I have done the update of the ceph software several times, but this is the first time I have to do an update (from quincy to reef) considering also ceph-fs I am not using cephadm I have 3 MDSs (2 active and one in standby) I'm looking at point 5 of: https://ceph.io/en/news/blog/2023/v18-2-0-reef-released/ If I have got it right It says to temporarily set max_mds to 1 and when you have only one active: 1 - turn off the MDS service on the 2 inactive MDS nodes 2- update and restart MDS on the active instance 3- restart (after updating the software: this is not written in the doc but it seems obvious to me) MDS on the other 2 instances 4 put max_mds back to 2 Did I understand correctly? But doesn't this mean to have a short period of time when the cluster is in error state (in 2, during the restart) and therefore with possible problems on the cephfs clients ? Thanks, Massimo