Horton

Re: Parallel Index Scan vs BTP_DELETED and BTP_HALF_DEAD

Thomas Munro <thomas.munro@enterprisedb.com>

View thread View on postgr.es
From: Thomas Munro <thomas.munro@enterprisedb.com>
To: Pg Hackers <pgsql-hackers@postgresql.org>
Date: 2017-12-11T03:07:23Z
Lists: pgsql-hackers 
On Mon, Dec 11, 2017 at 3:51 PM, Thomas Munro
<thomas.munro@enterprisedb.com> wrote:
> I heard a report of a 10.1 cluster hanging with several 'BtreePage'
> wait_events showing in pg_stat_activity.

I forgot to add, for bug reporter credit purposes: thanks to Patrick
Hemmer for the off-list report and backtrace.  He's able to work
around this recurring problem for now by disabling parallelism.

-- 
Thomas Munro
http://www.enterprisedb.com

Commits

  1. Fix parallel index scan hang with deleted or half-dead pages.