- Test - lists.nanog.org

RE: BGP quandary
by Leo Nelson 11 Oct '00

11 Oct '00

I do not claim to be a BGP guru but here is my take. If my take is way of base I would like to here from our BGP guru's here Thanks leo -----Original Message----- From: Timothy Brown [mailto:tcb@ga.prestige.net] Sent: Wednesday, October 11, 2000 10:15 AM To: nanog(a)merit.edu Subject: BGP quandary This may be an obvious answer, but, like some obvious answers, you just never get the clue, even though you struggle over it. I have four routers in a network. Two of the routers have no external peers, two of them have two peers a piece. The diagram looks like this: p1(e) p2(e) p1(e) p2(e) / / / / .-------. .--------. | rtr_1 | | rtr_2 | `-------' `--------' | (i) | (i) .-------. .--------. | swr_1 |--------| swr_2 | `-------' `--------' | | ------------------------------------- | | | fw fw fw p1 (AS12345) and p2 (AS12346) are the same AS across both routers, but connected to different pops. so on rtr_1, p1 may go to sjc, but on rtr_2, p1 may go to lax, even though it's the different AS. !# This is a bit confusing to me p1(AS12345) is p1 no matter what rtr connects to it or where it connects. Same goes for p2(As12346). My quandary is this: I want to advertise my best path to swr_1 and swr_2. the line between swr_1 and swr_2 is a trunk; e.g., they're one piece of broadcast media, but they both have routing cards in them to process IP packets (and BGP). my problem is, rtr_1 is going to select a best path from p1 and p2, and advertise it to whoever his peers are. rtr_2 will do the same, but with the additional caveat that rtr_1 may advertise his best path to rtr_2, giving rtr_2 three paths to choose from in this example (assuming p1 and p2 only advertise one path to me a piece for the same netblock). !# This may be difficult to do if you want any type of redundant pathing. Both swr's will need a peering session (tcp connection) to both rtr's for redundancy(unless you don't care). Therefore I would suggest iGBP peering betwween the rtr's and swr's and not worry about the swr's making a decision. You can use dampening if you are worried about excessive cpu utilization. further complicating the issue, swr_1 and swr_2 have to accomplish the same task. i don't want those two boxes to have to deal with selecting a best path; I would like the best path pushed down from rtr_1 and rtr_2, but i'm not sure where i'm pushing it. !# The only way to get one "best path" pushed is to have only one peer, igrp or egrp or default route to the eBGP rtr. Ohterwise the routing protocol will make a decision on what route to inject into it's table based on the metric if more than one is recieved. the fw device will route to the swr devices, probably via HSRP, but if they don't do that via HSRP, they have to select an additional best path (based on network load) via IGP, deciding which swr to connect to (assuming they are capable of that). the swr devices will have to identify a way to route externally, which i'm also not sure of. I can see several options, all of which have significant disadvantages and weird routing behaviour. does anyone have any suggestions on how to solve this situation? tim

1 0

BGP quandary
by Timothy Brown 11 Oct '00

11 Oct '00

This may be an obvious answer, but, like some obvious answers, you just never get the clue, even though you struggle over it. I have four routers in a network. Two of the routers have no external peers, two of them have two peers a piece. The diagram looks like this: p1(e) p2(e) p1(e) p2(e) / / / / .-------. .--------. | rtr_1 | | rtr_2 | `-------' `--------' | (i) | (i) .-------. .--------. | swr_1 |--------| swr_2 | `-------' `--------' | | ------------------------------------- | | | fw fw fw p1 (AS12345) and p2 (AS12346) are the same AS across both routers, but connected to different pops. so on rtr_1, p1 may go to sjc, but on rtr_2, p1 may go to lax, even though it's the different AS. My quandary is this: I want to advertise my best path to swr_1 and swr_2. the line between swr_1 and swr_2 is a trunk; e.g., they're one piece of broadcast media, but they both have routing cards in them to process IP packets (and BGP). my problem is, rtr_1 is going to select a best path from p1 and p2, and advertise it to whoever his peers are. rtr_2 will do the same, but with the additional caveat that rtr_1 may advertise his best path to rtr_2, giving rtr_2 three paths to choose from in this example (assuming p1 and p2 only advertise one path to me a piece for the same netblock). further complicating the issue, swr_1 and swr_2 have to accomplish the same task. i don't want those two boxes to have to deal with selecting a best path; I would like the best path pushed down from rtr_1 and rtr_2, but i'm not sure where i'm pushing it. the fw device will route to the swr devices, probably via HSRP, but if they don't do that via HSRP, they have to select an additional best path (based on network load) via IGP, deciding which swr to connect to (assuming they are capable of that). the swr devices will have to identify a way to route externally, which i'm also not sure of. I can see several options, all of which have significant disadvantages and weird routing behaviour. does anyone have any suggestions on how to solve this situation? tim

