Neohapsis is currently accepting applications for employment. For more information, please visit our website www.neohapsis.com or email email@example.com
From: Wietse Venema (wietseporcupine.org)
Date: Wed Nov 24 2010 - 14:26:21 CST
> Thank you for the detailed explanation Wietse. Given the low mail
> volume of this MX (<2000 connections/day) would increasing max_idle from
> 100s to something like 5m or 10m be sane, to keep proxymap alive longer,
> thus decreasing the frequency of table parsing, and thus total time per
> query? Would increasing ipc_idle above the default of 5s be beneficial?
With 40s of space between connections on average, I see no need to
worry about CPU performance. Postfix daemons will wait up to 100s
of idle time by default, so even in your case they will likely
reuse the result of parsing the expensive table several times.
My objection was to your suggestion that a single postmap command
is representative of Postfix performance. Given that there is easily
a factor 100 difference in compile time versus query time, a single
postmap command is typical only for machines that handle so little
mail that the daemons terminate when they reach the idle time limit.
A discussion of Postfix CPU performance makes more sense to me in
the context of a performance-critical scenario, like a server that
actually handles a significant amount of mail.