April 2001 - Test - lists.nanog.org

RE: Faster 'Net growth rate raises fears about routers
by Richard A. Steenbergen 03 Apr '01

03 Apr '01

On Mon, Apr 02, 2001 at 03:37:22PM -0700, Bill Woodcock wrote: > > > So, please explain to me how not being multi-homed is anything other than a > > bad-thing and high-risk? No, I am not including colo, because it is assumed > > that you know what their arrangements are before you "buy". Reputable colos > > are multi-homed, in spades. > > You say "responsible cab drivers must have not one, but two taxicabs, > in order to provide service in the event of a failure. Therefore, I > bought one from Fisher-Price, and one from Hot Wheels, and I'm > astounded to find that neither provides me with the luxury which I > expected." I think Patrik may have been suggesting that if you had a > Checker, you might not need to worry quite so much about redundancy. Only one transit? For a reliable internet, two transits is the minimium requirement, and you have but one, which is less then two, and two is what you need... Curses... You must immediately purchase some transit, which you need for internet, for without transit you cannot have the internet that you so require. -- 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)

5 4

RE: Faster 'Net growth rate raises fears about routers
by Roeland Meyer 03 Apr '01

03 Apr '01

> From: Eric A. Hall [mailto:ehall@ehsco.com] > Sent: Tuesday, April 03, 2001 12:10 PM > Convincing customers that it is cheaper/better to put their > main servers > somewhere off-site away from them is the challenge. Otherwise > more of them > would do it. You understate the problem. Service bureaus and ASPs have one problem in common, MIS/IT departments are loath to place their corporate data (jewels) under the control of another. Convincing them otherwise is an up-hill battle that you will mostly lose. To a lesser extent, this is also true of colos. The other part of the argument, that the local (fully burdened) square-footage is almost always cheaper than decent colo space ($1,200.00 per 1/4 rack, per month, typical). You therefore have two strong motives for locating corp datacenters on-site, rather than off-site. Either one, by themselves, are difficult to overcome. Then you have the third argument; most corporate data centers are used more internally than externally (road-warriors and B2B ops not withstanding). For performance reasons, you want the servers closest to their users. Unless you can get 100baseTX to the off-site facility, this means a strong incentive to mount servers at HQ, not in the colo. We are now back to multi-homing a business. BTW, this entire mess was played out before, in the 70's and 80's ... when the service bureau biz died. At that time SNA APPN/APPC was the network (my briefs were dyed "blue" at the time). The end-game was proprietary pre-internet data centers, connected by mutiple slow links.

1 0

RE: Faster 'Net growth rate raises fears about routers
by Irwin Lazar 03 Apr '01

03 Apr '01

> <quote> > "Half of the companies that are multihomed should have gotten > better service > from their providers," says Patrik Faltstrom, a Cisco > engineer and co-chair > of the IETF's Applications Area. "ISPs haven't done a good enough job > explaining to their customers that they don't need to multihome." > </quote> > > From what I just went through (DSLnetworks v Covad), this is straight > bulldust. Had I been properly multi-homed, MHSC would NOT > have been down for > 5 weeks. Good point, one of our customers multihomed to two different providers and had several instances where a BGP configuration change by one of their providers made their networks invisible to half of that provider's peers. These types of errors take quite a while to track down and are generally only found when the help desk gets calls that some people from around the world can't get to internal network resources (Looking glass sites are invaluable in these situations). Risk mitigation from these types of errors is easy justification for multi-homing. Irwin

2 1

RE: Faster 'Net growth rate raises fears about routers
by Barrows, Jeff 03 Apr '01

