Re: dynamic service registration

From: Howard J. Rogers <hjr_at_dizwell.com>
Date: Wed, 24 Mar 2004 17:32:46 +1100
Message-ID: <40612b90$0$3952$afc38c87@news.optusnet.com.au>

"srivenu" <srivenu_at_hotmail.com> wrote in message news:1a68177.0403232154.69e6e004_at_posting.google.com...
> > What may or may not happen when the listener is re-started doesn't
exactly
> > persuade either, because that's a special case with quite possibly a
special
> > code path.
> I dont think it is a seperate code path as the code for the Oracle
> server does not depend on the availability of the listener.

No, but maybe the start of the Listener invokes a prod to PMON if it's there?

> Coming to the facts, the fact is that the service resgitration with
> the listener never takes more than 1 minute.

So my log file wasn't a "fact", then? Or just an inconvenient one.

> Now deducing from this fact, there has to be some means of
> communication going on between the two every minute.

It doesn't mean that at all.

> I dont think the communication would be initiated by the listener as
> it has not idea about the PMON's address.

Fair enough. But "I don't think" still doesn't explain my listener log file, does it?

> So deducing from the fact, it shows that PMON contacts the listener
> every minute (which is not reflected in the log).

Ah. So you mean it's the logging that's at fault?

I don't think so either.

HJR Received on Wed Mar 24 2004 - 00:32:46 CST