connection to blizzard's servers blocked, and also has the wrong routing.

[msantore22]

hi, i have a problem from edmonton, my connection to blizzard's central ip location is being blocked on a router somewhere in the middle. and also this is the incorrect routing, it is supposed to take. blizzard's central ip is located in chicago.

here is what my tracert shows:

Tracing route to 24.105.62.129 over a maximum of 30 hops

1 <1 ms <1 ms <1 ms OpenWrt.lan [192.168.1.1]
2 9 ms 8 ms 7 ms 96.52.128.1
3 9 ms 8 ms 8 ms rc2ar-be117-1.ed.shawcable.net [64.59.186.161]
4 9 ms 8 ms 9 ms rc2we-be7.ed.shawcable.net [66.163.70.129]
5 12 ms 11 ms 11 ms rc3no-be6.cg.shawcable.net [66.163.64.69]
6 27 ms 27 ms 28 ms rc2wt-be100.wa.shawcable.net [66.163.75.233]
7 55 ms 51 ms 30 ms rc6wt-tge0-10-0-11.wa.shawcable.net [66.163.68.70]
8 57 ms 57 ms 58 ms ae1-br02-eqse2.as57976.net [137.221.73.35]
9 * * * Request timed out.
10 58 ms 57 ms 57 ms et-0-0-1-pe04-eqch2.as57976.net [137.221.69.79]
11 * * * Request timed out.
12 * * * Request timed out.
13 * * * Request timed out.
14 * * * Request timed out.
15 * * * Request timed out.
16 * * * Request timed out.
17 * * * Request timed out.
18 * * * Request timed out.
19 * * * Request timed out.
20 * * * Request timed out.
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.



why would it be blocked like that when blizzard advertises these servers as their pingable servers?


also the connection to that ip is incorrect, if its in chicago, should it not go through a different routing source?

i have a VPN in chicago and when i tracert that vpn, this is the correct routing:

Tracing route to 107.191.51.12.vultr.com [107.191.51.12]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms OpenWrt.lan [192.168.1.1]
2 8 ms 8 ms 8 ms 96.52.128.1
3 10 ms 10 ms 8 ms rc2ar-be117-1.ed.shawcable.net [64.59.186.161]
4 23 ms 24 ms 24 ms rc3sc-be1.wp.shawcable.net [66.163.70.77]
5 39 ms 38 ms 39 ms rc4ec-be25-1.il.shawcable.net [66.163.75.134]
6 55 ms 44 ms 40 ms 20473.chi.equinix.com [208.115.137.46]
7 * * * Request timed out.
8 * * * Request timed out.
9 * * * Request timed out.
10 * * * Request timed out.
11 40 ms 40 ms 39 ms 107.191.51.12.vultr.com [107.191.51.12]

Trace complete.

when i use my vpn to play on blizzard servers, i get significantly lower ping. about 30 ms lower.

that means im not being routed correctly.

the problem i am experiencing has been happening for some months now. just figured i'd ask about it now. thanks

[msantore22]

the post wouldnt let me edit to say that because my connection to blizzard's Central ip has the wrong routing and  its being blocked like that, means that the game is connecting me to Los Angeles, instead of chicago, which is almost unplayable speed for me with my latency in game reaching about 80 ms. i can actually feel the latency, and it doesnt feel as good as when i use my vpn to connect to blizzard's servers.

[mdk]

@msantore22 -- your trace:

3 8 ms rc2ar-be117-1.ed.shawcable.net [64.59.186.161]
4 9 ms rc2we-be7.ed.shawcable.net [66.163.70.129]
5 11 ms rc3no-be6.cg.shawcable.net [66.163.64.69]
6 28 ms rc2wt-be100.wa.shawcable.net [66.163.75.233]
7 30 ms rc6wt-tge0-10-0-11.wa.shawcable.net [66.163.68.70]
8 58 ms ae1-br02-eqse2.as57976.net [137.221.73.35]
9 * * * Request timed out.
10 57 ms et-0-0-1-pe04-eqch2.as57976.net [137.221.69.79]

seems to cross-over from Shaw's network to another network at Seattle -- "eqse2", and then it stops at a router in Chicago -- "eqch2" .

My trace is similar:

  5    11 ms  rc1wt-be40.wa.shawcable.net [66.163.68.18]
  6    18 ms  rc6wt-tge0-10-0-10.wa.shawcable.net [66.163.68.66]
  7    65 ms  ae1-br02-eqse2.as57976.net [137.221.73.35]
  8     *        *        *     Request timed out.
  9    62 ms  et-0-0-1-pe03-eqch2.as57976.net [137.221.69.57]

10    61 ms  24.105.62.129

Both traces go through 137.221.69.79 or its "neighbour" 137.221.69.57, but my trace reaches the target IP-address.

