Three lovely ST8000NM0075 Seagates drives supplied by Dell, all suddenly failed in a RaidZ3, that's now four within 3 years and 4 months.
I didn't anticipate this and only have one spare to plug in. Feels dangerous.
I would like to change the pool to read-only while the semi-offiste differential backup completes.
My reasoning is that the pool should be more stable if it's static, and I disable autosnapshots temporarily.
My concern is that resilvering will take days, and the multiuser system is always busy.
Is this reasonable or justifiable? Does it in fact hedge a little bit of caution?
The downside of not going read-only is yet another discontinuity of business within the same month that falls on me. I'm not sure if the impressions will be reparable, but that's fine if it's the wise decision to go read-only and repair the pool...after the backup.
Would taking the pool completely offline be even more safe that leaving it read-only?
$ dev/disk# zpool status -v
pool: darkpool
state: DEGRADED
status: One or more devices could not be used because the label is missing or
invalid. Sufficient replicas exist for the pool to continue
functioning in a degraded state.
action: Replace the device using 'zpool replace'.
see: http://zfsonlinux.org/msg/ZFS-8000-4J
scan: scrub in progress since Fri Nov 8 04:52:09 2019
1004G scanned out of 47.5T at 81.4M/s, 166h22m to go
0B repaired, 2.06% done
config:
NAME STATE READ WRITE CKSUM
darkpool DEGRADED 0 0 0
raidz3-0 DEGRADED 0 0 0
wwn-0x5000c5008581aafb ONLINE 0 0 0
wwn-0x5000c5008581b61b ONLINE 0 0 0
783034318520267027 FAULTED 0 0 0 was /dev/sdm1
7369503050985789936 FAULTED 0 0 0 was /dev/sdj1
wwn-0x5000c5008581b953 ONLINE 0 0 0
wwn-0x5000c5008581bdf7 ONLINE 0 0 0
wwn-0x5000c50085825ec7 ONLINE 0 0 0
11744243917579175290 FAULTED 0 0 0 was /dev/sdg1
wwn-0x5000c5008581e423 ONLINE 0 0 0
wwn-0x5000c5008581fd3f ONLINE 0 0 0
wwn-0x5000c50085820b93 ONLINE 0 0 0
wwn-0x5000c500858211b3 ONLINE 0 0 0
wwn-0x5000cca267ab0de4 ONLINE 0 0 0
spare-13 DEGRADED 0 0 0
11992420879588183985 FAULTED 0 0 0 was /dev/disk/by-path/pci-0000:03:00.0-scsi-0:0:10:0-part1
wwn-0x5000c500858252ef ONLINE 0 0 0
spares
wwn-0x5000c500858252ef INUSE currently in use
I have one more unboxed drive, not to be confused with the spare that's currently in use replacing another failed disk.
No, I was not involved in the planning of this setup.
