Re: Bad Pings to a California Server (bad hops?) in Cox HSI http://www.dslreports.com/forum/r20457343 en Thu, 08 Jan 2009 00:15:25 EDT Thu, 08 Jan 2009 00:15:25 EDT Re: Bad Pings to a California Server (bad hops?) http://www.dslreports.com/forum/remark,20465906 Wolf_sd : I'm actually in the same boat here, and a new user here so my apologies upfront. I've had this problem since back in November with WoW and I'm located in San Diego, connecting to the Daggerspine server which is a west coast server. Also to note, I've tried other west coast servers and the same thing happens.

My problem, which seems to be occurring a lot recently is that I'll be playing fine, no spikes or anything, and then the game lags out and I disconnect. Then my internet just lags for a few seconds (can't web browse or reconnect to wow) and then bam, I'm connected again like nothing happened. Afterwards, one of two things happen. Either 1, I relog into wow and instantly lagout again, or 2, the game plays fine like nothing happened. This fit can go on and on over the period of the day, or sometimes happen only once. Regardless it's guaranteed to happen every day.

I've been in contact with both Blizzard tech reps and Cox tech reps via email and even had Cox send someone out to look at my lines. I've done tons of troubleshooting that Blizzard has suggested, and even picked up a new modem and still the problem persists. What's even worse is it seems to only be WoW that the disconnecting occurs. Recently, I sent Blizzard reps my tracerts and pathpings to see if they can troubleshoot better. Below is the data that I emailed to the Blizzard rep.



Tracing route to 12.129.233.88 over a maximum of 30 hops

1 1 ms 1 ms 1 ms 192.168.1.1
2 9 ms 5 ms 6 ms 10.157.0.1
3 14 ms 8 ms 7 ms 68.6.11.154
4 8 ms 7 ms 9 ms fed1dsrj01-ge500.0.rd.sd.cox.net [68.6.8.0]
5 7 ms 9 ms 8 ms te-3-1.car1.SanDiego1.Level3.net [4.79.32.113]
6 25 ms 18 ms 17 ms ae-5-5.ebr1.LosAngeles1.Level3.net [4.69.133.206]
7 14 ms 17 ms 17 ms ae-61-61.csw1.LosAngeles1.Level3.net [4.69.137.2]
8 15 ms 12 ms 9 ms ae-1-69.edge1.LosAngeles1.Level3.net [4.68.20.7]
9 13 ms 12 ms 10 ms 192.205.33.229
10 11 ms 13 ms 13 ms 12.127.3.218
11 15 ms 11 ms 11 ms 12.127.3.185
12 14 ms 13 ms 12 ms 12.122.255.74
13 10 ms 11 ms 20 ms mdf001c7613r0004-gig-12-1.lax1.attens.net [12.129.193.246]

Tracing route to 12.129.233.88 over a maximum of 30 hops

0 CHRIS-II [192.168.1.102]
1 192.168.1.1
2 10.157.0.1
3 68.6.11.154
4 fed1dsrj01-ge500.0.rd.sd.cox.net [68.6.8.0]
5 te-3-1.car1.SanDiego1.Level3.net [4.79.32.113]
6 ae-5-5.ebr1.LosAngeles1.Level3.net [4.69.133.206]
7 ae-61-61.csw1.LosAngeles1.Level3.net [4.69.137.2]
8 ae-1-69.edge1.LosAngeles1.Level3.net [4.68.20.7]
9 192.205.33.229
10 12.127.3.218
11 12.127.3.185
12 12.122.255.74
13 mdf001c7613r0004-gig-12-1.lax1.attens.net [12.129.193.246]
14 * * *
Computing statistics for 350 seconds...