> why would it be blocked like that when blizzard advertises these servers as their pingable servers?

Is it "blocked" due to hardware failure somewhere between 137.221.69.xxx to the Blizzard server? Compare to a blizzard that causes an avalanche of snow that physically blocks a road -- no traffic (cars or network packets) can get through.

Does Blizzard have a web-page that reports its "current network status", including outages?

> the connection to that ip is incorrect, if its in chicago, should it not go through a different routing source?

The Internet is implemented as a "mesh", just like the US highway system. There are multiple highways from the Seattle (Washington) area to Chicago. How does one define "correct" ? Most scenic? Fewest mountain-passes subject to heavy snowfall? More high-speed interstate highways than country back-roads?  The routers on the Internet can dynamically adjust, when the connection between two routers is not working. There might be a "less-desirable" route -> Edmonton -> Winnipeg -> USA -> Chicago that will be used only when the "currently-best" route (Edmonton -> Vancouver -> Seattle -> Chicago) is not functioning.

> that means im not being routed correctly

The routing-tables inside each Internet router define "preferred", and make the inference that the preferred route is the "best" route. Blizzard's routers need to broadcast their "best" routing to their servers, and the "AS57976" network also needs to broadcast their "best" routing to Blizzard's routers.

> 39 ms 107.191.51.12.vultr.com [107.191.51.12]  ---- Trace complete.

That is the PING-times to the VPN-server. There are additional "hops" from that server over the Internet to Blizzard's servers. How much additional "delay" is added when traversing those network segments?

> the problem i am experiencing has been happening for some months now

Repeated temporarily-broken connections from the "AS57976" network to Blizzard is a problem, but since the problem is "outside" of Shaw's network, there is nothing that Shaw can do about it.

Blizzard could upgrade to "co-located" servers. Microsoft has connected their servers directly to the Internet "backbone", rather than keeping their servers inside their corporate offices in Redmond. Such servers reduce the "ping" times, because there will be fewer "routers" between their customers and their servers.

Example:

$ tracert www.microsoft.com

Tracing route to e13678.dscb.akamaiedge.net [23.11.242.9]

  4    12 ms  rc1bb-be20.vc.shawcable.net [66.163.75.245]
  5   18 ms  rc3ar-tge0-6-0-16.ed.shawcable.net [66.163.76.42]
  6     9 ms  a23-11-242-9.deploy.static.akamaitechnologies.com [23.11.242.9]

Trace complete.

The "host-name" is www.microsoft.com, but the co-located server is in Edmonton.

If Blizzard were to contract with Akamai Technologies to "co-locate" their servers, I'm sure that you would be very pleased.  🙂

[msantore22]

hi thanks for your reply.

its highly doubtful blizzard would co-locate like that. they have a large amount of data, and constantly need to fix things with their servers. so its all done in their corporate offices with connection very close to whatever datacenter is near there.

the vpn, i use gets 1ms - to 3ms of ping to blizzard's servers. since i figured out they are most likely using the same provider.

while connected to the vpn, and if you factor in that vpn software have latency themselves as overhead. i get 50ms to blizzard while i play overwatch. the routing to the vpn is the path that i want shaw to take with the blizzard ip.(going through winnipeg)

2 points i would like to point out:

1. i shouldnt have to use the vpn to get the speed of latency that i want.

2. a 30ms of latency difference between 2 paths is rather large when it comes to gaming specifically when youre trying to react fast to situations in the game.

the lowest ping route is always better for gaming. and in this situation happens to be as you described "less desirable"  but to me, and other gamers this would be more desirable than most other situations.

but im wondering why the new routing to vancouver became more desirable? i used to get 45 ms to overwatch atleast 1 to 2 years ago. ping was never a problem for me. i didnt have to use a vpn back then. but now i do?


speaking of avalanches, i remember in 2020 there was a avalanche in b.c., that entire week i had great wonderful ping. until they fixed whatever issue the avalanche caused.....i remember hearing from a third party provider it knocked out a shaw cable line.

is there someway that shaw can force my connection to go through winnipeg instead of vancouver, without using a vpn?

[msantore22]

there have been many posts about blizzards problems and shaw. that i believe something got misconstrued somewhere down the line.

please refer to this post here:
https://support.shaw.ca/t5/internet-discussions/ping-to-blizzard-servers-has-doubled/td-p/3486/page/...

on page 2 the user at the very bottom posted a picture of his "fixed connection", where he gets 25ms to blizzard servers. 25ms from winnipeg to blizzard sounds great, now how do i get routed from edmonton to winnipeg?

