Kubes is a deployment tool that makes it easier and more fun to work with Kubernetes. Kubes automates 3 main steps.
- Build a Docker image.
- Create the YAML files that declare the Kubernetes resources.
kubectl applyto create the resources on the cluster.
It boils everything down to one single command:
To get started, install kubes:
gem install kubes
Then simply create an empty directory and run
kubes init in it.
$ mkdir demo $ cd demo # ECR here but can be GCR, ACR, etc $ REPO=111111111111.dkr.ecr.us-west-2.amazonaws.com/demo $ kubes init --app demo --repo $REPO exist create Dockerfile exist create .kubes/config.rb create .kubes/config/env/dev.rb create .kubes/config/env/prod.rb exist create .kubes/resources/base/all.yaml create .kubes/resources/base/deployment.yaml create .kubes/resources/shared/namespace.yaml create .kubes/resources/web/deployment.yaml create .kubes/resources/web/deployment/dev.yaml create .kubes/resources/web/deployment/prod.yaml create .kubes/resources/web/service.yaml Initialized .kubes folder Updated .gitignore
Kubes init creates a starter template structure with a simple Dockerfile that runs the httpd webserver. Note, if your project already has a Dockerfile, then kubes uses the existing Dockerfile instead.
To deploy it’s one simple command:
$ kubes deploy => docker build -t 111111111111.dkr.ecr.us-west-2.amazonaws.com/demo:kubes-2020-08-15T17-46-02 -f Dockerfile . => docker push 111111111111.dkr.ecr.us-west-2.amazonaws.com/demo:kubes-2020-08-15T17-46-02 Pushed 111111111111.dkr.ecr.us-west-2.amazonaws.com/demo:kubes-2020-08-15T17-46-02 docker image. Compiled .kubes/resources files to .kubes/output Deploying kubes resources => kubectl apply -f .kubes/output/shared/namespace.yaml namespace/demo created => kubectl apply -f .kubes/output/web/service.yaml service/web created => kubectl apply -f .kubes/output/web/deployment.yaml deployment.apps/web created $
As you can see, kubes builds the docker image, generates the YAML files, and then runs
kubectl in the correct order.
Kubes also prints out the underlying
kubectl commands so it’s clear what it’s doing. If you wish, you can run the commands directly also. Or you can also run the 3 steps individually. See Kubes Run in Separate Steps.
There are more conveniences that kubes offers:
|kubes get||List all the resources defined and created by kubes from the .kubes folder. It also lists the pods associated with the deployment.|
|kubes exec||Hops into the first deployment’s pod’s container. If there are multiple containers, you can specify the container with -c. This avoids having to manually copy and paste the pod name with
|kubes logs||Shows the logs for the deployment pods.|
There much more features and customizations that kubes provides. Here are some of them:
- Ordering: When deploying, kubes runs
kubectl applyfor each Kubernetes resource in the correct order. When deleting, kubes will also run
kubectl deletein the correct reverse order.
- Layering: Use the same Kubernetes YAML to build multiple environments like dev and prod with layering.
- CLI Args and Hooks: You can customize the CLI args. You can also run hooks before and after kubectl commands.
- Kustomize Support: If you’re a kustomization user, you can use it with Kubes.
For more details check out the Kubes Intro Docs
Kubes is a Kubernetes deployment tool that saves you precious finger-typing energy. It automates the process and provides many additional conveniences. It also prints out the underlying the kubectl commands, so you know what’s happening. By reducing the Kubernetes mental overhead, you can focus on your code and app development. Learn more on the official kubes.guru site.