- Test - lists.nanog.org

Re: BBN Peering issues (fwd)
by Sean M. Doran 19 Aug '98

19 Aug '98

ftp.gatekeeper.com, like all content hosting sites, attracts users to one location, when it could instead attract users to the same content in multiple locations. This is inefficient. Many archive sites direct people to mirrors topologically closer to them, and as more actual content owners purchase duplicate sites in places like Europe (http://www.europe.cnn.com, http://www.yahoo.{se, .co.uk, .ca}) both for the technical gains and also as a means of strengthening one's brand globally. This is much more efficient. This should be carried on. It is in carriers' interests to encourage the trend too, and it is probably a good idea to work out means of hosting content more locally than data centres in California, as a means of increasing network efficiency. It is in content-owners' interests to encourage the trend also, as ultimately their brand is the one that is hurt by unreliability. Yes, it might hurt carrier X's business when carrier X cannot get to Popular Content Site, but some fraction of carrier X's customers will go away thinking, "performance to Popular Content Site sucks! Popular Company sucks!" Moreover, as the capacity market distortions are sorted out in many places, it almost certainly will be cheaper for content owners to spread the work of distribution around, so that the traffic stays as local as possible. Sean.

2 1

AOL spamming policy?!?
by Jakob 'sparky' Kaivo 18 Aug '98

18 Aug '98

I've gotten this same message now from four different AOL users today. Just wondering if anyone knows if this is just a group of people working together to fight spam, or if AOL is doing this, sending e-mail from their users without them knowing, or ... ??? ---------- Forwarded message ---------- Date: Sun, 16 Aug 1998 09:43:40 EDT From: *** FROM DELETED *** To: postmaster(a)nodomainname.com, *** OTHER ADDRESSESS REMOVED *** Subject: Polite Request Dear Sirs I have received Unsolicited Bulk E-Mail (UBE) apparently from yourselves or from one of your direct or indirect customers. I don't like it and would ask for your cooperation to put a stop to it. Most of the UBE I receive looks dishonest to me. I am sure your company isn't like that but you would do well to avoid using or permitting the same methods as these "spammers" lest you be tarred with the same brush. Also, you may be aware that a growing number of ISPs are taking to blocking incoming mail from "spam" domains. I don't want that to happen because I might lose legitimate mail and you might be inconvenienced. --snip-- The original message then followed, and it turns out that I am getting the mail because the spammer is trying to fake their host name, but it is simple to look and see their real IP address and reverse DNS, so, no, noone is spamming from my domain. I guess nodomainname.com is a popular domain name if you have no domain name. But that's beside the point. +-----------------------------+--------------------------------+ | Jakob 'sparky' Kaivo | jake(a)nodomainname.net | | NoDomainName Networks | http://www.nodomainname.net | +-----------------------------+--------------------------------+

3 2

Re: The Cidr Report
by Sean M. Doran 18 Aug '98

18 Aug '98

Tli asked: | Exercise for the reader: If prefixes of length F and longer are filtered, | and a domain has a prefix of length N, how many prefixes can they propagate | into the backbone? How many before people start proxy aggregating them? A/ (2 ** F) - 1. [assuming nobody imposes exact-match prefix filters] B/ When we can easily do: "router bgp N proxy-aggregate route-map Lazy-Types ! exceptions route-map Lazy-Types deny 10 match as-path 100 ! the rule route-map Lazy-Types permit 20 match originated-prefix-count 16" then the number 16 strikes me as a starting point, although I probably want to re-read Dennis's comments from a few years ago. Do I pass? Sean.

2 1

BBN Peering issues
by dlr＠bungi.com 18 Aug '98

18 Aug '98

If anyone would like to communicate privately about the BBN Peering issues, please drop me an email, or call. Dave Rand +1 408 254-7111 -- Dave Rand dlr(a)bungi.com http://www.bungi.com

24 39

Smurf Attack Reporting?
by Jason Weisberger 18 Aug '98

18 Aug '98

Where is the central repository these days for reporting a fairly large Smurf attack? Who is running the blackhole list? -- Jason Weisberger Chief Technology Officer SoftAware, Inc. 310/305-0275 "Trust in God and fear nothing." --Gen. Lewis A. Armistead

2 1

Re: BBN Peering issues (fwd)
by Sean Donelan 18 Aug '98

18 Aug '98

>It is in content-owners' interests to encourage the trend >also, as ultimately their brand is the one that is hurt >by unreliability. Yes, it might hurt carrier X's business >when carrier X cannot get to Popular Content Site, but some >fraction of carrier X's customers will go away thinking, >"performance to Popular Content Site sucks! Popular Company sucks!" It is interesting to see how things are sorting out. Yes, content owners are concerned about this, but the solution may not be good for the diversified carriers. I suspect you are going to see more deals like AOL's TeraPOP project. Networks with large content owners are connecting directly with networks with large users bypassing the usual backbones in the process. The diversified carriers are being relegated to managing the banks of modem pools. One thing which will really kill your stock price is if investors get the idea your portion of the market is about to turn into a commidity. The term 'disintermediation' is often used to describe how commerce will develope on the Internet. Diversified backbone providers serve a useful purpose while the market is fragmented. But the very centralization which makes the 'synergy' referenced in many of the mergers also sows the seeds of bypass. On one side you have the large access concentrators such as AOL, Earthlink, Mindspring; on the other side large content concentrators such as GlobalCenter, Exodus. What purpose does the carrier in the middle serve? >Moreover, as the capacity market distortions are sorted >out in many places, it almost certainly will be cheaper >for content owners to spread the work of distribution around, >so that the traffic stays as local as possible. Yes and no. There is a tension or balance between distribution and centralization. I don't think it is possible to make a blanket statement one style or another will be cheaper. -- Sean Donelan, Data Research Associates, Inc, St. Louis, MO Affiliation given for identification not representation

