August 1998 - Test - lists.nanog.org

Re: The Cidr Report
by Tony Li 17 Aug '98

17 Aug '98

> > | AS762 118 81 37 31.4% WELLFLEET-AS > > Oddly enough, they are announcing the aggregate *and* the specifics: > > > brd1>sho ip bgp reg _762_ > *> 192.32.0.0/16 207.106.96.5 0 4969 6461 1673 > *> 192.32.2.0 207.106.96.5 0 4969 6461 1673 > *> 192.32.3.0 207.106.96.5 0 4969 6461 1673 > *> 192.32.4.0 207.106.96.5 0 4969 6461 1673 > *> 192.32.6.0 207.106.96.5 0 4969 6461 1673 If one were paranoid, one might imagine that they are doing so intentionally to see how many Cisco routers they can tank through prefix overload. The problem with this is that a) they aren't also announcing the /17's, /18's, ... /25's and b) overloading a Cisco seems to mean that someone buys a bigger Cisco, thereby enhancing Cisco's revenue stream. One is then left with the theory that they are advertising more specifics for optimality, but haven't discovered communities yet. Exercise for the reader: If prefixes of length F and longer are filtered, and a domain has a prefix of length N, how many prefixes can they propagate into the backbone? How many before people start proxy aggregating them? Tony

1 0

Re: The Cidr Report
by Sean M. Doran 17 Aug '98

17 Aug '98

| 1) Gains by aggregating at the origin AS level | | --- 13Aug98 --- | ASnum NetsNow NetsCIDR NetGain % Gain Description | | AS3602 554 322 232 41.9% Sprint Canada Inc. | AS701 1133 933 200 17.7% Alternet | AS4293 442 255 187 42.3% IMCI | AS174 719 584 135 18.8% Performance Systems International | AS4200 192 100 92 47.9% AGIS (Apex Global Information Ser | AS2685 238 153 85 35.7% IBM Global Network - US | AS7046 229 163 66 28.8% UUNET-CUSTOMER | AS1239 535 487 48 9.0% SprintLink Backbone Sigh, bad trend. The big guys don't seem to care about aggregation any more. ob. NANOG: It's a *PLOT* to *CRUSH* the small providers, who are reliant on small and stable routing tables! Call the FTC!!!!!!! Funniest part: | AS762 118 81 37 31.4% WELLFLEET-AS Boy, *THEY* don't get it, that's for sure. Or maybe they want to be the ones to *SUPPLY THE BIGGER ROUTERS* to the poor small ISPs who will have to upgrade!! It's a PLOT! Sean.

2 1

Re: DNS Headaches.
by Mathias Koerber 17 Aug '98

17 Aug '98

-----BEGIN PGP SIGNED MESSAGE----- Why should only misspelt domains lookups go to him? If nameservers found the additional root-NSs during the time they were announced, they would have started to using them as much as all the other root-ns, at least until they had enough data to make the RTT decisions on which are preferred, no? - -----Original Message----- From: Ryan K. Brooks <ryan(a)inc.net> To: Craig A. Huegen <chuegen(a)quadrunner.com> Cc: max(a)inc.net <max(a)inc.net>; nanog(a)merit.edu <nanog(a)merit.edu> Date: Tuesday, August 18, 1998 4:20 AM Subject: Re: DNS Headaches. | | | | |Craig A. Huegen wrote: | |> This wouldn't be webteam.net, would it? =) |> |> This morning, Chris Yarnell from NASA pointed out the response |> to "dig @a.root-servers.net . soa": |> |> . 2D IN NS PANIC.WEBTEAM.NET. |> . 2D IN NS TORGO.WEBTEAM.NET. |> PANIC.WEBTEAM.NET. 2D IN A 207.67.50.8 |> TORGO.WEBTEAM.NET. 2D IN A 207.67.50.7 |> |> I don't know how this happened, but it doesn't look pretty. =) |> |> (The "standard" zone files were okay -- .COM, .NET, etc...) |> |> /cah | |How did this happen anyway? InterNIC? Postel? Doesn't this error imply that |a percentage of the Internet was unresolvable by the entire planet? | |It would really be great to find out what chain of events got us a root |nameserver on our network; not that we wouldn't mind donating the resources |to run one - it would just be nice to know in advance. :-) | | |Maybe we can get bilateral peering with BBN since we have a root server, | | |Ryan Brooks |ryan(a)inc.net |CTO, Internet Connect, Inc. | | | -----BEGIN PGP SIGNATURE----- Version: PGP 5.5.5 iQB1AwUBNdhMeVUjujkl4IK9AQGKdAL/YtGzleIfxiArvw7clyMDp/447uQg/FGB xkYr8JWdh+oxgR00zoParhGMtjK1PABf0UEIR3+fT/Us4PKJOIwJbqu1Ca/ZvSqP OSOcmxBSuwr8zXe1PNdzuT2lyvniVZZf =5cvS -----END PGP SIGNATURE-----

1 0

Re: BBN Peering issues
by owen＠DeLong.SJ.CA.US 17 Aug '98

17 Aug '98

Actually, I only put what I'd already seen arrive from NANOG in there. Since you made a public accusation, I made a public response. Owen > Owen - > > Is there any particular reason why you are quoting my private email > at the NANOG mailing list? > > Is this your personal or your company policy, or just 15 slips > of the finger? > > Sean. >

1 0

Re: BBN Peering issues (fwd)
by howland＠Priss.com 17 Aug '98

17 Aug '98

>From: Sean Donelan <SEAN(a)SDG.DRA.COM> >Where do you think senior corporate management comes from? Eggs.

