BUG #14054: "create index using gist ..." on large table never completes.

From: tsingle(at)muddypaddock(dot)com
To: pgsql-bugs(at)postgresql(dot)org
Subject: BUG #14054: "create index using gist ..." on large table never completes.
Date: 2016-03-30 08:38:19
Message-ID: 20160330083819.8934.34370@wrigleys.postgresql.org
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-bugs

The following bug has been logged on the website:

Bug reference: 14054
Logged by: Tim Singletary
Email address: tsingle(at)muddypaddock(dot)com
PostgreSQL version: 9.5.1
Operating system: osx 10.11.4
Description:

"create index using gist ..." ran for over three days on 100 million row
table; completes within a couple hours on a 10 million row table.

Had previously tried same experiment with 9.2.3 on a Linux box where the 100
million row table's index built within 5 hours.

The table looks like

CREATE TABLE mock (
handle INT NOT NULL
,lowest INT NOT NULL
,highest INT NOT NULL
,stuff TEXT
);

Data was randomly generated.

The command that never completes is:

create index on mock using gist (handle,(int4range(lowest,highest,'[]')));

Responses

Browse pgsql-bugs by date

  From Date Subject
Next Message Marc-Olaf Jaschke 2016-03-30 14:00:36 Re: Missing rows with index scan when collation is not "C" (PostgreSQL 9.5)
Previous Message maxim 2016-03-30 07:04:14 BUG #14053: postgres crashes in plpgsql under load of concurrent transactions