Re: Return of the Solaris vacuum polling problem -- anyone remember this?

Поиск
Список
Период
Сортировка
От Bruce Momjian
Тема Re: Return of the Solaris vacuum polling problem -- anyone remember this?
Дата
Msg-id 201008221651.o7MGplA28857@momjian.us
обсуждение исходный текст
Ответ на Re: Return of the Solaris vacuum polling problem -- anyone remember this?  (Tom Lane <tgl@sss.pgh.pa.us>)
Ответы Re: Return of the Solaris vacuum polling problem -- anyone remember this?  (Alvaro Herrera <alvherre@commandprompt.com>)
Список pgsql-hackers
Tom Lane wrote:
> Bruce Momjian <bruce@momjian.us> writes:
> > We often mention that we do vacuum freeze for anti-wraparound vacuum,
> > but not for pg_clog file removal, which is the primary trigger for
> > autovacuum vacuum freezing.  I have added the attached documentation
> > patch for autovacuum_freeze_max_age;  back-patched to 9.0.
> 
> This patch does not actually seem like an improvement.  The paragraph is
> all about transaction age, but you've inserted something entirely
> unrelated, and not only that but chosen to make the unrelated thing seem
> like the major consequence and anti-wraparound an afterthought.

Well, the reason that value is 200 million is for pg_clog cleanup, not
for xid wraparound protection.  The next sentence does relate to xid
wraparound, but it seems to fit because the previous sentence ends with
xid wraparound:
Note that the system will launch autovacuum processes toprevent wraparound even when autovacuum is otherwise disabled.

If we were worried about just xid wraparound I assume the value would be
2 billion.

Do you have a suggestion?  Reorder the items?

--  Bruce Momjian  <bruce@momjian.us>        http://momjian.us EnterpriseDB
http://enterprisedb.com
 + It's impossible for everything to be true. +


В списке pgsql-hackers по дате отправления:

Предыдущее
От: Tom Lane
Дата:
Сообщение: Re: More vacuum stats
Следующее
От: Peter Eisentraut
Дата:
Сообщение: Re: security label support, part.2