Received: from gamera.begeek.com (cr714364-a.lndn1.on.wave.home.com [24.114.87.48]) by postgresql.org (8.11.3/8.11.1) with ESMTP id f5R5nVa41655 for ; Wed, 27 Jun 2001 01:49:31 -0400 (EDT) (envelope-from craigl@begeek.com) Received: from jigra (IDENT:craigl@jigra [10.2.0.13]) by gamera.begeek.com (8.9.3/8.9.3) with ESMTP id BAA16201 for ; Wed, 27 Jun 2001 01:45:36 -0400 Subject: Re: strange plan From: Craig Longman To: pgsql-sql@postgresql.org In-Reply-To: <993619683.6506.12.camel@jigra.begeek.com> References: <993619683.6506.12.camel@jigra.begeek.com> Content-Type: text/plain X-Mailer: Evolution/0.10 (Preview Release) Date: 27 Jun 2001 01:45:35 -0400 Message-Id: <993620736.6508.13.camel@jigra.begeek.com> Mime-Version: 1.0 X-Archive-Number: 200106/427 X-Sequence-Number: 3724 On 27 Jun 2001 01:28:03 -0400, Craig Longman wrote: > > i've got two machines both with roughly the same database, one has a few > more rows in some of the tables, but only about 2% more. both databases > have the same tables and the same indexes. however, the slower machine > seemed to be much slower than i would have expected it to be, so i > started doing some looking, and found some very different and strange > query plans. i'm sorry. i had been under the mistaken impression that the vacuumdb command updated the table/index statistics. i returned to looking through the manual after sending the email and almost immediately stumbled upon the 'vacuum analyze' command. sigh. it optimizes as expected now and returns in a proportionate amount of time. humbly, -- CraigL->Thx(); Be Developer ID: 5852