[Oceanic] High Latency or Connecting to Wrong Servers

Technical Support
22

I think it’d be even better if there was a max ping option like in CSGO.

I would rather wait a bit longer than get put into a non-OCE server.

23

I do have more than a solid understanding of how it works. Please observe the forum rules. Your comment does not offer any insight to the OP, and its only purpose is to make insults.

1 Like
24

Because one of the peers probably has an error in their routing table – at least that has been one of the reasons this has happened several times over the last year. The other reason is overloaded networks not being able to handle traffic and pushing it off to other routes, which can also end up pushing you to a different Overwatch server.

1 Like
25

I definitely agree that can happen and obviously has before. I’m not sure most people realise how complicated the internet really is. Honestly I don’t and I’ve done network admin.

The current situation seems different in that there is consistent access to the SYD server in the practice range but not other modes. If it was a routing issue, my assumption would be that the practice range should follow the other modes in server location choosing.

When I’m home I’ll do some tests assuming my games are placed on the wrong servers tonight. I also have 2 ISP’s that use different peering networks so can test between them.

26

Sounds good, that info will likely help the staff pinpoint the issue. If it’s widespread enough, they sometimes reach out to ISPs to help them correct it.

1 Like
27

ADDED - disregard the winMTR part, was using v0.8 not v0.92

I’ve played three games (MH), so far on the right server (SYD2) but can’t get winMTR to work on that server IP or the IP Caterpepi provided.

I can use winMTR to find 8.8.8.8 for instance and it works, but fails for the OW servers. To continue that I can tracert without a problem. I can get to 37.244.40.21 (hop 15/16) before tracert goes cold.

Interestingly both ISP’s are now using the same peer network of Telstra to get from Brisbane to Sydney.

28

Heya Caterpepi,
I can’t fault it tonight in either QP or MH.
In future I’ll make sure I have winMTR info when posting :slight_smile:

Personally I really don’t like playing any games particularly in LAX, gameplay is quite different with such a different ping to normal. I would rather not be put into a game on the LAX or even PSE1 server even if it means no game, perhaps a disable server list or similar.

Nothing worse than entering a comp game when it puts you in a server half way across the world.

Anyway, at least for myself and the people in the same games as me (dozens), we were playing on PC, while normally on SYD2 (didn’t know there was a SYD1), we were put on LAX1 and PSE1.

Thanks for looking into it for us, it looks like it’s resolved for now :slight_smile:

31

The concern I have with this is that, as I’ve said before, during the time I have been placed on a server that’s not in my region (either LAX or PSE instead of SYD), I have Smokeping running and I have been pinging multiple SYD servers from games I’ve been in. During the affected period I don’t get any form of “bad routing” or high latency to any of these servers. The pings never go above 50ms.

Additionally, my comment had absolutely no intent of insulting anyone. I take offense to you thinking that. My comment was purely due to the fact that the underlying cause of the networking issues here is most likely quite complex, and making assumptions at this kind of stuff is a bad idea, because people who have a very good understanding of this (and have worked with networks for almost 10 years) can get frustrated at comments that are made which make no sense and aren’t possible. It’s nothing personal in any way.

32

Could we get an MTR test to BOTH the in-game IP shown in the netgraph as well as the one Caterpepi mentioned? This will help us compare the data. Even if there are “destination unreachable” or similar the test data may still help us identify the issue.

Thanks!

33

A post was split to a new topic: Azusu - Slow Download

34

Device: PC
Regional Server: LAX1

WinMTR to 24.105.13.122:

