The cluster became clean and then an osdmap update caused the PG to peer
again. This triggered the assertion failing the test. Minor RADOS
hiccups should not cause this to fail.
Fixes: https://tracker.ceph.com/issues/75201
Signed-off-by: Patrick Donnelly <pdonnell@ibm.com>
# the manager should be there
manager = ctx.managers[cluster_name]
manager.wait_for_clean()
- assert manager.is_clean()
mds_cluster = MDSCluster(ctx)
for fs in mds_cluster.status().get_filesystems():