----- Original Message -----
To: "Hector Santos" <winserver(_dot_)support(_at_)winserver(_dot_)com>
Cc: "IETF-SMTP" <ietf-smtp(_at_)imc(_dot_)org>
Sent: Monday, January 05, 2004 4:01 AM
Subject: Re: RFC 2821 Address Resolution
If you were *really* smart, you recorded how long the first lookup took,
and use that info to decide if you fire the second DNS when you start 7,
when you start 8, or when you're N seconds from deciding that 8 is dead -
if the first query took 2 seconds, you can wait till 3-4 seconds
before you think you need to start 9, if the first query took 30
seconds, you need to fire the DNS lookups 35-40 seconds ahead.
I wonder what would be the overall gain with this. With the new resolver
expanding the list, it will need to be augmented with a system policy option
to define the number of maximum servers per session. I probably will give
sysops a default of 1/2 the amount. In my off the cuff view, if a system
has trouble sending mail by 1/2+1, well, that means to mean that the IP list
is faulty to being with, i.e., operational or scalability issue at the
remote end. What do you think about this? I agree with Richard, 80% of
the time, it is successful in the first few servers.
But you now made me wonder about it, the R&D hat in me will make me explore
I have to give credit for much of that idea to Thom O'Connor and
Motonori Nakamura, both of whom have spent a lot more time
analyzing the DNS issues of SMTP for scaling high-volume delivery
than I have.
bulk spammers are probably experts at it too <g>
..... In other words, if your mail is
going to AOL and Yahoo, waiting for both sets of DNS queries to
finish before you connect to anybody can kill your performance...
Right. No, our system is a 100% multi-threaded Windows system. The sysop can
define the maximum number of worker threads - highly optimized high end
server system. :-)
Thanks for your input Valdis.
Hector Santos, Santronics Software, Inc.