Functional Tests porting to kubernetes (#553)

* Functional Tests porting to kubernetes

Signed-off-by: Tullio Sebastiani <tsebasti@redhat.com>

.github/workflows/build.yml

@@ -1,51 +0,0 @@
-name: Build Krkn
-on:
-  pull_request:
-
-jobs:
-  build:
-    runs-on: ubuntu-latest
-    steps:
-      - name: Check out code
-        uses: actions/checkout@v3
-      - name: Create multi-node KinD cluster
-        uses: redhat-chaos/actions/kind@main
-      - name: Install Python
-        uses: actions/setup-python@v4
-        with:
-          python-version: '3.9'
-          architecture: 'x64'
-      - name: Install environment
-        run: |
-          sudo apt-get install build-essential python3-dev
-          pip install --upgrade pip
-          pip install -r requirements.txt
-      - name: Run unit tests
-        run: python -m coverage run -a -m unittest discover -s tests -v
-      - name: Run CI
-        run: |
-          ./CI/run.sh
-          cat ./CI/results.markdown >> $GITHUB_STEP_SUMMARY
-          echo >> $GITHUB_STEP_SUMMARY
-      - name: Upload CI logs
-        uses: actions/upload-artifact@v3
-        with:
-          name: ci-logs
-          path: CI/out
-          if-no-files-found: error
-      - name: Collect coverage report
-        run: |
-          python -m coverage html
-      - name: Publish coverage report to job summary
-        run: |
-          pip install html2text
-          html2text --ignore-images --ignore-links -b 0 htmlcov/index.html >> $GITHUB_STEP_SUMMARY
-      - name: Upload coverage data
-        uses: actions/upload-artifact@v3
-        with:
-          name: coverage
-          path: htmlcov
-          if-no-files-found: error
-      - name: Check CI results
-        run: grep Fail CI/results.markdown && false || true
-      

