October 1996 - Test - lists.nanog.org

RE: Airplane crashing into Atlanta-NAP
by hcb＠clark.net 27 Oct '96

27 Oct '96

At 8:21 AM 10/27/96, Michael Dillon wrote: >On Sun, 27 Oct 1996, Chris A. Icide wrote: > >> Now that I've thrown in my share of late night sarcasm, It would interest >> me greatly to understand exactly why you came to the following >> conclusions: >> >> 1. The "operation" in Atlanta is Mickey Mouse > >Both the Internet and the POTS system are telecommunications networks that >are vital to the modern economy. POTS moreso than the Internet right now >but the Internet is certainly heading in that direction. POTS exchanges >are always in ground floor concrete buildings with no windows or >underground. But in Atlanta they stick it up on the 5th floor of some >office building?!?!?!? > >> 2. The floor # of a bulding affects the quality of the Exchange Point >> 3. The type of building affect the quality of the Exchange Point > >I cannot explain these ones but have reached this conclusion from >observing how the phone company builds and locates its exchanges. If an exchange needs to be in a shared building, there can be advantages to having it on an upper floor. I've built a fair number of computer rooms in my career, always taking great care with the fire protection system. [Andrew -- remember the entire staff watching through the windows when the fire marshal ran the halon test?] I've never actually had a fire in a computer or communications room with which I've been involved, but I've had several in-building floods -- caused by firefighting on higher floors or by malfunctioning sprinkler systems in or near the computer room. The second law of plumbing is "water runs downhill," the first being "if it doesn't leak, don't fix it." Keep these laws in mind when building facilities. > >> 4. A city the size of Atlanta needs more than 1 Exchange Point > >It has a lot more than one POTS exchange. Thus it will need more than >1 Internet exchange. Why should the packets from every video-call in the >city all travel downtown when frequently the two parties live in the same >neighborhood? I can't speak specifically to Atlanta, but in many metropolitan areas, it's vastly faster and cheaper to get space in shared buildings. Take the Wall Street area of New York -- there are some telco switches in the World Trade Center. Remember that many telco exchanges where constructed at a time when they had a guaranteed revenue stream from monopoly local service. Local telcos still often have advantages in financing new construction. [Aside -- the thing that really scares me about Atlanta is an airplane crashing into the Center for Disease Control Level 4 labs there.] > >> Truly I would be very interested in your thoughts on these items, as >> well may a few other folks on this list. > >I'm taking a long term view in which ISP's are just another form of >telephone company. Many ISP's are now getting to the size where they can >consider aquiring strategically located properties, building concrete >block exchange/colo buildings, wiring up entire office towers with >IP fiber and even running their own fibre in some case, especially in new >subdivisions. No argument. :-( I'm fighting with my local telco, not ISP, just to get a fourth pair run to my house. Many ISPs, however, are not quite at the stage when they can acquire properties as easily as they can get shared space. It might be a very useful thing for there to be some consensus guidelines to be developed for physical planning for small to medium ISPs in rapid growth phases. I'd be willing to put together a draft and edit in responses if there's interest in doing this; I have some slides I've put together for a new ISP planning course. I'm actually fairly familiar with the construction of Cheyenne Mountain and other hardened military sites. Not even the military suggests building that type of command post any longer; airborne ones are preferable. Now THAT's an approach for a NAP -- a 747. Running to the fiber to it, however, is something of a challenge. > >By today's standards Atlanta-NAP may be a really great thing, but too soon >we will discover that we aren't living in "today" any more and the >standards will be different. > No argument. But I suggest there is a meaningful intermediate level for which guidelines are useful, between the ordinary office and the purpose-built facility. Howard Berkowitz PSC International

