April 2000 - Test - lists.nanog.org

Re: peering wars revisited? PSI vs Exodus
by Steven M. Bellovin 05 Apr '00

05 Apr '00

In message <200004050511.e355BQp31490(a)black-ice.cc.vt.edu>, Valdis.Kletnieks@vt .edu writes: > > On Tue, 04 Apr 2000 21:16:55 PDT, Ehud Gavron said: > > Your list of people below are already emailing me to > > say "no no, we just gave him info" > > > > You're not journalist. > > (I'm assuming here that Gordon starts off with "Hi, I'm Gordon Cook, > with the Cook Report", and not social-engineering with false names etc ;) > > Umm.. given the list of names for 1998 through 2000, and the fact that > I *know* that at least some of them were aware of what Gordon writes > when they gave him info, I think it's safe to conclude that at least > a good fraction of Gordon's listed sources were giving him information > fully aware of where it was going to end up. If you tell somebody info > knowing it's going into a well-publicized newsletter, you're talking to > a journalist. It may not be Tom Brokaw, but it's a journalist. ;) Absolutely. As one of the people Gordon interviewed, there was and is no doubt in my mind that he's a journalist. In fact, I know of no reasonable definition of the term that would exclude him -- he gathers information and publishes it. Yes, Gordon expresses opinions -- but the set "columnist" is a subset of the set "journalist". Yes, I (and many others on this list) often disagree with his viewpoint. So what? --Steve Bellovin

1 0

Re: [nsp] IS-IS over IP?
by Danny McPherson 05 Apr '00

05 Apr '00

> I wasn't in Adelaide, but in earlier meetings, I had the impression > that there were at least some implementations using IPv4 as a > workaround both to lower layer specific problems (e.g., AAL SNAP/MUX) > as well as the hard MTU size limit due to lack of fragmentation > capability at the data link layer. I don't believe there were any meetings of the ISIS WG in Adelaide. I'm not aware of any implementations of ISIS over IPv4, much less in production networks. As for the MTU issue, most use maximum LSP size of 1492 bytes and even stop Hello padding after initialization. -danny

2 1

Re: peering wars revisited? PSI vs Exodus
by Gordon Cook 05 Apr '00

05 Apr '00

I have no time for your ilk and besides I am tired and making mistakes with my addressing. <plonk> >So like I said. You're not press. You're a leech. > >Gordon, I've been around before your swill, and I'll >be around after you're done in your leeching ground. > >Your list of people below are already emailing me to >say "no no, we just gave him info" > >You're not journalist. > >E **************************************************************** The COOK Report on Internet Index to 8 years of the COOK Report 431 Greenway Ave, Ewing, NJ 08618 USA http://cookreport.com (609) 882-2572 (phone & fax) Battle for Cyberspace: How cook(a)cookreport.com Crucial Technical . . . - 392 pages just published. See http://cookreport.com/ipbattle.shtml ****************************************************************

1 0

Re: peering wars revisited? PSI vs Exodus
by L KENNEDY 04 Apr '00

04 Apr '00

The thing I find most amusing about this is that people see fit to rag on the press when the level of speculation and misinformation in this forum is worse than the National Enquirer. For all we know, Exodus exactly wanted their "customer confidental" memo to leak. I think Elvis was called in to do it and PSI probably disconnected them because they hired the original inventor of the clipper chip. :-) Leave the speculation to the trade press and discuss the operational impact. Sean

1 0

Anyone received due dates from PB/SBC lately?
by Bill Woodcock 04 Apr '00

04 Apr '00

Has anyone received a due date on any DS1 local loop orders placed with Pac Bell or SBC since they switched their physical-layer provisioning groups on March 17/18? -Bill

1 0

Re: peering wars revisited? PSI vs Exodus
by Michael Shields 04 Apr '00

04 Apr '00

In article <00c601bf9e3a$5d0d2ea0$0a8d473f(a)redconnect.net>, "Dustin Goodwin" <dustin(a)clickthings.com> wrote: > Of course the one piece of information (peering status and utilization) that > would be good indicator of capacity of a ISP is always held as confidential. That is not universally true. > Thank god people work around the rules to let customers and perspective > customers know the truth about peering status. Personally I am ready for > government to step in and force public reporting of peering capacity and > utilization. Is there precedent for such things in other industries? There are many, many examples of organizations that perform tests on the finished products and services an industry produces without any mandated oversight of how they are made. Currently there is a lack of quality Internet service testing and reporting organizations, but that will not necessarily always be the case. -- Shields.

