Horton

Re: BUG #15577: Query returns different results when executed multiple times

Thomas Munro <thomas.munro@enterprisedb.com>

View thread View on postgr.es
From: Thomas Munro <thomas.munro@enterprisedb.com>
To: Bartosz Polnik <bartoszpolnik@gmail.com>
Cc: David Rowley <david.rowley@2ndquadrant.com>, PostgreSQL mailing lists <pgsql-bugs@lists.postgresql.org>
Date: 2019-01-07T23:38:48Z
Lists: pgsql-bugs 
On Tue, Jan 8, 2019 at 12:04 PM Bartosz Polnik <bartoszpolnik@gmail.com> wrote:
> On Mon, Jan 7, 2019 at 11:23 PM Thomas Munro <thomas.munro@enterprisedb.com> wrote:
>> As a sanity check, can you please see if the run-only-in-leader case
>> (max_parallel_workers = 0, so that we see "Workers Launched: 0")
>> produces the *same* 31 rows as the run-only-in-worker case
>> (force_parallel_mode = on, so that we see "Single Copy: true")?  That
>> is, the actual values of those 31 rows, in particular the columns
>> coming from table_b.
>
> Rows returned by two queries are the same.

Can we please also see the actual output in the broken case, where it
runs in two processes and produces fewer than 31 rows?

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

Commits

  1. Avoid sharing PARAM_EXEC slots between different levels of NestLoop.