Re: More vacuum stats

Поиск
Список
Период
Сортировка
От Tom Lane
Тема Re: More vacuum stats
Дата
Msg-id 16962.1282493851@sss.pgh.pa.us
обсуждение исходный текст
Ответ на Re: More vacuum stats  (Magnus Hagander <magnus@hagander.net>)
Ответы Re: More vacuum stats
Список pgsql-hackers
Magnus Hagander <magnus@hagander.net> writes:
> On Sun, Aug 22, 2010 at 17:29, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>> So I'd like to see a positive argument why this is important for users
>> to know, rather than merely "we should expose every conceivable detail
>> by default". �Why wouldn't a user care more about last AV time for a
>> specific table, which we already do expose?

> You need to connect to every database to do that. If you have many
> databases, that's a lot of overhead particularly if you're doing tihs
> for regular monitoring. Plus, those views will only track when
> autovacuum actually *did* something.

Well, the last-launch-time doesn't prove that autovacuum actually *did*
something ;-).

> Being able to see that autovacuum hasn't even touched a database for
> too long would be an early-indicator that you have some issues with
> it.

With the current AV launch algorithm, unless you have very serious
system-wide issues there will be a worker launched into each database
approximately every autovacuum_naptime seconds.  AFAICS this does not
tell you anything interesting about whether AV is getting its work done.
        regards, tom lane


В списке pgsql-hackers по дате отправления:

Предыдущее
От: Magnus Hagander
Дата:
Сообщение: Re: More vacuum stats
Следующее
От: Bruce Momjian
Дата:
Сообщение: Re: Return of the Solaris vacuum polling problem -- anyone remember this?