Hello list,
We have a Ceph cluster (v17.2.6 quincy) with 3 admin nodes and 6 storage nodes,
each connected to a JBOD enclosure. Each enclosure houses 28 HDD disks of size
18 TB, totaling 168 OSDs. The pool that houses the majority of the data is
erasure-coded (4+2). We have recently had one disk failure, which brought one
OSD down:
# ceph osd tree | grep down
2 hdd 16.49579 osd.2 down 0 1.00000
This OSD is out of the cluster, but we haven't replaced it physically yet. The
problem that we are facing is that the cluster was not in the best shape when
this OSD failed. Currently we have the following:
################
cluster:
id: <redacted>
health: HEALTH_ERR
1026 scrub errors
Possible data damage: 18 pgs inconsistent
2122 pgs not deep-scrubbed in time
2122 pgs not scrubbed in time
services:
mon: 5 daemons, quorum xyz-admin1,xyz-admin2,xyz-osd1,xyz-osd2,xyz-osd3
(age 17M)
mgr: xyz-admin2.sipadf(active, since 17M), standbys: xyz-admin1.nwaovh
mds: 2/2 daemons up, 2 standby
osd: 168 osds: 167 up (since 40h), 167 in (since 6w); 226 remapped pgs
data:
volumes: 2/2 healthy
pools: 9 pools, 2122 pgs
objects: 448.54M objects, 1.0 PiB
usage: 1.6 PiB used, 1.1 PiB / 2.7 PiB avail
pgs: 133905796/2676514497 objects misplaced (5.003%)
1880 active+clean
201 active+remapped+backfilling
23 active+remapped+backfill_wait
16 active+clean+inconsistent
1 active+remapped+inconsistent+backfill_wait
1 active+remapped+inconsistent+backfilling
io:
recovery: 703 MiB/s, 281 objects/s
progress:
Global Recovery Event (6w)
[=========================...] (remaining: 5d)
################
I have noticed the number of active+clean increasing, and objects misplaced
very slowly decreasing. My question is, should I wait until recovery is
complete, repair the 18 damaged pg, and then replace the disk? My thinking is
that replacing the disk will trigger more backfilling which will slow down the
recovering even more.
Another question, should I disable scrubbing while the recovery is not
finalized?
Thank you for any insights you may be able to provide!
-
Gustavo
_______________________________________________
ceph-users mailing list -- [email protected]
To unsubscribe send an email to [email protected]
