RE: Cogent now peering with Sprint?

That looks like a transit connection that Cogent bought at Ashburn, VA,
not SFI peering connection.

I have seen it at other places as well:

Tracing the route to portus.netsecdesign.com (66.6.208.6)

   1 sl-bb24-ana-13-0.sprintlink.net (144.232.1.138) 4 msec 4 msec 0
msec
   2 sl-st21-la-13-0.sprintlink.net (144.232.20.69) 4 msec 4 msec 4 msec
   3 p12-3.core01.lax05.atlas.cogentco.com (154.54.13.41) [AS 174] 48
msec 8 msec 40 msec
   4 t9-4.mpd01.lax01.atlas.cogentco.com (154.54.3.141) [AS 174] 4 msec
0 msec 0 msec
   5 g2-0-0.core01.lax01.atlas.cogentco.com (154.54.2.101) [AS 174] 4
msec 4 msec 4 msec
   6 g49.ba01.b002698-1.lax01.atlas.cogentco.com (66.250.12.130) [AS
174] 0 msec 4 msec 4 msec
   7 PAJO-Networks.demarc.cogentco.com (38.112.9.190) [AS 174] 4 msec 4
msec 4 msec
   8 dcap04.pcap.lax01.tierzero.net (216.31.128.14) [AS 11509] 4 msec 8
msec 4 msec
   9 mmic-gw.dcap6.lax.us.tierzero.net (216.31.188.94) [AS 11509] 12
msec 8 msec 8 msec
  10 dazedandconfused.netsecdesign.com (66.6.208.4) [AS 11509] 16 msec
16 msec 16 msec
  11 portus.netsecdesign.com (66.6.208.6) [AS 11509] 16 msec 16 msec 16
msec

I used to be getting 4 ms latency when my provider used Level3.

Hrmm, I can't tell by looking at a traceroute who paid whom, if anyone. Care to explain your magic? Is there a code in the in-addrs? Perhaps "sl-$FOO" means something in Sprint-speak?

Secondly, does anyone really give a rat's ass who is "SFI" any longer? There are at least 2 fully "SFI" networks who can't route half as well as a whole slew of non-SFI networks these days.

If [Cogent|Sprint] [buying|peering|whatever] [from|with] [Cogent|Sprint] makes their network better (i.e. lower latency, lower packet loss, higher throughput, and, if you care, lower jitter), I applaud them.

Anyone who thinks "X pays Y" is more important than any of the metrics above needs to reevaluate their priorities. (At least from a customer / engineering PoV. I wouldn't suggest $NETWORK's bean counters have the same priorities as their network engineers & customers. :slight_smile:

IMHO, of course.

> That looks like a transit connection that Cogent bought at Ashburn,
> VA,
> not SFI peering connection.

Hrmm, I can't tell by looking at a traceroute who paid whom, if
anyone. Care to explain your magic? Is there a code in the in-
addrs? Perhaps "sl-$FOO" means something in Sprint-speak?

Yea, but most of xfer-nets in this case appear to be assigned by Cogent
(then again, how does that make any difference for special buy|peer cases):

2 sl-st21-la-13-0.sprintlink.net (144.232.20.69) 4 msec 4 msec 4 msec
3 p12-3.core01.lax05.atlas.cogentco.com (154.54.13.41) [AS 174] 4 msec 4
msec 4 msec

11 sprint.iad01.atlas.cogentco.com (154.54.13.62) 12.032 ms 12.061 ms
12.103 ms

9 sprint.dfw03.atlas.cogentco.com (154.54.10.18) 21.964 ms 21.897 ms
21.756 ms

Etc..

And comm tags from Sprint side look similar to other SFI's:

  1239 174
    144.228.241.81 from 144.228.241.81 (144.228.241.81)
      Origin IGP, metric 4294967294, localpref 100, valid, external
      Community: 1239:321 1239:1000 1239:1011

  1239 209
    144.228.241.81 from 144.228.241.81 (144.228.241.81)
      Origin IGP, metric 4294967294, localpref 100, valid, external
      Community: 1239:321 1239:1000 1239:1011

Anyone who thinks "X pays Y" is more important than any of the
metrics above needs to reevaluate their priorities. (At least from a
customer / engineering PoV. I wouldn't suggest $NETWORK's bean
counters have the same priorities as their network engineers &
customers. :slight_smile:

IMHO, of course.

Indeed, at the end of the day, it really doesn't matter these days :slight_smile:

james

Patrick W. Gilmore schrieb:

Hrmm, I can't tell by looking at a traceroute who paid whom, if
anyone. Care to explain your magic? Is there a code in the in-addrs?
Perhaps "sl-$FOO" means something in Sprint-speak?

Secondly, does anyone really give a rat's ass who is "SFI" any
longer? There are at least 2 fully "SFI" networks who can't route
half as well as a whole slew of non-SFI networks these days.

If [Cogent|Sprint] [buying|peering|whatever] [from|with]
[Cogent|Sprint] makes their network better (i.e. lower latency, lower
packet loss, higher throughput, and, if you care, lower jitter), I
applaud them.

I heard some rumours at the Euro-IX forum last week.

Fact: Deutsche Telekom 3320 is playing the power play currently not
upgrading their peering interconnections in Europe. 174 is experiencing
up to 30% packet loss versus 3320 in Europe, making customers suffer.
This is first hand info (we are involved with some popular content site
sending out up to 800MBps peak to German-speaking users, and appx. 1/3
is 3320).

Think of 174 started to peer with 1239 and redirecting some outbound
traffic to 3320 over this new peer. Since 3320 is buying from 1239, they
will pay more to 1239, and 1239 accepts 174 as a new peer because they
get more money from 3320 ... as mentioned, this is just a rumour I
heard, but reading William B. Norton's theory (tactic #9), this would
make sense.

Fredy

Hi Fredy,

kuenzler@init7.net (Fredy Kuenzler) wrote:

Think of 174 started to peer with 1239 and redirecting some outbound
traffic to 3320 over this new peer. Since 3320 is buying from 1239, they
will pay more to 1239, and 1239 accepts 174 as a new peer because they
get more money from 3320 ... as mentioned, this is just a rumour I
heard, but reading William B. Norton's theory (tactic #9), this would
make sense.

In the very least it's eventually something to use to put pressure
on those AS3320 guys.

The idea is pretty smart :wink:

Elmar.

customers have some say over whats what.. at least one of my transits is using a /30 we assign

to be honest i'd be suspicious if 'most ' nets are from cogent as thats not indicative of an equal peering relationship

they still buy from verio and they have /30s with them also:
58.13.54.154.in-addr.arpa domain name pointer verio.sjc03.atlas.cogentco.com.

(of course, that may be a peer port too.. this is all conjecture! as patrick says the adjacencys are the interesting thing, not the $$)

Steve