Horton

Re: BUG #17485: Records missing from Primary Key index when doing REINDEX INDEX CONCURRENTLY

x4mmm@yandex-team.ru

View thread View on postgr.es
From: Andrey Borodin <x4mmm@yandex-team.ru>
To: Andres Freund <andres@anarazel.de>
Cc: Alvaro Herrera <alvherre@alvh.no-ip.org>, Robert Haas <robertmhaas@gmail.com>, Peter Geoghegan <pg@bowt.ie>, Michael Paquier <michael@paquier.xyz>, Петър Славов <pet.slavov@gmail.com>, PostgreSQL mailing lists <pgsql-bugs@lists.postgresql.org>
Date: 2022-05-31T06:47:11Z
Lists: pgsql-bugs 

> 31 мая 2022 г., в 00:08, Andres Freund <andres@anarazel.de> написал(а):
> 
> We're going to need to revert both, but I don't understand why c98763bf51bf
> would affect this your tests. Afaics it should just affect CIC, not RIC. Could
> you make sure you can reproduce the issue without c98763bf51bf reverted?

Here's the revision with tests v3 and revert of d9d0762 [0]. In CI you can see the tests pass on Linux and FreeBSD, but fail on MacOS [1].

Best regards, Andrey Borodin.

[0] https://github.com/x4m/postgres_g/commits/test
[1] https://cirrus-ci.com/task/4580203115577344?logs=test_world#L1566

Commits

  1. Revert changes to CONCURRENTLY that "sped up" Xmin advance

  2. Fix possible HOT corruption when RECENTLY_DEAD changes to DEAD while pruning.