.github/workflows/functional_tests.yaml

@@ -1,129 +0,0 @@
-on: issue_comment
-
-jobs:
-  check_user:
-    # This job only runs for pull request comments
-    name: Check User Authorization
-    env:
-      USERS: ${{vars.USERS}}
-    if: contains(github.event.comment.body, '/funtest') && contains(github.event.comment.html_url, '/pull/')
-    runs-on: ubuntu-latest
-    steps:
-      - name: Check User
-        run: |
-          for name in `echo $USERS`
-          do 
-            name="${name//$'\r'/}"
-            name="${name//$'\n'/}"
-            if [ $name == "${{github.event.sender.login}}" ]
-            then
-              echo "user ${{github.event.sender.login}} authorized, action started..."
-              exit 0
-            fi
-          done
-          echo "user ${{github.event.sender.login}} is not allowed to run functional tests Action"
-          exit 1
-  pr_commented:
-    # This job only runs for pull request comments containing /functional
-    name: Functional Tests
-    if: contains(github.event.comment.body, '/funtest') && contains(github.event.comment.html_url, '/pull/')
-    runs-on: ubuntu-latest
-    needs:
-      - check_user
-    steps:
-      - name: Check out Kraken
-        uses: actions/checkout@v3
-      - name: Checkout Pull Request
-        run: gh pr checkout ${{ github.event.issue.number }}
-        env:
-          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
-      - name: Install OC CLI
-        uses: redhat-actions/oc-installer@v1
-        with:
-          oc_version: latest
-      - name: Install python 3.9
-        uses: actions/setup-python@v4
-        with:
-          python-version: '3.9'
-      - name: Setup kraken dependencies
-        run:  pip install -r requirements.txt
-      - name: Create Workdir & export the path
-        run: |
-         mkdir workdir
-         echo "WORKDIR_PATH=`pwd`/workdir" >> $GITHUB_ENV
-      - name: Generate run id
-        run: |
-          echo "RUN_ID=`date +%s`" > $GITHUB_ENV
-          echo "Run Id: ${RUN_ID}"
-      - name: Write Pull Secret
-        env:
-          PULLSECRET_BASE64: ${{ secrets.PS_64 }}
-        run: |
-          echo "$PULLSECRET_BASE64" | base64 --decode > pullsecret.txt
-      - name: Write Boot Private Key
-        env:
-          BOOT_KEY: ${{ secrets.CRC_KEY_FILE }}
-        run: |
-          echo -n "$BOOT_KEY" > key.txt
-      - name: Teardown CRC (Post Action)
-        uses: webiny/action-post-run@3.0.0
-        id: post-run-command
-        with:
-          run: podman run --rm -v "${{ github.workspace }}:/workspace:z" -e AWS_ACCESS_KEY_ID="${{ secrets.AWS_ACCESS_KEY_ID }}" -e AWS_SECRET_ACCESS_KEY="${{ secrets.AWS_SECRET_ACCESS_KEY }}" -e AWS_DEFAULT_REGION=us-west-2 quay.io/crcont/crc-cloud:v0.0.2 destroy --project-name "chaos-funtest-${{ env.RUN_ID }}" --backed-url "s3://krkn-crc-state/${{ env.RUN_ID }}" --provider "aws"
-      - name: Create cluster
-        run: |
-          podman run  --name crc-cloud-create --rm \
-          -v ${PWD}:/workspace:z \
-          -e AWS_ACCESS_KEY_ID="${{ secrets.AWS_ACCESS_KEY_ID }}" \
-          -e AWS_SECRET_ACCESS_KEY="${{ secrets.AWS_SECRET_ACCESS_KEY }}" \
-          -e AWS_DEFAULT_REGION="us-west-2" \
-          quay.io/crcont/crc-cloud:v0.0.2 \
-          create aws \
-          --project-name "chaos-funtest-${RUN_ID}" \
-          --backed-url "s3://krkn-crc-state/${RUN_ID}" \
-          --output "/workspace" \
-          --aws-ami-id "ami-00f5eaf98cf42ef9f" \
-          --pullsecret-filepath /workspace/pullsecret.txt \
-          --key-filepath /workspace/key.txt
-
-      - name: Setup kubeconfig
-        continue-on-error: true
-        run: |
-          ssh  -o StrictHostKeyChecking=no -i id_rsa core@$(cat host) "cat /opt/kubeconfig" > kubeconfig
-          sed -i "s/https:\/\/api.crc.testing:6443/https:\/\/`cat host`.nip.io:6443/g" kubeconfig
-          echo "KUBECONFIG=${PWD}/kubeconfig" > $GITHUB_ENV
-
-      - name: Example deployment, GitHub Action env init
-        env:
-          NAMESPACE: test-namespace
-          DEPLOYMENT_NAME: test-nginx
-        run: ./CI/CRC/init_github_action.sh
-      - name: Setup test suite
-        run: |
-          yq -i '.kraken.port="8081"' CI/config/common_test_config.yaml
-          yq -i '.kraken.signal_address="0.0.0.0"' CI/config/common_test_config.yaml
-          yq -i '.kraken.kubeconfig_path="'${KUBECONFIG}'"' CI/config/common_test_config.yaml
-          echo "test_app_outages_gh" > ./CI/tests/my_tests
-          echo "test_container"      >> ./CI/tests/my_tests
-          echo "test_namespace"      >> ./CI/tests/my_tests
-          echo "test_net_chaos"      >> ./CI/tests/my_tests
-          echo "test_time"           >> ./CI/tests/my_tests
-
-      - name: Print affected config files
-        run: |
-          echo -e "## CI/config/common_test_config.yaml\n\n"
-          cat CI/config/common_test_config.yaml
-
-      - name: Running test suite
-        run: |
-          ./CI/run.sh
-      - name: Print test output
-        run:  cat CI/out/*
-      - name: Create coverage report
-        run: |
-          echo "# Test results" > $GITHUB_STEP_SUMMARY
-          cat CI/results.markdown >> $GITHUB_STEP_SUMMARY
-          echo "# Test coverage" >> $GITHUB_STEP_SUMMARY
-          python -m coverage report --format=markdown >> $GITHUB_STEP_SUMMARY
-
-