03 Apr '01

It is perhaps useful to note that there is at least one network out there with over 210,000 network entries in their tables today-- with multiple routers in that same AS having well over 100 BGP sessions on them-- comprised of around ~25% iBGP/eiBGP sessions in peer groups, and the remainder being eBGP sessions. These boxes push a lot of traffic, deal with a large/flat IGP, do IP filtering, and a variety of other random services. That same network had ~182,000 network entries in their tables in October of 2000. This could be reduced, and the rate of growth slowed, if necessary, but there apparently hasn't been a compelling enough reason for them to do so yet-- which doesn't necessarily indicate anything beyond the fact that there hasn't been enough pain in that area for them to change anything yet. Not to say that great pain isn't on the horizon. - jsb I don't doubt that this will continue to be a topic that is over-sensationalized, as is done here: http://www.nwfusion.com/news/2001/0402routing.html Which includes this tasty quote: "By the time the table gets to around 200,000 entries, we may be pushing a default-free router well beyond its processing capability." ...and this one: "Half of the companies that are multihomed should have gotten better service from their providers" [...] "ISPs haven't done a good enough job explaining to their customers that they don't need to multihome." > Granted, but are you saying that the 15 minutes it takes one of my BGP > sessions to reload has no relevance to the crusty, old processor doing > route calculations on 104,000 routes? Multiply the CPU > available by 5, and then look for bottlenecks. Seems sane to me. > Also seems a hell of a lot easier than trying to redesign the network > characteristics in 1 year and implement them ... IPv6 anyone? > > -travis > > > > > *shakes head* people keep forgetting this. Do you guys also > > think you can solve the internets problems by adding more bandwidth? > > > > > > > > Adrian > > > > -- > > Adrian Chadd "The fact you can download a > 100 megabyte file > > <adrian(a)creative.net.au> from half way around the world > should be viewed > > as an accident and not a right." > > -- Adrian Chadd and > Bill Fumerola > > > > > >

1 0

RE: Faster 'Net growth rate raises fears about routers
by Sean Donelan 03 Apr '01

03 Apr '01

On Mon, 02 April 2001, Roeland Meyer wrote: > From the article: > <quote> > "Half of the companies that are multihomed should have gotten better service > from their providers," says Patrik Faltstrom, a Cisco engineer and co-chair > of the IETF's Applications Area. "ISPs haven't done a good enough job > explaining to their customers that they don't need to multihome." > </quote> Why would a rational customer pay for a second connection (usually more than doubling their cost) if a single connection was satisfactory? Although providers try to vertically integrate their operations, time and time again, vertical integration tends to increase the risk for the customer. Mid-level providers serve an important function in the Internet hierarchy. Multi-homing works well with mid-level providers aggregating local routes, and managing redudancy between long-haul providers. If you don't use a mid-level provider, to achieve the same reliability you end up needing to be your own mid-level provider. Why can't a large provider operate their network as a set of mid-level networks, and connections to multiple long-haul networks. They could.

