Re: Corrupt index stopping autovacuum system wide - Mailing list pgsql-general

From Peter Geoghegan
Subject Re: Corrupt index stopping autovacuum system wide
Date
Msg-id CAH2-WzksZTcAChogQCfrDeqbHCBZC3iBA8+WCcNp5NodrKnOaw@mail.gmail.com
Whole thread Raw
In response to Re: Corrupt index stopping autovacuum system wide  (Aaron Pelz <aaronepelz@gmail.com>)
List pgsql-general
On Thu, Jul 18, 2019 at 9:06 AM Aaron Pelz <aaronepelz@gmail.com> wrote:
> It's a simple btree expression on a geometry(Point,4326) , no expression no partial no composite.

The cause of the corruption may be a bug in a Postgis B-Tree operator
class. I reported a bug in the Geography type that could lead to
corrupt B-Tree indexes (not Geometry):

https://trac.osgeo.org/postgis/ticket/3841

Though I also see what could be a comparable bug in Geometry:

https://trac.osgeo.org/postgis/ticket/3777

These bugs are from about 3 years ago. If I'm right you should be able
to isolate the bug using amcheck.

-- 
Peter Geoghegan



pgsql-general by date:

Previous
From: Aaron Pelz
Date:
Subject: Re: Corrupt index stopping autovacuum system wide
Next
From: Adrian Klaver
Date:
Subject: Re: PostgreSQL as a Service