What are the consequences of rebooting a Failover Cluster Node without any roles (passive / Auxiliary nodes)?

Question

I am now running a two-nodes MSCS failover cluster as VMs for my SQL Server.

I will patch/apply Windows Update to these VMs beginning with the passive/auxiliary node that does not run any Roles, followed by a reboot, and then a failover to the other node.

Could someone kindly confirm if the following action is required before rebooting the Passive / Auxiliary nodes after applying the Windows Update?

Process to drain stop and prevent the outage of the SQL Server due to the node reboot:

1. Open Failover Cluster Manager (CluAdmin.msc)
2. Click on “Roles” section, and verify the Owner Node name.
3. Click on “Nodes” section, and select the Node that does not run any roles.
4. Right-click on the node name and under ‘ Pause ’ click on ‘ Drain Roles ’
5. Under Status the node will appear as ‘Paused’.

When the node is once again powered on and ready to be put back into service, use the Resume action to re-enable the node to host roles. 

1. Open Failover Cluster Manager (CluAdmin.msc)
2. Click on “ Nodes ”
3. Right-click on the node name and select ‘ Resume ’, then select:
4. ‘ Do Not Fail Roles Back ’ – This will resume the node and not move any roles back.

What if I simply rebooted the passive/auxiliary node, which does not hold or run any roles?

Any assistance or feedback would be highly appreciated.

Answer 1

Hello there,

Rebooting a failover cluster node that does not currently have any active roles (such as passive or auxiliary nodes) generally has fewer consequences compared to rebooting a node that hosts active roles. However, there are still some important considerations to keep in mind:

Resource Failback: If there are any resources that are set to automatically fail back to the preferred node when it comes back online, the failback might occur when you reboot a passive node. This could impact resource distribution and availability.

Cluster Heartbeating: Rebooting a node affects the cluster's heartbeating and communication. When a node reboots, the cluster might need to recalculate the cluster's quorum and reach a new consensus about the cluster's state.

Cluster Quorum: Rebooting a node can lead to changes in the cluster quorum configuration. Depending on the quorum model being used (e.g., Node Majority, Node and Disk Majority, etc.), a node reboot might cause a shift in the quorum configuration, which could impact the cluster's ability to make decisions about resource ownership and failover.

Failover Potential: While the node you're rebooting might not have any active roles, there is always a potential for resources to fail over to it during the reboot process. This can happen if the cluster determines that the preferred owner is not available.

Resource Availability: If the node you're rebooting is part of a multi-site cluster, the remaining nodes in the other site will need to handle the workload until the node is back online. This might lead to increased resource utilization on those nodes.

Management and Monitoring Impact: During the reboot process, management tools and monitoring systems might temporarily lose connection to the node. This could impact your ability to monitor the node's status or perform administrative tasks.

I used AI provided by ChatGPT to formulate part of this response. I have verified that the information is accurate before sharing it with you.

Hope this resolves your Query !!

--If the reply is helpful, please Upvote and Accept it as an answer–