| From: | "Andrey M(dot) Borodin" <x4mmm(at)yandex-team(dot)ru> |
|---|---|
| To: | Salma Elsayed <salmasayed182003(at)gmail(dot)com> |
| Cc: | Zsolt Parragi <zsolt(dot)parragi(at)percona(dot)com>, Jim Jones <jim(dot)jones(at)uni-muenster(dot)de>, pgsql-hackers mailing list <pgsql-hackers(at)lists(dot)postgresql(dot)org>, Kirk Wolak <wolakk(at)gmail(dot)com>, Laurenz Albe <laurenz(dot)albe(at)cybertec(dot)at>, Pavel Stehule <pavel(dot)stehule(at)gmail(dot)com>, Kirill Reshke <reshkekirill(at)gmail(dot)com>, Nikolay Samokhvalov <samokhvalov(at)gmail(dot)com> |
| Subject: | Re: PoC: Simplify recovery after dropping a table by LOGGING the restore LSN |
| Date: | 2026-11-08 06:58:13 |
| Message-ID: | 36EE8B64-0908-40A3-AF53-B8C42EF831AA@yandex-team.ru |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Hi Salma,
The default text-log prefix has no database name. IDK, users might want
it: if you do a lot of CREATE DATABASE ... TEMPLATE ... table names and
oids are same all the way.
I tried Zsolt's SyncRep scenario with v13. Both messages were already
logged as expected. What do you think about adding a TAP check for the
message while the backend still waits in SyncRep?
I'm not convinced we need it, I don't foresee what kind of bugs could
hide there in future. Just raising the question about this tests.
Thank you!
Best regards, Andrey Borodin.
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Denis Smirnov | 2026-11-08 07:03:57 | Re: Detoast a column once per row instead of once per reference |
| Previous Message | Zhijie Hou | 2026-11-08 06:10:23 | Re: Avoid using raw xmin of frozen tuples in logicalrep conflict detection |