From e316840a7e1d2a72e3089ee194334244c959905a Mon Sep 17 00:00:00 2001 From: Ken Gaillot Date: Wed, 19 Jan 2022 09:53:53 -0600 Subject: [PATCH] Fix: pacemakerd: tweak systemd unit respawn settings If pacemaker exits immediately after starting, wait 1 second before trying to respawn, since the default of 100ms is a bit aggressive for a Pacemaker cluster. Also, allow 5 attempts in 25 seconds before giving up. --- daemons/pacemakerd/pacemaker.service.in | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/daemons/pacemakerd/pacemaker.service.in b/daemons/pacemakerd/pacemaker.service.in index 0363a2259c..3fd53d9ffb 100644 --- a/daemons/pacemakerd/pacemaker.service.in +++ b/daemons/pacemakerd/pacemaker.service.in @@ -31,6 +31,9 @@ After=rsyslog.service After=corosync.service Requires=corosync.service +# If Pacemaker respawns repeatedly, give up after this many tries in this time +StartLimitBurst=5 +StartLimitIntervalSec=25s [Install] WantedBy=multi-user.target @@ -57,6 +60,9 @@ TasksMax=infinity # resource. Sending -KILL will just get the node fenced SendSIGKILL=no +# Systemd's default of respawning a failed service after 100ms is too aggressive +RestartSec=1s + # If we ever hit the StartLimitInterval/StartLimitBurst limit, and the # admin wants to stop the cluster while pacemakerd is not running, it # might be a good idea to enable the ExecStopPost directive below. -- 2.27.0