When the worst happens it hits hard. Restoring a server is a lengthy process with users and managers alike breathing down your neck, so we try to help as much as possible where we can. The best advise I can give you is do this in a test environment a couple of times before you are forced to do it in a live environment! That way you will feel more comfortable and will be able to judge progress better than if you are still trying to figure out what you should do next….

Restoring a cluster consists of a couple of points that we’ll discuss in further detail later on:

  • Failover
  • Eviction if necessary
  • Preparation
  • joining the cluster
  • Installing exchange
  • Moving cluster nodes

Failover

When a node fails failover is needed to keep the resources available to the users on our cluster. Without this our cluster would be fairly pointless.

Eviction

If we need to evict the node that failed, f.e. we know it failed completely, we can evict the node using the windows cluster administrator.

Use due diligence where possible:

  1. Click Start, Administrative Tools, and then click Cluster Administrator.
  2. Stop Cluster Service running.
  3. Select the Evict Node command from the File menu item.
  4. Remove the node from the shared bus.

Preparation

This cover the process of setting up the new node to be joined in to the cluster! Installing windows, bringing it to the same patch level as the other nodes in the cluster. naming it as well as giving it an IP.

Joining the cluster

This is the process of joining your new node in to the exchange cluster

Installing exchange

Well, this is pretty straightforward, just remember  to not use the DR switch as it won’t work

Moving cluster nodes

Simply put, move the resources to the new node brought in to replace our failed node.

 

As you can see, it is pretty straightforward to replace a failed node. Other than having to evict the failed node it is consistent with bringing in an extra node to our cluster!