Why should I use Exchange Server maintenance mode?

Many admins just patch and reboot servers without failing over the Exchange databases and putting them into maintenance mode. But skipping this practice can create unnecessary headaches.

What is maintenance mode? Essentially, it's a script that drains the node from the cluster and suspends the server's activities so the database availability group (DAG) does not fail over. In Exchange 2016, you have a new setting that moves the database back to its preferred node, and the activation preference -- the default is one hour -- is set after a period of time.

There are times it takes longer than one hour to patch Exchange Server. After applying the update, the server needs to be rebooted, which requires stopping the services, to finish the patching process. In my experience, it can take about an hour to shut down the cluster service on a DAG node.

If you don't give the node enough time to complete its patching process, it can be disastrous. If a security update is underway but you didn't turn on maintenance mode, the DAG failover can break and get stuck so the database is unable to function. If you have two Exchange Servers in a DAG, you can end up with users unable to connect to the messaging platform if both servers are down.

Maintenance mode adds a few more minutes to the update process because it needs to shut down and pause activities on the Exchange Server, but it is recommended even when doing normal maintenance, such as server restarts.

Be sure to take the server out of maintenance mode when it is back online so you can fail over the other node that needs patching. Wait for your DAG logs to replay and show that everything is healthy before moving onto the next server.