This application (WaziApp) for the WaziGate predicts irrigation times with help of soil sensors and a weather API (open-meteo). A non-technical user guide with general instructions can be obtained from here.
Visit the WaziGates UI and open the App section. Press the plus button and click install custom app. Type into the textbox "waziup/irrigation-prediction:dev" to download/install the application from dockerhub.
For debugging there are 3 run configurations possible:
- Production mode: run as WaziApp in a docker container on the WaziGate
- Local debugging environment: Debug in Visual Studio with wazigate-edge and wazigate-dashboard running locally
- Local debugging environment against local gateway: Debug in Visual Studio with wazigate-edge and wazigate-dashboard running locally, using the API of a WaziGate that has its API exposed (e.g. local network).
You can change the "run configuration" in the ".env" file, just choose one of the three available options.
docker buildx build --platform linux/arm64/v8 -t waziup/irrigation-prediction:dev --no-cache --pull --build-arg CACHEBUST=$(date +%s) --load .
Issue the following cmd to push to dockerhub (you have to be logged in):
docker push waziup/irrigation-prediction:dev
Create a folder with the name of the app at the following path: "/var/lib/wazigate/apps/waziup.irrigation-prediction" and copy the files (docker-compose.yml & package.json) from the repository.
Issue the following cmd to push to dockerhub (you have to be logged in):
docker save {id of image} | gzip | pv | ssh pi@{ip of rpi} docker load
# Example:
docker save my-docker-image | gzip | pv | ssh [email protected] docker load