3 2

Re: BBN Peering issues
by shields＠crosslink.net 17 Aug '98

17 Aug '98

In article <Pine.BSF.3.96.980815012720.27167B-100000(a)workhorse.iMach.com>, "Forrest W. Christian" <forrestc(a)iMach.com> wrote: > After all, it's the BBN customers REQUESTING the data from exodus, not the > other way around. This argument goes around and around every time. Unless you have a customer relationship with Exodus or BBN there is no point in arguing about what should happen with this particular case, and there is no point at all in arguing about what is the morally correct thing for the general case. The only way to determine the value of connecting network A to network B, in either sign or magnitude, is for A and B to negotiate a price, which may even be zero. And that's what's happening here, albeit in a unusually painful and visible way. If you have an opinion and are a customer of BBN or Exodus, tell them your opinion. Otherwise there is no need to care. -- Shields, CrossLink.

5 5

re: BBN Peering issues
by Jerry Scharf 17 Aug '98

17 Aug '98

so it comes down to the fact that because YOU can't imagine UUNET continuing to give Exodus free peering, it must not be true. I guess life is supposed to follow your beliefs. For the record, until a couple years ago, UUNET was one of the most cooperative ISPs for peering. jerry

2 1

Anyone in the DC SMDS cloud?
by Avi Freedman 17 Aug '98

17 Aug '98

I'm trying to turn up an SMDS T3 and my customer isn't ready yet. If anyone is willing to throw a static map in to test with me, I'd greatly appreciate it. Sorry for the noise... Avi Freedman Net Access

1 0

Re: BBN Peering issues (fwd)
by howland＠Priss.com 17 Aug '98

17 Aug '98

Erik, I couldn't agree more. BBN (GTEI) is only doing what Sprint started doing about 3 years ago, when some associates of mine tried to turn up peering with them. All these arguments about traffic flow were used then, by so-called "engineers" no less! One place I worked I had to answer trouble calls from people who couldn't get to our customers, and customers who couldn't get to specific sites, to a particular four-letter network I won't mention. Their customers were NOT glad to hear that the reason they couldn't get to our customers was because the four-letter network had refused to peer. But the status-quo has changed. Remember the charter of the FCC, to regulate "scarce resource" communications. What happens when some brain-dead congress-crawler starts getting calls from their constituents that "we bought internet service but couldn't get to xyz!" Some staffer takes a 30 second look into the problem, and decides that peering is *so* overwhemed that in order for everyone to reach everyone else the Government should step in and regulate it. The 'Net gives some measure of universal connectivity because it was driven at its root by engineering. Who on NANOG, with any real world experience in networking, denies that maximum open peering benefits everyone? This decision, and the arguments in favor of pay-per-peer, has nothing to do with engineering. It has to do with paper-pushers who wouldn't know the difference between peering and transit if it were on 500 power-point slides, because they look at MONEY. They see "bits per second", and who they came from, and conclude that those packets are not sourced on their network and therefore someone should be paying for them specifically. And, when the engineering breaks because a political answer is trying to be imposed on a technical problem, and Little Jonhy can't get to www.Bob.com for his school assignment, the cry will go up to "protect the children" and regulation will fall faster than Madame LaFarge ever thought possible. Then those same "pay-per-peer" morons will get appointed to the governing body of Internet Communications, just like the Railroad Barrons of the late 19th Century, and history will endlessly repeat, again. Thanks, Erik, for the pointer on the AT&T monopoly. Do you have a source? Curt- ----- Begin Included Message ----- >From: "Erik E. Fair" <fair(a)clock.org> Subject: Re: BBN Peering issues Sender: owner-nanog(a)merit.edu Content-Length: 1119 There is a customer perception, dating from the earliest days of the Internet that when you connect to the Internet, you will be able to reach all sites that are up, everywhere. That this is still mostly true is a tribute to the hard work of a lot people on this list, and elsewhere. So far, the cases for which this is not true are small in both number and relative importance. If this perception breaks down, watch out. Theodore Vail was allowed to create the regulated monopoly AT&T in the early part of this century on the promise of Universal Service, which meant not only that everyone had a telephone, but that *all* telephones could call *all* other telephones - one big, happy, PSTN. The Internet presents this kind of universality today without the regulation, but don't doubt for a second that if the ISPs (of whatever size) begin destructive pissing matches of the form "I'm bigger than you, pay me or we disconnect" that the FCC will be pressured to regulate the ISPs in such a way to guarantee the universal connectivity aspect of the Internet. Your customers will demand it. Erik <fair(a)clock.org> ----- End Included Message -----

