Re: Crashes on a partition whose concurrent detach never finished

On Thu, Nov 12, 2026 at 4:40 PM Alvaro Herrera <alvherre(at)alvh(dot)no-ip(dot)org> wrote:
>
> On 2026-Nov-12, Zhijie Hou wrote:
>
> > Since there's only one publication in this file, I think the current name is
> > fine as we don't need to distinguish it from any other publication. Similar
> > names are also used in other tests.
>
> Right now, isolation tests run one after the other, so you're right that
> this name doesn't really matter much. However, someday it might be nice
> if we can make (some of) them run in parallel. We tried to do that eons
> ago, couldn't find the time to implement it, idea was forgotten. If
> that effort ever takes off again it would be nice if we don't have to
> fix a thousand pointless name conflicts of this kind. (We already have
> enough of those that we'll have to fix.) So I think Shveta had the
> right idea here.
>

Fair enough, now that I have committed the patch, is it worth
considering to change the name in HEAD-only?

BTW, it will be helpful to get some feedback from you on the 0004
patch in email [1] as that directly touches the detach partition ...
concurrently code path. You might want to read the email [2] to
understand the issue 0004 patch is trying to solve.

[1]: https://www.postgresql.org/message-id/CADzfLwU97r3QC5oZDeVcWvcvXhNy9VnoFNDVHw%2BERg_DEG4%3DFQ%40mail.gmail.com
[2]: https://www.postgresql.org/message-id/CABdArM4fJ8JAWyP0MAF%3D8n66M6%3DCouUmtSGwqizn2pD%3DgfPEZw%40mail.gmail.com

--
With Regards,
Amit Kapila.