allbery@ncoast.ORG (Brandon S. Allbery) (01/25/89)
I recently had the thought that if the lists of systems known to Internet
nameservers (and the paths, at least to the forwarders) was posted
occasionally, UUCP sites could then collect the information and add it to
the maps -- and pathalias could then properly deal with the Internet.
(Think of it as a more advanced version of "arpatxt".) Of course, this is
conditional, based on the assumption that it's possible to make such a list
in a rational way; not being all that familiar with nameservers and their
associated alphabet soup (you know: A, CNAME, MX, etc. ;-) I don't know if
it can be done. If it can, this would in effect provide a map of Internet
connections, so that pathalias can generate better paths to Internet sites
(e.g. it'll stop telling me that uunet is at hal!cwjcc!gatech!...!uunet when
hal!uunet.uu.net is faster (5 minutes to an hour, maximum!).
It might be preferred to leave off sites which are only on the Internet via
a forwarder (e.g. ncoast, or any of the .US sites), and RFC822 dictates that
some way be found to prevent mail paths from "hitching a ride" on the
Internet to travel between two non-Internet sites.
Any thoughts from the mail gurus?
++Brandon
--
Brandon S. Allbery, moderator of comp.sources.misc allbery@ncoast.org
uunet!hal.cwru.edu!ncoast!allbery ncoast!allbery@hal.cwru.edu
Send comp.sources.misc submissions to comp-sources-misc@<backbone>
NCoast Public Access UN*X - (216) 781-6201, 300/1200/2400 baud, login: makeuserpage@swan.ulowell.edu (Bob Page) (01/27/89)
If you used Erik Fair's 'mkglue' script recently posted you'll get what you want. ..Bob -- Bob Page, U of Lowell CS Dept. page@swan.ulowell.edu ulowell!page Have five nice days.
blarson@skat.usc.edu (Bob Larson) (01/27/89)
In article <13364@ncoast.ORG> allbery@ncoast.ORG (Brandon S. Allbery) writes: >I recently had the thought that if the lists of systems known to Internet >nameservers (and the paths, at least to the forwarders) was posted >occasionally, UUCP sites could then collect the information and add it to >the maps -- and pathalias could then properly deal with the Internet. >(Think of it as a more advanced version of "arpatxt".) The main problem with this idea is the problems that the Internet domain name servers were designed to fix reappear. A database of this size, with frequent changes being done from many locations, is to hard to keep up to date. Each entry now has its own exparation time, this allows anticipated changes to happen without any misuse of outdated information. There are also the minor problems of how to create and distribute the database. (The sri-nic hosts.txt file has over 6000 hosts listed, and USC has about 30 times as many hosts as are listed there!) -- Bob Larson Arpa: Blarson@Ecla.Usc.Edu blarson@skat.usc.edu Uucp: {sdcrdcf,cit-vax}!oberon!skat!blarson Prime mailing list: info-prime-request%ais1@ecla.usc.edu oberon!ais1!info-prime-request