Problem: I have some pgs with only two OSDs instead of 3 like all the other pgs have. This is causing active+undersized+degraded status.
History: 1. I started with 3 hosts, each with 1 OSD process (min_size 2) for a 1TB drive. 2. Added 3 more hosts, each with 1 OSD process for a 10TB drive. 3. Removed the original 3 1TB OSD hosts from the osd tree (reweight 0, wait, stop, remove, del osd&host, rm). 4. The last OSD to be removed would never return to active+clean after reweight 0. It returned undersized instead, but I went on with removal anyway, leaving me stuck with 5 undersized pgs. Things tried that didn't help: * give it time to go away on its own * Replace replicated default.rgw.buckets.data pool with erasure-code 2+1 version. * ceph osd lost 1 (and 2) * ceph pg repair (pgs from dump_stuck) * googled 'ceph pg undersized' and similar searches for help. Current status: $ ceph osd tree ID WEIGHT TYPE NAME UP/DOWN REWEIGHT PRIMARY-AFFINITY -1 19.19119 root default -5 1.00000 host osd1 3 1.00000 osd.3 up 1.00000 1.00000 -6 9.09560 host osd2 4 9.09560 osd.4 up 1.00000 1.00000 -2 9.09560 host osd3 0 9.09560 osd.0 up 1.00000 1.00000 $ ceph pg dump_stuck ok PG_STAT STATE UP UP_PRIMARY ACTING ACTING_PRIMARY 88.3 active+undersized+degraded [4,0] 4 [4,0] 4 97.3 active+undersized+degraded [4,0] 4 [4,0] 4 85.6 active+undersized+degraded [4,0] 4 [4,0] 4 87.5 active+undersized+degraded [0,4] 0 [0,4] 0 70.0 active+undersized+degraded [0,4] 0 [0,4] 0 $ ceph osd pool ls detail pool 70 'default.rgw.rgw.gc' replicated size 3 min_size 2 crush_rule 0 object_hash rjenkins pg_num 4 pgp_num 4 last_change 548 flags hashpspool stripe_width 0 pool 83 'default.rgw.buckets.non-ec' replicated size 3 min_size 2 crush_rule 0 object_hash rjenkins pg_num 8 pgp_num 8 last_change 576 owner 18446744073709551615 flags hashpspool stripe_width 0 pool 85 'default.rgw.control' replicated size 3 min_size 2 crush_rule 0 object_hash rjenkins pg_num 8 pgp_num 8 last_change 652 flags hashpspool stripe_width 0 pool 86 'default.rgw.data.root' replicated size 3 min_size 2 crush_rule 0 object_hash rjenkins pg_num 8 pgp_num 8 last_change 653 flags hashpspool stripe_width 0 pool 87 'default.rgw.gc' replicated size 3 min_size 2 crush_rule 0 object_hash rjenkins pg_num 8 pgp_num 8 last_change 654 flags hashpspool stripe_width 0 pool 88 'default.rgw.lc' replicated size 3 min_size 2 crush_rule 0 object_hash rjenkins pg_num 8 pgp_num 8 last_change 600 flags hashpspool stripe_width 0 pool 89 'default.rgw.log' replicated size 3 min_size 2 crush_rule 0 object_hash rjenkins pg_num 8 pgp_num 8 last_change 655 flags hashpspool stripe_width 0 pool 90 'default.rgw.users.uid' replicated size 3 min_size 2 crush_rule 0 object_hash rjenkins pg_num 8 pgp_num 8 last_change 662 flags hashpspool stripe_width 0 pool 91 'default.rgw.users.email' replicated size 3 min_size 2 crush_rule 0 object_hash rjenkins pg_num 8 pgp_num 8 last_change 660 flags hashpspool stripe_width 0 pool 92 'default.rgw.users.keys' replicated size 3 min_size 2 crush_rule 0 object_hash rjenkins pg_num 8 pgp_num 8 last_change 659 flags hashpspool stripe_width 0 pool 93 'default.rgw.buckets.index' replicated size 3 min_size 2 crush_rule 0 object_hash rjenkins pg_num 8 pgp_num 8 last_change 664 flags hashpspool stripe_width 0 pool 95 'default.rgw.intent-log' replicated size 3 min_size 2 crush_rule 0 object_hash rjenkins pg_num 4 pgp_num 4 last_change 656 flags hashpspool stripe_width 0 pool 96 'default.rgw.meta' replicated size 3 min_size 2 crush_rule 0 object_hash rjenkins pg_num 4 pgp_num 4 last_change 657 flags hashpspool stripe_width 0 pool 97 'default.rgw.usage' replicated size 3 min_size 2 crush_rule 0 object_hash rjenkins pg_num 4 pgp_num 4 last_change 658 flags hashpspool stripe_width 0 pool 98 'default.rgw.users.swift' replicated size 3 min_size 2 crush_rule 0 object_hash rjenkins pg_num 4 pgp_num 4 last_change 661 flags hashpspool stripe_width 0 pool 99 'default.rgw.buckets.extra' replicated size 3 min_size 2 crush_rule 0 object_hash rjenkins pg_num 4 pgp_num 4 last_change 663 flags hashpspool stripe_width 0 pool 100 '.rgw.root' replicated size 3 min_size 2 crush_rule 0 object_hash rjenkins pg_num 4 pgp_num 4 last_change 651 flags hashpspool stripe_width 0 pool 101 'default.rgw.reshard' replicated size 3 min_size 2 crush_rule 0 object_hash rjenkins pg_num 8 pgp_num 8 last_change 1529 owner 18446744073709551615 flags hashpspool stripe_width 0 pool 103 'default.rgw.buckets.data' erasure size 3 min_size 2 crush_rule 1 object_hash rjenkins pg_num 256 pgp_num 256 last_change 2106 flags hashpspool stripe_width 8192 I'll keep on googling, but I'm open to advice! Thank you, Roger
_______________________________________________ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
