[Mailman-Developers] Some notes on my stuff...
Nigel Metheringham
Nigel.Metheringham@VData.co.uk
23 Jul 2001 17:05:17 +0100
On 22 Jul 2001 22:15:15 -0700, Chuq Von Rospach wrote:
> My testing showed the major processing delay was the speed of uptake by
> sendmail, primarily due to DNS delays. And given that you can't turn off DNS
> lookups without turning off a lot of the anti-spam stuff, you're kinda
> stuck.
I assume that applies to sendmail only - since there is no need to do
in-line resolving of either recipient or sender addresses for messages
generated from Mailman (since sender will be a local list related
address, and you can do the recipients later and report em back by
mail).
As for disk throughput... my favoured theoretical solution is a
journaled filesystem with journal on NVRAM - great for mail spools where
a lot traffic is transitory and you never need to actually commit those
journal transactions to the platters. [In theory all the stuff is in
place to do that with ext3 now... its just I don't have the impetous to
get the kit to test this config]
Nigel.