WinMTR statistics

Host % Sent Recv Best Avrg Wrst Last
192.168.1.254 0 11 11 0 0 1 0
lo0.bras2.syd2.on.ii.net 0 11 11 19 24 55 19
be2.cr2.syd2.on.ii.net 0 11 11 19 25 62 24
be51.cr2.syd7.on.ii.net 0 11 11 199 203 219 199
te0-2-0-3.br2.sjc2.on.ii.net 0 11 11 174 178 203 177
eqix-sv8.attblizzard.com 0 11 11 177 192 229 229
ae1-br02-eqsv5.as57976.net 0 11 11 175 185 236 178
xe-0-0-1-1-br02-eqla1.as57976.net 0 11 11 207 218 255 211
No response from host 100 3 0 0 0 0 0
No response from host 100 3 0 0 0 0 0
No response from host 100 3 0 0 0 0 0
No response from host 100 3 0 0 0 0 0
No response from host 100 3 0 0 0 0 0
No response from host 100 3 0 0 0 0 0
No response from host 100 3 0 0 0 0 0
No response from host 100 3 0 0 0 0 0
No response from host 100 3 0 0 0 0 0
No response from host 100 3 0 0 0 0 0
No response from host 100 3 0 0 0 0 0
No response from host 100 3 0 0 0 0 0
No response from host 100 3 0 0 0 0 0
No response from host 100 3 0 0 0 0 0
No response from host 100 3 0 0 0 0 0
No response from host 100 3 0 0 0 0 0
No response from host 100 3 0 0 0 0 0
No response from host 100 3 0 0 0 0 0
No response from host 100 3 0 0 0 0 0
No response from host 100 3 0 0 0 0 0
No response from host 100 3 0 0 0 0 0
No response from host 100 3 0 0 0 0 0

WinMTR to 37.244.40.1:

WinMTR statistics

Host % Sent Recv Best Avrg Wrst Last
192.168.1.254 0 10 10 0 0 1 0
lo0.bras2.syd2.on.ii.net 0 10 10 18 24 33 21
be2.cr2.syd2.on.ii.net 0 10 10 21 27 34 22
be51.cr2.syd7.on.ii.net 0 10 10 20 25 33 23
ae6.cr2.syd4.on.ii.net 0 10 10 19 23 30 25
57976.syd.equinix.com 0 10 10 25 34 82 27
et-0-0-49-br02-eqsy4.as57976.net 0 10 10 22 27 36 28
et-0-0-0-pe04-eqsy4.as57976.net 0 10 10 21 28 48 24
137.221.66.139 0 10 10 19 25 34 27
37.244.40.1 0 10 10 21 26 35 23
35

RE: connecting to wrong servers - Its been nearly a month, and we’ve had no hint of any form of actual assistance from your support team.

You guys are closing other threads about this issue, moving everyone to this thread, but offer no help?

This is unlikley to be an ISP issue, when the fix most of your players are doing is just to quit a match when they see the bad server cos they know they’ll be put in a good one within a few match quits.
There are hundreds of players over the past month who I know have complained about it, so there are probably thousands of others who are affected.

In fact, it’s got so wide spread that its basically become a standard to play “server roulette” and join matches and players see the 200 to 300 ping people just “F” or “RIP Aussies” or whatever in match chat, thats even if they stay, most just leave when they see they’ve been put on a bad server - because you know the game will be unplayable. :persevere:

You cant just keep closing threads and sweeping this under the rug like its not an issue :roll_eyes:

Your Australian player base have got to the point of trying their own work around of quitting matches when they see they’ve been put on a trash server, until they get put on a normal one. Literally most games where its 200-300 ping you see 6-12 people leave within a few minutes because they cant play on those server.
My issue with the quitting matches workaround is that I don’t want to get penalized for having to play server roulette quitting due to your lack of assistance with this problem.

This is a widespread ongoing issue that the community are getting pretty fed up with, especially with winter wonderland coming - so… can you either remove the leaving penalty so that the Aussies can quit matches until they get placed on normal servers so that the game is playable (like I said, I don’t want to get penalized for leaving a server that’s unplayable), or you can fix the issue?

