Re: [PATCH] Provide more information to filter_prepare

Поиск
Список
Период
Сортировка
От Markus Wanner
Тема Re: [PATCH] Provide more information to filter_prepare
Дата
Msg-id 590e0d87-8a7d-9310-ff6d-b484f15522d0@enterprisedb.com
обсуждение исходный текст
Ответ на Re: [PATCH] Provide more information to filter_prepare  (Amit Kapila <amit.kapila16@gmail.com>)
Ответы Re: [PATCH] Provide more information to filter_prepare  (Amit Kapila <amit.kapila16@gmail.com>)
Список pgsql-hackers
On 29.03.21 08:23, Amit Kapila wrote:
> On Mon, Mar 29, 2021 at 11:42 AM Amit Kapila <amit.kapila16@gmail.com> wrote:
>> Why do you think that this callback can be invoked several times per
>> transaction? I think it could be called at most two times, once at
>> prepare time, then at commit or rollback time. So, I think using
>> 'multiple' instead of 'several' times is better.

Thank you for reviewing.

That's fine with me, I just wanted to provide an explanation for why the 
callback needs to be stable.  (I would not want to limit us in the docs 
to guarantee it is called only twice.  'multiple' sounds generic enough, 
I changed it to that word.)

> What exactly is the node identifier here? Is it a publisher or
> subscriber node id? We might want to be a bit more explicit here?

Good point.  I clarified this to speak of the origin node (given this is 
not necessarily the direct provider when using chained replication).

An updated patch is attached.

Regards

Markus

Вложения

В списке pgsql-hackers по дате отправления:

Предыдущее
От: Markus Wanner
Дата:
Сообщение: Re: [PATCH] add concurrent_abort callback for output plugin
Следующее
От: Markus Wanner
Дата:
Сообщение: Re: [PATCH] Provide more information to filter_prepare