January 2012 - Test - lists.nanog.org

10GE TOR port buffers (was Re: 10G switch recommendaton)
by bas 30 Jan '12

30 Jan '12

Hi, Is there a reason switch vendors 1U TOR 10GE aggregation switches are all cut-through and there are no models with deep buffers? I've ben looking at all vendors I can think of and all have the same models. TOR switches as cut-through with little buffers, and chassis based boxes with deep buffers. TOR: Juniper EX4500 208KB/10GE (4MB shared per PFE) Cisco 4900M 728KB/10GE (17.5MB shared) Cisco Nexus 3064 140KB/10GE (9MB shared) Cisco Nexus 5000 680KB/10GE Force10 S2410 I can't find it anymore, but it wasn't much Arista 7148SX 123KB/10GE (80KB per port plus 5MB dynamic) Arista 7050S 173KB/10GE (9MB shared) Brocade VDX 6730-32 170KB/10GE Brocade TurboIron 24X 85KB/10GE HP 6600-24XG 4500KB/10GE HP 5820-24XG-SFP+ 87KB/10GE Extreme Summit X650 375KB/10GE Chassis: Juniper EX8200-8XS 512MB/10GE Cisco WS-X6708-10GE 32MB/10GE (or 24MB) Cisco N7K-M132XP-12 36MB/10GE Arista DCS-7548S-LC 48MB/10GE Brocade BR-MLX-10Gx8-X 128MB/10GE (not sure) 1GE aggregation. Force10 S60 1250MB shared HP 5830 3000MB shared I am at a loss why there are no 10GE TOR switches with deep buffers. Apparently there is a need for deep buffers as the vendors make them available in the chassis linecards. There also are deep buffer 1GE aggregation switches. Is there some (technical) reason for this? I can imagine some vendors would say that you need to scale up to a chassis if you need deep buffers, but at least one vendor should be able to get quite some customers with a 10G deep buffer TOR switch. I understand that flow-control should prevent loss with microbursts, but in my customers get adverse effects, with strong negative performance if they let flow-control do its thing. Any pointers why this is, or if there is a solution for microburst loss would be greatly appreciated. Thanks, Bas

11 36

Re: Paypal outage?
by Chris 29 Jan '12

29 Jan '12

I switched browsers and it seemed to clear it up. Just never seen an odd error like that before..

3 2

LX sfp minimum range
by jon Heise 29 Jan '12

29 Jan '12

we are moving a router between 2 data centers and we only have LX sfp's for connection, is there any issue using LX sfp's in a short range deployment ?

15 16

MPLS Traffic Engineering Help
by Jeremy 29 Jan '12

29 Jan '12

Hi Everyone, I could use a little help on MPLS and Traffic Engineering. Right now I'm just trying to wrap my head around it. I currently have a couple tunnels going in either direction, those are working fine (but certainly took me long enough to get them working!) and I can direct traffic over them easy enough. Now I'm looking into allocating/reserving bandwidth for a given tunnel and if possible have it react to increased network loads and recalculate its path if need be. (Poor) Example: I have two paths that two different tunnels (A and B) that will go over a T-1 and 100mbps ethernet. A is more important than B. When traffic is low, I'd like them both to go over the 100mbps link so either tunnel can fill the pipe, but if Tunnel A requires more bandwidth, Tunnel B should react and move to the T-1. Is this possible? or am I horribly confused? I'm not really looking for the exact commands or the 'answer' to this problem, but some guidance would be greatly appreciated. I'm working with Cisco gear, 2800's and such. This is purely an academic exercise. Thanks! Jeremy

1 0

Re: US DOJ victim letter
by John Peach 29 Jan '12

29 Jan '12

On Sat, 28 Jan 2012 16:30:47 +0000 bmanning(a)vacation.karoshi.com wrote: > On Fri, Jan 27, 2012 at 10:20:08PM -0500, Martin Hannigan wrote: > > On Fri, Jan 27, 2012 at 1:32 PM, Randy Epstein <nanog(a)hostleasing.net> wrote: > > > [snip] > I missed the part where ARIN turned over its address database w/ associatedd > registration information to the Fed ... I mean I've always advocated for > LEO access, but ther has been significant pushback fromm the community on > unfettered access to that data. As I recall, there are even policies and > processes to limit/restrict external queries to prevent a DDos of the whois > servers. And some fairly strict policies on who gets dumps of the address > space. As far as I know (not very far) bundling the address database > -and- the registration data are not available to mere mortals. > > So - just how DID the Fed get the data w/o violating ARIN policy? > > /bill > > Ours came from our whois information. -- John

