Re: agent, clusters and failover

Are you talking about the Intelligent Agent?

Or something else??

If it's the Intelligent Agent, then of course you need one on every node, otherwise you can't manage the cluster *as* a cluster, but merely as a collection of separate instances. Enterprise Manager, for example, uses GSD on one node to talk to the other GSDs on the other nodes to perform, for example, cluster-wide shutdowns or backups or exports. GSD pulls off that trick by in turn talking to the Intelligent Agents on each node to handle the actual scheduling of a job. No Intelligent Agent, no cluster-wide scheduling.

Maybe I've misunderstood your question though.

For example, the question "should I bind the agent at the service or at the node" means nothing if we're talking about the Intelligent Agent, since that is intrinsically a "node process", running independently of whether or not the instances comprising your RAC are actually up or not.

Regards
HJR "FM" <fabrizio.magni_at_mycontinent.com> wrote in message news:mRG9c.114528$z23.4862907_at_news3.tin.it...
> Maybe a trivial question:
> how would you set oracle agents (10g in particular) in a cluster
> environment?
>
> In a simple failover system? Would you keep an agent active on every
> node or only on the one which is giving service?
>
> And for mutual failover? Would you bind the agent at the service(s) or
> at the node?
>
> I see more advantages having an agent on every node but I'd welcome
> different opinions.
> Anybody with experience in clustered services monitored by agents?
>
> Thank you
> Fabrizio Magni
>
>
> --
>
> Fabrizio Magni
>
> fabrizio.magni_at_mycontinent.com
>
> replace mycontinent with europe
Received on Sun Mar 28 2004 - 16:58:58 CST