I have two 10 gig connections into a Virtual Distributed switch. We are using this for iSCSI, vMotion, and VM traffic. The Juniper switch ports that provide connectivity are configured as trunks ports, no LACP or aggregation.
- We use iSCSI port binding and both VMkernel ports are configured as "Route based on originating virtual port"
- We use multi-nic vMotion and both VMkernel ports are configured as "Route based on originating virtual port"
- Our VM portgroup is configured with "Route based on physical NIC load"
Everything has been working fine for the past 6 months when we put these new HPE servers in. We decided to enable Virtual Distributed Switch Health check and it came up with several problems including Non-IP Hash Mismatch under teaming and failover. This is either a mis-configuration on our part or the vDS health check is buggy and not reliable. We are running ESXi 6.7 update 1. What are you seeing in your vDS health check?