Hi Arun, How did you end up with a 'working' cluster with so many pgs per OSD?
"too many PGs per OSD (2968 > max 200)" To (temporarily) allow this kind of pgs per osd you could try this: Change these values in the global section in your ceph.conf: mon max pg per osd = 200 osd max pg per osd hard ratio = 2 It allows 200*2 = 400 Pgs per OSD before disabling the creation of new pgs. Above are the defaults (for Luminous, maybe other versions too) You can check your current settings with: ceph daemon mon.ceph-mon01 config show |grep pg_per_osd Since your current pgs per osd ratio is way higher then the default you could set them to for instance: mon max pg per osd = 1000 osd max pg per osd hard ratio = 5 Which allow for 5000 pgs per osd before disabling creation of new pgs. You'll need to inject the setting into the mons/osds and restart mgrs to make them active. ceph tell mon.* injectargs ‘--mon_max_pg_per_osd 1000’ ceph tell mon.* injectargs ‘--osd_max_pg_per_osd_hard_ratio 5’ ceph tell osd.* injectargs ‘--mon_max_pg_per_osd 1000’ ceph tell osd.* injectargs ‘--osd_max_pg_per_osd_hard_ratio 5’ restart mgrs Kind regards, Caspar Op vr 4 jan. 2019 om 04:28 schreef Arun POONIA < arun.poo...@nuagenetworks.net>: > Hi Chris, > > Indeed that's what happened. I didn't set noout flag either and I did > zapped disk on new server every time. In my cluster status fre201 is only > new server. > > Current Status after enabling 3 OSDs on fre201 host. > > [root@fre201 ~]# ceph osd tree > ID CLASS WEIGHT TYPE NAME STATUS REWEIGHT PRI-AFF > -1 70.92137 root default > -2 5.45549 host fre101 > 0 hdd 1.81850 osd.0 up 1.00000 1.00000 > 1 hdd 1.81850 osd.1 up 1.00000 1.00000 > 2 hdd 1.81850 osd.2 up 1.00000 1.00000 > -9 5.45549 host fre103 > 3 hdd 1.81850 osd.3 up 1.00000 1.00000 > 4 hdd 1.81850 osd.4 up 1.00000 1.00000 > 5 hdd 1.81850 osd.5 up 1.00000 1.00000 > -3 5.45549 host fre105 > 6 hdd 1.81850 osd.6 up 1.00000 1.00000 > 7 hdd 1.81850 osd.7 up 1.00000 1.00000 > 8 hdd 1.81850 osd.8 up 1.00000 1.00000 > -4 5.45549 host fre107 > 9 hdd 1.81850 osd.9 up 1.00000 1.00000 > 10 hdd 1.81850 osd.10 up 1.00000 1.00000 > 11 hdd 1.81850 osd.11 up 1.00000 1.00000 > -5 5.45549 host fre109 > 12 hdd 1.81850 osd.12 up 1.00000 1.00000 > 13 hdd 1.81850 osd.13 up 1.00000 1.00000 > 14 hdd 1.81850 osd.14 up 1.00000 1.00000 > -6 5.45549 host fre111 > 15 hdd 1.81850 osd.15 up 1.00000 1.00000 > 16 hdd 1.81850 osd.16 up 1.00000 1.00000 > 17 hdd 1.81850 osd.17 up 0.79999 1.00000 > -7 5.45549 host fre113 > 18 hdd 1.81850 osd.18 up 1.00000 1.00000 > 19 hdd 1.81850 osd.19 up 1.00000 1.00000 > 20 hdd 1.81850 osd.20 up 1.00000 1.00000 > -8 5.45549 host fre115 > 21 hdd 1.81850 osd.21 up 1.00000 1.00000 > 22 hdd 1.81850 osd.22 up 1.00000 1.00000 > 23 hdd 1.81850 osd.23 up 1.00000 1.00000 > -10 5.45549 host fre117 > 24 hdd 1.81850 osd.24 up 1.00000 1.00000 > 25 hdd 1.81850 osd.25 up 1.00000 1.00000 > 26 hdd 1.81850 osd.26 up 1.00000 1.00000 > -11 5.45549 host fre119 > 27 hdd 1.81850 osd.27 up 1.00000 1.00000 > 28 hdd 1.81850 osd.28 up 1.00000 1.00000 > 29 hdd 1.81850 osd.29 