Thanks a lot for the time you take Justin here. Bellow information you requested
> Le 15 mars 2023 à 20:10, Justin Bertram <jbert...@apache.org> a écrit : > > - How many clients will be using the edge broker, and how will they be > using it (e.g. consumer, producer, protocol, multicast, anycast, etc.)? On the central one, one consumer for all feedbacks, and one producer of orders One each edge one, one consumer for all orders, one producer for feedbacks, and 2 to 3 services for local sync (each produce and consume local signals) > - How many addresses & queues will be used on the edge broker? Is the > configuration static in broker.xml or dynamic via auto-creation? One for upstream, one for local, all statics > - Which queues will be read to send messages to the cloud broker? Only one > - Which addresses will receive messages from the cloud broker? Not yet at that level of knowledge on my side (still learning) > Other considerations... > > - What happens if the edge broker crashes? Nothing breaks and no one get hurts. It won't be used for active piloting but for feedback and signaling. No actions can't be started without full sync of requests on each clients, and loss of signals will mean loss of observability but not loss of safety systems. > - What happens if connectivity to the cloud broker is absent for an > extended period of time? Current work is finished without issue, no new work can be sent to the cell, and we are blind to the status of the last one from the central point. If the messages are kept until the network recovery, and sent once connectivity is back, we will just resume where we were. > - Will you use any kind of flow-control for the producers to ensure the > broker isn't overwhelmed with messages? Not yet planned but can be That's clearly the kind of help and wondering I was hopping here: questions to challenge our idea from people that know the technology and can lead us to appropriate config / training source / eventually consultant or support (once we pass the PoC). Thanks a lot
