Skip to content

Latest commit

 

History

History
152 lines (110 loc) · 4.96 KB

CONTRIBUTING.md

File metadata and controls

152 lines (110 loc) · 4.96 KB
Raw

Contributing to Skiperator

Getting started with development

Installing dependencies

You're going to want to make sure to have the following things installed:

Running the operator

Check out the project from git. If you want to get to know the project hierarchy, have a look at the kubebuilder documentation.

$ git clone [email protected]:kartverket/skiperator.git
$ cd skiperator/

Local setup with make

make setup-local
make run-local

make setup-local will create a kind cluster called kind-skiperator, with all dependencies installed. make run-local will build and run the operator. This step can be replaced with building through your IDE if you need debugging.

If for any reason you need a clean local environment, delete your local kind cluster with kind delete cluster --name skiperator

Manual setup

Start a cluster on docker using kind.

$ kind create cluster

Make sure Kind is the active context

$ kubectx kind-kind

Optionally you may create a new namespace for Skiperator, for example

$ kubectl create namespace skiperator-system
$ kubens skiperator-system

Install Istio to make sure all CRDs are available

$ kubectl create ns istio-gateways && istioctl install

Cert-manager (version in link above) must be installed locally on your cluster with

$ kubectl apply -f https://github.com/cert-manager/cert-manager/releases/download/<VERSION>/cert-manager.yaml

Prometheus Operator CRDs must be installed locally:

$ kubectl apply -f https://github.com/prometheus-operator/prometheus-operator/releases/download/<VERSION>/stripped-down-crds.yaml

Roles and CRDs are autogenerated and can be generated with

$ make generate

and installed in the cluster with

$ kubectl apply -f config/ --recursive

A few ConfigMaps for a few settings in Skiperator is needed. These vary from cluster to cluster, but a few dummies have been added in the samples/ folder, together with some sample applications. These can be added using the following command

$ kubectl apply -f samples/ --recursive

To start skiperator locally on your computer, you can use the the make target run-local which will generate CRD's, apply them in your local kind-cluster, build and run skiperator. The default kubectl context is kind-kind. To override, set the env-variable SKIPERATOR_CONTEXT to the name of the desired context.

$ make run-local

Now you should have a running app in your namespace. Run the following command to see all the created resources.

$ kubectl get Application,all,networkpolicies,PeerAuthentication,Gateway,VirtualService,Sidecar

Changes to the api requires the CRD to be generated and applied again, and changes to the application controllers requires skiperator to be rebuilt and startet again.

Use the command make run-local to do this in a safe way without mixing up kubectl contexts.

Tests

This project is using Chainsaw for testing.

To run tests against your running kind cluster, with skiperator running in your IDE:

# All tests
make test

# Single test
make test-single dir=tests/application/hpa

If you want to run tests without running skiperator yourself you can use

# All tests
make run-test

# Single test
make run-test TEST_DIR=tests/application/hpa

You can find more handy parameters here.

Known issues

Parallel tests can cause some concurrency issues. Try lowering the amount of tests running in parallel if you experience tests that occasionally fail. Concurrent tests means a unique namespace for each test. Chainsaw will create a unique namespace, but in some cases you might need to specify a namespace. In that case name your namespace -ns so we dont get overlapping namespaces.

Writing code

Have a look at the following files:

Also look at the following documentation pages: