June 2011 - Test - lists.nanog.org

coresite in reston?
by Christopher Morrow 22 Jun '11

22 Jun '11

<http://bits.blogs.nytimes.com/2011/06/21/f-b-i-seizes-web-servers-knocking-…> apparently had a bad day? fbi raid? (happy to lend a hand if folk who aren't local need assistance getting things put back together... wkumari used to run a list at: newhere.com ) -chris

1 0

Re: ICANN to allow commercial gTLDs
by brunner＠nic-naa.net 21 Jun '11

21 Jun '11

ray, > ... only trust ".band" and that ".com" et. al. are "less secure". "secure" is not a well-defined term. as the .com registry access model accepts credit card fraud risk, a hypothetical registry, say .giro, with wholesale registration at the same dollar price point but an access mechanism accepting less risk than credit card fraud would have less "insecure" registration events. as john levine pointed out, the hstld advisory group attempted to address a property of "zone file(s)." as a member of that advisory group i made public comments on the issues, technical and process, it encountered. > With a $185,000 application fee this tends to really kill small > businesses and conditions the public to favor ecommerce with the > giants, not to mention a nice revenue boost for ICANN. > > Would love to hear the dirt on backroom conversations that led to this > decision... a mainer has been invovled in policy development since, before there was an icann. a vermonter is on the current icann board. when looking for root causes, while the policy recommendation made by the policy development body did not restrict the implementation of the new gtld application process to a single event, staff adversity to law suit risk precluded distinguishing between types of applications based on policy -- say "high policy" applications like the original sponsored applications before "low policy" applications like the original standard applications -- and evaluating one type before the other. i suggest to you that institutional risk adversity (there exists a litigation history with the legacy monopoly operator) is the answer to questions of the form "wny one single, indivisible, wicked expensive, evaluation process for all?" > ... there will be enough public outcry to reverse it... but I'm > not optimistic. i would prefer "participation" over "outcry", and the act of "involvment" seems to be more on point than the mental state of being "optimistic", but milage always varies. on thursday there will be a text from the governmental, and the at large, advisory groups, on applicant support from developing economies. -e

3 2

Re: ICANN to allow commercial gTLDs
by brunner＠nic-naa.net 21 Jun '11

21 Jun '11

> I was talking about public perception and the ability to change it > through marketing; not any actual security. > > It's like the difference between ".com" and ".biz", "people" don't > understand when something isn't a ".com" and don't trust it. When I > say "people" I'm talking about the average non-technical consumer. > > That is all. it isn't likely that we could arrive at a useful definition of some subjective view held by others, so a discussion of "security" as a belief held by others is a waste of list and subscribers. there are broad claims: o zone file security requires registrant data correctness, a claim advanced by the set of actors advancing "WHOIS" policy at icann and elsewhere, o zone file security requires digital signature and little else, a claim implicit in the broad advocacy of dnssec and no other requirement relating to zone files, o zone file security requires registry applicant vetting and little else, a claim implicit in the broad advocacy for vetting and no other requirement relating to zone files, o zone file security requires registry security, and that would be me, channeling for roland on availability as an architecture consequence, followed by the other two legs of the cannonical CIA triad, for distributed systems. pick one or write your own and run with it. the folks at nominet did something useful and generous during the life in hell of the hstld ag, they attempted to cost compliance to the baroque set of requirements the aba/bits salted the mine with before there was an agenda, and the whois fanatics added. they came up with a six figure sum, which if applied to all registries, would put a nice "gold standard" logo on .com, and kill off (the whois fanatics wanted it mandatory on all new registries) all of the registries with less than that amount in wasteful excess annual income. -e

1 0

VMware ESX LACP Support
by Manu Chao 21 Jun '11

21 Jun '11

I would like to design VSS LACP based MECs with ESX hosts. Does VMware ESX support LACP? Do we need Nexus 1000 for ESX LACP support? R/ Manu

