feat: RunnerSet backed by StatefulSet (#629)

* feat: RunnerSet backed by StatefulSet

Unlike a runner deployment, a runner set can manage a set of stateful runners by combining a statefulset and an admission webhook that mutates statefulset-managed pods with required envvars and registration tokens.

Resolves #613
Ref #612

* Upgrade controller-runtime to 0.9.0

* Bump Go to 1.16.x following controller-runtime 0.9.0

* Upgrade kubebuilder to 2.3.2 for updated etcd and apiserver following local setup

* Fix startup failure due to missing LeaderElectionID

* Fix the issue that any pods become unable to start once actions-runner-controller got failed after the mutating webhook has been registered

* Allow force-updating statefulset

* Fix runner container missing work and certs-client volume mounts and DOCKER_HOST and DOCKER_TLS_VERIFY envvars when dockerdWithinRunner=false

* Fix runnerset-controller not applying statefulset.spec.template.spec changes when there were no changes in runnerset spec

* Enable running acceptance tests against arbitrary kind cluster

* RunnerSet supports non-ephemeral runners only today

* fix: docker-build from root Makefile on intel mac

* fix: arch check fixes for mac and ARM

* ci: aligning test data format and patching checks

* fix: removing namespace in test data

* chore: adding more ignores

* chore: removing leading space in shebang

* Re-add metrics to org hra testdata

* Bump cert-manager to v1.1.1 and fix deploy.sh

Co-authored-by: toast-gear <15716903+toast-gear@users.noreply.github.com>
Co-authored-by: Callum James Tait <callum.tait@photobox.com>

acceptance/checks.sh

@@ -1,32 +1,84 @@
 #!/usr/bin/env bash
 
-set -e
+set +e
 
-runner_name=
+repo_runnerdeployment_passed="skipped"
+repo_runnerset_passed="skipped"
 
-while [ -z "${runner_name}" ]; do
-  echo Finding the runner... 1>&2
-  sleep 1
-  runner_name=$(kubectl get runner --output=jsonpath="{.items[*].metadata.name}")
-done
+echo "Checking if RunnerDeployment repo test is set"
+if [ "${TEST_REPO}" ] && [ ! "${USE_RUNNERSET}" ]; then
+  runner_name=
+  count=0
+  while [ $count -le 30 ]; do
+    echo "Finding Runner ..."
+    runner_name=$(kubectl get runner --output=jsonpath="{.items[*].metadata.name}")
+    if [ "${runner_name}" ]; then
+      while [ $count -le 30 ]; do
+        runner_pod_name=
+        echo "Found Runner \""${runner_name}"\""
+        echo "Finding underlying pod ..."
+        runner_pod_name=$(kubectl get pod --output=jsonpath="{.items[*].metadata.name}" | grep ${runner_name})
+        if [ "${runner_pod_name}" ]; then
+          echo "Found underlying pod \""${runner_pod_name}"\""
+          echo "Waiting for pod \""${runner_pod_name}"\" to become ready..."
+          kubectl wait pod/${runner_pod_name} --for condition=ready --timeout 270s
+          break 2
+        fi
+        sleep 1
+        let "count=count+1"
+      done
+    fi
+    sleep 1
+    let "count=count+1"
+  done
+  if [ $count -ge 30 ]; then
+    repo_runnerdeployment_passed=false
+  else
+    repo_runnerdeployment_passed=true
+  fi
+echo "Checking if RunnerSet repo test is set"
+elif [ "${TEST_REPO}" ] && [ "${USE_RUNNERSET}" ]; then
+  runnerset_name=
+  count=0
+  while [ $count -le 30 ]; do
+    echo "Finding RunnerSet ..."
+    runnerset_name=$(kubectl get runnerset --output=jsonpath="{.items[*].metadata.name}")
+    if [ "${runnerset_name}" ]; then
+      while [ $count -le 30 ]; do
+        runnerset_pod_name=
+        echo "Found RunnerSet \""${runnerset_name}"\""
+        echo "Finding underlying pod ..."
+        runnerset_pod_name=$(kubectl get pod --output=jsonpath="{.items[*].metadata.name}" | grep ${runnerset_name})
+        echo "BEFORE IF"
+        if [ "${runnerset_pod_name}" ]; then
+          echo "AFTER IF"
+          echo "Found underlying pod \""${runnerset_pod_name}"\""
+          echo "Waiting for pod \""${runnerset_pod_name}"\" to become ready..."
+          kubectl wait pod/${runnerset_pod_name} --for condition=ready --timeout 270s
+          break 2
+        fi
+      sleep 1
+      let "count=count+1"
+      done
+    fi
+    sleep 1
+    let "count=count+1"
+  done
+  if [ $count -ge 30 ]; then
+    repo_runnerset_passed=false
+  else
+    repo_runnerset_passed=true
+  fi
+fi
 
