Re: Lengthy deletion

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Herouth Maoz <herouth(at)unicell(dot)co(dot)il>
Cc: pgsql-general(at)postgresql(dot)org
Subject: Re: Lengthy deletion
Date: 2011-11-29 09:10:03
Message-ID: 11988.1322557803@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

Herouth Maoz <herouth(at)unicell(dot)co(dot)il> writes:
> On 29/11/2011, at 09:13, Tom Lane wrote:
>> Yup, that's a clue all right. I'll bet a nickel that you don't
>> have an index on the foreign key's referencing column (ie,
>> sent_messages.subscription_id). That means each delete in
>> the referenced table has to seqscan the referencing table to
>> see if the delete would result in an FK violation.

> Makes sense. But shouldn't that be figured into the EXPLAIN plan?

If you run EXPLAIN ANALYZE, on a reasonably recent release, you'll see
the foreign-key trigger eating a lot of time. Plain EXPLAIN doesn't
show triggers because it has no idea how much time they'll take.

regards, tom lane

In response to

Browse pgsql-general by date

  From Date Subject
Next Message Albe Laurenz 2011-11-29 09:34:30 Re: odbc_fdw
Previous Message Herouth Maoz 2011-11-29 08:51:46 Re: Lengthy deletion