Aroberts64's profile

New Contributor

 • 

10 Messages

Closed

Problem with traffic to Blizzard servers (World of Warcraft) via Cogentco

The traffic from my house in AZ is sent to Phoenix and then sent out to a Cogentco connection between Cox and ATT (Blizzards provider).  At the Cogentco interface there appears to be a problem with saturation and our data is being slowed down considerably by this saturation.  I am not the only one experiencing this issue.  Below are the links to multiple threads showing many people with problems connecting from the AZ area and others (some on different ISPs but still routing through Cogentco).

http://us.battle.net/wow/en/forum/topic/7006347308

http://us.battle.net/wow/en/forum/topic/7593741008

http://us.battle.net/wow/en/forum/topic/7592401496

http://us.battle.net/wow/en/forum/topic/7593741878

http://us.battle.net/wow/en/forum/topic/6552145272

When I call and try to get some sort of help from customer service, they have no clue what Im talking about and cant give me any means of contacting someone who does.  When I contacted someone from the home networking department (which I was told to do via the online customer chat) I was passed around to someone else after explaining what the problem was to one person, and then while trying to explain the issue to this person I was hung up on.

Im sure Im going to be told this isnt an ISP problem but rather a problem on Blizzards end.  And it isnt exactly a Cox Comm problem but it is a problem with how you route our traffic.  Everyone who is being routed through these Cogentco connections is having the same issue, regardless of who their ISP is.  Cogentco wont do anything to address the issue because we arent their customers and have told us that we need to contact our ISPs so that the ISP can address the issue with Cogentco.

Please do something to fix this.

New Contributor

 • 

12 Messages

Long time since we had input from Cox on this thread

Come on Cox, time is up.

After two weeks of excuses I said I'd give until the end of the week.

Tomorrow I will be checking here and then swapping to Century Link.

Last chance

New Contributor

 • 

15 Messages

I thought maybe a pathping could help show where the problem is. The packet loss starts at an AT&T IP adrress in Kansas City:

192.205.37.173

This  node seems to have a history of high latency/bad connections with the blizzard servers dating back to 2011, I believe. At least those are the records google search reveals.I'm sorry if the original format got messed up pasting into this box, but the info is there, even if it's skewed a little.

