[Savagex]

Hello All,

I'm looking for some input to the problem I have below. I'm not an expert network user or anything but I will see if I can explain it as best I can.

I have Telus Optik TV/Internet service (not the turbo, Actiontec V1000H router). I have had the connection for almost 2 years now and it has always been very stable.

A couple of weeks ago I got a phone call from Telus saying that the following friday they were going to be doing some upgrades to their network and I should expect a 5 min interuption in service, after which I would have to restart all my gear and all would be good.

On that day all was good that morning, I had no issues and service was business as usual. About mid afternoon I had the service go out as expected so I restarted my gear. First thing I noticed was that Telus had assigned me an IP address in the 75.xx... range (different from before) no biggie, as I knew it should not effect my service in any way.

Later that night I logged into World of Warcraft and noted my ping was about 50-80ms higher than it has been for the last year and a half. At first I didn't sweat it, I was sure it would settle out after a couple of days like these little burps always do. Well it has now been almost 2 weeks and I'm still getting the same increased latency.

I have tracert'ed and pinged my servers IP daily since then and keep coming back with the same results. (while doing these I powered down all other devices on my network and pluged the computer directly into my router)

I have also tracert'ed and pinged the same IP from my mom's house as well as my buddys (both Optik TV service) and they are showing results like I used to get with no hint of increased latency.

Whats puzzling to me is that when I tracert from home, the latency spike comes at around hop 19, but when I tracert from my moms or buddys, the same hop, to the same IP gives no such spike in latency.

Today I called Telus Tech Support. The lady said she did some tests and said the connection looked good. She said she would reset the connection and that I should wait a day or two to see if it helps.

Any experts with any thoughts ?

Thanks in advance.

 

[scrooloose]

Are you saying you only experience the high latency to this one server, or everything? If it's just the one server, perhaps the traffic is being routed differently than it was before (you mentioned a new IP address). I don't suppose you have a before trace route? What is the hop where you begin seeing the high latency? If it's outside your ISP network, there might not be much anyone can do. Does WOW allow you to find a server closer to you (I don't know much about that game)?

-Mike

 

[Savagex]

WoW stores all its game data server side, once you create a toon on that server its stuck there unless you pay $25 to move it. (Moving servers is out of the question as I have friends that play there).

I don't play on any of the other servers so I wouldn't know the norm on them.

I don't have a tracert from before, as there wasn't an issue.

I have done the same tracerts at my house, my moms and my buddys (all on telus optik) and compaired each and every hop. All 3 take about the same route and have the same response times up to hop 19, which is the same IP across all 3, but in my case I get a latency spike and not the other 2. What I'm having a hard time understanding is why my tracerts show a large spike at that point. I'm pretty sure this point is not inside Telus' network but why does my connection take a hit ?

 

[Savagex]

Just an addition. Here is a jpeg of pinging and tracerts. The left is from my moms house and the right is from mine. As you can see the latency jump happens at hop 19 (yes the IP's are slightly different, so I ping the other one at the end with the same result on the right). Still not sure why this spike is happening and not sure who to contact to get it fixed ?

http://i53.photobucket.com/albums/g67/Savagex128/Tracert-Ping.jpg

 

[scrooloose]

This is mine from work but I don't see the high latency at the 192.205.37.105 hop like you do. That appears to be where Cogent meets AT&T. I'm afraid you're going to have a real hard time finding anyone interested in troubleshooting this, especially since it's out in the carrier networks! I don't know what to tell you.

4 4 ms 3 ms 3 ms te0-3-0-0.ccr21.yyz02.atlas.cogentco.com [154.54
.40.137]
5 16 ms 15 ms 15 ms te0-1-0-6.ccr21.jfk02.atlas.cogentco.com [154.54
.28.33]
6 23 ms 21 ms 22 ms te0-3-0-3.ccr21.dca01.atlas.cogentco.com [154.54
.26.185]
7 22 ms 22 ms 22 ms te0-0-0-5.ccr21.iad02.atlas.cogentco.com [154.54
.30.122]
8 23 ms 22 ms 22 ms te0-0-0-4.mpd22.iad02.atlas.cogentco.com [154.54
.31.106]
9 23 ms 24 ms 23 ms 192.205.37.105
10 24 ms 23 ms 23 ms cr2.wswdc.ip.att.net [12.122.134.162]
11 24 ms 23 ms 23 ms cr2.n54ny.ip.att.net [12.122.3.37]
12 23 ms 29 ms 23 ms cr84.n54ny.ip.att.net [12.122.115.94]
13 23 ms 22 ms 22 ms gar20.n54ny.ip.att.net [12.122.130.237]
14 24 ms 24 ms 23 ms 12-122-254-114.attens.net [12.122.254.114]
15 23 ms 24 ms 23 ms mdf001c7613r0003-gig-10-1.nyc3.attens.net [63.24
0.65.10]

-Mike

 

[Newby]

That is freaky.

