VMware Cloud Community
Anthony1Robinso
Contributor
Contributor

VM Unresponsive problems.

Alright, I will try to make as much sense out of this as I can. I have a VXRail with 4 Nodes, tons of space, power, etc. From vSphere Web Client, I have a VM named "Ticketsystem" , I can move it between nodes all day without any issues. However, anytime I move around or do any work on any other VM, it will sometimes cause Ticketsystem to become unresponsive. Sometimes I can console in and do a proper reboot, other times I have to Reset the VM from vSphere. I cannot find an answer for this happening, nor a fix that seems to take. No other VM I have seems to be affected by this, the TicketSystem VM is the only one having this issue.

Ticketsystem Specs: Win Server 2012r2 - 8 Sockets, 8 Cores per Socket. 16GB Ram & 275GB HD. (Started with CPU 4/2 & memory 8GB, gave it more power a couple months ago hoping that would help)

I'm sure there is a ton of more info I should have put, let me know what else can help, and thank you!

2 Replies
itsmattmagbee
Contributor
Contributor

Have you looked at the NUMA core/socket ratio? I have run into similar issues with my vm's until I started to follow this rule.  I have found large 12 cpu vm's are best with 1 socket and 12 cores. After 12 cores you go up 1 socket to 2 sockets 2 cores and so on.

If you can give that a shot ... I think you will find a possible solution.

Anthony1Robinso
Contributor
Contributor

Thank you! I will give this a shot and let you know.

Reply
0 Kudos