X-Git-Url: https://gerrit.opnfv.org/gerrit/gitweb?a=blobdiff_plain;f=extraconfig%2Ftasks%2Fmajor_upgrade_controller_pacemaker_3.sh;h=6748f891eb3bdb5c5717ae60a5833bdde8eeb5b4;hb=6998edc64a16e1dd859008caba719baeebf43d37;hp=b653c7c7503340170997b7fd9e794233debe5c23;hpb=a1b7af553de02e79f47af4110955e5fc73c385f3;p=apex-tripleo-heat-templates.git

diff --git a/extraconfig/tasks/major_upgrade_controller_pacemaker_3.sh b/extraconfig/tasks/major_upgrade_controller_pacemaker_3.sh
index b653c7c7..6748f891 100755
--- a/extraconfig/tasks/major_upgrade_controller_pacemaker_3.sh
+++ b/extraconfig/tasks/major_upgrade_controller_pacemaker_3.sh
@@ -2,25 +2,67 @@
 
 set -eu
 
-start_or_enable_service rabbitmq
-check_resource rabbitmq started 600
+cluster_form_timeout=600
+cluster_settle_timeout=1800
+galera_sync_timeout=600
+
+if [[ -n $(is_bootstrap_node) ]]; then
+    pcs cluster start --all
+
+    tstart=$(date +%s)
+    while pcs status 2>&1 | grep -E '(cluster is not currently running)|(OFFLINE:)'; do
+        sleep 5
+        tnow=$(date +%s)
+        if (( tnow-tstart > cluster_form_timeout )) ; then
+            echo_error "ERROR: timed out forming the cluster"
+            exit 1
+        fi
+    done
+
+    if ! timeout -k 10 $cluster_settle_timeout crm_resource --wait; then
+        echo_error "ERROR: timed out waiting for cluster to finish transition"
+        exit 1
+    fi
+
+    for vip in $(pcs resource show | grep ocf::heartbeat:IPaddr2 | grep Stopped | awk '{ print $1 }'); do
+      pcs resource enable $vip
+      check_resource_pacemaker $vip started 60
+    done
+fi
+
+start_or_enable_service galera
+check_resource galera started 600
 start_or_enable_service redis
 check_resource redis started 600
-start_or_enable_service openstack-cinder-volume
-check_resource openstack-cinder-volume started 600
-
+# We need mongod which is now a systemd service up and running before calling
+# ceilometer-dbsync. There is still a race here: mongod might not be up on all nodes
+# so ceilometer-dbsync will fail a couple of times before that. As it retries indefinitely
+# we should be good.
+# Due to LP Bug https://bugs.launchpad.net/tripleo/+bug/1627254 am using systemctl directly atm
+systemctl start mongod
+check_resource mongod started 600
 
-# Swift isn't controled by pacemaker
-systemctl_swift start
+if [[ -n $(is_bootstrap_node) ]]; then
+    tstart=$(date +%s)
+    while ! clustercheck; do
+        sleep 5
+        tnow=$(date +%s)
+        if (( tnow-tstart > galera_sync_timeout )) ; then
+            echo_error "ERROR galera sync timed out"
+            exit 1
+        fi
+    done
 
-# We need to start the systemd services we explicitely stopped at step _1.sh
-# FIXME: Should we let puppet during the convergence step do the service enabling or
-# should we add it here?
-services=$(services_to_migrate)
-if [[ ${keep_sahara_services_on_upgrade} =~ [Ff]alse ]] ; then
-    services=${services%%openstack-sahara*}
+    # Run all the db syncs
+    # TODO: check if this can be triggered in puppet and removed from here
+    ceilometer-upgrade --config-file=/etc/ceilometer/ceilometer.conf --skip-gnocchi-resource-types
+    cinder-manage db sync
+    glance-manage --config-file=/etc/glance/glance-registry.conf db_sync
+    heat-manage --config-file /etc/heat/heat.conf db_sync
+    keystone-manage db_sync
+    neutron-db-manage --config-file /etc/neutron/neutron.conf --config-file /etc/neutron/plugin.ini upgrade head
+    nova-manage db sync
+    nova-manage api_db sync
+    nova-manage db online_data_migrations
+    sahara-db-manage --config-file /etc/sahara/sahara.conf upgrade head
 fi
-for service in $services; do
-    manage_systemd_service start "${service%%-clone}"
-    check_resource_systemd "${service%%-clone}" started 600
-done