Re: Make query cancellation keys longer

Поиск
Список
Период
Сортировка
От Jelte Fennema-Nio
Тема Re: Make query cancellation keys longer
Дата
Msg-id CAGECzQS2+meELiT2ZwrTXvRDpnCyioPUT0Yp8htSfnaV03uykQ@mail.gmail.com
обсуждение исходный текст
Ответ на Re: Make query cancellation keys longer  (Peter Eisentraut <peter@eisentraut.org>)
Список pgsql-hackers
On Fri, 1 Mar 2024 at 15:19, Peter Eisentraut <peter@eisentraut.org> wrote:
> > One complication with this was that because we no longer know how long
> > the key should be, 4-bytes or something longer, until the backend has
> > performed the protocol negotiation, we cannot generate the key in the
> > postmaster before forking the process anymore.
>
> Maybe this would be easier if it's a protocol version number change,
> since that is sent earlier than protocol extensions?

Protocol version and protocol extensions are both sent in the
StartupMessage, so the same complication applies. (But I do agree that
a protocol version bump is more appropriate for this type of change)



В списке pgsql-hackers по дате отправления:

Предыдущее
От: Andy Fan
Дата:
Сообщение: Re: Shared detoast Datum proposal
Следующее
От: Andy Fan
Дата:
Сообщение: a wrong index choose when statistics is out of date