rolling_update: clarify "serial" usage

Prior to this commit the serial variable was poorly documented. Now we
are making clear that this value should be left untouched as the rolling
update mechanism should happen serially.

Solves: bz-1396742
Signed-off-by: Sébastien Han <seb@redhat.com>

diff --git a/infrastructure-playbooks/rolling_update.yml b/infrastructure-playbooks/rolling_update.yml
index 3a4b39f6e62fb4468208315622e112ce815cc7ee..9b74f7048128654235bcf4940e7ec4e46d4c1603 100644 (file)
--- a/infrastructure-playbooks/rolling_update.yml
+++ b/infrastructure-playbooks/rolling_update.yml
@@ -1,6 +1,10 @@
 ---
 # This playbook does a rolling update for all the Ceph services
-# Change the value of 'serial:' to adjust the number of server to be updated.
+#
+# The value of 'serial:' adjusts the number of servers to be updated simultaneously.
+# We recommend a value of 1, which means hosts of a group (e.g: monitor) will be
+# upgraded one by one. It is really crucial for the update process to happen
+# in a serialized fashion. DO NOT CHANGE THIS VALUE.
 #
 # The four roles that apply to the ceph hosts will be applied: ceph-common,
 # ceph-mon, ceph-osd and ceph-mds. So any changes to configuration, package updates, etc,