1 0

Root-64 Weekly Status Report (fwd)
by Michael Dillon 27 Oct '96

27 Oct '96

This Root-64 activity is widespread enough that I think all network operators should take note of it. Read this message carefully, there are people working on this project in many countries. If you want to keep tabs on what is happening there is a URL in the message for a mailing list to subscribe to. For the record, I am opposed to what this group is doing. Jim Fleming, the author of the forwarded message, is in favor. If you want to discuss the matter I think NANOG is not the right list given that there is a list mentioned in the message below. Michael Dillon - ISP & Internet Consulting Memra Software Inc. - Fax: +1-604-546-3049 http://www.memra.com - E-mail: michael(a)memra.com ---------- Forwarded message ---------- Date: Sat, 26 Oct 1996 14:28:31 -0500 From: Jim Fleming <JimFleming(a)unety.net> To: 'New Newdom' <newdom(a)vrx.net> Subject: Root-64 Weekly Status Report Saturday, October 26, 1996 Root-64 Weekly Status Report Several people have expressed interest in some sort of weekly summary. This will help busy people get an overview of the progress being made without getting involved in the daily discussions. This past week was very busy, the NSF and the FNC (Federal Networking Council) met and based on the early reports from that meeting, the FNC is recommending that the NSF work hard to divest itself from the top level domain issues and the management of the popular root name servers. Hopefully, the NSF will see that projects like the Root-64 Project will help accelerate this process. Attached is the growing list of root name servers which are being deployed around the world. This list helps to illustrate that the commercial world as well as non profit organizations are willing to step in as the NSF and the U.S. Government withdraw support. Additional root name servers will provide the stability needed to grow the Internet beyond the current R&D architecture supported by the NSF. As shown below, the major recent changes are: England: Keith Mitchell - keith(a)linx.org has expressed interest in the project and appears to have a very strategic location and facilties in the U.K. to provide support for not only the U.K. but surrounding regions. France: Laurent BERNARD - lbernard(a)artinternet.fr is very enthusiastic about getting involved from his Paris based ISP. He has volunteered to help develop a root DNS FAQ, which hopefully may also be available in French. Australia: Andrew Khoo - andrew(a)aussie.net has indicated that his management is supportive of the project and would like more information to help them bring a root name server to Australia. ISOC/IAHC: The Internet Society, under Don Heath's (heath(a)linus.isoc.org) direction, is starting to make progress. They have appointed a committee to discuss all of the issues of the top level domains. Only one of the members of the committee, Perry Metzger, was active in the "newdom" discussions this past year. It is still not clear if they intend to host a root name server. That may be one of the topics that their committee discusses. It is still too early to tell. @@@@@ 0 - Legacy Internet, R&D, Education, Etc. 0 - IANA - U.S.- Southern California - NS1.ISI.EDU - 128.9.0.107 1 - InterNIC - U.S. - Virginia - NS.INTERNIC.NET - 198.41.0.4 2 - NANOG - ??? 3 - ISP/C - ??? 4 - MERIT - ??? 5 - IETF - U.S. - California - NS.ISC.ORG - 192.5.5.241 6 - ISOC - Contact Don Heath - heath(a)linus.isoc.org IAHC - Contact ??? 1 - Perry Metzger - perry(a)piermont.com 2 - David Crocker - 3 - Jeff Houston 4 - Hank Nusbacher 5 - David Maher 6 - Sally Abel 7 - Robert Shaw 8 - WIPO representative 9 - ??? 7 - WIA - ??? 1 - North America (U.S., Mexico, Canada) 0 - U.S. - Northwestern - AlterNIC - MX.ALTERNIC.NET - 204.94.42.1 1 - U.S. - Illinois - MCS Net - ROOT-NS.MCS.NET - 192.160.127.86 2 - U.S. - Michigan - AGN Net - SIMBA.AGN.NET - 160.79.1.3 3 - U.S. - Wisconsin - SPARKNET.NET - ROOT-NS.THENIC.NET - 207.67.22.81 4 - U.S. - Maryland - TERP.UMD.EDU - 128.8.10.90 5 - Canada - Ontario - TORONTO.ALTERNIC.NET - 207.107.232.106 6 - Mexico - http://www.nic.mx ??? 7 - U.S. - Southeastern - C.PSI.NET - 192.33.4.12 2 - South America, Central America and the Caribbean 0 - U.S. Virgin Islands - USVI.NET - 204.199.0.4 1 - British Virgin Islands - ??? 