- Test - lists.nanog.org

Start ISPs as Registries
by Jim Fleming 25 Feb '97

25 Feb '97

On Tuesday, February 25, 1997 2:06 AM, Avi Freedman[SMTP:freedman@netaxs.com] wrote: @ > But now we are receiving many petitions from much smaller ISP's @ > asking for BGP connections. These small customers are going to announce @ > few networks (a couple of /24, or some /23). We have no problems to @ > configure our routers with more BGP peerings, as our routers can support @ > it, but I'm a little bit concerned about the impact of such small @ > announcements to the whole Internet. @ @ I think the general feeling is that if the customer is multi-homed, there's @ going to be another route announcement for them anyway - whether it's a @ /23, /24, or /16 or /17. @ In my opinion, ISPs start at the wrong end of the business. We should try to encourage future ISPs to start as registries BEFORE they buy a lot of modems, routers and other stuff. If the airlines ran like the Internet, they would encourage pilots to buy their own planes, to transport customers, and after getting enough hours, they would teach them about air traffic control, navigation, weather, radios, etc. Because there has been a lack of emphasis on growing the IP address registries, and domain registries by the InterNIC, people do not get an opportunity to train in those important functions first, before jumping into the cockpit. As a net result of this, it is hard to find ISPs who are growing into the registry business. I suppose some ISPs do not see that there is any money in that end of the business. I also feel that there is a lack of education in this critical area. If NANOG members have any influence on the direction the Internet takes, I suggest that they help new ISPs first learn the registry end of the business. One way to do this is to help establish IP address registries in your area and the training programs needed to allow people to handle allocations as skilled professionals. There is not much capital required to launch someone as a registry, compared to the capital required to be an ISP. The key ingredient needed is education and NANOG seems like a natural group to help people get started. With the various charging models coming out of the ARIN discussions <http://www.arin.net>, there should be a lot of food for thought on how a few people in an office with some PCs, paper, pencils and telephones can become registries and support themselves. Once established as a registry, these skilled professionals could grow and branch out to add other services. Who knows, they could even become ISPs, and if they do, then they would have a firm foundation of registry training to support their growth and to better serve the Internet -- Jim Fleming Unir Corporation e-mail: JimFleming(a)unety.net JimFleming(a)unety.s0.g0 (EDNS/IPv8)

1 0

Re: BGP announcements and small providers
by Paul Ferguson 25 Feb '97

25 Feb '97

If the downstream customers are not multihomed, there's no reason to run BGP with them; a simple default/static will work just fine. Also, if you suballocate addresses to them from within your address space, you can aggregate them. This is generally considered a Good Thing. - paul At 08:06 AM 2/25/97 -0500, Avi Freedman wrote: > >I think the general feeling is that if the customer is multi-homed, there's >going to be another route announcement for them anyway - whether it's a >/23, /24, or /16 or /17. > >If the customer isn't multi-homed, and you're their only path, then: > >a) If you BGP with them and pass it on to the global 'net, that path > will flap if the line goes up or down. This is considered bad. > >b) There's no point, if that route is out of your address space, in > announcing the more specific if there's no extra path to them if > their connection to you goes down. >

