Horton
Menu

Re: Bug in ALTER SUBSCRIPTION ... SERVER / ... CONNECTION with broken old server

Jeff Davis <pgsql@j-davis.com>

View thread
From: Jeff Davis <pgsql@j-davis.com>
To: Amit Kapila <amit.kapila16@gmail.com>
Cc: Chao Li <li.evan.chao@gmail.com>, Zsolt Parragi <zsolt.parragi@percona.com>, "Hayato Kuroda (Fujitsu)" <kuroda.hayato@fujitsu.com>, Ajin Cherian <itsajin@gmail.com>, PostgreSQL-development <pgsql-hackers@postgresql.org>
Date: 2026-07-06T23:46:10Z
Lists: pgsql-hackers 
On Mon, 2026-06-29 at 09:57 +0530, Amit Kapila wrote:
> wrconn = walrcv_connect(conninfo, true, true, must_use_password,
> subname, &err);

...

> So, if the connection is not successful then we simply return without
> checking tablesync slots (when slotname is NONE) whereas with FDW
> server connection, the ERROR will be raised from
> construct_subserver_conninfo->ForeignServerConnectionString(). Isn't
> that different from current situation?

It's easier to cause an error in ForeignServerConnectionString(); but
walrcv_connect() can throw errors, too. Also, a network connection is a
side-effect that you may not want to have for whatever reason.

I think it's better to have an escape that prevents any connection to
the publisher, as the documentation says, to deal with these kinds of
edge cases.

Regards,
	Jeff Davis

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

  1. Avoid errors during DROP SUBSCRIPTION when slot_name is NONE.

  2. Avoid errors during ALTER SUBSCRIPTION.

  3. Check retain_dead_tuples for ALTER SUBSCRIPTION ... SERVER.

  4. Allow multiple xacts during table sync in logical replication.