Hi,
We have had problems with a SQL server cluster. As a result of this we have been advised by our SAN hardware manufacturer to replace the Win 2003 servers with Windows 2000 servers. I have replaced one server, and it will successfully host all of the databases in the cluster except one. We see the following errors in the Application Event Log on the Windows 2000 server when we try to move this virtual server onto the Windows 2000 server:
EventID 17055 Computer FDSQL01 (The Virtual Server Name)
17113 :
initconfig: Error 32(error not found) opening '\\DBDOLLY02\U$\Program Files\Microsoft SQL Server\MSSQL$FDSQL01\data\master.mdf' for configuration information.
EventID 17052 Computer DBDOLLY02 (The Real Server Name)
17113 :
initconfig: Error 32(error not found) opening '\\DBDOLLY02\U$\Program Files\Microsoft SQL Server\MSSQL$FDSQL01\data\master.mdf' for configuration information.
EventID 17052 Computer DBDOLLY02
[sqsrvres] OnlineThread: ResUtilsStartResourceService failed (status 435)
EventID 17052 Computer DBDOLLY02
[sqsrvres] OnlineThread: Error 435 bringing resource online.
We can see the other SAN drives OK, and the physical disk comes online without problems. (If we take the SQL services offline we can move the group to the Windows 2000 server without problems).
There appear to be no configuration differences between the different SQL Service definitions (other than pointing to different installations). Service Pack 3A is installed on both nodes and all SQL virtual servers.
I have been googling all night and not come up with anything. I have tried removing the Database from the node (using the failover maintenance when running the SQL Server install) and re-adding it (several times!!). I have managed to move the other virtul servers and from my notes I have not done anything different with this instance.
Please advise actions that we can take to enable the service to run on the new node, any other information you need just let me know! As this is a customer critical database I cannot move it until all of the customers' users have finished accessing it so I may not be able to try anything until I get a maintance window tonight.
I look forward to any replies.

Cluster Node not running all databases