Re: [(known) BUG] DELETE/UPDATE more than one row in partitioned foreign table

Hi Nikita,

On Fri, May 15, 2026 at 2:23 AM Nikita Malakhov <hukutoc(at)gmail(dot)com> wrote:
> CFbot was unhappy with previous patch set, so here's updated one

Thanks for working on this issue!

I took a quick look at the patch set. IIUC I think it's created based
on what I proposed in the original thread, which is invasive and thus
not back-patchable, so what you are proposing here isn't
back-patchable, either, I think.

I think we should first work on a back-patchable solution. So I'd
like to re-propose the patch that I proposed in this thread before to
disallow UPDATE/DELETE in problematic cases [1]. Attached is a new
version of the patch. Changes are:

* Renamed the new table option inherited to remotely_inherited, to
avoid confusion with local inheritance.
* Moved the logic to prevent problematic UPDATE/DELETE from a planner
function to an executor function, to avoid throwing an error
unnecessarily when there are no target rows to update/delete.
* Added docs to postgres-fdw.sgml.

I'm planning to add the postgresImportForeignSchema() support in the
next version.

I think the remotely_inherited option would be useful when adding the
support for the UPDATE/DELETE, as it could be used to address one of
Tom Lane's comments about what I proposed in the original thread that
it adds the tabloid condition to a remote UPDATE/DELETE query whether
the target table is inherited or not: that could be avoid if the
option is set to false.

What do you think about that?

Best regards,
Etsuro Fujita

[1] https://www.postgresql.org/message-id/CAPmGK15CQK-oYFMAyq%2BrR0rQapUHtvAGuGgY5ahERHzZ4tmC8g%40mail.gmail.com