Обсуждение: The max value of autovacuum_vacuum/analyze_scale_factor.

Поиск
Список
Период
Сортировка

The max value of autovacuum_vacuum/analyze_scale_factor.

От
Masahiko Sawada
Дата:
Hi all,

Is there any reason why the max values of
autovacuum_vacuum/analyze_scale_factor are 100.0? These max values are
defined since when the parameters has been introduced but I think that
1.0 is enough.

Patch attached.

Regards,

--
Masahiko Sawada
NIPPON TELEGRAPH AND TELEPHONE CORPORATION
NTT Open Source Software Center

Вложения

Re: The max value of autovacuum_vacuum/analyze_scale_factor.

От
Ashutosh Bapat
Дата:
On Mon, Dec 5, 2016 at 4:44 PM, Masahiko Sawada <sawada.mshk@gmail.com> wrote:
> Hi all,
>
> Is there any reason why the max values of
> autovacuum_vacuum/analyze_scale_factor are 100.0? These max values are
> defined since when the parameters has been introduced but I think that
> 1.0 is enough.
>

Yes, at least from
https://www.postgresql.org/docs/devel/static/runtime-config-autovacuum.html,
it looks like this is supposed to be a "fraction of table size".
anything higher than 1.0 isn't a fraction. If at all any value > 1.0
has a meaning, I am wondering whether it's to account for bloat. But
then who would want to place the threashold in bloated area.

-- 
Best Wishes,
Ashutosh Bapat
EnterpriseDB Corporation
The Postgres Database Company



Re: The max value of autovacuum_vacuum/analyze_scale_factor.

От
Masahiko Sawada
Дата:
On Mon, Dec 5, 2016 at 9:15 PM, Ashutosh Bapat
<ashutosh.bapat@enterprisedb.com> wrote:
> On Mon, Dec 5, 2016 at 4:44 PM, Masahiko Sawada <sawada.mshk@gmail.com> wrote:
>> Hi all,
>>
>> Is there any reason why the max values of
>> autovacuum_vacuum/analyze_scale_factor are 100.0? These max values are
>> defined since when the parameters has been introduced but I think that
>> 1.0 is enough.
>>
>
> Yes, at least from
> https://www.postgresql.org/docs/devel/static/runtime-config-autovacuum.html,
> it looks like this is supposed to be a "fraction of table size".
> anything higher than 1.0 isn't a fraction. If at all any value > 1.0
> has a meaning, I am wondering whether it's to account for bloat. But
> then who would want to place the threashold in bloated area.
>

I understand that use case.
Thank you!

Regards,

--
Masahiko Sawada
NIPPON TELEGRAPH AND TELEPHONE CORPORATION
NTT Open Source Software Center