A Server using the Paho client connects to the Apollo broker on an unsecure internal connection tcp://127.0.0.1:8440. A remote client connects to the Apollo broker using a mutual TLS connection https://192.168.127.107:8445.
The server subscribes for message responses and connection status topics. The remote client connects, subscribes, publishes a connection status event and awaits for messages. All topics are at QoS level 2 and the clean session flag is false. No stay alive pings. The server receives the remote client's connection status event. The server is then able to publish message topics to the remote client and the client receives them. The remote client shuts down and then restarts. It uses saved information to reconnect to the MQTT server and publishes the connection status event. It does not resubscribe; it should not have to. The server receives the event. However, the server is unable to publish events (well, the remote client does not see them). Using the Administrative console what I see on the tcp://127.0.0.1:8440 connection is a queue of each publish the server tries to do; all 'in' and one out. On the secure https://192.168.127.107:8445 connection I see 1 in and 0 out and with the currently connected value set to one. Why are the messages getting queued by the broker and not being forwarded to the client? Is this a miss-configuration on the client or server side? Any ideas on how I can further solve this problem? -- View this message in context: http://activemq.2283324.n4.nabble.com/Apollo-MQTT-server-queuing-and-NOT-publishing-topics-tp4679478.html Sent from the ActiveMQ - User mailing list archive at Nabble.com.
