Why multiple aggregates?

[SysAdmin1]

Customer has two nodes, both connected to two racks/shelves of disks (total 24 disks).  One node has an aggregate using all 24 disks and the other node has another aggregate using all those same disks.  Each aggregate has an SVM.  Why not have one aggregate with two SVMs instead?

[andris]

I'm not really following your question... perhaps you can provide more details (what is the FAS/AFF model, is advanced disk partitioning (ADP) being used, clarify data aggregates v.s. root aggregates).

Each node will have its own root aggregate (might be the small partitions on the physical disks).

If you're active/passive configuration, you'd only have one large data aggregate on one of the nodes and the other node doesn't serve data during normal operation

If you want both nodes to serve data, then each node would have its own data aggregate.

Aggregates can only belong to one storage virtual machine (SVM) a.k.a. Vserver in ONTAP CLI.

Unless you need to completely partition data access among multiple tenants, you only need one data SVM and all of the aggregates would be mapped to that single SVM.

https://docs.netapp.com/us-en/ontap/concepts/storage-virtualization-concept.html

[SysAdmin1]

These are data aggregates that were created by the site, not the Netapp guy.  Maybe the nodes are both active for each aggregate (why?)?  The two nodes are in cluster and can fail over to one another as well.  Where do I see the active/passive info?  I'm in the GUI, but can connect via Putty if needed.

[andris · Accepted solution]

In an active/passive configuration, only one of the nodes in the HA pair will own a data aggregate.

If you like, collect the "sysconfig -r" output from each node.

See: https://kb.netapp.com/Advice_and_Troubleshooting/Data_Storage_Software/ONTAP_OS/How_to_run_sysconfig_-a_or_sysconfig_-r_in_ONTAP_9