- NANOG - lists.nanog.org

Internet Traffic Data
by Robert Olshansky 28 Apr '97

28 Apr '97

Hi, K Claffy at LANR suggested I contactb you. At GTE Labs we are working on design of IP backbone facilities for ADSL access services. One of the challenges is to be able to project traffic growth per ADSL useer for the next 3-5 years. Can you direct me to sources of the followinmg: (1) Measured internet traffic statitics at NAPs for the past two years. (Info available at NAP Web sites that I have found is only for the last 5-30 days.) (2) Some measurements of growth of traffic/user over the past two years. So, for a fixed or known user population ( eg a Univ campus or an ISP's modem pool) how has traffic per user grown? (3) Anything else that would be useful. So I would greatly appreciate it if you could direct me to some source. TX Bob *************************** Bob Olshansky Manager, Advanced Service Platforms Service Research Laboratory GTE Laboratories Inc. 40 Sylvan Road Waltham, MA 02254 Tel 617-466-2565 Fax 617-466-2650 ****************************

2 1

Wow, AS7007!
by Stephen A Misel 28 Apr '97

28 Apr '97

I happened to be in one of our 7505 routers this afternoon when POP -- all of a sudden most of the internet disappeared! I immediately thought it was me, but looked around and saw this AS7007 broadcasting MY routes! It wasn't for all of our network space -- We have several /18's here, and it seemed only the first /24 of each CIDR was affected. When I found a workstation at the end of the /18, we got the whois info for 7007 -- Florida Internet Exchange, and called them. They claimed to have a customer broadcasting some bad routing information and unplugged their router. A few moments later, the internet stabilized and I started seeing real routes. Correct me if I'm wrong, but: (1) We're going to read about this in EVERY computer magazine, newspaper and TV as "the end of the internet?" (2) Access lists by backbone providers *should* have prevented this. (3) Does or does not the RADB and other routing registries (MCI's, etc) prevent this? I bet this hole will be patched up real soon! Steve

4 3

Re: Ascend GRF400
by garyz＠savvis.com 28 Apr '97

28 Apr '97

Kent, Savvis only has the GRF1600 in productions, we have ordered a couple of the GRF400 for our test bed. The GRF1600 was the first and only router from NetStar, but NetStar already had planned the GRF400 b4 the Ascend days..... Gary Zimmerman V.P. of Network Engineering Savvis Communications Corp. email: garyz(a)savvis.com http://www.savvis.com Office: 314.719.2423 Address: 7777 Bonhomme Suite 1000 St. Louis, MO 63105 ---------- > From: Neil J. McRae <neil(a)domino.org> > To: Kent W. England <kwe(a)6SigmaNets.com> > Cc: salo(a)msc.edu; nanog(a)merit.edu > Subject: Re: Ascend GRF400 > Date: Monday, April 28, 1997 2:32 AM > > > Correct me if I am wrong, but I thought the GRF-400 was an Ascend product > > and the GRF-1600 is the NetStar GigaRouter, with some additional developments. > > > > The GRF-400 is out there in networks like AGIS, NetRail, PSI, DOMINO and > > Savvis. No word on any NetStar-cum-GRF1600 deployments. > > s/DOMINO/COLT. > > > > > > Am I right? > > > Look at www.ascend.com. > > Neil. > > -- > Neil J. McRae - Alive and Kicking. D O M I N O > neil(a)DOMINO.ORG NetBSD/sparc - 100% SpF (Solaris protection Factor) > Free the daemon in your <A HREF="http://www.NetBSD.ORG/">Computer!</A>

1 0

Re: Ascend GRF400
by salo＠msc.edu 28 Apr '97

28 Apr '97

> Date: Thu, 24 Apr 1997 06:12:43 -0400 (EDT) > From: Joe Provo - Network Architect <jprovo(a)ma.ultranet.com> > To: neil(a)domino.org > Subject: Re: Ascend GRF400 > Cc: nanog(a)merit.edu > [...] > - the grf is not all of ascend. maybe they finally got things > right? > [...] It is probably useful to differentiate between Netstar, which was recently acquired by Ascend, and the rest of Ascend. The GigaRouter was developed by Netstar prior to its acquisition by Ascend. -tjs

