We have a 2 node exchange 2010 HT / CAS setup that uses windows Load balancing in unicast mode.
Last 2 days one of the nodes becomes un responsive, when this happens I keep a constant ping and its response is 250ms, it gradually comes down to about 8ms then shoots straight back to 250ms and repeats the cycle. that's one issue and I cant work out what's causing it to do this as system event log don't say much, this happened yesterday at mid day and again this morning at 11.30.
next issue is that when this happens users start logging calls for not being able to connect, it seems the load balancing is not working correctly as I would expect the other node to continue running things.
The node that is failing is CAS02, in WNLB my host priority has this as priority 2 and CAS01 as priority 1
Can anyone point me in the direction of why the NLB may not be working correctly or is my expectations that this should be seamless incorrect.