kubernetes · vishnuchalla · May 19, 2026
diff --git a/tests/e2e/scenarios/scalability/run-kube-burner-test.sh b/tests/e2e/scenarios/scalability/run-kube-burner-test.sh
@@ -0,0 +1,253 @@
+#!/usr/bin/env bash
+
+# Copyright 2026 The Kubernetes Authors.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+set -e
+set -x
+
+make test-e2e-install
+
+REPO_ROOT=$(git rev-parse --show-toplevel)
+if [[ -z "${K8S_VERSION:-}" ]]; then
+  K8S_VERSION=https://storage.googleapis.com/k8s-release-dev/ci/latest.txt
+fi
+
+# Default cloud provider to aws
+if [[ -z "${CLOUD_PROVIDER:-}" ]]; then
+  CLOUD_PROVIDER="aws"
+fi
+echo "CLOUD_PROVIDER=${CLOUD_PROVIDER}"
+if [[ -z "${ADMIN_ACCESS:-}" ]]; then
+  ADMIN_ACCESS="0.0.0.0/0" # Or use your IPv4 with /32
+fi
+echo "ADMIN_ACCESS=${ADMIN_ACCESS}"
+
+KOPS_SCHEDULER_QPS="${KOPS_SCHEDULER_QPS:-500}"
+KOPS_SCHEDULER_BURST="${KOPS_SCHEDULER_BURST:-500}"
+KOPS_CONTROLLER_MANAGER_QPS="${KOPS_CONTROLLER_MANAGER_QPS:-500}"
+KOPS_CONTROLLER_MANAGER_BURST="${KOPS_CONTROLLER_MANAGER_BURST:-500}"
+KOPS_APISERVER_MAX_REQUESTS_INFLIGHT="${KOPS_APISERVER_MAX_REQUESTS_INFLIGHT:-640}"
+ETCD_QUOTA_BACKEND_BYTES="${ETCD_QUOTA_BACKEND_BYTES:-8589934592}"
+echo "KOPS_SCHEDULER_QPS=${KOPS_SCHEDULER_QPS} KOPS_SCHEDULER_BURST=${KOPS_SCHEDULER_BURST}"
+echo "KOPS_CONTROLLER_MANAGER_QPS=${KOPS_CONTROLLER_MANAGER_QPS} KOPS_CONTROLLER_MANAGER_BURST=${KOPS_CONTROLLER_MANAGER_BURST}"
+echo "KOPS_APISERVER_MAX_REQUESTS_INFLIGHT=${KOPS_APISERVER_MAX_REQUESTS_INFLIGHT}"
+
+# cilium does not yet pass conformance tests (shared hostport test)
+#create_args="--networking cilium"
+create_args=()
+if [[ "${CLOUD_PROVIDER}" == "aws" ]]; then
+  create_args+=("--network-cidr=10.0.0.0/16,10.1.0.0/16,10.2.0.0/16,10.3.0.0/16,10.4.0.0/16,10.5.0.0/16,10.6.0.0/16,10.7.0.0/16,10.8.0.0/16,10.9.0.0/16,10.10.0.0/16,10.11.0.0/16,10.12.0.0/16")
+  create_args+=("--node-size=${NODE_SIZE:-t3a.medium,t3.medium,t3a.large,c5a.large,t3.large,c5.large,m5a.large,m6a.large,m5.large,c7a.large,r5a.large,r6a.large,m7a.large}")
+  create_args+=("--node-volume-size=20")
+  create_args+=("--control-plane-volume-size=500")
+  create_args+=("--zones=us-east-2a,us-east-2b,us-east-2c")
+  create_args+=("--image=${INSTANCE_IMAGE:-ssm:/aws/service/canonical/ubuntu/server/24.04/stable/current/amd64/hvm/ebs-gp3/ami-id}")
+  # TODO: track failures of tests (HostPort & OIDC) when using `--dns=none`
+  create_args+=("--dns=none")
+fi
+if [[ "${CLOUD_PROVIDER}" == "gce" ]]; then
+  create_args+=("--zones=us-east1-b,us-east1-c,us-east1-d")
+  create_args+=("--node-size=${NODE_SIZE:-e2-medium}")
+  create_args+=("--node-volume-size=30")
+  create_args+=("--control-plane-volume-size=1000")
+  create_args+=("--gce-service-account=default")
+  create_args+=("--topology=private")
+  create_args+=("--image=${INSTANCE_IMAGE:-ubuntu-os-cloud/ubuntu-2404-noble-amd64-v20251001}")
+  create_args+=("--set spec.networking.podCIDR=10.64.0.0/10")
+  create_args+=("--set spec.networking.subnets[0].cidr=10.128.0.0/15")
+  create_args+=("--set spec.networking.serviceClusterIPRange=10.130.0.0/15")
+  create_args+=("--set spec.networking.nonMasqueradeCIDR=10.64.0.0/10")
