Cogent peering issues with Sprint

Cogent is experiencing two problems right now. Their automated message reports that they have a backbone problem causing latency, but they also seem to be experiencing peering problems with Sprint.

Output over Cogent:
sarglund-cogent>traceroute www.duke-energy.com
Translating “www.duke-energy.com”…domain server (66.28.0.45) [OK]

Type escape sequence to abort.
Tracing the route to www.duke-energy.com (192.234.122.137)

1 g0-2.na01.b000629-0.ord01.atlas.cogentco.com (38.112.0.193) [AS 174] 0 msec 0 msec 0 msec
2 g1-0.core01.ord01.atlas.cogentco.com (66.28.6.73) [AS 174] 0 msec 0 msec 0 msec
3 t4-1.mpd01.ord01.atlas.cogentco.com (154.54.1.98) [AS 174] 4 msec 0 msec 0 msec
4 v3488.mpd01.ord03.atlas.cogentco.com (154.54.5.26) [AS 174] 0 msec 0 msec 0msec
5 g6-0-0-3492.core01.ord03.atlas.cogentco.com (154.54.3.241) [AS 174] 0 msec 0 msec 0 msec
6 sprint.ord03.atlas.cogentco.com (154.54.13.46) [AS 174] 0 msec 4 msec 0 msec
7 sl-bb20-chi-4-0.sprintlink.net (144.232.8.218) [AS 174] 0 msec 0 msec 0 msec
8 sl-bb21-chi-15-0.sprintlink.net (144.232.26.2) [AS 174] 4 msec 4 msec 0 msec
9 sl-bb25-chi-13-0.sprintlink.net (144.232.26.90) [AS 174] 0 msec 0 msec 0 msec
10 * * *
11 * * *
12 * * *
13 * * *
14 * *
sarglund-cogent>

vs.

Output over Verizon:
sarglund-gw>traceroute www.duke-energy.com
Translating “www.duke-energy.com”…domain server (198.6.1.2) [OK]

Type escape sequence to abort.
Tracing the route to www.duke-energy.com (192.234.122.137)

1 500.Serial4-10.GW1.CHI2.ALTER.NET (157.130.165.213) [AS 701] 8 msec 0 msec 4msec
2 0.so-0-0-0.XT1.CHI2.ALTER.NET (152.63.64.130) [AS 701] 4 msec 0 msec 0 msec
3 0.so-7-0-0.XL1.CHI13.ALTER.NET (152.63.64.206) [AS 701] 4 msec 16 msec 4 msec
4 0.so-6-0-0.BR2.CHI13.ALTER.NET (152.63.73.25) [AS 701] 4 msec 0 msec 44 msec
5 sl-st21-chi-3-0-0.sprintlink.net (144.232.18.141) [AS 701] 32 msec 12 msec 12 msec
6 sl-crs2-chi-0-1-0-0.sprintlink.net (144.232.18.154) [AS 701] 8 msec 16 msec 20 msec
7 sl-bb25-chi-8-0.sprintlink.net (144.232.26.113) [AS 701] 12 msec 28 msec 12 msec
8 sl-bb26-rly-10-0.sprintlink.net (144.232.20.88) [AS 701] 116 msec 216 msec *
9 sl-bb21-dc-12-0-0.sprintlink.net (144.232.9.212) [AS 701] 24 msec 32 msec 36 msec
10 sl-gw21-dc-15-0.sprintlink.net (144.232.15.9) [AS 701] 32 msec 28 msec 24 msec
11 sl-dukenergy-57200-0.sprintlink.net (144.228.23.170) [AS 701] 40 msec 36 msec 40 msec
sarglund-gw>

enjoy,
-carl

We're actually seeing the same isses @ NTT/AS2914 peering with
Sprint/AS1239:

traceroute to www.sprint.net (199.0.234.48), 64 hops max, 40 byte packets
1 10.30.1.1 (10.30.1.1) 0.504 ms 0.221 ms 0.234 ms
2 ge-2-2-0.r00.chcgil08.us.bb.gin.ntt.net (129.250.208.1) 0.298 ms 0.290
ms 0.284 ms
3 p16-0-2-3.r21.chcgil09.us.bb.gin.ntt.net (129.250.3.17) 0.469 ms 0.474
ms 0.452 ms
4 ae-0.r20.chcgil09.us.bb.gin.ntt.net (129.250.3.97) 0.423 ms 0.421 ms
0.418 ms
5 sl-st21-chi-14-1-1.sprintlink.net (144.232.8.221) 0.408 ms 0.439 ms
0.414 ms
6 sl-crs2-chi-0-0-0-3.sprintlink.net (144.232.26.9) 1.661 ms 1.521 ms
1.407 ms
7 sl-bb25-chi-8-0.sprintlink.net (144.232.26.113) 1.264 ms 1.212 ms
1.075 ms
8 * * *
^C

-Basil @ CIFNet

Cogent is experiencing two problems right now. Their automated message reports that they have a backbone problem causing latency, but they also seem to be experiencing peering problems with Sprint.

