VMware

ESXi 6.5u3 iSCSI redundancy not working

Hello all,

We have two domains with B series UCS servers. Recently I’ve been upgrading and fixing various issues. While working with Cisco TAC, we had to bring down one of our Fabric Interconnects and when that was done, all storage communication stopped. Our first thought was that the hosts weren’t configured correctly network wise but as soon as I triggered a rescan, all my ESXi hosts immediately started talking to the VMs again

I’m not sure what’s going on but with two FIs and no redundancy, it’s kinda silly.. so here’s a quick breakdown.

Two FIs
Many hosts
Two iSCSI networks on different subnets
If one FI goes down, storage communication is lost until a rescan is triggered
No networking on either iSCSI network fails


View Reddit by t3hWheezView Source

Related Articles

4 Comments

  1. How are your NICs and VMkernel ports set up for iSCSI port bindin? It’s one of the things I often see set up incorrectly. Failing that then look at the path selection policy.

  2. Another thing to check is the vNIC configuration. Make sure you have both A and B side FIs configured. Either as a failover or with a seperate vNIC fixed to side A and a second fixed to side B.

  3. The iSCSI port binding is not to be used in this setup.
    It’s a common misconception that port binding has to be configured all the time.

    The port binding feature is for iSCSI setups where initiator and target ports in the SAN reside in the SAME subnet / broadcast domain.
    That multi fabric design you’re describing doesn’t need port binding.

    For more information look here:
    https://kb.vmware.com/s/article/2038869

Leave a Reply

Your email address will not be published. Required fields are marked *

Close