6 6

routes to hit 50,000 soon? or filter AS 7007....
by Christian Nielsen 28 Apr '97

28 Apr '97

sorry if someone thinks this is the wrong message for the group, but it does effect us all... we are seeing lots of routes coming in sho ip bgp sum BGP table version is 6699646, main routing table version 6699646 49793 network entries (94886/99644 paths) using 8897576 bytes of memory 11894 BGP path attribute entries using 1541824 bytes of memory but the routes seem to be coming from AS 7007... They are advertising many routes, even our own. Sorry, but email and phones to them isn't working. *> 156.5.0.0 137.39.167.29 50 701 701 701 1239 1790 7007 ? * 208.131.13.29 0 5650 6176 1239 1790 7007 ? *> 167.32.0.0 137.39.167.29 50 701 701 701 1239 1790 7007 ? * 208.131.13.29 50 5650 3561 3561 1239 1790 7007 ? *> 192.56.218.0 137.39.167.29 0 701 701 701 3561 194 3909 1 1239 17 90 7007 ? *> 192.72.0.0 137.39.167.29 50 701 701 701 1239 1790 7007 ? * 208.131.13.29 50 5650 3561 3561 1239 1790 7007 ? *> 192.73.26.0 137.39.167.29 50 701 701 701 1239 1790 7007 ? * 208.131.13.29 50 5650 3561 3561 1239 1790 7007 ? *> 192.73.28.0 137.39.167.29 50 701 701 701 1239 1790 7007 ? * 208.131.13.29 50 5650 3561 3561 1239 1790 7007 ? *> 192.76.170.0 137.39.167.29 50 701 701 701 1239 1790 7007 ? * 208.131.13.29 0 5650 6176 1239 1790 7007 ? * 192.83.0.0 208.131.13.29 0 5650 6176 1239 1790 7007 ? *> 137.39.167.29 50 701 701 701 1239 1790 7007 ? * 192.83.64.0 208.131.13.29 0 5650 6176 1239 1790 7007 ? whois 156.5.0.0 Unilever (NET-UNILEVER2) Sprint International Christian

7 7

"Which Internet?"
by Robert E. Seastrom 28 Apr '97

28 Apr '97

Whilst investigating why my routing table size suddenly jumped by 5k routes, I discovered that Advantis (AS 2685 at MAE-E, origin AS 3082) was leaking all sorts of more-specifics for various networks into the bgp mesh. Yeah, I know I should be running bogon filters... Anyway, I called up their NOC to complain, and when I explained the situation I was asked which Internet I was calling about. Then when I asked to be transferred to a supervisor, the "support" person became argumentative and didn't want to transfer me. I had to be rudely direct and demand to be transferred before I was put on hold for five minutes and then informed that the supervisor was out to dinner. Come on Advantis, get it together. ---Rob

2 2

MAI.net filters.
by Jason Swain 28 Apr '97

28 Apr '97

For those of you following the "incident" on Friday, I have now implemented ip-based filters for MAI.NET and their downstream neighbors. This will prevent MAI or their downstream neighbors from advertising routes other than their own. As well, Vince from MAI has made considerable effort (along with Jack Dolan from Bay Networks) to add access-lists on their router to ensure if there is a problem, it doesn't propagate outside their network. Keep on keepin' on, Alex Sprintlink

1 0

Correction in Thanks
by Vincent J. Bono 28 Apr '97

28 Apr '97

Just to make small correction to my Apology post of yesterday, it was not Net Access Corporation, but Net Access of Philadelphia that offered assistance, aka Avi Freedman. I didn't realize that there was a "Net Access the Lesser". -Vincent

1 0

Re: NSI SAYS FCC SHOULD ASSUME INTERNET REGISTRATION FUNCTIONS
by Joseph T. Klein 27 Apr '97

