How to set up Tomcat high availability with PCS

High availability for Tomcat keeps a Java application reachable during node failure by moving the service endpoint to a healthy cluster member.

On a Pacemaker cluster managed with pcs, the endpoint is commonly a floating IP (VIP) provided by the IPaddr2 OCF agent, while a systemd resource starts and monitors the Tomcat service unit. Grouping the resources keeps ordering predictable, so the VIP comes up before the application starts.

A VIP-based design is active/passive, meaning only one node serves traffic at a time. Application binaries and configuration must match across nodes, and session state or writable data should live outside local disks (database, shared storage, or explicit session replication) to avoid losing state during failover.

1. Confirm the cluster is online with quorum.

   $ sudo pcs status
   Cluster name: clustername
   Cluster Summary:
     * Stack: corosync (Pacemaker is running)
     * Current DC: node-01 (version 2.1.6-6fdc9deea29) - partition with quorum
     * 3 nodes configured
     * 0 resource instances configured
   ##### snipped #####

   Related: How to check Pacemaker cluster status

2. Identify the Tomcat service unit name.

   $ systemctl list-unit-files --type=service | grep -E '^tomcat.*\.service'
   tomcat10.service                             enabled         enabled

   Related: How to manage Pacemaker cluster services with systemctl in Linux

3. Disable the Tomcat unit autostart with an immediate stop on every cluster node.

   $ sudo systemctl disable --now tomcat10
   Removed "/etc/systemd/system/multi-user.target.wants/tomcat10.service".
   ##### snipped #####

   Stopping Tomcat outside PCS control interrupts service until the resource group starts on an active node.

   Related: How to manage Pacemaker cluster services with systemctl in Linux

4. Create a floating IP resource for the application endpoint.

   $ sudo pcs resource create tomcat_ip ocf:heartbeat:IPaddr2 ip=192.0.2.61 cidr_netmask=24 op monitor interval=30s

   Use an unused IP on the target subnet to avoid address conflicts and intermittent traffic loss.

   Related: How to create a floating IP address in Pacemaker

5. Create the Tomcat service resource.

   $ sudo pcs resource create tomcat_service systemd:tomcat10 op monitor interval=30s

   Use the systemd unit base name (tomcat10 from tomcat10.service) in the systemd: resource identifier.

6. Group the IP resource with the Tomcat service resource.

   $ sudo pcs resource group add tomcat-stack tomcat_ip tomcat_service

   Pacemaker starts grouped resources in listed order, so the VIP becomes available before the Tomcat unit is started.

   Related: How to create a Pacemaker resource group

7. Verify the resource group placement.

   $ sudo pcs status resources
     * Resource Group: tomcat-stack:
       * tomcat_ip	(ocf:heartbeat:IPaddr2):	 Started node-01
       * tomcat_service	(systemd:tomcat10):	 Started node-01

8. Confirm VIP reachability using an address listing plus an HTTP request to the endpoint.

   $ ip -brief address show | grep -F '192.0.2.61'
   eth0@if456       UP             192.0.2.11/24 192.0.2.61/24
   
   $ curl -I http://192.0.2.61:8080/
   HTTP/1.1 200 
   Accept-Ranges: bytes
   ETag: W/"1905-1767248950263"
   Last-Modified: Thu, 01 Jan 2026 06:29:10 GMT
   Content-Type: text/html
   Content-Length: 1905
   Date: Thu, 01 Jan 2026 06:34:49 GMT

   Set the port to the connector exposed by Tomcat, such as 8080, or the fronting proxy port when the VIP points at a reverse proxy.

9. Run a failover test with the group in a running state.

   Related: How to run a Pacemaker failover test with PCS