2 - Brazil - ??? 3 - Argentina ??? 4 - Venezula - Contact Peter de Blanc - pdeblanc(a)usvi.net 5 - Bolivia ??? 6 - Costa Rica??? 7 - Panama ??? 3 - England, Europe, Scandanavia and Russia 0 - RIPE - Netherlands - http://www.eu.net ??? 1 - Sweden - NIC.NORDU.NET - 192.36.148.17 2 - England - Contact Keith Mitchell - keith(a)linx.org 3 - France - Contact Laurent BERNARD - lbernard(a)artinternet.fr 4 - Germany - ??? 5 - Switzerland - ??? 6 - Italy - ??? 7 - Ukraine - NS.WW.NET - 193.124.73.100 4 - Japan, Korea, China and The Pacific Rim 0 - APNIC - http://www.apnic.net ??? 1 - Japan - http://www.nic.ad.jp ??? 2 - Korea - http://www.krnic.net ??? 3 - 4 - Taiwan - http://www.twnic.net ??? 5 - China - http://www.cnc.ac.cn ??? 6 - Philippines - http://www.ph.net ??? 7 - Hong Kong - http://www.cuhk.hk/hkwww.html ??? 5 - Asia, Africa and the Middle East 0 - 1 - India - http://www.iisc.ernet.in/innic.html ??? 2 - Pakistan - http://www.ar.pk/public/pknic.html ??? 3 - Bangladesh - http://www.bangla.org/bdinet ??? 4 - Israel - ??? 5 - Saudi Arabia - ??? 6 - South Africa - ??? 7 6 - Australia, New Zealand, Southeast Asia and the South Pacific 0 - Australia - Contact Andrew Khoo - andrew(a)aussie.net 1 - New Zealand - http://servius.waikato.ac.nz/isocnz ??? 2 - Singapore - http://www.nic.net.sg ??? 3 - Thailand - http://www.thnic.net ??? 4 - Indonesia - http://www.iptek.net.id/ipteknet_eng.html ??? 5 - Guam - http://ns.gov.gu ??? 6 - Malaysia - NS.ALPHAQUE.COM - 202.185.254.12 7 - 7 - Vehicles, Boats, Spaceships, Ham Radio, etc. 0 - NS.NASA.GOV - 192.203.230.10 1 - NS.NIC.DDN.MIL - 192.112.36.4 2 - AOS.ARL.ARMY.MIL - 128.63.2.53 3 - NS.UNETY.NET - 207.32.128.1 4 - 5 - 6 - 7 - @@@@ A variety of people are starting to write software, scripts, documentation, etc. that will be needed to coordinate the updates for all of the root name servers. For people that want a Java view of the servers, the temporary web site, http://www.unety.net/Java/root.html has a demo. Chris Sevcik - chris(a)sparknet.net has been exploring some interesting ideas about a web sites, etc. Chris has pointed out the need for a web site that helps document who the contacts are for the above root name servers. Alexis Yushin - alexis(a)dawn.ww.net of Ukraine has expressed interest in moving the discussion forward on the technology needed to keep the world wide collection of root name servers in synch. I would encourage everyone to work with Alexis on these important projects. John Palmer - newdom(a)Taka.AGN.NET reported that their "whois" database is up and working for the .EARTH and .USA top level domains. John provided both URLs as just another example of how the new TLDs can coexist with the "popular" TLDs. http://www.earth/whois.html http://www.agn.net/whois.html Peter deBlanc -pdeblanc(a)usvi.net reports that he is making good progress in Venezula and hopes to have some news in early November. South America is clearly an area where more progress is needed. If anyone wants to explore those regions, cruise on down there. The NANOG group met in Ann Arbor this week. I did receive a response from them, but I do not anticipate any news until the results of their meetings are digested. The operator of the OLD newdom mailing list has suggested closing the list. Several of the members of that list have now subscribed to the NEW newdom list. Richard J. Sexton, the operator of the new list wants to remind everyone that there is a web site for people to subscribe. Please circultate this URL <http://www.newdom.com/lists/>. On the International front, there are meetings next week in Geneva, Switzerland. Stay tuned to the following web site for more info http://www.wia.org It should be clear from that web site that many stakeholders are now involved in the Internet, and some of them will want to take part in helping to provide stability via the deployment of root name servers. Stay tuned for next week's report...inputs are welcome...;-) P.S. As far as I know, none of the above work was funded by the NSF, or the U.S. government. All of the work is being done by private citizens and companies who volunteer their time and resources to these efforts. -- Jim Fleming UNETY Systems, Inc. Naperville, IL e-mail: JimFleming(a)unety.net JimFleming(a)unety.net.s0.g0 (EDNS/IPv8)

