Re: killed select?

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: jtp <john(at)akadine(dot)com>
Cc: pgsql-general(at)postgresql(dot)org
Subject: Re: killed select?
Date: 2002-02-28 21:48:21
Message-ID: 14991.1014932901@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general pgsql-hackers

jtp <john(at)akadine(dot)com> writes:
> A question was asked which i through to the database to see how it was
> able to handle the question at hand and it failed . . . after 50 minutes
> of processing it flopped to the ground killed: out of swap space.

My guess is that what actually bombed out was psql, which tries to
buffer the entire result of a query. (Well, actually it's libpq not
psql that does that, but anyway the client side is what's failing.)

I suspect that your query is insufficiently constrained and will return
many millions of rows --- are you sure you have the WHERE clauses right?

If you actually do need to process a query that returns gazillions of
rows, the best bet is to declare a cursor so you can fetch the result
in bite-size chunks, say a few hundred rows at a time.

regards, tom lane

In response to

Browse pgsql-general by date

  From Date Subject
Next Message Richard Emberson 2002-03-01 00:30:49 bulk insertions in PL/pgsql functions
Previous Message Andrew Sullivan 2002-02-28 21:36:59 vsnprintf and 64-bit Solaris 7 (was: abnormal exits)

Browse pgsql-hackers by date

  From Date Subject
Next Message Greg Sabino Mullane 2002-02-28 22:23:46 Database Caching
Previous Message Bruce Momjian 2002-02-28 21:35:05 Re: elog() patch