
1.29f 0 0 0 0 0 0 0 active+remapped 2013-08-06 05:40:07.882865 0'0 21920'388 [4,7] [4,7,8] 0'0 2013-08-04 09:01:40.075441 0'0 2013-07-29 01:53:10.068503
1.118 0 0 0 0 0 0 0 active+remapped 2013-08-06 05:50:34.081067 0'0 21920'208 [8,15] [8,15,5] 0'0 2034-02-12 23:20:03.933842 0'0 2034-02-12 23:20:03.933842
0.119 0 0 0 0 0 0 0 active+remapped 2013-08-06 05:50:34.095446 0'0 21920'208 [8,15] [8,15,5] 0'0 2034-02-12 23:18:07.310080 0'0 2034-02-12 23:18:07.310080
4.115 248 0 226 0 987364352 0 0 active+remapped 2013-08-06 05:50:34.112139 21920'6840 21920'42982 [8,15] [8,15,5] 0'0 2013-08-05 06:59:18.303823 0'0 2013-08-05 06:59:18.303823
4.4a 241 0 286 0 941573120 0 0 active+degraded 2013-08-06 12:00:47.758742 21920'85238 21920'206648 [4,6] [4,6] 0'0 2013-08-05 06:58:36.681726 0'0 2013-08-05 06:58:36.681726
0.4e 0 0 0 0 0 0 0 active+remapped 2013-08-06 12:00:47.765391 0'0 21920'489 [4,6] [4,6,1] 0'0 2013-08-04 08:58:12.783265 0'0 2013-07-28 14:21:38.227970
Can anyone suggest a way to clear this up?
Thanks!
Jeff
--
Wido den Hollander
42on B.V.
Phone: +31 (0)20 700 9902
Skype: contact42on
_______________________________________________
ceph-users mailing list
ceph-users@xxxxxxxxxxxxxx
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
