Generally speaking, how a message is published has nothing to do with how it is consumed. This is one of the main benefits of the decoupled nature of messaging.
Furthermore, every protocol has its own semantics when it comes to guarantees for consuming and/or producing messages. Therefore, when you produce a message from an MQTT client using QoS 2 that only applies to the production of that message, not the consumption. The AMQP client can consume that message however it wants, and it should expect guarantees (if any) according to the AMQP specification. Justin On Tue, Mar 14, 2023 at 3:40 AM Modanese, Riccardo <riccardo.modan...@eurotech.com.invalid> wrote: > Hello, > I'm pretty ignorant about AMQP protocol and Artemis broker > internals so please excuse me if my question is inappropriate. > > I didn’t find a clear documentation telling what happens about the MQTT > QoS (subscribed or published messages) while crossing the connector > boundary. > I mean, if I configure Artemis broker to open 2 different connectors (A > and B) supporting only one protocol each (A -> MQTT and B -> AMQP) if my > device connected to MQTT connector A publishes a message with QoS = 2 and I > have a subscriber for that address connected through AMQP connector B what > should I expect? > To receive the message just once as MQTT QoS 2 tells? > > Regards, > Riccardo >
