Now that we have our code properly built/compiled, we can create a Dockerfile. This file defines how to build the Docker image for the service by declaring the environment, building stages, and copying any binaries or assets needed to run the service to the Docker image filesystem.
In other words, the Dockerfile serves as a recipe for creating a Docker image, from which Docker containers are spun up. This is ultimately what runs an instance of your service on the Embassy.
Create the necessary Docker files:
touch Dockerfile touch docker_entrypoint.sh
We’ll start editing the
Dockerfile by importing a base image, in this case Alpine, as recommended.
Next, we issue some commands to setup the filesystem. Here we update repositories and install required system packages.
RUN apk update RUN apk add tini curl
Next, we add the cross-compiled binary of
/usr/local/bin/ and add the
docker_entrypoint.sh file from the project root. Then, we set permissions for
ADD ./hello-world/target/aarch64-unknown-linux-musl/release/hello-world /usr/local/bin/hello-world ADD ./docker_entrypoint.sh /usr/local/bin/docker_entrypoint.sh RUN chmod a+x /usr/local/bin/docker_entrypoint.sh
Next, we set a working directory, and set the location of the entrypoint. Exposing ports is not necessary for EOS, but its often useful to leave this line for quick reference and clarity.
WORKDIR /root EXPOSE 80 ENTRYPOINT ["/usr/local/bin/docker_entrypoint.sh"]
That’s it! Let’s take a look at our final
FROM arm64v8/alpine:3.12 RUN apk update RUN apk add tini ADD ./hello-world/target/aarch64-unknown-linux-musl/release/hello-world /usr/local/bin/hello-world ADD ./docker_entrypoint.sh /usr/local/bin/docker_entrypoint.sh RUN chmod a+x /usr/local/bin/docker_entrypoint.sh WORKDIR /root # not necessary for EmbassyOS, but often left for quick reference and clarity EXPOSE 80 ENTRYPOINT ["/usr/local/bin/docker_entrypoint.sh"]
Finally, add the following code to the
#!/bin/sh exec tini hello-world
This is a script that defines how the service starts, and often acts as an init system. It will need to complete any environment setup (such as folder substructure), set any environment variables, and execute the run command. It’s also PID 1 in the Docker container, so should do all of the signal handling for container exits.