April 2008 - NANOG - lists.nanog.org

AS7132 issues in Chicago to Cogent and Teleglobe ?
by Mike Tancsa 30 Apr '08

30 Apr '08

One of my customers (out of AS11647) is seeing a lot of packet loss to their office in SBC's network (AS7132). The site is behind one of SBC's BASes (67.36.65.16). Pinging 67.36.65.16 when the path is through Cogent and Teleglobe is showing periodic packet loss. It will be clean, and then all of a sudden I cant ping 67.36.65.16 (and anything past it) for about 15 seconds and then its clean again.... Both my Teleglobe and Cogent paths meet SBC/ATT in Chicago, so I wonder if there is a common fault there or near there ? My customer's US operations tried opening a ticket but was not able to get any info to SBC/ATT past first level support. Anyone know of any issues ? ---Mike -------------------------------------------------------------------- Mike Tancsa, tel +1 519 651 3400 Sentex Communications, mike(a)sentex.net Providing Internet since 1994 www.sentex.net Cambridge, Ontario Canada www.sentex.net/mike _______________________________________________ NANOG mailing list NANOG(a)nanog.org http://mailman.nanog.org/mailman/listinfo/nanog

2 1

AT&T DSL problems?
by Roy 30 Apr '08

30 Apr '08

I am hearing of problems from AT&T DSL customer to some destinations on the Internet. Is anyone else hearing these reports? _______________________________________________ NANOG mailing list NANOG(a)nanog.org http://mailman.nanog.org/mailman/listinfo/nanog

4 3

Re: [NANOG] AT&T DSL problems?
by Chris Riling 30 Apr '08

30 Apr '08

Forgot to copy the list on this one... Chris ---------- Forwarded message ---------- From: Chris Riling <criling(a)gmail.com> Date: Apr 30, 2008 2:14 PM Subject: Re: [NANOG] AT&T DSL problems? To: Roy <r.engehausen(a)gmail.com> I've had a couple reports from some customers that have eluded to AT&T having routing problems in general... I think our call center confirmed this with AT&T as well... Chris On 4/30/08, Roy <r.engehausen(a)gmail.com> wrote: > > I am hearing of problems from AT&T DSL customer to some destinations on > the Internet. Is anyone else hearing these reports? > > _______________________________________________ > NANOG mailing list > NANOG(a)nanog.org > http://mailman.nanog.org/mailman/listinfo/nanog > _______________________________________________ NANOG mailing list NANOG(a)nanog.org http://mailman.nanog.org/mailman/listinfo/nanog

1 0

Purpose of Internap's PNET AS22212
by Thurber 29 Apr '08

29 Apr '08

Can anybody shed some light on Internap's PNET AS22212? Specifaly how it relates to their PNAP architecture? Is Internap now doing peering? I was under the impression that their entire business model was based around isolated PNAPs and being a backboneless provider. Attempts at getting an explanation from Internap have been fruitless. CT _______________________________________________ NANOG mailing list NANOG(a)nanog.org http://mailman.nanog.org/mailman/listinfo/nanog

