From 3cb9ed7e20f59eec036908eed4cfdbc61e990ee2 Mon Sep 17 00:00:00 2001
From: "Kaleb S. KEITHLEY" <[email protected]>
Date: Tue, 11 Dec 2018 10:09:42 -0500
Subject: [PATCH 081/124] ganesha-ha: ensure pacemaker is enabled after setup

There appears to be a race between `pcs cluster setup ...` early
in the setup and the `systemctl enable pacemaker` at the end. The
`pcs cluster setup ...` disables pacemaker and corosync. (Now, in
pacemaker-1.1.18. Was it always the case?)

I am not able to reproduce this on my devel system. I speculate that
on a busy system that the `pcs cluster setup ...` disable may, under
the right conditions, not run until after the setup script enables
it. It must require the right alignment of the Sun, Moon, and all
the planets.

Regardless, we'll use the --enable option to `pcs cluster setup ...`
to ensure that the cluster (re)starts pacemaker.

Label: DOWNSTREAM ONLY

Change-Id: I771ff62c37426438b80e61651a8b4ecaf2d549c3
Signed-off-by: Jiffin Tony Thottan <[email protected]>
Reviewed-on: https://code.engineering.redhat.com/gerrit/167166
Reviewed-by: Soumya Koduri <[email protected]>
Tested-by: RHGS Build Bot <[email protected]>
Reviewed-by: Sunil Kumar Heggodu Gopala Acharya <[email protected]>
---
 extras/ganesha/scripts/ganesha-ha.sh | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/extras/ganesha/scripts/ganesha-ha.sh b/extras/ganesha/scripts/ganesha-ha.sh
index 4a98f32..32af1ca 100644
--- a/extras/ganesha/scripts/ganesha-ha.sh
+++ b/extras/ganesha/scripts/ganesha-ha.sh
@@ -195,9 +195,9 @@ setup_cluster()
 
     pcs cluster auth ${servers}
     # pcs cluster setup --name ${name} ${servers}
-    pcs cluster setup ${RHEL6_PCS_CNAME_OPTION} ${name} --transport udpu ${servers}
+    pcs cluster setup ${RHEL6_PCS_CNAME_OPTION} ${name} --enable --transport udpu ${servers}
     if [ $? -ne 0 ]; then
-        logger "pcs cluster setup ${RHEL6_PCS_CNAME_OPTION} ${name} ${servers} failed"
+        logger "pcs cluster setup ${RHEL6_PCS_CNAME_OPTION} ${name} --enable --transport udpu ${servers} failed"
         #set up failed stop all ganesha process and clean up symlinks in cluster
         stop_ganesha_all "${servers}"
         exit 1;
-- 
1.8.3.1