.github/workflows/tests.yml

@@ -0,0 +1,102 @@
+name: Functional & Unit Tests
+on:
+  pull_request:
+jobs:
+  tests:
+    name: Functional & Unit Tests
+    runs-on: ubuntu-latest
+    steps:
+      - name: Check out code
+        uses: actions/checkout@v3
+      - name: Create multi-node KinD cluster
+        uses: redhat-chaos/actions/kind@main
+      - name: Install Helm & add repos
+        run: |
+          curl https://raw.githubusercontent.com/helm/helm/main/scripts/get-helm-3 | bash
+          helm repo add prometheus-community https://prometheus-community.github.io/helm-charts
+          helm repo add stable https://charts.helm.sh/stable
+          helm repo update
+      - name: Deploy prometheus & Port Forwarding
+        run: |
+          kubectl create namespace prometheus-k8s
+          helm install \
+          --wait --timeout 360s \
+          kind-prometheus \
+          prometheus-community/kube-prometheus-stack \
+          --namespace prometheus-k8s \
+          --set prometheus.service.nodePort=30000 \
+          --set prometheus.service.type=NodePort \
+          --set grafana.service.nodePort=31000 \
+          --set grafana.service.type=NodePort \
+          --set alertmanager.service.nodePort=32000 \
+          --set alertmanager.service.type=NodePort \
+          --set prometheus-node-exporter.service.nodePort=32001 \
+          --set prometheus-node-exporter.service.type=NodePort
+
+          SELECTOR=`kubectl -n prometheus-k8s get service kind-prometheus-kube-prome-prometheus -o wide --no-headers=true | awk '{ print $7 }'`
+          POD_NAME=`kubectl -n prometheus-k8s get pods --selector="$SELECTOR" --no-headers=true | awk '{ print $1 }'`
+          kubectl -n prometheus-k8s port-forward $POD_NAME 9090:9090 &
+          sleep 5
+      - name: Install Python
+        uses: actions/setup-python@v4
+        with:
+          python-version: '3.9'
+          architecture: 'x64'
+      - name: Install environment
+        run: |
+          sudo apt-get install build-essential python3-dev
+          pip install --upgrade pip
+          pip install -r requirements.txt
+#      - name: Run unit tests
+#        run: python -m coverage run -a -m unittest discover -s tests -v
+      - name: Deploy test workloads
+        run: |
+          kubectl apply -f CI/templates/container_scenario_pod.yaml
+          kubectl wait --for=condition=ready pod -l scenario=container --timeout=300s
+          kubectl create namespace namespace-scenario
+          kubectl apply -f CI/templates/time_pod.yaml
+          kubectl wait --for=condition=ready pod -l scenario=time-skew --timeout=300s
+      - name: Get Kind nodes
+        run: |
+          kubectl get nodes --show-labels=true
+
+      - name: Setup Functional Tests
+        run: |
+            yq -i '.kraken.distribution="kubernetes"' CI/config/common_test_config.yaml
+            yq -i '.kraken.port="8081"' CI/config/common_test_config.yaml
+            yq -i '.kraken.signal_address="0.0.0.0"' CI/config/common_test_config.yaml
+            yq -i '.kraken.performance_monitoring="localhost:9090"' CI/config/common_test_config.yaml
+            # echo "test_app_outages_gh" > ./CI/tests/my_tests
+            echo "test_container"      >> ./CI/tests/my_tests
+            echo "test_namespace"      >> ./CI/tests/my_tests
+            echo "test_net_chaos"      >> ./CI/tests/my_tests
+            echo "test_time"           >> ./CI/tests/my_tests
+      - name: Run Functional tests
+        run: |
+          ./CI/run.sh
+          cat ./CI/results.markdown >> $GITHUB_STEP_SUMMARY
+          echo >> $GITHUB_STEP_SUMMARY
+      - name: Run Unit tests
+        run: python -m coverage run -a -m unittest discover -s tests -v
+      - name: Upload CI logs
+        uses: actions/upload-artifact@v3
+        with:
+          name: ci-logs
+          path: CI/out
+          if-no-files-found: error
+      - name: Collect coverage report
+        run: |
+          python -m coverage html
+      - name: Publish coverage report to job summary
+        run: |
+          pip install html2text
+          html2text --ignore-images --ignore-links -b 0 htmlcov/index.html >> $GITHUB_STEP_SUMMARY
+      - name: Upload coverage data
+        uses: actions/upload-artifact@v3
+        with:
+          name: coverage
+          path: htmlcov
+          if-no-files-found: error
+      - name: Check CI results
+        run: grep Fail CI/results.markdown && false || true
+      

