up vote 2 down vote favorite
1
share [g+] share [fb]

I'm trying to set up VMware HA for the cluster, and having trouble understanding how resource monitoring works. We overcommit memory as a general practice, so out provisioned memory is always ~1.5 higher on individual VMs.

So I created a cluster with 2 Hosts in it, and one was ~90% full in terms of memory (and I mean used memory, as provisioned was ~140%). Second host ran no VMs. Tried powering on one VM - and I got an error, saying that would make it impossible to tolerate one host failure.

Reading more, I found that when this happens and you disable the policy to prevent power on, VMware will not guarantee failover for all hosts.

  • But does it mean that it will just not try if it thinks that there is not enough resources?
  • Or does it mean that something bad might happen because memory usage will go over all available, and it'll have to start swapping?
  • How does it make such decisions?
link|improve this question
100% accept rate
feedback

1 Answer

up vote 2 down vote accepted

This is normal behavior. From vCenter Server right click on Cluster > Edit Settings > VMware HA and check "Disable: Power on VMs that violate availability constraints." That would fix your issue.

Basically in 2 node HA cluster, when one node is down, VMs have no machine to failover if the second node also fails and you are disabling that check and that is normal for 2 node clusters.

If you had 3 or more nodes in a cluster, then you can keep that option enabled.

link|improve this answer
Well, that is clear. The question is that will VMware try to power on VMs on a machine that "doesn't have enough resources?" And to what point? – Serge Jul 27 '11 at 16:44
Resource Pools and priority settings for individual VMs would help you there. If you have resource pools setup properly, then all the high priority machines should start up first and if resources are available other machines would be powered on. – Raj J Jul 27 '11 at 18:15
How is "resources are available" determined? Would it start up machines until memory usage hits 100%? Or how is it determined? – Serge Jul 27 '11 at 22:54
If you have 5 VMs each set with 2GB memory reservation in a resource pool with 9GB memory reservation, after 4 VMs are powered on, fifth one would not come up. So it would depend on/vary by resource pool memory limit. – Raj J Jul 27 '11 at 23:50
Ok, that's when you have DRS. But we have just "Standard" license. So no resource pools. Any insight on how that's going to be handled? ANd if you can point me to the right documentation would be great. – Serge Jul 28 '11 at 1:18
show 1 more comment
feedback

Your Answer

 
or
required, but never shown
discard

Not the answer you're looking for? Browse other questions tagged or ask your own question.