Re: Keepalives win32

Поиск
Список
Период
Сортировка
От Kevin Grittner
Тема Re: Keepalives win32
Дата
Msg-id 4C2B1C7E0200002500032D81@gw.wicourts.gov
обсуждение исходный текст
Ответ на Re: Keepalives win32  (Magnus Hagander <magnus@hagander.net>)
Ответы Re: Keepalives win32  (Tom Lane <tgl@sss.pgh.pa.us>)
Список pgsql-hackers
Magnus Hagander <magnus@hagander.net> wrote:
> Windows default for idle is 2 hours, for interval 1 second.
And it defaults to five retries.  With these settings, you could
have a TCP connection break with as little as a five second network
outage, if it happened to come after two hours of silence on the
connection; although an outage of up to two hours could go totally
unnoticed.  The RFC values have a total of nine tries at 75 second
intervals, so for a single network outage to break a connection, it
would have to last at least ten minutes; but again, an outage of up
to two hours could occur before it started to check for problems.
I'm inclined toward option 2 (previously described on this thread),
because the Windows defaults are dumb.  Wait two hours and then test
for five seconds???
I also think we may want to suggest that for most environments,
people may want to change these settings to something more
aggressive, like a 30 to 120 second initial delay, with a 10 or 20
second retry interval.  The RFC defaults seem approximately right
for a TCP connection to a colony on the surface of the moon, where
besides the round trip latency of 2.5 seconds they might have to pay
by the byte.  In other words, it is *so* conservative that I have
trouble seeing it ever causing a problem compared to not having
keepalive enabled, but it will eventually clean things up.  In
practice people usually want something more aggressive.
-Kevin


В списке pgsql-hackers по дате отправления:

Предыдущее
От: Tom Lane
Дата:
Сообщение: Re: Keepalives win32
Следующее
От: Robert Haas
Дата:
Сообщение: Re: Check constraints on non-immutable keys