Received: from sss2.sss.pgh.pa.us (sss.pgh.pa.us [209.114.166.2])
	by hub.org (8.9.3/8.9.3) with ESMTP id KAA32909
	for <pgsql-interfaces@postgreSQL.org>;
	Wed, 16 Feb 2000 10:12:05 -0500 (EST)
	(envelope-from tgl@sss.pgh.pa.us)
Received: from sss2.sss.pgh.pa.us (tgl@localhost [127.0.0.1])
	by sss2.sss.pgh.pa.us (8.9.3/8.9.3) with ESMTP id KAA02773;
	Wed, 16 Feb 2000 10:10:41 -0500 (EST)
To: Peter Mount <petermount@it.maidstone.gov.uk>
cc: "'M.Mazurek@poznan.multinet.pl'" <M.Mazurek@poznan.multinet.pl>,
	pgsql-interfaces@postgreSQL.org
Subject: Re: [INTERFACES] invalid lo descriptorc in begin commit fraze 
In-reply-to: 
 <1B3D5E532D18D311861A00600865478C70C1E9@exchange1.nt.maidstone.gov.uk>
References: 
 <1B3D5E532D18D311861A00600865478C70C1E9@exchange1.nt.maidstone.gov.uk>
Comments: In-reply-to Peter Mount <petermount@it.maidstone.gov.uk>
	message dated "Wed, 16 Feb 2000 08:52:15 +0000"
Date: Wed, 16 Feb 2000 10:10:41 -0500
Message-ID: <2770.950713841@sss.pgh.pa.us>
From: Tom Lane <tgl@sss.pgh.pa.us>

Peter Mount <petermount@it.maidstone.gov.uk> writes:
> I'll look into it. There shouldn't be anything in the driver to limit
> the number of concurrent objects.
> Does anyone else see this type of problem (I'm thinking of those who use
> libpq rather than JDBC)?

I got the impression that what he was talking about was multiple
independent threads issuing commands across a single backend connection.
When one thread commits its transaction, all the others lose their open
LO descriptors.  I don't see any way to fix that short of changing the
client code so that each thread is talking to a different backend.

If I'm all wrong about this, I'll crawl back in my hole now...

			regards, tom lane