-echo Found runner ${runner_name}.
-
-# Wait a bit to make sure the runner pod is created before looking for it.
-sleep 2
-
-pod_name=
-
-while [ -z "${pod_name}" ]; do
-  echo Finding the runner pod... 1>&2
-  sleep 1
-  pod_name=$(kubectl get pod --output=jsonpath="{.items[*].metadata.name}" | grep ${runner_name})
-done
-
-echo Found pod ${pod_name}.
-
-echo Waiting for pod ${runner_name} to become ready... 1>&2
-
-kubectl wait pod/${runner_name} --for condition=ready --timeout 270s
-
-echo All tests passed. 1>&2
+if [ ${repo_runnerset_passed} == true ] || [ ${repo_runnerset_passed} == "skipped" ] && \
+   [ ${repo_runnerdeployment_passed} == true ] || [ ${repo_runnerdeployment_passed} == "skipped" ]; then
+  echo "INFO : All tests passed or skipped"
+  echo "RunnerSet Repo Test Status : ${repo_runnerset_passed}"
+  echo "RunnerDeployment Repo Test Status : ${repo_runnerdeployment_passed}"
+else
+  echo "ERROR : Some tests failed"
+  echo "RunnerSet Repo Test Status : ${repo_runnerset_passed}"
+  echo "RunnerDeployment Repo Test Status : ${repo_runnerdeployment_passed}"
+  exit 1
+fi

acceptance/deploy.sh

@@ -35,6 +35,7 @@ if [ "${tool}" == "helm" ]; then
     --set image.repository=${NAME} \
     --set image.tag=${VERSION} \
     -f ${VALUES_FILE}
+  kubectl apply -f charts/actions-runner-controller/crds
   kubectl -n actions-runner-system wait deploy/actions-runner-controller --for condition=available --timeout 60s
 else
   kubectl apply \
@@ -47,8 +48,13 @@ fi
 sleep 20
 
 if [ -n "${TEST_REPO}" ]; then
-  cat acceptance/testdata/runnerdeploy.yaml | envsubst | kubectl apply -f -
-  cat acceptance/testdata/hra.yaml | envsubst | kubectl apply -f -
+  if [ -n "USE_RUNNERSET" ]; then
+      cat acceptance/testdata/repo.runnerset.yaml | envsubst | kubectl apply -f -
+  else
+    echo 'Deploying runnerdeployment and hra. Set USE_RUNNERSET if you want to deploy runnerset instead.'
+    cat acceptance/testdata/repo.runnerdeploy.yaml | envsubst | kubectl apply -f -
+    cat acceptance/testdata/repo.hra.yaml | envsubst | kubectl apply -f -
+  fi
 else
   echo 'Skipped deploying runnerdeployment and hra. Set TEST_REPO to "yourorg/yourrepo" to deploy.'
 fi

acceptance/testdata/org.hra.yaml

@@ -24,6 +24,7 @@ spec:
     minReplicas: 0
   minReplicas: 0
   maxReplicas: 5
+  # Used to test that HRA is working for org runners
   metrics:
   - type: PercentageRunnersBusy
     scaleUpThreshold: '0.75'

acceptance/testdata/repo.runnerset.yaml

@@ -0,0 +1,51 @@
+apiVersion: actions.summerwind.dev/v1alpha1
+kind: RunnerSet
+metadata:
+  name: example-runnerset
+spec:
+  # MANDATORY because it is based on StatefulSet: Results in a below error when omitted:
+  #   missing required field "selector" in dev.summerwind.actions.v1alpha1.RunnerSet.spec
+  selector:
+    matchLabels:
+      app: example-runnerset
+
+  # MANDATORY because it is based on StatefulSet: Results in a below error when omitted:
+  # missing required field "serviceName" in dev.summerwind.actions.v1alpha1.RunnerSet.spec]
+  serviceName: example-runnerset
+
+  #replicas: 1
+  ephemeral: false
+
+  repository: ${TEST_REPO}
+  #
+  # Custom runner image
+  #
+  image: ${RUNNER_NAME}:${RUNNER_TAG}
+  #
+  # dockerd within runner container
+  #
+  ## Replace `mumoshu/actions-runner-dind:dev` with your dind image
+  #dockerdWithinRunnerContainer: true
+  #
+  # Set the MTU used by dockerd-managed network interfaces (including docker-build-ubuntu)
+  #
+  #dockerMTU: 1450
+  #Runner group
+  # labels:
+  # - "mylabel 1"
+  # - "mylabel 2"
+
+  #
+  # Non-standard working directory
+  #
+  # workDir: "/"
+  template:
+    metadata:
+      labels:
+        app: example-runnerset
+    spec:
+      containers:
+      - name: runner
+        imagePullPolicy: IfNotPresent
+      #- name: docker
+      #  #image: mumoshu/actions-runner-dind:dev