5 4

Re: [NANOG] DWDM
by Frank Coluccio 29 Apr '08

29 Apr '08

ps - a 2007 updated prezo on SURFnet6's DWDM network presented by Erik-Jan Bos: http://www.ces.net/doc/seminars/cef2007/p/bos.ppt -- On Tue Apr 29 15:05 , Frank Coluccio sent: >re: "the latest CPL technology from Nortel is a real breakthrough, for once a >vendor has listened and understood the challenges for operators and actively >addressed our concerns." > >When CPL was appeared I found its architecture refreshingly straightforward, as >you've also noted, which was best demonstrated (to me, at least) by Kees Neggers >within SURFnet6 throughout the Netherlands about three years ago (thus allowing >him to offload boatloads of Layer3 gear). > >I'm not sure which issue of the Cook Report it was in, but the details of this >SURFnet6 deployment were captured in an interview of Neggers quite nicely by >Gordon Cook in one of his reports during that era. > >During the intervening three years, however, advancements in dispersion >management have been introduced and opened the door for some new players, such as >Infinera, whose use of electronic dispersion compensation (EDC) in its WDM >platforms has now given it somewhat of an edge, or so it seems at the moment, >based on the market share it has gained in a very short time. I'm hardly >suggesting that Infinera trumps the former CPL approach across all areas of >design consideration, but only that it's an approach that merits investigation. > >I should note that Infinera went diametrically against the grain of the >"all-optical" holy grail that is commonly sought by many, and actually leverages >its ability to perform functions in the 'electronic' domain a lot more >efficiently than other vendors have been capable of achieving in the purely >optical domain, thus positioning itself in a category that's pretty much all its >own today. I'd be eager to read comments and criticisms from anyone who's had >actual experience with its wares. > >Frank A. Coluccio >DTI Consulting Inc. >212-587-8150 Office >347-526-6788 Mobile > >On Fri Apr 25 9:31 , "Neil J. McRae" sent: > >>The current technologies for DWDM have really made it technology >>that's reasonably straight forward to deploy. The last >>generation was a nightmare! >> >>Tuneable optics, dispersion compensation, and ROADM have made >>a substantial difference to deploying and operating DWDM networks. >> >>I had experience with the former generation system from Nortel >>which although very reliable it was very resource intensive >>to deploy new services, the latest CPL technology from Nortel >>is a real breakthrough, for once a vendor has listened and understood >>the challenges for operators and actively addressed our concerns. >> >>-----Original Message----- >>From: Scott E. MacKenzie [smackenzie(a)qf.org.qa','','','')">smackenzie(a)qf.org.qa] >>Sent: Friday, April 25, 2008 4:00 AM >>To: NANOG >>Subject: [NANOG] DWDM >> >> >>Does anyone know where I can locate a list of DWDM networks deployed for >>Education, Science & Research, and Commercialization? >> >>We need to determine the practicality of DWDM use... >> >> >>Scott >> >> >> >>_______________________________________________ >>NANOG mailing list >>NANOG(a)nanog.org >>http://mailman.nanog.org/mailman/listinfo/nanog > > > >_______________________________________________ >NANOG mailing list >NANOG(a)nanog.org >http://mailman.nanog.org/mailman/listinfo/nanog _______________________________________________ NANOG mailing list NANOG(a)nanog.org http://mailman.nanog.org/mailman/listinfo/nanog

2 1

Re: [NANOG] DWDM
by Frank Coluccio 29 Apr '08

29 Apr '08

re: "the latest CPL technology from Nortel is a real breakthrough, for once a vendor has listened and understood the challenges for operators and actively addressed our concerns." When CPL was appeared I found its architecture refreshingly straightforward, as you've also noted, which was best demonstrated (to me, at least) by Kees Neggers within SURFnet6 throughout the Netherlands about three years ago (thus allowing him to offload boatloads of Layer3 gear). I'm not sure which issue of the Cook Report it was in, but the details of this SURFnet6 deployment were captured in an interview of Neggers quite nicely by Gordon Cook in one of his reports during that era. During the intervening three years, however, advancements in dispersion management have been introduced and opened the door for some new players, such as Infinera, whose use of electronic dispersion compensation (EDC) in its WDM platforms has now given it somewhat of an edge, or so it seems at the moment, based on the market share it has gained in a very short time. I'm hardly suggesting that Infinera trumps the former CPL approach across all areas of design consideration, but only that it's an approach that merits investigation. I should note that Infinera went diametrically against the grain of the "all-optical" holy grail that is commonly sought by many, and actually leverages its ability to perform functions in the 'electronic' domain a lot more efficiently than other vendors have been capable of achieving in the purely optical domain, thus positioning itself in a category that's pretty much all its own today. I'd be eager to read comments and criticisms from anyone who's had actual experience with its wares. Frank A. Coluccio DTI Consulting Inc. 212-587-8150 Office 347-526-6788 Mobile On Fri Apr 25 9:31 , "Neil J. McRae" sent: >The current technologies for DWDM have really made it technology >that's reasonably straight forward to deploy. The last >generation was a nightmare! > >Tuneable optics, dispersion compensation, and ROADM have made >a substantial difference to deploying and operating DWDM networks. > >I had experience with the former generation system from Nortel >which although very reliable it was very resource intensive >to deploy new services, the latest CPL technology from Nortel >is a real breakthrough, for once a vendor has listened and understood >the challenges for operators and actively addressed our concerns. > >-----Original Message----- >From: Scott E. MacKenzie [smackenzie(a)qf.org.qa','','','')">smackenzie(a)qf.org.qa] >Sent: Friday, April 25, 2008 4:00 AM >To: NANOG >Subject: [NANOG] DWDM > > >Does anyone know where I can locate a list of DWDM networks deployed for >Education, Science & Research, and Commercialization? > >We need to determine the practicality of DWDM use... > > >Scott > > > >_______________________________________________ >NANOG mailing list >NANOG(a)nanog.org >http://mailman.nanog.org/mailman/listinfo/nanog _______________________________________________ NANOG mailing list NANOG(a)nanog.org http://mailman.nanog.org/mailman/listinfo/nanog

