HPE Nimble iSCSI Discovery Connectivity

[SteveJ1]

Hi, I have installed two Nimble AF60 storage arrays across two datacentres. Included in each datacenter are two HPE SN201M iSCSI switches and Synergy 12000 frames with SY480 servers. 

We have configured the two Nimble storage arrays as a group with storage array A as group leader and storage array B as BGL.  The Synergy servers have ESX installed with volumes presented via the Nimble storage with the Synergy frame and Nimble storage array both connected directly to the two iSCSI switches at each site.

In addition we have configured peer persistence  to replicate the volumes between both sites using the mgmt ports.

The issue is that the storage presented via the group leader (storage array A) can be 'seen' by the ESX hosts in Site A but the storage presented via the BGL (storage array B) cannot be 'seen' by the ESX hosts in Site B.

I bleieve this is because when adding the secondary storage array to the group it uses the network configuration of the group leader including discovery IP addresess. As the iSCSI switches are directly connected to each storage array and Synergy frame at each site they will need a Discovery IP for each iSCSI SAN network i.e.

• iSCSI Discovery IP for site A e.g. 192.168.1.2
• iSCSI Discovery IP for site B e.g. 192.168.2.2

Is this correct?

If so how do we configure that with a group, do we simply create a new subnet for site B and assign the Site B storage array ports to it?

Thanks in advance for any guidance on this.

[support_s]

System recommended content:

1. Advisory: HPE Synergy Converged Network Adapter - HPE Nimble Storage May Experience Poor Performance When an HPE Synergy Compute Module Is Configured With Both Hardware iSCSI and Software iSCSI

2. HPE OneView 6.6 User Guide for HPE Synergy | HPE Nimble Storage systems

Please click on "Thumbs Up/Kudo" icon to give a "Kudo".

Thank you for being a HPE valuable community member.

[Thomas_Lam_HK]

Please note that PP is just a form of storage clustering which 2 arrays under same group with Sync Repl enabled, and all of the managment activities are carried by Group Leader.

And under PP configuration, all host need to "see" all arrays' iSCSI port for multipathing.  The upstream volume will propogate "Active-Optimized" path and downstream array will propogate as "Backup"" path.

For iSCSI discovery IP address, regardless on Site-A or Site-B,  all server should use the same discover IP address from the data network which is hosted on the Group Leader.

https://infosight.hpe.com/InfoSight/media/cms/active/public/HPE_Nimble_Storage_Peer_Persistence_Deployment_Considerations.pdf