How infinite volume physically distributed?

[netappmagic]

I created a vserver for the volume infinite volume. I understand that the volume could be located on different nodes in the cluster. However, I don't understand the structure / architecture of how the volume internally created. why two mirrors are created on n02_aggr1, and  1 data  n01_aggr1 and 2 data on n02_aggr1 respectively? how do mirror and data  work together?  Thanks for your inputs.

cluster1::*> vol create -vserver vsinf -volume bigvolume -size 6gb -junction-path /big -state online -policy repos_namespace_export_policy -data-aggr-list n01_aggr1,n02_aggr1

  (volume create)

[Job 36] Creating constituent "bigvolume_ns_mirror0001" on aggregate "n02_aggr1"

[Job 36] Creating constituent "bigvolume_ns_mirror0001" on aggregate "n02_aggr1"

[Job 36] Creating constituent "bigvolume_1024_data0001" on aggregate "n01_aggr1"

[Job 36] Creating constituent "bigvolume_1024_data0002" on aggregate "n02_aggr1"

[Job 36] Creating constituent "bigvolume_1024_data0002" on aggregate "n02_aggr1"

[Job 36] Job succeeded: Created Infinite Volume successfully.

[georgevj]

Could you post the output of " vol show -is-constituent true" ?

[Darkstar]

I suggest you read the TR 4178, "Infinite Volume Deployment and Implementation Guide", which describes what the mirrors are for and how the containing aggregates are selected:

http://www.netapp.com/us/media/tr-4178.pdf