VMware Cloud Community
Cosmic_Charlie
Contributor
Contributor

Lost Network Connectivity After Performing VMotion Migration

Host Configuration[/u]

2 Dell PE2950s 16 GB RAM

EMC CX3-20 Drive array (fiber connected)

vmware 3.0.1 Enterprise (no patches)

VC 2.0.1

Problem

We're in the test/dev process of our first taste of vmware. Everything is setup and appears to be working well. We're in the middle of testing vmotion and have run into a snag.

After setting up a test guest vm, migrating between ESX hosts normally work fine. We've ran continuous pings to the guest and we're pleased to hardly notice any sort of loss of connectivity.

When we try to copy a large file onto the guest, and start the vmotion migration during the copy, the guest vm drops all network connectivity. The copy hangs, the pings begin to timeout, and we can no longer control the vm from the VC console.

The only resolution we have found so far is to power down the vm and start it back up. Everything is fine at that point.

This problem is reproducible 10 out of 10 times.

We've done some searching and can't come up with anything. Any ideas?

0 Kudos
5 Replies
MR-T
Immortal
Immortal

Is your VMotion connection 1GB?

Also, do you have your VMkernel switch on it's own VLAN or the same LAN as the VM's?

0 Kudos
Cosmic_Charlie
Contributor
Contributor

1GB yes.

The VMKernel switch is not on its own VLAN. It shares the same VLAN and NIC as the VM.

0 Kudos
zemotard
Hot Shot
Hot Shot

Hi, I have the same problem with the same configuration ...

Do you have solved this problem ?

Regards.

Best Regards If this information is useful for you, please consider awarding points for "Correct" or "Helpful".
0 Kudos
Dave_C1
Contributor
Contributor

Hi guys,

I had this problem recently and couldn't seem to figure it out from looking here in the forums. I opened an SR with VMware and after trying a few things that didn't work, they recommended a particular patch:

http://kb.vmware.com/kb/2066306 (the part about high CPU load was the relevant part for me)

Applied the patch last week and then tested it today. Still getting timeouts more frequently than we'd like, but the issue with losing networking/hanging has not returned.

Hope this helps!

0 Kudos
jasonboche
Immortal
Immortal

Can I guess your VMs might be running Microsoft Network Load Balancing in unicast or multicast mode?

VCDX3 #34, VCDX4, VCDX5, VCAP4-DCA #14, VCAP4-DCD #35, VCAP5-DCD, VCPx4, vEXPERTx4, MCSEx3, MCSAx2, MCP, CCAx2, A+
0 Kudos