First I would like to say that SVM-DR is the greatest thing since slided bread. (Okay, sandwich bread).
When I first read and implemented it, it made life so much easier, because, it automated the entire business continuance and backup and recovery process. (Having to clone to a new RW volume wasn't as much trouble with FlexClones and I've been able to recover data for my users and admin groups perfectly).
Now I've come to the Lifecycle end of the source SVM. Wouldn't you know management wants to, a) end volume lifecycle in a staggered fashion, b) they want to maintain an archive in the destination for the volumes at end-of-life for a year or two.
Here is the issue: SVM-DR does not allow you to take destination volumes out of the cycle. With SVM-DR, it is all or nothing according to Support. That is not very smart; after committing to SVM-DR's promises we find that it is a bear. Support suggests moving the source-volume to a new archival SVM and creating a new volume SnapMirror. Kind of defeats the purpose of having a single point of Backup/DR-BC.
Now, I did find that you can get around that, but no thanks to NetApp. It is a process I discovered after a few weeks of playing around. But I think the Project Manager should be made aware that SVM-DR is positioned as a single point of DR, BC, SnapVault-Backup. It should allow for movement of both source and destination volumes, even if that means converting from an SVM-DR to a Volume-SM policy.
I hope this message reaches the proper ONTAP Protection Managers.