Time / Warner RoadRunner Cable Modem Service | Local Tests

The following informations details tests that have been run from a machine directly connected to the Time / Warner RoadRunner network via a cable modem, as any customer would have a cable modem connected to the network.

Traceroute and Ping: Network Connectivity Tools Reveal Extreme Slackness in Time/Warner RoadRunner Network Backbone

Traceroute of Wednesday, September 13th, 2000, 2300 Hours
traceroute to 63.83.58.1 (63.83.58.1), 30 hops max, 40 byte packets

1  10.162.226.1 (10.162.226.1)  7.476 ms  8.523 ms  7.91 ms
2  24.25.1.129 (24.25.1.129)  8.178 ms  7.786 ms  8.099 ms
3  24.25.1.49 (24.25.1.49)  8.375 ms  8.163 ms  8.044 ms
4  24.93.64.97 (24.93.64.97)  8.099 ms  8.006 ms  8.002 ms
5  sgarden-roc-gsr.carolina.rr.com (24.93.64.18)  10.333 ms  9.363 ms  9.961 ms
6  sa-gsr-sgarden-gsr.carolina.rr.com (24.93.64.29)  12.704 ms  13.007 ms  13.004 ms
7  24.93.64.58 (24.93.64.58)  54.713 ms  56.46 ms  61.016 ms
8  24.88.3.149 (24.88.3.149)  599.545 ms  593.435 ms  612.187 ms
9  pos1-1.atlaga1-brt1.rr.com (24.218.190.122)  590.623 ms  597.773 ms  571.984 ms
10  pos1-1.vinnva1-brt1.rr.com (24.218.188.246)  585.43 ms  580.252 ms  576.442 ms
How embarrasing for RoadRunner

This traceroute shows that it takes almost 600ms for an ICMP ping packet to get to Atlanta and back. That's worse than a 14.4k regular modem. Who is managing this network? They should hire an experienced network troubleshooter to diagnose and solve the problem. This particular latency issue has been going on now for at least two weeks, and is especially evident during the evening hours. To their credit, the network is a bit faster during the day. How backwards is this?

Traceroute of Wednesday, September 14th, 2000, 1800 Hours
traceroute to wcarchive.cdrom.com (209.155.82.18), 30 hops max, 40 byte packets

1  10.162.226.1 (10.162.226.1)  7.141 ms  8.136 ms  7.929 ms
2  24.25.1.129 (24.25.1.129)  8.138 ms  8.015 ms  7.928 ms
3  24.25.1.49 (24.25.1.49)  8.454 ms  11.645 ms  10.761 ms
4  24.93.64.97 (24.93.64.97)  8.321 ms  8.48 ms  8.017 ms
5  sgarden-roc-gsr.carolina.rr.com (24.93.64.18)  10.332 ms  9.966 ms  9.99 ms
6  sa-gsr-sgarden-gsr.carolina.rr.com (24.93.64.29)  13.843 ms  13.591 ms  14.746 ms
7  24.93.64.58 (24.93.64.58)  19.383 ms  18.382 ms  18.19 ms
8  24.88.3.149 (24.88.3.149)  300.015 ms  305.424 ms  303.871 ms
9  pos1-1.atlaga1-brt1.rr.com (24.218.190.122)  308.537 ms  318.489 ms  319.182 ms
Simple incompetence or illegal Denial of Service?

All I know is, I'm not getting the throughput I'm paying for. Given Time/Warner's reputation in this area, I'd go with incompetence. Still 300ms to routers on their own backbone. This is closer to the speed of a 33.6k modem; they're getting better.

Traceroute of Sunday, September 17th, 2000, 2330 Hours
traceroute to 24.218.188.246 (pos1-1.vinnva1-brt1.rr.com), 30 hops max, 40 byte packets

1  10.162.226.1 (10.162.226.1)  8.953 ms  7.347 ms  7.807 ms
2  24.25.1.129 (24.25.1.129)  8.753 ms  7.746 ms  8.03 ms
3  24.25.1.49 (24.25.1.49)  9.443 ms  8.961 ms  7.998 ms
4  24.93.64.97 (24.93.64.97)  8.141 ms  8.898 ms  7.988 ms
5  sgarden-roc-gsr.carolina.rr.com (24.93.64.18)  10.332 ms  10.037 ms  9.94 ms
6  sa-gsr-sgarden-gsr.carolina.rr.com (24.93.64.29)  12.724 ms  13.315 ms  14.006 ms
7  24.93.64.58 (24.93.64.58)  21.31 ms  20.334 ms  19.747 ms
8  24.88.3.149 (24.88.3.149)  499.046 ms  509.395 ms  502.497 ms
9  pos1-1.atlaga1-brt1.rr.com (24.218.190.122)  499.875 ms  507.411 ms  514.041 ms
10  pos1-1.vinnva1-brt1.rr.com (24.218.188.246)  524.62 ms  526.184 ms  538.115 ms
Time/Warner RoadRunner is back up to 500ms again.

