I'm trying to figure out what is wrong with a server at work that was recently virtualized. When pinging the VM from anywhere in the network, there is a lot of variance in the ping times. I wanted to eliminate the network as the issue, so I ran an extended ping from the VM to the loopback IP and am seeing consistent spikes. Average response is 1ms and then will jump to 15ms for one or two pings. This happens roughly once every 1 - 2 minutes. I've several occasional spikes up to 600ms!

We've upgraded the NIC to VMXNET3. Verified the NIC settings and made sure all offloading functions to the NIC are enabled. The VM has plenty of resources available and is not starving for bandwidth. Most of the time the server is sitting idle with very little load on it.

I have no idea what else could be causing this issue other than a P2V conversion gone bad. Any thoughts?

So you're pinging 127.0.0.1 from the server and seeing spikes? If so, sounds like it needs more CPU.

_________________
http://blog.alwaysthenetwork.com

Yes, pinging 127.0.0.1. The VM looks fine on resources. CPU usage is never higher than 25%. We've moved the VM to a host that didn't have any other VM's on it to rule out CPU ready or any other resource contention, but still the issue persists.