6 5

Re: ICANN to allow commercial gTLDs
by Paul Vixie 21 Jun '11

21 Jun '11

David Conrad <drc(a)virtualized.org> writes: > I believe the root server operators have stated (the equivalent of) that > it is not their job to make editorial decisions on what the root zone > contains. They distribute what the ICANN/NTIA/Verisign gestalt > publishes. yes. for one example, see: http://www.icann.org/en/announcements/announcement-04jan08.htm other rootops who have spoken about this have said similar/compatible things. -- Paul Vixie KI6YSY

14 29

Address Assignment Question
by Steve Richardson 21 Jun '11

21 Jun '11

Hello NANOG, I work for a medium-sized ISP with our own ARIN assignments (several /18 and /19 netblocks) and I've got a question about a possibly dubious customer request. I know a lot of you have experience on a much grander scale than myself, so I'm looking for some good advice. We have a customer who, over the years, has amassed several small subnet assignments from us for their colo. They are an email marketer. They have requested these assignments in as many discontiguous netblocks as we can manage. They are now asking for more addresses (a /24s worth) in even more discontiguous blocks. What I'd like to know is whether there is a legitimate use for so many addresses in discontiguous networks besides spam? I am trying my best to give them the benefit of the doubt here, because they do work directly with Spamhaus to not be listed (I realize reasons on both sides why this could be) and searches on Google and spam newsgroups for their highest traffic email domains yield next to nothing, given the amount of email they say they send out. I strongly believe that their given justification for so many addresses is not a good one (many addresses on an MTA, off-chance one gets blocked, etc), especially now that IPv4 addresses are becoming more of a scarce resource. However, if they *are* legitimate, which certainly is possible, are discontiguous networks a common practice for even legit operators, as it's quite likely that even legit email marketers will end up being blocked because someone accidentally hit 'Spam' instead of 'Delete' in their AOL software? Thanks, steve Note: I hate spammers as much as anyone out there, but I *do* know that not everyone who sends out massive amounts of email is a spammer. While it's possible they don't deserve it, I'm trying to give my customer the benefit of the doubt.

22 38

Re: ICANN to allow commercial gTLDs
by brunner＠nic-naa.net 21 Jun '11

21 Jun '11

> 185K is just the application few, the process includes some > requirements to have a given amount of dough for operations in escrow, > add what you need to pay attorneys, "experts" > , lobbyists, and setup and staff a small corporation even if you plan > to outsource part of the dayt-2-day operations to a back end operator, > you can easy reach a $2M figure just to start playing on that level. correct that the $185k is just the application fee, and there is a schedule of additional fees which may be applicable, e.g., extended evaluation if the application fails some evaluation criteria, and objections should any party with standing pay a fee to file an objection, etc., the continuity cost is undefined and the subject of lengthy communication between primarily myself and icann staff on the tdg-legal mailing list, though a vendor also commented on an aspect of continuity there as well. opinions on what a dead registry has to do (necessary functions) while waiting to time out and go "poof" or find a new operator sought. clue, in units of business organizations, finance, human resources, leasing, (law) and consensus policy, existing and advocacy, and registry constituency issues (icann policy), and registry and registrar contractual (law) issues are useful. the if-outsource-then-2m back-of-an-bar-napkin figure may be beer dampened, as standard offers by platform operators offer revenue sharing (whether all are equally exploitive of applicants as eventual registry contract holders is a question of projection, expectation, and taste) terms, resulting in significantly lower initial cost-to-acquire outcomes. a data point is .cat, which started with two thousand euros as its total marketing budget, and what i personally consider reasonable commercial terms from one non-profit (platform operator) to another (registry contract holding foundation). another is .museum, which started on a desktop in a kitchen. comments critical of their post-initial investment outcome should, though rarely are, be informed by restrictions placed upon the respective access to registrants, a subject just commented upon recently by both the antitrust division of the department of justice, and the european commission. a counter-example of course is the public record of a 2000 round standard gtld applicant, which spent approximately $20m on its buildout, before down-sizing its head-count by approximately 100 in late 2001, and which ten years of operational art has obtained an approximately 4% market share of generic names. > Yesterday's vote was just for the cameras, the program was approved > long time ago, the staff just got the directive to start implementing, > but there are still many holes in the process. broadly true, though missing both the ongoing gac-board dynamic, and the senior leadership transition(s) as driving this particular date and place of announcement. > All those part now the ICANN ecosystem are celebrating in fantastic > parties while the developing world supposedly has to be happy because > they reserved $2M (when the organization has at least a $70M/yr > running budget) for assistance... this is still an area of active work, i was working on it ... yesterday and the day before, today, and tomorrow and the day after tomorrow ... there will be a gac+alac statement in about 24 hours from now on the subject of support for applicants meeting the qualifications contained in the mile stone 2 report of the joint application support working group. it is a matter of record that i contribute to the jas-wg, and that the alac formed a drafting team to work with the gac, and that i am one of the members of that drafting team. what i'd like to see is some "i can help" notes cluttering my inbox, or the list -- real soon now this is going to be blades or 1us in cages with 4 and 6 provisioning and cash and clue (see above). hell, the only reason i'm in singapore is that a high-status privacy-only issue advocacy guy quit and left a unit of travel support for re-allocation. i don't want to have to walk to senegal in three months time to finalize the forms of support available to applicants, and icann wasn't funding me, they were funding the guy who quit. -e

