VCS 2.0 GAB heartbeat Issues

[Rajesh]

Hellow
We are running VCS 2.0 on a two node (Sun Fire 280R) Cluster. Both these node have the same hardware 2 USIII CPUs and 4 Gig RAM.

The Machines are absolutely slow in response. If we tail the engine log then some thing like this appear every now and then;
TAG_E VCS:10486: Excessive delay within GAB heartbeat call(6seconds)
The CPU is not busy at all, neither HAD is being restarted by Shadow HAD, as the process-id of HAD remains same. We actually have 30 Service Groups on this Cluster.

Any help as why is this happening, would be appreciated as I really do not have any idea.

Regards, Rajesh
THUS Plc.
UK

[Jim_Senicka]

You need to make sure you are on latest Maintenance pack (and really need to get to 3.5 or 4.0)
Missing GAB heartbeat is indicitive of a couple things.
First is usualy CPU load.
Second is some other process tying up something GAB needs.
There were several fixes to HAD/GAB thorughout 2.0 product to address this issue.

I would highly recommend getting to 3.5 MP3 or 4.0 MP1, as 2.0 is end of life

[Rajesh]

Thanks Jim, your help is much appreciated.

Cheers, Rajesh