As a part of your IoT solution, you have an edge device which runs a Stream Analytics job as a custom module. Based on the actual data from the previous periods, you updated the job’s logic and now you want to update your deployment on the device itself. As part of the process, you have published the module to your container registry on ACR, sent the deployment manifest to the device. Everything looked fine, the edge runtime processed the deployment and the device is up and running. After a short time you notice that the module’s behavior is the same as before the update. 
What can be the probable reason for that?

Answer options:

A.The device must be restarted so that the changes take effect
B.The version number in module.json is the same as its version number in the ACR
C.The version number in module.json differs from its version number in the ACR
D.The update of the modules should be executed as part of updating the edge runtime

Correct Answer: B
Option A is incorrect because the device doesn’t need to be restarted. During new deployment the edge runtime is processing the deployment manifest deploying and restarting only the modules affected. The device itself remains up and running.
Option B is CORRECT because if you publish a new version of a module’s container to your container registry using the same version tags as the previous module, the runtime will not download the new version. So, the new version number must differ from the previous one.
Option C is incorrect because if you publish a new version of a module’s container to your container registry using the same version tags as the previous module, the new version won’t be downloaded.
Option D is incorrect because there is no such constraint. Any one of the modules can be deployed/updated independently of the other components on the device.
References: 

https://docs.microsoft.com/en-us/azure/iot-edge/tutorial-machine-learning-edge-06-custom-modules?view=iotedge-2018-06#build-publish-and-deploy-modules
https://docs.microsoft.com/en-us/azure/iot-edge/how-to-deploy-modules-cli?view=iotedge-2018-06