The next phase for getting your custom Docker images deployed is to push them to your very own private container registry.
By now you will have provisioned a container registry on Azure so now we have somewhere to push our Docker images.
Firstly we'll need to make sure we have a docker image of our application ready to push.
Let's push up an image of the Python app we worked with in session 3.
Navigate to the directory for that application (that you cloned in session 3).
It's on the link below if you need to re-clone
https://github.com/techreturners/devops-bookstore-api
Make sure Docker is running on your computer and build the docker image by running this command:
docker build -t devops-bookstore-api:1.0 .
That will produce a local docker image. One that we can PUSH up to your container registry.
Next we need to make sure your local docker instance has the right credentials to push up to your container registry.
To authenticate with the container registry, replace the REGISTRY_NAME in the following command with your registry name and then run:
az acr login --name REGISTRY_NAME
It should so something similar to:
Login Succeeded
The way you tell Docker about your container registries is through the tagging system.
You can tag the image you built in step 2 with a new tag for your container registry. To do this (replace the REGISTRY_NAME with your Azure registry name):
docker tag devops-bookstore-api:1.0 REGISTRY_NAME.azurecr.io/devopsupskill/devops-bookstore-api:1.0
You can see your Docker images by running docker images
and you should see your tagged version.
Once the image has been tagged and you have configured docker to be authenticated with your account you should be able to push your image.
Remember the registry name will be slightly different to the example shown below.
docker push REGISTRY_NAME.azurecr.io/devopsupskill/devops-bookstore-api:1.0
If all works you should see it show progress of your "push"
If you then open up the Azure web console you should be able to navigate to the container registry and see your image. It should look similar to the image below:
You can now head back to the README and lets work through getting that image deployed.