5 8

RE: Airplane crashing into Atlanta-NAP
by William Sommers 27 Oct '96

27 Oct '96

On Sun, 27 Oct 1996 08:21:57 -0800 (PST) Michael Dillon wrote: > POTS exchanges are always in ground floor concrete buildings with no > windows or underground. Not the one 2-1/2 blocks from here. As a matter of fact, it has long provided an entrepreneurial opportunity for those who know where to dispose of copper by the pound. I can only surmise that it's cheaper to let things continue than to invest in window bars, but it doesn't exactly give me warm fuzzies to know that folks regularly wander through the aisles housing my circuits. William Sommers San Francisco Online Televolve, Inc. sommers(a)sfo.com

1 0

RADB route object "advisory" field
by Ken Lindahl 27 Oct '96

27 Oct '96

hi, this is substantially different from the usual threads i see on this list, and i trust someone will let me know if it's completely inappropriate. can anyone more experienced with the RADB than i (that's probably most everyone here) point me to documentation describing the "advisory" field in RADB route objects? i've looked through the docs at www.ra.net and found no mention of this field. thanks, ken

2 1

Re: Should NSF declare victory and leave?
by Scott Bradner 27 Oct '96

27 Oct '96

> Could or should ISOC step in if NSF vacates these shoes? sorta depends on just what the "co-operative agreement" says - come to think of it, I would think that changing the term of the agreement sorta depends on what the existing agreement says - there are not all that many contracts that allow one party to change the terms in that basic a way without the agreement of the other party Scott

1 0

the Union makes us Strong
by William Allen Simpson 27 Oct '96

27 Oct '96

> From: Vadim Antonov <avg(a)quake.net> > Now, let's imagine that wolves learned to eat grass and started to enjoy it. > What they do then is simply catch and eat all rabbits -- to eliminate > competition for grass. Of course, some rabbits may be real quick-pawed > (hint, hint). > ... Several have noted in the past that Vadim has a religious predeliction toward "robber baron" oligopolies. We've had plenty of historic experience with them here in the US. The wolves have learned to eat grass (IP) in addition to their old fare (overpriced direct links, X.25, and OSI). Gentlefolk, if we are to avoid the doom that Vadim predicts, the large number of rabbits need to band together to fend off the wolves. And develop BIG TEETH. (Holy Grail :-) There was a great deal of talk among small providers in the halls and restaurants around NANOG last week. Several different models of cooperation were discussed. Much of the effort so far has been all volunteer. There seems to be a general fear of actual financial cooperation and management. As I raised as an issue with the large providers during their presentation, the very nature of IP statistical multiplexing improves throughput when several providers join together and use the same pipe. This could also help ameliorate the cost and peering issues raised here by our international brethren. (As an aside, the objection was that the US Government would object. This is silly! The automotive industry isn't required to have different size screws made by different suppliers for every car, or use different railroads and trucking firms to get parts to the plants. Instead, the screws are standardized and all can buy from the same pool of parts and transit suppliers.) Yet, if we are to survive, we will need to build a shared cooperative infrastructure tailored to our own needs. The writing is on the wall. I propose that ISPs join together, build, and fund regional and metropolitan exchanges, and each of these exchanges make a group purchase of transit to other exchanges, instead of each ISP purchasing their own transit. Eventually, the independent exchanges together will constitute a super provider which meets the draconian peering requirements promulgated by the big providers to limit competition. I believe that we need an umbrella organization to make this happen. There is apprehension that the CIX might involve conflict of interest among its membership. The new ISPC might be a good choice. WSimpson(a)UMich.edu Key fingerprint = 17 40 5E 67 15 6F 31 26 DD 0D B9 9B 6A 15 2C 32 BSimpson(a)MorningStar.com Key fingerprint = 2E 07 23 03 C5 62 70 D3 59 B1 4F 5E 1D C2 C1 A2