 

[Savagex]

Just an update on this item:

The increased latency continued until February 24, 2011, at which time I came home from work and found that things had finally returned to normal. I ran a tracert (first one below) and found that my routing had greatly improved. I noticed that I was taking one more hop inside of Telus' network and less hops in Cogentco's.

I came home last night (March 3, 2011) to find that things had went bad again (second tracert).

I'm wondering who decides what route my traffic takes ? Telus did make a change (for the better) in their network for the time between Feb 24 and Mar 3, then reverted that change. Who do I contact to get this looked into ?

I know that the latency spike occurs at the peering node between Cogentco and AT&T, however, the change that was made to my routing made that spike go away.

Anybody able to give some insight ?

Thanks in advance,
Adrian


(PS - The final IP I'm tracking below is the actual WoW server and is protected from ping/tracert traffic - hence the timeouts once it reaches there)


First - Feb 24 to Mar 3 (good)

Tracing route to 199.107.6.232 over a maximum of 30 hops

1 <1 ms <1 ms <1 ms . [192.168.1.254]
2 10 ms 9 ms 8 ms 10.231.208.1
3 9 ms 8 ms 8 ms 173.182.206.134
4 11 ms 9 ms 9 ms 209.115.197.53
5 10 ms 10 ms 10 ms 154.11.94.252
6 * * * Request timed out.
7 * * * Request timed out.
8 39 ms 38 ms 38 ms chcgildtgr00.bb.telus.com [154.11.11.30]
9 39 ms 38 ms 38 ms 173.182.200.2
10 39 ms 39 ms 39 ms te0-3-0-0.ccr21.ord01.atlas.cogentco.com [154.54.5.25]
11 57 ms 58 ms 57 ms te0-0-0-1.ccr21.dca01.atlas.cogentco.com [154.54.40.185]
12 58 ms 58 ms 58 ms te0-1-0-5.ccr21.iad02.atlas.cogentco.com [154.54.2.50]
13 59 ms 59 ms 59 ms te0-2-0-0.mpd22.iad02.atlas.cogentco.com [154.54.31.102]
14 77 ms 76 ms 76 ms 192.205.35.105
15 80 ms 79 ms 80 ms cr2.wswdc.ip.att.net [12.122.134.166]
16 78 ms 79 ms 79 ms cr2.n54ny.ip.att.net [12.122.3.37]
17 80 ms 80 ms 79 ms cr84.n54ny.ip.att.net [12.122.115.94]
18 79 ms 78 ms 78 ms gar20.n54ny.ip.att.net [12.122.130.237]
19 79 ms 80 ms 79 ms 12-122-254-114.attens.net [12.122.254.114]
20 80 ms 80 ms 80 ms mdf001c7613r0004-gig-12-1.nyc3.attens.net [63.240.65.14]
21 * * * Request timed out.
22 * * * Request timed out.
23 * * * Request timed out.
24 * * * Request timed out.
25 * * * Request timed out.
26 * * * Request timed out.
27 * * * Request timed out.
28 * * * Request timed out.
29 * * * Request timed out.
30 * * * Request timed out.

Trace complete.


Second - Mar 3 until who knows when...


