US Cable Support → Charter Internet/TV > [HSI] Blizzard Games (WoW/D3/SC) routing lag SoCal

[HSI] Blizzard Games (WoW/D3/SC) routing lag SoCal

Just curious, does anyone using Charter HSI in the SoCal market experiencing lags when playing Blizzard games such as World of Warcraft, Diablo 3 & Star Craft?

Generally speaking, in my case with Diablo 3, it USED to be less than 40ms, now it's above 100ms. At Blizzard tech support forum they asked people to do a trace route to the server and post the results.

So here's my trace route:


1 GATEWAY IP 0.503 ms 0.450 ms 0.365 ms
2 10.87.192.1 (10.87.192.1) 7.036 ms 9.357 ms 10.926 ms
3 dtr03mtpkca-tge-0-1-0-6.mtpk.ca.charter.com (96.34.96.188) 11.715 ms 8.037 ms 10.149 ms
4 crr01mtpkca-tge-0-3-0-4.mtpk.ca.charter.com (96.34.97.230) 10.594 ms 9.767 ms 8.945 ms
5 crr02mtpkca-bue-100.mtpk.ca.charter.com (96.34.99.235) 10.503 ms 9.225 ms 11.064 ms
6 bbr01mtpkca-bue-3.mtpk.ca.charter.com (96.34.2.26) 10.113 ms 16.016 ms 18.981 ms
7 be4016.ccr21.lax05.atlas.cogentco.com (38.104.84.101) 11.595 ms 8.222 ms 8.460 ms
8 be2023.ccr21.lax01.atlas.cogentco.com (154.54.30.185) 12.096 ms 9.810 ms 11.035 ms
9 be2065.ccr21.iah01.atlas.cogentco.com (154.54.5.65) 53.794 ms 46.360 ms
be2067.mpd21.iah01.atlas.cogentco.com (154.54.7.161) 49.357 ms
10 be2144.ccr21.dfw01.atlas.cogentco.com (154.54.25.106) 52.576 ms 55.518 ms
be2145.ccr21.dfw01.atlas.cogentco.com (154.54.25.218) 52.891 ms
11 be2031.ccr21.dfw03.atlas.cogentco.com (154.54.7.46) 49.925 ms 53.776 ms 52.776 ms
12 192.205.36.221 (192.205.36.221) 88.577 ms 95.288 ms 97.597 ms
13 cr2.dlstx.ip.att.net (12.122.138.18) 93.103 ms 87.836 ms 91.186 ms
14 cr2.la2ca.ip.att.net (12.122.28.178) 90.549 ms 86.930 ms 92.003 ms
15 gar29.la2ca.ip.att.net (12.122.129.241) 91.747 ms 88.096 ms 90.056 ms
16 12.122.251.190 (12.122.251.190) 92.926 ms
12-122-254-234.attens.net (12.122.254.234) 85.994 ms
12-122-254-238.attens.net (12.122.254.238) 92.428 ms
17 mdf001c7613r0002.lax1.attens.net (206.16.68.54) 95.044 ms 96.512 ms 91.511 ms


hops 18 and the rest are the Blizzard servers.

If I am correct, the packet took a trip from LA to Dallas, then back to south of LA (Orange County) where the servers sit.

Just for giggle I did a trace route from work in downtown LA, with WAN-link provided by AT&T and here's the result:


1 1 ms 1 ms 1 ms GATEWAY IP
2 * * * Request timed out.
3 41 ms 21 ms 22 ms 76-218-72-2.lightspeed.irvnca.sbcglobal.net [76.218.72.2]
4 * * * Request timed out.
5 22 ms 23 ms 23 ms 12.83.38.153
6 23 ms 22 ms 23 ms gar20.la2ca.ip.att.net [12.122.128.181]
7 24 ms 24 ms 25 ms 12.122.251.190
8 23 ms 23 ms 31 ms mdf001c7613r0002.lax1.attens.net [206.16.68.54]


It amazes me the huge difference in routing with two ISPs.

I have submitted a ticket with Blizzard, their response was Charter's routing problem. I haven't called Charter bx I'm pretty sure they'll tell me it's Blizzard's problem.

So if you are using Charter, do a trace route to 12.129.209.68 and report back with your geological location.

