Re: [Patch] Fix check_pub_rdt bypass when origin is set in same ALTER SUBSCRIPTION - Mailing list pgsql-hackers

From Amit Kapila
Subject Re: [Patch] Fix check_pub_rdt bypass when origin is set in same ALTER SUBSCRIPTION
Date
Msg-id CAA4eK1KEPGO+pyQPe48UYUgh0Cu9D+PsVf5iftTuT6fF0kpwiA@mail.gmail.com
Whole thread
In response to [Patch] Fix check_pub_rdt bypass when origin is set in same ALTER SUBSCRIPTION  (SATYANARAYANA NARLAPURAM <satyanarlapuram@gmail.com>)
Responses RE: [Patch] Fix check_pub_rdt bypass when origin is set in same ALTER SUBSCRIPTION
List pgsql-hackers
On Tue, May 26, 2026 at 3:50 PM SATYANARAYANA NARLAPURAM
<satyanarlapuram@gmail.com> wrote:
>
> HI hackers,
>
> When ALTER SUBSCRIPTION SET (retain_dead_tuples = true, origin = 'none')
> is used, the publisher version/recovery check was bypassed because the
> origin handling unconditionally overwrote check_pub_rdt to false.
>
> Fix by using |= instead of = so that the flag set by retain_dead_tuples
> cannot be cleared by a subsequent origin assignment in the same command.
>

Thanks for catching this. Your fix LGTM. I won't be able to commit
today as I have to travel. In the meantime, let's see if Hou-San or
others have any comments.

--
With Regards,
Amit Kapila.



pgsql-hackers by date:

Previous
From: Álvaro Herrera
Date:
Subject: Re: logical: fix recomputation required LSN on restart_lsn-only advancement
Next
From: Alvaro Herrera
Date:
Subject: Re: Adding REPACK [concurrently]