Delete Volumes in SVM DR
Delete Volumes in SVM DR
2018-01-25 10:21 AM
Two FAS 8020, NetApp Release 9.1P6 with SVM DR configured. When a volume gets created on Site A SVM it then gets automatically replicated to Site B SVM. Site B SVM is not running until failover. A few other things happend in a WFA workflow but that is basically what is going on.
I need to delete a few volumes but cannot because they are in a SVM DR relationship. When I attempt quiese, break or release a volume I get the message, "Error: command failed: Cannot complete the operation because the destination volume "VOLUMES" belongs to Vserver "SVM" which is configured as the destination endpoint of Vserver DR relationship."
I can exclude a volume from SVM DR with this command,
source_cluster::> volume modify -vserver vs1 -volume test_vol1 -vserver-dr-protection unprotected.
This will remove the relationship from System Manager but if you run list destination from the command line it still appears that relics of relationship are still present. Interesting enough, I can delete the destination volume but not the source volume. It it gives a similar message to the one above. Here is the message.
Data ONTAP API Failed: Volume "VolumeName" in Vserver "SVM" is the source endpoint of one or more SnapMirror relationships. Before you delete the volume, you must release the source information of the SnapMirror relationships using "snapmirror release". To display the destinations to be used in the "snapmirror release" commands, use the "snapmirror list-destinations -source-vserver SVM -source-volume Volume" command.
I have attempted the suggested commands and that is where I find the relics of the relationship even though I have excluded it from protection. I also can not doing anything on volume and will received the message above.
Support suggest I break the entire SVM DR relationship, then I will be able to interact with the volumes and eventually delete them. This seems excessive. Plus, while I am deleting the volumes, my other volumes are syncing on their scheduled interval.
The next procedure worked for me:
1) Run the next command on the source SVM:
source_cluster::> volume modify -vserver vs1 -volume test_vol1 -vserver-dr-protection unprotected
2) Wait for at least 15 mins, after that the snapmirror relationship should not exist anymore
3) No need to run the release command, proceed to delete the snapmirror destination volume
4) Unmount the source volume
5) Offline the source volume
6) Delete the source volume (including all its data)