How to set maintenance mode in Pacemaker

Putting a Pacemaker cluster into maintenance-mode pauses automated resource management during planned maintenance, preventing unexpected restarts, migrations, and recovery actions while infrastructure changes are in progress.

The maintenance-mode cluster property disables monitoring and start/stop decisions for all resources, leaving their current state unchanged until the property is cleared. Cluster-wide maintenance mode overrides node- and resource-level maintenance settings.

The setting applies to the entire cluster and persists until explicitly disabled, so automatic failover and recovery are effectively suspended for the duration. Clearing maintenance-mode can immediately trigger reconciliation actions (start/stop/recover) so services match the configured desired state.

1. Enable cluster-wide maintenance mode.

   $ sudo pcs property set maintenance-mode=true

   The change is stored in the cluster configuration and only needs to be run once from any node with pcs access.

2. Verify the maintenance-mode property is enabled.

   $ sudo pcs property show --all | grep -E 'maintenance-mode[=:]'
     maintenance-mode=true

   Older pcs output may appear as maintenance-mode=true.

3. Confirm Pacemaker reports resource management as disabled.

   $ sudo pcs status
   Cluster name: clustername
   Cluster Summary:
     * Stack: corosync (Pacemaker is running)
     * Current DC: node-03 (version 2.1.6-6fdc9deea29) - partition with quorum
     * Last updated: Wed Dec 31 09:47:55 2025 on node-01
     * Last change:  Wed Dec 31 09:47:54 2025 by root via cibadmin on node-01
     * 3 nodes configured
     * 7 resource instances configured
   
                 *** Resource management is DISABLED ***
     The cluster will not attempt to start, stop or recover services
   
   ##### snipped #####
   Full List of Resources:
     * Resource Group: web-stack (maintenance):
       * cluster_ip (ocf:heartbeat:IPaddr2): Started node-02 (maintenance)
     * Clone Set: stateful-demo-clone [stateful-demo] (promotable, maintenance):
       * stateful-demo (ocf:pacemaker:Stateful): Promoted node-03 (maintenance)
       * stateful-demo (ocf:pacemaker:Stateful): Unpromoted node-01 (maintenance)
       * Stopped: [ node-02 ]
   ##### snipped #####

4. Disable maintenance mode after planned work is complete.

   $ sudo pcs property set maintenance-mode=false

   Leaving maintenance-mode enabled disables automated recovery and failover until the property is cleared.

5. Verify the maintenance-mode property is disabled.

   $ sudo pcs property show --all | grep -E 'maintenance-mode[=:]'
     maintenance-mode=false

6. Confirm Pacemaker has resumed managing resources.

   $ sudo pcs status
   Cluster name: clustername
   Cluster Summary:
     * Stack: corosync (Pacemaker is running)
     * Current DC: node-03 (version 2.1.6-6fdc9deea29) - partition with quorum
     * Last updated: Wed Dec 31 09:47:56 2025 on node-01
     * Last change:  Wed Dec 31 09:47:55 2025 by root via cibadmin on node-01
     * 3 nodes configured
     * 7 resource instances configured
   
   ##### snipped #####

   If resources were manually stopped during maintenance, clearing maintenance-mode may start them again to match cluster configuration.