Re: PHJ file leak.

On Fri, Mar 6, 2020 at 9:43 AM Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:

> Jeremy Finzel <finzelj(at)gmail(dot)com> writes:
> > We are hitting this leak in production on an 11.6 system for a query that
> > is using a parallel hash join. Was this fix pushed in 11.7? I can't tell
> > clearly from the release notes for 11.7 or this thread.
>
> It looks like you're asking about this commit:
>
> Author: Thomas Munro <tmunro(at)postgresql(dot)org>
> Branch: master [76cbfcdf3] 2019-11-16 10:11:30 +1300
> Branch: REL_12_STABLE Release: REL_12_2 [24897e1a1] 2019-11-16 10:18:45
> +1300
> Branch: REL_11_STABLE Release: REL_11_7 [bc049d0d4] 2019-11-16 10:19:16
> +1300
>
> Always call ExecShutdownNode() if appropriate.
>
> which is documented thus in the 11.7 release notes:
>
> <listitem>
> <!--
> Author: Thomas Munro <tmunro(at)postgresql(dot)org>
> Branch: master [76cbfcdf3] 2019-11-16 10:11:30 +1300
> Branch: REL_12_STABLE [24897e1a1] 2019-11-16 10:18:45 +1300
> Branch: REL_11_STABLE [bc049d0d4] 2019-11-16 10:19:16 +1300
> -->
> <para>
> Ensure parallel plans are always shut down at the correct time
> (Kyotaro Horiguchi)
> </para>
>
> <para>
> This oversight is known to result in <quote>temporary file
> leak</quote> warnings from multi-batch parallel hash joins.
> </para>
> </listitem>
>
>
> regards, tom lane
>

Thank you! Yep, pretty clear :).