1 0

Re: peering wars revisited? PSI vs Exodus
by Sean Donelan 04 Apr '00

04 Apr '00

On Mon, 03 April 2000, Randy Bush wrote: > >> Because one party -- the originator -- marks an electronic communique as a > >> confidential communication, does that really require the reciever to keep > >> it confidential? > > Professional courtesy. > > or one nanoclue if you think you want anyone to talk to you. Gordon is a reporter, professionally what would you expect? Did Gordon do anything different than the New York Times did with the Pentagon Papers? Besides the law of large numbers is working against you. Although a few people may not talk to Gordon, if you send a "confidential" communication to large group of people with no duty to keep it secret, someone will talk. Since peering is listed as a significant risk factor in many SEC filings, I'm wondering whether it will be considered a material piece of information for one or both of the parties. If it was important enough to notify Exodus' customers, was it also important enough to notify investors? Was Gordon shorting the stock?

1 0

Community advertisement
by Greene, Dylan 04 Apr '00

04 Apr '00

Hello there, All of the backbone NSPs I've spoken with (UUNET, Sprint, etc..) have offered to provide me with either full tables or partial (customer) routes. What would actually be interesting to me would be full tables w/ that customer list tagged as a separate community. It isn't difficult to config, does anyone know if this is actually offered by any providers? If not, why not? It would be very helpful to know which routes stay on a single backbone, and which traverse peering links, etc.. The partial tables provide this, but there are situations where one wants to route other traffic to that provider, so a full table is necessary (default won't really work if you've got advertisements from other providers you _don't_ want to send that traffic to.) I'd be interested to hear if others are doing anything similar to this, have requested it, etc. The standard community offering I've seen from anyone are the MED like 'set local_pref on our backbone' tags which helps direct inbound traffic, but there is little available for outbound. Thanks, ..Dylan .+'''+. D B 7 0 @loa.com - Please do not bend, fold, `+.,.+' dgreene(a)navisite.com - spindle or mutilate.

3 2

Re: [nsp] IS-IS over IP?
by Howard C. Berkowitz 04 Apr '00

04 Apr '00

>The ISIS WG in the IETF is (/was) working on an "ISIS over IPv4" >specification, though it's value has become somewhat questionable. > >Initially, I beleive there were a few drivers such as removing the >requirement for the OSI CLNP support, and the need for more efficient >"ISIS over [something] over ATM" so that AAL5 SNAP could be replaced >with AAL5 MUX or the like to alleviate overhead required for protocol >identifier. > >However, workarounds have been provided for the ATM overhead issue and >the "ISIS over IPv4" specification itself seems to have become more of >an academic exercise. > >A few good things did come out of the WG, however. See > >http://www.ietf.org/html.charters/isis-charter.html > >for current working group information. > >-danny I wasn't in Adelaide, but in earlier meetings, I had the impression that there were at least some implementations using IPv4 as a workaround both to lower layer specific problems (e.g., AAL SNAP/MUX) as well as the hard MTU size limit due to lack of fragmentation capability at the data link layer. Disclaimer: I'm developing requirements for Nortel's carrier routing products. > > > > ISIS is a routing protocol that runs over ISO/clns - so what is your > > question ? > > > > /Jesper > > > > -- > > Jesper Skriver, jesper(at)skriver(dot)dk - CCIE #5456 > > Work: Network manager @ AS3292 (Tele Danmark DataNetworks) > > Private: Geek @ AS2109 (A much smaller network ;-) > > > > One Unix to rule them all, One Resolver to find them, > > One IP to bring them all and in the zone to bind them. > >

1 0

Re: [nsp] IS-IS over IP?
by Howard C. Berkowitz 04 Apr '00

04 Apr '00

>On Tue, Apr 04, 2000 at 09:35:30AM -0400, Howard C. Berkowitz wrote: > > Is anyone running IS-IS over IPv4, rather than a data link protocol, > > in production? If so, does it appear to solve the jumbo LSP and > > medium dependency problems? > >ISIS is a routing protocol that runs over ISO/clns - so what is your >question ? > >/Jesper As opposed to OSPF, which runs over IP, and BGP, which runs over TCP, (Integrated) ISIS is defined to run directly over the data link layer. There is an internet draft about enabling ISIS to run over IP. I'm trying to find operational experience with it. >-- >Jesper Skriver, jesper(at)skriver(dot)dk - CCIE #5456 >Work: Network manager @ AS3292 (Tele Danmark DataNetworks) >Private: Geek @ AS2109 (A much smaller network ;-) > >One Unix to rule them all, One Resolver to find them, >One IP to bring them all and in the zone to bind them.

1 0