Tracing route to 199.107.6.232 over a maximum of 30 hops

1 <1 ms <1 ms <1 ms . [192.168.1.254]
2 10 ms 9 ms 9 ms 10.231.208.1
3 9 ms 9 ms 9 ms 173.182.206.130
4 12 ms 9 ms 9 ms 209.115.197.57
5 10 ms 10 ms 11 ms 154.11.94.252
6 * * * Request timed out.
7 * * * Request timed out.
8 24 ms 24 ms 23 ms 154.11.10.230
9 24 ms 24 ms 24 ms te4-2.ccr01.sea02.atlas.cogentco.com [154.54.10.45]
10 23 ms 24 ms 24 ms te3-4.ccr01.sea01.atlas.cogentco.com [154.54.1.197]
11 27 ms 27 ms 27 ms te4-1.ccr01.pdx02.atlas.cogentco.com [154.54.40.121]
12 45 ms 45 ms 44 ms te3-4.ccr01.slc01.atlas.cogentco.com [154.54.5.34]
13 56 ms 55 ms 56 ms te3-3.mpd02.den01.atlas.cogentco.com [154.54.3.13]
14 59 ms 59 ms 59 ms te0-4-0-3.ccr21.mci01.atlas.cogentco.com [154.54.7.129]
15 60 ms 59 ms 59 ms te0-2-0-3.ccr21.ord01.atlas.cogentco.com [154.54.7.186]
16 78 ms 78 ms 78 ms te0-2-0-1.ccr21.dca01.atlas.cogentco.com [154.54.40.193]
17 78 ms 78 ms 79 ms te0-2-0-0.mpd22.dca01.atlas.cogentco.com [154.54.26.37]
18 79 ms 79 ms 79 ms te0-2-0-1.mpd22.iad02.atlas.cogentco.com [154.54.5.42]
19 112 ms 112 ms 111 ms 192.205.37.105
20 116 ms 115 ms 115 ms cr2.wswdc.ip.att.net [12.122.134.170]
21 117 ms 116 ms 116 ms cr2.n54ny.ip.att.net [12.122.3.37]
22 116 ms 115 ms 115 ms cr84.n54ny.ip.att.net [12.122.115.94]
23 114 ms 114 ms 114 ms gar20.n54ny.ip.att.net [12.123.2.141]
24 115 ms 115 ms 115 ms 12-122-254-114.attens.net [12.122.254.114]
25 115 ms 116 ms 116 ms mdf001c7613r0004-gig-12-1.nyc3.attens.net [63.240.65.14]
26 * * * Request timed out.
27 * * * Request timed out.
28 * * * Request timed out.
29 * * * Request timed out.
30 * * * Request timed out.

Trace complete.

 

[markv]

One option you have, though it can be detrimental to speeds of other downloads, is disable Nagle's algorithm. It will tend to increase latency for gaming and I personally haven't had an issue with anything else I've been using, but that's not to say you won't have problems yourself down the line.

It is easy to undo, but if you don't like registry hacks then steer clear of it.

http://lifeandcode.net/2009/05/reduce-game-network-latency-in-windows-7-or-vista/

 

[Savagex]

markv,

I have applyed that reg hack long ago and was very happy with the results, however my new post is regarding how my internet traffic is being routed and how it seems to be causing a spike in latency that is not needed.

Thank you for your reply tho.

 

[rewrite]

I got similar result as yours. High latency at 192.205.37.105. I don't think the end user can do anything about it. There must be something going on with 192.205.37.105 and 192.205.35.105. The traffic is probably very congested at those nodes including the last two nodes in cogentco.com. In my linux trace route result, it jumps around those two nodes in the same trace route.

From my linux machine:

