Add Support for Simulating Error Conditions and Handling by VPA & Kruize

monitoring/local_monitoring/common.sh

@@ -350,6 +350,22 @@ function update_vpa_roles() {
 	fi
 }
 
+function apply_limitrange_and_namespace_quotas() {
+    echo "🔄 Applying LimitRange and Namespace Quotas..."
+
+    kubectl apply -f "${current_dir}/manifests/minimal_limit_range.yaml" -n demo-limitrange-violation
+    if [ $? -ne 0 ]; then
+        echo "❌ Error applying LimitRange manifest."
+    fi
+
+    kubectl apply -f "${current_dir}/manifests/minimal_ns_quota.yaml" -n demo-quota-violation
+    if [ $? -ne 0 ]; then
+        echo "❌ Error applying Namespace Quota manifest."
+    fi
+
+    echo "✅ LimitRange and Namespace Quotas applied!"
+}
+
 function kruize_local_demo_setup() {
 	bench=$1
 	# Start all the installs
@@ -409,6 +425,17 @@ function kruize_local_demo_setup() {
 			benchmarks_install ${APP_NAMESPACE} ${bench} >> "${LOG_FILE}" 2>&1
 			apply_benchmark_load ${APP_NAMESPACE} ${bench} >> "${LOG_FILE}" 2>&1
 			echo "✅ Completed!"
+
+			if [[ -n "${vpa_install_required}" && "${vpa_install_required}" -eq 1 && -n "${CREATE_ALL_EXPERIMENTS}" && "${CREATE_ALL_EXPERIMENTS}" -eq 1 ]]; then
+            	echo -n "🔄 Creating additional namespaces and installing benchmarks..."
+            	create_namespace "demo-quota-violation" >> "${LOG_FILE}" 2>&1
+           		create_namespace "demo-limitrange-violation" >> "${LOG_FILE}" 2>&1
+            	benchmarks_install "demo-quota-violation" ${bench} >> "${LOG_FILE}" 2>&1
+				apply_benchmark_load "demo-quota-violation" ${bench} >> "${LOG_FILE}" 2>&1
+            	benchmarks_install "demo-limitrange-violation" ${bench} >> "${LOG_FILE}" 2>&1
+				apply_benchmark_load "demo-limitrange-violation" ${bench} >> "${LOG_FILE}" 2>&1
+            echo "✅ Completed!"
+        	fi
 		fi
 		echo "" >> "${LOG_FILE}" 2>&1
 	fi
@@ -461,6 +488,10 @@ function kruize_local_demo_setup() {
 		echo -n "🔄 Updating cluser-roles for VPA..."
 		update_vpa_roles >> "${LOG_FILE}" 2>&1
 		echo "✅ Done!"
+
+		if [ "${CREATE_ALL_EXPERIMENTS:-}" == "1" ]; then
+            apply_limitrange_and_namespace_quotas
+        fi
 	}
 	fi
 

monitoring/local_monitoring/vpa_demo/README.md

@@ -32,7 +32,8 @@ cd kruize-demos/monitoring/local_monitoring/vpa_demo
 ```
 
 ```
-Usage: ./bulk_service_demo.sh [-s|-t] [-c cluster-type] [-l] [-p] [-r] [-i kruize-image] [-u kruize-ui-image]
+Usage: ./bulk_service_demo.sh [-s|-t] [-c cluster-type] [-l] [-p] [-r] [-i kruize-image] [-u kruize-ui-image] [-a create-all-experiments]
+a = create all experiments simulating error conditions demo
 c = supports minikube, kind and openshift cluster-type
 i = kruize image.
 p = expose prometheus port

monitoring/local_monitoring/vpa_demo/experiments/container_vpa_experiment_limitrange_violation.json

@@ -0,0 +1,30 @@
+[{
+  "version": "v2.0",
+  "experiment_name": "vpa-limit-violation",
+  "cluster_name": "default",
+  "performance_profile": "resource-optimization-local-monitoring",
+  "metadata_profile": "cluster-metadata-local-monitoring",
+  "mode": "recreate",
+  "target_cluster": "local",
+  "datasource": "prometheus-1",
+  "kubernetes_objects": [
+    {
+      "type": "deployment",
+      "name": "sysbench",
+      "namespace": "demo-limitrange-violation",
+      "containers": [
+        {
+          "container_image_name": "quay.io/kruizehub/sysbench",
+          "container_name": "sysbench"
+        }
+      ]
+    }
+  ],
+  "trial_settings": {
+    "measurement_duration": "2min"
+  },
+  "recommendation_settings": {
+    "threshold": "0.1"
+  }
+}]
+

monitoring/local_monitoring/vpa_demo/experiments/container_vpa_experiment_quota_violation.json

