Your basic ITPro blog... What's going on at work, what I'm interested in.

Thursday, December 17, 2009

Odd TCP/IP Behavior in Hyper-V Virtual Machines

I have two Hyper-V hosts running a total of around 20 VMs. I recently came across some odd behavior that ended up in a call to Microsoft Support, as I couldn’t figure it out on my own and we didn’t want to spend any more time on it ourselves. Basically, I was seeing the following:

cap1

As you can see, ping times were all over the place. We found a solution in a combination of KB articles and blog posts.

 

RESOURCES:

KB938448

KB895980

http://fawzi.wordpress.com/2009/10/28/hyper-v-domain-controller-negative-ping-results/

http://joystickjunkie.blogspot.com/2009/04/erratic-or-negative-ping-times-on-hyper.html

http://blogs.msdn.com/tvoellm/archive/2009/02/18/why-does-my-avg-disk-write-sec-counter-keep-climbing.aspx

 

I have three VMs that are multi-proc and all three of them were doing this. All three VMs are running a flavor of Windows Server 2003. I don’t know if this happens with other OSs on multi-proc VMs… I am guessing not. With the /usepmtimer switch added in the boot.ini file, all three are now working as expected. I hope that the Hyper-V team is working on a solution to this so that boot.ini file manipulation is not required in the future.

No comments:

Additional Info

My photo
email: support (AT) mangrumtech (DOT) com
mobile: 480-270-4332