+  create_args+=("--set spec.etcdClusters[*].etcdMembers[*].volumeIOPS=10000")
+  create_args+=("--set spec.etcdClusters[*].etcdMembers[*].volumeThroughput=1000")
+  create_args+=("--set spec.etcdClusters[*].etcdMembers[*].volumeSize=120")
+  create_args+=("--set spec.etcdClusters[*].etcdMembers[*].volumeType=hyperdisk-balanced")
+fi
+create_args+=("--networking=${CNI_PLUGIN:-calico}")
+if [[ "${CNI_PLUGIN}" == "amazonvpc" ]]; then
+  create_args+=("--set spec.networking.amazonVPC.env=ENABLE_PREFIX_DELEGATION=true")
+fi
+create_args+=("--set spec.etcdClusters[0].manager.listenMetricsURLs=http://localhost:2382")
+create_args+=("--set spec.etcdClusters[*].manager.env=ETCD_QUOTA_BACKEND_BYTES=${ETCD_QUOTA_BACKEND_BYTES}")
+create_args+=("--set spec.etcdClusters[*].manager.env=ETCD_ENABLE_PPROF=true")
+create_args+=("--set spec.cloudControllerManager.concurrentNodeSyncs=10")
+create_args+=("--set spec.kubelet.maxPods=96")
+create_args+=("--set spec.kubelet.kubeAPIQPS=100")
+create_args+=("--set spec.kubeScheduler.authorizationAlwaysAllowPaths=/healthz")
+create_args+=("--set spec.kubeScheduler.authorizationAlwaysAllowPaths=/livez")
+create_args+=("--set spec.kubeScheduler.authorizationAlwaysAllowPaths=/readyz")
+create_args+=("--set spec.kubeScheduler.authorizationAlwaysAllowPaths=/metrics")
+create_args+=("--set spec.kubeScheduler.qps=${KOPS_SCHEDULER_QPS}")
+create_args+=("--set spec.kubeScheduler.burst=${KOPS_SCHEDULER_BURST}")
+create_args+=("--set spec.kubeScheduler.enableProfiling=true")
+create_args+=("--set spec.kubeScheduler.enableContentionProfiling=true")
+create_args+=("--set spec.kubeControllerManager.endpointUpdatesBatchPeriod=500ms")
+create_args+=("--set spec.kubeControllerManager.endpointSliceUpdatesBatchPeriod=500ms")
+create_args+=("--set spec.kubeControllerManager.kubeAPIQPS=${KOPS_CONTROLLER_MANAGER_QPS}")
+create_args+=("--set spec.kubeControllerManager.kubeAPIBurst=${KOPS_CONTROLLER_MANAGER_BURST}")
+create_args+=("--set spec.kubeControllerManager.enableProfiling=true")
+create_args+=("--set spec.kubeControllerManager.enableContentionProfiling=true")
+# inflight requests are bit higher than what currently upstream uses for GCE scale tests
+create_args+=("--set spec.kubeAPIServer.maxRequestsInflight=${KOPS_APISERVER_MAX_REQUESTS_INFLIGHT}")
+create_args+=("--set spec.kubeAPIServer.maxMutatingRequestsInflight=0")
+create_args+=("--set spec.kubeAPIServer.enableProfiling=true")
+create_args+=("--set spec.kubeAPIServer.enableContentionProfiling=true")
+create_args+=("--set spec.kubeAPIServer.logLevel=3")
+create_args+=("--set spec.kubeAPIServer.deleteCollectionWorkers=16")
+create_args+=("--set spec.kubeAPIServer.compactionInterval=150s")
+
+# this is required for Prometheus server to scrape metrics endpoint on APIServer
+create_args+=("--set spec.kubeAPIServer.anonymousAuth=true")
+# this is required for kindnet to use nftables
+create_args+=("--set spec.kubeProxy.proxyMode=${KUBE_PROXY_MODE:-iptables}")
+# this is required for prometheus to scrape kube-proxy metrics endpoint
+create_args+=("--set spec.kubeProxy.metricsBindAddress=0.0.0.0:10249")
+# bump coredns memory on large clusters