CI/CRC/deployment.yaml

@@ -1,44 +0,0 @@
-apiVersion: v1
-kind: Namespace
-metadata:
-  name: $NAMESPACE
-
----
-
-apiVersion: v1
-kind: Service
-metadata:
-  name: $DEPLOYMENT_NAME-service
-  namespace: $NAMESPACE
-spec:
-  selector:
-    app: $DEPLOYMENT_NAME
-  ports:
-    - name: http
-      port: 80
-      targetPort: 8080
-  type: ClusterIP
-
---- 
-
-apiVersion: apps/v1
-kind: Deployment
-metadata:
-  namespace: $NAMESPACE
-  name: $DEPLOYMENT_NAME-deployment
-spec:
-  replicas: 3
-  selector:
-    matchLabels:
-      app: $DEPLOYMENT_NAME
-  template:
-    metadata:
-      labels:
-        app: $DEPLOYMENT_NAME
-    spec:
-      containers:
-        - name: $DEPLOYMENT_NAME
-          image: nginxinc/nginx-unprivileged:stable-alpine
-          ports:
-            - name: http
-              containerPort: 8080

CI/CRC/init_github_action.sh

@@ -1,58 +0,0 @@
-#!/bin/bash
-SCRIPT_PATH=./CI/CRC
-DEPLOYMENT_PATH=$SCRIPT_PATH/deployment.yaml
-
-[[ ! -f $DEPLOYMENT_PATH ]] && echo "[ERROR] please run $0 from GitHub action root directory" && exit 1
-[[ -z $DEPLOYMENT_NAME ]] && echo "[ERROR] please set \$DEPLOYMENT_NAME environment variable" && exit 1
-[[ -z $NAMESPACE ]] && echo "[ERROR] please set \$NAMESPACE environment variable" && exit 1
-
-
-OPENSSL=`which openssl 2>/dev/null`
-[[ $? != 0 ]] && echo "[ERROR]: openssl missing, please install it and try again" && exit 1
-OC=`which oc 2>/dev/null`
-[[ $? != 0 ]] && echo "[ERROR]: oc missing, please install it and try again" && exit 1
-SED=`which sed 2>/dev/null`
-[[ $? != 0 ]] && echo "[ERROR]: sed missing, please install it and try again" && exit 1
-JQ=`which jq 2>/dev/null`
-[[ $? != 0 ]] && echo "[ERROR]: jq missing, please install it and try again" && exit 1
-ENVSUBST=`which envsubst 2>/dev/null`
-[[ $? != 0 ]] && echo "[ERROR]: envsubst missing, please install it and try again" && exit 1
-
-
-API_PORT="6443"
-API_ADDRESS="https://api.`cat host`.nip.io:${API_PORT}"
-FQN=$DEPLOYMENT_NAME.apps.$API_ADDRESS
-
-
-echo "[INF] deploying example deployment: $DEPLOYMENT_NAME in namespace: $NAMESPACE"
-$ENVSUBST < $DEPLOYMENT_PATH | $OC apply -f - > /dev/null 2>&1
-
-echo "[INF] creating SSL self-signed certificates for route https://$FQN"
-$OPENSSL genrsa -out servercakey.pem > /dev/null 2>&1
-$OPENSSL req -new -x509 -key servercakey.pem -out serverca.crt -subj "/CN=$FQN/O=Red Hat Inc./C=US" > /dev/null 2>&1
-$OPENSSL genrsa -out server.key > /dev/null 2>&1
-$OPENSSL req -new -key server.key -out server_reqout.txt -subj "/CN=$FQN/O=Red Hat Inc./C=US" > /dev/null 2>&1
-$OPENSSL x509 -req -in server_reqout.txt -days 3650 -sha256 -CAcreateserial -CA serverca.crt -CAkey servercakey.pem -out server.crt > /dev/null 2>&1
-echo "[INF] creating deployment: $DEPLOYMENT_NAME public route: https://$FQN"
-$OC create route --namespace $NAMESPACE  edge --service=$DEPLOYMENT_NAME-service --cert=server.crt --key=server.key --ca-cert=serverca.crt --hostname="$FQN" > /dev/null 2>&1
-
-
-echo "[INF] setting github action environment variables"
-
-NODE_NAME="`$OC get nodes  -o json | $JQ -r '.items[0].metadata.name'`"
-COVERAGE_FILE="`pwd`/coverage.md"
-echo "DEPLOYMENT_NAME=$DEPLOYMENT_NAME" >> $GITHUB_ENV
-echo "DEPLOYMENT_FQN=$FQN" >> $GITHUB_ENV
-echo "API_ADDRESS=$API_ADDRESS" >> $GITHUB_ENV
-echo "API_PORT=$API_PORT" >> $GITHUB_ENV
-echo "NODE_NAME=$NODE_NAME" >> $GITHUB_ENV
-echo "NAMESPACE=$NAMESPACE" >> $GITHUB_ENV
-echo "COVERAGE_FILE=$COVERAGE_FILE" >> $GITHUB_ENV
-
-echo "[INF] deployment fully qualified name will be available in \${{ env.DEPLOYMENT_NAME }} with value $DEPLOYMENT_NAME"
-echo "[INF] deployment name will be available in \${{ env.DEPLOYMENT_FQN }} with value $FQN"
-echo "[INF] OCP API address will be available in \${{ env.API_ADDRESS }} with value $API_ADDRESS"
-echo "[INF] OCP API port will be available in \${{ env.API_PORT }} with value $API_PORT"
-echo "[INF] OCP node name will be available in \${{ env.NODE_NAME }} with value $NODE_NAME"
-echo "[INF] coverage file will ve available in \${{ env.COVERAGE_FILE }} with value $COVERAGE_FILE"
-