1 0

[ga] Verisign Agreement Adopted (fwd)
by Patrick Greenwell 03 Apr '01

03 Apr '01

For those of you not following this, the following decision means effectively NSI has been regranted a registry monopoly over .com (they spin out .net and .org) and they do not have to split the registry operation from the registrar operation. From: Andy M�ller-Maguhn <andy(a)ccc.de> To: icann-news(a)lists.ccc.de Date: Mon, 2 Apr 2001 16:02:27 +0200 Subject: [icann board telco] verisign agreement adopted Mailing-List: contact icann-news-help(a)lists.ccc.de; run by ezmlm just a short note after the boards telephone call today - the new versign agreement with some soft changes arrangend between the new ICANN ceo Stuart Lynn in cooperation with ICANN lawyers Joe Sims and Louis Touton has been adopted by the board with 12 votes, 3 against (amadeu abril i abril, karl auerbach, myself) and 1 abstain (ken fockler). The style of the discussion - which was about 40 minutes - was rushed by vint cerf cause he wanted to keep the length of the telconf < 1 h. Ken�s argument, that the NC gave a clear vote against the new proposal ("B") was rejected by the president with the argument, that the arguments of the NC where mentioned and lead to the changes. My suggestion, to give the NC�s constituencys at least 2 days for commenting the work from the Lynn/Sims/Touton that was send in a confidential note sunday (yesterday) to the board, was rejected by the Lynn/Sims/Tounton with the argument that the finalized statement would still mean a lot of work until have it as a written contract and any delay would make it worse. ICANN website will give an overview of the adopted decision soon. sorry for only giving you a brief summary here, sitting on the airport of duesseldorf and have to rush to my airplane now... A. http://www.icann.org/minutes/prelim-report-02apr01.htm Preliminary Report Special Meeting of the Board 2 April 2001 (Posted 2 April 2001) The ICANN Board of Directors held a meeting by teleconference on 2 April 2001. Directors Vint Cerf (chairman), Amadeu Abril i Abril, Karl Auerbach, Ivan Moura Campos, Jonathan Cohen, Phil Davidson, Ken Fockler, Masanobu Katoh, Hans Kraaijenbrink, Sang-Hyon Kyong, Stuart Lynn, Andy Mueller-Maguhn, Jun Murai, Alejandro Pisanty, Nii Quaynor, Helmut Schink, and Linda S. Wilson participated. The Board adopted the following resolutions: Contract for IANA Function (Adopted 16-0, with Andy Mueller-Maguhn not yet present due to telephone difficulties) Whereas, as provided by resolutions 00.5 and 00.6, ICANN entered a contract with the United States Government on 9 February 2000 under which it performed the IANA functions through 30 September 2000; Whereas, as provided in resolution 00.71 that contract was extended to 31 March 2001; Whereas, the United States Government has requested that the contract be renewed to provide a continuing contractual basis for ICANN's provision of the IANA function as the transition of Internet-coordination functions from the United States Government to ICANN progresses; Whereas, ICANN's General Counsel has reviewed the terms of the renewal contract with the Board and recommended that the Board enter into the agreement on those terms; Resolved [01.45], the Board accepts the terms of the renewal contract as reviewed with the Board; and Resolved further [01.46], the Board authorizes the General Counsel to sign the renewal contract on behalf of the Corporation in the form presented. Revision to Agreements with VeriSign (Adopted 12-3, with one abstension. Vint Cerf, Ivan Moura Campos, Jonathan Cohen, Phil Davidson, Masanobu Katoh, Hans Kraaijenbrink, Sang-Hyon Kyong, Stuart Lynn, Jun Murai, Alejandro Pisanty, Nii Quaynor, and Linda S. Wilson voted in favor. Amadeu Abril i Abril, Karl Auerbach, and Andy Mueller-Maguhn voted against. Ken Fockler abstained. Helmut Schink had left the call by the time of the vote.) Whereas, a proposal has been presented to the Board for various revisions in the agreements among ICANN, Network Solutions, Inc., and the United States Department of Commerce that were approved on 4 November 1999 in resolutions 99.132 and 99.133 and were signed on 10 November 1999; Whereas, the proposal was forwarded to all three supporting organizations with a request for comments; Whereas, in resolution 01.22 the Board requested all members of the Internet community, including the Names Council and any of the constituencies and other participants in the Domain Name Supporting Organization, to provide comments on the substantive merits of the proposal no later than 31 March 2001; Whereas, on 9 March 2001 the Protocol Supporting Organization notified the Board that it had considered the proposal only with regard to potential protocol-related technical issues as a result of splitting .com, .net and .org into three registries, and saw no problems with this approach providing that the stability of the DNS resolution is protected; Whereas, on 30 March 2001 the Names Council of the Domain Name Supporting Organization (DNSO) forwarded a four-part resolution accompanied by statements of the seven DNSO constituencies and participants in the DNSO general assembly; Whereas, numerous comments of the Internet community have been received on ICANN's web-based public comment forum, by e-mail, and otherwise; Whereas, in response to the comments ICANN has asked VeriSign for improvement of the proposed agreements through the addition of some text changes and other related undertakings between ICANN and VeriSign; Whereas, VeriSign has stated its commitment to make those improvements; Whereas, the Board has carefully considered the proposal in connection with the input from the supporting organizations and Internet community and the improvements agreed by VeriSign and finds that acceptance of the proposal is necessary and appropriate to further ICANN's purposes; Resolved [01.47] that the President is authorized on behalf of the Corporation to sign the proposed agreements with VeriSign as presented, with suitable revisions and additions to the provisions of the agreements to reflect the text changes and additional commitments set forth in the posted correspondence with Stratton Sclavos and with any further corrections and modifications consistent with the proposal that the President determines are appropriate, such signing to be subject to any necessary approvals of the U.S. Department of Commerce; Further resolved [01.48], the President is authorized to take such actions as appropriate to implement the agreements. - List archives can be found at: <http://www.moongroup.com/inet.php> Send 'unsubscribe' in the body to 'list-request(a)inet-access.net' to leave. Eat sushi frequently. inet(a)inet-access.net is the human contact address.

