NServiceBus.RabbitMQ and Streams?

Does NServiceBuse supports RabbitMQ streams?

Hi @dnf,

No, we do not support using RabbitMQ streams.

@bording Any plans for this in future?

Not at this time, no. Streams are very different from queues, so they serve a different purpose.

See Let's talk about Kafka • Particular Software for more information about what we think about streams vs queues. While that is about Kafka, everything in there would also apply to RabbitMQ streams.

Hi @dnf

Following up on what Brandon said, what is the use case you are thinking about for the RabbitMQ streams? We are tracking a couple of different possible usage scenarios to be able to make technical decision in the future and these are:

  • I need to integrate with an existing system that uses event streaming (be it Kafka or RabbitMQ streams) for messaging
  • I need to discover business-significant events in streams of events
  • I need to build a messaging system over an event-streaming technology because those are the organization rules

@dnf one more thing. We have a sample on how to process Kafka streams in order to extract business-meaningful events from the stream and “uplift” them: Using NServiceBus and Kafka in Azure Functions (isolated) • NServiceBus.AzureFunctions.Worker.ServiceBus Samples • Particular Docs

Is it similar to what you intend to use RabbitMQ streams for?

Hi @SzymonPobiega. First of all I have to point out that I’m beginning with messaging and I’m doing some research on what tool we will use to event driven scenarios. One scenario I’m thinking about is building read models from events and streams solution like kafka looks perfect fit for this. On the other hand there are scenarios where message broker like RabbitMQ is better fit so it would be great to have one service installed (RabbitMQ) for both scenarios and it would be even better not to work on rabbit native API but use some enterprise solution like NServiceBus. Maybe may needs are not realistic because I’m new in the topic but for now it make sense :slight_smile: Another point is that this solution will be on prem so we will not use Azure functions etc.