Re: Prepared statements, parameters and logging

From: Csaba Nagy <nagy(at)ecircle-ag(dot)com>
To: Dave Cramer <pg(at)fastcrypt(dot)com>
Cc: Postgres JDBC <pgsql-jdbc(at)postgresql(dot)org>
Subject: Re: Prepared statements, parameters and logging
Date: 2007-06-25 13:34:35
Message-ID: 1182778475.13328.79.camel@coppola.muc.ecircle.de
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-jdbc

On Mon, 2007-06-25 at 15:11, Dave Cramer wrote:
> I think the text should be specifying that once you exceed
> preparedThreshold executions the driver uses a named prepared
> statement, and can re-use it. Otherwise it uses unnamed prepared
> statements which need to be prepared for each execution.

OK, but I also thought that the value 0 means no reuse at all, i.e. it
should always prepare for each execution... and 1 should mean always
reuse, from the first statement on.

Before I was using prepareThreshold='1', and I was seeing only such
entries in the server log:

duration: 5437.499 ms statement: EXECUTE <unnamed> [PREPARE: SELECT
... ]

Note the <unnamed> in the log entry, it was always there for each
statement.

After setting prepareThreshold='0', I started getting 2 kinds of log
entries, the ones like above, and the likes:

duration: 2200.570 ms statement: EXECUTE C_220047 [PREPARE: SELECT
... ]

Note the C_220047, that's what's changed...

I distinctly remember discussing this issue some time ago on this list,
back then I needed the opposite action, to disable the one-time-prepare
by default... now I just need the opposite, to enable it by default even
for repetitive queries... is that possible ?

Cheers,
Csaba.

In response to

Responses

Browse pgsql-jdbc by date

  From Date Subject
Next Message Bady, Brant RBCM:EX 2007-06-25 16:18:37 Can MD5 sums be posted for the downloadable files?
Previous Message Heikki Linnakangas 2007-06-25 13:20:04 Re: a simple example of XA (not working)