Canada - Latency Issues With Shaw ISP

Technical Support
1

Was told to make a new thread rather than use the [Nov 11] Canada - Mid Match Latency Spikes thread. Blizzard Tech Support sent me to the forums from the ticket I had sent in. I’ve already had a tech from Shaw come by and they’ve said there’s nothing else to fix on their end, but the lag spikes are still happening and it’s very evident when it occurs.

ISP: Shaw
Location: NA - Sault Ste Marie, Ontario, Canada

Here’s a couple MTRs showing what is occurring:

|------------------------------------------------------------------------------------------|

| WinMTR statistics |

| Host - % | Sent | Recv | Best | Avrg | Wrst | Last |

|------------------------------------------------|------|------|------|------|------|------|

| No response from host - 100 | 42 | 0 | 0 | 0 | 0 | 0 |

| No response from host - 100 | 42 | 0 | 0 | 0 | 0 | 0 |

| 24.244.8.141 - 0 | 210 | 210 | 9 | 12 | 28 | 12 |

| rc4fs-tge0-14-0-15.mt.shawcable.net - 0 | 210 | 210 | 17 | 23 | 36 | 22 |

| rc3ec-be10.il.shawcable.net - 0 | 210 | 210 | 28 | 32 | 42 | 30 |

| rc4ec-be6-1.il.shawcable.net - 0 | 210 | 210 | 35 | 38 | 72 | 37 |

| eqix-ix-ch1.blizzard.com - 0 | 210 | 210 | 35 | 39 | 65 | 37 |

| ae1-br02-eqch2.as57976.net - 0 | 210 | 210 | 35 | 43 | 149 | 39 |

| be2-pe02-eqch2.as57976.net - 0 | 210 | 210 | 69 | 72 | 82 | 72 |

| chi-eqch2-ia-bons-04.as57976.net - 0 | 210 | 210 | 69 | 73 | 83 | 72 |

| 24.105.62.129 - 0 | 210 | 210 | 68 | 72 | 82 | 69 |

|________________________________________________|______|______|______|______|______|______|

WinMTR v0.92 GPL V2 by Appnor MSP - Fully Managed Hosting & Cloud Provider

|------------------------------------------------------------------------------------------|

| WinMTR statistics |

| Host - % | Sent | Recv | Best | Avrg | Wrst | Last |

|------------------------------------------------|------|------|------|------|------|------|

| No response from host - 100 | 208 | 0 | 0 | 0 | 0 | 0 |

| No response from host - 100 | 208 | 0 | 0 | 0 | 0 | 0 |

| 24.244.8.141 - 0 | 1044 | 1044 | 8 | 12 | 29 | 10 |

| rc4fs-tge0-14-0-15.mt.shawcable.net - 0 | 1044 | 1044 | 16 | 22 | 37 | 22 |

| rc3ec-be10.il.shawcable.net - 0 | 1044 | 1044 | 27 | 31 | 72 | 32 |

| rc4ec-be6-1.il.shawcable.net - 0 | 1044 | 1044 | 35 | 39 | 72 | 41 |

| eqix-ix-ch1.blizzard.com - 0 | 1044 | 1044 | 34 | 39 | 78 | 44 |

| ae1-br02-eqch2.as57976.net - 0 | 1043 | 1043 | 35 | 49 | 427 | 103 |

| be2-pe02-eqch2.as57976.net - 0 | 1044 | 1044 | 68 | 72 | 91 | 74 |

| chi-eqch2-ia-bons-04.as57976.net - 0 | 1044 | 1044 | 69 | 73 | 88 | 76 |

| 24.105.62.129 - 0 | 1044 | 1044 | 67 | 71 | 86 | 70 |

|________________________________________________|______|______|______|______|______|______|

WinMTR v0.92 GPL V2 by Appnor MSP - Fully Managed Hosting & Cloud Provider

2

There’s not much happening in these WinMTR tests. Did you experience the issue while they were running? The large spikes on the eq* links can be ignored since those hops will be slow to respond as a protection mechanism.

Is it possible you’re experiencing a hardware issue and not a connection one, since the WinMTRs aren’t showing a problem?

3

I don’t think it’s a hardware issue because it occurs on every PC in my house and only to Blizzard. I have 35 ping and no spikes for example to Riot’s League of Legends servers, and no issues to anything else. I’m not sure what hardware issue it could be that wouldn’t cause a problem otherwise or have been caught by a tech?

4

Well, the WinMTR tests aren’t showing any latency issues, so that’s why I asked about hardware. Since your local computer IP (for your internal network) isn’t showing in the tests, it could be that your modem or router are responding slowly, but we don’t have the data to figure that out.