2 2

re: BBN Peering issues
by Gordon Cook 17 Aug '98

17 Aug '98

John levine <johnl(a)iecc.com>: >>I will point out, briefly, that the issue is not 'free peering' versus paid >>peering. The issue seems to be the desire for BBN to have traffic come in >>its network indirectly (through another transit provider). > >Exodus has claimed several times on nanog, that although they peer >with many providers, they don't buy transit from anyone, and if GTE >won't accept their packets as a peer, they're not going to get their >packets through anyone else. > john, I am sorry but I am really surprised at the paucity of research done by those who are bashing BBN. People really ought to read http://www.edgar-online.com/quotecom/gdoc/?choice=2-885584&nad=0 there they will find exodus stating in january 98 to the SEC that: "While the Company began operations in 1992, it did not offer server hosting services until 1995 and did not open its first dedicated Internet Data Center until August 1996, at which time it refocused its business strategy on providing Internet system and network management solutions for enterprises' mission-critical Internet operations. As a result, the Company's business model is still in an emerging state." COOK: translation it became a web hosting company and as the out put of its large content providers increased to the point where a two to one imbalance in traffic between BBN and exodus is likely more like 10 packets or more into BBN for every one packet in the other direction. Levine: >Since Exodus hosts many of the largest and most popular web farms on >the net such as Geocities, a lot of us expect that BBN's customers >won't be pleased when they find that they've lost access to many of >their favorite web sites. > COOK: John, this has been a point of contention on the internet at least since 94, but please show me where the consenus has been reached that web hosting companies need to be interconnected to the internet for free. Later on the Exodus S1 admits (p10) S1: "In particular, the Company is dependent on WorldCom and certain other telecommunications providers for its backbone capacity and is therefore dependent on such companies to maintain the operational integrity of its backbone. In addition, the Company relies on a number of public peering interconnections and private peering interconnections to deliver its services. Many of the operators of the private peering interconnections are competitors of the Company. Currently, ***the Company does not pay a fee for many of these interconnections,*** and if these organizations were to begin to charge the Company for utilizing these interconnections, or, in the cases where the Company currently pays a fee, to increase the pricing associated with utilizing these networks, the Company may be required to identify alternative methods through which it can distribute its customers' content." COOK: What is not clear here is how many of the top 5, or 10 or 15 exodus has private peerings with and how many it PAYS for transit. It is clear from the above that Exodus PAYS - that it is NOT entirely transit free - despite the claims of Lawrence Delskii and others. what *IS* clear from the S1 is that Exodus is HIGHLY leveraged and that unexpected expenses could cause it grief. At any rate if exodus doesn't pay a fee for many interconnects it must pay a fee for some and, if it does, why should BBN, by giving exodus a FREE connection, subsidize its competitors who *ARE* charging exodus? levine: >BBN's argument about the imbalance in number of packets makes no sense >to me. It's as though my grocery store claimed that since the >groceries I bought weighed 20 pounds, while the check I gave them in >exchange weighed a fraction of an ounce, they're getting the short end >of the deal. Networks peer because the two sides both get similar >value from the connection, not because they get the same number of >packets. > COOK: no, not the same, but at least there generally is some degree of proportion. I would like to know how exodus can tell the SEC in january 98 that only some of its connections are free peering while others are purchased, and now in august whine that BBN is screwing it because *ONLY* BBN is breaking the peering relationship since exodus is a customer of NO ONE else. if exodus had gone from paid transit to free peering with one or more companies since the beginning of this year, one would think that exodus's ellen hancock would have figured out a way to let the world know. of course ellen ran networking for IBM in the late 80s and early 90s when IBM was left standing on the platform as the internet train left the station. then she was CTO at Apple while that company went deeper into hell. now she's CEO of a company that made a 69 million dollar IPO in march, issued 200 million in bonds in July, has just bought its OWN fiber national back bone from Qwest for an undisclosed sum and, when BBN stops giving it a free ride, (read multiple hundreds of thousands worth of free interconnections per year) whines and snivels over the whole internet. Gee: why am I not impressed with ellen's latest leadership? Exodus' idea of how to do PR puzzles me. But I am even more puzzled that no one seems who complains on these lists seems to think to read its 10Qs or S1. *************************************************************************** The COOK Report on Internet New Special Report: Building Internet 431 Greenway Ave, Ewing, NJ 08618 USA Infrastructure ($395) available. See (609) 882-2572 (phone & fax) http://www.cookreport.com/building.html cook(a)cookreport.com Index to 6 years of COOK Report, how to subscribe, exec summaries, special reports, gloss at http://www.cookreport.com ***************************************************************************

1 0