I have found that if I set norecovery then I can get almost all OSDs to come up
and stay up, but the moment I unset norecovery so that the cluster can heal
itself, tons of OSDs go offline again.
The OSD host servers have plenty of available RAM, they are not maxing on CPU
or I/O near as I can t
On 8/4/22 09:17, Torkil Svensgaard wrote:
Hi
We have a lot of OSDs flapping during recovery and eventually they don't
come up again until kicked with "ceph orch daemon restart osd.x".
This is the end of the log for one OSD going down for good:
"
2022-08-04T09:57:31.752+ 7f3812cb2700 1
Hi Torkil,
it looks like you're facing pretty well known problem with RocksDB
performance degradation caused by bulk data removal. This has been
discussed multiple times in this mailing list.
And here is one of the relevant tracker:
https://tracker.ceph.com/issues/40741
To eliminate the ef
Hi Cassiano,
the backtrace you've provided relates to the bug fixed by:
https://github.com/ceph/ceph/pull/37793
This fix is going to be releases with the upcoming v14.2.17.
But I doubt that your original crashes have the same root cause - this
issue appears during shutdown only.
Anyway yo
Hi, really this error was only showing up when I've tried to run
ceph-bluestore-tool repair, In my 3 OSDs that keeps crashing, it show the
following log... please let me know if there is something I can do to get
the pool back to a functioning state.
Uptime(secs): 0.0 total, 0.0 interval
Flush(GB)
Thanks again Igor, using the ceph-bluestore-tool with the
CEPH_ARGS="--bluestore_hybrid_alloc_mem_cap=0" I was able to detect two
OSDs returning IO errors.
This OSDs crashing has caused backfills operations that triggered some OSDs
marking others as down due to some kind of slow operations and the