We have esxi set up to have two vSwitches. One has a Virtual Machine port group, and is 2 fiber connections that are set up as an etherchannel group on the Cisco stack. The other has a VMkernel port group, and is two ethernet connections that are also set up as an etherchannel group on the Cisco stack. This second one is configured with the management IP we use to access the host in vSphere.
This has worked fine for years, no problem. However, today we had an issue with one of the network cards in one of the servers, which caused one of the ethernet ports to only be 100 instead of 1000Mbps. This caused the switch to suspend the port group, which then meant that I couldn't see the host in vSphere. Fine. Irritating that an issue on one port would cause the interface to get suspended, but I'm going to figure out if there's anything I can do about that. Regardless, if the management port is down, I can't see it in vSphere...this makes sense.
However, the other issue is this - when the management network went down, I could also no longer access the guest VMs themselves. This seems...not at all ideal. Is this expected behaviour? If so, is it changeable? If not, any idea how I can track down what's going wrong and fix it?
Thanks!