Are you sure that this is not in Sprint, or even Duke Energy ?

I can't ping to 192.234.122.137 from either home or work, and I don't see any signs of Cogent problems
from Tyco Road / Tysons Corner.

Regards
Marshall

From Cogent in Tysons Corner, Virginia :

Type escape sequence to abort.
Tracing the route to www.duke-energy.com (192.234.122.137)

   1 g0-7.na21.b002176-1.dca01.atlas.cogentco.com (38.99.206.153) 0 msec 0 msec 0 msec
   2 g2-1-3587.core01.dca01.atlas.cogentco.com (38.20.32.13) [AS 174] 4 msec 0 msec 4 msec
   3 t3-1.ccr02.dca01.atlas.cogentco.com (154.54.3.158) [AS 174] 0 msec 0 msec 0 msec
   4 t4-1.mpd01.dca02.atlas.cogentco.com (154.54.2.182) [AS 174] 4 msec 0 msec 4 msec
   5 t2-2.mpd01.iad01.atlas.cogentco.com (154.54.1.78) [AS 174] 0 msec 4 msec 0 msec
   6 g14-0-0.core01.iad01.atlas.cogentco.com (154.54.5.57) [AS 174] 4 msec 0 msec 4 msec
   7 sprint.iad01.atlas.cogentco.com (154.54.13.62) [AS 174] 0 msec 0 msec 4 msec
   8 sl-bb20-dc-8-0-0.sprintlink.net (144.232.9.120) [AS 1239] 0 msec 4 msec 4 msec
   9 sl-gw21-dc-14-0.sprintlink.net (144.232.15.7) [AS 1239] 0 msec 0 msec 0 msec
10 sl-dukenergy-57200-0.sprintlink.net (144.228.23.170) [AS 1239] 12 msec 8 msec 12 msec
11 * * *
12 * * *
13 * * *
14 * * *
15 * * *
16 * * *
17 * * *
18 * * *
19 * * *
20 * * *
21 * * *
22 * * *
23 * * *
24 * * *
25 * * *
26 * * *
27 * * *
28 * * *
29 * * *
30 * * *

From Cox Cable in Clifton, Virginia

[TME-Laptop-2:~] tme% traceroute www.duke-energy.com
traceroute to www.duke-energy.com (192.234.122.137), 64 hops max, 40 byte packets
1 * * *
2 ip70-179-104-1.dc.dc.cox.net (70.179.104.1) 12.776 ms 11.714 ms 11.778 ms
3 ip72-196-255-137.dc.dc.cox.net (72.196.255.137) 12.472 ms 10.686 ms 11.940 ms
4 mrfddsrj01-ge110.rd.dc.cox.net (68.100.0.161) 13.513 ms 13.235 ms 11.577 ms
5 xe-9-2-0.edge1.washington1.level3.net (4.79.228.61) 12.426 ms 12.197 ms 11.816 ms
6 ae-24-79.car4.washington1.level3.net (4.68.17.70) 13.260 ms ae-34-89.car4.washington1.level3.net (4.68.17.134) 12.958 ms ae-24-79.car4.washington1.level3.net (4.68.17.70) 13.032 ms
7 sprint-level3-te.washington1.level3.net (4.68.111.170) 14.570 ms 12.670 ms 13.926 ms
8 sl-bb20-dc-8-0-0.sprintlink.net (144.232.9.120) 16.061 ms 13.807 ms 13.416 ms
9 sl-gw21-dc-14-0.sprintlink.net (144.232.15.7) 14.193 ms 15.085 ms 13.761 ms
10 sl-dukenergy-57200-0.sprintlink.net (144.228.23.170) 25.361 ms 22.298 ms 22.464 ms
11 * * *
12 * * *
13 * * *
14 * * *
15 * * *
16 * * *
17 * * *
18 * * *
19 * * *

Cogent is experiencing two problems right now. Their automated message
reports that they have a backbone problem causing latency, but they also
seem to be experiencing peering problems with Sprint.

