Re: Standard terminology for a dark fiber path?

2016-02-24 Thread Larry Sheldon
On 2/24/2016 14:55, Fletcher Kittredge wrote: What is the standard terminology for strands of dark fiber spliced together to form a continuous path between points A and Z? I have seen: - *fiber circuit* [but also seen used to denote a connection at the network layer over a physical

Standard terminology for a dark fiber path?

2016-02-24 Thread Fletcher Kittredge
What is the standard terminology for strands of dark fiber spliced together to form a continuous path between points A and Z? I have seen: - *fiber circuit* [but also seen used to denote a connection at the network layer over a physical fiber connection. This definition of circuit would

Re: Cogent & Google IPv6

2016-02-24 Thread Valdis . Kletnieks
On Wed, 24 Feb 2016 16:51:55 -0500, "Patrick W. Gilmore" said: > Or do you think Cogent is paying all of them? That is a possibility, but it > means that Cogent is not getting paid - by definition. All depends how creative their accountants are... :) pgpW8dCKWjsxu.pgp Description: PGP

Re: Cogent & Google IPv6

2016-02-24 Thread Patrick W. Gilmore
On Feb 24, 2016, at 4:48 PM, Ricky Beam wrote: > On Wed, 24 Feb 2016 15:48:22 -0500, Patrick W. Gilmore > wrote: >> And Ricky is wrong, the vast majority of prefixes Cogent routes have zero >> dollars behind them. Cogent gets paid by customers, not peers.

Re: Cogent & Google IPv6

2016-02-24 Thread Ricky Beam
On Wed, 24 Feb 2016 15:48:22 -0500, Patrick W. Gilmore wrote: And Ricky is wrong, the vast majority of prefixes Cogent routes have zero dollars behind them. Cogent gets paid by customers, not peers. (At least not the big ones.) Show me a single connection to Cogent for

Re: Cogent & Google IPv6

2016-02-24 Thread Brandon Butterworth
> From nanog-boun...@nanog.org Wed Feb 24 21:03:17 2016 > In one's situation, does Cogent have enough pros to overcome the > cons? Same for HE or any other carrier. Who cares, with everyone trying to be IPv6 transit free and covering it with a settlement free peering policy it may accidentally

Re: Cogent & Google IPv6

2016-02-24 Thread Paras Jha
Transit providers are the mdidlemen of the internet, I see no problem with the concept of "double dipping". It's their fiber and infrastructure, if you want access to everything on their network, including other people on their network, pay for it or find a way to get access. On Wed, Feb 24, 2016

Re: Cogent & Google IPv6

2016-02-24 Thread Mike Hammett
*nods* and everything is pros and cons. In one's situation, does Cogent have enough pros to overcome the cons? Same for HE or any other carrier. If I get full tables (v4 and b6) from multiple networks and\or I peer with the networks that are missing from a particular provider's offering, I may

Re: Cogent & Google IPv6

2016-02-24 Thread Patrick W. Gilmore
“Tier One” used to mean SFI or customer downstream to every prefix on the ‘Net. Today it is more like “transit free”, since some “tier one” providers have paid peering. And Ricky is wrong, the vast majority of prefixes Cogent routes have zero dollars behind them. Cogent gets paid by customers,

Re: Cogent & Google IPv6

2016-02-24 Thread Patrick W. Gilmore
Agreed on all points. “Double dipping” is not morally abhorrent, or even slightly slimy. However, Cogent customers paid Cogent to connect to The Internet, not “The other networks that are paying Cogent”. So in this case, if I had to make a choice of which provider to drop, I’d stick with

Re: Cogent & Google IPv6

2016-02-24 Thread Mike Hammett
Isn't that how "Tier 1s" have always operated? Like, always? Customers or peers with peers subject to various requirements. - Mike Hammett Intelligent Computing Solutions http://www.ics-il.com Midwest-IX http://www.midwest-ix.com - Original Message - From: "Ricky Beam"

RE: Cogent & Google IPv6

2016-02-24 Thread Damien Burke
I have already shut down peering with cogent over ipv6 entirely (two weeks ago) over this issue. Cogent needs to get it together and work it out. Google is our overlord - you cannot refuse them. -Original Message- From: NANOG [mailto:nanog-boun...@nanog.org] On Behalf Of Patrick W.

