This post is part of the Manual Deployment Openstack HA and Ceph series.
Preparation (Exec on all controller nodes)
1. Install pcs package
apt install pacemaker corosync fence-agents pcs resource-agents -y
2. Change user hacluster password
echo 'hacluster:dama!pcs' | chpasswd
3. Enable pcs service
systemctl enable --now pcsd pacemaker corosync
Initiate PCS Cluster (Exec only on controller-01)
1. Authenticate both the nodes using pcs command
pcs host auth os-controller-01 os-controller-02 os-controller-03
2. Configure cluster
pcs cluster setup os-ha os-controller-01 os-controller-02 os-controller-03 --force
3. Set pcs cluster property
pcs property set pe-warn-series-max=1000 pe-input-series-max=1000 pe-error-series-max=1000 cluster-recheck-interval=5min
4. Create pcs resource vip and haproxy
pcs resource create internal_vip ocf:heartbeat:IPaddr2 ip="10.10.10.100" cidr_netmask="24" op monitor interval="30s"
pcs resource create public_vip ocf:heartbeat:IPaddr2 ip="202.10.10.100" cidr_netmask="24" op monitor interval="30s"
pcs resource create lb-haproxy systemd:haproxy op monitor interval="30s"
5. Define ordering and colocation constraints
pcs constraint order start internal_vip then public_vip
pcs constraint colocation add public_vip with internal_vip INFINITY
pcs constraint colocation add lb-haproxy with internal_vip INFINITY
6. Show cluster status
pcs status
7. Restart haproxy from pcs
pcs resource restart lb-haproxy