We've had some issues using the same SVCC for managing both the protected and recovery sites which went away when we deployed two SVCCs - one for managing each site.

Just a thought

Cormac

We also just tried a full run thinking maybe it is just haning up on the flashcopy process but the full failover fails as well.

As I put in a previous posting, this exact same error occurs if you use a single SVC Console to manage both the protected and recovery site.

To address, you need to deploy an SVC Console to manage just a single site.

However, the problem is that this failure does not clean up after itself on the recovery site SVC cluster, and leaves the FlashCopy instance used for the TestFailover hanging around.

So you will have to login to the recovery site SVC cluster, and run the following set of commands:

1. Check if there is there a flashcopy instance currently?
svcinfo lsvdisk
svcinfo lsfcmap

1. Unpresent the flashcopy VDisk from the recovery site ESX (this might have been done by the SRA):
svctask rmvdiskhostmap

2. Stop the flashcopy between the replicated VDisk and the FlashCopy VDisk:
svctask stopfcmap

3. Remove the FlashCopy map:
svctask rmfcmap

4. Remove the FlashCopy VDisk:
svctask rmvdisk

After running these commands, and setting up the SVC consoles to manage just a single cluster instance each, I could successfully run a TestFailover on my SVC environment using the same version of SRA that you are using (Version 1.0.1-2 | Released 04/27/2009).

HTH

Cormac