3 3

RE: [ga] Verisign Agreement Adopted (fwd)
by Roeland Meyer 03 Apr '01

03 Apr '01

> From: Steve Sobol [mailto:sjsobol@NorthShoreTechnologies.net] > Sent: Tuesday, April 03, 2001 12:12 AM > > Patrick Greenwell wrote: > > > For those of you not following this, the following decision means > > effectively NSI has been regranted a registry monopoly over > .com (they > > spin out .net and .org) and they do not have to split the registry > > operation from the registrar operation. > > ?! > > uhhh, so what happens to all the alternative dotcom registrars? > > I mean, is ICANN asking to be hauled into court or what? Nothing of the sort. It simply means that the NSI registrar can continue as-is. The prior agreement would have forced NSI to divest of either the registrar or the registry. Now, they get to keep both but have to divest themselves of ORG and NET registries. BTW, their are legitimate COM/NET/ORG registrars. They are legitimate re-sellers. If you are speaking towards the independent inclusive TLD registries, that remains status quo, as well. What this will mean is that someone has to deploy ORG and NET registries. How much you wanna bet that ICANN contracts NSI to do so?

1 0

RE: dsl providers that route /24- DNS Surveys
by Greg Estabrooks 03 Apr '01

03 Apr '01

> And a survey of DNS servers quite recently showed that 16% still haven't > upgraded to non-hackable versions of BIND. A lot of people drive > without seat belts too. Just because 2/3 of a group do something > doesn't mean it's a good idea. I actually did a few surveys of this sort of thing recently for any who are interested: http://www.phaze.org/dnsaudit PS: Sorry if this is a duplicate post. I don't think my post earlier made it out to the list since I had the wrong reply address set :) -- "And he piled upon the whales white hump, the sum of all the rage and hate felt by his whole race. If his chest had been a cannon, he would have shot his heart upon it."

1 0

DMCA Enforcement
by Kevin Day 03 Apr '01

03 Apr '01

Sorry to those North American operators who aren't in the US for this slightly off topic query. How do upstream providers here deal with DMCA violation reports? When several tiers seperate the backbone from the end customer, should everyone respond? Should the top tiers pass it down to their customers to deal with? The DMCA itself seems pretty vague here, how are you guys dealing with this, and how have your respective legal teams interpreted who needs to get involved? Thanks, Kevin (P.S. Any ISDN gurus here who don't mind answering a quick question, if you could contact me off list, it would be very appreciated)

2 1

RE: Faster 'Net growth rate raises fears about routers
by Paul G. Donner 02 Apr '01

02 Apr '01

it also works the other way around. letting clueless networks manage multihoming with BGP can be a bad thing for the internet in general, especially if the upstreams do not enforce some *discipline* on their customers in these cases. this can be as detrimental to the customer as having lousy upstream service. At 03:37 PM 4/2/2001 -0700, Bill Woodcock wrote: > > So, please explain to me how not being multi-homed is anything > other than a > > bad-thing and high-risk? No, I am not including colo, because it is > assumed > > that you know what their arrangements are before you "buy". > Reputable colos > > are multi-homed, in spades. > >You say "responsible cab drivers must have not one, but two taxicabs, in >order to provide service in the event of a failure. Therefore, I bought >one from Fisher-Price, and one from Hot Wheels, and I'm astounded to find >that neither provides me with the luxury which I expected." I think >Patrik may have been suggesting that if you had a Checker, you might not >need to worry quite so much about redundancy. > > -Bill

1 0