- Test - lists.nanog.org

/24s run amuck again
by Richard A. Steenbergen 08 Jun '01

08 Jun '01

We all know that /24s have run amuck. According to the APNIC bgp statistics (http://www.apnic.net/stats/bgp) view of the internet, the current prefix distribution looks: Routing Table Report June 08, 2001 BGP routing table entries examined: 106727 Number of prefixes announced per prefix length (Global) ------------------------------------------------------- /1:0 /2:0 /3:0 /4:0 /5:0 /6:0 /7:0 /8:20 /9:5 /10:6 /11:9 /12:32 /13:77 /14:207 /15:346 /16:6905 /17:1175 /18:2174 /19:6720 /20:5247 /21:4509 /22:6779 /23:8684 /24:61402 /25:278 /26:450 /27:463 /28:489 /29:374 /30:317 /31:1 /32:58 This mean over 57% of the routing table consists of /24s. So in order to facilitate getting this fixed, I have prepared a list of the worst offenders and their specific blocks which (most likely) have no reason for not being aggregated. If all of these entries could be aggregated (and I'm certain this simple script missed some), it would reduce the routing table by over 29,000 entries, to less the 78k. I think it's pretty obvious that a lot of these simply have no excuse. http://www.e-gerbil.net/ras/projects/ipaddr/24amuck.txt A list of ASNs with the largest sized offending blocks (not overall count, #1 is UU #2 is BBN #3 is Telstra nothing unexpected there) is: 2711 - Rock Hill Telephone Company 8112 - Bell Atlantic Internet Solutions 701 - UUNet Technologies, Inc. 11371 - Rhythms NetConnections 226 - Los Nettos 7029 - Alltel Information Services, Inc. 18687 - MGC Communications, Inc. 6595 - DoD Education Activity Network Assistance Center 1677 - ANS Communications 1785 - AppliedTheory Corporation 1326 - Advanced Networks & Services Inc. 6499 - Brooks Fiber Properties, Inc. I highly encourage anyone interested to yell at these people. -- Richard A Steenbergen <ras(a)e-gerbil.net> http://www.e-gerbil.net/ras PGP Key ID: 0x138EA177 (67 29 D7 BC E8 18 3E DA B2 46 B3 D8 14 36 FE B6)

3 2

Bell Labs' Discovery May Lead to Efficient Networks
by Irwin Lazar 08 Jun '01

08 Jun '01

FYI: Bell Labs recently conducted an analysis of the nature of Internet traffic. Some pretty interesting findings. http://Networking.ittoolbox.com/browse.asp?c=NetworkingNews&r=/news/dispnews .asp?i=44100 Through the use of sophisticated new software programs that analyzed and simulated data traffic in "unprecedented detail," Bell Labs researchers found that the "burstiness" seen in traffic at the edges of the Internet disappears at the core. The discovery that traffic on heavily loaded, high-capacity network links is unexpectedly regular may point the way to more efficient system and network designs with better performance at lower cost, Bell Labs said.

6 6

backup relationships possibly harmful
by Randy Bush 08 Jun '01

08 Jun '01

in the paper by gao, griffin, and rexford, it is shown that backup transit low-pref relationships as done today can lead to routing loops. it's worth a look just for that. while the authors propose a protocol approach, i wonder if there is an operational approach. randy --- From: Jennifer Rexford <jrex(a)research.att.com> To: randy(a)research.att.com Cc: griffin(a)research.att.com Subject: backup relationships Date: Wed, 6 Jun 2001 15:24:31 -0400 (EDT) Randy, Following up on our conversation this afternoon about backup routes and convergence problems, see Lixin Gao, Tim Griffin, and Jennifer Rexford, "Inherently safe backup routing with BGP," Proc. IEEE INFOCOM, April 2001. http://www.research.att.com/~jrex/papers/infocom01.ps and the slides at http://www.research.att.com/~jrex/talks/infocom01.ps http://www.research.att.com/~jrex/talks/infocom01.ppt The slides give a relatively theory-free overview. The intro of the paper presents the problem and the key results of the paper, and the body of the paper has some examples worked out and a fair amount of Greek to prove that our proposals handle the problem. The paper builds on earlier work by Tim on the stable paths problem (INFOCOM'00) and the non-convergence of some BGP configurations (SIGCOMM'99), and by Lixin and me on showing how policies based on peer-peer and customer-provider relationships can prevent these problems (SIGMETRICS'00). Backup routes throw a wrench into the work that Lixin and I did; this INFOCOM'01 paper proposes a way to deal with that problem. See the section on "global sanity of BGP routing" at http://www.research.att.com/~griffin/interdomain.html#divergepapers for the papers. -- Jen

2 1

Re: Why so little traffic from C&W
by George William Herbert 08 Jun '01

08 Jun '01

JD wrote: >Mikael Abrahamsson <swmike(a)swm.pp.se> wrote: >> I believe the whole structure of tier1, tier2 etc is breaking down and >> everybody is going after all customers, and that this will have >> interesting implications in the future. > > Breaking down? It used to be that anyone connected directly > to an exchange point was tier one, and the tiers are pretty > obvious beyond that. Now that everyone's at the exchanges, > "tier one" is simply a marketing term. To "an" exchange point? It certainly didn't start out that way; in the 93-94 timeframe it was pretty clear tier 1's brought their bandwidth around the country and connected wherever possible, not just at one point. I have to agree that the "structure" is breaking down, if nothing else because people don't agree with what the definitions are (though, a coherent case can be made that we never did... like peering policies spinning in the wind, definitions have changed a lot over the years). -george william herbert gherbert(a)retro.com

