From: Michele Baldessari <michele@acksyn.org>
Date: Tue, 23 Feb 2016 07:42:55 +0000 (+0100)
Subject: Add meta notify=true to rabbitmq resource
X-Git-Tag: opnfv-6.0.0~2161^2
X-Git-Url: https://gerrit.opnfv.org/gerrit/gitweb?a=commitdiff_plain;h=a70db5a907ad9d310a5ec890c159c92ee2122208;p=apex-tripleo-heat-templates.git

Add meta notify=true to rabbitmq resource

See RHBZ 1311005 and 1247303. In short: sometimes when a controller
node gets fenced, rabbitmq is unable to rejoin the cluster. To fix this
we need two steps:
1) The fix for the RA in BZ 1247303
2) Add notify=true to the meta parameters of the rabbitmq resource on
   fresh installs and updates

Note that if this change is applied on systems that do not
have the fix for the rabbitmq resource agent, no action is taken.
So when the resource agent will be updated, the notify
operation will start to work as soon as the first monitor
action will take place.

Fixes RH Bug #1311005

Change-Id: I513daf6d45e1a13d43d3c404cfd6e49d64e51d5a
---

diff --git a/extraconfig/tasks/yum_update.sh b/extraconfig/tasks/yum_update.sh
index 869b1a42..59e4be45 100755
--- a/extraconfig/tasks/yum_update.sh
+++ b/extraconfig/tasks/yum_update.sh
@@ -128,6 +128,9 @@ openstack-nova-scheduler"
     # mongod start timeout is higher, setting only stop timeout
     pcs -f $pacemaker_dumpfile resource update mongod op start timeout=370s op  stop timeout=200s
 
+    echo "Making sure rabbitmq has the notify=true meta parameter"
+    pcs -f $pacemaker_dumpfile resource update rabbitmq meta notify=true
+
     echo "Applying new Pacemaker config"
     if ! pcs cluster cib-push $pacemaker_dumpfile; then
         echo "ERROR failed to apply new pacemaker config"
diff --git a/puppet/manifests/overcloud_controller_pacemaker.pp b/puppet/manifests/overcloud_controller_pacemaker.pp
index b57bcffb..471e1b1c 100644
--- a/puppet/manifests/overcloud_controller_pacemaker.pp
+++ b/puppet/manifests/overcloud_controller_pacemaker.pp
@@ -356,6 +356,7 @@ if hiera('step') >= 2 {
       ocf_agent_name  => 'heartbeat:rabbitmq-cluster',
       resource_params => 'set_policy=\'ha-all ^(?!amq\.).* {"ha-mode":"all"}\'',
       clone_params    => 'ordered=true interleave=true',
+      meta_params     => 'notify=true',
       require         => Class['::rabbitmq'],
     }