4 3

BGP announcements and small providers
by Javier Gonzalez 25 Feb '97

25 Feb '97

Dear all, I'm writting to this list because I think this is a good forum for this question, as my questions relates to the amount of routing announcements that are sent to all Internet, and actually I'm a beginner on these points. But first let me introduce myself: I'm currently working for Telefonica, at Spain, giving internet access to spanish ISP's. We've got many customers here, and so the numbers of networks being announced by our AS is currently quite big, but this number is getting bigger and bigger as we are experiencing an exponential growing of the Internet market at Spain (due also to a special service called "Infovia" specially designed here to make Internet access through normal telephone lines much easier for final users). As you will likely imagine, I'm currently in charge of routing issuess at our network (which is called Ibernet). As part of this routing issues, I'm in charge of the BGP configuration, and always following CIDR rules for peerings. Up to now, we used to have, apart from the peering with our providers, BGP peering with a couple of customers; but these customers where big enough, and they were announcing several /17, /18 and /19. But now we are receiving many petitions from much smaller ISP's asking for BGP connections. These small customers are going to announce few networks (a couple of /24, or some /23). We have no problems to configure our routers with more BGP peerings, as our routers can support it, but I'm a little bit concerned about the impact of such small announcements to the whole Internet. I know there are some controversials about network announcements flapping on the Internet, and that there is something called "dampening", whose way of working I don't know, but I guess it is for avoiding flapping; and I think that people is also worried about the amount of networks that are being announced through Internet, and indeed that some providers (Sprint ?) are filtering high prefixes (maybe higher than /19, or at least they did ?). I don't know if other providers are now, or are going in the future, to establish such kind of policies, but I'm affraid that accepting BGP peerings with small provider that are to announce only /23 or /24 could create problems in the future, first for the own customer (who could have his announcements suppressed by other AS's) and second, to the whole Internet if all carriers are doing the same. So I would like to know what is the point of view from other people, with much more experience than us in these points, about accepting these small BGP peerings. Is it really a problem, or I am worrying about nothing ? Are there any specific rules (apart, of course, from the CIDR rules) that the routing administrators of carrier providers, like us, should follow all over the world in order to make the routing at Internet much easier, or everyone can do whatever he wants ? I would very much appreciate any comments on that (and sorry for the long of this e-mail :-) Best regards ----------------------------------------------------------------------------- Nombre/Name: Javier Gonzalez Vela. Empresa/Company: Telefonica Transmision de Datos // IBERNET E-mail: jgonz(a)ibernet.es Telefono/Phone: + 34 1 584 08 18 Fax: + 34 1 523 44 99 -----------------------------------------------------------------------------

2 1

demon.co.uk spam policies? [sorry if off-topic]
by Scott Call 25 Feb '97

25 Feb '97

In light of the recent spam/nospam discussions on the list, I thought I'd share something one of my customers got with the rest of you: >From info(a)senora.demon.co.uk Fri Feb 21 17:29:07 1997 Received: from home.glasscity.net (home.glasscity.net [208.13.0.1]) by ccnet4.ccnet.com (8.7.1/8.6.12) with SMTP id RAA05294; Fri, 21 Feb 1997 17:29:06 -0800 (PST) From: info(a)senora.demon.co.uk Received: from home.glasscity.net (194.72.198.24) by home.glasscity.net (EMWAC SMTPRS 0.81) with SMTP id <B0000772250(a)home.glasscity.net>; Fri, 21 Feb 1997 20:28:01 -0500 Date: Fri, 21 Feb 1997 20:28:01 -0500 Message-ID: <B0000772250(a)home.glasscity.net> Subject: Dear spam hater. X-UIDL: 23485792837498720384760912734067 Priority: normal X-Flags: PXU2-4R65-2264:R Apparently-To: <anon(a)anon.net> X-Status: Status: RO HA HA HA HA You are on a published list of peolple who hate commercial mail! How STUPID can you get! Commercial email is just taking off - and I'm going to distribute this list everywhere I can to shut you whining people with nothing better to do than complain up! It's BASTARDS like you who got me shut down last time - but demon won't touch me! HA HA HA! I can go thru DIFFERENT servers! ANYTIME........... Alan -- Alan K Baker (MailWurliTzer Willy) www.senora.demon.co.uk alan(a)senora.demon.co.uk -- ___ / ~-. Scott Call | \ TCP/IP Network Administrator/Holder of Past Knowledge \\j`j; / , CCnet, Inc http://www.ccnet.com/ (^, )bxr~ Walnut Creek, CA ---' http://www.devolution.com/ 46. Outrageous vehicles only make the hero cooler. 101. Mallets can be stored anywhere on anybody. -All I Need to Know I Learned From Anime