And if any Charter tech support is aware of the issue, could you please look into this?

How long has this been occurring? Could be a temporary routing change that revert back to the optimal route soon.

I'd say about half year now. I didn't notice the latency issue until recently. Then as Blizzard suggested to do a trace route.

My friend who uses Cox had similar route to my AT&T trace, less than 10 hops and all stayed within SoCal.

Looks like the traceroute actually went from LA (LAX) to Houston (IAH) to Dallas (DFW) for a few hops then back to LA....ick.

Would be helpful if other Charter users in the SoCal area would post traceroutes to the IP you specified. Wouldn't be the first time I've seen differences in routing with a specific IP block.

I am not a Charter customer, nor in Southern California. I am intrigued that I am going through Kansas City and Dallas en route to that IP address. Seems to be something going on in the Cogentco transit?

Tracing route to 12.129.209.68 over a maximum of 30 hops
 
  1     2 ms    <1 ms     7 ms  Chihiro [192.168.102.1]
  2    36 ms    39 ms    67 ms  173-228-7-1.dsl.static.sonic.net [173.228.7.1]
  3    34 ms    39 ms    36 ms  gig1-4.cr1.lsatca11.sonic.net [70.36.243.13]
  4    31 ms    40 ms    39 ms  0.xe-5-1-0.gw.pao1.sonic.net [69.12.211.1]
  5    44 ms    39 ms    36 ms  te0-0-0-15.ccr21.sjc04.atlas.cogentco.com [38.104.141.81]
  6    47 ms    39 ms    47 ms  be2018.mpd22.sfo01.atlas.cogentco.com [154.54.28.81]
  7    73 ms    77 ms    79 ms  be2134.mpd21.mci01.atlas.cogentco.com [154.54.6.38]
  8    86 ms    88 ms    92 ms  be2064.ccr21.dfw01.atlas.cogentco.com [154.54.5.218]
  9    86 ms    88 ms    89 ms  be2032.ccr21.dfw03.atlas.cogentco.com [154.54.6.54]
 10   128 ms   123 ms   138 ms  192.205.36.221
 11   116 ms   121 ms   125 ms  cr2.dlstx.ip.att.net [12.122.138.18]
 12   121 ms   125 ms   120 ms  cr2.la2ca.ip.att.net [12.122.28.178]
 13   128 ms     *      116 ms  gar29.la2ca.ip.att.net [12.122.129.241]
 14   125 ms   119 ms   129 ms  12-122-254-238.attens.net [12.122.254.238]
 15   127 ms   128 ms   127 ms  mdf001c7613r0003-gig-12-1.lax1.attens.net [12.129.193.254]
 16     *        *        *     Request timed out.
 

I called Charter today, they have NO idea, nor do they have a Tier 2 anymore according to the phone and live chat.

I'm tempting to call Cogento Communications but I'm sure they'll just go WE DON'T SUPPORT END USERS or LOL GTFO.

Please if you have experiencing lags or even spikes please post your trace route results to the server at 12.129.209.68 thanks.

EDIT:

I decided to ditch Charter and move with uVerse... half the price with half the downstream but same upstream. Sorry Charter but I had fun.

This is one of those cases where you basically have no option but to switch providers, or wait it out and hope it gets better. Any Charter support you'd be able to contact is too low level to have any idea what you're talking about, or have any control over getting it resolved. And even if you did get someone who knew what you were talking about, I doubt they would even consider it an issue. All they have to do is provide you with internet access, they don't guarantee good routing or low ping.
Plus, it's not really in their control anyway. It mainly looks like an issue between AT&T and Cogent, and as you said, they would only look into it if contacted by one of their own large customers (like Charter).

The issue is with Charter's upstream provider, Cogent. AT&T does not pay Cogent anything, I can tell you that right now.

It is crazy stupid how it's going from LA Internet Exchange to Dallas, to hand off to AT&T, to head back to the LA Exchange.

But that's peering disputes for you. Cogent probably has a ratio issue with AT&T in LAX or Dallas, so they rerouted traffic to bring their ratios more inline to maintain peering.

Thanks for your information!

After some research I came to the same conclusion.

Anyway I have setup an appointment to have uVerse installed coming Monday.

Thanks Charter for the fun.