Horton

Re: Race condition in TransactionIdIsInProgress

Alvaro Herrera <alvherre@alvh.no-ip.org>

View thread View on postgr.es
From: Alvaro Herrera <alvherre@alvh.no-ip.org>
To: Andrey Borodin <x4mmm@yandex-team.ru>
Cc: Konstantin Knizhnik <knizhnik@garret.ru>, PostgreSQL Hackers <pgsql-hackers@postgresql.org>
Date: 2022-02-10T19:33:15Z
Lists: pgsql-hackers 
On 2022-Feb-10, Andrey Borodin wrote:

> > On 10 Feb 2022, at 21:46, Konstantin Knizhnik <knizhnik@garret.ru> wrote:
> > As a result "t_xmin %u is uncommitted in tuple (%u,%u) to be updated in table"
> > error is reported.
> 
> Wow, cool, that seem to be a solution to one more mysterious
> corruption thread - “reporting TID/table with corruption error" [0].
> Thank you!

Ooh, great find.  I'll have a look at this soon.  Thanks for CCing me.

-- 
Álvaro Herrera              Valdivia, Chile  —  https://www.EnterpriseDB.com/
"I love the Postgres community. It's all about doing things _properly_. :-)"
(David Garamond)

Commits

  1. Fix visibility check when XID is committed in CLOG but not in procarray.

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

  3. Make TransactionIdIsInProgress check transam.c's single-item XID status cache