6 5

policy
by Tim Keanini 25 Feb '97

25 Feb '97

I am in search of a real live Co-Location Policy for an Exchange point. Could someone send me a pointer please. Thanks in advance, -Tim Keanini

1 0

Developing a broader view...
by Jim Fleming 25 Feb '97

25 Feb '97

Many NANOG members have been around longer than most. If any NANOG members can help fill in the blanks on this time-table, it might help your members develop a broader view. JF @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ March 1992 Original NSF Solicitation <http://rs.internic.net/nsf/solicitation.html> May 1992 Original NSI Proposal October 1992 Modified NSI Proposal <http://rs.internic.net/nsf/nis/proposal-toc.html> "Network Solutions believes NSF's objectives will be met most effectively by the award of the bulk of the services to a single contractor." <http://rs.internic.net/nsf/nis/sectionM.html> "Network Solutions proposes Mr. Jon Postel as the IANA Manager and Chairman of the Advisory Panel for the NREN NIS Manager project. He will provide services as an employee of USC's Information Sciences Institute (ISI), subcontractor to Network Solutions." AT&T Proposal <http://ds1.internic.net/internic.info/proposal/> January 1, 1993 Cooperative Agreements General Atomics (GA) ??? AT&T ??? Network Solutions, Inc. <http://rs.internic.net/nsf/agreement/> Estimated Total Amount: $4,219,339 Effective Date: January 1, 1993 Expiration Date: September 30, 1998 <1993...period of "cooperation"..1994> General Atomics was supposed to be the NIC of NICs and to help coordinate the activities of all three contractors. More NICs were supposed to be formed. December 1994 Midterm Evaluation <http://www.rs.internic.net/nsf/review/review-toc.html> "The InterNIC awards set the precedent of requiring significant self-coordination among a team of awardees, and requiring outreach to other Network Information Centers. The panel suggests that the NSF critically consider whether it is viable to expect significant self-coordination among a team of awardees in future awards. The panel also notes that the NSF's program management was not able to correct GA's problems early on despite excellent efforts by the NSF staff, primarily because the NSF staff were overextended by monitoring at least two major projects (the InterNIC and the NSFNET backbone) at once. The panel recommends that for future large scale efforts in the rapidly changing Internet environment, the NSF should form an ongoing advisory panel of outside experts or employ some external consultants to help manage such cooperative agreements, rather than waiting two years to call for a review." December 1994 General Atomics Dismissed January 1995 NSF gives NSI more money... Amendment 1 <http://rs.internic.net/nsf/agreement/amendment1.html> This amendment increases the funds available under Cooperative Agreement No. NCR-9218742 by $1,258,457. NSI changes key personnel... Amendment 2 <http://rs.internic.net/nsf/agreement/amendment2.html> February-March 1995 (???estimated date???) SAIC purchases NSI http://www.saic.com http://www.netsol.com Circumstances, terms, previous owners....???? not known ???? May 1995 NSF gives NSI more money...AND reviews plan to start charging Amendment 3 <http://rs.internic.net/nsf/agreement/amendment3.html> "The National Science Foundation (NSF) hereby awards $1,948,632 to Network Solutions, Inc. for additional support..." September 