Real time reply to Class A device

I’m trying to implement a device that needs to be able to receive a processed payload every time it send data to the app server.
Something like this:

  • Device sends payload
  • Application receives data from MQTT
  • Application send processed answer to MQTT
  • Answer is sent to device in the RX1 or RX2.

It usually takes 100~200ms from the received message to my application publish the answer in MQTT, so RX windows should be still “open”.
However, the answer are being queued and only being delivered in the next communication, making everything out of sync.
I was thinking about delaying the RX1 window, however those configuration don’t show for OTAA devices. I’m missing something?
Do you have any suggestion how can I archive something like this?
Thanks :smiley:

The RX delay can be configured in the LoRa Server configuration. Please see also the get_downlink_data_delay="100ms" setting. By default LoRa Server will wait 100ms before pulling an item from the downlink queue. When you increase this, your application has more time to enqueue a payload to be scheduled in the receive-window of the same “transaction”. Note that this also increases the roundtrip time, so you might need to increase de rx-delay too in this case.