Re: Extreme bloating of intarray GiST indexes

Поиск
Список
Период
Сортировка
От Tom Lane
Тема Re: Extreme bloating of intarray GiST indexes
Дата
Msg-id 11677.1304399234@sss.pgh.pa.us
обсуждение исходный текст
Ответ на Re: Extreme bloating of intarray GiST indexes  (Josh Berkus <josh@agliodbs.com>)
Ответы Re: Extreme bloating of intarray GiST indexes
Список pgsql-hackers
Josh Berkus <josh@agliodbs.com> writes:
> Tom, Alexander,
> So we are using gist_intbig_ops, so that's not the issue.

> Using pgstattuple might be a bit of a challenge.  The client doesn't
> have it installed, and I can't pull it from Yum without also upgrading
> PostgreSQL, since Yum doesn't stock old versions AFAIK.

And updating Postgres to latest minor release is a bad thing why?
I can't believe you're not holding your client's feet to the fire
about running an old version, quite independently of the fact that
they need that contrib module.

But having said that, what you say makes no sense at all.  They have
intarray installed, so they have postgresql-contrib.  I know of no
Yum-accessible distributions in which intarray and pgstattuple wouldn't
be delivered in the same RPM.
        regards, tom lane


В списке pgsql-hackers по дате отправления:

Предыдущее
От: Vlad Arkhipov
Дата:
Сообщение: Re: Predicate locking
Следующее
От: "Johann 'Myrkraverk' Oskarsson"
Дата:
Сообщение: DLL export with mingw-w64: currently a no-op