Question: what is the recommended practice for doing dry-run builds?

[starpit]

Hello, nice project here!

The docker/build-and-push action supports push: false which allows for doing test builds without a push to the upstream container registry. How would you recommend we do so with this workflow?

[ngeorger]

Hello, nice project here!

The docker/build-and-push action supports push: false which allows for doing test builds without a push to the upstream container registry. How would you recommend we do so with this workflow?

Hi! Thank you :)

• Well, it will depend on which stage of the building process you'd like to "stop". If you don't want to push anything to the registries, you should start by disabling every login action, ie:

multi-arch-docker-github-workflow/.github/workflows/multi-build.yaml

Lines 111 to 118 in 0fe24c8

	- name: Login to GitHub Container Registry
	# This step logs in to the GitHub Container Registry (GHCR) using the docker/login-action.
	# It uses the GitHub actor's username and the GITHUB_TOKEN secret for authentication.
	uses: docker/login-action@5e57cd118135c172c3672efd75eb46360885c0ef # v3.6.0
	with:
	registry: ghcr.io
	username: ${{ github.actor }}
	password: ${{ secrets.GITHUB_TOKEN }}

• As you mentioned, the docker/build-push-action supports disabling pushing, but in this case we use a digests-pushing approach in order to merge them later into a single manifest which references those digests created by separated runners for each architecture.

multi-arch-docker-github-workflow/.github/workflows/multi-build.yaml

Lines 121 to 140 in 0fe24c8

	- name: Build and push by digest
	# This step builds and pushes the Docker image using Buildx.
	# It uses the docker/build-push-action to build the image with the specified context and platforms.
	# The image is built with the labels and annotations generated in the previous steps.
	# The outputs are configured to push the image by digest, which allows for better caching and versioning.
	# The cache-from and cache-to options are used to enable caching for the build process.
	# The cache is stored in GitHub Actions cache and is scoped to the repository, branch, and platform.
	id: build
	uses: docker/build-push-action@263435318d21b8e681c14492fe198d362a7d2c83 # v6.18.0
	env:
	DOCKER_BUILDKIT: 1
	with:
	context: .
	platforms: ${{ matrix.platform }}
	labels: ${{ steps.meta.outputs.labels }}
	annotations: ${{ steps.meta.outputs.annotations }}
	outputs: type=image,name=${{ env.GHCR_IMAGE }},push-by-digest=true,name-canonical=true,push=true,oci-mediatypes=true
	cache-from: type=gha,scope=${{ github.repository }}-${{ github.ref_name }}-${{ matrix.platform }}
	cache-to: type=gha,scope=${{ github.repository }}-${{ github.ref_name }}-${{ matrix.platform }}

• Maybe if you could provide more details or context for your use case, I could provide some ideas, but at this time and without more context, I could only suggest to skip/disable the merge step, but this won't prevent to build/push the digests itself.

multi-arch-docker-github-workflow/.github/workflows/multi-build.yaml

Lines 165 to 167 in 0fe24c8

	merge:
	# This job merges the Docker manifests for the different platforms built in the previous job.
	name: Merge Docker manifests

Please, if you could provide more details, I could suggest a better solution.

Regards!

[ngeorger]

Answered in #33 (comment)