You will need roughly 4 minutes to read this article.
We’ve got quickstart repos, sample apps and a getting started guide available to make starting out with Codeship Pro faster and easier.
Any Go service or tool that can run inside a Docker container will run on Codeship Pro. This documentation article will highlight simple configuration files for a Go-based Dockerfile and project.
When accessing other containers please be aware that those services do not run on
localhost, but on a different host, e.g.
mysql. If you reference
localhost in any of your configuration files you will have to change that to point to the service name of the service you want to access. Setting them through environment variables and using those inside of your configuration files is the cleanest approach to setting up your build environment.
project_name: build: image: organisation_name/project_name dockerfile: Dockerfile depends_on: - redis - postgres environment: - DATABASE_URL=postgres://postgres@postgres/YOUR_DATABASE_NAME - REDIS_URL=redis://redis redis: image: healthcheck/redis:alpine postgres: image: healthcheck/postgres:alpine
Note that in this example we are using the healthcheck version of our Redis and PostgreSQL images to avoid startup timing issues.
The following is an example of a Codeship Steps file.
Note that every step runs in isolated containers, so changes made on one step do not persist to the next step. Because of this, any required setup commands, such as migrating a database, should be done via a custom Dockerfile, via a
entrypoint on a service or repeated on every step.
- service: project_name command: bash -c "go build ./... && go test ./..."
Following is an example Dockerfile with inline comments describing each step in the file. The Dockerfile shows the different ways you can install extensions or dependencies so you can extend it to fit exactly what you need. Also take a look at the Golang image documentation on the Docker Hub.
# Starting from the latest Golang image FROM golang:1.9 # INSTALL any further tools you need here so they are cached in the docker build # Set the WORKDIR to the project path in your GOPATH, e.g. /go/src/github.com/go-martini/martini/ WORKDIR /go/src/your/package/name # Copy the content of your repository into the image COPY . ./ # Install dependencies through go get, unless you vendored them in your repository before # Vendoring can be done with an external tool like godep or glide # Go versions after 1.5.1 include support for a vendor directory RUN go get
Using Docker’s multi-stage build feature, you can implement some changes to you Dockerfile to allow you to build and use a Go binary from a single Dockerfile, outputting a Docker image with the Go binary but none of the Golang build tools - meaning a smaller and more efficient image with a less complex setup.
Multi-stage builds allow you to specify multiple
FROM lines in a Dockerfile, where each
FROM line begins a new stage. The final image is the result of the last stage, which means any previous stages are not saved in the final image. This is great for creating “builder” workflows easily.
Here’s an example using Go in a Dockerfile:
# phase one, labeled as build-stage # first stage does the building FROM golang:1.9 as build-stage WORKDIR /go/src/github.com/codeship/go-hello-world COPY hello-world.go . RUN go build -o hello-world . # starting second stage FROM alpine:3.6 # copy the binary from the `build-stage` COPY --from=build-stage /go/src/github.com/codeship/go-hello-world/hello-world /bin CMD hello-world
Notice that the second
FROM line begins the second stage, and this second stage is what the final image will consist of.
Because of version and test dependency issues, it is advised to try using the Jet CLI to debug issues locally via
You can enable caching per service in your Services file.
Do you think we need to improve this article? If so, please submit our feedback form to help us improve this article!