I have (I believe) for the most part successfully created an embedded broker. It has been configured with two connectors, one addressed externally from tls/ssl://ip:61684 and the other from tcp://ip:61683. The idea actually is that the mutually authenticated url is to be used by external clients and the unsecure url is to be used by the internal client. The internal client is sitting on the same machine as the apollo broker.
My question is how to address the embedded broker in order to optimize out a network hop? My internal client is using the paho mqtt client library and I tried "vm://localhost:61683" and got an illegal argument exception (which makes sense I guess) as it is expecting "ssl" or "tcp" as the protocol identifier. First, is "vm://localhost:61683" the correct address to use for an embedded broker? That is what is documented for the activeMQ broker but apollo is not activeMQ and there is nothing in the apollo document that specifies how to address the embedded broker. Second, if that is the case then I assume that one will need to modify code in order to use the "vm" prefix with libraries like the paho mqtt client? -- View this message in context: http://activemq.2283324.n4.nabble.com/Address-to-an-embedded-apollo-broker-tp4674803.html Sent from the ActiveMQ - User mailing list archive at Nabble.com.
