After upgrading our Ceph cluster from 17.2.7 to 17.2.8 using `cephadm`, all OSDs are reported as unreachable with the following error:
HEALTH_ERR 32 osds(s) are not reachable [ERR] OSD_UNREACHABLE: 32 osds(s) are not reachable osd.0's public address is not in '172.20.180.1/32,172.20.180.0/24' subnet 1. Has anyone else encountered this issue after upgrading to 17.2.8?
Yes, we ran into it also. It "only" reads the first entry and skips the second (and third and so on)
2. Is this a known regression? (This seems similar to issue #67517.)
This matches my experience better than your case: https://tracker.ceph.com/issues/69175
3. Would upgrading to Ceph 18.x (Reef) resolve the problem? 4. Is there any solution other than muting the health warning?
We widened the netmask to cover both nets so we would have them in one definition.
..and we did not see any .1/32 entries being added, only that ceph would only consider the first network in the list. So sorry, I have no more help to give for your odd situation. -- May the most significant bit of your life be positive.