|
Neohapsis is currently accepting applications for employment. For more information, please visit our website www.neohapsis.com or email hr@neohapsis.com |
Subject: Re: Relaying Performance
From: Greg Stark (gsstark
mit.edu)Date: Sun Feb 06 2000 - 10:13:08 CST
- Next message: Greg Stark: "Re: Relaying Performance"
- Previous message: Wietse Venema: "Re: reject_unknown_sender_domain - still relevant?"
- Next in thread: Greg Stark: "Re: Relaying Performance"
- Maybe reply: Greg Stark: "Re: Relaying Performance"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
wietse
porcupine.org (Wietse Venema) writes:
> In a nutshell, it introduces a stiffer coupling between the receiving side
> and the delivering side. I hope to implement this solution shortly.
Would this mean there would be a configurable parameter controlling the ratio
of accepted messages to delivered messages? That would be very useful, if you
have multiple mail servers then having one that's bogged down for whatever
reason on the outgoing side start pushing back would mean the others would
pick up the slack (assuming the senders try another server).
It would also allow people to limit the size of the queue, if they set the
ratio to 1 then the queue should never grow. It might also be useful to have
an explicit setting for the queue size so if the queue grows too large new
connections are refused until mail is delivered.
Just to make it clear, I'm not talking about giving SMTP errors. I would like
to see an option to just delay giving the 250 responses to the mail or data
commands until the mail can be accepted. I suppose the more generally useful
method would be to give SMTP errors though.
-- greg
- Next message: Greg Stark: "Re: Relaying Performance"
- Previous message: Wietse Venema: "Re: reject_unknown_sender_domain - still relevant?"
- Next in thread: Greg Stark: "Re: Relaying Performance"
- Maybe reply: Greg Stark: "Re: Relaying Performance"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]