- Test - lists.nanog.org

Re: Houston problems?
by Richard A. Steenbergen 12 Jun '01

12 Jun '01

On Mon, Jun 11, 2001 at 05:39:39PM -0700, Jason Legate wrote: > > I am getting word that the Qwest Houston TeraPOP is down, due to > flooding. > > Is this true, and does anyone know of anything else from down there? They mean water flooding Jason, not anything you'd be interested in. :P -- 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)

2 1

Re: non-op (Re: Definition of Tier-1)
by Richard A. Steenbergen 11 Jun '01

11 Jun '01

On Mon, Jun 11, 2001 at 09:46:05AM -0400, Travis Pugh wrote: > > The part that really drives me crazy is that nobody seems to have > played the "tier 2 and proud" card from a marketing standpoint. I can > think of a few reasons why I'd rather not be transit free right now, > and could probably successfully pitch those reasons to customers if I > wanted to change careers. InterNAP. Personally I think the whole "tier 1" craze is overrated. I'd rather have multiple good paths to my destination, and the ability to divert traffic elsewhere in the event of a problem. It can cost a lot of time and money to get all the peering you need in all locations (or at least enough to keep from bouncing traffic across the country because thats where your peer is, or thats where your private peer is, etc). Then again it does solve the problem of path selection by making it a non-issue, there is only one path. -- 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)

4 3

Wave length for OC48 : Summary
by Vincent Gillet 11 Jun '01

11 Jun '01

Regarding all the inputs i received for the request, here is a summary . POS card send laser to a specific wave length but used to have receivers with a large-band that include both 1310 and 1550 nm. . Wave length have nothing to do with laser power nor attenuators but 1550 nm frequency has better performance (less attenuation) in fiber. . Some people reported successful implementation with different wave-length on Cisco. . 1 guy reported successful implementation with Juniper / Cisco using oc48. (It was private mail, off-list) I will keep nanog-list informed for my own implementation !!! Thanks to everyone for the inputs. V

1 0

Re: Multicast Traffic on Backbones
by Sean Donelan 11 Jun '01

11 Jun '01

Folks have given a number of interesting multimedia applications using multicast, but I'm not sure any of them really need multicast. Akamia/iBeam/Skycache, or any other set of large memory buffer boxes around the network edge, seem to be a much better way to replicate multimedia streams, and even provide the popular VCR start/pause/stop effect. This works well for single sourced (or limited source) multimedia streams that have very little interaction with the recipient (aka TV for the Internet). Multicast seems to be more efficient for updating large buffer edge boxes than delivering the content directly to the end-user.

2 1

Cisco HA (High Availability) for Cat6509
by Steve Robinski 11 Jun '01

11 Jun '01

Does anyone on here have any experience with using the high availabity design/configuration of a Cisco Catalyst 6509? Please reply offline regarding your thoughts and experiences. Thanks -Steve Robinski _______________________________________________________ Send a cool gift with your E-Card http://www.bluemountain.com/giftcenter/

1 0

RE: /24s run amuck again
by Hank Nussbacher 11 Jun '01

11 Jun '01

At 14:10 09/06/01 -0400, Richard A. Steenbergen wrote: > > 11371 307 Rhythms NetConnections Email sent to gharmon(a)rhythms.net, cgreen(a)rhythms.net, doroberts(a)rhythms.net on Feb 8 - no response. > > 3491 651 CAIS Internet > >DSL providers are becoming very bad about this. Someone pointed out to me >off list that CAIS had carved up PSI's /8 into over 500 /24s. > > > 690 502 Merit Network > >Well at least we don't have to go too far to find the guilty party. :P > > > 18994 468 Global Crossing > > 15870 436 Global Center Frankfurt > > 18993 325 Global Crossing > >Those are the GlobalCenter datacenters being converted into the Exodus >network. It looks like they are leaking a sizable number of /32s /30s etc, >and since its GBLX space I'm assuming its stuff that used to be aggregated >into a single announcement. Email sent to: ipadmin(a)gblx.net, huberman(a)gblx.net, ip-eng(a)gblx.net, scarter(a)gblx.net, bgp(a)gblx.net, bp(a)gblx.net on June 4 - everyone responded that the problem was forwarded to Exodus and from there it disappeared into a black hole. Basically, after having sent out dozens of emails over the past 6 months I have come to the conclusion that there are a few out there that will fix things when presented with the problem. But the vast majority either don't have a clue, don't want to have a clue or couldn't give a damn that the routing tables are increasing in size. -Hank > > There is no attempt to measure aggregation - that's the job of the > > CIDR Report. This simply looks at the prefix announced and if it is > > outside the above limits, it is counted. Makes very interesting > > reading... > >The one interesting pattern I noticed in the rampant /24 abuse was non- >contiguous announcements. It's likely that this kept them off the CIDR >Report and any other scans which only looked for contiguous announcements. >For example: > >1.2.3.0/24 >1.2.5.0/24 >1.2.7/0.24 > >-- >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)

6 5

Re: Multicast Traffic on Backbones
by brandon＠rd.bbc.co.uk 11 Jun '01

11 Jun '01

> HOWEVER, most of the 24x7 media sites (CNN, etc.) are doing on-demand > video, rather than streaming a constant feed. 1) rights 2) unicast doesn't scale to the sort of audiences of radio/tv broadcasts and if you try it becomes financially unviable too (even with people attempting to fake multicast, e.g. Akamai) One way to manage both is to use the VCR model > On the one > hand, content providers aren't offering 24x7 multicast feeds because there > isn't enough multicast access at the end-points. We offer it regardless but there's not many can use it > Apart from 24x7 broadcast there isn't an obvious killer app. And there's reasonable argument over making the intenet = tv brandon

