Announcement

Collapse
No announcement yet.

MQTTS and QOS on BRX

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts


  • #16
    To give you an idea of the application - we have controllers in fairly hostile locations that tend to be in remote areas. It's not unusual for their network and power connections to be out on a regular basis. Regulatory requirements decree that we keep permanent logs of events (that typically get sent to the cloud via MQTT and then discarded on the controller). Thus we can't throw away messages unless we're sure they've been accepted at the far end. It's an interesting scenario but I suspect not terribly unusual in the IoT world.

    One other thing I just though of - would it be hard to expose the clean-session flag? Our requirement is to set it to false (ie persistent sessions) - I didn't see this in the BRX/MQTT video.

    Comment



    • #17
      Originally posted by Andrew S View Post
      To give you an idea of the application - we have controllers in fairly hostile locations that tend to be in remote areas. It's not unusual for their network and power connections to be out on a regular basis. Regulatory requirements decree that we keep permanent logs of events (that typically get sent to the cloud via MQTT and then discarded on the controller). Thus we can't throw away messages unless we're sure they've been accepted at the far end. It's an interesting scenario but I suspect not terribly unusual in the IoT world.

      One other thing I just though of - would it be hard to expose the clean-session flag? Our requirement is to set it to false (ie persistent sessions) - I didn't see this in the BRX/MQTT video.
      It's hardcoded, but easy enough to change since we're already messing with the UI.

      Comment



      • #18
        Great. Thanks Bob.

        Comment



        • #19
          Regarding the MQTT Client ID - it's exactly what you set in the Device Name field, right?

          Comment



          • #20
            Fyi...


            Click image for larger version

Name:	image_3784.png
Views:	23
Size:	36.9 KB
ID:	120495

            Comment



            • #21
              Looking great Bob. This is exciting. Will all of these changes make it into 2.4?

              Comment



              • #22
                Originally posted by Andrew S View Post
                Looking great Bob. This is exciting. Will all of these changes make it into 2.4?
                That is the hope, barring testing issues.

                Comment



                • #23
                  Can you confirm that the MQTT Client-ID is the device name? I need to set specific client-ids which cannot change over time.

                  Comment



                  • #24
                    Originally posted by Andrew S View Post
                    Can you confirm that the MQTT Client-ID is the device name? I need to set specific client-ids which cannot change over time.
                    DOMORE_mmmmmm_iiii

                    Where mmmmmm is the MAC address of the unit and iiii is the device ID of the MQTT device. Since there can be more than one device, it was necessary to use the device ID to make it unique.

                    Comment



                    • #25
                      Gotcha. That's a reasonable format. As a possible future feature enhancement may I suggest allowing that field to be overridable. Some back-end systems require MQTT devices to have a specific format for their client-ID.

                      Comment

                      Working...
                      X