- NANOG - lists.nanog.org

Re: NTP Issues Today
by Scott Weeks 19 Nov '12

19 Nov '12

On 11/19/12 6:32 PM, "Scott Weeks" <surfer(a)mauigateway.com> wrote: >--- vanwolfe(a)gmail.com wrote: >From: Van Wolfe <vanwolfe(a)gmail.com> > >Did anyone else experience issues with NTP today? We had our server >times update to the year 2000 at around 3:30 MT, then revert back to 2012. >----------------------------------------- >You need to provide more information. For example, what NTP >source are you using? ------------------------------------------ --- chaynes(a)centracomm.net wrote: From: Clay Haynes <chaynes(a)centracomm.net> I can confirm this had happened on one of my test servers - it was pointing to tick.usno.navy.mil and tock.usno.navy.mil at the time. ------------------------------------------- That's not a very diverse set of NTP servers. In the future if you think it might be an outage, you might try on the 'outages' list: http://puck.nether.net/mailman/listinfo/outages For this one, you might ask the server contact if there was a problem: http://support.ntp.org/bin/view/Servers/TockUsnoNavyMil That assumes you've done your homework first and made sure it wasn't something in your network. scott

1 0

Re: NTP Issues Today
by Scott Weeks 19 Nov '12

19 Nov '12

--- vanwolfe(a)gmail.com wrote: From: Van Wolfe <vanwolfe(a)gmail.com> Did anyone else experience issues with NTP today? We had our server times update to the year 2000 at around 3:30 MT, then revert back to 2012. ----------------------------------------- You need to provide more information. For example, what NTP source are you using? scott

2 1

Re: Google/Youtube problems
by Nick Olsen 19 Nov '12

19 Nov '12

I stand corrected. That's what I get for going off memory. Nick Olsen Network Operations (855) FLSPEED x106 ---------------------------------------- From: "Scott Whyte" <swhyte(a)gmail.com> Sent: Monday, November 19, 2012 4:48 PM To: nick(a)flhsi.com Subject: Re: Google/Youtube problems On Mon, Nov 19, 2012 at 11:10 AM, Nick Olsen <nick(a)flhsi.com> wrote: > I think this would be true if they offered some form of paid peering. > > Google want's a good fast route to your customers, And your customers want > a good fast route to Google. > > IF Google ran its transit at or near congestion. This could degrade your > customers performance. After so long, You'd contact Google and attempt to > troubleshoot. And they would say if you want good peering with them, You > should pay them to peer. Where you could control just how much traffic was > on your port and expand it if needed. Pretty sure this was Comcast and > level3/Netflix did. But Comcast had the winning leverage (more eyeballs) in > the discussion. > > But, I don't think Google does this. My knowledge on AS15169 is limited. > But I recall them having a very strict peering policy. Strict? Really? https://peering.google.com/about/peering_policy.html > > Nick Olsen > Network Operations (855) FLSPEED x106 > > ---------------------------------------- > From: "Joly MacFie" <joly(a)punkcast.com> > Sent: Monday, November 19, 2012 1:21 PM > To: "joel jaeggli" <joelja(a)bogus.com> > Subject: Re: Google/Youtube problems > > WIth my limited understanding of such topics I've long been confused by > something I read a couple of years back - in an Arbor report perhaps - to > the effect that by being the originator of so much traffic, and as they > built out their own network, Google were making money on transit. > > Can anyone elaborate or refute? > > On Mon, Nov 19, 2012 at 11:55 AM, joel jaeggli <joelja(a)bogus.com> wrote: > >> On 11/19/12 5:59 AM, Saku Ytti wrote: >> >>> What I'm trying to say, I can't see youtube generating anywhere nearly >>> enough revenue who shift 10% (or more) of Internet. And to explain this >>> conundrum to myself, I've speculated accounting magic (which I'd frown >>> upon) and leveraging market position to get free capacity (which is ok, > I'd >>> do the same, had I the leverage) >>> >> Or there's a simpler explanation. Which is that it makes money either >> directly or as part of a salubrious interaction with other google >> properties. >> >> They had about 2.5Billion left over for their trouble in the quarter >> ending 9/30 which isn't too shabby on a gross of 14 billion. >> >> > > -- > --------------------------------------------------------------- > Joly MacFie 218 565 9365 Skype:punkcast > WWWhatsup NYC - http://wwwhatsup.com > http://pinstand.com - http://punkcast.com > VP (Admin) - ISOC-NY - http://isoc-ny.org > -------------------------------------------------------------- > - >

