In Exchange whether it is Exchange 2010 or 2016 or maybe even 2019, when you setup a Database Availability Group (DAG) requires Failover Cluster Manager to be installed and when the DAG is formed you can see it in Failover Cluster Manager.

In normal circumstances your DAG members will come online after a reboot but you may end up with a single member that just does not want to come online.

This was the case in a customer situation today. No reboot fixed the error. This involved some deeper troubleshooting.

As you can see above, the member shows as down.

To fix this error here are the steps that were performed to bring the node back online and the cluster online:

  • Checked the event log, the server was complaining about Microsoft Failover Cluster Virtual Adapter.

In this case the adapter wasn’t showing as disabled or in an error state. I removed the adapter from Device Manager and rebooted to let windows pick it up again.

After doing this and a reboot, that event was no longer logging, yet the cluster was still down.

  • Check Services.msc

The DHCP Client service was not started on this server, running 2008 R2. Started it and set the Startup Type to Automatic.

Again the server was rebooted and after the reboot the node came back online and the cluster was showing as up.

After the reboot, all the events logged in Cluster Events stopped and the DAG was in a healthy state.

Hope it helps.

Advertisements