Neohapsis is currently accepting applications for employment. For more information, please visit our website www.neohapsis.com or email firstname.lastname@example.org
Subject: RE: Co-located server
From: Nick Rout (nickrihug.co.nz)
Date: Tue Mar 21 2000 - 20:40:09 CST
- Next message: Lourdes Jones: "RE: Duplicate local deliveries via aliases..."
- Previous message: Dave Young: "Re: hello"
- Maybe in reply to: James Thompson: "Co-located server"
- Maybe reply: Nick Rout: "RE: Co-located server"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
The sending of outgoing mail is covered in the faq under "Running postfix on a dialup machine", see eg http://postfix.merit.edu/faq.html
The other bit is beyond me :-)
From: James Thompson [SMTP:jamestmath.ksu.edu]
Sent: Wednesday, 22 March 2000 13:27
Subject: Co-located server
I've been a happy postfix user since the first public beta, always using a
static IP connected 24/7 to the net. But now I'm attempting to do
something a little more involved and to be honest after reading the docs,
faq, and archives I don't know where to start.
I'm installing a co-located server at a local ISP since the company I work
for cannot get better than 56K dialup at the site of business. The ISP
will provide us with a static IP for the server at their location.
However the system at the site of business has to deal with a dynamic IP.
What I'd like to do.....
Outgoing mail would originate at the dynamic IP and relay out to the
static IP server at set intervals. Inter-company mail would simply be
delivered on site and not relayed to the static IP server. Mail comming
into the server at the ISP would simply be queued (no validation) until
the dynamic IP server contacts it and requests mail download.
The list archives suggest that UUCP over tcp/ip is the way to go. However,
I'm also looking into the possibilities of setting up a VPN between the
static IP and dynamic IP servers. If so then I'll end up with two static
IPs and that suggests ETRN as an option.
I've never used UUCP or ETRN and the archives suggested that each has it's
set of problems. I guess I'm asking which method would cause me the least
amount of grief in the long run. Or is there some other method I've
If I've missed something in the on-line docs, faq, or elsewhere on the web
please give me a URL and tell me to RTFM. Thanks
James Thompson 138 Cardwell Hall Manhattan, Ks 66506 785-532-0561
Kansas State University Department of Mathematics