1 Like
36

It happened a few days ago for me and I recorded winMTR’s to both a known SYD2 server and the server it put me on at the same time as playing a game on the PSE1 server. Since it happened multiple times today I figured it would be good to post the winMTR results. Looking at the winMTR I see no reason why the PSE1 server was chosen over the SYD2 server.
PSE1
|------------------------------------------------------------------------------------------|
| WinMTR statistics |

Host - % Sent Recv Best Avrg Wrst Last
192.168.0.250 - 49 229 119 0 0 1 0
lo40.lns04.brispri.qld.vocus.network - 0 668 668 1 1 3 1
ae2-14.per01.brispri.qld.vocus.network - 0 668 668 1 2 24 1
ae6.per03.brispri.qld.m2core - 0 668 668 1 6 20 2
BE-1.bdr01.bne03.qld.VOCUS - 0 668 668 1 1 7 1
e-1-24.cer01.bne01.qld.VOCUS - 0 668 668 106 106 109 108
BE-109.cor01.syd11.nsw.VOCUS - 0 668 668 106 106 118 107
BE-102.cor01.per02.wa.VOCUS - 0 668 668 108 108 110 109
BE-201.bdr01.sin01.sin.VOCUS - 0 668 668 108 109 117 108
16509.sgw.equinix - 0 668 668 106 107 122 107
________________________________________________ ______ ______ ______ ______ ______ ______

SYD2
|------------------------------------------------------------------------------------------|
| WinMTR statistics |

Host - % Sent Recv Best Avrg Wrst Last
192.168.0.250 - 4 641 619 0 0 1 0
lo40.lns04.brispri.qld.vocus.network - 0 725 725 1 1 3 2
ae2-14.per01.brispri.qld.vocus.network - 0 725 725 1 2 28 2
ae6.per03.brispri.qld.m2core.net.au - 0 725 725 1 8 23 2
BE-1.bdr01.bne03.qld.VOCUS.net.au - 0 725 725 1 1 13 2
Bundle-Ether15.cha-edge901.brisbane.telstra - 0 725 725 14 15 19 15
bundle-ether6.cha-core4.brisbane.telstra - 0 725 725 15 16 34 16
bundle-ether20.ken-core10.sydney.telstra - 0 725 725 17 18 24 18
bundle-ether1.oxf-gw10.sydney.telstra - 0 725 725 18 19 23 19
bundle-ether1.sydo-core04.sydney.reach - 0 725 725 18 19 25 18
i-91.sydo10.telstraglobal - 0 725 725 15 15 50 16
unknown.telstraglobal - 1 722 721 26 29 86 27
et-0-0-48-br01-eqsy4.as57976 - 0 725 725 18 22 147 19
et-0-0-0-pe01-eqsy4.as57976 - 0 725 725 17 20 122 19
137.221.66.133 - 0 725 725 18 18 23 19
37.244.40.41 - 0 725 725 18 18 21 18
________________________________________________ ______ ______ ______ ______ ______ ______

ADDED: My own router (192.168.0.250) is set to reject too many ICMP at a time, hence the data loss there

37

Thanks for the continued reports! We are tracking the issue on our end, but the data request for the WinMTR test help greatly with the reports. This issue is also impacting our other games and for the SEA region as well. Here’s one of the World of WarCraft threads that we’re currently tracking along side this one here.

There was a cable cut on the SEA-ME-WE3 a few months back that speculated to cause higher latency and a longer connection route than normal for Telstra backbone connections to work around the cable cut. There’s still no ETA on when this cable may be fully fixed. It’s my personal opinion, but it seems likely to what’s been causing the issue for higher latency, latency spikes, and connecting to the wrong servers.

The following information can help us determine if that’s the specific cause to compare the data with:

  • Device (PC, PSN, Xbox, Switch):
  • Regional Server (Example: LAX1, ORD1, SYD1) - Press CTRL+SHIFT+N to open up the Network Graph, this will be next to the IP address at the top left corner of the graph.
  • WinMTR test while the issue occurs:
  • Let’s also grab a WinMTR test for the Oceanic realm to compare what’s happening. Use the IP Address: 37.244.40.1

In the mean time, either seeing if another backbone peering network or using a VPN to reroute the connection should help lower the latency and hopefully, have everyone back on the appropriate home servers. I’ll dig through some more data on our end to see what’s going on with the servers concurrency as well as these connection backbone issues. Thanks again for your reports!

