- Test - lists.nanog.org

Re: Hey ISC, thanks for providing free wifi to intercage!
by hobbit＠avian.org 02 Oct '08

02 Oct '08

see, *this* is why google needs to WAKE UP and start putting real headers in their gmail spew. _H*

1 0

paix palo attitude
by Randy Bush 02 Oct '08

02 Oct '08

we're looking at dropping 2/3 of a rack of routing research measurement kit into paix palo alto, and have some questions. considering the price of space and power, we want to kinda make sure we'll get the data we need. we are looking specifically for a large amount of bgp peering data, we hope that folk will give us bgp peering though we expect no in- or out-bound traffic. think analogously to route views, though the kit on our end will be grinding the beans much more finely, espressoBGP :). pnis are expensive. if we plug into the peering mesh, are any larger folk on it? if not, would we be better off at the linx, at de-cix, at amsix, in stockholm? private replies are probably better as this is not of general interest. thanks. randy

2 2

once again, network hardware vendors spamming...
by Robert E. Seastrom 01 Oct '08

01 Oct '08

Anyone else gotten mail from John Lucania at atlantixglobal.com today? He lists his phone numbers as: 770.582.7248 Direct and 404.287.2603 Cell Why not give him a ring (preferably on his cell phone, maybe tonight?) and tell him what you think of spammers? :-) -r

6 5

RE: Google's PUE
by Skywing 01 Oct '08

01 Oct '08

Maybe, but I suspect that it is more complex than that. Most of the real environmental costs are still externalized in today's day and age. - S -----Original Message----- From: Deepak Jain <deepak(a)ai.net> Sent: Wednesday, October 01, 2008 15:08 To: Patrick W. Gilmore <patrick(a)ianai.net> Cc: NANOG list <nanog(a)nanog.org> Subject: Re: Google's PUE Patrick W. Gilmore wrote: > On Oct 1, 2008, at 2:04 PM, Martin Hannigan wrote: > >>> Personally, I think only a self-owned DC could get that low. A >>> general purpose DC would have too many inefficiencies since someone >>> like Equinix must have randomly sized cages, routers and servers, >>> custom-built suites, etc. By owning both sides, GOOG gets a boost. >>> But it's still frickin' amazing, IMHO. >> >> I wonder what it cost? :-) > > What cost to the environment of not doing it? > > OK, green hat off. :) Seriously, I doubt GOOG isn't seeing serious > savings from this over time. If they weren't why would they do it? > Not talking down this PR release.... Without comparing locations, sizes of floor plates, etc. I am sure Google has more than A-F, so one has to wonder which data centers they left off the map. I think I can submit without proof that a PUE of 1.2 is far more impressive in New Mexico or Arizona than it is in Vancouver, BC since you are essentially measuring the energy to keep the datacenter at temperature throughout seasonal (or external) ambient heat deltas. Likewise, a 10,000 sq ft single customer DC is far less impressive than a 200,000 sq ft general purpose (colo) DC. (they say large scale -- is that number of cores, or sq ft?, but I don't have a number for that). And to address Patrick's rhetorical question - if it costs you $400MM (like a datacenter proposed underground in Japan) to save $15MM / yr in energy costs, one could easily argue that the environment "savings" is not sufficient to overcome the upfront investment. If you spent $40MM in trees (instead of $400MM on an investment to save $15MM/yr), you could argue the environment would be far better off. Deepak

1 0

About.com/NYTimes admins about?
by Christopher Morrow 01 Oct '08

01 Oct '08

Is there perhaps an about.com/nytimes.com admin around? I was wondering if they perhaps knew that their loadbalancer for www.nytimes.com is fairly broken wrt answering AAAA queries: (who's NS for nytimes.com) dig NS nytimes.com +short ns1t.nytimes.com. nydns2.about.com. nydns1.about.com. (who do they think is the NS for www.nytimes.com) dig www.nytimes.com @ns1t.nytimes.com. NS ;; QUESTION SECTION: ;www.nytimes.com. IN NS ;; AUTHORITY SECTION: www.nytimes.com. 60 IN NS nss1.sea1.nytimes.com. www.nytimes.com. 60 IN NS nss1.lga2.nytimes.com. (what is the AAAA for www.nytimes.com ?? ) dig www.nytimes.com @nss1.sea1.nytimes.com. AAAA ;www.nytimes.com. IN AAAA ;; AUTHORITY SECTION: . 3600000 IN NS k.root-servers.net. . 3600000 IN NS l.root-servers.net. . 3600000 IN NS m.root-servers.net. . 3600000 IN NS a.root-servers.net. . 3600000 IN NS b.root-servers.net. . 3600000 IN NS c.root-servers.net. . 3600000 IN NS d.root-servers.net. . 3600000 IN NS e.root-servers.net. . 3600000 IN NS f.root-servers.net. . 3600000 IN NS g.root-servers.net. . 3600000 IN NS h.root-servers.net. . 3600000 IN NS i.root-servers.net. . 3600000 IN NS j.root-servers.net. ;; ADDITIONAL SECTION: k.root-servers.net. 3600000 IN A 193.0.14.129 l.root-servers.net. 3600000 IN A 198.32.64.12 m.root-servers.net. 3600000 IN A 202.12.27.33 ;; Query time: 89 msec ;; SERVER: 170.149.172.35#53(170.149.172.35) wha??? <ricky voice>Lucy, your loadbalancer is foobar'd</ricky voice> In an effort to make v6 things work a tad better in this hostile world, could the NYTimes folks let us know what sort of LB that is? and why it wants to not be a good Intenet Citizen?? -Chris

