vCenter

Hello,

If I have 2 VM Hosts in a cluster environment. Under the VMware HA summary, I see this:

Current Failover Capacity: 1 host

Configured Failover Capacity: 1 host

Shouldn't the Current Failover Capacity be 2 hosts instead?

Thanks,

Daniel

Hello,

It basically says you have capacity to lose 1 ESX host. If you lost two you would have no ESX host to failover to with only two hosts.

That's correct, if one host failed you would only have one other working host to fail over to.

Thank you for the responses....in other words, my HA is reporting the right info and it is configured correctly?

The reason I was looking into this was for some reason, HA had an error of

"Insufficient resources to satisfy HA failover level on cluser..." and seconds later, HA reconfigured itself and back to normal. All this was going on in less than 1 minute. I checked the Cluster resources but everything was normal and showed no signed of insufficient resources. Is there anything else I can check to see why HA though there was insufficient resources?

Thanks,

Daniel

in other words, my HA is reporting the right info and it is configured correctly?

Yes, it looks good.

As far as where to look for a potential issue or why you saw a brief error message you could try:

/var/log/vmware/vpx/vpxa.log

/var/log/vmware/hostd.log

It may have been that across the two hosts you were running more virtual machines that could be handled on one if you were to lose a host.

Hope that helps!

Insufficient resources to satisfy HA failover -

> You can configure DRS to avoid this situation by powering Off unecasssary VM's or yu just can do the adminission process for them... You need to manage the resource pool some how that all the VM can accomodat at time of failover.

Make sure you don't have much memory and CPU reserved for VM's.Try enabling the extended resource for adminission.

Is your ESX really went down when HA start or it was just a network glitch..... in this you can avoid suddent pressure on resources.

Essentially, the "Insufficent resources" message meant that you were over capacity should one host fail.

Example: You have one host running at 50% CPU, and another running at 60%. This is 110% total, and therefore if one host dies, there are not sufficient resources available to run all VM's at their current service levels.

As you add hosts to the cluster you can boost your average utilization (2 hosts = 50%, 3 hosts = 66%, etc).

Thanks for all the help folks. I appreciate it