Shutting down and restarting a cluster

There may be times when you need to gracefully shut down a cluster. The following procedure walks you through it. You must have SSH access to each node in a cluster in order to perform these steps. If additional assistance is needed, you can contact Scale Computing Support.

Connect to the primary node in the cluster and perform the following steps.
1. Use the SC//Reliant™ Virtual Controller (RVC) CLI to ensure that the cluster isn't in maintenance mode. Verify you see Maintenance.: false in the output.
```
rvc cluster status
```
2. Disable STONITH for the cluster.
```
sudo crm configure property stonith-enabled=false
```
3. Place the node in standby. This gracefully shuts down any containers and virtual machines on the node that are running inside the cluster. Replace <node_1_hostname> in the following command with the hostname of the node you are placing in standby.
```
sudo pcs node standby <node_1_hostname>
```
4. If you have any containers running outside the cluster, use the following command to stop them. Replace <container_name> with the name of your container, ensuring that the full name follows the docker-<container_name> format.
```
sudo systemctl stop docker-<container_name>
```
5. Monitor the resources being shut down.
```
sudo crm_mon
```
Connect to the other node in the cluster and perform the following steps.
1. Place the node in standby. This gracefully shuts down any containers and virtual machines running on the node. Replace <node_2_hostname> with the hostname of the node.
```
sudo pcs node standby <node_2_hostname>
```
2. If you have any containers running outside the cluster, use the following command to stop them. Replace <container_name> with the name of your container, ensuring that the full name follows the docker-<container_name> format.
```
sudo systemctl stop docker-<container_name>
```
3. Monitor the resources being shut down.
```
sudo crm_mon
```

Verify that all resources are shut down on each node.

sudo crm_mon

The output should look similar to this:

Cluster Summary:
 * Stack: corosync
 * Current DC: sit0006red1 (version 2.1.5-a3f44794f94) - partition with quorum
 * Last updated: Fri Jun 27 15:13:37 2025
 * Last change:  Fri Jun 27 15:00:05 2025 by root via cibadmin on sit0006red0
 * 2 nodes configured
 * 23 resource instances configured
Node List:
 * Node sit0006red0: standby
 * Node sit0006red1: standby
Active Resources:
 * No active resources

Once all resources are gracefully shut down, you can power off the nodes.

Bringing a cluster back up

After you have gracefully shut down a cluster, use the following steps when you are ready to bring that cluster back up.

Power on each node.
Connect to one of the nodes in the cluster, usually the one you intend to be the primary node, and perform the following steps.
1. Remove the node from standby.
```
sudo pcs node unstandby <node_1_hostname>
```
2. Verify the node is showing as Active and is no longer in standby.
```
rvc cluster status
```
3. Enable STONITH for the cluster.
```
sudo crm configure property stonith-enabled=true
```
4. Use the SC//Reliant Virtual Controller (RVC) CLI to ensure that the cluster resources come back up.
```
rvc cluster status
rvc docker containers
rvc kvm list
```
Connect to the other node in the cluster and perform the following steps.
1. Remove the node from standby.
```
sudo pcs node unstandby <node_2_hostname>
```

Shutting down and restarting a cluster

Shutting down and restarting a cluster

Bringing a cluster back up

Related documentation