home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!haven.umd.edu!decuac!pa.dec.com!vixie
- From: vixie@pa.dec.com (Paul A Vixie)
- Newsgroups: comp.protocols.tcp-ip.domains
- Subject: Re: Why no MX records for .BITNET hosts?
- Date: 26 Dec 92 03:12:09
- Organization: DEC Network Software Lab
- Lines: 25
- Distribution: world
- Message-ID: <VIXIE.92Dec26031209@cognition.pa.dec.com>
- References: <NELSON.92Dec25002433@cheetah.clarkson.edu>
- NNTP-Posting-Host: cognition.pa.dec.com
- In-reply-to: nelson@sun.soe.clarkson.edu's message of 25 Dec 92 05:24:33 GMT
-
- > So why not have MX records whose preference value is based on the number
- > of hops from the gateway machines?
-
- Name service is not quite deterministic enough to create a fake domain
- unless you have a screening IP router between you and the core and are
- willing to run with a limited set of "forwarders". Since the NIC has
- been historically unwilling to create real domains for .BITNET (and .UUCP
- for that matter), this would have to be done as a fake domain.
-
- We do it here, of course :-). uucp-gw-1.pa.dec.com's name server believes
- that it is primary for .BITNET; the data is pulled over from BITSRV once
- every second blue moon. I bolt on a hack SOA and off it goes. This
- greatly simplifies connecting to .BITNET hosts since many of them are
- dual-hosted (IP and BITNET) or have dual-hosted close neighbors -- it's
- nice to use these hack MX's to force the mail to a host or its close
- neighbor rather than gating all of it through a single relay.
-
- If the new GSI NIC is willing to entertain a .BITNET domain, this could
- be made a lot easier on everyone. Likewise a .UUCP domain. I will
- personally guarantee primary and secondary nameservers for each of them.
- --
- Paul Vixie, DEC Network Systems Lab
- Palo Alto, California, USA "Don't be a rebel, or a conformist;
- <vixie@pa.dec.com> decwrl!vixie they're the same thing, anyway. Find
- <paul@vix.com> vixie!paul your own path, and stay on it." -me
-