1 0

Re: [NANOG] Level3 not honoring Broadwing contracts?
by David Hubbard 29 Apr '08

29 Apr '08

From: Chris Owen [mailto:owenc@hubris.net] > > It was clear from the conversation that he never considered for a > moment that we might actually pay the new rate. He just assumed we > were gone as soon as it happened. Obviously I don't have a business > degree because I don't understand the business model of buying up > business and then going out of your way to chase off their > customers. > The Level3 higher ups must see something I don't. > I think the idea is that raise the rates dramatically on everyone, those who stay, now you're making more profit off of them, those that go were probably small enough that they were able to easily leave so you're clearing up room for the customers who are too large to profitably move and who have no choice but to continue growing in the existing space at the new inflated rates. Can't say I like it as a customer since I'm on the receiving end of a re-rate that increased costs by 50%, but makes business sense obviously since I didn't leave, as I'm sure many others didn't either. If you use a lot of power or floor space though, and not massive amounts of bandwidth, their new rates are very close to making it profitable to build your own facility, maybe they just want to sell bandwidth now. David _______________________________________________ NANOG mailing list NANOG(a)nanog.org http://mailman.nanog.org/mailman/listinfo/nanog

1 0

Need Hananet Contact
by Giuseppe Ravasio 29 Apr '08

29 Apr '08

Hi, I'm having troubles accessing Hananet DNS (AS9318) from AS44567. Probably because our IP Address space (93.92 etc etc) was listed in a very old version of Bogon Networks Block List ( http://www.team-cymru.org/?sec=8&opt=25 ) I tried some Hananet Whois Contacts but I haven't had any feedback. Do someone has a hananet working contact? Best regards Giuseppe Ravasio _______________________________________________ NANOG mailing list NANOG(a)nanog.org http://mailman.nanog.org/mailman/listinfo/nanog

2 1

Juniper vs Foundry
by Andrew Matthews 28 Apr '08

28 Apr '08

I'm looking to replace some old aging extreme black diamonds. I run a semi decent size data center and I'm looking at a few options. right now we run cables from patch to patch to a 2 core switches. So it takes 3 patch cables to get it all the way back to the core from the customers rack/shared rack. Our new plan is to put a few switches per row. for a total of 8 rows. So two switches per row. Now here are my options. Foundry: Two core switches which do all layer 3, edge switches on racks doing layer 2 vlans up to the core switch. Juniper: two layer 2 switches for distro, using Virtual Chassis groups for each set of switches on rows which do layer 3. Yes my OSPF grows because now instead of 2 layer 3 switches i now i have 8 VC groups. Anyone have any suggestions? Anyone using foundry? Anyone using Juniper Switches. What do you think? I know the juniper are pretty new and i'm looking at for some some success stories, but i'm also looking for some pitfalls, i know juniper has been in the routing industry for a long time, but as for switches this is a new product for them. Thanks Andrew _______________________________________________ NANOG mailing list NANOG(a)nanog.org http://mailman.nanog.org/mailman/listinfo/nanog

1 0

OT: Planetary-Scale Views on a Large Instant-Messaging Network.
by Roland Dobbins 27 Apr '08

27 Apr '08

Interesting Microsoft Research piece on macro-scale user behaviors: <http://research.microsoft.com/~horvitz/leskovec_horvitz_www2008.pdf> ----------------------------------------------------------------------- Roland Dobbins <rdobbins(a)cisco.com> // +66.83.266.6344 mobile History is a great teacher, but it also lies with impunity. -- John Robb _______________________________________________ NANOG mailing list NANOG(a)nanog.org http://mailman.nanog.org/mailman/listinfo/nanog

1 0