Ceph

I think the problem is that https://github.com/ceph/ceph/pull/38118 has merged in nautilus while https://github.com/ceph/ceph/pull/38345 hasn't merged in octopus. When doing an upgrade test from N->O, the nautilus version adds extra health detail to clog but does not set "mon health detail to clog = false" because the test suite being used is octopus, which still doesn't have that change.

The fact that this warning only appears while the mons are running nautilus proves the above theory.

2020-12-10 21:56:51.662 7f4ec0c3e700  0 log_channel(cluster) log [WRN] : Health detail: HEALTH_WARN 1/3 mons down, quorum c,a
.
.
2020-12-10 21:56:51.806 7f4ec2441700  7 mon.c@1(peon).log v68 update_from_paxos applying incremental log 68 2020-12-10 21:56:51.749969 mon.b (mon.0) 8 : cluster [INF] Health check cleared: MON_DOWN (was: 1/3 mons down, quorum c,a)
.
.
2020-12-10T22:24:33.202+0000 7f776b6b2540  0 ceph version 15.2.7-678-gd0234e1e (d0234e1ede269851a029cd700e16721cae756cc6) octopus (stable), process ceph-mon, pid 16286

https://github.com/ceph/ceph/pull/38118 merged 11 days ago which aligns with "Seems to be introduced after 11/27/20 https://pulpito.ceph.com/?suite=upgrade%3Anautilus-x&branch=octopus"

@Neha ., so this is fixed by https://github.com/ceph/ceph/pull/38345 ?