Re: Postgres DB Slowness

Juan José Santamaría Flecha <[email protected]> Fri, 23 Aug 2019 13:08:24 +0200

Newsgroups	gmane.comp.db.postgresql.admin
Message-ID	<CAC+AXB2=EA-iCdURZDxALoJUHsRi+dn0a=J6L26m=wPCSqnY5A@mail.gmail.com>

On Fri, Aug 23, 2019 at 10:17 AM <[email protected]> wrote:
>
> Query
> ===========
> SELECT
>     i.*
> FROM
>     npcurren.num_cps_instelling i,
>     npcurren.num_aangesloten_nr n
> WHERE
>     n.fk_exploit_nop_int_oper_id = 'PTT'
>     AND i.telefoonnummer != n.anr_nummer_hoog
>     AND i.telefoonnummer != n.anr_nummer_laag;
>
> Explain Plan
> ======================
> "Nested Loop  (cost=2068.47..1769226624.33 rows=101090505159 width=73)"
>

This query looks wrong, resulting in a cartesian join.

As other people have pointed out, the execution times do not seem
reasonable, could you change the 'SELECT *' into a 'SELECT count(1)'
for the comparison?

Regards,

Juan José Santamaría Flecha