Re: [HACKERS] Potential data loss of 2PC files

Поиск
Список
Период
Сортировка
От Michael Paquier
Тема Re: [HACKERS] Potential data loss of 2PC files
Дата
Msg-id CAB7nPqTo1hDJSSKUxTFPKubcwB+Uq4Ce_68DoyU+f0mcA=CX-g@mail.gmail.com
обсуждение исходный текст
Ответ на Re: [HACKERS] Potential data loss of 2PC files  (Ashutosh Bapat <ashutosh.bapat@enterprisedb.com>)
Ответы Re: [HACKERS] Potential data loss of 2PC files  (Ashutosh Bapat <ashutosh.bapat@enterprisedb.com>)
Список pgsql-hackers
On Tue, Jan 3, 2017 at 8:41 PM, Ashutosh Bapat
<ashutosh.bapat@enterprisedb.com> wrote:
> Are you talking about
>     /*
>      * Now we can mark ourselves as out of the commit critical section: a
>      * checkpoint starting after this will certainly see the gxact as a
>      * candidate for fsyncing.
>      */
>     MyPgXact->delayChkpt = false;
>
> That's while creating the file. I do not see similar code in
> FinishPreparedTransaction() where the 2PC file is removed.

RecordTransactionCommitPrepared() does the same work for COMMIT PREPARED.
-- 
Michael



В списке pgsql-hackers по дате отправления:

Предыдущее
От: Simon Riggs
Дата:
Сообщение: Re: [HACKERS] Measuring replay lag
Следующее
От: Heikki Linnakangas
Дата:
Сообщение: Re: pg_authid.rolpassword format (was Re: [HACKERS] Passwordidentifiers, protocol aging and SCRAM protocol)