Tracing route to 63.240.161.189 over a maximum of 30 hops

  0  JOEBLOW2.ph.cox.net [xx.xxx.xx.xxx]
  1  10.112.56.1
  2  172.21.0.18
  3  mcdlcorc01-pos-0-7-0-0.ph.ph.cox.net [70.169.72.72]
  4  mcdldsrj01-ae2.0.rd.ph.cox.net [70.169.76.225]
  5     *     68.1.5.139
  6  te0-7-0-21.ccr22.lax05.atlas.cogentco.com [154.54.13.129]
  7  te0-4-0-3.mpd22.lax01.atlas.cogentco.com [154.54.88.198]
  8  te0-3-0-6.mpd22.iah01.atlas.cogentco.com [154.54.0.254]
  9  te0-2-0-1.ccr22.dfw01.atlas.cogentco.com [154.54.25.97]
 10  te0-2-0-1.ccr22.mci01.atlas.cogentco.com [154.54.46.217]
 11  te0-3-0-2.ccr22.ord01.atlas.cogentco.com [154.54.6.213]
 12  te0-1-0-5.ccr22.ord03.atlas.cogentco.com [154.54.5.2]
 13  192.205.37.173
 14  cr2.cgcil.ip.att.net [12.122.84.94]
 15     *     gar2.clboh.ip.att.net [12.122.133.197]
 16  12.122.251.22
 17  63.240.130.214
 18     *        *        *     
Computing statistics for 425 seconds...
            Source to Here   This Node/Link
Hop  RTT    Lost/Sent = Pct  Lost/Sent = Pct  Address
  0                                           JOEBLOW2.ph.cox.net [xx.xxx.xx.xxx]
                                0/ 100 =  0%   |
  1    7ms     0/ 100 =  0%     0/ 100 =  0%  10.112.56.1
                                0/ 100 =  0%   |
  2   11ms     0/ 100 =  0%     0/ 100 =  0%  172.21.0.18
                                0/ 100 =  0%   |
  3   13ms     0/ 100 =  0%     0/ 100 =  0%  mcdlcorc01-pos-0-7-0-0.ph.ph.cox.net [70.169.72.72]
                                0/ 100 =  0%   |
  4   12ms     0/ 100 =  0%     0/ 100 =  0%  mcdldsrj01-ae2.0.rd.ph.cox.net [70.169.76.225]
                                0/ 100 =  0%   |
  5   26ms     0/ 100 =  0%     0/ 100 =  0%  68.1.5.139
                                0/ 100 =  0%   |
  6   23ms     0/ 100 =  0%     0/ 100 =  0%  te0-7-0-21.ccr22.lax05.atlas.cogentco.com [154.54.13.129]
                                0/ 100 =  0%   |
  7   22ms     0/ 100 =  0%     0/ 100 =  0%  te0-4-0-3.mpd22.lax01.atlas.cogentco.com [154.54.88.198]
                                0/ 100 =  0%   |
  8   59ms     0/ 100 =  0%     0/ 100 =  0%  te0-3-0-6.mpd22.iah01.atlas.cogentco.com [154.54.0.254]
                                0/ 100 =  0%   |
  9   63ms     0/ 100 =  0%     0/ 100 =  0%  te0-2-0-1.ccr22.dfw01.atlas.cogentco.com [154.54.25.97]
                                0/ 100 =  0%   |
 10   74ms     0/ 100 =  0%     0/ 100 =  0%  te0-2-0-1.ccr22.mci01.atlas.cogentco.com [154.54.46.217]
                                0/ 100 =  0%   |
 11   90ms     0/ 100 =  0%     0/ 100 =  0%  te0-3-0-2.ccr22.ord01.atlas.cogentco.com [154.54.6.213]
                                0/ 100 =  0%   |
 12   87ms     0/ 100 =  0%     0/ 100 =  0%  te0-1-0-5.ccr22.ord03.atlas.cogentco.com [154.54.5.2]
                               13/ 100 = 13%   |
 13  103ms    13/ 100 = 13%     0/ 100 =  0%  192.205.37.173
                               87/ 100 = 87%   |
 14  ---     100/ 100 =100%     0/ 100 =  0%  cr2.cgcil.ip.att.net [12.122.84.94]
                                0/ 100 =  0%   |
 15  ---     100/ 100 =100%     0/ 100 =  0%  gar2.clboh.ip.att.net [12.122.133.197]
                                0/ 100 =  0%   |
 16  ---     100/ 100 =100%     0/ 100 =  0%  12.122.251.22
                                0/ 100 =  0%   |
 17  ---     100/ 100 =100%     0/ 100 =  0%  63.240.130.214

Trace complete.

New Contributor

 • 

19 Messages

5     *     68.1.5.139  That is the same node I have been having problems with and it is not in KC it turns out it is really in Macon GA, and it belongs to Cox.

New Contributor

 • 

15 Messages

If you look closer at the pathping I submitted, you'll see j was able to ping the node you are talking about, and it was pretty good latency. The node I pointed at IS in Kansas City, and it IS an AT&T node, unless whois lied to me, and that's where it started losing packets and degraded

New Contributor

 • 

1 Message

Any new updates on if anything is going to get done? If not I'll also be happy to jump on the century link bandwagon. 

New Contributor

 • 

2 Messages

Is anyone seeing an issue with more than just Blizzard games?


Never mind, I see that others said that they are having issues with other connections.

New Contributor

 • 

4 Messages

Still, nothing but crickets chirping...tried the email approach, too...just got the insulting form letter=the equivalent of "Did you turn it off and on again, then?"

I have a t-shirt that says that, now...

Crickets------

New Contributor

 • 

2 Messages

When an ISP peers with a transport level carrier the traffic past the point of peering is out of the ISPs control.  So what would have to happen is Cox would need to circumvent that carrier entirely.  If they're using any kind of conventional protocols it could cause twice the latency if not worse by bottle-necking with another ISPs throughput over what would become another over saturated link.  Conversely it could cause traffic to simply take a much longer and unusual route, which would also add latency to the connection.  In addition to that, if they were to move to a different physical path, the routing protocols would potentially pass traffic back through the same ATT/Cogentco link that has problems now.  Networks are self healing when they experience irreparable issues so if this path is being chosen now - regardless of the latency issues - the odds that the backbone would route the traffic through this link anyways is very high. 

