Re: Sync Rep Design

Поиск
Список
Период
Сортировка
От Stefan Kaltenbrunner
Тема Re: Sync Rep Design
Дата
Msg-id 4D1DDF39.1030307@kaltenbrunner.cc
обсуждение исходный текст
Ответ на Re: Sync Rep Design  (Robert Haas <robertmhaas@gmail.com>)
Ответы Re: Sync Rep Design  (Robert Haas <robertmhaas@gmail.com>)
Список pgsql-hackers
On 12/31/2010 02:39 PM, Robert Haas wrote:
> On Fri, Dec 31, 2010 at 7:57 AM, Heikki Linnakangas
> <heikki.linnakangas@enterprisedb.com>  wrote:
>> On 31.12.2010 14:40, Robert Haas wrote:
>>>
>>> Someone may have proposed this before, but one way of getting standby
>>> naming "for free" would be to make the standby names the same as the
>>> roles used to log in, rather than adding a separate parameter.  We
>>> could just recommend to people that they use a separate, descriptive
>>> role name for each standby.  Then the synchronous_standbys parameter -
>>> when added - would mean "a standby from one of these roles".
>>
>> Seems a bit weird. It's not a lot of effort to give each standby a name. But
>> if you want something automatic, how about gethostname() ?
>
> Uh, wow, no way.  That would be making a critical aspect of system
> reliability depend on something way, way outside of our configuration.

+1

>
> What's weird about using the role name?  That's our standard way of
> distinguishing between two or more users.  Why invent something new?

wel a user is not a host/server for me - given there is no real benefit 
from using distinct roles for each standby yet I don't see why we should 
complicate the replication setup procedure by requiring 1 role per standby.
So I'm all for giving each standby a name but please make it an explicit 
one and not something that is only vaguely related to the actual standby 
host.


Stefan


В списке pgsql-hackers по дате отправления:

Предыдущее
От: Peter Eisentraut
Дата:
Сообщение: Re: and it's not a bunny rabbit, either
Следующее
От: Peter Eisentraut
Дата:
Сообщение: Re: and it's not a bunny rabbit, either