Source to Here This Node/Link

Hop RTT Lost/Sent = Pct Lost/Sent = Pct Address

0 CHRIS-II [192.168.1.102]

0/ 100 = 0% |

1 0ms 0/ 100 = 0% 0/ 100 = 0% 192.168.1.1

0/ 100 = 0% |

2 7ms 13/ 100 = 13% 13/ 100 = 13% 10.157.0.1

0/ 100 = 0% |

3 10ms 1/ 100 = 1% 1/ 100 = 1% 68.6.11.154

0/ 100 = 0% |

4 9ms 1/ 100 = 1% 1/ 100 = 1% fed1dsrj01-ge500.0.rd.sd.cox.net [68.6.8.0]

0/ 100 = 0% |

5 16ms 1/ 100 = 1% 1/ 100 = 1% te-3-1.car1.SanDiego1.Level3.net [4.79.32.113]

0/ 100 = 0% |

6 16ms 1/ 100 = 1% 1/ 100 = 1% ae-5-5.ebr1.LosAngeles1.Level3.net [4.69.133.206]

0/ 100 = 0% |

7 17ms 1/ 100 = 1% 1/ 100 = 1% ae-61-61.csw1.LosAngeles1.Level3.net [4.69.137.2]

0/ 100 = 0% |

8 --- 100/ 100 =100% 100/ 100 =100% ae-1-69.edge1.LosAngeles1.Level3.net [4.68.20.7]

0/ 100 = 0% |

9 11ms 0/ 100 = 0% 0/ 100 = 0% 192.205.33.229

1/ 100 = 1% |

10 --- 100/ 100 =100% 99/ 100 = 99% 12.127.3.218

0/ 100 = 0% |

11 --- 100/ 100 =100% 99/ 100 = 99% 12.127.3.185

0/ 100 = 0% |

12 --- 100/ 100 =100% 99/ 100 = 99% 12.122.255.74

0/ 100 = 0% |

13 20ms 1/ 100 = 1% 0/ 100 = 0% mdf001c7613r0004-gig-12-1.lax1.attens.net [12.129.193.246]

99/ 100 = 99% |

14 --- 100/ 100 =100% 0/ 100 = 0% CHRIS-II [0.0.0.0]

Trace complete.

This is what a rep had to say:

Thank you for the update. Your traces do not appear to show any problems along the connection just as they did before. However, your ping tests show a complete loss of signal at hops 8 and 10-12. These hops appear to be in and after the Level3 network and are probably located in AT&T's net as well. However, you are a direct customer of Cox so you would need to contact them if your connection was experiencing any issues. They can forward the problem on to the other networks for investigation.


So I contacted Cox via email and they responded with:

Thank you for your e-mail. I understand you are having difficulties
with a trace route. I apologize for any inconvenience this may cause
you.

Unfortunately, We are not allowed to support this difficulty because we
do not have the license agreement to support it. I suggest you contact
the manufacturer for a further assistance regarding this difficulty.

I hope the information above was useful in resolving your difficulty. If
you would like to inquire about other information, do not hesitate to
contact us for further assistance.


I'm really at a loss here. I've tried just about everything I can think of. I've tweaked WoW settings, updated drivers, replaced NIC cards, replaced modem, reinstalled WoW, by passed router, and still nothing. Could anyone offer any insight?]]> http://www.dslreports.com/forum/remark,20465906 Sun, 11 May 2008 15:43:22 EDT Re: Bad Pings to a California Server (bad hops?) http://www.dslreports.com/forum/remark,20460008 Goofball : Edit: Note, I am in the Phoenix area as well (Avondale).

I spoke too soon. I'm still seeing the loss, it's just very intermittent at the moment.

Looks like there is some route flap or a piece of iffy hardware messing with things somewhere along the way and causing retransmits. I'm watching my response times go from 60-70ms -> lost packet or two -> steady 300-400ms for a bit -> lost packet or two -> steady 60-70ms.

All testing so far is ICMP ping directly to fairly idle devices on the AT&T connection and to devices directly inside my work LAN via VPN that I know aren't configured to or busy enough to be de-prioritizing/dropping ICMP. The VPN eliminates any devices along the way that might be configured to drop ICMP as it encapsulates everything in IPSEC packets before sending.