13, 1995 NSF approves NSI's plan to charge for domain name registrations <http://rs.internic.net/nsf/agreement/amendment4.html> October 1995 Many people and companies debate the issues surrounding the commercialization of domain name registrations. The NSF and the IANA prevent commercial registries from having new Top Level Domains. November 1995 Debates December 1995 Debates January 1996 Debates February 1996 Debates and delays March 1996 Debates and delays April 1996 Debates and promises of plans May 1996 Debates and more promises of plans June 1996 ISOC Board of Trustees - ANNUAL GENERAL MEETING Montreal, Quebec, Canada <http://www.isoc.org/trustees/96-004.htm> "RESOLVED, that the Board of Trustees of the Internet Society endorse in principle the proposal "New Registries and the Delegation of International Top Level Domains", dated June 1996 by Jon Postel, and approve the role assigned to the Internet Society in this proposal. The Board authorises Postel, in his IANA role, to refine the proposal to include a business plan for review and approval by the Board." July 1996 Debates and refinement of plans August 1996 Debates, meetings and claims of an October 1st start September 1996 Operational registries start to appear October 1996 The balance of the Intellectual Infrastructure Fund is disclosed... <http://rs.internic.net/announcements/iif-update.html> "..through October 31, 1996, $8,542,200.00 has been deposited into the account." The Federal Networking Council Advisory Committee advises the NSF <http://www.fnc.gov/FNCAC_10_96_minutes.html> "The FNCAC reiterates and underscores the urgency of transferring responsibility for supporting U.S. commercial interests in ITLD administration from the NSF to an appropriate entity." November 1996 "...through November 30, 1996, $9,911,000.00 has been deposited into the account." Commercial TRUE Root Name Servers start to appear... the NSF does not allow commercial registries to have TLD entries in the Root Name Servers they control... The NSF awards USC/ISI $1.5 million <http://www.nsf.gov/ftp/awards96/awd9615/a9615927.txt> December 1996 "...through December 31, 1996: $12,685,450.00 has been desposited into the account." The ISOC's IAHC gets started to provide recommendations to IANA http://www.iahc.org NSI ends 1996 registering about 80,000 names per month which is an approximate gross revenue of $8,000,000 per month at $100 per registration. That covers the first 2 years. On an annual basis, this amounts to approximately $48,000,000 per year. January 1997 IAHC debates continue... Plans are announced by NSI to launch ARIN to charge for IP addresses <http://www.arin.net> NSF gives NSI more money... <http://rs.internic.net/nsf/agreement/amendment5.html> NSI and the IANA announce plans to deploy TRUE Root Name Servers February 1997 IAHC recommendations to the IANA finalized <http://www.iahc.org/draft-iahc-recommend-00.html> Commercial registries which are now fully operational are still restricted from having their names entered into the widely used Root Name Servers controlled by the National Science Foundation. Companies are now actively deploying commercial Root Name Servers to replace the NSF supported servers. < YOU ARE HERE> September 1988 Cooperative Agreement Ends... @@@@@@@@@@@@@@@@@@@@@@@@@@ -- Jim Fleming Unir Corporation e-mail: JimFleming(a)unety.net JimFleming(a)unety.s0.g0 (EDNS/IPv8)

