I am currently building a UDT to publish via system.cirruslink.transmission.publish() into a custom UNS structure.
In testing I have disabled the broker and on restoring the connection I am getting a large amount of data published at one time. Does this script interface with the default history store and follow the flush parameters as configured? It would be good to maintain a flush period to slowly deplete my buffer and make sure the receiving end has time to process each data point.
Were you able to store messages in the history when using the “system.cirruslink.transmission.publish()” function?
I’m trying to use the same function, but want messages to be stored when the broker is offline.
No, after testing it appears that the Python scripting is held in the gateway memory due to a failed RPC to the broker. Upon restoration the data flow will resume to the broker but with no ability to throttle. If the gateway goes offline this would result in data loss as no on disk store is utilized. The store and forward appears limited to using the built-in tag transmitters.