+if [[ "${KUBE_NODE_COUNT:-100}" -ge 5000 ]]; then
+  create_args+=("--set spec.kubeDNS.memoryRequest=340Mi")
+  create_args+=("--set spec.kubeDNS.memoryLimit=340Mi")
+fi
+create_args+=("--node-count=${KUBE_NODE_COUNT:-100}")
+create_args+=("--control-plane-count=${CONTROL_PLANE_COUNT:-1}")
+create_args+=("--control-plane-size=${CONTROL_PLANE_SIZE:-c5.2xlarge}")
+
+# Enable HTTP for events etcd to reduce TLS overhead in scale tests
+KOPS_FEATURE_FLAGS="EtcdEventsHTTP,${KOPS_FEATURE_FLAGS:-}"
+
+# AWS ONLY feature flags
+if [[ "${CLOUD_PROVIDER}" == "aws" ]]; then
+  # AWS doesn't run dedicated addons node
+  export CL2_PROMETHEUS_TOLERATE_MASTER="true"
+  # Enable creating a single nodes instance group
+  KOPS_FEATURE_FLAGS="AWSSingleNodesInstanceGroup,${KOPS_FEATURE_FLAGS:-}"
+  create_args+=("--set spec.etcdClusters[*].etcdMembers[*].volumeIOPS=10000")
+  create_args+=("--set spec.etcdClusters[*].etcdMembers[*].volumeSize=120")
+  create_args+=("--set spec.etcdClusters[*].etcdMembers[*].volumeType=io2")
+
+fi
+echo "KOPS_FEATURE_FLAGS=${KOPS_FEATURE_FLAGS}"
+
+# Note that these arguments for kubetest2
+KUBETEST2_ARGS=()
+KUBETEST2_ARGS+=("-v=2")
+KUBETEST2_ARGS+=("--max-nodes-to-dump=${MAX_NODES_TO_DUMP:-5}")
+KUBETEST2_ARGS+=("--cloud-provider=${CLOUD_PROVIDER}")
+KUBETEST2_ARGS+=("--cluster-name=${CLUSTER_NAME:-}")
+KUBETEST2_ARGS+=("--admin-access=${ADMIN_ACCESS:-}")
+KUBETEST2_ARGS+=("--env=KOPS_FEATURE_FLAGS=${KOPS_FEATURE_FLAGS}")
+KUBETEST2_ARGS+=("--pre-test-cmd=${REPO_ROOT}/tests/e2e/scenarios/scalability/pre-test.sh")
+if [[ -n "${KUBE_FEATURE_GATES:-}" ]]; then
+  KUBETEST2_ARGS+=("--kubernetes-feature-gates=${KUBE_FEATURE_GATES}")
+fi
+if [[ "${JOB_TYPE}" == "presubmit" && "${REPO_OWNER}/${REPO_NAME}" == "kubernetes/kops" ]]; then
+  KUBETEST2_ARGS+=("--build")
+  KUBETEST2_ARGS+=("--kops-binary-path=${GOPATH}/src/k8s.io/kops/.build/dist/linux/$(go env GOARCH)/kops")
+elif [[ "${JOB_TYPE}" == "presubmit" && "${REPO_OWNER}/${REPO_NAME}" == "kubernetes/kubernetes" ]]; then
+  KUBETEST2_ARGS+=("--build")
+  KUBETEST2_ARGS+=("--build-kubernetes=true")
+  # Scale clusters run linux/amd64 nodes only, so build a single arch instead of
+  # the full multi-arch release matrix.
+  KUBETEST2_ARGS+=("--target-build-arch=linux/amd64")
+  KUBETEST2_ARGS+=("--kops-version-marker=${KOPS_VERSION_MARKER:-https://storage.googleapis.com/k8s-staging-kops/kops/releases/markers/master/latest-ci.txt}")
+  cd "${GOPATH}/src/k8s.io/kubernetes"
+else
+  KUBETEST2_ARGS+=("--kops-version-marker=${KOPS_VERSION_MARKER:-https://storage.googleapis.com/k8s-staging-kops/kops/releases/markers/master/latest-ci.txt}")
+fi
+
+if [[ "${CLOUD_PROVIDER}" == "gce" ]]; then
+  if [[ -n "${GCP_PROJECT:-}" ]]; then
+    KUBETEST2_ARGS+=("--gcp-project=${GCP_PROJECT}")
+  else
+    KUBETEST2_ARGS+=("--boskos-resource-type=${BOSKOS_RESOURCE_TYPE:-scalability-project}")
+  fi
+  KUBETEST2_ARGS+=("--control-plane-instance-group-overrides=spec.rootVolume.type=hyperdisk-balanced")
+  KUBETEST2_ARGS+=("--control-plane-instance-group-overrides=spec.rootVolume.iops=10000")
+  KUBETEST2_ARGS+=("--control-plane-instance-group-overrides=spec.rootVolume.throughput=1000")
