Take the 2-minute tour ×
Server Fault is a question and answer site for professional system and network administrators. It's 100% free, no registration required.

I'm having a wierd issue setting up a HA group in vSphere. I have created the cluster, and added both machines to it, but only one machine at a time will reconfigure properly to the HA group.

One host is fine, the other simply reports "HA Agent has an error"

enter image description here

However, if i put the 'good' machine into maintenance mode, the other VM will successfully activate the HA agent.

enter image description here

So now the other machine is fine, so why not take the orignal "good" machine out of maintenance so all is right... nope, now that HA agent has an error.

enter image description here

I can swap these back and for and the behavior is consistant. I can create another HA group, and move a non-working server into it, and the HA agent will reconfigure just fine. If I move another server to it though, then the new server again cannot configure its HA agent. Also, the HA group warning is just a no management network redundancy.

The error on the server tasks simply says "HA agent has an error" and doesn't give any type of description. The related events all pertain to disabling/enabling the HA agent itself, with no other connected errors.

Has anyone seen this before? What am i doing wrong?

VMware Infrastructure Client 2.5 B-119598

VMware Virtual Center 2.5 B-119598

share|improve this question
    
Can you give us the specific errors? Also, what version of vSphere and servers are you using? And at this point, what's preventing you from adding management network redundancy? –  ewwhite Feb 7 '13 at 22:56
    
added it to the end of the original post –  Lee Harrison Feb 7 '13 at 22:59
    
1. It's not called a group, it's called a cluster. 2. It's not a VM, it's a host. 3. What do you see under Alarms for the cluster and for the hosts? Do you see only the management network redundancy warning? –  joeqwerty Feb 7 '13 at 23:01
    
Alarms are empty for all three items. The only warning I recieve is the network redundancy issue. Also, corrected the original post. –  Lee Harrison Feb 7 '13 at 23:04
    
I believe you need redundant networks for heartbeat here. –  SpacemanSpiff Feb 7 '13 at 23:15
show 1 more comment

2 Answers 2

This could be a dns issue. Make sure you can ping the two hosts using their fqdn from each other. SSH to each host, and add the hostname entries in /etc/hosts

192.168.1.152   test1.example.net  test1
192.168.1.153   test2.example.net  test2

If that doesn't work, make sure the clocks in the two hosts are synced. If that one fails, please follow the instructions in this blog.

share|improve this answer
add comment

To participate in a HA Cluster the two hosts require identical vSwitches and also common storage for VMs. There's other requirements also. I think you'll find that there is something missing, and it's probably related to your network configuration.

Please refer to this for details on what is required to configure the HA cluster if you have not already. https://www.vmware.com/files/pdf/techpaper/vmw-vsphere-high-availability.pdf

share|improve this answer
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

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