Ping to Blizzard servers has doubled

[tahlee]

We have several people in our guild that live in Calgary as well still experiencing the same issues as the people above.  We all normally had 35-40ms and everyones ping has doubled or more to 80-120 ms.  This is a tracert I just ran that shows the same issues the people have above.

Tracing route to 24.105.62.129 over a maximum of 30 hops

  1     1 ms     1 ms     1 ms  10.0.0.1
  2    10 ms    11 ms     8 ms  184.64.0.1
  3    12 ms    10 ms    10 ms  64.59.134.201
  4    35 ms    32 ms    35 ms  rc1wt-be82.wa.shawcable.net [66.163.76.9]
  5    27 ms    27 ms    27 ms  rc6wt-tge0-10-0-10.wa.shawcable.net [66.163.68.66]
  6    77 ms    76 ms    85 ms  ae1-br01-eqse2.as57976.net [137.221.73.33]
  7    59 ms    72 ms    62 ms  xe-0-0-1-1-br02-eqch2.as57976.net [137.221.65.42]
  8    82 ms    79 ms    81 ms  be2-pe02-eqch2.as57976.net [137.221.69.73]
  9    78 ms    79 ms    83 ms  chi-eqch2-ia-bons-04.as57976.net [137.221.66.15]
 10    81 ms    79 ms    78 ms  24.105.62.129

Trace complete.

C:\Windows\system32>

Never had issues for the past few years until now.  I really hope this can be looked into further and fixed.  

[mjducharme]

Hello,

Based on your tracert the problem looks to be between ae1-br02-eqch2.as57976.net and be2-pe02-eqch2.as57976.net, where you can see the latency increases from around 33ms to 68ms. Those two routers are inside Blizzard's network.

Here is my traceroute from a non-Shaw connection that takes a very similar route for comparison:

Tracing route to 24.105.62.129 over a maximum of 30 hops

  1    <1 ms    <1 ms    <1 ms  redacted
  2     1 ms     1 ms     1 ms  redacted
  3     1 ms    <1 ms    <1 ms  redacted
  4     1 ms     1 ms     1 ms  redacted.core1.ywg1.he.net [redacted]
  5     9 ms     9 ms    12 ms  100ge14-1.core1.msp1.he.net [184.105.64.85]
  6    17 ms    17 ms    17 ms  100ge13-1.core2.chi1.he.net [184.105.223.177]
  7    17 ms    17 ms    18 ms  eqix-ix-ch1.blizzard.com [208.115.136.85]
  8    17 ms    17 ms    19 ms  ae1-br02-eqch2.as57976.net [137.221.69.35]
  9    18 ms    17 ms    17 ms  be2-pe01-eqch2.as57976.net [137.221.69.71]
 10    19 ms    20 ms    18 ms  chi-eqch2-ia-bons-02.as57976.net [137.221.66.11]
 11    20 ms    19 ms    19 ms  24.105.62.129

While both the Shaw traceroute and mine (going through he.net service) have about the same latency to ae1-br02, Shaw's traffic is getting sent to be2-pe02 as the next hop after that, while HE.net's is getting sent to be2-pe01 instead. Based on your traceroute screenshot, it seems the latency issue is between ae1-br02 and be2-pe01 at Blizzard (Blizzard is as57976.net). The latency between ae1-br02 and be2-pe01 is fine, as you can see in my traceroute.

I think Blizzard needs to figure out the cause for the latency between those two routers. There should not be a 33 ms latency between one Blizzard router in Chicago and another Blizzard router in Chicago. They have something messed up.

[#3]

I play Blizzard's overwatch game regularly. i used to get 40 ping even on that game from edmonton, but recently its jumped to 60 to 65 ping.   i know very well that it is a routing issue.

what was strange is that shortly after the recent patch, my ping went back down to 40 for a brief day. then it went back up to 60 with no hope of going down.

the only thing i can do to get proper ping is to connect to my VPN server in Chicago from there i get around 45-50 ping to overwatch. on the bonus side it includes ddos protection.

but how can we fix this without a vpn?

[machnee]

Same issue from Winnipeg, getting 80ms or higher when it was previously ~40ms

[clockwise]

Thank you for providing the comparison, this is helpful.

shaw-tony‌, let’s engage Blizzard to see what’s happening here.

[mjducharme]

Here is my own test from a Shaw Winnipeg connection, which shows the latency as being caused within Blizzard's network as I explain above:

It looks like traffic that goes from ae1-br02 to be2-pe01 is fine (which is why my HE.net connection that takes that path does not have this issue), but the traffic that goes from ae1-br02 to be2-pe02 experiences a tremendous increase in latency, as does the traffic from ae1-br01 to be1-pe02 in my screenshot above. All of those routers are in Chicago and in Blizzard's network so again I think something must be messed up on their end because 30 ms between two Blizzard routers in Chicago should not be normal.

[shaw-tony]

tahlee, pharrow, mjducharme thank you for the details. We will work with Blizzard to find out what is happening and work on a resolution. We appreciate your patience. Keep us posted on any changes.

[clockwise]

drkcrix‌. When you have a second, can you run your trace again? I just had a conversation with Blizzard and they made some changes.

[clockwise]

b.bentley1‌ can you run your trace again as well?

[mjducharme]

Shaw Winnipeg latency to the blizzard IP has decreased substantially, now only 25ms. That is a huge improvement vs. the 58ms before:

Compare with my screenshot from three posts up. I would say they fixed it, the question is whether it is fixed for everybody.