4 3

Re: Why so little traffic from C&W
by Sean Donelan 11 Jun '01

11 Jun '01

On Sun, 10 June 2001, Sean M. Doran wrote: > Rachel Warren <rachel(a)telco-bitch.com> quotes Bill Manning and replies: > | > There was never any governmental sanction of the term or > | > concept of tier anything associate with the NSFnet or its > | > transition. > | > | And this is definately a good thing. > > It's also not true. NSF 93-52 did indeed specify > precisely what type of backbone the funded regionals could > connect to, and it required those distinguished backbones > to interconnect in specific locations (the "priority" NAPs). > > The name "tier" wasn't used wrt eligible and ineligible > backbones, but there was an explicit ranking of preference by the > NSF that was based on specific interconnectivity, and that mapped > very cleanly onto the then tier-1 backbones, with the exception of > UUNET, which didn't play (and MFS's NAP wasn't even a "priority" one). The NSF only specified connectivity to the NAP, not interconnectivity between providers. Due to problems with various NAPs, who connected to what was always in flux. For example, the Chicago NAP was designated a "priority NAP." But networks service providers (NSPs) were very slow to connect to chicago. I believe ANS went through various periods of not connecting, connecting, and disconnecting at the Chicago NAP. MCI's connection to the chicago NAP didn't work for months at a time. Well, the MCI vBNS connection worked, but the InternetMCI connection didn't. There were some networks which didn't connect to either of the ATM NAPs (Chicago and San Francisco) because their engineers didn't like ATM. They only used the Gigaswitch based NAPs. UUNET and Genuity are examples of networks which declined to connect to the ATM networks for a long, long time. If I recall my history, PSI and AGIS were the first two networks to connect to all the priority and non-priority NAPs of the day (Chicago, New York, San Francisco, MAE-East and the de facto CIX router). A Sprint engineer was one of the most vocal opponents to the ATM naps.

1 0

Re: Why so little traffic from C&W
by smd＠clock.org 11 Jun '01

11 Jun '01

Rachel Warren <rachel(a)telco-bitch.com> quotes Bill Manning and replies: | > There was never any governmental sanction of the term or | > concept of tier anything associate with the NSFnet or its | > transition. | | And this is definately a good thing. It's also not true. NSF 93-52 did indeed specify precisely what type of backbone the funded regionals could connect to, and it required those distinguished backbones to interconnect in specific locations (the "priority" NAPs). The name "tier" wasn't used wrt eligible and ineligible backbones, but there was an explicit ranking of preference by the NSF that was based on specific interconnectivity, and that mapped very cleanly onto the then tier-1 backbones, with the exception of UUNET, which didn't play (and MFS's NAP wasn't even a "priority" one). | I don't think I would want to see my tax money being used by the | Department of Standards to define "Tier-1". Been there, done that, got the bruises (thanks pford & priscilla, I like the "I went to Danvers and all I got was a meeting behind my back" t-shirt souvenir). Sean.

1 0

RE: /24s run amuck again
by Dave Israel 10 Jun '01

10 Jun '01

[Rant below. My apologies. Sometimes, your spleen needs a-ventin'.] I know I'm behind on this conversation, but I want to say that, for the record, Bellovin's presentation, and the resultant "You're not paying me for my routing table, you bastard" sentiment, drives me up a wall. BGP is for setting and sharing policy. If you don't want to listen to somebody else's legitimate policy, the right answer is not to filter that policy; the right answer is not to peer. If you're not interested in policy decisions, you can get static routes by examining the routing arbiters or even the registries' records. Despite what some venders want you to believe, memory is cheap, processors that can handle the Internet routing table are cheap, and the only really expensive things to build in a router are the high speed line cards and the forwarding engine(s) to push them. If we were running the Internet on 68020 processors over 56k and T1 lines still, you would have a point. But the additional load of allowing other people to set their policy as they see fit is not a relevant concern on modern equipment. There are lots of legitimate reasons for making announcements like this. One is backup connectivity; if the primary fails, and you're filtering the backup, you might not be taking the best path to the end user. In fact, if you're well connected, you're passing it off to the primary, who, if he is filtering as well, won't pass it off to the secondary, and if he's not filtering, you're transiting when you don't need to. At a bare minimum, you're putting the decision off to your provider, in which case, why are you peering in the first place? Another is load sharing; if I have 60MB of traffic for my site, and I have two T3s, I might set policy to try to equalize that data. If you ignore my policy, you might overload one of the links and not be able to reach my site. I suffer, your customers suffer, eveybody gets pissy, because you didn't have 2k of RAM to share with a fellow member of the Internet whose down on his luck. You should do this to be a good neighbor. You should do this to be charatable. You should do this because you might need it yourself someday. Now, before anybody gets the wrong idea, yes, I know there are people who announce their supernets as /24s or smaller because of technical incompetance. The ISPs should do what they can to prevent that; I know that we have historically been rat bastards to our customers who wanted to make smaller announcements and couldn't tell us why. And there are people who are running old hardware, or are being victimized by 800% markup by the vendor, or otherwise really cannot spare the cycles and the RAM to support other people's policy. You gotta do what you gotta do. But hogging cycles you aren't using, and memory that is sitting idle, on some sort of "You're not paying me" principle, is just absurd. -Dave On 6/9/2001 at 13:47:15 -0700, Randy Bush said: > > >>> There are lots of reasons to have some diverse /22 announcements in your > >>> network for example. > >> but if you are not paying me, what reasons are there for me to spend > >> my resources (route bloat) so you can engineer your traffic? > > The fact that your other customers are paying you to reach everyone on the > > internet? > > red herring alert! the point is not reachability. the aggregate is still > available. see bellovin's phoenix presentation. > > randy -- Dave Israel Senior Manager, IP Backbone Intermedia Business Internet

1 0