Last post Dec 15, 2017 07:27 PM by Ivan P
Dec 14, 2017 11:59 AM|Ivan P|LINK
I have docker container with asp.net core app webapi including signalr -> everything is working fine when I am running it on my local machine, events are dispatched with no problems and clients are receiving it. However problem starts when I deploy this
container to amazon behind loadbalancer. Now when even trying to connect to the hub i get following exception stack
at System.Runtime.CompilerServices.TaskAwaiter.ThrowForNonSuccess(Task task)
at System.Runtime.CompilerServices.TaskAwaiter.HandleNonSuccessAndDebuggerNotification(Task task)
The operation identifier is not valid. (Exception from HRESULT: 0x800710DD)
any advice? Should I include Redis? I would understand if this will happen when trying to dispatch messages outside of hub, but why its on connect? Thanks
Dec 15, 2017 02:47 AM|Fei Han - MSFT|LINK
Hi Ivan P,
According to your description, your project work as expected on your local development environment, but not work amazon cloud environment. Perhaps you do not do some configurations or settings on amazon cloud environment, and that cause the issue, you can
get help in amazon cloud services forums or contact amazon supporter to know if you need to do specific settings on amazon cloud environment.
Besides, based on your error message, the issue is related to WebSockets. And as we know, SignalR supports these four transports (WebSocket, Server Sent Events, Forever Frame and long polling), for detecting if it is transports related issue, you can try
to explicitly specify a transport while you create and start a connection in your code. You can also
enable SignalR Tracing to get more helpful diagnostic information.
Dec 15, 2017 07:27 PM|Ivan P|LINK
Hi There, actually we used what you suggested and find out our load balancer is not publishing TCP ports, just HTTP, so publishing TCP fixed the issue.