1 router (192.168.0.150) 0.342 ms 0.604 ms 0.785 ms
2 10.196.128.1 (10.196.128.1) 14.729 ms 15.146 ms 15.664 ms
3 209.121.73.146 (209.121.73.146) 23.975 ms 24.190 ms 24.618 ms
4 209.121.72.252 (209.121.72.252) 19.800 ms 19.982 ms 20.383 ms
5 154.11.22.116 (154.11.22.116) 22.091 ms 22.557 ms 23.239 ms
6 * * *
7 * * *
8 204.225.243.22 (204.225.243.22) 18.737 ms 18.757 ms 20.083 ms
9 te4-2.ccr01.sea02.atlas.cogentco.com (154.54.10.45) 21.298 ms 21.625 ms 22.489 ms
10 te8-3.ccr01.sea01.atlas.cogentco.com (154.54.44.89) 23.705 ms 24.022 ms te7-4.ccr01.sea01.atlas.cogentco.com (154.54.3.33) 24.934 ms
11 te4-1.ccr01.pdx02.atlas.cogentco.com (154.54.40.121) 28.878 ms 29.756 ms 30.516 ms
12 te7-1.ccr01.slc01.atlas.cogentco.com (154.54.5.22) 48.980 ms te3-4.ccr01.slc01.atlas.cogentco.com (154.54.5.34) 50.137 ms te7-1.ccr01.slc01.atlas.cogentco.com (154.54.5.22) 38.003 ms
13 te4-3.mpd01.den01.atlas.cogentco.com (154.54.0.41) 49.337 ms 50.217 ms te7-2.mpd01.den01.atlas.cogentco.com (154.54.45.42) 50.481 ms
14 te0-4-0-2.mpd22.mci01.atlas.cogentco.com (154.54.45.50) 67.230 ms 67.372 ms 67.484 ms
15 te0-2-0-3.mpd22.ord01.atlas.cogentco.com (154.54.6.254) 67.248 ms te0-0-0-3.mpd22.ord01.atlas.cogentco.com (154.54.25.78) 67.783 ms 67.693 ms
16 te0-0-0-1.mpd22.dca01.atlas.cogentco.com (154.54.40.233) 85.869 ms te0-1-0-1.mpd22.dca01.atlas.cogentco.com (154.54.40.237) 84.769 ms te0-3-0-4.mpd22.dca01.atlas.cogentco.com (154.54.40.245) 85.561 ms
17 te0-0-0-5.mpd22.iad02.atlas.cogentco.com (154.54.30.118) 85.962 ms te0-0-0-1.mpd22.iad02.atlas.cogentco.com (154.54.26.118) 85.237 ms te0-1-0-5.mpd22.iad02.atlas.cogentco.com (154.54.5.62) 85.919 ms
18 192.205.35.105 (192.205.35.105) 105.168 ms 192.205.37.105 (192.205.37.105) 105.489 ms 192.205.35.105 (192.205.35.105) 105.425 ms
19 cr2.wswdc.ip.att.net (12.122.134.166) 108.868 ms 108.869 ms 108.716 ms
20 cr2.n54ny.ip.att.net (12.122.3.37) 108.884 ms 108.861 ms 108.870 ms
21 cr84.n54ny.ip.att.net (12.122.115.94) 107.648 ms 107.916 ms 107.526 ms
22 gar20.n54ny.ip.att.net (12.123.2.141) 108.080 ms 108.618 ms 107.099 ms
23 12-122-254-114.attens.net (12.122.254.114) 108.618 ms 108.809 ms 108.375 ms
24 mdf001c7613r0004-gig-12-1.nyc3.attens.net (63.240.65.14) 108.323 ms 108.198 ms 108.420 ms

From Windows machine:

1 1 ms 1 ms 1 ms 192.168.0.150
2 14 ms 14 ms 15 ms 10.196.128.1
3 14 ms 13 ms 14 ms 209.121.73.146
4 14 ms 15 ms 15 ms 209.121.72.252
5 15 ms 14 ms 15 ms 154.11.22.116
6 * * * Request timed out.
7 * * * Request timed out.
8 18 ms 18 ms 17 ms 204.225.243.22
9 18 ms 19 ms 18 ms te4-2.ccr01.sea02.atlas.cogentco.com [154.54.10.45]
10 19 ms * 17 ms te3-4.ccr01.sea01.atlas.cogentco.com [154.54.1.197]
11 21 ms 21 ms 22 ms te4-1.ccr01.pdx02.atlas.cogentco.com [154.54.40.121]
12 39 ms 38 ms 39 ms te3-4.ccr01.slc01.atlas.cogentco.com [154.54.5.34]
13 50 ms 49 ms 49 ms te7-2.mpd01.den01.atlas.cogentco.com [154.54.45.42]
14 67 ms 67 ms 67 ms te0-4-0-2.mpd22.mci01.atlas.cogentco.com [154.54.45.50]
15 68 ms 67 ms 68 ms te0-4-0-3.mpd22.ord01.atlas.cogentco.com [154.54.30.177]
16 86 ms 86 ms 85 ms te0-1-0-1.mpd22.dca01.atlas.cogentco.com [154.54.40.237]
17 86 ms 86 ms 86 ms te0-2-0-1.mpd22.iad02.atlas.cogentco.com [154.54.5.42]
18 107 ms 106 ms 106 ms 192.205.37.105
19 109 ms 109 ms 109 ms cr2.wswdc.ip.att.net [12.122.134.166]
20 110 ms 110 ms 110 ms cr2.n54ny.ip.att.net [12.122.3.37]
21 109 ms 109 ms 109 ms cr84.n54ny.ip.att.net [12.122.115.94]
22 108 ms 108 ms 108 ms gar20.n54ny.ip.att.net [12.123.2.141]
23 109 ms 115 ms 110 ms 12-122-254-114.attens.net [12.122.254.114]
24 120 ms 109 ms 129 ms mdf001c7613r0004-gig-12-1.nyc3.attens.net [63.240.65.14]