Re: Comments solicited: assignment of non-routed network numbers
I'm sorry, but I just don't get it. If the number isn't unique and can't be used and the network in question can NEVER attach to the Internet, why bother to get an officially assigned number at all? -Jeff Ogden Merit/MichNet ------- Forwarded message Received: from merit.edu by um.cc.umich.edu via MTS-Net; Wed, 26 Jan 94 20:48:32 EST Return-Path: <medin@nsipo.nasa.gov> Received: from dscs.arc.nasa.gov by merit.edu (4.1/1123-1.0) id AA12653; Wed, 26 Jan 94 20:46:11 EST Received: from localhost.arc.nasa.gov by dscs.arc.nasa.gov (4.1/1.5T) id AA00964; Wed, 26 Jan 94 17:46:09 PST Message-Id: <9401270146.AA00964@dscs.arc.nasa.gov> To: Vince Fuller <vaf@valinor.stanford.edu> Cc: regional-techs@merit.edu Subject: Re: Comments solicited: assignment of non-routed network numbers In-Reply-To: Your message of "Wed, 26 Jan 94 16:50:17 PST." <CMM.0.90.2.759631817.vaf@Valinor.Stanford.EDU> Date: Wed, 26 Jan 94 17:46:09 -0800 From: "Milo S. Medin" (NASA ARC NSI Office) <medin@nsipo.nasa.gov> AN RFC is being drafted to assign a "non-globally unique" network number out of class A,B, and C space, for the NIC to respond to these queries with. You should use one of them if it's a disconnected segment which will never require access. Note people really need to understand the meaning of that - including that they won't be able to talk to root nameservers, etc... Thanks, milo
Jeff: Did you check a workstation manual about how to configure an IP address for your machine? Last time I checked, the ones I looked at claimed that you should get an officially assigned network number. That's it. Now, explain the difference to them poor users. It would be better if the manuals would say "if you need an official number, get one. If not, use xyz." If xyz is well documented as a reusable number, then there should be a defensible situation not to route it later. Hans-Werner
Thats what 192.0.2.0 is. I understand that it was allocated for just such use... to be used in vendor documentation & as a reusable number. -- Regards, Bill Manning bmanning@rice.edu PO Box 1892 713-285-5415 713-527-6099 Houston, Texas R.U. (o-kome) 77251-1892
I'd also suggest net 10, may the ARPA net live forever! Note tha many of the examples by Cisco, et al use net 10. I think there should be exactly one each class A, B, and C. This makes it clearer to the people who are using such nets the they can not interoperate with the rest of the Internet. And we can more easily detect firewall leaks! Yes, the formal owner of net 10 should be approached. --MM--
I think there should be exactly one each class A, B, and C. This makes it clearer to the people who are using such nets the they can not interoperate with the rest of the Internet. And we can more easily detect firewall leaks!
Hmmm, but I thought we were moving toward classlessness? allen
Date: Thu, 27 Jan 94 09:24:28 -0500 From: mathis@pele.psc.edu Yes, the formal owner of net 10 should be approached. FYI, the formal "owner" of net 10 is IANA as it is presently not assigned (it got turned in :-). I'm pretty sure Jon and Joyce won't reassign it, though, without a lot of advance publicity :-). -MAP
The point is to have well-known numbers assigned, that can be configured into martian filters, and if anyone ever tries to add it to some routing filter, people will immediately recognize it as bogus. It's a human factors thing. thanks, Milo PS Sort of like 127.0.0.0 and 192.9.200 now. :-)
-------- ] From: "Milo S. Medin" (NASA ARC NSI Office) <medin@nsipo.nasa.gov> ] Subject: Re: Comments solicited: assignment of non-routed network numbers ] Date: Wed, 26 Jan 94 19:42:21 -0800 ] ] The point is to have well-known numbers assigned, that can be configured ] into martian filters, and if anyone ever tries to add it to some routing ] filter, people will immediately recognize it as bogus. It's a human factors ] thing. Likewise, if you let the people at each site just choose their "local" network number, they may find out 5 months later that they need to talk to hosts on the corresponding "real" network. By having agreed upon network(s), sites don't have to worry about a need for connectivity to the bona fide owner. /John
Milo S. Medin sez:
PS Sort of like 127.0.0.0 and 192.9.200 now. :-)
Whoops! Wrong number Milo. Thats 192.0.2.0 -- Regards, Bill Manning bmanning@rice.edu PO Box 1892 713-285-5415 713-527-6099 Houston, Texas R.U. (o-kome) 77251-1892
I thought 192.9.200 was also essentially that way. It's been the default network number you get when you hit <return> when SunOS is configuring it's network number... :-) thanks, Milo
Milo S. Medin
I thought 192.9.200 was also essentially that way. It's been the default network number you get when you hit <return> when SunOS is configuring it's network number... :-)
Nope, Sun turned that one in sometime back. Its been re-assigned. brazos[bmanning]: whois 192.9.200 Central Data Corporation (NET-CENDATA) 1602 Newton Drive Champaign, IL 61821 Netname: CENDATA Netnumber: 192.9.200.0 Coordinator: McGrath, Robert (RM524) cendata!cendata!mcgrath@uunet.uu.net (217) 359-8010 ext. 247 Record last updated on 10-Jun-91. -- Regards, Bill Manning bmanning@rice.edu PO Box 1892 713-285-5415 713-527-6099 Houston, Texas R.U. (o-kome) 77251-1892
bmanning@is.rice.edu (William Manning) writes * Milo S. Medin * > * > * > I thought 192.9.200 was also essentially that way. It's been the default * > network number you get when you hit <return> when SunOS is configuring * > it's network number... :-) * > * * Nope, Sun turned that one in sometime back. Its been re-assigned. My two cents in this discussion. Daniel Karrenberg and Yakov are drafting such an RFC. The implications are the bits that need to be well documented. The matter is not as simple as "not connecting to the Internet", because one should realize that one also cannot connect to another organisation doing the same thing with the same number. It basically assumes "no connectivity" to any other organisation .... Daniel will be at the regional techs next week, I think those involved and interested should take things up with him. -Marten
participants (9)
-
Allen Robel -
bmanning＠is.rice.edu -
hwb＠upeksa.sdsc.edu -
Jeff.Ogden＠um.cc.umich.edu -
John Curran -
Marten Terpstra -
mathis＠pele.psc.edu -
Michael A. Patton -
Milo S. Medin