TimeoutDisconnect while connecting to: app-us.exitgamescloud

We are using PUN+ to handle unity networking. We are connected to the US region. We are getting the following error about 80% of the time on 2 of our 3 test machines. The two test machines that failed are MAC and Windows (with the third successful one being PC). We are getting close to our release date and would very much like to use PUN.

Other posts point to the firewall / router as being a possible issue. How can we verify this ? If this is the case what is a work around ?

Please assist.

TimeoutDisconnect while connecting to: app-us.exitgamescloud.com:5055. Check if the server is available.
UnityEngine.Debug:LogWarning(Object)
NetworkingPeer:OnStatusChanged(StatusCode) (at Assets/Photon Unity Networking/Plugins/PhotonNetwork/NetworkingPeer.cs:1437)
ExitGames.Client.Photon.EnetPeer:SendOutgoingCommands()
ExitGames.Client.Photon.PhotonPeer:SendOutgoingCommands()
PhotonHandler:Update() (at Assets/Photon Unity Networking/Plugins/PhotonNetwork/PhotonHandler.cs:96)

Comments

  • Did you connect and play or does this error happen while the client does it's initial connect?
    The error's message points me to "initial connect fails" and this in turn means you machine can't connect at all. It's either not getting TO the server or anything FROM the server.

    Afaik, we did not have outages on 25. May but there is also no known client side issue that will cause this with that probability.
    Please rise the debug level and send us a full log.
    Set: PhotonNetwork.logLevel = PhotonLogLevel.Full and PhotonNetwork.networkingPeer.DebugOut = DebugLevel.INFO;
  • gekido
    gekido ✭✭
    I'm also seeing this as of today.

    was working last night, but haven't been able to connect at all today.

    nslookup & tracert can find it properly, but the exported game can't connect to the photon cloud master server at all.

    This is on the initial startup call to connect to the master server.
  • gekido
    gekido ✭✭
    Here's the full output of the connection failure that I'm seeing, FYI.
  • We are also experiencing initial connection problems with the US Photon Cloud server on about 20% of attempts as of yesterday.

    We have been using the EU server in the meantime to resolve the issue for testing. It doesn't seem affected by this issue.
  • I'm getting this too!

    Using the US server, it only connects about 10% of the time. It never used to be this bad, when I used Photon a couple months ago.
    Its terrible now, can hardly ever even connect..
  • gekido
    gekido ✭✭
    I'll give the EU servers a shot to see if that improves the situation, good idea.
  • Hi all,

    thanks for your bug reports and for your feedback. We are currently investigating the issues on the US server. Sorry for the inconvenience so far!
  • Hi all,

    we see that there is still an issue with the Photon Cloud US, we are running various checks, but have not yet been able to resolve it.

    Could you do me a favor please?

    If you see this again on your machine, could you run a WinMTR trace to "app-us.exitgamescloud.com" and post the result here? WinMTR can be downloaded here: http://winmtr.net/download-winmtr/

    If we could get 3-4 traces from different users, it might really help. It would also be interesting to know your location (country or state).

    Thanks!
  • gekido
    gekido ✭✭
    WinMTR results (Vancouver, BC):
    |------------------------------------------------------------------------------------------|
    
    |                                      WinMTR statistics                                   |
    
    |                       Host              -   %  | Sent | Recv | Best | Avrg | Wrst | Last |
    
    |------------------------------------------------|------|------|------|------|------|------|
    
    |                             10.100.70.2 -    0 |   23 |   23 |    0 |    0 |    0 |    0 |
    
    |                          162.210.32.147 -    0 |   23 |   23 |    0 |    1 |   25 |    0 |
    
    |                            10.100.64.65 -    0 |   23 |   23 |    0 |    0 |   14 |    0 |
    
    |             ae1-103.van10.ip4.tinet.net -    0 |   23 |   23 |    0 |    1 |   18 |    0 |
    
    |           xe-11-0-0.sea23.ip4.tinet.net -    0 |   23 |   23 |    5 |    5 |   10 |    5 |
    
    |              as3356.sea21.ip4.tinet.net -    0 |   23 |   23 |    5 |    7 |   39 |    5 |
    
    |       ae-32-52.ebr2.seattle1.level3.net -    0 |   23 |   23 |   72 |   72 |   73 |   72 |
    
    |          ae-2-2.ebr2.denver1.level3.net -    0 |   23 |   23 |   72 |   72 |   74 |   72 |
    
    |         ae-3-3.ebr1.chicago2.level3.net -    0 |   23 |   23 |   72 |   72 |   74 |   73 |
    
    |       ae-1-100.ebr2.chicago2.level3.net -    0 |   23 |   23 |   73 |   73 |   76 |   73 |
    
    |     ae-6-6.ebr2.washington12.level3.net -    0 |   23 |   23 |   72 |   72 |   74 |   72 |
    
    |ae-203-3603.edge1.washington12.level3.net -    0 |   23 |   23 |   72 |   72 |   73 |   73 |
    
    |fiberring-b.edge1.washington12.level3.net -    0 |   23 |   23 |   78 |   78 |   79 |   78 |
    
    |                           108.59.15.145 -    0 |   23 |   23 |   77 |   77 |   78 |   77 |
    
    |                  hosted-by.leaseweb.com -    0 |   23 |   23 |   77 |   77 |   78 |   77 |
    
    |________________________________________________|______|______|______|______|______|______|
    
       WinMTR v0.92 GPL V2 by Appnor MSP - Fully Managed Hosting & Cloud Provider
    
    

    I did speak to the IT folks here and they think that it might be an ACL issue on inbound UDP traffic from the master server. We're investigating to see.

