Re: Parallel scan with SubTransGetTopmostTransaction assert coredump

Поиск
Список
Период
Сортировка
От Pavel Borisov
Тема Re: Parallel scan with SubTransGetTopmostTransaction assert coredump
Дата
Msg-id CALT9ZEHh6AiK4=Dh1XwBfJ7Cj1ru4Uzpj892La9KDPcFzjXD0Q@mail.gmail.com
обсуждение исходный текст
Ответ на Re: Parallel scan with SubTransGetTopmostTransaction assert coredump  (Ranier Vilela <ranier.vf@gmail.com>)
Ответы Re: Parallel scan with SubTransGetTopmostTransaction assert coredump  (Ranier Vilela <ranier.vf@gmail.com>)
Список pgsql-hackers

Please, avoid using decimal based values.

128 is multiple of 64.

It's true that 128 is better to use than 120 but the main problem is not in the value but in the fact we never get 
CurrentRunningXacts->subxid_overflow = suboverflowed; with value more than 120. This solves the problem but it doesn't seem the right way to fix the issue. Instead it's better to process suboverflowed state which is legit itself  not resulting getting the crash on the Assert. So the discussion of "better" value doesn't seem related to the problem. It is for demonstration only.

--
Best regards,
Pavel Borisov

Postgres Professional: http://postgrespro.com

В списке pgsql-hackers по дате отправления:

Предыдущее
От: Robert Haas
Дата:
Сообщение: Re: Race condition in recovery?
Следующее
От: Aleksander Alekseev
Дата:
Сообщение: Re: Add ZSON extension to /contrib/