Thread

Re: Fix slot synchronization with two_phase decoding enabled

Nisha Moond <nisha.moond412@gmail.com> — 2025-06-12T07:40:00Z

On Tue, Jun 10, 2025 at 2:29 PM shveta malik <shveta.malik@gmail.com> wrote:
>
> On Fri, Jun 6, 2025 at 12:37 PM Nisha Moond <nisha.moond412@gmail.com> wrote:
> >
> > Attached v18 patch.
> >  - patch-001: modified error messages as suggested above.
> >  - patch-002: improved pg_dump docs as per Shveta's off-list suggestions.
> >
>
> Thanks for the patches. Please find few comments:
>

Thanks for the review.

> 1)
> + * However, we allow this combination in binary upgrade mode, where
> + * pg_upgrade guarantees that all slot changes are consumed and no
> + * prepared transactions exist.
> + */
>
> Can we please mention on which system pg_upgrade ensures that no
> prepared txn exists. Is it at the source system or target system or
> both? The similar comment is there at other place too, please update
> that as well.
>

pg_upgrade checks for prepared transactions on both source and target
systems. I've updated the respective comments.
~~~

Attached v19 patches addressing all the comments in [1] and [2].

[1] https://www.postgresql.org/message-id/CAJpy0uCGE0fi4oMb9A_L27C-fJq%2BVpSO9ZbcVxW5s5F3%2BsX0Lw%40mail.gmail.com
[2] https://www.postgresql.org/message-id/CAFPTHDYXjq_%3Dr2ss6YMQBr%3DwbSfkH%3DuGJFrJzVpaW0eqPhy0RQ%40mail.gmail.com
--
Thanks,
Nisha