  2 g1-0.core01.ord01.atlas.cogentco.com (66.28.6.73) [AS 174] 0 msec 0
msec 0 msec
  3 t4-1.mpd01.ord01.atlas.cogentco.com (154.54.1.98) [AS 174] 4 msec 0
msec 0 msec
  4 v3488.mpd01.ord03.atlas.cogentco.com (154.54.5.26) [AS 174] 0 msec 0
msec 0msec
  5 g6-0-0-3492.core01.ord03.atlas.cogentco.com (154.54.3.241) [AS 174] 0
msec 0 msec 0 msec
  6 sprint.ord03.atlas.cogentco.com (154.54.13.46) [AS 174] 0 msec 4 msec
0 msec
  7 sl-bb20-chi-4-0.sprintlink.net (144.232.8.218) [AS 174] 0 msec 0 msec
0 msec
  8 sl-bb21-chi-15-0.sprintlink.net (144.232.26.2) [AS 174] 4 msec 4 msec
0 msec
  9 sl-bb25-chi-13-0.sprintlink.net (144.232.26.90) [AS 174] 0 msec 0 msec
0 msec
10 * * *
11 * * *
12 * * *
13 * * *
14 * *

We're actually seeing the same isses @ NTT/AS2914 peering with
Sprint/AS1239:

traceroute to www.sprint.net (199.0.234.48), 64 hops max, 40 byte packets
1 10.30.1.1 (10.30.1.1) 0.504 ms 0.221 ms 0.234 ms
2 ge-2-2-0.r00.chcgil08.us.bb.gin.ntt.net (129.250.208.1) 0.298 ms 0.290
ms 0.284 ms
3 p16-0-2-3.r21.chcgil09.us.bb.gin.ntt.net (129.250.3.17) 0.469 ms 0.474
ms 0.452 ms
4 ae-0.r20.chcgil09.us.bb.gin.ntt.net (129.250.3.97) 0.423 ms 0.421 ms
0.418 ms
5 sl-st21-chi-14-1-1.sprintlink.net (144.232.8.221) 0.408 ms 0.439 ms
0.414 ms
6 sl-crs2-chi-0-0-0-3.sprintlink.net (144.232.26.9) 1.661 ms 1.521 ms
1.407 ms
7 sl-bb25-chi-8-0.sprintlink.net (144.232.26.113) 1.264 ms 1.212 ms
1.075 ms
8 * * *
^C

I can't ping www.sprint.net, but I can reach their name servers just fine through Cogent and Cox, and
I can get to www.sprint.net on the web. I am not really sure that there is a problem as I can
also get to http://www.duke-energy.com/ over the web from both Cogent and Cox.

Regards
Marshall

From Cogent :

multicasttech-1>trace ns1-auth.sprintlink.net
Translating "ns1-auth.sprintlink.net"...domain server (63.105.122.7) [OK]

Type escape sequence to abort.
Tracing the route to ns1-auth.sprintlink.net (206.228.179.10)

   1 g0-7.na21.b002176-1.dca01.atlas.cogentco.com (38.99.206.153) 0 msec 0 msec 0 msec
   2 g2-1-3587.core01.dca01.atlas.cogentco.com (38.20.32.13) [AS 174] 4 msec 0 msec 0 msec
   3 t3-1.ccr02.dca01.atlas.cogentco.com (154.54.3.158) [AS 174] 0 msec 0 msec 0 msec
   4 t4-1.mpd01.dca02.atlas.cogentco.com (154.54.2.182) [AS 174] 4 msec 0 msec 0 msec
   5 t2-2.mpd01.iad01.atlas.cogentco.com (154.54.1.78) [AS 174] 0 msec 0 msec 4 msec
   6 g14-0-0.core01.iad01.atlas.cogentco.com (154.54.5.57) [AS 174] 0 msec 4 msec 0 msec
   7 sprint.iad01.atlas.cogentco.com (154.54.13.62) [AS 174] 4 msec 0 msec 4 msec
   8 sl-bb20-dc-8-0-0.sprintlink.net (144.232.9.120) [AS 1239] 4 msec 0 msec 0 msec
   9 sl-crs2-dc-0-4-0-0.sprintlink.net (144.232.15.19) [AS 1239] 4 msec 4 msec 4 msec
10 sl-bb21-nsh-1-0-0.sprintlink.net (144.232.18.184) [AS 1239] 20 msec 24 msec 20 msec
11 sl-bb27-fw-13-0.sprintlink.net (144.232.8.64) [AS 1239] 44 msec 48 msec 44 msec
12 sl-bb20-fw-11-0.sprintlink.net (144.232.11.13) [AS 1239] 48 msec 44 msec 44 msec
13 sl-s1-fw-0-0.sprintlink.net (144.232.0.136) [AS 1239] 44 msec 48 msec 44 msec
14 ns1-auth.sprintlink.net (206.228.179.10) [AS 1239] 40 msec 40 msec 36 msec

It would seem that this issue with Sprint's been resolved.

