Migration Manager for Active Directory is capable of re-permissioning a Microsoft Windows 2000-based server cluster or a Windows Server 2003-based server cluster, however, it requires a more involved procedure than what is required by non-clustered servers. This article describes detailed steps for a successful cluster migration.
PRE-WINDOWS 2008 CLUSTERS ONLY
Microsoft does not support moving an already installed and configured Windows Server 2008 or Windows Server 2008 R2 failover cluster from one domain to another.
PRE-WINDOWS 2008 CLUSTERS ONLY
Microsoft does not support moving an already installed and configured Windows Server 2008 or Windows Server 2008 R2 failover cluster from one domain to another.
The correct procedure involves a three step process:
1. Processing physical nodes with Resource Updating Manager.
2. Processing virtual servers with VMOVER.EXE (remotely).
3. Joining physical nodes to target domain.
There are two variations on the steps that can be taken. See below for the two options:
Option 1:
1. Process each node using the Resource Updating Manager (update resources); only the actual nodes and not the virtual servers.
2. In Migration Manager Select Tools ->Export to-> INI File for Resource Updating on the File menu and specify the required options.
3. Run the following command remotely from the console machine against each virtual server (this needs to be performed directly from the folder where vmover.exe and vmover.ini files reside):
Vmover.exe /c /system=/ini=vmover.ini
OR
Process the virtual server name remotely by adding it to a RUM group that is configured with "Process Computers Remotely" GUI option.
Note: When processing is initiated from command line, it will do it remotely without deploying an agent (also see NAS processing) unlike in the scenario where resource updating is started regular way via GUI. This is in particular important when resource updating is executed against the virtual server (cluster). Such remote processing was also implemented in the RUM console itself via an option "Process computers Remotely" and now can also be utilized as GUI alternative to executing via command line.
4. Move the nodes using the Resource Updating Manager to the target domain (without rebooting).
5. Reboot the Passive Node.
6. Restart the Cluster service on the Active Node.
NOTE: During the restart of the service the resources will not be available.
7. After a successful start of the Cluster service on the active node, start the cluster service on the passive node.
8. Move the resources to the passive node and reboot the active node.
9. After the node restarts move the resources back.
Option 2:
Follow steps 1 - 4 above and instead of taking the last 5-9 steps, reboot both nodes at the same time or select the move with the reboot option in step 4.
In either case the resources will be unavailable at a certain time, because the cluster service cannot run using two accounts (source/target). Both of the nodes should be running using the same account (either source or target )
Note in the article it states: "The Cluster service on all nodes must be stopped and restarted during this procedure. The Cluster service must use the same account and password at all times on all nodes within the cluster".
© 2024 Quest Software Inc. ALL RIGHTS RESERVED. Terms of Use Privacy Cookie Preference Center