Re: [PATCH] Btree BackwardScan race condition on Standby during VACUUM

Michail Nikolaev <michail.nikolaev@gmail.com>

From: Michail Nikolaev <michail.nikolaev@gmail.com>

To: Peter Geoghegan <pg@bowt.ie>, PostgreSQL Hackers <pgsql-hackers@postgresql.org>

Date: 2020-07-10T20:25:15Z

Lists: pgsql-hackers

Hello, Peter.

Thanks for the update.

Yes, it is the right decision.
I have started to spot that bug only while working on a faster scan
using hint bits on replicas [1], so it is unlikely to hit it in
production at the moment.

Thanks,
Michail.

[1]: https://www.postgresql.org/message-id/CANtu0ojmkN_6P7CQWsZ%3DuEgeFnSmpCiqCxyYaHnhYpTZHj7Ubw%40mail.gmail.com

Commits

Fix replica backward scan race condition.
- 9a9db08ae462 14.0 landed
Justify nbtree page split locking in code comment.
- 9945ad6e9040 13.0 cited
Remove unneeded "pin scan" nbtree VACUUM code.
- 9f83468b3536 13.0 cited
Compute XID horizon for page level index vacuum on primary.
- 558a9165e081 12.0 cited
Make heap TID a tiebreaker nbtree index column.
- dd299df8189b 12.0 cited