-Basil @ CIFNet

We're seeing very similar issues with prefixes that only have transit
via Cogent. 66.28.0.0/16, however, is fine from my location:

traceroute to 66.28.0.45 (66.28.0.45), 30 hops max, 40 byte packets
1 10.30.0.1 (10.30.0.1) 0.363 ms 0.427 ms 0.526 ms
2 excore.dc2.hostmysite.net (65.36.160.1) 2.324 ms 2.326 ms 2.328 ms
3 border1.dc3.hostmysite.net (67.59.145.17) 2.328 ms 2.382 ms 2.518 ms
4 208.24.180.58 (208.24.180.58) 4.010 ms 4.079 ms 4.081 ms
5 sl-bb24-pen-2-0.sprintlink.net (144.232.16.105) 4.491 ms 4.651 ms 4.623 ms
6 sl-bb20-rly-15-0-0.sprintlink.net (144.232.20.145) 7.064 ms 5.450 ms 5.610 ms
7 sl-bb24-rly-9-0.sprintlink.net (144.232.14.122) 5.710 ms 5.709 ms *
8 sl-st22-ash-6-0.sprintlink.net (144.232.20.189) 7.568 ms 7.567 ms 7.899 ms
9 p15-2.core01.iad01.atlas.cogentco.com (154.54.13.61) 6.608 ms 6.695 ms 6.895 ms
10 v3491.mpd01.iad01.atlas.cogentco.com (154.54.5.34) 6.941 ms 7.175 ms 7.215 ms
11 t7-4.mpd01.dca02.atlas.cogentco.com (154.54.7.157) 7.486 ms 7.714 ms 7.800 ms
12 v3490.mpd01.iad03.atlas.cogentco.com (154.54.7.98) 6.273 ms 6.270 ms 6.540 ms
13 res1.dns.cogentco.com (66.28.0.45) 6.737 ms 7.023 ms 7.033 ms

Yes, it looks like this might have been resolved. Duke Energy, Teco Energy, Accenture, and the Washington Post are all now accessible via Cogent.

-carl

Basil Kruglov basil@cifnet.com
Sent by: owner-nanog@merit.edu

10/10/2007 10:17 AM

Please respond to
nanog@merit.edu

To
nanog@merit.edu

cc

Subject
Re: Cogent peering issues with Sprint

On Wed, Oct 10, 2007 at 11:00:13AM -0400, Marshall Eubanks wrote:
> Are you sure that this is not in Sprint, or even Duke Energy ?
>
> I can't ping to 192.234.122.137 from either home or work, and I don't
> see any signs of Cogent problems
> from Tyco Road / Tysons Corner.

It would seem that this issue with Sprint's been resolved.

-Basil @ CIFNet

We shut off Cogent around 9:00 because of routing issues to Telia and other
parts of the Net.

Cogent is experiencing two problems right now. Their automated message reports
that they have a backbone problem causing latency, but they also seem to be
experiencing peering problems with Sprint.

There may be some internal problems on the Sprint network; I'm unable to hit
www.duke-energy.com from either my Covad T1 or my AT&T T1 and my users with
Sprint wireless cards can't reach my VPN concentrator (Covad-connected).

From Covad:

traceroute to 209.163.178.6 (209.163.178.6), 64 hops max, 40 byte packets
1 drawbridge.alfordmedia.com (192.168.67.254) 0.698 ms 0.279 ms 0.313
ms
2 h-74-1-2-209.dllatx37.covad.net (74.1.2.209) 2.219 ms 2.317 ms 2.177
ms
3 172.22.0.1 (172.22.0.1) 27.102 ms 16.690 ms 15.209 ms
4 192.168.38.173 (192.168.38.173) 33.027 ms 15.440 ms 12.087 ms
5 66.236.128.245.ptr.us.xo.net (66.236.128.245) 20.182 ms 16.088 ms
37.398 ms
6 p4-3-0.mar1.dallas-tx.us.xo.net (207.88.82.9) 13.000 ms 13.187 ms
15.036 ms
7 65.106.4.245.ptr.us.xo.net (65.106.4.245) 31.468 ms 13.498 ms 14.839
ms
8 p0-0.ir1.dallas2-tx.us.xo.net (65.106.4.182) 14.992 ms 18.729 ms
15.151 ms
9 peer-01-ge-3-1-0-15.dlfw.twtelecom.net (66.192.252.145) 14.835 ms
14.095 ms 20.054 ms
10 hagg-03-ge-0-0-0-460.dlfw.twtelecom.net (64.129.234.4) 45.161 ms
16.826 ms 28.240 ms
11 64-129-174-30.static.twtelecom.net (64.129.174.30) 24.740 ms 16.621 ms
14.966 ms
12 * * *
(ad infinitum)

Wow, pasted the wrong traceroute AND hit send instead of delete. It's
another great day around here. More caffeine, stat....

Sprint EVDO network still seems flaky from here, but I'm waiting for one of
my more technical users to get online for troubleshooting purposes. The
duke-energy.com IP is reachable now via both providers.