1 0

Re: BBN Peering issues (fwd)
by Sean M. Doran 18 Aug '98

18 Aug '98

Owen and company - Forgive me this indulgence of my more immature side, but I will not reply to you. Should someone else ask these questions or make these observations, I will be happy to provide a careful response. Sean.

1 0

Re: BBN Peering issues (fwd)
by owen＠DeLong.SJ.CA.US 18 Aug '98

18 Aug '98

> ftp.gatekeeper.com, like all content hosting sites, attracts > users to one location, when it could instead attract users to > the same content in multiple locations. > Please explain why you believe this to be universally true of content hosting sites. Especially when you consider that Exodus has been a leader in promoting distributed fault tolerant services to its customers. > This is inefficient. > Agreed. However, assertion 1 being false, the corollary seems less relevant. > Many archive sites direct people to mirrors topologically closer > to them, and as more actual content owners purchase duplicate sites > in places like Europe (http://www.europe.cnn.com, > http://www.yahoo.{se, .co.uk, .ca}) both for the technical gains > and also as a means of strengthening one's brand globally. > And many content providers use various products for redirecting requests to the topologically closest mirror. The result is actually better in many cases, as no manual intervention is required by the user, and the system usually measures topological closeness instead of making geographical recommendations. > This is much more efficient. > > This should be carried on. > Agreed. > It is in carriers' interests to encourage the trend too, > and it is probably a good idea to work out means of hosting > content more locally than data centres in California, > as a means of increasing network efficiency. > Also agreed. What makes you think Exodus, for example, only has datacenters in California? We currently have customers live in Seattle, Santa Clara, Irvine, Waltham, Jersey City, Herndon, and London. > It is in content-owners' interests to encourage the trend > also, as ultimately their brand is the one that is hurt > by unreliability. Yes, it might hurt carrier X's business > when carrier X cannot get to Popular Content Site, but some > fraction of carrier X's customers will go away thinking, > "performance to Popular Content Site sucks! Popular Company sucks!" > True. However, a good percentage of them will call carrier X or Popular Company, or Popular Company's ISP first. > Moreover, as the capacity market distortions are sorted > out in many places, it almost certainly will be cheaper > for content owners to spread the work of distribution around, > so that the traffic stays as local as possible. > Yep. At least three hosting companies I know of are trying to lead the way on technologies and business models to encourage this. Given that, I simply cannot see your statement that the hosting companies prevent this as being anything but specious. > Sean. > Owen

1 0

BBN Peering issues
by cheops POP 18 Aug '98

18 Aug '98

If your traffic is largely based on "client" side applications then guess what, your gonna have a lot of in-bound traffic. This is like AT&T whining about MCI clients calling their moms who have AT&T long distance and not collecting on that usage! If your business model doesn't account for such asymetrical behaviour what are ya gonna do?, ask mom for a charge on her incoming calls?) It looks to me like someone ought to be looking at what is good for the customer and what makes good business sense even though the two hardly ever merge.=) -pete Peter E. Giza Black Pearl Software & Consulting Inc. www.blackpearl.net 603.883.9774

1 0

Re: BBN Peering issues
by Dean Robb 18 Aug '98

18 Aug '98

At 14:20 8/13/98 -0500, you wrote: >That is one thing I don't understand about the aversion to peering. We >peer with anyone who will either pay the line costs or connect to us over >Bell FR, where the costs are negligible. Granted, we are small potatoes in >the ISP field, but to my way of thinking both parties benefit. There is >a customer here communicating with a customer there. Both ISPs are getting >money from their customer to get connectivity to the other ISP's customer. >Dropping the peering connection degrades connectivity for both ISPs' >customers. On the macro, non-technical side of things...this move also degrades the very cooperative nature of the Internet (which is getting less cooperative by the day). It used to be "we're all one big family...c'mon in!" but now it's "you must not only pay your own way, but pay part of our way as well or you can't come in!". Pay per play, all the way. Wanna be on the Internet? Bring your Platinum card. The next step will be "you must pay a hefty price for every packet transitting our pipes", directing transitting (packets only going through specified routes) and strong-arm tactics on the smaller players to sign up with one specific network. Then you'll see regulation akin to what telcos have, with the concommitant expense. Sorry, Pandora...the box is open and Woe is on the loose. Wabbit season!..duck season!..wabbit season!..duck season!..SPAMMER SEASON! Dean Robb PC-EASY computer services (757) 495-EASY [3279]

3 2