[mdk]

@msantore22 -- its highly doubtful blizzard would co-locate like that. they have a large amount of data, and constantly need to fix things with their servers. so its all done in their corporate offices with connection very close to whatever datacenter is near there.

I disagree. By co-locating their servers, the advantages include:

* servers that are just one "hop" away from the "backbone" of the Internet;

* better performance, due to more bandwidth available over that one "hop"

* professional physical security, backup power, 24/7 monitoring, backup of files/folders, and maybe "hot-swap" capability

* reliable & fast high-end servers

Given the speed of the Internet between their corporate offices and their co-located servers, their "constant" updates will be just as fast as when their servers are in a nearby data-centre.

> the lowest ping route is always better for gaming. and in this situation happens to be as you described "less desirable"  but to me, and other gamers this would be more desirable than most other situations.

You cannot please all the people all the time.

If Blizzard were to use co-located servers, they could "load-balance", and use servers in multiple locations, e.g., Edmonton & Seattle & Anaheim & Chicago & Tampa, for "best" latency for customers anywhere north of Mexico. Of course, you, in/near Edmonton, would "game" against others who also connect to the co-located server in Edmonton.

But, as a first step, Blizzard should concentrate on one co-located server, in Chicago, so that you could "game" against any Blizzard user, not just those in/near Edmonton.

> but I am wondering why the new routing to vancouver became more desirable?

All the routers on the Internet use "routing tables", because the Internet is a "mesh", meaning that there are multiple routes. For example, when driving your automobile from Edmonton to Camrose, which is the "correct" route:

East on #2, then south on #630
South-east on #216, then east on #14, then south on #21
South-east on #216, then east on #14, then south on #833
South on #2, south on #2A, then east on #13

and which is the "best" route?

The routing tables in each router factor-in the speed of each "hop" on the Internet, just like the speed-limit signs on those roads, and the number of lanes on each road, help you choose the "best" route.

Blizzard provides routing-tables to reach their range of IP-addresses, and Shaw provides routing-tables that "prefer" the Edmonton -> Vancouver -> Seattle segments of their network.

> speaking of avalanches, i remember in 2020 there was a avalanche in b.c., that entire week i had great wonderful ping. until they fixed whatever issue the avalanche caused.....i remember hearing from a third party provider it knocked out a shaw cable line.

Presumably, Shaw's router in Edmonton could not connect to Shaw's router in Vancouver, except by using an alternate path.

That "redundancy" is inherit in the design of the Internet -- it was originally designed (by ARPA - part of the USA Department of Defense) to withstand a nuclear bomb in a specific target city, by planning for alternate, but "less desirable" routes.

To show the "mesh" of the Internet, compare these two "partial" trace-routes.
The first is to the University of North Dakota, through Seattle,
and the second is to the University of Chicago, through Calgary & Winnipeg & Montreal

Tracing route to www.und.edu [134.129.183.70]

  5    12 ms  rc1wt-be40.wa.shawcable.net [66.163.68.18]
  6    12 ms  100ge14-2.core1.sea1.he.net [206.81.80.40]
  7    35 ms  dcn-llc.e0-53.switch1.sea2.he.net [66.160.133.166]
  8    37 ms  66.97.224.102
  9    35 ms  bis-internet-119-l3-gateway.nd.gov [165.234.119.1]
 10    65 ms  gfmfh-i2.itd.nd.gov [165.234.98.9]

Tracing route to wsee2.elb.uchicago.edu [34.225.113.202]

  4    21 ms  rc1bb-be20.vc.shawcable.net [66.163.75.245]
  5    13 ms  rc1st-be25.vc.shawcable.net [66.163.69.198]
  6    25 ms  rc3no-be11-1.cg.shawcable.net [66.163.72.69]
  7    46 ms  rc2nr-be110-1.wp.shawcable.net [66.163.76.58]
  8    67 ms  rc3fs-be25.mt.shawcable.net [66.163.76.22]

So, some of Shaw's "traffic" from Edmonton goes "west", and some goes "east". Different routing-tables contain different information.  Try tracing those two sites from your computer.

[mdk]

@msantore22 -- please check another similar thread on this discussion forum.

The post you referenced on Sunday uses the routing-tables in Winnipeg (not from your location in Edmonton) to connect to Blizzard. Those routing tables "prefer" some other route into the USA -- instead of Winnipeg -> Vancouver -> Seattle -> Chicago.