    The master server connects out via UDP:5055 - do you know what port(s) it uses for the response? Is it the same? Just trying to check out possibilities from my end.

    Thanx
  • gekido
    gekido ✭✭
    K just switched to use the EU master server and it connects fine, so definitely something wierd going going on with the US master server.
  • Thanks for the trace! That's looking quite good, no obvious issue in that trace.

    On your side? The client uses a random port for the outgoing "connection". The Master server sends the response to that port. There is no fixed port on your side. It might be an ACL issue, but beyond that, there IS an issue with the US server (we see that disconnect rates are approximately twice as high as usual), but every component seems to work fine. We'll keep on investigating (and will probably switch to another server in the next step).

    I' also d be happy if someone else could post a WinMTR trace :)
  • I've been having this issue all day...I'm in Seattle.

    Never had a problem with this until now It constantly hangs and fails to connect more often than it connects....even on a pc with cat5 directly to the router, not just wireless.


    WinMTR:
    |
    |
    | WinMTR statistics |
    | Host - % | Sent | Recv | Best | Avrg | Wrst | Last |
    |
    |
    |
    |
    |
    |
    |
    |
    | 192.168.20.1 - 0 | 33 | 33 | 0 | 0 | 2 | 1 |
    | 73.98.110.1 - 0 | 33 | 33 | 9 | 13 | 22 | 18 |
    |te-0-2-0-20-ur08.burien.wa.seattle.comcast.net - 0 | 33 | 33 | 11 | 16 | 36 | 18 |
    |ae-21-0-ar03.seattle.wa.seattle.comcast.net - 0 | 33 | 33 | 12 | 31 | 65 | 29 |
    |he-1-4-0-0-11-cr01.seattle.wa.ibone.comcast.net - 0 | 33 | 33 | 12 | 18 | 39 | 13 |
    |ix-7-2-2-0.tcore1.00S-Seattle.as6453.net - 0 | 33 | 33 | 10 | 16 | 37 | 15 |
    | if-1-0-0-3.core1.00S-Seattle.as6453.net - 0 | 33 | 33 | 11 | 16 | 38 | 20 |
    |if-8-2-3-0.tcore2.CT8-Chicago.as6453.net - 0 | 33 | 33 | 95 | 101 | 109 | 105 |
    | if-22-2.tcore1.CT8-Chicago.as6453.net - 0 | 33 | 33 | 96 | 101 | 120 | 102 |
    | No response from host - 100 | 7 | 0 | 0 | 0 | 0 | 0 |
    | if-11-2.tcore1.NYY-New-York.as6453.net - 0 | 33 | 33 | 95 | 101 | 125 | 105 |
    | if-4-16.tcore2.AEQ-Ashburn.as6453.net - 0 | 33 | 33 | 96 | 101 | 124 | 97 |
    | No response from host - 100 | 7 | 0 | 0 | 0 | 0 | 0 |
    | po3.cs2.wdc1.leaseweb.net - 0 | 33 | 33 | 96 | 101 | 108 | 104 |
    | hosted-by.leaseweb.com - 0 | 33 | 33 | 97 | 101 | 122 | 101 |
    |________________________________________________|______|______|______|______|______|______|
    WinMTR v0.92 GPL V2 by Appnor MSP - Fully Managed Hosting & Cloud Provider

    8
  • gekido
    gekido ✭✭
    After I switched over to the EU server, we had about 15 people in the office connecting & playing fine. With the US server, none of us could connect for some reason over those 2 days.

    From my home machine, the US server has seemed fine throughout.
  • Hello everyone,

    thanks for your help & your input so far.

    Today, we have replaced the hardware of the US Master server. We are still having a close look at performance data - unfortunately, it does not seem to have improved as much as I hoped. :( We'll keep on investigating.

    Can you let me know if this is still an issue for you?
  • I'm still experiencing the issue. First attempt on US failed when I tried it just now. Haven't had a failure in nearly three days straight testing on EU server so it seems likely related.

    I'll post some more information tomorrow when I get some more time to spend on this!

    EDIT: Can anyone confirm that the issue is resolved? I can't replicate it any more (on 6/2) so it seems resolved for me! I'm going to start my daily testing on the US server and see if it happens again.
  • We have made another hardware / network update today, 6/2.

    I'm glad to hear that it's working better for you now. As far as we see, there are signifcantly fewer disconnects now.

    @all: please let me know if it's okay for you as well now, or if you still have trouble.

    Thanks a lot for your valuable feedback and your patience, anyways!