@@ -0,0 +1,30 @@
+[{
+  "version": "v2.0",
+  "experiment_name": "vpa-quota-violation",
+  "cluster_name": "default",
+  "performance_profile": "resource-optimization-local-monitoring",
+  "metadata_profile": "cluster-metadata-local-monitoring",
+  "mode": "recreate",
+  "target_cluster": "local",
+  "datasource": "prometheus-1",
+  "kubernetes_objects": [
+    {
+      "type": "deployment",
+      "name": "sysbench",
+      "namespace": "demo-quota-violation",
+      "containers": [
+        {
+          "container_image_name": "quay.io/kruizehub/sysbench",
+          "container_name": "sysbench"
+        }
+      ]
+    }
+  ],
+  "trial_settings": {
+    "measurement_duration": "2min"
+  },
+  "recommendation_settings": {
+    "threshold": "0.1"
+  }
+}]
+

monitoring/local_monitoring/vpa_demo/manifests/minimal_limit_range.yaml

@@ -0,0 +1,14 @@
+apiVersion: v1
+kind: LimitRange
+metadata:
+  name: minimum-limit-range
+  namespace: demo-limitrange-violation
+spec:
+  limits:
+  - type: Container
+    min:
+      cpu: "500m"
+      memory: "24Mi"
+    max:
+      cpu: "1"
+      memory: "32Gi"

monitoring/local_monitoring/vpa_demo/manifests/minimal_ns_quota.yaml

@@ -0,0 +1,12 @@
+apiVersion: v1
+kind: ResourceQuota
+metadata:
+  name: minimal-quota
+  namespace: "demo-quota-violation"
+spec:
+  hard:
+    pods: "20"
+    requests.cpu: "10m"
+    requests.memory: "0.2Mi"
+    limits.cpu: "10m"
+    limits.memory: "0.2Mi"

monitoring/local_monitoring/vpa_demo/vpa_demo.sh

@@ -36,7 +36,7 @@ KRUIZE_UI_PORT=8081
 TECHEMPOWER_PORT=8082
 
 function usage() {
-	echo "Usage: $0 [-s|-t] [-c cluster-type] [-f] [-i kruize-image] [-u kruize-ui-image] [-e experiment_type] [ [-b] [-m benchmark-manifests] [-n namespace] [-l] [-d load-duration] ] [-p]"
+	echo "Usage: $0 [-s|-t] [-c cluster-type] [-f] [-i kruize-image] [-u kruize-ui-image] [-e experiment_type] [ [-b] [-m benchmark-manifests] [-n namespace] [-l] [-d load-duration] ] [-p] [-a]"
 	echo "s = start (default), t = terminate"
 	echo "c = supports minikube, kind, aks and openshift cluster-type"
 	echo "f = create environment setup if cluster-type is minikube, kind"
@@ -48,7 +48,7 @@ function usage() {
 	echo "l = Run a load against the benchmark"
 	echo "d = duration to run the benchmark load"
 	echo "p = expose prometheus port"
-
+    echo "a = creates all experiments including error scenarios"
 	exit 1
 }
 
@@ -67,10 +67,15 @@ export APP_NAMESPACE="default"
 export LOAD_DURATION="1200"
 export BENCHMARK_MANIFESTS="resource_provisioning_manifests"
 export EXPERIMENT_TYPE=""
+export CREATE_ALL_EXPERIMENTS=0
+
 # Iterate through the commandline options
-while getopts bc:d:e:fi:lm:n:pstu: gopts
+while getopts abc:d:e:fi:lm:n:pstu: gopts
 do
 	case "${gopts}" in
+	    a)
+            export CREATE_ALL_EXPERIMENTS=1
+            ;;
 		b)
 			start_demo=2
 			benchmark=1
@@ -118,9 +123,14 @@ export demo="local"
 export vpa_install_required="1"
 
 EXPERIMENT_TYPE="container"
-export EXPERIMENTS=("container_vpa_experiment_sysbench")
 BENCHMARK="sysbench"
 
+if [ ${CREATE_ALL_EXPERIMENTS} -eq 1 ]; then
+	export EXPERIMENTS=("container_vpa_experiment_sysbench" "container_vpa_experiment_quota_violation" "container_vpa_experiment_limitrange_violation")
+else
+	export EXPERIMENTS=("container_vpa_experiment_sysbench")
+fi
+
 
 if [ ${start_demo} -eq 1 ]; then
 	echo > "${LOG_FILE}"