Re: Why is vacuum_freeze_min_age 100m?

От: Tom Lane
Тема: Re: Why is vacuum_freeze_min_age 100m?
Дата: ,
Msg-id: 1849.1250038484@sss.pgh.pa.us
(см: обсуждение, исходный текст)
Ответ на: Why is vacuum_freeze_min_age 100m?  (Josh Berkus)
Ответы: Re: Why is vacuum_freeze_min_age 100m?  ("Kevin Grittner")
Re: Why is vacuum_freeze_min_age 100m?  (Torsten Zühlsdorff)
Список: pgsql-performance

Скрыть дерево обсуждения

Why is vacuum_freeze_min_age 100m?  (Josh Berkus, )
 Re: Why is vacuum_freeze_min_age 100m?  (Josh Berkus, )
  Re: Why is vacuum_freeze_min_age 100m?  (Robert Haas, )
   Re: Why is vacuum_freeze_min_age 100m?  (Josh Berkus, )
    Re: Why is vacuum_freeze_min_age 100m?  (Robert Haas, )
     Re: Why is vacuum_freeze_min_age 100m?  (Josh Berkus, )
      Re: Why is vacuum_freeze_min_age 100m?  (Robert Haas, )
 Re: Why is vacuum_freeze_min_age 100m?  (Tom Lane, )
  Re: Why is vacuum_freeze_min_age 100m?  ("Kevin Grittner", )
   Re: Why is vacuum_freeze_min_age 100m?  (Tom Lane, )
    Re: Why is vacuum_freeze_min_age 100m?  ("Kevin Grittner", )
     Re: Why is vacuum_freeze_min_age 100m?  (Tom Lane, )
      Re: Why is vacuum_freeze_min_age 100m?  (Robert Haas, )
       Re: Why is vacuum_freeze_min_age 100m?  ("Kevin Grittner", )
  Re: Why is vacuum_freeze_min_age 100m?  (Torsten Zühlsdorff, )
 freezing tuples ( was: Why is vacuum_freeze_min_age 100m? )  (Jeff Davis, )
  Re: [HACKERS] freezing tuples ( was: Why is vacuum_freeze_min_age 100m? )  (Alvaro Herrera, )
   Re: [HACKERS] freezing tuples ( was: Why is vacuum_freeze_min_age100m? )  ("Kevin Grittner", )
   Re: Re: [HACKERS] freezing tuples ( was: Why is vacuum_freeze_min_age 100m? )  (Josh Berkus, )
  Re: [HACKERS] freezing tuples ( was: Why is vacuum_freeze_min_age 100m? )  (Robert Haas, )

Josh Berkus <> writes:
> I've just been tweaking some autovac settings for a large database, and
> came to wonder: why does vacuum_max_freeze_age default to such a high
> number?  What's the logic behind that?

(1) not destroying potentially useful forensic evidence too soon;
(2) there's not really much to be gained by reducing it.

            regards, tom lane


В списке pgsql-performance по дате сообщения:

От: Robert Haas
Дата:
Сообщение: Re: Why is vacuum_freeze_min_age 100m?
От: Nickolay
Дата:
Сообщение: transaction delays to apply