No Guest OS heartbeats are being received...

[RaglanDan]

I have just added a third ESX 3.0 server to our infrastructure. I have setup the networking identically to the existing esx servers and have Vmotioned 4 machines across to the new server. All went fine for this and I Vmotioned a couple back and forth a few times.

Today when I try to Vmotion any of the machines off I receive the following error message:

Yellow warning.

Migration from esx.host: No guest OS heartbeats are being received. Either the guest OS is not responding or VMware tools is not configured properly.

The migration completes fine but I'm sure this is something that I need to address. Also on the new esx server I can get no performance data and on the summary page there is no CPU or Memory usage information.

Any ideas anyone?

[jftwp]

I have just opened a case wherein these events were noted in virtualcenter.

In my situation, this message was a very bad thing. The guest OS/vm in question was dead (could not ping it, manage it, connect to its resources, etc.). It was 'on' but just dead. Was fine prior to this, then just 'died'. Only option available via VI3 client was to 'Power Off' the VM, which was fine (for now). It came back up okay, but I opened case cuz want to know 'why' this happened in the first place.

Running ESX 3.0.1 patched through March released patches. Waiting to hear back / work with VMware on this. Cluster with numerous other VMs otherwise fine.

[thedafa]

Hi,

Did you ever figure out why this happened? Same thing is happening here

\--

dafa

[aarggh]

Hi,

I'm also very curious if you had some positive results regarding this issue? I thought at first it seems to be a problem with my Linux servers not being fully compatible with vmtools, but upon testing HA with the SAN (by pulling out a Fibre cable), all servers auto migrated fine from one node to the other, although the Linux servers were powered off and not powered back up.

Trying it futher from the server none could migrate back to the original node, and later a win2k3 std server could migrate fine to either of two nodes, but couldn't later still. Trying again later the win2k3 was fine again but the Linux ones still wouldn't. DNS doesn't seem to be an issue. I did selct the "reconfigure HA" option on the nodes but this seemed to just allow the win2k3 to migrate occasionally.

[nbrowne1]

I'm getting this in ESX 4.1... Migrations still work fine though. And the KB article from VMware says not to worry about it:

http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=100380...

[REED201110141]

I also see this problem on vSphere 4.1.  In our case, there is a specific VM that becomes isolated on its VM network interface if the warning is ignored and the VM migrates to a different host.  I can migrate back to the original host without warnings or issues (in this case, it is a management cluster of two hosts).  Unfortunately, it is fairly catastrophic as the VM is a SQL server supporting vCenter.

I already tried the VMware KB that blames VMware tools for the heartbeat warning and recommended reinstalling VMware tools.  That seemed to work for one migration, but now the VM has reverted to its previous behaviour.  We are using a vmnext3 virtual NIC.

[dvbrenner]

This may not be a serious issue in all cases.  I tried what was suggested (first suggestion) in the following blog and it did indeed work:

http://www.frostbyte.com.au/Blog/tabid/64/EntryId/10/VMware-vCenter-and-No-guest-OS-heartbeats-error...