1 0

New mailing list: model-peer
by Sean Donelan 08 Jun '01

08 Jun '01

I tried sending this earlier, but I think the list server caught it as a command. So here goes again. I have set up a new mailing list for discussing topics releated to peering, and what should go into a model peering agreement between providers. This is partially related to the BOF Stephan Stuart held at the last NANOG and recent discussions on the NANOG mailing list. > Post message: model-peer(a)yahoogroups.com > Subscribe: model-peer-subscribe(a)yahoogroups.com > Unsubscribe: model-peer-unsubscribe(a)yahoogroups.com > List owner: model-peer-owner(a)yahoogroups.com

1 0

C&W Network
by Moe Allen 08 Jun '01

08 Jun '01

06.07.01 Well, after another great phone call with the Cable & Wireless staff, I have to agree with Sean. C&W is now informing their ISP's with T1 connections that C&W is going to be disconnecting the T1's that are currently being used and "Cut Over" to a much better C&W (New) network. C&W will setup a time that this is going to occur. As the conversation goes, this is going to be a much faster and better network. Since C&W provides us with our T1 backbone, I had to ask. What is going on? Is this (New) network, going to just be an ATM circuit or what? It appears that no one has an answer. They were just told to contact us or in our case this was a letter in our statement, which indicated we needed to contact C&W, ASAP. It went on to say, in so many words, failure to contact C&W could cause the "Cut Over" to occur, without notice. After what has occurred this week, it really makes you worry. What is C&W up to? And is this going to be good for my business? Or, we're going to get the shaft, as the song goes. Anyone else heard anything about this New and Improved, faster network that C&W is telling us about? Just a thought... Morris Allen VidcomNet, Inc.

4 3

Re: HTML-format postings
by Andy Bradford 08 Jun '01

08 Jun '01

Thus said "M. David Leonard" on Wed, 06 Jun 2001 14:10:05 EDT: > Am I the only one who objects to HTML-formatted postings on a > listserv? I thought there was a rule that postings should be in plain > text. Am I mistaken? You are not the only one who objects. Alas, it is getting harder and harder to convince people of the benefits of this. My general rule is that if it has a text/plain part along with the HTML then I might read it. If it is HTML-only without the accompanying text/plain then I simply discard the message without further wasting of time. :-) Andy -- [-----------[system uptime]--------------------------------------------] 8:19pm up 29 days, 22:57, 7 users, load average: 1.05, 1.11, 1.16

1 0

HTML-format postings
by M. David Leonard 08 Jun '01

08 Jun '01

Am I the only one who objects to HTML-formatted postings on a listserv? I thought there was a rule that postings should be in plain text. Am I mistaken? David Leonard ShaysNet Curmudgeon-at-large

13 15

Re: 95th Percentile again (was RE: C&W Peering Problem?)
by George Michaelson 07 Jun '01

07 Jun '01

I believe Geoff glosses one or two interesting cases. 1) the (semi) regulated market where an external body defines an exchange rate which can differ in each of A,B's {dis}favour, and requires A and B to settle inbound,outbound according to that tarrif. In that case, A and B can be required or expected to exchange differences between net A->B and B->A traffic for traffic in some set of {calls,time,initiater-end} Of course, A and B might also be expected to pay tax or other impost on gross amounts! 2) A and B form a regulated {combination,confederation} and seek to acquire leverage off C for exchange, arriving at some interior settlement which is not peering, but the entity D, which is some relationship of A and B has peering power against C, E, F ... What I'm trying to intrude is that this is not the 1980s and we are not in some Milton Friedman nightmare of completely unregulated exchanges between entitities. We might differ about the desirability of unregulated exchange between competing parties, but even in the international arena, albiet in a very ineffectual sense, other bodies dictate what happens. Like, forcing MCI to divest of some activity in forming Concert with BT (which fell apart) or like preventing the RBOC from being in the long distance game (which fell apart) or like going to the FCC to winge about international phone settlement rates (which is falling apart) The falling apart part is about political apathy as much as a realpolitik view of what will work. That these things fall apart doesn't mean inside the fine-grained 5 year window they can't work, or that better regulation might not shore them up for another interesting 5 year window. A lot of things happen in a 5 year window. Off to one side of this is another dimension. USA people may not have seen a story in the Australian press pointing out that the sale of C&W Optus to Singapore raises the prospect of the Australian national security agencies satellite communications channels being owned and operated by a non-resident entity. Of course its a beat-up, and from a not disinterested source, but it makes for interesting considerations in how entities might be constrained to offer communication services, and by implication settlements. cheers -George PS this is a personal mail and doesn't reflect APNIC policy views. -- George Michaelson | APNIC Email: ggm(a)apnic.net | PO Box 2131 Milton QLD 4064 Phone: +61 7 3367 0490 | Australia Fax: +61 7 3367 0482 | http://www.apnic.net

1 0

Re: HTML-format postings (Re: NANOG Digest V1 #748)
by Crist Clark 07 Jun '01

07 Jun '01

On the thread dealing with the evils of HTML posting, I just wanted to add that HTML and all MIME messages are both annoying and useless for those of us who receive the NANOG mailings in digest format. -- Crist J. Clark Network Security Engineer crist.clark(a)globalstar.com Globalstar, L.P. (408) 933-4387 FAX: (408) 933-4926 The information contained in this e-mail message is confidential, intended only for the use of the individual or entity named above. If the reader of this e-mail is not the intended recipient, or the employee or agent responsible to deliver it to the intended recipient, you are hereby notified that any review, dissemination, distribution or copying of this communication is strictly prohibited. If you have received this e-mail in error, please contact postmaster(a)globalstar.com

2 1