2 1

F.Root-Servers.net and bogus data
by Eric Germann 25 Feb '97

25 Feb '97

Folks, In the process of switching ISP's, I sent some updates off to InterNIC last friday. The new SOA's are in place on all root servers except for f.root-servers.net. Mail to hostmaster(a)internic.net yields no response (who knows, they could be using f.....) The domain name is cctec.com. The correct answers are 207.87.243.20 and 207.87.243.60, not 206.21.31.xxx as listed on f. So who runs F, and how do I politely ask them to check and see if they picked up their updates last Friday. Contact listed for root-servers.net is hostmaster(a)internic.net. At the risk of starting another flame war, does this make F the One FALSE Root Server? Eric

5 4

Re: Flapping /16 prefix, why?
by Jim Rowh 25 Feb '97

25 Feb '97

The degree of prepending you show is probably more for Sprint's benefit than anyone else's. 1793, 1239, and 1790 are all Sprint AS's. If you're lucky, you can find someone familiar with Sprint's confederated routing scheme to explain the details. AS3505 belongs to VNet, who uses both Sprint and Gridnet as providers. Prepending allows them (VNet) to effectively designate Gridnet as their primary (all things being equal, better--shorter--AS path wins). As for hold-down routes (to Null0) and damping, yes, they're both good and generally practiced ideas. As to whether and to what extent Sprint uses either or both of them, you will have to take that up with them: noc(a)sprint.net or 1-800-230-5108 jim > Sorry for littering NANOG with this email, but I'm hoping for a little > education on this matter from some of the larger National Service > Providers. > > This morning we had a customer call in concerning a network hole. > Upon investigation, I found the following.. > > BGP routing table entry for 166.82.0.0/16, version 541808 > Paths: (1 available, best #1, advertised over IBGP, EBGP) > 1793 1239 1790 3505 3505 > 144.228.159.17 from 144.228.159.17 (144.228.159.1) > Origin IGP, metric 0, valid, external, best > Community: 1 > Dampinfo: penalty 656, flapped 18 times in 02:19:29 > > Two immediate things came to mind here. > > 1. This is a /16 that is flapping. > 2. The end user is using AS path prepending for some reason. > > With these two facts in mind, the following questions arise. > > Anchoring or tying down of a route with a static route to null seems > to be a very basic idea (far more basic that AS path prepending), so > why isn't this route tied down? > > I only see one route to this prefix, why are they prepending? (I admit > that only the user of this prefix can answer this question, and I also > admit that with 53 flaps on the books so far, the other provider may > be dampening the route for us *thank you*) > > Finally, does Sprint not apply *any* dampening to customer routes? > > > > My cat is long dead (10 years at least), so it's safe from my curiosity. > > Chris A. Icide > Nap.Net, L.L.C. > >

2 1

RE: F.Root-Servers.net and bogus data
by Jim Fleming 25 Feb '97

25 Feb '97

On Monday, February 24, 1997 5:51 PM, Eric Germann[SMTP:ekgermann@cctec.com] wrote: @ At 05:23 PM 2/24/97 -0600, Jim Fleming wrote: @ >If you are using the new TRUE Root Name Servers... @ > @ >J.ROOT-SERVERS.NET @ >K.ROOT-SERVERS.NET @ >L.ROOT-SERVERS.NET @ >M.ROOT-SERVER.NET @ > @ @ @ I can't change the whole Internet overnight. I obviously can resolve @ myself since I go to my own DNS, not needing a root server. But @ statistically, people have an even chance of hitting the Heretics. @ Henceforth, the need to update it. I like the idea of a TXT record or a @ page for contacts at www.root-servers.net Seem like logical solutions. @ I also like the idea of using TXT records and the DNS system in general for more "whois-like" functions. If you ask me, I will provide more detailed proposals. I will not post them to this list to keep the traffic to a minimum. As for as "Heretics", I think the servers above are going to be RFC 2010 compliant. It may not be good to resort to name calling until you have read 2010. @@@@@@@@@@ Network Working Group B. Manning Request for Comments: 2010 ISI Category: Informational P. Vixie ISC October 1996 Operational Criteria for Root Name Servers Status of this Memo This memo provides information for the Internet community. This memo does not specify an Internet standard of any kind. Distribution of this memo is unlimited. Abstract This document specifies the operational requirements of root name servers, including host hardware capacities, name server software revisions, network connectivity, and physical environment. @@@@@@@@@@@@@-- Jim Fleming Unir Corporation e-mail: JimFleming(a)unety.net JimFleming(a)unety.s0.g0 (EDNS/IPv8)

1 0

RE: F.Root-Servers.net and bogus data
by Eric Germann 24 Feb '97

24 Feb '97

At 05:23 PM 2/24/97 -0600, Jim Fleming wrote: >If you are using the new TRUE Root Name Servers... > >J.ROOT-SERVERS.NET >K.ROOT-SERVERS.NET >L.ROOT-SERVERS.NET >M.ROOT-SERVER.NET > I can't change the whole Internet overnight. I obviously can resolve myself since I go to my own DNS, not needing a root server. But statistically, people have an even chance of hitting the Heretics. Henceforth, the need to update it. I like the idea of a TXT record or a page for contacts at www.root-servers.net Seem like logical solutions.

1 0