We are having trouble with both our servers. Here's our setup and the problem:
2 identical Asus
RS520-E8-RS12-EV2 servers, fully updated, with Server 2019 Standard Hyper-V. Both servers have an Intel X710 2-port networkcard (SFP+) (latest Intel driver) These cards have been added later for faster networking. So they don't come standard onboard.
The servers are directly connected together using one SPF+ cable for quick Hyper-V replication that works fine.
Both servers are also connected to the LAN network on one RJ45 port 1GB (no teaming) This network card is also attached to a Virtual Switch (1GB)
Both servers are also connected to the LAN network on one SPF+ port using a 10GB Fibre cable to a Huawei switch (no teaming), also connected to a Virtual Switch (10GB)
The problem: We keep getting this error 10400 in the event logs every 30 seconds:
The network interface "Intel(R) Ethernet Converged Network Adapter X710-2" has begun resetting. There will be a momentary disruption in network connectivity while the
hardware resets. Reason: The network driver detected that its hardware has stopped responding to commands. This network interface has reset 455 time(s) since it was last initialized.
When this event occurs, the connectivity on the 10GB Fibre port (to the LAN) is lost for about 3 or 4 pings. All the virtual machines (6) connected to this Virtual Switch (10GB) also lose connection so the whole network is crippled that way.
Once I switched all VM's to the Virtual Switch (1GB) RJ45, the problem stops and the host itself can be pinged again on the 10GB port, no problem all day. At this point I don't really know what to look for. We also get this message all day long
every second of the day, despite the problem went away,
Event 25, Hyper-V-VmSwitch
The MAC address 3C-FD-FE-CD-FE-5D has moved from port 515A1F44-F3BA-49FB-8E2D-2F4D970F472E (Friendly Name: Virtual Switch - DAN_External) to port C40C1E38-A55D-4415-B7ED-2D29C6BEE007 (Friendly Name: Virtual Switch - DAN - SPF - 10GB)
The MAC address 3C-FD-FE-CD-FE-5D has moved from port C40C1E38-A55D-4415-B7ED-2D29C6BEE007 (Friendly Name: Virtual Switch - DAN - SPF - 10GB) to port 515A1F44-F3BA-49FB-8E2D-2F4D970F472E (Friendly Name: Virtual Switch - DAN_External).
What I notice is that the friendly name "Virtual Switch - DAN - SPF - 10GB" is the one I created, the friendly nameVirtual Switch - DAN_External is unknown to me, this is not a Virtual Switch present in Hyper-V there is no switch with that name.
Both ports of the Intel X710 show this message all day filling the log (direct attached cable and the 10GB LAN connection)
We checked the switch and there are 0 errors or collissions on all switch ports so it does not seem to be switch related.
When the VM's are not connected to the 10GB Virtual Switch the card is 100% stable
To me this appears to be a driver problem or maybe a config error in Hyper-V that we are unaware of. VMQ is enabled by default in Hyper-V and also the drivers of the Intel card.
Any clue?