Not enough SRM Datastore groups

[brianroehm]

I have an issue with SRM where I am trying to create protection groups for individual servers but all of my servers are getting lumped into two datasource groups.

Here's a Screenshot : ,

The two drives that I have red arrows pointed at are test volumes. They are both directly mapped to the ESX servers and have no other associations to any other virtual machine or volumes. ONe is VMFS and the other is to test SRM + RDM, not currently mapped. My understanding and what I've seen in the past is that these volumes should be in their own datastore group unless they have something shared with the other VM's. I checked swapfiles but they are set to stay with the virtual machine on its datastore. Any ideas?

[notorious_bdg]

Can you verify that your storage replication shows those LUNs in different consistency group?

[brianroehm]

Not sure what you mean by that? Supposedly the SRA just hands a list of replicated LUNs and then VMWare groups them together into those datastore groups. I can verify that all the luns are properly replicated.

[notorious_bdg]

A datastore group corresponds to a replication group or consistency group within your storage replication. You should see a separate datastore group for each consistency group. What storage replication technology are you using?

[brianroehm]

We're using Compellent. They supposedly just pass the info via their SRA of ALL replicated LUNs and then VMWare groups them into datasource groups. So, they would pass thin for a virtual machine with 2 LUNS as two datasources then let VMWare group them together. i.e. 1 LUN for OS VMDK and 1 LUN as an RDM stored in that VMFS with the VMDK this should come up as one datasource group. Is that incorrect?