1 0

Re: Google/Youtube problems
by Nick Olsen 19 Nov '12

19 Nov '12

I think this would be true if they offered some form of paid peering. Google want's a good fast route to your customers, And your customers want a good fast route to Google. IF Google ran its transit at or near congestion. This could degrade your customers performance. After so long, You'd contact Google and attempt to troubleshoot. And they would say if you want good peering with them, You should pay them to peer. Where you could control just how much traffic was on your port and expand it if needed. Pretty sure this was Comcast and level3/Netflix did. But Comcast had the winning leverage (more eyeballs) in the discussion. But, I don't think Google does this. My knowledge on AS15169 is limited. But I recall them having a very strict peering policy. Nick Olsen Network Operations (855) FLSPEED x106 ---------------------------------------- From: "Joly MacFie" <joly(a)punkcast.com> Sent: Monday, November 19, 2012 1:21 PM To: "joel jaeggli" <joelja(a)bogus.com> Subject: Re: Google/Youtube problems WIth my limited understanding of such topics I've long been confused by something I read a couple of years back - in an Arbor report perhaps - to the effect that by being the originator of so much traffic, and as they built out their own network, Google were making money on transit. Can anyone elaborate or refute? On Mon, Nov 19, 2012 at 11:55 AM, joel jaeggli <joelja(a)bogus.com> wrote: > On 11/19/12 5:59 AM, Saku Ytti wrote: > >> What I'm trying to say, I can't see youtube generating anywhere nearly >> enough revenue who shift 10% (or more) of Internet. And to explain this >> conundrum to myself, I've speculated accounting magic (which I'd frown >> upon) and leveraging market position to get free capacity (which is ok, I'd >> do the same, had I the leverage) >> > Or there's a simpler explanation. Which is that it makes money either > directly or as part of a salubrious interaction with other google > properties. > > They had about 2.5Billion left over for their trouble in the quarter > ending 9/30 which isn't too shabby on a gross of 14 billion. > > -- --------------------------------------------------------------- Joly MacFie 218 565 9365 Skype:punkcast WWWhatsup NYC - http://wwwhatsup.com http://pinstand.com - http://punkcast.com VP (Admin) - ISOC-NY - http://isoc-ny.org -------------------------------------------------------------- -

2 1

Re: Plages d'adresses IP Orange
by Scott Weeks 19 Nov '12

19 Nov '12

--- jlewis(a)lewis.org wrote: From: Jon Lewis <jlewis(a)lewis.org> Pourquoi demandez-vous des questions NANOG que Wanadoo peut rÃ©pondre? Hopefully google translate hasn't butchered that too badly. ----------------------------------------------------- It said something about eggs? Were you hungry for some RA eggs? ;-) "Why do you ask questions that NANOG Wanadoo © RA can lay eggs?" scott

1 0

Google/Youtube problems
by Daniel Suchy 19 Nov '12

19 Nov '12

Hello, for approx. last 14 days we're seeing problems with video playing from youtube (page loads without problems, but player shows error), and also other applications like maps are having problems. As these problems were only for some of prefixes announced out of AS 8251, we recognised that as problem with Google's CDN and reported it to Google. As workaround, filtering affected prefixes from peering in Prague partially helped. We already tried communicate problem with Google from the beginning (in our network, around 50000 end users are directly afected), and they're claiming, that problem is related only to our network and there's no global issue. But similar issues we're seeing from some other networks, which are peering with Google and have nothing common with our AS8251. We sent emails to Google NOC/NST, also tried to phone them about the problems, but according to end-user claims, problem persist. Problem is isolated only to peak hours, when something seems to be saturated. If I recall past optional IPv6 from Google, they said: "It is very important to us that our users always receive the best possible experience". But majority of end users still uses IPv4 and we're seeing problems here - and response is minimal. At least information about cause of the problem and expected time for problem resolution. Is anyone else seeing similar problems with Google/Youtube? With regards, Daniel

