From: pgsql-sql@ruby.sartorelli.gen.nz
To: pgsql-sql@hub.org
Subject: Problem with SQL query (eats swap)
Date: Fri, 11 Aug 2000 14:46:21 +1200 (NZST)
Message-ID: <m13N4px-003bmOC@ruby.sartorelli.gen.nz> (raw)
Hi, I have two tables into which I dump mail statistics. The two tables are:
create table attachments (
id text,
attachment text
);
create table mail (
id text,
size int,
whofrom text,
subject text,
date datetime,
inout char
);
The table mail holds information about each mail message. The table attachments holds the name of any attachments. Linking the attachment(s) and the message is the id.
I want to delete all records relating to mail that is over two months old. I tried:
select count(*) from attachments a where a.id in (select m.id from mail m where m.date < now()-62);
but ran out of swap. The mail table is 19Mb, the attachment one 1Mb. I was up to 380Mb of swap used on a machine with 128Mb RAM, and over 15 minutes run time. At that point I killed the query.
I next tried is as follows:
drop table temp;
select id into temp from mail where date < now()-62;
delete from attachments where id in (select id from temp);
delete from mail where date < now()-62;
This worked fine and deleted the records as intended in a few minutes. Can anyone see why the initial query ate so much swap?
Cheers
Kevin
kevin@sartorelli.gen.nz
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Reply to all the recipients from your mail client; the link fills in
the headers below: reply via email
Some clients, webmail especially, drop In-Reply-To and References from
a link, and the reply then starts a new thread: use the mbox below.
To: pgsql-sql@postgresql.org
Cc: pgsql-sql@ruby.sartorelli.gen.nz, pgsql-sql@hub.org
Subject: Re: Problem with SQL query (eats swap)
In-Reply-To: <m13N4px-003bmOC@ruby.sartorelli.gen.nz>
References: <m13N4px-003bmOC@ruby.sartorelli.gen.nz>
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox