Re: [BUG] Assert failure in ReorderBufferReturnTXN during logical decoding due to leaked specinsert change

Álvaro Herrera <alvherre@kurilemu.de>

From: Álvaro Herrera <alvherre@kurilemu.de>

To: "Hayato Kuroda (Fujitsu)" <kuroda.hayato@fujitsu.com>

Cc: 'Vishal Prasanna' <vishal.g@zohocorp.com>, pgsql-bugs <pgsql-bugs@lists.postgresql.org>

Date: 2026-06-16T18:22:16Z

Lists: pgsql-bugs

On 2026-Jun-10, Álvaro Herrera wrote:

> Thanks.  As far as I can tell this is just a leak, right?  There's
> nothing that misbehaves when assertions are disabled, even with
> MEMORY_CONTEX_CHECKING -- it's just a small, very improbably mem leak.
> 
> I'm tempted not to touch REL_14_STABLE with it in that case.

I ended up pushing to 14 after all, after I noticed that the code in
that branch is pretty much identical to what's in 15 -- it was only the
test code that required a separate patch file.

-- 
Álvaro Herrera         PostgreSQL Developer  —  https://www.EnterpriseDB.com/

Commits

logical decoding: Correctly free speculative insertion
- f50c329f538f 19 (unreleased) landed
- bd7b2184390f 18 (unreleased) landed
- ac445069002c 17 (unreleased) landed
- 813b0d8f455d 16 (unreleased) landed
- 1add4a41bc31 15 (unreleased) landed
- 0939aad56402 14 (unreleased) landed
Fix memory counter update in ReorderBuffer.
- dbed2e36625d 17.0 cited