doc: update MDS upgrade procedure

Now that CompatSet changes to the FSMap no longer cause old MDS to
suicide.

Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>

diff --git a/PendingReleaseNotes b/PendingReleaseNotes
index 627cec6760b5c2dfcb3e413cd94da17a47f33a74..0b1f7623ff76aad3585357a087dc1b1e968882a1 100644 (file)
--- a/PendingReleaseNotes
+++ b/PendingReleaseNotes
@@ -30,6 +30,8 @@
   using these re-authorized IDs may be disrupted, this command requires the
   "--yes-i-really-mean-it" flag. Also, mirroring is expected to be disabled
   on the file system.
+* MDS upgrades no longer require stopping all standby MDS daemons before
+  upgrading the sole active MDS for a file system.
 
 * RGW: `radosgw-admin realm delete` is now renamed to `radosgw-admin realm rm`. This
   is consistent with the help message.

diff --git a/doc/cephfs/upgrading.rst b/doc/cephfs/upgrading.rst
index e9df8c0c74300738ef751caaf2c24d680ff85a46..2dc29e1298bb1fc5c7815f5f09bacb63ab8e51b6 100644 (file)
--- a/doc/cephfs/upgrading.rst
+++ b/doc/cephfs/upgrading.rst
@@ -6,13 +6,11 @@ flags to support seamless upgrades of the MDSs without potentially causing
 assertions or other faults due to incompatible messages or other functional
 differences. For this reason, it's necessary during any cluster upgrade to
 reduce the number of active MDS for a file system to one first so that two
-active MDS do not communicate with different versions.  Further, it's also
-necessary to take standbys offline as any new CompatSet flags will propagate
-via the MDSMap to all MDS and cause older MDS to suicide.
+active MDS do not communicate with different versions.
 
 The proper sequence for upgrading the MDS cluster is:
 
-1. Disable and stop standby-replay daemons.
+1. For each file system, disable and stop standby-replay daemons.
 
 ::
 
@@ -27,7 +25,7 @@ command. Older versions of Ceph require you to stop these daemons manually.
     ceph mds fail mds.<X>
 
 
-2. Reduce the number of ranks to 1:
+2. For each file system, reduce the number of ranks to 1:
 
 ::
 
@@ -39,43 +37,20 @@ command. Older versions of Ceph require you to stop these daemons manually.
 
     ceph status # wait for MDS to finish stopping
 
-4. Take all standbys offline, e.g. using systemctl:
-
-::
-
-    systemctl stop ceph-mds.target
-
-5. Confirm only one MDS is online and is rank 0 for your FS:
-
-::
-
-    ceph status
-
-6. Upgrade the single active MDS, e.g. using systemctl:
-
-::
-
-    # use package manager to update cluster
-    systemctl restart ceph-mds.target
-
-7. Upgrade/start the standby daemons.
+4. For each MDS, upgrade packages and restart. Note: to reduce failovers, it is
+   recommended -- but not strictly necessary -- to first upgrade standby daemons.
 
 ::
 
     # use package manager to update cluster
     systemctl restart ceph-mds.target
 
-8. Restore the previous max_mds for your cluster:
+5. For each file system, restore the previous max_mds and allow_standby_replay settings for your cluster:
 
 ::
 
     ceph fs set <fs_name> max_mds <old_max_mds>
-
-9. Restore setting for ``allow_standby_replay`` (if applicable):
-
-::
-
-    ceph fs set <fs_name> allow_standby_replay true
+    ceph fs set <fs_name> allow_standby_replay <old_allow_standby_replay>
 
 
 Upgrading pre-Firefly file systems past Jewel