8 10

Youtube
by Glauber Derlland 19 Nov '12

19 Nov '12

Hi, Anyone know what's going to inform with Youtube here in Brazil this horrible, super slow it is already 14 days. -- Glauber Derlland 81-3497-7250 / 81-8859-3306 / 81-4062-9207 / 11-4063-0189 INOC-DBA.br: 262792*100 www.vescnet.com.br msn: vesc.net@hotmail. <vesc.net(a)hotmail.com> http://as262792.peeringdb.com/

2 1

25Mbps vs 4 Mbps
by Glen Kent 19 Nov '12

19 Nov '12

Hi, The service provider(s) pipe that takes all web traffic from my laptop to the central servers (assume youtube) remain same whether i take a 4Mbps or a 25Mbps connection from my service provider. This means that the internet connection that i take from my service provider only affects the last mile -- from my home network to my service providers first access router. Given this, would one really see a 6 times improvement in a 25Mbps connection over a 4Mbps connection? I assume that the service providers rate limit the traffic much more aggressively in a 4Mbps connection. But this would only matter if the traffic from my youtube server is greater than 4Mbps, which i suspect would be the case. The question then is that how does going for a higher BW connection from the service provider help? Glen

2 1

Generic IP Connectivity Provisioning Profile
by mohamed.boucadair＠orange.com 19 Nov '12

19 Nov '12

Dear all, Ron suggested I share this document in this mailing list in order to assess whether there is interest from the nanog community to carry out this effort within IETF. If you are supporting this effort, please voice it. Your feedback will help Ron in making a decision to AD-sponsor this document in the IETF. This is an effort trying to capture and expose the IP connectivity service provided by the network layer to upper services, to adjacent networks, offered/delivered to customers, etc. The document focuses on the definition of the CPP itself and does not discuss for instance how this is translated into configuration data/policies to be enforced in underlying nodes. Abstract: This document describes the Connectivity Provisioning Profile (CPP) and proposes a CPP Template to capture IP connectivity requirements to be met in the context of delivery of services (e.g. Voice over IP or IP TV) which are to be plugged upon an IP/MPLS infrastructure. The CPP defines the set of IP transfer parameters to be supported by the underlying IP/MPLS transport network together with a reachability scope and bandwidth/capacity needs. Appropriate performance metrics such as one-way delay, one-way delay variation are used to characterize an IP transfer service. Both global and restricted reachability scopes can be captured in the CPP. Having the generic CPP template allows for (1) automating the process of service negotiation and activation, thus accelerating service provisioning, (2) setting the (traffic) objectives of Traffic Engineering functions and service management functions and (3) enriching service and network management systems with 'decision- making' capabilities based on negotiated/offered CPPs. https://datatracker.ietf.org/doc/draft-boucadair-connectivity-provisioning-… http://tools.ietf.org/html/draft-boucadair-connectivity-provisioning-profil… Comments are more than welcome. Cheers, Med

1 0

Dns sometimes fails using Google DNS / automatic dnssec
by MailPlus| David Hofstee 19 Nov '12

19 Nov '12

Hi, We've been seeing automatic RRSIG records on Google DNS lately, the 8.8.8.8 en 8.8.4.4. They are not always provided. They cause problems for some of our customers in a weird way I cannot explain. For them these records do not resolve but I cannot reproduce it. So when I run dig command dig @8.8.8.8 m1.mailplus.nl it often provides the RRSIG record (but e.g. the TXT record will not be signed). I've heard that DNS may fall back to TCP and/or may be filtered by firewalls if UDP is over 512 bytes. However, the request is not that long, about 200 bytes if I interpret the answer correctly. Can someone come up with a good explanation why a tiny percentage of our customers cannot resolve (some of) our domains? Btw, our nameservers (transip.nl) only provide DNSSEC records if explicitly asked. What is standard here? Thanks, David Hofstee

4 6