I am hard pressed to find much documentation on using/coding an embedded broker. The single example that comes with the distribution instantiates a Broker() object, but in the FAQ it is illustrated using a BrokerService() object. I don't understand the difference.
In the end I need to configure the service to support client-authenticated TLSV1. There will be a local keystore containing the server's private key/certificate and truststore containing the trusted client's certificates. The keystore and truststore will also be used by the 'main' server which is using the MQTT broker to publish commands to trusted clients. The idea of using the embedded appoach is to use an unsecure 'vm//' binding skipping the overhead of a real 'tcp//' binding. The external clients will use the mutually authenticated tls binding to communicate with both the MQTT apollo broker and the 'main' server. If I ever get this to work I will document it and put it somewhere on this site so others may more easily get started using the embedded option. (I would like to control ALL from code and use no xml; hopefully also learning how to include only those dependencies that are needed for MQTT. Appreciate pointers from anyone who has managed to get an embedded secured service up and running! Especially if you know of a good source of documentation. Thanks. -- View this message in context: http://activemq.2283324.n4.nabble.com/Embedded-approach-Broker-versus-BrokerService-tp4674755.html Sent from the ActiveMQ - User mailing list archive at Nabble.com.