CI/run.sh

@@ -1,15 +1,14 @@
 #!/bin/bash
-set -x
 MAX_RETRIES=60
 
-OC=`which oc 2>/dev/null`
-[[ $? != 0 ]] && echo "[ERROR]: oc missing, please install it and try again" && exit 1
+KUBECTL=`which kubectl 2>/dev/null`
+[[ $? != 0 ]] && echo "[ERROR]: kubectl missing, please install it and try again" && exit 1
 
 wait_cluster_become_ready() {
   COUNT=1
-  until `$OC get namespace > /dev/null 2>&1`
+  until `$KUBECTL get namespace > /dev/null 2>&1`
   do
-    echo "[INF] waiting OpenShift to become ready, after $COUNT check"
+    echo "[INF] waiting Kubernetes to become ready, after $COUNT check"
     sleep 3
     [[ $COUNT == $MAX_RETRIES ]] && echo "[ERR] max retries exceeded, failing" && exit 1
     ((COUNT++))
@@ -20,7 +19,7 @@ wait_cluster_become_ready() {
 
 ci_tests_loc="CI/tests/my_tests"
 
-echo "running test suit consisting of ${ci_tests}"
+echo -e "********* Running Functional Tests Suite *********\n\n"
 
 rm -rf CI/out
 
@@ -37,9 +36,31 @@ echo 'Test                   | Result | Duration' >> $results
 echo '-----------------------|--------|---------' >> $results
 
 # Run each test
+failed_tests=()
 for test_name in `cat CI/tests/my_tests`
 do
   wait_cluster_become_ready
-  ./CI/run_test.sh $test_name $results
+  return_value=`./CI/run_test.sh $test_name $results`
+  if [[ $return_value == 1 ]]
+  then
+    echo "Failed"
+    failed_tests+=("$test_name")
+  fi
   wait_cluster_become_ready
 done
+
+if (( ${#failed_tests[@]}>0 ))
+then
+  echo -e "\n\n======================================================================"
+  echo -e "\n     FUNCTIONAL TESTS FAILED  ${failed_tests[*]} ABORTING"
+  echo -e "\n======================================================================\n\n"
+
+  for test in "${failed_tests[@]}"
+  do
+    echo -e "\n********** $test KRKN RUN OUTPUT **********\n"
+    cat "CI/out/$test.out"
+    echo -e "\n********************************************\n\n\n\n"
+  done
+
+  exit 1
+fi