38

Thanks for the reply Caterpepi.
Some more information for you.
Device : PC
Also I’ve asked my ISP (FibreMax a reseller of Vocus) why they are peering with Telstra when they have their own network, they said they would look into it as that shouldn’t be the case.
Also Optus (I have a 2nd ISP) and TPG (according to PewPew on these forums) are using the same Telstra backbone at least from QLD (Brisbane). So it sounds like all/most traffic from QLD to Overwatch uses the same Telstra backbone.

39

Im with Aussie Broadband and i play on PC have been having similar issues
every so often i will be put into PSE1 when my usual local server is SYD2
Below is results from a good connection what i usually get will post results from bad connection when it pops up again.

WinMTR Statistics

WinMTR statistics

Host % Sent Recv Best Avrg Wrst Last
10.1.1.1 0 267 267 0 0 1 0
loop11918380.bng1.otw.bne.aussiebb.net 0 267 267 30 31 50 31
be1.core2.nextdc-s1.syd.aussiebb.net 0 267 267 41 42 49 43
hundredgige0-0-0-3.core1.otw.bne.aussiebb.net 0 267 267 41 41 44 41
be2.core1.nextdc-b1.bne.aussiebb.net 0 267 267 41 41 44 42
hundredgige0-0-0-18.core1.nextdc-s1.syd.aussiebb.net 0 267 267 41 41 43 41
40

So there’s 2 of us on the same connection, in the same room, and usually we get around ~40ms pings. But every few games, one of us will get a ~200ms ping, whilst the other still has the usual 40ms.

What’s up with that?

Usually connected to syd2.

(Internode, Tasmania)

1 Like
41

Results from getting routed to PSE1 13.229.250.88

WinMTR Statistics

WinMTR statistics

Host % Sent Recv Best Avrg Wrst Last
10.1.1.1 0 483 483 0 0 2 0
loop11918380.bng1.otw.bne.aussiebb.net 0 483 483 30 31 56 34
be3.core1.sg1-equinix.sin.aussiebb.net 0 483 483 135 136 201 136
hundredgige0-0-0-3.core1.otw.bne.aussiebb.net 0 483 483 135 135 210 135
be2.core1.nextdc-b1.bne.aussiebb.net 0 483 483 135 135 204 136
hundredgige0-0-0-18.core1.nextdc-s1.syd.aussiebb.net 0 483 483 135 135 221 136
hundredgige0-0-0-4.core1.nextdc-m1.mel.aussiebb.net 0 483 483 135 136 215 136
be3.core2.ia-dcb.portmel.aussiebb.net 0 483 483 135 135 201 135
be2.core1.ia-dcb.portmel.aussiebb.net 0 483 483 135 135 202 135
hundredgige0-0-0-8.core1.vdc01.per.aussiebb.net 0 483 483 135 135 192 136
be1.core2.vdc01.per.aussiebb.net 0 483 483 135 135 170 135
be2.core2.nextdc-p1.per.aussiebb.net 0 483 483 135 136 220 135
42

Bumping this because its still an ongoing issue its not just me multiple people from Australia in my games keep getting routed to PSE1 server.

43

Hey there Julian,

I ran a report on the ISP you’re using and did not see any major spikes in latency. Basically we just need to root out why a single user connection on the ISP is being treated differently than other users.

It might be best to start with an extended power cycle on the networking devices by leaving them unplugged for ~30 minutes then plugging them back in.

Next we’d want to check on software on the computer itself. Some software has “Network Prioritization,” “Quality of Service,” or similar features. Here are a few examples:

  • Killer Control Center (stream detect option)
  • Alienware Control Center
  • SmartByte
  • HP Omen Connection Optimization

Keep in mind these are just examples. There are a huge number of software out there that will do this.

Another thing to try is contacting the ISP and asking about firmware updates and device assistance (Quality of Service settings, factory reset, etc.)

Hope the info helps!