2 1

Re: XBOX 720: possible digital download mass service.
by J. M. D. Patterson, CEO 28 Jan '12

28 Jan '12

Ahem: "Never underestimate the bandwidth of a carrier pidgeon with a 4GB USB stick" http://news.bbc.co.uk/2/hi/8248056.stm John Patterson a.k.a "InetDaemon" -----Original Message----- Date: Sat, 28 Jan 2012 11:35:59 -0500 From: Lamar Owen <lowen(a)pari.edu> To: nanog(a)nanog.org Subject: Re: XBOX 720: possible digital download mass service. Message-ID: <201201281136.00093.lowen(a)pari.edu> Content-Type: Text/Plain; charset="US-ASCII" On Friday, January 27, 2012 05:56:19 AM Randy Bush wrote: > > Can internet in USA support that? Call of Duty 15 releases may 2014 > > and 30 million gamers start downloading a 20 GB files. Would the > > internet collapse like a house of cards?. > not a problem. the vast majority of the states is like a developing > country [0], the last mile is pretty much a tin can and a string. so > this will effectively throttle the load. Being in 'the middle of nowhere' as I write, even we are a few notches above RFC1149 capabilities. As one visitor to our site (who had been recently to NRAO Greenbank) said 'if this isn't the middle of nowhere, you can probably see it from here.' Our base DSL is 7Mb/s down, 0.5Mb/s up, with 11Mb/s down and 1Mb/s up available to over 99% of our very rural county. We (work) have 1Gb/s on the local loop fiber pair, throttled to the amount of IP we actually pay for at the ISP's PoP. Now if RFC1149 supported jumbo frames, it might give tin-cans-and-string a run for its money....

1 0

XBOX 720: possible digital download mass service.
by Tei 28 Jan '12

28 Jan '12

This seems the right mail list to ask this. Consoles have a lifespan of 5 years. The current generation is 6 years old. There are rumours that production of the GPU's for the new generation has started in asia. So the new consoles can come in 2013. Theres also a rumour that these new consoles will require internet to download games. These games can weigth 9 to 20 GB. That may be 30 million users in USA, maybe 50 worldwide. The question is: Can internet in USA support that? Call of Duty 15 releases may 2014 and 30 million gamers start downloading a 20 GB files. Would the internet collapse like a house of cards?. If not, will be internet USA ready for the next next generation? ( 2018 ). -- -- ℱin del ℳensaje.

21 31

Craigslist outage
by Matthew Black 28 Jan '12

28 Jan '12

Accessing from Southern California. Cannot get any pages to view, except a few "about" pages. http://www.craigslist.org/about/help/system-status.html Status says runs good, but cannot pull up any city sites or the basic home page http://www.craigslist.org. Anyone else having trouble or are you able to get in? matthew black information technology services california state university, long beach

4 10

Paypal outage?
by Chris 28 Jan '12

28 Jan '12

Anyone getting a 400 Bad Request from Paypal when you try to login to your account or make a transaction? -- --C "The dumber people think you are, the more surprised they're going to be when you kill them." - Sir William Clayton

2 1

photonic buffer bloat
by Eugen Leitl 28 Jan '12

28 Jan '12

In future photonic networks (which will do relativistic cut-through directly in a photonic crossbar without converting photons to electrons and back) the fiber is not just a transport channel but also a photonic buffer (e.g. at 10 GBit/s Ethernet a short reach fiber already buffers a standard 1500 MTU). Of course photonic gates are expensive, individual delays do add up so even with slow light buffers or optical delay loops taken into consideration current TCP/IP header layout has not been optimized for leading edge containing most significant switching/routing information, or even local-knowledge routing (with no global routes). It's too bad IPv6 was not radical enough, so today's legacy protocols have to be tunneled through the networks of the future. I presume this future is some 20-30 years away still.

2 1