The Kubernetes Package Manager
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
 
 
 
Go to file
vaikas-google b5edff70f6
Merge pull request #155 from runseb/patch-2
9 years ago
common Deployment now has top-level error for deployment-level failures 9 years ago
dm Add help text for manifest command 9 years ago
docs Fix some document errors. 9 years ago
examples Broke guestbook. 9 years ago
expandybird Reintroduce replicatedservice.py as it is used by tests. 9 years ago
manager Resource errors should result in a failed deployment with good errors. 9 years ago
registry address CR comments 9 years ago
resourcifier Merge pull request #155 from runseb/patch-2 9 years ago
util address CR comments 9 years ago
version Initial commit for DeploymentManager on k8s. 9 years ago
.gitignore Streamline getting started. 9 years ago
.travis.yml Simpler travis config. 9 years ago
CONTRIBUTING.md Initial commit for DeploymentManager on k8s. 9 years ago
LICENSE Initial commit for DeploymentManager on k8s. 9 years ago
Makefile Edit Makefile to actually return failure code on test failures. 9 years ago
README.md Fix some document errors. 9 years ago
install.yaml Put DM into its own namespace. 9 years ago

README.md

Deployment Manager

Go Report Card

Deployment Manager (DM) dm makes it easy to create, describe, update and delete Kubernetes resources using declarative configuration. A configuration is just a YAML file that configures Kubernetes resources or supplies parameters to templates. Templates are just YAML files with Jinja mark up or Python scripts.

For example, this simple configuration deploys the Guestbook example:

resources:
- name: frontend
  type: github.com/kubernetes/application-dm-templates/common/replicatedservice:v1
  properties:
    service_port: 80
    container_port: 80
    external_service: true
    replicas: 3
    image: gcr.io/google_containers/example-guestbook-php-redis:v3
- name: redis
  type: github.com/kubernetes/application-dm-templates/storage/redis:v1
  properties: null

It uses two templates. The front end is a replicated service, which creates a service and replication controller with matching selectors, and the back end is a Redis cluster, which creates a Redis master and two Redis slaves.

Templates can use other templates, making it easy to create larger structures from smaller building blocks. For example, the Redis template uses the replicated service template to create the Redis master, and then again to create each Redis slave.

DM runs server side, in your Kubernetes cluster, so it can tell you what templates you've instantiated there, what resources they created, and even how the resources are organized. So, for example, you can ask questions like:

  • What Redis instances are running in this cluster?
  • What Redis master and slave services are part of this Redis instance?
  • What pods are part of this Redis slave?

Because DM stores its state in the cluster, not on your workstation, you can ask those questions from any client at any time.

Templates live in ordinary Github repositories called template registries. See the Kubernetes Template Registry for curated Kubernetes applications using Deployment Manager templates.

For more information about configurations and templates, see the design document.

Please hang out with us in the Slack chat room and/or the Google Group for the Kubernetes configuration SIG.

Installing Deployment Manager

Follow these 3 steps to install DM:

  1. Make sure your Kubernetes cluster is up and running, and that you can run kubectl commands against it.
  2. Clone this repository into the src folder of your GOPATH, if you haven't already. See the Kubernetes developer documentation for information on how to setup Go and use the repository.
  3. Use kubectl to install DM into your cluster kubectl create -f install.yaml

That's it. You can now use kubectl to see DM running in your cluster:

kubectl get pod,rc,service --namespace=dm

If you see expandybird-service, manager-service, resourcifier-service, and expandybird-rc, manager-rc and resourcifier-rc with pods that are READY, then DM is up and running!

Using Deployment Manager

Setting up the client

The easiest way to interact with Deployment Manager is through the dm tool hitting a kubectl proxy. To set that up:

  1. Build the tool by running make in the deployment-manager repository.
  2. Run kubectl proxy --port=8001 --namespace=dm & to start a proxy that lets you interact with the Kubernetes API server through port 8001 on localhost. dm uses http://localhost:8001/api/v1/proxy/namespaces/dm/services/manager-service:manager as the default service address for DM.

Using the client

The DM client, dm, can deploy configurations from the command line. It can also pull templates from a template registry, generate configurations from them using parameters supplied on the command line, and deploy the resulting configurations.

Deploying a configuration

dm can deploy a configuration from a file, or read one from stdin. This command deploys the Guestbook example using the configuration shown above from the examples directory in this project:

dm deploy examples/guestbook/guestbook.yaml

You can now use kubectl to see Guestbook running:

kubectl get service

Look for frontend-service. If your cluster supports external load balancing, it will have an external IP assigned to it, and you can navigate to it in your browser to see the guestbook in action.

For more information about this example, see examples/guestbook/README.md

Deploying a template directly

You can also deploy a template directly, without a configuration. This command deploys a redis cluster with two slaves from the redis template in the Kubernetes Template Registry:

dm deploy storage/redis:v1

You can optionally supply values for template parameters on the command line, like this:

dm --properties workers=3 deploy storage/redis:v1

When you deploy a template directly, without a configuration, dm generates a configuration from the template and the supplied parameters, and then deploys the configuration.

For more information about deploying templates from a template registry or adding types to a template registry, see the template registry documentation.

Additional commands

Here's a list of available dm commands:

expand              Expands the supplied configuration(s)
deploy              Deploys the named template or the supplied configuration(s)
list                Lists the deployments in the cluster
get                 Retrieves the named deployment
delete              Deletes the named deployment
update              Updates a deployment using the supplied configuration(s)
deployed-types      Lists the types deployed in the cluster
deployed-instances  Lists the instances of the named type deployed in the cluster
templates           Lists the templates in a given template registry
describe            Describes the named template in a given template registry

Uninstalling Deployment Manager

You can uninstall Deployment Manager using the same configuration:

kubectl delete -f install.yaml

Building the Container Images

This project runs Deployment Manager on Kubernetes as three replicated services. By default, install.yaml uses prebuilt images stored in Google Container Registry to install them. However, you can build your own container images and push them to your own project in the Google Container Registry:

  1. Set the environment variable PROJECT to the name of a project known to GCloud.
  2. Run make push

Design of Deployment Manager

There is a more detailed design document available.

Status of the Project

This project is still under active development, so you might run into issues. If you do, please don't be shy about letting us know, or better yet, contribute a fix or feature.

Contributing

Your contributions are welcome.

We use the same workflow, License and Contributor License Agreement as the main Kubernetes repository.

Relationship to Google Cloud Platform

DM uses the same concepts and languages as Google Cloud Deployment Manager, but creates resources in Kubernetes clusters, not in Google Cloud Platform projects.