Re: Performance Tuning / RAM Usage

From: Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us>
To: Oliver Elphick <olly(at)lfix(dot)co(dot)uk>
Cc: Andrew Sullivan <andrew(at)libertyrms(dot)info>, pgsql-general(at)postgresql(dot)org
Subject: Re: Performance Tuning / RAM Usage
Date: 2002-08-28 16:50:39
Message-ID: 200208281650.g7SGod720232@candle.pha.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

Oliver Elphick wrote:
> On Wed, 2002-08-28 at 15:55, Andrew Sullivan wrote:
> > On Wed, Aug 28, 2002 at 10:40:46AM -0400, Tom Lane wrote:
> > > Andrew Sullivan <andrew(at)libertyrms(dot)info> writes:
> > > > I know that the Debian package does not -- or at least didn't used to
> > > > -- check for deadlocks at all,
> > >
> > > [blinks] What?
> >
> > Hmm. I just checked my home box (where I have the Debian package
> > installed), and it seems no longer to be the case. There's no entry
> > for deadlock_timeout, so the default (1000, IIRC) is what gets used,
> > I guess. I have a dim memory, however, of a comment in the old
> > config file to the effect that the (Debian) default setting of
> > deadlock_timeout was infinite, i.e. effectively no deadlock checking.
> >
> > Hmm. I just checked an ancient box with 6.5 on it, and it's not
> > there. Maybe I'm on crack.
>
> I fear so :-)
>
> I'm sure I've never made any such change to the package.

Remember Debian had the pg_hba.conf 'peer' patch in their distribution a
few years ago, so it is possible they mucked up the deadlock timeout
too --- not likely, but possible.

--
Bruce Momjian | http://candle.pha.pa.us
pgman(at)candle(dot)pha(dot)pa(dot)us | (610) 359-1001
+ If your life is a hard drive, | 13 Roberts Road
+ Christ can be your backup. | Newtown Square, Pennsylvania 19073

In response to

Browse pgsql-general by date

  From Date Subject
Next Message Bruce Momjian 2002-08-28 16:54:19 Re: Tuple identifier (tid) - object identifiers
Previous Message Tourtounis Sotiris 2002-08-28 16:42:03 What is the size of Tuple identifier (tid) - object identifiers (fwd)