Re: Transaction timeout

x4mmm@yandex-team.ru

From: "Andrey M. Borodin" <x4mmm@yandex-team.ru>

To: Japin Li <japinli@hotmail.com>

Cc: Junwang Zhao <zhjwpku@gmail.com>, 邱宇航 <iamqyh@gmail.com>, Fujii Masao <masao.fujii@oss.nttdata.com>, Andrey Borodin <amborodin86@gmail.com>, Andres Freund <andres@anarazel.de>, Michael Paquier <michael.paquier@gmail.com>, Nikolay Samokhvalov <samokhvalov@gmail.com>, pgsql-hackers <pgsql-hackers@postgresql.org>, "pgsql-hackers@lists.postgresql.org" <pgsql-hackers@lists.postgresql.org>

Date: 2024-01-04T08:41:43Z

Lists: pgsql-hackers

Attachments

v22-0001-Introduce-transaction_timeout.patch (application/octet-stream) patch v22-0001
v22-0002-Add-tests-for-transaction_timeout.patch (application/octet-stream) patch v22-0002
v22-0003-Try-to-enable-transaction_timeout-before-next-co.patch (application/octet-stream) patch v22-0003
v22-0004-fix-reschedule-timeout-for-each-commmand.patch (application/octet-stream) patch v22-0004

> On 4 Jan 2024, at 07:14, Japin Li <japinli@hotmail.com> wrote:
> 
> Does the timeout is too short for testing? I see the timeouts for lock_timeout
> and statement_timeout is more bigger than transaction_timeout.

Makes sense. Done. I've also put some effort into fine-tuning timeouts Nik's case tests. To have 100ms gap between check, false positive and actual bug we had I had to use transaction_timeout = 300ms. Currently all tests take more than 1000ms!
But I do not see a way to make these tests both stable and short.

Best regards, Andrey Borodin.

Commits

Add TAP tests for timeouts
- eeefd4280f6e 17.0 landed
Remove flaky isolation tests for timeouts
- a661bf7b0f56 17.0 landed
Followup fixes for transaction_timeout
- bf82f43790a6 17.0 landed
Introduce transaction_timeout
- 51efe38cb92f 17.0 landed
On systems that have setsid(2) (which should be just about everything except
- 3ad0728c817b 8.2.0 cited