1 0

Re: New Peering Point Mailing Lists
by Simon Lockhart 11 Oct '00

11 Oct '00

>BTW, it would be *real* good if the NAP's and ISP's can provide more >traffic statistics. It may help the customers to decide where and whom >to join, and help people who want to have a better understanding about >the network. I recall we used to have more traffic data a few years ago, >now, there is almost none in US. Currently, from what I have seen >recently, NORDUnet (http://www.nordu.net/) is the best ISP by far that >provides detailed traffic information.... I think you'll find that there's a few that are open (e.g. Abovenet http://www.above.net/traffic/ and ourselves [although we're not US based] http://support.bbc.co.uk/support/mrtg/internet/) A lot of the big corporates consider it confidential information (it gives away how good/bad they are or something). Simon -- Simon Lockhart | Tel: +44 (0)1737 839676 Internet Engineering Manager | Fax: +44 (0)1737 839516 BBC Internet Services | Email: Simon.Lockhart(a)bbc.co.uk Kingswood Warren,Tadworth,Surrey,UK | URL: http://support.bbc.co.uk/

2 1

Re: New Peering Point Mailing Lists
by Sean Donelan 11 Oct '00

11 Oct '00

Good idea. People interested in information about various exchange points, such as pricing, how much are cross-connects, who to contact, etc should start with http://www.ep.net/ It has links to essentially all of the functioning exchange points in the world. Most exchange points have web sites with contact information, and sometimes pricing and other details. I think more information is better than less information. On the other hand, it is very difficult to keep lots of different places up to date. Nevertheless, I'll end up subscribing to any list which has information. Disclaimer: I've run an XP in the past, and have a financial interest in companies operating several XPs. On Sun, 08 October 2000, Jeffrey Meltzer wrote: > Correct. I've created this list because in asking around a bit there > wasn't a resource for public discussions for those not currently in > the exchanges, and people seemed to not know where to post questions > to. (inet-access, nanog, isp-lists, etc). > > They're not necessarily just for NANOG type 'outage' messages, but for > general discussion on people looking for peers, finding out how much it > costs for a cable run, who to contact for info, and other general > chit-chat about the points.

2 1

Network Engineer (Cisco/StrataCom) Available
by Stephen K Davis 11 Oct '00

11 Oct '00

[If this information is not of interest to you, I apologize for the intrusion. As you draft your flame, please remember that I only sent a link and not an attachment. :) ] I am available for contract or sub-contract opportunities. Please find my resume at: http://www.geocities.com/skdavis HTML http://www.geocities.com/skdavis/skdres01.doc MS-Word Thanks, -Stephen -=*=-=*=-=*=-=*=-=*=-=*=-=*=- Stephen K Davis mailto:SKDavis@ParkAveTech.com Park Avenue Technologies, Inc. http://www.ParkAveTech.com 817-252-0792 office 817-252-0724 fax "This company is pledged to settle customer disputes through an arbitration program operated by the Better Business Bureau at Fort Worth"