I suppose I'll go eat some dinner, test to make sure it occurs on the office cable modem and then get Cox business support on the phone and see if they can get this annoyance fixed.

Traceroutes below if anyone cares.

Home Traceroute:

Traceroute from the Comcast - Located in Colorado Springs:
]]> http://www.dslreports.com/forum/remark,20460008 Sat, 10 May 2008 00:37:43 EDT Re: Bad Pings to a California Server (bad hops?) http://www.dslreports.com/forum/remark,20459870 Goofball : Somebody found a problem somewhere in there and fixed it in the last few minutes. I've been messing around with tracking down an intermittent packet loss problem with my work VPN (Cox to AT&T via Level3) that's been going on for the last few days that sounds exactly like the original poster's issue. It wasn't high on my priority list as it wasn't a complete outage, it just made things a little "chunky" for most of the tools I use.

I was just getting ready to see if I could recreate the problem on the backup Cox business service cable modem we have at work between it and our primary AT&T internet service. I wanted to make sure the issues weren't with my local modem and node before I started calling support.

I'm pretty sure the problem was somewhere in the Cox network or Cox/Level3 peering as I have 8 other diverse ISP/Geographical sites connected via VPN over the same AT&T connection that had absolutely no issues. One of those sites is on Comcast business service cable modem that routes to AT&T via Comcast/Level3 peering with no issues.

In my testing at home if anyone cares: Smallish (up to 500 bytes) ICMP packets seemed to pass with only marginally bad 6% loss both inside and outside the VPN. Large (5000 bytes plus) ICMP packets were at about 20% loss outside the VPN and around 50% inside due to the nature of the connection and packet fragmentation.

I'd be interested to know what the problem was if anyone has any insight.]]> http://www.dslreports.com/forum/remark,20459870 Fri, 09 May 2008 23:49:14 EDT Re: Bad Pings to a California Server (bad hops?) http://www.dslreports.com/forum/remark,20458346 Dimp : PM Sent! i really appreciate the fast and friendly support! And hope that we can get this resolved (if there is anything at all COX can do)]]> http://www.dslreports.com/forum/remark,20458346 Fri, 09 May 2008 17:40:34 EDT Re: Bad Pings to a California Server (bad hops?) http://www.dslreports.com/forum/remark,20457847 AntiSnipe : As an ex-wow player, I can tell you a WoW server is no place to trouble shoot. You could ping 20 to the server and still have HORRID lag in game due to their data base servers (or whatever they call them) being slow. It's not network lag, but CPU overload on their server.

That said, there may be packet loss or something within the Cox network giving you trouble as well. Your tracert actually looks really good. You go through a lot of the same hops as me, so I assume you're in Tempe, Mesa, chandler, or there abouts like me, in Mesa.

If your not having connection drops on your modem and you can get decent round trip pings, I'd have to blame Blizzard and their laggy servers.

Can you check your modem's signal quality? I am having a problem with very poor upstream power on mine right now from Mesa causing total connection loss, but modem lights are on as usual. Could be related?]]> http://www.dslreports.com/forum/remark,20457847 Fri, 09 May 2008 15:50:30 EDT Re: Bad Pings to a California Server (bad hops?) http://www.dslreports.com/forum/remark,20457769 CoxTOC1 : As a WOW player myself I can honestly say that I understand your frustration.

First thing to note is that your round trip time is in the 30ms range so no problems their. As for the *'s in hop 7 and 8 that's probably ICMP de-prioritization (For more info on what that is try here: »[OK] Major packet loss--- can't watch video online?)

Now what makes this very frustrating is that Blizard blocks ICMP on their network. So we have no way to test end to end connectivity once we reach their network. And to make it even more fun blizzard does not use ICMP to calculate their ping times (They use something else that totally escapes me right now) so where does this leave you? As far as we (The ISP) are concerned we are getting you their in about 30ms which is rely good. Which puts you back on Blizards door step.