CI/run_test.sh

@@ -1,5 +1,4 @@
 #!/bin/bash
-set -x
 readonly SECONDS_PER_HOUR=3600
 readonly SECONDS_PER_MINUTE=60
 function get_time_format() {
@@ -14,9 +13,7 @@ ci_test=`echo $1`
 
 results_file=$2
 
-echo -e "\n======================================================================"
-echo -e "     CI test for ${ci_test}                    "
-echo -e "======================================================================\n"
+echo -e "test: ${ci_test}" >&2
 
 ci_results="CI/out/$ci_test.out"
 # Test ci
@@ -28,13 +25,16 @@ then
   # if the test passes update the results and complete
   duration=$SECONDS
   duration=$(get_time_format $duration)
-  echo "$ci_test: Successful"
+  echo -e  "> $ci_test: Successful\n" >&2
   echo "$ci_test | Pass | $duration" >> $results_file
   count=$retries
+  # return value for run.sh
+  echo 0
 else
   duration=$SECONDS
   duration=$(get_time_format $duration)
-  echo "$ci_test: Failed"
+  echo -e "> $ci_test: Failed\n" >&2
   echo "$ci_test | Fail | $duration" >> $results_file
-  echo "Logs for "$ci_test
+  # return value for run.sh
+  echo 1
 fi

CI/scenarios/container_scenario.yml

@@ -1,8 +1,8 @@
 scenarios:
-- name: "kill machine config container"
-  namespace: "openshift-machine-config-operator"
-  label_selector: "k8s-app=machine-config-server"
-  container_name: "hello-openshift"
-  action: "kill 1"
+- name: "kill test container"
+  namespace: "default"
+  label_selector: "scenario=container"
+  container_name: "fedtools"
+  action: 1
   count: 1
   retry_wait: 60

CI/scenarios/network_chaos.yaml

@@ -1,6 +1,8 @@
 network_chaos:                   # Scenario to create an outage by simulating random variations in the network.
   duration: 10                   # seconds
   instance_count: 1
+  node_name: kind-worker2
   execution: serial
   egress:
     bandwidth: 100mbit
+

CI/scenarios/network_diagnostics_namespace.yaml

@@ -1,6 +1,6 @@
 scenarios:
   - action: delete
-    namespace: "^$openshift-network-diagnostics$"
+    namespace: "^namespace-scenario$"
     label_selector:
     runs: 1
     sleep: 15

CI/scenarios/time_scenarios.yml

@@ -1,5 +1,5 @@
 time_scenarios:
 - action: skew_time
   object_type: pod
-  label_selector: k8s-app=etcd
+  label_selector: scenario=time-skew
   container_name: ""

CI/templates/container_scenario_pod.yaml

@@ -0,0 +1,16 @@
+apiVersion: v1
+kind: Pod
+metadata:
+  name: container
+  labels:
+    scenario: container
+spec:
+  hostNetwork: true
+  containers:
+  - name: fedtools
+    image: docker.io/fedora/tools
+    command:
+    - /bin/sh
+    - -c
+    - |
+      sleep infinity

CI/templates/time_pod.yaml

@@ -0,0 +1,16 @@
+apiVersion: v1
+kind: Pod
+metadata:
+  name: time-skew
+  labels:
+    scenario: time-skew
+spec:
+  hostNetwork: true
+  containers:
+  - name: fedtools
+    image: docker.io/fedora/tools
+    command:
+    - /bin/sh
+    - -c
+    - |
+      sleep infinity

CI/tests/common.sh

@@ -13,6 +13,6 @@ function error {
 }
 
 function get_node {
-  worker_node=$(oc get nodes --no-headers | grep worker | head -n 1)
+  worker_node=$(kubectl get nodes --no-headers | grep worker | head -n 1)
   export WORKER_NODE=$worker_node
 }