Usually tier 1 network carriers are fairly quick to repair saturation issues as it affects the primary service they sell.  The only thing that could slow them down is the ISPs that would be affected by the work as it usually involves disabling the connection in question for as long as it takes to make the change.  

New Contributor

 • 

15 Messages

Try googling "Cogentco high latency", they've had complaints and issues with high latency clear back to 2005. As far as online games goes, I've read posts in guild wars 2 and rift, complaining about high latency. Guess who the ISP's name is. I'm posting in their forum now. Cox originally wanted to send a technician to my home to check my wiring. When I sent the link to blizzards forum posts , citing Cox as the common denominator, then Cox responded that it was a node close to my house in Phoenix causing the issue, It;s been stated in this thread that Cox changed it's routing the beginning of this month, right when everyone started having the high latency, and they changed it to the service who has a history of high latency complaints. Why would they do this, since that service's reputation can be googled, and is considered a good cheap backup service if your primary routing service goes down for maintenance. I think the answer is economics. They thought they could serve their customers and pay less for routing at the same time. There's posts in other forums suggesting what was done last month when latency got out of hand with other Cox customers, reroute around the problem node, and service was good again. Cox doesn't seem to want to do this again, and they don't want to be honest with their customers about what the real problem is. They'd rather waste money, sending a tech out to each of our homes, trying to put the blame on our wiring,, than state the obvious, they have a problem in their system. Century customers are having no connection or latency issues, at least here in Phoenix. Their 12 mag service is getting half the latency and twice the framerate playing world of warcraft. I'm paying for Cox Ultimate Internet and getting latency up to 4000ms before i get disconnected completely and have to logon again. But at 3 AM, my latency is rock solid at 80 to 85ms, even running 4 accounts at once(I multibox when my connection is good). Thats a BIG variation and it wasn't there a month ago.

New Contributor

 • 

7 Messages

I am seriously fed up. Cox has stopped replying to this thread completely. When I call in they are anything and everything BUT helpful. They refuse to accept responsibility. I will be switching ISPs immediately and I suggest you all do the same. There is no reason we should give an abhorrent business with complete and utter disregard for their customer's intelligence (by telling us we don't know what we're talking about) and satisfaction;  like Cox our money.

New Contributor

 • 

7 Messages

KipK, On behalf of the Cox Customers that posted here I would like to thank you for posting here and especially for fixing the issue. I just wish all the pushback we (the customers) received from Cox hadn't happened. The number of times I was told by Cox representatives that its not their problem was abhorrent.

New Contributor

 • 

2 Messages

Thanks all you want to Cox, many of us have already moved and are getting much improved speed, lower cost and better customer service.

 

The best thing I did was move to FiOS and saw the huge difference in every aspect of my account. Adios Cox I had you for over 10 years and you just crapped on me more and more.

 

Hope those of you that remain enjoy your increased prices for the same old service ROFLMAO.

New Contributor

 • 

15 Messages

I would be thrilled to have FIOS available here in Phoenix. I would never have even wasted my time here, trying to prove the issue existed to an ISP that already knows theres a problem, but prefers to lie to customers, rather than accepting blame  and working to fix it. Not until the problem had improved, did they even admit there was a problem. Every time somebody posted a pathping, Cox's "YES" man, chris, pointed to a different post or a different part of the pathping and said that he saw no problem.  Cox, just in case you really are concerned now, when I connect to blizzards public test server, somewhere around Los Angeles, I have around 25 latency, Today, playing on my server, bbased in Chicago, my latency was around 120. Improved? yes! The speed that I'm paying $100 a month for? not even close! Please tell us. Is this as good as its gonna get?

Related Content

  • Closed

    8

    0

  • Closed

    9

    0

  • Closed

    19

    0

  • Closed

    2

    0

  • Closed

    2

    0

Recent Discussions

View More

Loading...