I had just a few (4) vms on and in one of them I was installed oracle (free edition) and in the middle of the install my vic connection was loosing response to the server. All the vms where running fine and I was still able to load web sites zippy quick that where hosted from vms on the box. Ping response was fine to all the vms ips as well as the managment ip I was having the slow response time from. A few minutes later, I did an nmap (sniffed the ports) and noticed almost all the TCP/IP ports where closed. Onlys 2 or three where open. ssh wasn't one of them. Every few tries I was able to reconnect but most items I tried to view couldn't pull any details before my connection got cut again.
I did a reboot and all is well now. When I tried to check out the graphs, I got this weird spike in system, datastore, disk and datastore graphs (the attachment). I'm asuming I didn't allocate enough resources to vmware so I checked and the resources allocated to vmware where the defaults, 500Mhz of cpu and a tiny amount of ram. I was doing a lot of I/O with that oracle install so could the stall be from vmware not having enough resources to juggle a i/o demanding vm?