Re: Permanent settings

Поиск
Список
Период
Сортировка
От Dawid Kuroczko
Тема Re: Permanent settings
Дата
Msg-id 758d5e7f0802201327j16b378fat7e9c5e76f5f73a58@mail.gmail.com
обсуждение исходный текст
Ответ на Re: Permanent settings  (Tom Lane <tgl@sss.pgh.pa.us>)
Список pgsql-hackers
On Wed, Feb 20, 2008 at 7:34 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Josh Berkus <josh@agliodbs.com> writes:
>
> > 2) allow *commenting* of pg_settings / SET PERMANENT.  Thus:
>
>  > SET PERMANENT work_mem = '12mb' COMMENT '16mb too high; OOM';
>
>  Ugh :-(
>
>  I think that putting this into SET is a pretty bad idea in any case.
>  SET is, and always has been, a session-local operation.  Providing a
>  secondary option that transforms it into something completely different

I think that's valid argument.

We already have ALTER USER foo SET bar = baz, so why not something like:

ALTER CLUSTER SET shared_buffers TO '2GB';

...perhaps with some other word than CLUSTER?

Regards,  Dawid


В списке pgsql-hackers по дате отправления:

Предыдущее
От: Andrew Dunstan
Дата:
Сообщение: Re: Why does pg_ctl.c have its own copy of PM_VERSIONSTR?
Следующее
От: Tom Lane
Дата:
Сообщение: Re: Why does pg_ctl.c have its own copy of PM_VERSIONSTR?