Starwind in a SAN Cluster, failover failing

[DarranC]

I'm having a problem with failing Starwind service over between nodes.

We have the starwind service running in a cluster between two file servers in it's own resource group.
The Starport service is running on two print server nodes also clustered. The Data and Quorum drives are stored on iSCSI disks.

When we 'move group' on the file server to move the Starwind service the active print server iSCSI disks go into a 'reconnecting' state and stop there until I manually restart the Starwind service. On occasion I also have to restart the Starport service and cluster service on the print server.
The issue happens on both print server nodes independent of which file server is active.

The Starwind service is stopping on the file server when the resource is moved away so the service has already restarted once when it is moved back. Is there any reason a second restart of the service is needed? I have made the IP Address, Network name and Disk dependencies of the Starwind service in case of any startup issues.

Is clustering supported for the Starwind service?

Any help would be gratefully appreciated.
Darran

[anton (staff)]

Yes, of course StarWind is cluster-capable. You need to replace StarPort with MS iSCSI initiator however as it's not... Also please make sure you do run the most recent version of our software as there could be some issues with the early ones (cluster environment related). If you'll still have problems please grab logs from both StarWind nodes, zip and either send them to support@starwindsoftware.com with your comments and reference to this thread or just post them as an attachment here (e-mail preferred as logs could be quite large). Thanks!

I'm having a problem with failing Starwind service over between nodes.

We have the starwind service running in a cluster between two file servers in it's own resource group.
The Starport service is running on two print server nodes also clustered. The Data and Quorum drives are stored on iSCSI disks.

When we 'move group' on the file server to move the Starwind service the active print server iSCSI disks go into a 'reconnecting' state and stop there until I manually restart the Starwind service. On occasion I also have to restart the Starport service and cluster service on the print server.
The issue happens on both print server nodes independent of which file server is active.

The Starwind service is stopping on the file server when the resource is moved away so the service has already restarted once when it is moved back. Is there any reason a second restart of the service is needed? I have made the IP Address, Network name and Disk dependencies of the Starwind service in case of any startup issues.

Is clustering supported for the Starwind service?

Any help would be gratefully appreciated.
Darran