update_identifier=${update_identifier//[^a-zA-Z0-9-_]/}
# seconds to wait for this node to rejoin the cluster after update
-cluster_start_timeout=360
+cluster_start_timeout=600
+galera_sync_timeout=360
+cluster_settle_timeout=1800
timestamp_file="$timestamp_dir/$update_identifier"
if [[ -a "$timestamp_file" ]]; then
fi
pacemaker_status=$(systemctl is-active pacemaker)
+pacemaker_dumpfile=$(mktemp)
if [[ "$pacemaker_status" == "active" ]] ; then
+SERVICES="memcached
+httpd
+neutron-dhcp-agent
+neutron-l3-agent
+neutron-metadata-agent
+neutron-openvswitch-agent
+neutron-server
+openstack-ceilometer-alarm-evaluator
+openstack-ceilometer-alarm-notifier
+openstack-ceilometer-api
+openstack-ceilometer-central
+openstack-ceilometer-collector
+openstack-ceilometer-notification
+openstack-cinder-api
+openstack-cinder-scheduler
+openstack-cinder-volume
+openstack-glance-api
+openstack-glance-registry
+openstack-heat-api
+openstack-heat-api-cfn
+openstack-heat-api-cloudwatch
+openstack-heat-engine
+openstack-keystone
+openstack-nova-api
+openstack-nova-conductor
+openstack-nova-consoleauth
+openstack-nova-novncproxy
+openstack-nova-scheduler"
+
+ echo "Dumping Pacemaker config"
+ pcs cluster cib $pacemaker_dumpfile
+
+ echo "Checking for missing constraints"
+
+ if ! pcs constraint order show | grep "start openstack-nova-novncproxy-clone then start openstack-nova-api-clone"; then
+ pcs -f $pacemaker_dumpfile constraint order start openstack-nova-novncproxy-clone then openstack-nova-api-clone
+ fi
+
+ if ! pcs constraint order show | grep "start rabbitmq-clone then start openstack-keystone-clone"; then
+ pcs -f $pacemaker_dumpfile constraint order start rabbitmq-clone then openstack-keystone-clone
+ fi
+
+ if ! pcs constraint order show | grep "promote galera-master then start openstack-keystone-clone"; then
+ pcs -f $pacemaker_dumpfile constraint order promote galera-master then openstack-keystone-clone
+ fi
+
+ if pcs resource | grep "haproxy-clone"; then
+ SERVICES="$SERVICES haproxy"
+ if ! pcs constraint order show | grep "start haproxy-clone then start openstack-keystone-clone"; then
+ pcs -f $pacemaker_dumpfile constraint order start haproxy-clone then openstack-keystone-clone
+ fi
+ fi
+
+ if ! pcs constraint order show | grep "start memcached-clone then start openstack-keystone-clone"; then
+ pcs -f $pacemaker_dumpfile constraint order start memcached-clone then openstack-keystone-clone
+ fi
+
+ if ! pcs constraint order show | grep "promote redis-master then start openstack-ceilometer-central-clone"; then
+ pcs -f $pacemaker_dumpfile constraint order promote redis-master then start openstack-ceilometer-central-clone require-all=false
+ fi
+
+ # ensure neutron constraints https://review.openstack.org/#/c/229466
+ # remove ovs-cleanup after server and add openvswitch-agent instead
+ if pcs constraint order show | grep "start neutron-server-clone then start neutron-ovs-cleanup-clone"; then
+ pcs -f $pacemaker_dumpfile constraint remove order-neutron-server-clone-neutron-ovs-cleanup-clone-mandatory
+ fi
+ if ! pcs constraint order show | grep "start neutron-server-clone then start neutron-openvswitch-agent-clone"; then
+ pcs -f $pacemaker_dumpfile constraint order start neutron-server-clone then neutron-openvswitch-agent-clone
+ fi
+
+
+ if ! pcs resource defaults | grep "resource-stickiness: INFINITY"; then
+ pcs -f $pacemaker_dumpfile resource defaults resource-stickiness=INFINITY
+ fi
+
+ echo "Setting resource start/stop timeouts"
+ for service in $SERVICES; do
+ pcs -f $pacemaker_dumpfile resource update $service op start timeout=200s op stop timeout=200s
+ done
+ # 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 "Applying new Pacemaker config"
+ if ! pcs cluster cib-push $pacemaker_dumpfile; then
+ echo "ERROR failed to apply new pacemaker config"
+ exit 1
+ fi
+
echo "Pacemaker running, stopping cluster node and doing full package update"
- pcs cluster stop
+ node_count=$(pcs status xml | grep -o "<nodes_configured.*/>" | grep -o 'number="[0-9]*"' | grep -o "[0-9]*")
+ if [[ "$node_count" == "1" ]] ; then
+ echo "Active node count is 1, stopping node with --force"
+ pcs cluster stop --force
+ else
+ pcs cluster stop
+ fi
+
+ # clean leftover keepalived and radvd instances from neutron
+ # (can be removed when we remove neutron-netns-cleanup from cluster services)
+ # see https://review.gerrithub.io/#/c/248931/1/neutron-netns-cleanup.init
+ killall neutron-keepalived-state-change 2>/dev/null || :
+ kill $(ps ax | grep -e "keepalived.*\.pid-vrrp" | awk '{print $1}') 2>/dev/null || :
+ kill $(ps ax | grep -e "radvd.*\.pid\.radvd" | awk '{print $1}') 2>/dev/null || :
else
echo "Excluding upgrading packages that are handled by config management tooling"
command_arguments="$command_arguments --skip-broken"
exit 1
fi
done
+
+ tstart=$(date +%s)
+ while ! clustercheck; do
+ sleep 5
+ tnow=$(date +%s)
+ if (( tnow-tstart > galera_sync_timeout )) ; then
+ echo "ERROR galera sync timed out"
+ exit 1
+ fi
+ done
+
+ echo "Waiting for pacemaker cluster to settle"
+ if ! timeout -k 10 $cluster_settle_timeout crm_resource --wait; then
+ echo "ERROR timed out while waiting for the cluster to settle"
+ exit 1
+ fi
+
pcs status
else
Code Review / apex-tripleo-heat-templates.git / blobdiff