27 Apr '97

Keep the agencies of the United States government out of this. This sounds like NSI is trying to wiggle out of participation in CORE. NSI does not like a solution that provides no guarantee that NSI will be one of the new registries. This is Washington politics - SNAFU and FUD. Operators, read the documentation and decide for yourself. http://www.iahc.org/ The IETF COREdb working group is defining a technical solution to "too much bureaucracy." Who thinks the FCC can be less burocratic than a shared database? Do I see any hands? If you agree with the IAHC solution then show your support and advocate that your company sign the MoU. --- On Mon, 21 Apr 1997 23:50:18 -0700 "Charles R. Hoynowski" <charles(a)etak.com> wrote: > NSI SAYS FCC SHOULD ASSUME INTERNET REGISTRATION FUNCTIONS > Network Solutions Inc., which currently registers all top-level domain > names under contract to the National Science Foundation, has suggested that > the Federal Communications Commission temporarily assume that function > until an international legal authority can be created to manage the system. > The transition period would allow for public comment on the plan in order > to incorporate any new processes or structures deemed necessary. The plan > is in contrast to an earlier proposal announced by the Internet > International Ad Hoc Commission to create seven new shard generic top-level > domains to be administered by 28 new registrars. NSI's president says the > IAHC plan risks Internet instability, creates "too much bureaucracy," and > will contribute to increased domain name legal disputes. (BNA Daily Report > for Executives 16 Apr 97) ---------------End of Original Message----------------- -- From: Joseph T. Klein, Titania Corporation http://www.titania.net E-mail: jtk(a)titania.net Sent: 10:07:06 CST/CDT 04/22/97 If the Internet stumbles, it will not be because we lack for technology, vision, or motivation. It will be because we cannot set a direction and march collectively into the future. -- http://info.isoc.org/internet-history/#Future

7 7

Network Solutions Dropped as Registrar Of Internet Domains
by Aleph One 27 Apr '97

27 Apr '97

Network Solutions Dropped as Registrar Of Internet Domains By David S. Hilzenrath Washington Post Staff Writer Thursday, April 24 1997; Page E01 The Washington Post The National Science Foundation said yesterday it will not renew Network Solutions Inc.'s deal to register Internet domains when the Herndon company's agreement with the government expires next year. But Network Solutions responded by saying it does not plan to relinquish its exclusive hold on the registration of ".com" domains and several other established forms of Internet addresses. The two announcements leave the future of one of the Internet's central resources in confusion and contention. Domain names serve as something of a Zip-code system for the Internet, enabling users to address electronic mail and locate pages on the graphical World Wide Web. Corporations and organizations pay to register their locations. A coalition of organizations has been pushing for competition in the registration of Internet domains, including those ending in .com for commercial users, the most popular form of Internet address. Critics have argued that competition could promote better service and lower prices. As of Dec. 31, Network Solutions had collected $42.6 million in fees from the enterprise, charging $100 to register new addresses ending in .com, .org, .gov, .net and .edu. It also charges annual renewal fees of $50 after two years. Seventy percent of the money is revenue to Network Solutions, a subsidiary of California-based federal government contractor Science Applications International Corp., and the rest is set aside for improvement of the Internet. The NSF inspector general recently estimated that registrations would generate annual fees of about $200 million by mid-1999. National Science Foundation spokeswoman Beth Gaston said it "has not been determined yet" whether the registry Network Solutions operates belongs to the company. The agency put Network Solutions in charge of the registry in 1993, and it has grown to include about 1.2 million domains. Networks Solutions seemed more definite. "It is not our intention to share .com or the others [domains] we register," Network Solutions spokesman Christopher Clough said. "Those would obviously [be] assets that we've developed . . . much as Microsoft wouldn't share DOS," its proprietary software. Network Solutions favors competition, but only in the registration of new types of domains, Clough said. A coalition of groups led by the Internet Society is trying to create a system in which competing registrars could process the same types of domains, including those controlled by Network Solutions. "They've taken the low road and tried to protect their monopoly instead of taking a leadership role in the best interest of the Internet," said Donald Heath, president of the Internet Society. The National Science Foundation's inspector general recently recommended that the federal government maintain oversight of Internet addresses and continue earmarking a portion of registration fees for development of the Internet. But Joseph Bordogna, the agency's acting deputy director, issued a statement yesterday saying the commercialization of the Internet leaves the NSF less reason to stay involved. Bordogna said the NSF "has no plans to renew or re-compete" the agreement with Network Solutions. A group of federal agencies, including the NSF, the Federal Communications Commission and the White House's Office of Science and Technology Policy, has been discussing what role the government should play. Copyright 1997 The Washington Post Company Aleph One / aleph1(a)dfw.net http://underground.org/ KeyID 1024/948FD6B5 Fingerprint EE C9 E8 AA CB AF 09 61 8C 39 EA 47 A8 6A B8 01

15 22