Re: Adding a '--clean-publisher-objects' option to 'pg_createsubscriber' utility.

On Fri, Mar 14, 2025 at 12:26 AM Zhijie Hou (Fujitsu) <
houzj(dot)fnst(at)fujitsu(dot)com> wrote:

> For logical replication, there is a frequent need to automatically delete
> all
> objects (including publications) on replicas that are no longer needed.
> This
> requirement comes from a common use case in logical replication, which is
> to
> replicate only a subset of database objects.

I don't see us implementing an API to provide an alternative to "ALL
TABLES"...

Personally, I am uncertain about
> the use case for retaining only some of the publications while dropping
> others.
>

To assume there is nothing between All or None seems unwise even if we
cannot imagine what that may be.

> Besides, I am unsure if the rationale behind pg_upgrade's output script is
> applicable here.

Yeah, I probably shouldn't have mentioned it. I don't need it to explain
or support my case.

I think the option proposed in this thread is not to handle the
> version mismatch between the publisher and the subscriber.

Certainly not given that this requires a physical standby as a base and
those are not cross-major-version capable.

Overall,
> I favor automatically deleting publications rather than offering a script
> for
> manual execution.
>
>
I'm done arguing the counter-point to this. I cannot give this a +1 unless
the DBA is given a chance to review and edit the drop commands that we
create. But as of now I'm in the minority and this has a committer willing
to proceed without this capability. I take some comfort in that it seems
at least if they use --dry-run they can see what objects will be dropped.
But the editor capability seems more useful. There is always v19 and
beyond; at least nothing here precludes adding such a feature in the future.

David J.