I am running Network Load Balancing on two Windows Server 2012 R2 Datacenter VMs in a VMWare ESXi Host environment. The cluster and hosts are configured to use multicast and all are on the same VLAN and have both a dedicated LAN card as well as an NLB LAN card. When creating the cluster and joining the hosts to the cluster, the hosts converge and work properly for a seemingly random amount of time. After some time, however, one of the hosts goes to a "converging" state (it is always the same host). Stopping the host that is converged causes the converging host to converge at that time. However, I cannot get both hosts to converge simultaneously until I delete and recreate the cluster and re-add the hosts, and the cycle repeats. I enabled logging but I can't discern any error messages or problems from the information it gives me. When this happens, my load balanced site usually becomes unavailable until I either stop the converged host or delete and recreate the cluster.
I have created static ARP entries in my router and assigned static MAC addresses to the MAC table on my switches in order to make sure the packets are properly routed throughout my network. Like I stated above, it works, but only for a while.
Are there any NLB wizards that may know what is happening here? I will be happy to provide any additional information. Thanks in advance.