4 6

DSL at MAE-East
by George William Herbert 01 Oct '08

01 Oct '08

Is there anyone who has successfully gotten DSL for out of band back channel into the MAE-East building? We're being told it's impossible, and somehow this strikes me as somewhat improbable... Thanks. -george william herbert gherbert(a)retro.com

7 9

Exchange point tax-exempt status
by Chris Caputo 01 Oct '08

01 Oct '08

The Seattle IX (SIX) recently applied for and received an IRS tax-exemption due to its operation as a non-profit exchange point. In case there are any other non-profit exchange points that want to take this route, feel free to copy the relevant parts from our filing: http://www.seattleix.net/docs/20080703_IRS_Non_Profit_Filing.pdf Apparently the filing was sufficient, as there was no back and forth with the IRS. Three months passed and the IRS granted the exemption, per: http://www.seattleix.net/docs/20080911_IRS_501(c)(6)_Exemption_Letter.pdf Enjoy, Chris SIX Secretary/Treasurer

1 0

rackmount managed PDUs
by Justin M. Streiner 30 Sep '08

30 Sep '08

As much as I hate to tear people away from the Intercage/Atrivo debacle and semi-tangential rants, I'll take one for the team and do it :) I have an opportunity coming up to rebuild an existing machine room space to an extent. It's not a total gut-and-refit, but I'll at least get to put in some new infrastructure. That said, I'd be interested in hearing about peoples' experiences with various rackmountable managed PDUs. I have some Tripp Lite PDUMH30NETs that work well and are reasonably priced, but they have a few quirks (no RS-232 console port, web interface seems to be a little shaky with Firefox, etc) that would become more annoying when scaled up to several rows of new rack footprints. I'm also open to using managed vertically mounted PDUs. The plan is for each footprint to have "A" and B" feeds, so two PDUMH30NETs would take up 4U per footprint, which is a bit much... I don't need to worry about distributing DC power - just AC. This site will be lights-out most of the time, so robust remote management capabilities are a must. Any thoughts/insight are greatly appreciated. jms

15 16

RE: [nanog] Advice in dealing with BGP prefix hijacking
by mack 29 Sep '08

29 Sep '08

First I would contact them, you have obviously done this and it didn't work since it happened again. Second I would contact their major transit providers, assuming they aren't a major transit provider themselves. If you share mutual transit providers then you will be most likely to get satisfactory results. If you aren't a customer of their transit provider then you may not get anywhere. Third you can contact ARIN to talk to the perpetrator and/or their transit providers. This is one of their primary functions. The final option is legal. IANAL (usual disclaimer) CMU probably has a legal department that may be able to request an injunction or file suit for damages. People tend to forget that when prefixes are hijacked there is legal recourse after the fact. Of course with the international nature of the internet legal recourse may be difficult or impossible. In this case with a US based ISP, the legal remedy might be feasible. This is long slow and doesn't help at all during an incident, but it will probably keep it from happening again. ISPs are businesses and when it hits the pocket book they pay attention. -- LR Mack McBride Network Administrator Alpha Red, Inc. > > Message: 2 > Date: Mon, 29 Sep 2008 11:10:48 -0400 > From: "L. Gabriel Somlo" <mikebenden(a)gmail.com> > Subject: Advice in dealing with BGP prefix hijacking > To: nanog(a)nanog.org > Message-ID: <20080929151048.GA19783(a)hedwig.net.cmu.edu> > Content-Type: text/plain; charset=us-ascii > > Hi, > > I'm looking for advice on how to deal with a US ISP (names withheld to > protect the guilty) which seems to repeatedly fat-finger our IP space > into their BGP announcements, and then gives us the run-around when we > call them on the phone, since we're not their clients... > > Happened twice already, took several hours to fix each time, and I'm > looking for any ideas, opinions, and war stories that would help me > figure out how to discourage them from repeating this type of thing > in the future... > > Thaks much, > > Mike > >

1 0

Re: BGP for disaster recovery site
by Scott Weeks 29 Sep '08

29 Sep '08

Inline... --- CHRISTINE.M.BERNS(a)sargentlundy.com wrote: We currently have a routable block (class B) of IP addresses. We are in the process of designing a disaster recovery site. Our main site is already dual homed to two different Internet service providers via BGP. ------------------------------------------------ First, unless the first two bits of the first octet are "10" it's not a "Class B". It's a /16, which is the same size as a "Class B" but the first two bits of the first octet are not necessarily "10". Are these two sites (main and DR) in the same ASN? ------------------------------------------------- A consultant told us that in order to allow us to test access to the DR site without affecting the production environment, we should get another block of addresses from ARIN and advertise those addresses out the DR site's Internet connection. Can we even expect to get another block from ARIN if we already have a class B... ------------------------------------------------- I would consider another consultant if you have not used 80% of your current allocation. ARIN will not allocate anymore space until you hit 80% utilization ------------------------------------------------- ...and could we not accomplish the same thing by advertising a subnet of our existing Class B at the DR site? I would actually prefer to advertise a subnet of our class B, but am wondering if there are any reasons why this is not a good idea. ------------------------------------------------- If the two sites are not in the same AS it's not a good idea unless you're experienced with such things. See past posts in the archives about the subject. ----------------------------------------------- Also, I have seen reference to some Internet service providers possibly not accepting /24 BGP routes and either dropping them or aggregating them to a /21 or /20 or /19. Are there recommendations as to what is the longest prefix that we should advertise to guarantee that the prefix will be advertised throughout the Internet? --------------------------------------------------- Nothing is ever guaranteed WRT routing certain prefix sizes on the internet, but you should have no problems with a /24. Any longer (a /25) and you will not have good connectivity on the internet. scott

1 0