Original-Subject: Re: Why isn't PG using an index-only scan?
Original-From: Jean-Christophe BOGGIO <postgresql@thefreecat.org>
Original-Cc: 
  "pgsql-performance@lists.postgresql.org"
 <pgsql-performance@lists.postgresql.org>
Message-ID: <63d65775-0130-4c5c-a151-8d7728b6d4e8@thefreecat.org>
Date: Thu, 18 Sep 2025 09:45:13 +0200
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Subject: Re: Why isn't PG using an index-only scan?
To: David Rowley <dgrowleyml@gmail.com>
Cc: 
  "pgsql-performance@lists.postgresql.org"
 <pgsql-performance@lists.postgresql.org>
References: <8f25c08c-acb2-47ca-b8c1-6664fc31a361@thefreecat.org>
 <CAApHDvoUdvw68oHnfACQE=u8XPJ+Te49dZsfC2yW+DVfzdT2+Q@mail.gmail.com>
Content-Language: en-US, fr
From: Jean-Christophe BOGGIO <postgresql@thefreecat.org>
In-Reply-To: 
 <CAApHDvoUdvw68oHnfACQE=u8XPJ+Te49dZsfC2yW+DVfzdT2+Q@mail.gmail.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
Archived-At: 
 <https://www.postgresql.org/message-id/63d65775-0130-4c5c-a151-8d7728b6d4e8%40thefreecat.org>
Precedence: bulk

Thanks David,

Le 18/09/2025 à 09:20, David Rowley a écrit :
> Yes. Since *all* records of "oeu" are required and they're not
> required in any particular order, then Seq Scan should be the fastest
> way to access those records.

Ok but then why is it doing it on the AD table? Is it because of the 
number of rows?

> 5 hours seems very slow for the estimated number of records. Have you
> tried running the SELECT using EXPLAIN ANALYZE without the INSERT
> part? Even if the 84 million Merge Join row estimate is accurate, 5
> hours seems excessively long.

I tried this, it took 1 second. And then I discovered that I had an old 
trigger on the copyrightad table.

The insert took 2 seconds.

Very sorry for wasting your time and thanks again for putting me on the 
right track.

Have a great day!