Yeah, this is pretty weird. Can you run "df" on each of your OSDs and see if any of them individually have a strange output? -Greg Software Engineer #42 @ http://inktank.com | http://ceph.com On Tue, Aug 12, 2014 at 6:52 AM, Karan Singh <karan.singh@xxxxxx> wrote: > Forgot to mention i am observing EB in ceph -s output , does it mean > Exabyte ;-) > > > # ceph -s > cluster 009d3518-e60d-4f74-a26d-c08c1976263c > health HEALTH_WARN 'cache-pool' at/near target max > monmap e3: 3 mons at > mdsmap e14: 1/1/1 up {0=storage0101-ib=up:active} > osdmap e194215: 402 osds: 402 up, 402 in > pgmap v743051: 31168 pgs, 22 pools, 8 EB data, 378 kobjects > 17508 GB used, 1284 TB / 1301 TB avail > 31168 active+clean > > > - Karan Singh - > > > On 12 Aug 2014, at 16:45, Karan Singh <karan.singh@xxxxxx> wrote: > > Hello Developers > > I have encountered some wired output of ceph df command , suddenly > > > When i was writing some data on cache-pool , and checked its used % , i > found some used as 8E ( don’t know what is this ) and the used % for > cache-pool was 0 > > > # ceph df > GLOBAL: > SIZE AVAIL RAW USED %RAW USED > 1301T 1284T 17518G 1.31 > POOLS: > NAME ID USED %USED OBJECTS > data 0 801M 0 2 > metadata 1 801M 0 22 > rbd 2 0 0 0 > .rgw 3 3524 0 26 > .rgw.root 4 778 0 3 > .rgw.control 5 0 0 8 > .rgw.buckets 6 8201M 0 2298 > .rgw.buckets.index 7 0 0 13 > .users.swift 8 7 0 1 > volumes 9 1106G 0.08 283387 > images 10 40960k 0 8 > backups 11 0 0 0 > .rgw.gc 12 0 0 32 > .users.uid 13 848 0 5 > .users 14 16 0 2 > .log 15 153k 0 37 > 16 0 0 0 > hpsl4540 21 110G 0 28152 > hpdl380 22 245G 0.02 62688 > EC-2-2 23 6338G 0.48 4859 > cache-pool 24 8E 0 5849 ## What > is the meaning of E here , also please note used % for cache-pool is 0 here > ssd 25 25196M 0 5464 > > > After some time when cache-poo used value changed to 7E and used % as > 644301.19 . While there were no objects in the cache-pool > > > # ceph df > GLOBAL: > SIZE AVAIL RAW USED %RAW USED > 1301T 1284T 17508G 1.31 > POOLS: > NAME ID USED %USED OBJECTS > data 0 801M 0 2 > metadata 1 801M 0 22 > rbd 2 0 0 0 > .rgw 3 3524 0 26 > .rgw.root 4 778 0 3 > .rgw.control 5 0 0 8 > .rgw.buckets 6 8201M 0 2298 > .rgw.buckets.index 7 0 0 13 > .users.swift 8 7 0 1 > volumes 9 1106G 0.08 283387 > images 10 40960k 0 8 > backups 11 0 0 0 > .rgw.gc 12 0 0 32 > .users.uid 13 848 0 5 > .users 14 16 0 2 > .log 15 153k 0 37 > 16 0 0 0 > hpsl4540 21 110G 0 28152 > hpdl380 22 245G 0.02 62688 > EC-2-2 23 6338G 0.48 4843 > cache-pool 24 7E 644301.19 1056 > ## The used % for cache-pool has become 644301.19 > ssd 25 25196M 0 5464 > # > > > # rados -p cache-pool ls > # > > > > > Is this a bug , if yes , then is is already known. Do you want me to raise a > bug ticket in tracker.ceph.com ? > > > > **************************************************************** > Karan Singh > Cloud computing group > CSC - IT Center for Science, > Keilaranta 14, P. O. Box 405, FIN-02101 Espoo, Finland > tel. +358 9 4572001 > fax +358 9 4572302 > http://www.csc.fi/ > **************************************************************** > > _______________________________________________ > ceph-users mailing list > ceph-users@xxxxxxxxxxxxxx > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com > > > > _______________________________________________ > ceph-users mailing list > ceph-users@xxxxxxxxxxxxxx > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com > -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
