We are seeing degraded performance from one switch and checking it, there is some packetloss due to this on some servers across winvps, and clusters 1 and 2.
Issue looks to be resolved, one of our power strips seems to have gone bad on the network for us, and it took a while to resolve because it wasn't showing obvious signs of any trouble.
We are working on this, performance is going up and down in spikes, we are working to isolate the exact cause and sometimes may have a port offline for a minute or two.