5

Would that not have been caught by the tech that was here? We were told by the tech the ISP sent that the modem is functioning perfectly and that there is nothing at all wrong with it.

6

I’ve been told that by my ISP when tests clearly showed they were wrong, and it took 6 techs once for them to recognize the issue. I can’t speak for the techs coming to your home, I can only guess at the situation with the information we have. But I don’t have any other guesses as to what else would be causing this.

7

Would a modem problem not cause the issue in other things though? This is only happening on games through Battlenet and doesn’t occur in any other games or anything else on the internet. I’ve tried every single step and suggestion offered by tech support and the support help links. I’m failing to see how it could be the modem when it’s only happening with the connection to Blizzard.

I don’t really know what other information I can provide. The Blizzard tech support person who answered my ticket told me to post the MTR to the forums and said there’s a known issue with routing problems and Canadian ISPs. I don’t think they’re going to replace my modem when any tests they run according to them indicate there’s nothing wrong with it. And not being a tech person, all I can say for myself is that since there’s no problems with anything else it’s hard to believe there is something wrong with the modem that would cause the issues to only occur with the Battlenet games.

8

Hey LoneWolf,

Out of curiousity, whats the make and model of your modem? There are some potential configurations in the modem’s settings that have caused issues with our traffic, depending on how its setup.

Something that may help would be disabling QoS, UPnP, and SIP ALG if they’re active in the modem’s settings. Since ICMP traffic doesn’t usually get hit by these settings, it wouldn’t typically show in the MTR results, but the game traffic may be getting caught.

-Lantrasau

1 Like
9

Hitron CGNM-2250 appears to be the model of the modems Shaw supplies. All 3 of those things were enabled by default on the modem. I’ve turned them all off now. Will have to see if that makes a difference on anything.

Edit: Tested it, lag spiking still occurring. Are we certain it’s not a routing issue? I used to have 55 ping to the ord1 server a couple months ago. Ever since the lag spikes started, it’s now always between 85-120, often meaning I have a lower ping to lax1 despite Chicago being ~473 miles from my location and Los Angeles being ~2400 miles away.

10

With Shaw in Winnipeg, I was previously getting 40ms, now it’s a consistent 80ms. I wouldn’t call this a lag spike, since it’s consistent. Other Shaw users across Canada are reporting a similar issue.

community.shaw.ca/thread/42912

1 Like
11

It’s a higher average ping as well yes, but the latency spikes are also occurring for me. It’d be great if both were fixed though. The game feels frustrating and unresponsive on higher ping compared to the old ping. Coupled with latency spikes means I can’t play competitive unless I want to handicap my teammates.

12

Lonewolf,

Now that you’ve disabled those settings can you run the WinMTR again and see if we can see anything on either of the first two hops? If we still can’t see anything, do you have a phone you can use as a mobile hotspot using 4g? Do the latency spikes even out some on that network, if so?

13

Hello,

I am a network engineer for an ISP, not really a gamer, just signed up due to following this problem.

I have access to a few routers on Winnipeg, one is on Shaw, another is connected via Hurricane Electric. The HE service taking almost the same route as Shaw is fine, but the Shaw sees a pronounced latency between two of the Blizzard routers that are ostensibly in Chicago, the same two shown in the OP’s ping results. Going from ae1-br02-eqch2.as57976.net to be2-pe02-eqch2.as57976.net is a 30 ms jump which is not what I would expect going from Chicago to Chicago (I assume eqch is Equinix Chicago). This 30ms jump shows up in the OP’s ping results as well. I do not see this same latency increase between hops on the non-Shaw connection, but that is a different pair of routers (ae1-br02 and be2-pe01) and that seems to be fine. The problem looks to be within Blizzard’s network and not Shaw’s, based on the MTR results.

And yes I do see the latency spikes that Nicole mentioned above due to the protection mechanism, but those are unrelated to the delay that I mention here. That 30 ms delay is consistent and follows all future hops.

14

Hey folks, I didn’t realize we were looking at increases of only 30 ms all around until earlier today. Previously we were working on a larger issue, but it looks like the resolution of that problem changed the issue a little.

Let’s try to keep this all on one thread - can those affected send the information requested at this thread? I agree that a ~30 ms spike within chicago is odd, and I expect the increase in overall latency has something to do with changes we made to fix a worse issue we were having previously. If we can get a decent number of examples we should be able to have our GNOC team tweak things.

Locking this to redirect to the thread I just linked. Thanks for the time!