Re: PostGIS 3.5.2 ST_Contains() query on gist is slow
Paul Ramsey via postgis-users <[email protected]> Tue, 9 Sep 2025 10:25:12 -0700
| Newsgroups | gmane.comp.gis.postgis |
|---|---|
| Message-ID | <CACowWR1OYVpMeKtsCmegesQJQSenptoknPrTjCvh=TvY6kz7Xg@mail.gmail.com> |
On Mon, Sep 8, 2025 at 8:34 PM Sebastiaan Couwenberg via postgis-users < [email protected]> wrote: > may be a regression in PostGIS 3.5.2, but I'm at a loss on how to confirm > and fix that. > > So, this seems to be a change in planning behaviour primarily, and you have a lot of variability in your upgrade, any piece of which could be implicated: - Have you run ANALYZE on the relevant tables? - You have a PostgreSQL upgrade from 15 to 17, can you hold your PostGIS version constant while testing PostgreSQL version changes - You have a PostGIS upgrade from 3.3 to 3.5, can you hold your PostgreSQL version constant while testing PostGIS version changes - Can you check the PostGIS selectivity estimate on the old and the new installations, using SELECT _postgis_selectivity ('nodes', 'geom', ST_GeomFromEWKT('SRID=4326;POLYGON((6.11982 52.6121955,6.11982 53.2038323,7.0927397 53.2038323,7.0927397 52.6121955,6.11982 52.6121955))'), '2'); P