Re: Cogent & Google IPv6

2016-02-24 Thread Ricky Beam
On Wed, 24 Feb 2016 14:46:56 -0500, Matt Hoppes wrote: Isn't that how the Internet is suppose to work? Perhaps. But that's not how *Cogent* works. They have a very idiotic view of "Tier 1". They have no transit connections with anyone; someone is paying them

Re: Cogent & Google IPv6

2016-02-24 Thread Mike Hammett
Whomever hurts the most will blink first. I don't really care who that is. I have no ill will towards "double dipping". Either they do or they don't offer the desired connectivity and I'm moving on. - Mike Hammett Intelligent Computing Solutions http://www.ics-il.com Midwest-IX

Re: Cogent & Google IPv6

2016-02-24 Thread Patrick W. Gilmore
Are HE & Google the new L3 & FT? Nah, L3 would never have baked Cogent a cake. :) Shall we start a pool? Only problem is, should the pool be “who will disconnect from Cogent next?” or “when will Cogent blink?” I’m voting for the former. -- TTFN, patrick > On Feb 24, 2016, at 3:08 PM, Baldur

Re: Cogent & Google IPv6

2016-02-24 Thread Baldur Norddahl
This is Google saying that Google does not want to pay for traffic to Cogent. If Cogent wants to exchange any traffic with Google, Cogent is invited to peer directly with Google. Of course Cogent refuses. And now Cogent is not only missing the part of IPv6 internet that is Hurricane Electric

Re: Cogent & Google IPv6

2016-02-24 Thread Patrick W. Gilmore
To answer Matt’s question, NO. Assume Cogent peers with NTT. Assume Google peers with NTT. NTT has very good v6 connectivity (not an assumption). Cogent cannot send a packet to NTT and say “please hand this to Google”. Nor can Google hand a packet to NTT with a destination of Cogent. Under

Re: Cogent & Google IPv6

2016-02-24 Thread Max Tulyev
If you connected to Internet ONLY through Cogent - there is no other way. If you have another upstreams - Google should be reachable. On 24.02.16 21:46, Matt Hoppes wrote: > Correct me if I'm wrong, but if Cogent isn't peering with Google IPv6, > shouldn't the traffic flow out to one of their

Re: Cogent & Google IPv6

2016-02-24 Thread Matt Hoppes
Correct me if I'm wrong, but if Cogent isn't peering with Google IPv6, shouldn't the traffic flow out to one of their peer points where another peer DOES peer with Google IPv6 and get you in? Isn't that how the Internet is suppose to work? On 2/24/16 2:43 PM, Damien Burke wrote: Not sure. I

RE: Cogent & Google IPv6

2016-02-24 Thread Damien Burke
Not sure. I got the same thing today as well. Is this some kind of ipv6 war? -Original Message- From: NANOG [mailto:nanog-boun...@nanog.org] On Behalf Of Ian Clark Sent: Wednesday, February 24, 2016 10:25 AM To: NANOG Subject: Cogent & Google IPv6 Anyone know what's actually going on

Cogent & Google IPv6

2016-02-24 Thread Ian Clark
Anyone know what's actually going on here? We received the following information from the two of them, and this just started a week or so ago. *From Cogent, the transit provider for a branch office of ours:* Dear Cogent Customer, Thank you for contacting Cogent Customer Support for

SIP strangeness on T-W cable ?

2016-02-24 Thread John Levine
Today I am unable to connect to my usual SIP servers from my T-W cable account. I've tried two Sipura terminal adapters and the softphone in Jitsi, and I can't connect to either callcentric.com or voipdiscount.com, and the network connection is otherwise looking normal. The SIP providers are up,

SunGard On List?

2016-02-24 Thread James Bensley
Hi All, Any SunGard on list? Having a path issue from multiple ISPs in the UK. Cheers, James.

Re: APC vs UPC?

2016-02-24 Thread Niels Bakker
* baldur.nordd...@gmail.com (Baldur Norddahl) [Tue 23 Feb 2016, 14:25 CET]: SFP modules will generally have UPC connectors. You therefore need to use cables with UPC at one end and APC at the other end. If you use a APC-APC cable you will have 3-6 dB of optical loss. If it is a short