Re: Impact of vacuum full...

Поиск
Список
Период
Сортировка
От Erik Jones
Тема Re: Impact of vacuum full...
Дата
Msg-id 44C10380.20907@myemma.com
обсуждение исходный текст
Ответ на Re: Impact of vacuum full...  (Scott Marlowe <smarlowe@g2switchworks.com>)
Ответы Re: Impact of vacuum full...  (Bill Moran <wmoran@collaborativefusion.com>)
Re: Impact of vacuum full...  (Scott Marlowe <smarlowe@g2switchworks.com>)
Список pgsql-general
Scott Marlowe wrote:
> On Fri, 2006-07-21 at 10:13, Erik Jones wrote:
>
>> Hello, I was wondering if someone could enlighten me as to the impact to
>> the entire database of running VACUUM FULL against a single table.  The
>> reason I ask is that at company we work for we have a very large number
>> of queue type tables that fill up and empty out on a regular basis
>>
>
> HOLD ON! Do you empty them by doing something like
>
> delete from table
>
> with no where clause?
>
> If so, then try truncating the table.  That will clean it completely and
> reclaim all the dead space, plus it's faster than delete anyway.
>
> If that doesn't help, look at scheduling more aggressive plain vacuums
> (no just autovacuum, but cron job vacuum on specific tables that you
> know have a high turnover).
>
> Vacuum full is basically admitting your regular vacuum schedule isn't /
> can't be aggressive enough.
>
No!!!  The table is filled and entries are deleted one at a time, or in
groups, but definitely not all at once.    So, then what is the
difference between scheduling regular vacuum on specific tables v.
scheduling vacuum full on specific tables?  Basically, what I want to do
is to ensure that when I clean out a table row or rows at a time, the
space is immediately freed up.

--
erik jones <erik@myemma.com>
software development
emma(r)


В списке pgsql-general по дате отправления:

Предыдущее
От: Dan Strömberg
Дата:
Сообщение: Re: Column info without executing query
Следующее
От: Bill Moran
Дата:
Сообщение: Re: Impact of vacuum full...