December 2002 - NANOG - lists.nanog.org

RE: UseNet Feeds...(OT?)
by Ringdahl, Dwight (WebUseNet) 19 Dec '02

19 Dec '02

I don't really think that a free peering session seems like a 'sales job' and I don't agree that the original use/protocol of the internet would be off topic. As for Satellite feeds... Show me a feed provider who's satellite pushes more than 45mb/s which is now only 60% of a full feed. Yesterday a full feed from our system would have filled up 1.046TB of disk space, a whole lot more than 200gB. This is a growing issue on most networks, at 73mb/s just for the feed not including any reading traffic. One of our customers receives over 800mb/s 24/7 just in NNTP reader traffic... And that's not the largest I have heard of out there. I think Verizon and RR are 2 of the larger one.

2 1

UseNet Feeds.
by Ringdahl, Dwight (WebUseNet) 19 Dec '02

19 Dec '02

If anyone has a usenet server and would like a feed let me know. (FYI current full feed is 72mb/s) We also can peer bandwidth at any of our sites which you can see at http://www.webusenet.com/peering.html (PAIX ATL and Palo Alto are too) Thanks Dwight

2 1

Cogent Thrashing Box Available. IPERF and Web
by Mike (meuon) Harrison 19 Dec '02

19 Dec '02

On Thu, 19 Dec 2002, ______ _________ wrote: > Ehh...something isn't quite right... I managed '573 kBytes/sec' on a > 1.5/384 consumer ADSL... 573 kBytes/sec is > 1.5mbit by a good margin :) > Though it is a nice ego boost LOL :) Running the test again as I'm typing > this message so we'll see what it gets me (hopefully more realistic). I do > have path MTU on as well as large TCP windows which might be affecting the > numbers some but I can't be sure...Second test yielded even wierder > results...Locally another machine started a download so the results > should've been much lower, instead my result was like 3272.73 this time, > even less possible... :D I took out the name, as it was sent to me privately. We saw this last time testing with them as well. Speed tests on this box locally on 100mbps ethernet show normal realistic settings. What we are theorizing is that Cogent is caching outgoing/incoming web traffic, the web server sees the buffered speed.. and reports it.

1 0

Cogent and Level3 Peering Issues
by Dale Levesque 19 Dec '02

19 Dec '02

Anyone happen to have more information on the problems that have been happening with the peering between Cogent and Level3. Cogent gives the standard answer when you call support, but some more specific information would be great. Thanks Dale

10 19

(no subject)
by Ringdahl, Dwight (WebUseNet) 19 Dec '02

19 Dec '02

>> Thing is if your connection is completely full one way, it'll effect >> traffic the other way too. My thoughts are Cogents primary customers are sites that are looking for very cheap bandwidth, which most likely is adult content. Therefore they would look more like a content provider than a transit provider. My question, being a content network, is how would AOL expect them to have a balanced pipe? AOL is all eyeballs, and really doesn't have much content which is useful outside of their user base. Especially if you already peer with Time Warner in other sites. When ISP's peer I would have thought it is to prevent having to pay transit companies like Level3 for the bandwidth. This leads me to believe there might be something more to this, like maybe the spam spewed from the adult sites. Just a guess, anyone have any hard data? Dwight

5 4

Semi off-topic - DC power plant contractors
by Brian Wallingford 19 Dec '02

19 Dec '02

Looking for an electrical contractor in the northeast (preferably in the Providence/Boston area) capable of placing an existing Lorain rectifier & battery array in service. All work must comply to local (Providence) code. Job will include running cabling from the plant to Telect 60A dist panels. Off-list feedback is most welcome. cheers, brian

2 1

Cogent / AOL peering issues
by Gil Cohen 19 Dec '02

19 Dec '02

Some possibly good information about what's going on: Cogent had a "test" peering arrangement with AOL. That means that it was just a test to see what the ratio of cogent/aol traffic is to see whether or not they qualify for a real peering arrangement. Cogent did not meet that requirement, so AOL terminated the agreement. AOL offered Cogent a purchased peering program, but Cogent does not want to take it, because it costs money. Cogent purchases transit from another provider, MFN, that as a peering arrangement with AOL, but Cogent has chosen not to allow traffic to/from AOL to go through MFN. I would imagine this is because it'd cost Cogent money, whereas if it went through the L(3) peering arrangement, it won't. So Cogent is now pushing all the traffic over through their L(3) pipes, therefore, congesting them. That's the reason for all the high latency. Cogent, several months ago started the process to get more pipes into L(3) for peering. Now they're trying to make the process go a lot faster. The following is going to be my interpretation: I think AOL purchases transit from L(3), so by Cogent pushing all their AOL-bound traffic through their L(3) peering pipes, they're hoping that it'll push AOL's transit bill up. This is going to be their playing card. I think. Phew. That's about it. Signed, Gil -- Gil Cohen <gcohen(a)saturnbandwidth.net> | SENIOR NETWORK ENGINEER | SATURN BANDWIDTH & FIBER, INC. | (o) 954-438-7000x124 (c) 954-347-0615 |

2 1

Cogent
by Ringdahl, Dwight (WebUseNet) 19 Dec '02

19 Dec '02

>> But they let them >> all go. So ones the chef(s) were out the door, who maintains it? >> It's not exactly a status quo situation. It was treated in house as, >> well we have the peering so why do we need all these pain in the butt >> expensive guys? We could use another one or two of these expensive guys... Send them my way. Do you think this will hurt them and their other peers? I know L3 has been trimming peers down and if they are filling up their pipes now might cause problems.

2 1

Any good fiber cuts today?
by Sean Donelan 19 Dec '02

19 Dec '02

So what is the story on fiber cuts today? Several people asked me, but I didn't have anything.

1 0

NANOG 27 Information
by Carol Wadsworth 18 Dec '02

18 Dec '02

Meeting registration for NANOG 27, February 9-11, 2003, in Phoenix will open on January 2. Hotel information is now posted at: http://www.nanog.org/mtg-0302/hotel.html Call for Presentations: http://www.nanog.org/mtg-0302/call27.html (Note: deadline for proposals is January 6.)

1 0