Some things that may help you on your road to solving your problem are:

»www.wowwiki.com/US_Realm_List_By_Datacenter

and

»forums.worldofwarcraft.com/threa···41&sid=1

And as always if you want to PM me your modem MAC address I'll check stuff out. In fact if you get me on a slow enough day I'll even conference into a call with Blizard tech support and help you through it.]]> http://www.dslreports.com/forum/remark,20457769 Fri, 09 May 2008 15:37:05 EDT Re: Bad Pings to a California Server (bad hops?) http://www.dslreports.com/forum/remark,20457385 Dimp : Very sorry for triple post! I saw that people like it if your registered so that PMs can be sent soooo ... I registered under this name]]> http://www.dslreports.com/forum/remark,20457385 Fri, 09 May 2008 14:11:15 EDT Re: Bad Pings to a California Server (bad hops?) http://www.dslreports.com/forum/remark,20457343 anon : Sorry! forgot to mention (as well as put in subject) that it is Phoenix AZ]]> http://www.dslreports.com/forum/remark,20457343 Fri, 09 May 2008 14:03:38 EDT Bad Pings to a California Server (bad hops?) http://www.dslreports.com/forum/remark,20457338 anon : Hey guys, Brand new to this forum so i am very sorry if i am doing something incorrectly!

I have been having very bad ping problems the past 3 weeks with a World of Warcraft server that i connect to.
I have talked to multiple people having this problem, and there is even a thread of phoenix and surrounding cox customers with problems connecting to the California servers.

Before you tear into me, trust me i have contacted both COX and Blizzard (the world of warcraft company) about this problem, and both say it is out of their control.

Cox sent me to a level 2 tech somewhere in San Diego, who basically told me that because i am only a normal residential user they could only do so much for me.

People have told me that it has something to do with the hop at Level 3, but being that that is neither mine nor Blizzard's ISP i do not have any DIRECT support with them. Blizzard is telling everyone to have their ISPs contact Level 3 because Cox is Level 3's customer, so they should be able to get SOMETHING done.

I dont know what to do at this point. The game is almost unplayable, and i do realize its only a game however it is important to me as i play with family on it.

The ping seems to spike to 700-2000 ping almost constantly during play.

Help! What can/should i do about this issue

My Tracert:
1 10 ms 9 ms 12 ms 10.145.128.1
2 11 ms 10 ms 9 ms 68.2.8.17
3 10 ms 13 ms 12 ms 68.2.12.78
4 21 ms 12 ms 10 ms 68.2.12.1
5 11 ms 9 ms 11 ms chnddsrj02-ae1.0.rd.ph.cox.net [68.2.14.13]
6 24 ms 204 ms 27 ms chnddsrj01-so010.91.rd.ph.cox.net [70.169.76.5]
7 13 ms 21 ms * ae-11-11.car2.Phoenix1.Level3.net [4.69.133.34]
8 29 ms * 23 ms ae-4-4.ebr2.LosAngeles1.Level3.net [4.69.133.38]
9 30 ms 35 ms 21 ms ae-62-62.csw1.LosAngeles1.Level3.net [4.69.137.18]
10 31 ms 29 ms 20 ms ae-1-69.edge1.LosAngeles1.Level3.net [4.68.20.7]
11 29 ms 21 ms 23 ms 192.205.33.225
12 29 ms 23 ms 29 ms 12.127.3.210
13 33 ms 20 ms 29 ms 12.127.3.185
14 21 ms 33 ms 30 ms 12.122.255.74
15 22 ms 29 ms 30 ms mdf001c7613r0004-gig-12-1.lax1.attens.net [12.129.193.246] ]]> http://www.dslreports.com/forum/remark,20457338 Fri, 09 May 2008 14:02:06 EDT