2 1

Re: unqualified domains, was ICANN to allow commercial gTLDs
by brunner＠nic-naa.net 20 Jun '11

20 Jun '11

> Another avenue could be At-Large. The North American Regional At-Large > Organization (NARALO) - uniquely amongst the RALO's - accepts individual > members. as the elected unaffiliated member representative (or "umr") i suppose i should point out that (a) yes, the structural feature of individual membership exists in the naralo, and (b) it is unique to this ralo, and (c) these members do elect an officer to the ralo leadership, in some cases by accliamation or apathy, depending upon point of view, and (d) redundently, i am that stuckee. points (c) and (d) are not terribly important to the issue of how any number of persons having no other "at large structure" (als) membership, say a local isoc chapter, may, if they choose, lobby for what they each, jointly or severally -- to express involvement as a liability -- think is in the public interest. i simply mention (c) and (d) for completeness. i do have a caveat to offer. when i switched from the contracted parties to the naralo mailing lists i found a "technical" working group and hoped right on over. i foud that its purpose was not to provide a venue for the technical evaluation of policy issues, such as the sanity of v6-uber-alles as a non-negotiable requirement for new registries located where there is no v6, but to educate others. at that point i hoped right out. i don't think "policy for dummies" is any more attractive than "tech for dumies" as process and competency models. > http://naralo.org as joly's comment implies, there's a link to click, and consequences in the form of works, not faith. -e

1 0

Cogent depeers ESnet
by Nick Hilliard 20 Jun '11

20 Jun '11

Slightly old news, but it looks like Cogent depeered ESnet last week: > http://www.es.net/news-and-publications/esnet-news/2011/important-status-an… Current traceroutes indicate that ESnet is reaching Cogent via 6939_1299. In other news, automatically dropping interconnects at around the time of your HQ's close of business while your peering co-ordinator is on vacation seems to be the new gold standard. Nick

7 10

So... is it time to do IPv6 day monthy yet?
by Jay Ashworth 20 Jun '11

20 Jun '11

It certainly sounds like it might be. Cheers, -- jra -- Jay R. Ashworth Baylink jra(a)baylink.com Designer The Things I Think RFC 2100 Ashworth & Associates http://baylink.pitas.com 2000 Land Rover DII St Petersburg FL USA http://photo.imageinc.us +1 727 647 1274

14 21