Thread

Re: nbtree VACUUM's REDO routine doesn't clear page's VACUUM cycle ID

Andrey Borodin <x4mmm@yandex-team.ru> — 2025-11-19T04:36:19Z

> On 18 Nov 2025, at 23:54, Peter Geoghegan <pg@bowt.ie> wrote:
> 
> Are you arguing that the xl_btree_split record should include the cycleid?

Yes.

> I see that systems that are built on this architecture do something
> along these lines:
> https://github.com/neondatabase/postgres/commit/a9b92820c5d14dbff8f59ab65ffdaae92ab9c3c8

Thanks for the link. This solution seems surprisingly complicated.

> However, that seems well out of scope for core Postgres. At least for
> the foreseeable future.

Yes, I agree. It's just a case when redo routines do not match original execution, and it seemed interesting to me. I understand that for core Postgres it's extra overhead in WAL record only for a beautification.

Thanks!


Best regards, Andrey Borodin.