The objective of the project is to have a mesh network connected to the edge device (Raspberry PI) and this edge connected to the IoT Hub and azure portal.
Because our Mesh module is already in python, we wanted to try to be coherent if possible and develop all other modules in python as well. The problem is:
I want to create multiple twins, one for each device in the mesh network, but the python sdk, contrary to the java one, doesn't seem to have a devicetwin class, for which we can start multiple twins depending on the object used.
I tried passing the devices as a json array in the normal device twin reporter properties, but it doesn't support arrays.
Anyone as any experience with this or a different solution? Or should I just give up and have a java module if I want to mess around with multiple device twins?
So I think I misunderstood something, the java code I analyzed as base is acting as a identity translation gateway, it opens multiple device clients that I think are signed with different keys, therefore not creating an error.
I didn't notice this and in python tried to open 2 device clients but with same keys, and that's probably why it didn't work.
So my question now is another one, lets imagine I scan the mesh network and find 2 devices, I want to automatically add them to the IoTHub from the edge, is this possible? Do I have to manually add devices if I want to have a device twin on the Hub? These devices would probably have to keep their own keys stored in their memory?