+  KUBETEST2_ARGS+=("--control-plane-instance-group-overrides=spec.associatePublicIP=true")
+elif [[ "${CLOUD_PROVIDER}" == "aws" ]]; then
+  KUBETEST2_ARGS+=("--control-plane-instance-group-overrides=spec.rootVolume.type=io2")
+fi
+
+# More time for bigger clusters
+KUBETEST2_ARGS+=("--validation-wait=75m")
+KUBETEST2_ARGS+=("--validation-count=3")
+KUBETEST2_ARGS+=("--validation-interval=60s")
+
+# The caller can set DELETE_CLUSTER=false to stop us deleting the cluster
+if [[ -z "${DELETE_CLUSTER:-}" ]]; then
+  DELETE_CLUSTER="true"
+fi
+
+if [[ "${DELETE_CLUSTER:-}" == "true" ]]; then
+  KUBETEST2_ARGS+=("--down")
+fi
+
+export PROMETHEUS_KUBE_PROXY_SELECTOR_KEY="k8s-app"
+
+# Download and install kube-burner if path not already specified
+if [[ -z "${KUBE_BURNER_PATH:-}" ]]; then
+  if [[ -z "${KUBE_BURNER_VERSION:-}" ]]; then
+    KUBE_BURNER_VERSION=$(curl -fsSL https://api.github.com/repos/kube-burner/kube-burner/releases/latest | grep -o '"tag_name": "v[^"]*"' | sed 's/"tag_name": "v\(.*\)"/\1/')
+  fi
+  KUBE_BURNER_OS=$(uname -s | tr '[:upper:]' '[:lower:]')
+  KUBE_BURNER_ARCH=$(uname -m)
+  if [[ "${KUBE_BURNER_ARCH}" == "aarch64" ]]; then
+    KUBE_BURNER_ARCH="arm64"
+  fi
+  KUBE_BURNER_DIR=$(mktemp -d)
+  KUBE_BURNER_TARBALL="kube-burner-V${KUBE_BURNER_VERSION}-${KUBE_BURNER_OS}-${KUBE_BURNER_ARCH}.tar.gz"
+  KUBE_BURNER_URL="https://github.com/kube-burner/kube-burner/releases/download/v${KUBE_BURNER_VERSION}/${KUBE_BURNER_TARBALL}"
+  echo "Downloading kube-burner ${KUBE_BURNER_VERSION} from ${KUBE_BURNER_URL}"
+  curl -fsSL "${KUBE_BURNER_URL}" -o "${KUBE_BURNER_DIR}/${KUBE_BURNER_TARBALL}"
+  tar -xzf "${KUBE_BURNER_DIR}/${KUBE_BURNER_TARBALL}" -C "${KUBE_BURNER_DIR}"
+  KUBE_BURNER_PATH="${KUBE_BURNER_DIR}/kube-burner"
+  chmod +x "${KUBE_BURNER_PATH}"
+  echo "kube-burner ${KUBE_BURNER_VERSION} installed at ${KUBE_BURNER_PATH}"
+fi
+
+KUBE_BURNER_ARGS=()
+KUBE_BURNER_ARGS+=("--workdir=${KUBE_BURNER_WORKDIR:-k8s.io/perf-tests}")
+KUBE_BURNER_ARGS+=("--workload=${KUBE_BURNER_WORKLOAD}")
+KUBE_BURNER_ARGS+=("--kube-burner-path=${KUBE_BURNER_PATH}")
+if [[ -n "${KUBE_BURNER_UUID:-}" ]]; then
+  KUBE_BURNER_ARGS+=("--uuid=${KUBE_BURNER_UUID}")
+fi
+if [[ "${KUBE_BURNER_SKIP_TLS_VERIFY:-}" == "true" ]]; then
+  KUBE_BURNER_ARGS+=("--skip-tls-verify")
+fi
+if [[ -n "${KUBE_BURNER_KUBECONFIG:-}" ]]; then
+  KUBE_BURNER_ARGS+=("--kubeconfig=${KUBE_BURNER_KUBECONFIG}")
+fi
+if [[ -n "${KUBE_BURNER_LOG_LEVEL:-}" ]]; then
+  KUBE_BURNER_ARGS+=("--log-level=${KUBE_BURNER_LOG_LEVEL}")
+fi
+if [[ -n "${KUBE_BURNER_EXTRA_ARGS:-}" ]]; then
+  KUBE_BURNER_ARGS+=("--extra-args=${KUBE_BURNER_EXTRA_ARGS}")
+fi
+
+kubetest2 kops "${KUBETEST2_ARGS[@]}" \
+  --up \
+  --kubernetes-version="${K8S_VERSION}" \
+  --create-args="${create_args[*]}" \
+  --test=kube-burner \
+  -- \
+  "${KUBE_BURNER_ARGS[@]}"
+
+if [[ -n "${KUBE_BURNER_REPORT_DIR:-}" ]]; then
+  mkdir -p "${ARTIFACTS}/${KUBE_BURNER_REPORT_DIR}"
+  if ls collected-metrics* 1>/dev/null 2>&1; then
+    mv collected-metrics* "${ARTIFACTS}/${KUBE_BURNER_REPORT_DIR}/"
+    echo "Kube-burner locally indexed metrics moved to ${ARTIFACTS}/${KUBE_BURNER_REPORT_DIR}"
+  fi
+fi