Re: [HACKERS] Replication/backup defaults

Поиск
Список
Период
Сортировка
От Peter Eisentraut
Тема Re: [HACKERS] Replication/backup defaults
Дата
Msg-id 1f24df97-fbde-5319-6782-f0c5fcbd561f@2ndquadrant.com
обсуждение исходный текст
Ответ на Re: [HACKERS] Replication/backup defaults  (Peter Eisentraut <peter.eisentraut@2ndquadrant.com>)
Ответы Re: [HACKERS] Replication/backup defaults  (Andres Freund <andres@anarazel.de>)
Список pgsql-hackers
On 1/4/17 2:44 PM, Peter Eisentraut wrote:
> On 1/4/17 9:46 AM, Magnus Hagander wrote:
>>     How about we default max_replication_slots to -1, which means to use the
>>     same value as max_wal_senders?
> 
>> But you don't necessarily want to adjust them together, do you? They are
>> both capped by max_connections, but I don't think they have any other
>> direct relation between each other? 
> 
> I think the most usual case is that you use approximately one
> replication slot per wal sender slot.  So it would be a good default to
> make them equal.

Well, let's worry about that later.

I think everyone is now in agreement with your original change proposal.

My suggestion would be to make the defaults of max_wal_senders and
max_replication_slots the same, so we don't open an opportunity for
ongoing discussions about why they are different and how different they
should be.  Ideally, we would make max_replication_slots go away at some
point similar to my suggestion above.

I also suggest making the defaults for both 20 instead of 10.  That
leaves enough room that almost nobody ever has to change them, whereas
10 can be a bit tight for some not-outrageous installations (8 standbys
plus backup?).

Your patch looks OK, documentation changes look good.  I wouldn't be
surprised if we found another place or two that will need updating, but
that is not a big deal.

-- 
Peter Eisentraut              http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services



В списке pgsql-hackers по дате отправления:

Предыдущее
От: Peter Eisentraut
Дата:
Сообщение: Re: pg_authid.rolpassword format (was Re: [HACKERS] Passwordidentifiers, protocol aging and SCRAM protocol)
Следующее
От: Alvaro Herrera
Дата:
Сообщение: Re: [HACKERS] ALTER TABLE .. ALTER COLUMN .. ERROR: attribute .. haswrong type