1 0

Re: Airplane crashing into Atlanta-NAP
by William Allen Simpson 27 Oct '96

27 Oct '96

> From: Michael Dillon <michael(a)memra.com> > The only joke here is that some people think that the Internet exchange in > a city the size and importance of Atlanta can be adequately housed on the > 5th story of an office building. > > I could understand it if Atlanta had 20 or 30 exchange points, but when > there is only one exchange the city deserves better than a mickey-mouse > 5th floor operation. > Michael, I often agreed with your comments on this list in the past, but I think this one was un-called for. I may joke at times about the chronological age, experience, and general cluefulness of this particular vendor, but it appears that they have already looked hard at the general size and survivability issues. Indeed, I think that they have done a better job (on paper) than any of the original NSF NAPs. Wait and see, the proof is in the pudding.... Moreover, it takes time for an exchange to grow. I would not expect that the location of any NAP be fixed for all time to come. I am a long-time advocate of building more regional exchanges, large and small. This area of the continent _needs_ a good exchange. If this a good geographical location, where the links of many carriers already pass nearby, it behooves us (NANOG) to encourage the placement of an exchange there. So, let's all get together and help make it a reality. Or, would you rather become totally dependent on MFS in this region of the world? I'm still waiting to see what the peering and pricing model will be. Anything like PAIX? WSimpson(a)UMich.edu Key fingerprint = 17 40 5E 67 15 6F 31 26 DD 0D B9 9B 6A 15 2C 32 BSimpson(a)MorningStar.com Key fingerprint = 2E 07 23 03 C5 62 70 D3 59 B1 4F 5E 1D C2 C1 A2

1 0

Re: GigaRouter (Was Re: Cisco as Big Brother))
by doleary＠cisco.com 27 Oct '96

27 Oct '96

Hey, how are things going? I missed you at NANOG, but I heard a rumor you were these even though you didn't do the presentation on Thursday - Also Sam Madani said he got voicemail from you and he paged you sometime on Thursday and again on Friday morning, I don't know if he actually got in touch. Also I talked with Dave Dukinfield, he said that there is finally an account manager working with you - anyway, I hope life is okay and not too crazy, and that if you did miss NANOG it was something innocuous.... dave At 9:15 AM 10/21/96, Nathan Stratton wrote: >On Mon, 21 Oct 1996 alex(a)relcom.eu.net wrote: > >> (1) when PC-based router became out of memory, I have to add some more >>memory - >> I pay about 200$ for extra 16Mb of ram, and that's all; > >Yep, right now I am using 64 megs. > >> (2) when PC-based router became out of CPU, it can be upgraded to the >> faster CPU easy. Intel's power increases draqmatically every month, and >>I have'not >> pay extra 100,000$ for the new super/giga/huge-ROUTER (as 7513) - >> I pay new 1,500$ and get new PC with Pentium/200, for example. >> And I know there would be available better processor in next 6 month - >> and I would'not have to pay next 100,000$ (or I there have to pay >> new 20,000$ for the new CS4700, for example - why can't I change >> CPU in CS4500, or why can't I add extra 32Mb of the RAM into my CS4500, >> and WHY have I to pay 3,500$ for the 32Mb ram if this RAM costs >> 600$ on the free market???). > >Yes, but you can't get the speed out of the PC. I don't care how fast your >CPU is, it will be slower then port to port switching. Don't get me wrong, >I am a big fan of PC routers. I think I am the only one with a PC router >at MAE-East (will be upgraded to a NetStar or cisco in 4 days). I started >Netrail when I was 18, and no one would give me big money and I wanted to >peer at MAE-East and play with the big guys. I did not have the money to >upgrade all of our 4000s to 7000s, so we built PC routers. I think it was >the best thing we did. > >Ok, now we are building a OC3 and DS3 backbone and a PC can't come close >to the PPS a cisco or NetStar can. My only question is should I go with >the NetStar/Cascade or just cisco network? > >Nathan Stratton CEO, NetRail, Inc. Tracking the future today! >--------------------------------------------------------------------------- >Phone (703)524-4800 NetRail, Inc. >Fax (703)534-5033 2007 N. 15 St. Suite 5 >Email sales(a)netrail.net Arlington, Va. 22201 >WWW http://www.netrail.net/ Access: (703) 524-4802 guest >--------------------------------------------------------------------------- >"Therefore do not worry about tomorrow, for tomorrow will worry about >itself. Each day has enough trouble of its own." Matthew 6:34

