robotframework-kubelibrary 0.7.0

Kubernetes library for Robot Framework

KubeLibrary

RobotFramework library for testing Kubernetes cluster

Quick start

# install library itself
pip install robotframework-kubelibrary

# export KUBECONFIG
export KUBECONFIG=~/.kube/config

# run example tests
pip install robotframework-requests
git clone https://github.com/devopsspiral/KubeLibrary.git
cd KubeLibrary
robot -e prerelease testcases

Example testcase

Pods in kube-system are ok
    [Documentation]  Test if all pods in kube-system initiated correctly and are running or succeeded
    [Tags]    cluster    smoke
    Given kubernetes API responds
    When getting all pods in  "kube-system"
    Then all pods in "kube-system" are running or succeeded

Grafana has correct version
    [Documentation]  Test if Grafana container image is in correct version
    [Tags]    grafana
    Given kubernetes API responds
    When accessing "grafana-" excluding "svclb" container images version in "default"
    Then "grafana/grafana:6.5.0" version is used

More examples in testcases/ directory.

To see all the tests passing execute below commands.

Documentation

Library docs

Cluster Tests

# run cluster tests
robot -i cluster -e prerelease testcases/

Grafana Tests

helm repo add grafana https://grafana.github.io/helm-charts
helm repo update
helm install grafana grafana/grafana -f testcases/grafana/values.yaml

# run grafana tests
export KLIB_POD_PATTERN='grafana.*'
export KLIB_POD_ANNOTATIONS='{"kubelibrary":"testing"}'
export KLIB_POD_NAMESPACE=default

robot -i grafana -e prerelease testcases/

Octopus Tests

git clone https://github.com/kyma-incubator/octopus
helm install octopus octopus/chart/octopus/

# run octopus tests
export KLIB_RESOURCE_LIMITS_MEMORY=30Mi
export KLIB_POD_PATTERN='octopus.*'
export KLIB_RESOURCE_REQUESTS_CPU=100m
export KLIB_POD_LABELS='{"app":"octopus"}'
export KLIB_RESOURCE_LIMITS_CPU=100m
export KLIB_ENV_VARS='{"SECRET_NAME":"webhook-server-secret"}'
export KLIB_POD_NAMESPACE=default
export KLIB_RESOURCE_REQUESTS_MEMORY=20Mi

robot -i octopus -e prerelease testcases/

Other Tests

These tests require the kubelib-test helm-chart to be installed in your test cluster.

# run other library tests
export KLIB_POD_PATTERN='busybox.*'
export KLIB_POD_NAMESPACE=kubelib-tests
export KLIB_POD_LABELS='job-name=busybox-job'

kubectl create namespace $KLIB_POD_NAMESPACE
kubectl label namespaces kubelib-tests test=test
helm install kubelib-test ./test-objects-chart -n $KLIB_POD_NAMESPACE

robot -i other -e prerelease testcases/

Multi Cluster Tests

These tests require more than one cluster and utilize KinD as a setup. Download KinD and install it.

# Create Test Cluster 1
kind create cluster --kubeconfig ./cluster1-conf --name kind-cluster-1

# Create namespace in Test Cluster 1
kubectl create namespace test-ns-1 --context kind-kind-cluster-1 --kubeconfig ./cluster1-conf
# For bearer token auth
kubectl apply -f testcases/reload-config/sa.yaml
MYSA_TOKEN_SECRET=$(kubectl get sa mysa -o jsonpath="{.secrets[0].name}")
export K8S_TOKEN=$(kubectl get secret $MYSA_TOKEN_SECRET --template={{.data.token}} | base64 -d)
kubectl get secret $MYSA_TOKEN_SECRET -o jsonpath="{.data.ca\.crt}" | base64 -d > ca.crt
export K8S_API_URL=$(kubectl config view -o jsonpath='{.clusters[0].cluster.server}')
export K8S_CA_CRT=./ca.crt

# Create Test Cluster 2
kind create cluster --kubeconfig ./cluster2-conf --name kind-cluster-2

# Create namespace in Test Cluster 2
kubectl create namespace test-ns-2 --context kind-kind-cluster-2 --kubeconfig ./cluster2-conf

robot -i reload-config -e prerelease testcases/

# Clean up
kind delete cluster --name kind-cluster-1
kind delete cluster --name kind-cluster-2

Keywords documentation

Keywords documentation can be found in docs/.

Further reading

DevOps spiral article on KubeLibrary

KubeLibrary: Testing Kubernetes with RobotFramework | Humanitec

RobotFramework User Guide

Development

# clone repo
git clone https://github.com/devopsspiral/KubeLibrary.git
cd KubeLibrary

# create virtualenv
virtualenv .venv
. .venv/bin/activate
pip install -r requirements

Create keyword and test file, import KubeLibrary using below to point to library under development.

*** Settings ***

Library    ../src/KubeLibrary/KubeLibrary.py

For development cluster you can use k3s/k3d as described in DevOps spiral article on K3d and skaffold.

Generate docs

(
    # To generate keyword documentation a connection
    # to a cluster is not necessary. Skip to load a
    # cluster configuration.
    #
    # Set the variable local for the libdoc call only
    export INIT_FOR_LIBDOC_ONLY=1
    python -m robot.libdoc src/KubeLibrary/KubeLibrary.py docs/index.html
)