| From: | "Andrey M(dot) Borodin" <x4mmm(at)yandex-team(dot)ru> |
|---|---|
| To: | Peter Geoghegan <pg(at)bowt(dot)ie> |
| Cc: | PostgreSQL Hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org> |
| Subject: | Re: nbtree VACUUM's REDO routine doesn't clear page's VACUUM cycle ID |
| Date: | 2024-11-20 09:40:09 |
| Message-ID: | 81659D22-A31A-4866-A124-4E55E382E5D0@yandex-team.ru |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
> On 15 Nov 2024, at 21:33, Peter Geoghegan <pg(at)bowt(dot)ie> wrote:
>
> Attached patch teaches btree_xlog_vacuum, nbtree VACUUM's REDO
> routine, to reset the target page's opaque->btpo_cycleid to 0. This
> makes the REDO routine match original execution, which seems like a
> good idea on consistency grounds.
>
> I propose this for the master branch only.
The change seems correct to me: anyway cycle must be less than cycle of any future vacuum after promotion. I cannot say anything about beauty of resetting or not resetting the field.
I'd suggest renaming the field into something like "btpo_split_vaccycleid". I was aware of index vacuum backtracking, but it took me a while to build context again.
Best regards, Andrey Borodin.
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Toto guyoyg | 2024-11-20 10:11:04 | Planner picks n² query plan when available |
| Previous Message | Daniel Gustafsson | 2024-11-20 09:12:27 | Re: A way to build PSQL 17.1 source on AIX platform |