1 0

Re: Provider credibility - does it matter? was Re: Inter-provider relations
by Peter Lothberg 27 Oct '96

27 Oct '96

> So what's the problem? Basically, one organisation, Ebone, which > operates a fairly large pan-European net, won't peer with others. This > means that European customers connected to non-Ebone connected > providers like Global One lack European connectivity without going > through the US. Ebone provides great connectivity at cost, but the big > US NSPs making presences in Europe naturally don't want to buy > bandwidth from someone like Ebone. We'll see what happens. > > - Jakob Faarvang > cybernet.dk I don't think this is correct. Ebone peers with other pan-european networks of comparable size, IBM, BT and EUnet comes to my mind. G1 is a customer of Ebone and connects in Paris. --Peter

3 2

Strangeness at MAE-East
by Chris A. Icide 27 Oct '96

27 Oct '96

I seem to have lost connectivity across MAE-East to a group of peers. It appears to have come back at about 1:00PM Eastern time. I've attached some snippets from our BGP table, anyone know what may have happened, and what is common to these peers that dropped thier sessions? Neighbor V AS MsgRcvd MsgSent TblVer InQ OutQ Up/Down State [snip] 192.41.177.65 4 6205 32618 72819 2428592 0 0 07:48:39 192.41.177.73 4 6082 16026 56940 2428592 0 0 07:47:15 192.41.177.75 4 3951 48247 56693 2428593 0 0 1w4d [snip] 192.41.177.113 4 3549 16133 56706 2428593 0 0 1w3d 192.41.177.117 4 6079 16245 56869 2428593 0 0 07:47:49 192.41.177.118 4 6196 16160 56661 2428593 0 0 6d16h 192.41.177.122 4 1299 17874 56841 2428593 0 0 3d07h 192.41.177.132 4 3257 16163 57035 2428593 0 0 07:47:45 192.41.177.147 4 3847 18249 56666 2428593 0 0 2d20h [snip] 192.41.177.160 4 3407 48456 56732 2428593 0 0 4d09h 192.41.177.166 4 2885 20740 56618 2428593 0 0 07:47:28 192.41.177.169 4 2885 20599 56815 2428593 0 0 1d07h [snip] 192.41.177.182 4 5683 18340 56875 2428593 0 0 4d08h 192.41.177.183 4 5683 18370 57018 2428593 0 0 07:47:47 192.41.177.192 4 2548 41322 52346 2428593 0 0 2d12h [snip] 192.41.177.210 4 2551 21995 56687 2428593 0 0 1w4d 192.41.177.215 4 5413 165794 58802 2428593 0 0 07:47:51 192.41.177.228 4 4006 16269 56949 2428593 0 0 1d21h [snip] 192.41.177.242 4 1239 184562 56686 2428593 0 0 1w4d 192.41.177.245 4 174 57149 57069 2428593 0 0 07:47:29 NOTE the 07:47:XX. All appear to have dropped at the same time. Hrm, one CWIX router stayed up, one went down. CWIX, what is the difference? Chris A. Icide Nap.Net, L.L.C.

2 1