gdal version for Postgis 2.4?

From: "Zwettler Markus (OIZ)" <Markus(dot)Zwettler(at)zuerich(dot)ch>
To: "pgsql-general(at)lists(dot)postgresql(dot)org" <pgsql-general(at)lists(dot)postgresql(dot)org>
Subject: gdal version for Postgis 2.4?
Date: 2020-03-09 15:50:34
Message-ID: bf18b22341b7411d89e8513fd2d27440@zuerich.ch
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

I am in the same situation:
<quote>
I found that Postgis 24 depends on gdal30-libs.x86_64 0:3.0.4-2.
But this installation creates an error :
dbname=# create extension postgis;
ERROR: could not load library "/usr/pgsql-11/lib/rtpostgis-2.4.so":
/usr/gdal30/lib/libgdal.so.26: undefined symbol: GEOSMakeValid_r
And I found that postgis 24 actually depends on gdal23.
</quote>
https://www.postgresql.org/message-id/CAAMgDXks6XSX7NxVhggEQvtPQ%2BE1niZsKyK1%2Bwzki8jwuW1wAQ%40mail.gmail.com

My Postgis 2.4 rpm also seems to depend on gdal30-libs-3.0.4:
<quote>
BuildArch: x86_64
Name: postgis24_96
Version: 2.4.8
Release: 7.rhel7
...
Requires: gdal30-libs >= 3.0.4
</quote>

QUESTIONS:
Does Postgis 2.4 depend on gdal23?
Is there a bug with the rpm dependency?
How to fix this?

$ uname -a
Linux szhm49251 3.10.0-1062.12.1.el7.x86_64 #1 SMP Thu Dec 12 06:44:49 EST 2019 x86_64 x86_64 x86_64 GNU/Linux
$ rpm -aq | grep postgres
postgresql96-contrib-9.6.17-1PGDG.rhel7.x86_64
postgresql96-libs-9.6.17-1PGDG.rhel7.x86_64
postgresql96-server-9.6.17-1PGDG.rhel7.x86_64
postgresql96-devel-9.6.17-1PGDG.rhel7.x86_64
postgresql96-9.6.17-1PGDG.rhel7.x86_64
postgresql96-plpython-9.6.17-1PGDG.rhel7.x86_64
$ rpm -aq | grep postgis
postgis24_96-2.4.8-7.rhel7.x86_64
postgis24_96-client-2.4.8-7.rhel7.x86_64
$ rpm -qa | grep gda
gdal-libs-1.11.4-12.rhel7.x86_64
gdal30-libs-3.0.4-1.rhel7.x86_64
gdal23-libs-2.3.2-8.rhel7.x86_64

Responses

Browse pgsql-general by date

  From Date Subject
Next Message Andres Freund 2020-03-09 18:34:40 Re: Is it safe to rename an index through pg_class update?
Previous Message Kouber Saparev 2020-03-09 15:47:23 Re: Is it safe to rename an index through pg_class update?