up 1.00000 1.00000 > -12 5.45549 host fre121 > 30 hdd 1.81850 osd.30 up 1.00000 1.00000 > 31 hdd 1.81850 osd.31 up 1.00000 1.00000 > 32 hdd 1.81850 osd.32 up 1.00000 1.00000 > -13 5.45549 host fre123 > 33 hdd 1.81850 osd.33 up 1.00000 1.00000 > 34 hdd 1.81850 osd.34 up 1.00000 1.00000 > 35 hdd 1.81850 osd.35 up 1.00000 1.00000 > -27 5.45549 host fre201 > 36 hdd 1.81850 osd.36 up 1.00000 1.00000 > 37 hdd 1.81850 osd.37 up 1.00000 1.00000 > 38 hdd 1.81850 osd.38 up 1.00000 1.00000 > [root@fre201 ~]# > [root@fre201 ~]# > [root@fre201 ~]# > [root@fre201 ~]# > [root@fre201 ~]# > [root@fre201 ~]# ceph -s > cluster: > id: adb9ad8e-f458-4124-bf58-7963a8d1391f > health: HEALTH_ERR > 3 pools have many more objects per pg than average > 585791/12391450 objects misplaced (4.727%) > 2 scrub errors > 2374 PGs pending on creation > Reduced data availability: 6578 pgs inactive, 2025 pgs down, > 74 pgs peering, 1234 pgs stale > Possible data damage: 2 pgs inconsistent > Degraded data redundancy: 64969/12391450 objects degraded > (0.524%), 616 pgs degraded, 20 pgs undersized > 96242 slow requests are blocked > 32 sec > 228 stuck requests are blocked > 4096 sec > too many PGs per OSD (2768 > max 200) > > services: > mon: 3 daemons, quorum ceph-mon01,ceph-mon02,ceph-mon03 > mgr: ceph-mon03(active), standbys: ceph-mon01, ceph-mon02 > osd: 39 osds: 39 up, 39 in; 96 remapped pgs > rgw: 1 daemon active > > data: > pools: 18 pools, 54656 pgs > objects: 6050k objects, 10942 GB > usage: 21900 GB used, 50721 GB / 72622 GB avail > pgs: 0.002% pgs unknown > 12.050% pgs not active > 64969/12391450 objects degraded (0.524%) > 585791/12391450 objects misplaced (4.727%) > 47489 active+clean > 3670 activating > 1098 stale+down > 923 down > 575 activating+degraded > 563 stale+active+clean > 105 stale+activating > 78 activating+remapped > 72 peering > 25 stale+activating+degraded > 23 stale+activating+remapped > 9 stale+active+undersized > 6 stale+activating+undersized+degraded+remapped > 5 stale+active+undersized+degraded > 4 down+remapped > 4 activating+degraded+remapped > 2 active+clean+inconsistent > 1 stale+activating+degraded+remapped > 1 stale+active+clean+remapped > 1 stale+remapped+peering > 1 remapped+peering > 1 unknown > > io: > client: 0 B/s rd, 208 kB/s wr, 22 op/s rd, 22 op/s wr > > > > Thanks > Arun > > > On Thu, Jan 3, 2019 at 7:19 PM Chris <bitskr...@bitskrieg.net> wrote: > >> If you added OSDs and then deleted them repeatedly without waiting for >> replication to finish as the cluster attempted to re-balance across them, >> its highly likely that you are permanently missing PGs (especially if the >> disks were zapped each time). >> >> If those 3 down OSDs can be revived there is a (small) chance that you >> can right the ship, but 1400pg/OSD is pretty extreme. I'm surprised the >> cluster even let you do that - this sounds like a data loss event. >> >> Bring back the 3 OSD and see what those 2 inconsistent pgs look like with >> ceph pg query. >> >> On January 3, 2019 21:59:38 Arun POONIA <arun.poo...