cc @radcortez @gsmet

@mocenas 👋 the test is failing and requires a fix or it's all fine?

@mocenas 👋 the test is failing and requires a fix or it's all fine?

That's weird it passes for me no problem, executing in using mvn -pl integration-tests/reactive-messaging-kafka clean verify -Dtest-containers -Dtest=TransmitterTest

What failure do you have with it?

I rebased the PR on up-to-date main.

@mocenas it was a question :). But I think you answered it and it's all fine.

Basically I wanted to know if it was a reproducer for an issue to fix or an additional test. I understand it's an additional test.

I decided to put this test into separate module in the end. When I tried to add it into existing integration-tests in just kept breaking stuff. E.g. The module I tried to put it in originally integration-tests/reactive-messaging-kafka turned out to be really sensitive about order of execution of testClasses. It was IMHO not worth to refactor the entire module or hack it to enforce order of execution.

Can you please update the title of the PR, as it currently requires someone looking at the issue to figure out even the basic context

I'll let @ozangunalp decide how he wants to handle this, but bear in mind a couple things:

• We usually don't create new integration test modules unless we have to
• I believe it would be better to use *Emitter instead of *Transmitter as it aligns with the Quarkus Messaging terminology

I agree that we can add a test like this. Probably we also need to add one with the usage of InMemoryConnector for tests. That's where we had the regression.

We can add this test to an existing test module with more descriptive bean & test names such as MessageProducer and UsingKafkaCompanionTest. Naming is hard :)

OK, I renamed the classes and moved the reproducer into integration-tests/reactive-messaging-kafka module.