Caused by: Bad CONNECT

[MAXBAF1]

Hi! I'm using your library in an android app.

Since I need to connect using a token, I use KtorWebSocketClient, and then I create a session using the .stomp() method. The connectToWebSocket() method works well, the connection really happens, it returns 101, but as soon as I try to switch to stomp, an error occurs in the sendCellToWebSocket() method Caused by: org.hildan.krossbow.stomp.StompErrorFrameReceived: Bad CONNECT.

Do you know what the problem might be?

implementation("org.hildan.krossbow:krossbow-websocket-ktor:7.1.0")
implementation("org.hildan.krossbow:krossbow-stomp-kxserialization-json:7.1.0")

I am attaching screenshots from the client and logs from the server:

Backend Logs.txt
Rabbit Logs.txt

[joffrey-bion]

It looks like you're passing /topic/chunk.changes as a host in the .stomp() call. This is supposed to be a host name, not a subscription destination.

By default, it will be the same as the one from the web socket connection. Did you try not to override it?

[joffrey-bion]

Then my guess is that the host needs to be different from aroundgame.ru. Maybe your server uses something different for STOMP. Did you check the RabbitMQ errors now? I don't know RabbitMQ, but from a quick Google search it looks like you should be able to see the virtual hosts in the RabbitMQ management console (see https://stackoverflow.com/a/51213694/1540818).

Or, if the host is valid, maybe it doesn't have the right permissions for your user.

[joffrey-bion]

I meant exactly what I wrote above, no additional accesses are required for the STOMP level

I understand what you said, I was wondering what the source of the information was, so I could get aligned. You said the backend says, so it was unclear where you found this information (if it was from some docs, or from some error response or something).

[MAXBAF1]

Then my guess is that the host needs to be different from aroundgame.ru. Maybe your server uses something different for STOMP. Did you check the RabbitMQ errors now? I don't know RabbitMQ, but from a quick Google search it looks like you should be able to see the virtual hosts in the RabbitMQ management console (see https://stackoverflow.com/a/51213694/1540818).

Or, if the host is valid, maybe it doesn't have the right permissions for your user.

I'll ask you about it now.

In the meantime, I will send you the following information, I hope you can learn something useful from this:

My man who wrote the aroundgame backend says the following:

In short, it looks like you're sending something crooked.

apparently, the stomp message curves or something like that, because the websocket connection itself is successfully upgraded

New StandardWebSocketSession[id=162de323-ad7b-6c 4d6c 4d6c 4d6c 4d6c 4d6c 4d-3186-731fec52731e, uri=ws://around game.ru/ws]

CONNECT user=1df25f1d-6135-47d9-92b0-5ff39127de77 session=162de323-ad7b-6c4d-3186-

here you are legally detected by the websocket

and then, when subscribing to /topic/chunk.changes via STOMP, this stomp message somehow sends you wrong

and rabbitmq reads
/topic/chunk.changes
as a vhost, not as a chunk.changes queue for exchange um.topic

and gives the Virtual host server error '/topic/chunk.changes' access denied

because my user (my backend as a user in rabbitmq) does not have such a vhost

[joffrey-bion]

The response from the person from the backend is exactly what I told you in my first message. It was about sending /topic/chunk.changes as a host in the stomp() call. Now you have changed it, and you're sending aroundgame.ru. Yet you still get a similar error, so the host is still wrong.

From the screenshot of the Virtual Hosts list, it looks like the only supported virtual host is /.

Could you please try setting host = "/" in the stomp() call?

[joffrey-bion]

By the way, I opened #580 to provide the body of the error frame by default in the exception, so you would immediately get this kind of information and save time.

[MAXBAF1]

Now the mistake is different, we will dig

Process: com.example.aroundcompose, PID: 30783 (Ask Gemini)
                 org.hildan.krossbow.stomp.ConnectionTimeout: Timed out waiting for 15s when connecting to /
                 	at org.hildan.krossbow.stomp.StompWsExtensionsKt.stomp(StompWsExtensions.kt:64)
                 	at org.hildan.krossbow.stomp.StompWsExtensionsKt.access$stomp(StompWsExtensions.kt:1)
                 	at org.hildan.krossbow.stomp.StompWsExtensionsKt$stomp$2.invokeSuspend(Unknown Source:13)
                 	at kotlin.coroutines.jvm.internal.BaseContinuationImpl.resumeWith(ContinuationImpl.kt:33)
                 	at kotlinx.coroutines.internal.ScopeCoroutine.afterResume(Scopes.kt:28)
                 	at kotlinx.coroutines.AbstractCoroutine.resumeWith(AbstractCoroutine.kt:100)
                 	at kotlin.coroutines.jvm.internal.BaseContinuationImpl.resumeWith(ContinuationImpl.kt:46)
                 	at kotlinx.coroutines.DispatchedTask.run(DispatchedTask.kt:100)
                 	at kotlinx.coroutines.internal.LimitedDispatcher$Worker.run(LimitedDispatcher.kt:113)
                 	at kotlinx.coroutines.scheduling.TaskImpl.run(Tasks.kt:96)
                 	at kotlinx.coroutines.scheduling.CoroutineScheduler.runSafely(CoroutineScheduler.kt:589)
                 	at kotlinx.coroutines.scheduling.CoroutineScheduler$Worker.executeTask(CoroutineScheduler.kt:816)
                 	at kotlinx.coroutines.scheduling.CoroutineScheduler$Worker.runWorker(CoroutineScheduler.kt:720)
                 	at kotlinx.coroutines.scheduling.CoroutineScheduler$Worker.run(CoroutineScheduler.kt:707)
                 	Suppressed: kotlinx.coroutines.internal.DiagnosticCoroutineContextException: [StandaloneCoroutine{Cancelling}@3efe82a, Dispatchers.Main.immediate]

[joffrey-bion]

Ok at least there is progress! Good to hear. Please let me know if you find the cause of the timeout.

For the investigation, you can try to increase the timeout by changing the defaults. Change the StompConfig() you pass to stomp() for this.

[MAXBAF1]

Everything is working, thank you very much for your prompt help ))