We're currently running two groups of EqualLogics (let's call the groups A and B for simplicity), with a PS4100 and four PS6100's in each. When we started, it made sense to have half our live volumes on one group and half on the other, with replication occurring between each group. Now that we've increased our group size and number of volumes, we're looking to move all of the live volumes onto one group and have the other group to hold the replicas, making the failover process a lot more simple.
I'm trying to move all the volumes from group A onto group B, and was hoping to make use of the replicas. All volumes are configured with a replication schedule from Group A to Group B, keeping a failback snapshot. The process I've tried so far is:-
1) Take a volume offline from the Windows server it's connected to
2) Create a final manual replica from group A to Group B so I have the latest data
3) Promote the replica set on Group B to a volume
4) Demote the volume on Group A to a replica set
5) Make the promotion on Group B permanent - during this process I'm keeping the volume name the same as the original, e.g. Live volume is called Test, replica set is called Test.1, then when I make the promotion permanent I call the volume Test
6) Demote the original volume on Group A to a replica set
That's all worked well so far, but I've hit a minor snag. I was hoping that I could use the demoted volume on Group A as a starting point for the new replication from Group B to Group A, but when I configure replication it creates a new replica set "Test.1" and the original volume, now a replica set is still called "Test" and appears to be orphaned.
Is it possible to to link the newly promoted volume to the original volume back on Group A, or am I going to have to just setup a new replication schedule and shift terabytes of data on the first pass for each volume? Or is there a better, more streamlined way of doing this?
Thanks,
Michael