gcloud docker will return the following message when executing:

WARNING: gcloud docker will not be supported for Docker client versions above 18.03. Please use gcloud auth configure-docker to configure docker to use gcloud as a credential helper, then use docker as you would for non-GCR registries, e.g. docker pull gcr.io/project-id/my-image. Add --verbosity=error to silence this warning, e.g. gcloud docker --verbosity=error -- pull gcr.io/project-id/my-image. See: https://cloud.google.com/container-registry/docs/support/deprecation-notices#gcloud-docker

I'm following up here from a support ticket we submitted to CodeShip. Recently, it seems Docker has been removed from this image. In the ticket, we asked if it was possible to add Docker ack in.

Here's was Drew's reply to our ticket:

Moving forward, we're aiming to keep the images as lean as possible here. Nothing to preclude you from building off the base image as you've recently done. Apologies again for the trouble with the update.

I'd like to respectfully request you guys reconsider this policy. Anyone who's deploying to GKE is going to need Docker installed in their image, and I'd bet that's a large group of people. Our deployment scripts depend on docker being present, and this change broke them unexpectedly.

Yes, we could extend the image and install our own tools. However, one of the biggest benefits of using CodeShip for us is that we don't have to maintain our own images—we can pull in a standard CodeShip image and run deploy with minimal effort. Pulling Docker out creates a distributed maintenance cost for anyone who wants to use these images and needs Docker.

Could you elaborate a little more on why you're striving for lean images? Is it an infrastructure cost issue? Is it due to performance? When it comes to deployment, my team's preference is for functionality. We use docker and kubectl extensively in our deployments, and we'd prefer to have those tools within arms reach by default.

Thanks for taking the time to consider.

Getting this warning when using your image. You should switch to:

gcloud auth configure-docker

Hey there

For some reason the kubectl client installed is outdated, which means we don't have kubectl apply which is the recommended way for updating deployments.

If you run kubectl version inside this container you'll get back:

Client Version: version.Info{Major:"1", Minor:"0", GitVersion:"v1.0.6", GitCommit:"388061f00f0d9e4d641f9ed4971c775e1654579d", GitTreeState:"clean"}
Server Version: version.Info{Major:"1", Minor:"3", GitVersion:"v1.3.7", GitCommit:"a2cba278cba1f6881bb0a7704d9cac6fca6ed435", GitTreeState:"clean"}

The latest version of the client isv1.3.7.

Its strange, because I can see you have a run directive for gcloud components update kubectl inside the Dockerfile. Is the image stored on dockerhub the same as what's here on github?