@nuagenetworks.net> >> wrote: >> >>> Hi, >>> >>> Recently I tried adding a new node (OSD) to ceph cluster using >>> ceph-deploy tool. Since I was experimenting with tool and ended up deleting >>> OSD nodes on new server couple of times. >>> >>> Now since ceph OSDs are running on new server cluster PGs seems to be >>> inactive (10-15%) and they are not recovering or rebalancing. Not sure what >>> to do. I tried shutting down OSDs on new server. >>> >>> Status: >>> [root@fre105 ~]# ceph -s >>> 2019-01-03 18:56:42.867081 7fa0bf573700 -1 asok(0x7fa0b80017a0) >>> AdminSocketConfigObs::init: failed: AdminSocket::bind_and_listen: failed to >>> bind the UNIX domain socket to >>> '/var/run/ceph-guests/ceph-client.admin.4018644.140328258509136.asok': (2) >>> No such file or directory >>> cluster: >>> id: adb9ad8e-f458-4124-bf58-7963a8d1391f >>> health: HEALTH_ERR >>> 3 pools have many more objects per pg than average >>> 373907/12391198 objects misplaced (3.018%) >>> 2 scrub errors >>> 9677 PGs pending on creation >>> Reduced data availability: 7145 pgs inactive, 6228 pgs down, >>> 1 pg peering, 2717 pgs stale >>> Possible data damage: 2 pgs inconsistent >>> Degraded data redundancy: 178350/12391198 objects degraded >>> (1.439%), 346 pgs degraded, 1297 pgs undersized >>> 52486 slow requests are blocked > 32 sec >>> 9287 stuck requests are blocked > 4096 sec >>> too many PGs per OSD (2968 > max 200) >>> >>> services: >>> mon: 3 daemons, quorum ceph-mon01,ceph-mon02,ceph-mon03 >>> mgr: ceph-mon03(active), standbys: ceph-mon01, ceph-mon02 >>> osd: 39 osds: 36 up, 36 in; 51 remapped pgs >>> rgw: 1 daemon active >>> >>> data: >>> pools: 18 pools, 54656 pgs >>> objects: 6050k objects, 10941 GB >>> usage: 21727 GB used, 45308 GB / 67035 GB avail >>> pgs: 13.073% pgs not active >>> 178350/12391198 objects degraded (1.439%) >>> 373907/12391198 objects misplaced (3.018%) >>> 46177 active+clean >>> 5054 down >>> 1173 stale+down >>> 1084 stale+active+undersized >>> 547 activating >>> 201 stale+active+undersized+degraded >>> 158 stale+activating >>> 96 activating+degraded >>> 46 stale+active+clean >>> 42 activating+remapped >>> 34 stale+activating+degraded >>> 23 stale+activating+remapped >>> 6 stale+activating+undersized+degraded+remapped >>> 6 activating+undersized+degraded+remapped >>> 2 activating+degraded+remapped >>> 2 active+clean+inconsistent >>> 1 stale+activating+degraded+remapped >>> 1 stale+active+clean+remapped >>> 1 stale+remapped >>> 1 down+remapped >>> 1 remapped+peering >>> >>> io: >>> client: 0 B/s rd, 208 kB/s wr, 28 op/s rd, 28 op/s wr >>> >>> Thanks >>> -- >>> Arun Poonia >>> >>> _______________________________________________ >>> ceph-users mailing list >>> ceph-users@lists.ceph.com >>> http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com >>> >>> >> > > -- > Arun Poonia > > _______________________________________________ > ceph-users mailing list > ceph-users@lists.ceph.com > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com >
_______________________________________________ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com