/opt/LGTOaam512/bin/ft_startup failed to complete within 3 minutes......

[myjess1]

HA reconfig error:

/opt/LGTOaam512/bin/ft_startup failed to complete within 3 minutes.

Anyway:

11 server cluster.

1st server now down hard. cannot even get into RILO.

Attempting to patch other servers to get some work done.

2nd server patched and rebooted, but would not reconfigure for HA for a long time, but eventually did.

3rd server patched and rebooted. Cannot get this server into HA mode at all.

In VVC ViC I get the message stated in at the top of this post.

In the /opt/LGTOaam512/config/vmware-sites file, it references only one other server, and that just happens to be esx01, which is the one that is down hard. I tried copying the file from the other 2nd server on which HA eventually worked, but when I reconfigured for HA on the 3rd server, the file went back to just referencing the 1st (hard down) server!

I know from scant previous posts, that there is a master HA server.

Questions:

1. Is the Master HA server the 1st one, and does this vmware-sites file make ft_startup go looking for that 1st server.

2. How can I make it that another server becomes the Master server for HA?

3. How do I fix this fault, if I am completely wrong and off track with questions 1 and 2.

Thank you all.

[myjess1]

Earlier message not too clear.

1st server is down hard for an unknown reason before I started the patching. I even disconnected this 1st server from within the cluster before working on the others.

I unchecked HA on the cluster and then re-enabled it again. ALL servers in the cluster now will not reconfigure for HA. It gone from bad to worse!

[uslacker99]

Check that DNS is working - that you can pinging each host by short/long name would suffice for this. You don't use CNAMEs when adding the hosts to VC, do you? You don't want to do that. You can also try a new cluster if you don't have resource pools and stuff set up or if you can recreate them easily.

[Chamon]

We have had this before and just had to remove HA and add it again 2 or 3 times before it would finally work. HA seams a bit flaky. I would just give that a try. I know that it doesn't sound right but it worked for us.

What is the error in the cluster or for the Host now?

[myjess1]

What are CNAMES?

Message in VVC for hosts is exact same for all and is same as I first posted.

I will do as you suggest, disabling and enabling HA a number of times.

I will also make a new cluster and put some in there to see what happens.

Thanks for the replys.

[Rajeev_S]

cnames are CANONICAL names.. A computer can have only one hostname. But for a single host or IP address we can multiple cnames. This would be useful in web hosting ( to resolve to different names for the same IP).

For this issue as already mentioned you can check DNS (short and FQDN). I recommend you to manually do a entry in /etc/hosts file. If you are using VC2.0.2 make sure that u type it in lower case. This should mostly fix the problem. HA is completely dependent on DNS. Is DRS working in the cluster??

If you planning to remove and add it to a new cluster.Before removing it from existing cluster make sure that u put it on Maintenance mode.

Hope this helps

[myjess1]

I think I found out what the problem is.

I created a new cluster, shut down all the VM's, maintenance'd all the Hosts, and then moved them over to the new cluster. I then created new resource pools exactly like the old ones, and moved the relevant VM's into the relevant resource pools.

Exited maintenance mode and all the hosts came up configured for HA with no errors.

I moved a host back to old cluster. Still would not HA. Go figure!

[marklemon]

If yo have a bogus IP address in your /etc/hosts file, you get this error as well.

correcting it works a charm.

[Nashwood]

did you receive an answer for this question? I have the exact same question for the community. Thanks.