Legal Information
PC Knowledge Base -Disaster Recovery for Microsoft Exchange 2000 Server - Preparing a Server to Replace a Failed Node

Good Knowledge Is Good2Use

If one of the nodes of an Exchange 2000 cluster fails, replace that node as soon as possible. The following are three methods you can use to replace a failed node:

For all three methods, the only hardware in a replacement node that must be identical to the hardware in the other nodes of the cluster is the processor type and amount of RAM. The computer (NetBIOS) name of the replacement node can be different than that of the failed node, although the replacement node must belong to the same domain. To prepare a cluster node recovery server
  1. Install Windows 2000 Server, including the latest service pack that the server was running, software updates, and software you run on your Exchange 2000 cluster nodes (such as anti-virus software).
  2. Join it to the same domain as the other nodes in the cluster.
  3. Connect the computer to the shared SCSI bus being used by the cluster.
  4. If the cluster already has the maximum number of nodes, you may need to temporarily evict one of the nodes from the cluster and remove it from the shared SCSI bus so you can attach the stand-by cluster node while you configure it.
    To evict a node from the cluster:
    1. Open Cluster Administrator (click Start, point to Programs, point to Administrative Tools, and then click Cluster Administrator). If Cluster Administrator cannot find the cluster the node belongs to by its name, type the name of an active node in the cluster. You can also connect to the cluster by opening Cluster Administrator on a cluster node and entering a PERIOD (.) in Cluster or server name.
    2. Stop the Cluster service on the node you wish to evict. To stop the Cluster service, click Start, point to Programs, point to Administrative Tools, and then click Services. In Services, double-click Cluster Service, and then click Stop. When the Cluster service has stopped successfully, click OK.
    3. On the File menu, in Cluster Administrator, click Evict Node.
    4. Uninstall the Cluster service from the Add-Remove Windows Components part of Add/Remove Programs in the Windows 2000 Control Panel.
    5. If you use SCSI, make sure that the SCSI bus is terminated before you remove the evicted node from the SCSI bus.
    Start the stand-by cluster node computer, and join the cluster by adding the Cluster service from the Add-Remove Windows Components part of Add/Remove Programs in the Windows 2000 Control Panel. Use the wizard to add the computer as a node to the cluster.
  5. Install Exchange 2000 Server on the stand-by cluster node. Exchange automatically notifies you it is installing the "cluster-aware" version of Exchange 2000. (Applications that support the Cluster API are defined as "cluster-aware.") After Exchange 2000 Setup completes, install any Exchange 2000 Service Packs running on the other nodes in the cluster.
  6. Evict the stand-by node from the cluster by following procedures in Step 4 of this procedure.
  7. Power off the stand-by cluster node, it is now prepared to be joined back to the cluster in the event of a disaster occurring to one of the online nodes after the failed node is removed.
  8. Store this computer in a safe place until it is needed to replace a node in the cluster that fails due to a disaster.
  9. . If you temporarily evicted a node from the cluster to configure the stand-by cluster node, you can now reconnect the node to the shared bus and turn on the computer. Install the cluster software on the node that was temporarily evicted so it can rejoin the cluster. Reconfigure the preferred owners for groups in the cluster, if necessary.

The information in this article applies to:



Search Knowledge Base Feedback
If you like our web site refer a friend.
Your friends name.
Your friends email address.
Your Name
Your Email Address


© Copyright 1998-1999 GOOD2USE