2 1

Down UUNet DS3 in the SF area...
by Craig Holland 09 Oct '00

09 Oct '00

UUNet is having problems with a DS3 to the San Francisco area. All they would tell me is the TT # is 293854 and the remote router is in the LAX area. craig

1 0

Re: New Peering Point Mailing Lists
by Sean Donelan 09 Oct '00

09 Oct '00

On Sun, 08 October 2000, Jeffrey Meltzer wrote: > There doesn't seem to be any mailing lists existing specifically > for the public Peering Points. So, I've created a few. In my experience every public peering point has a mailing list associated with it. I've been subscribed to mailing lists for mae-east, mae-west, all the NAPs with address space managed by Bill Manning and ISI. I'm less familar with mailing lists for XPs outside the US, but most also have a mailing list associated with each exchange point. Sometimes the XP operator makes a big deal of only allowing certain people to subscribe to their list. But it is rarely difficult to find a member willing to forward messages. I'm subscribed to relatively few lists now, but I have messages forwarded to me all the time. I would prefer to subscribe directly, if only so I could manage my aliases better. But I understand a long time ago the XP operator promised its membership it would only allow other members to subscribe to the list, so now they feel they can't change the rules.

4 4

Re: New Peering Point Mailing Lists
by Sean Donelan 09 Oct '00

09 Oct '00

On Sun, 08 October 2000, Simon Lockhart wrote: > Where I see benefit in having independant lists about NAPs (in general, I > think it's a bit premature to have separate lists for each NAP) if for > gaining preliminary information about the NAP, for those people > considering joining. Historically, NAP disccussion took place on a relatively few mailing lists. In The Beginning.... all the public discussions took place on com-priv(a)psi.net com-priv has been pretty much defunct for a few years. Later the discussions migrated to various regional mailing lists such as nanog and ripe. Later they seperated into different groups, for "big" isps and for "tiny" isps. Small ISPs generally ask the questions on the mailing lists they already use, such as inet-access. Big ISPs have private mailing lists, but they don't say much. The problem, such as what happened when someone set up a whole bunch of new york colo lists, is the subject areas really aren't that different. So people start cross-posting to all the lists, which means subscribers such as myself who subscribed to all of them, receive the same message over and over and over again. Just like when the first message in this thread was cross-posted to multiple lists. If you feel there is a need for a public nap mailing list, please create one. Please don't create lots of new tiny lists.

3 2

New Peering Point Mailing Lists
by Jeffrey Meltzer 09 Oct '00

09 Oct '00

Hello Everyone, There doesn't seem to be any mailing lists existing specifically for the public Peering Points. So, I've created a few. I've setup seperate lists for most of the major Peering Points in the US (Mae East, Mae West, NYIIX, etc), as well as lists for Europe and Asia-Pacific, and a general discussion list (11 lists in total). The lists are for anyone that is currently peering at these Exchange Points, or are thinking of doing so, and looking for more information. I've created some aliases to make it easier to subscribe to multiple lists. Check it out at http://www.publicnap.net Jeff

2 1

Re: Power Outage in Chicago Loop
by Sean Donelan 09 Oct '00

09 Oct '00

I asked the "top" four ISP's, and none reported any problems nor any other affects due to the power outage. None reported any power loss at all. The Internet performance graphs at Matrix shows the world-wide packet loss for the Internet and WWW sites doubled at the time of the Chicago Loop lost power. The packet loss increased from about 3.9% to 7.9%. http://average.miq.net/ The power anomaly (aka blackout) began at 12:45 CDT when a circuit breaker exploded at the 828 S Jefferson Street Commonwealth Edison power station, the resulting fire also damaged other breakers in the station. Power was interrupted between 24th and Chicago Ave; and Lake Shore Drive and Ashland Ave. Power was fully restored shortly after 6pm CDT. I believe this is the same substation which had problems last year. The Associated Press reported 12,000 direct customers were affected. The Chicago Tribune has a picture of the substation and fire on their web site. http://www.chicagotribune.com/news/metro/chicago/article/0,2669,ART-47366,F…

2 1