ANY network that pings over 500ms across its own backbone needs to refund it's customers money. I'm starting to think that Time/Warner purposely restricts bandwidth during non-business hours. I can't imagine why. Whatever the reason, it's really inept management.

Traceroute of Tuesday, September 19th, 2000, 2000 Hours
traceroute to 24.128.191.13 (pos11-0-0.chcgil1-brt1.rr.com), 30 hops max, 40 byte packets

 1  10.162.226.1 (10.162.226.1)  7.958 ms  8.192 ms  7.858 ms
 2  24.25.1.129 (24.25.1.129)  8.983 ms  8.009 ms  7.985 ms
 3  24.25.1.49 (24.25.1.49)  10.669 ms  8.181 ms  7.987 ms
 4  24.93.64.97 (24.93.64.97)  8.105 ms  8.46 ms  7.989 ms
 5  sgarden-roc-gsr.carolina.rr.com (24.93.64.18)  10.324 ms  10.948 ms  10.054 ms
 6  sa-gsr-sgarden-gsr.carolina.rr.com (24.93.64.29)  12.67 ms  13.597 ms  15.18 ms
 7  24.93.64.58 (24.93.64.58)  24.432 ms  21.627 ms  20.127 ms
 8  24.88.3.149 (24.88.3.149)  642.775 ms  662.47 ms  655.869 ms
 9  pos1-1.atlaga1-brt1.rr.com (24.218.190.122)  669.902 ms  674.088 ms  662.863 ms
10  srp3-0.atlaga1-brt2.rr.com (24.218.189.149)  675.281 ms  658.143 ms  646.598 ms
11  pos0-1.chcgil1-brt2.rr.com (24.218.188.149)  675.322 ms  675.764 ms  661.49 ms
12  srp5-0.chcgil1-brt3.rr.com (24.218.189.101)  660.094 ms  658.805 ms  669.19 ms
13  pos11-0-0.chcgil1-brt1.rr.com (24.128.191.13)  666.059 ms  676.437 ms  680.379 ms 
680ms at 30% packet loss seems to be the norm at Time/Warner RoadRunner.

In kindergarten, we once made a communications network of string and aluminum cans. This proven technology seems to have been adopted for use in the Time/Warner RoadRunner network. I'm going to E-Mail the "help" folx at RoadRunner and advise them of the fact this web page exists. Will it help? Doubtful, but my next step is to start naming the names of everyone at Time/Warner with whom I speak regarding this issue.

Traceroute of Tuesday, September 30th, 1540 Hours
traceroute to pos1-1.vinnva1-brt1.rr.com (24.218.188.246), 30 hops max, 40 byte packets

 1  10.162.226.1 (10.162.226.1)  8.42 ms  7.88 ms  7.996 ms
 2  24.25.1.129 (24.25.1.129)  9.056 ms  7.842 ms  7.994 ms
 3  24.25.1.49 (24.25.1.49)  8.304 ms  7.975 ms  8.043 ms
 4  24.93.64.97 (24.93.64.97)  8.036 ms  8.58 ms  8.007 ms
 5  sgarden-roc-gsr.carolina.rr.com (24.93.64.18)  10.301 ms  10.443 ms  10.217 ms
 6  sa-gsr-sgarden-gsr.carolina.rr.com (24.93.64.29)  12.712 ms  13.244 ms  12.176 ms
 7  24.93.64.58 (24.93.64.58)  22.911 ms  21.781 ms  23.099 ms
 8  24.88.3.149 (24.88.3.149)  232.598 ms  234.247 ms  231.721 ms
 9  pos1-1.atlaga1-brt1.rr.com (24.218.190.122)  240.126 ms  244.447 ms  242.262 ms
10  pos1-1.vinnva1-brt1.rr.com (24.218.188.246)  285.261 ms  269.898 ms  276.829 ms
It's the 30th. Still laggy. RR's tech staff will lie to clients. Are we surprized?

I suppose I shouldn't be too shocked to find that the backbone is still laggy. I have a very important reality check for Time/Warner RoadRunner: NO ONE WILL PAY TO PEER WITH YOU IF YOU CAN'T MAINTAIN YOUR BACKBONE. So the tech guys James M. and Jon S, as well as the local Technical Manager, lied to me about the situation and when it will be repaired. So, if you're a Network Design engineer for an Internet carrier and want to peer with Time/Warner, don't, or you have no reason to expect different treatment.