Re: BUG #19416: Backend SIGSEGV in ExecShutdownHashJoin/ExecHashTableDetach/dsa_free - Mailing list pgsql-bugs

From Michael Paquier
Subject Re: BUG #19416: Backend SIGSEGV in ExecShutdownHashJoin/ExecHashTableDetach/dsa_free
Date
Msg-id adW4FhgrcxmBQ_yI@paquier.xyz
Whole thread Raw
In response to Re: BUG #19416: Backend SIGSEGV in ExecShutdownHashJoin/ExecHashTableDetach/dsa_free  (Chris Hofstaedtler <ch+pg@zeha.at>)
List pgsql-bugs
On Tue, Apr 07, 2026 at 10:53:36AM +0200, Chris Hofstaedtler wrote:
> Sorry for the lack of followup. Currently we are thinking our problem
> matches the bug described in
> https://postgr.es/m/044401dcabac$fe432490$fac96db0$@icloud.com
> ("[BUG + PATCH] DSA pagemap out-of-bounds in make_new_segment odd-sized
> path").

That smells like the same problem to me.

> For now our plan is to wait for the next 18.x stable release, and then reenable
> enable_parallel_hash. Then see if it occours again.

The next minor release is planned for the 14th of May, as per:
https://www.postgresql.org/developer/roadmap/

Please let us know how it goes, in case we have a different problem at
hand.  You could as well cherry-pick 4da2afd01f93 in your builds of
v18, and give it a try in a non-production environment.
--
Michael

Attachment

pgsql-bugs by date:

Previous
From: PG Bug reporting form
Date:
Subject: BUG #19452: heap-buffer-overflow in `pg_basebackup --format=tar --gzip`
Next
From: Richard Guo
Date:
Subject: Re: BUG #19405: Assertion in eval_windowaggregates() fails due to integer overflow