diff --git a/charts/generic/experiments.yaml b/charts/generic/experiments.yaml index 5f56672..6c6d483 100644 --- a/charts/generic/experiments.yaml +++ b/charts/generic/experiments.yaml @@ -1,213 +1,12 @@ apiVersion: litmuschaos.io/v1alpha1 description: message: | - Injects network packet duplication on pods belonging to an app deployment + Scale the application replicas and test the node autoscaling on cluster kind: ChaosExperiment metadata: - name: pod-network-duplication + name: pod-autoscaler labels: - name: pod-network-duplication - app.kubernetes.io/part-of: litmus - app.kubernetes.io/component: chaosexperiment - app.kubernetes.io/version: latest -spec: - definition: - scope: Namespaced - permissions: - - apiGroups: - - "" - - "batch" - - "apps" - - "apps.openshift.io" - - "argoproj.io" - - "litmuschaos.io" - resources: - - "jobs" - - "pods" - - "pods/log" - - "replicationcontrollers" - - "deployments" - - "statefulsets" - - "daemonsets" - - "replicasets" - - "deploymentconfigs" - - "rollouts" - - "pods/exec" - - "events" - - "chaosengines" - - "chaosexperiments" - - "chaosresults" - verbs: - - "get" - - "list" - - "patch" - - "create" - - "update" - - "delete" - - "deletecollection" - image: "litmuschaos/go-runner:latest" - args: - - -c - - ./experiments -name pod-network-duplication - command: - - /bin/bash - env: - - name: TOTAL_CHAOS_DURATION - value: '60' - - - name: RAMP_TIME - value: '' - - - name: TARGET_CONTAINER - value: '' - - - name: TC_IMAGE - value: 'gaiadocker/iproute2' - - - name: NETWORK_INTERFACE - value: 'eth0' - - - name: NETWORK_PACKET_DUPLICATION_PERCENTAGE - value: '100' # in percentage - - # lib can be litmus or pumba - - name: LIB - value: 'litmus' - - - name: TARGET_PODS - value: '' - - ## percentage of total pods to target - - name: PODS_AFFECTED_PERC - value: '' - - - name: LIB_IMAGE - value: 'litmuschaos/go-runner:latest' - - # provide the name of container runtime - # for litmus LIB, it supports docker, containerd, crio - # for pumba LIB, it supports docker only - - name: CONTAINER_RUNTIME - value: 'docker' - - # provide the destination ips - # chaos injection will be triggered for these destination ips - - name: DESTINATION_IPS - value: '' - - # provide the destination hosts - # chaos injection will be triggered for these destination hosts - - name: DESTINATION_HOSTS - value: '' - - # provide the socket file path - - name: SOCKET_PATH - value: '/var/run/docker.sock' - - ## it defines the sequence of chaos execution for multiple target pods - ## supported values: serial, parallel - - name: SEQUENCE - value: 'parallel' - - labels: - name: pod-network-duplication - app.kubernetes.io/part-of: litmus - app.kubernetes.io/component: experiment-job - app.kubernetes.io/version: latest - ---- -apiVersion: litmuschaos.io/v1alpha1 -description: - message: | - Deletes a pod belonging to a deployment/statefulset/daemonset -kind: ChaosExperiment -metadata: - name: k8-service-kill - labels: - name: k8-service-kill - app.kubernetes.io/part-of: litmus - app.kubernetes.io/component: chaosexperiment - app.kubernetes.io/version: latest -spec: - definition: - scope: Namespaced - permissions: - - apiGroups: - - "" - - "apps" - - "batch" - - "litmuschaos.io" - resources: - - "deployments" - - "jobs" - - "pods" - - "configmaps" - - "chaosengines" - - "chaosexperiments" - - "chaosresults" - verbs: - - "create" - - "list" - - "get" - - "patch" - - "update" - - "delete" - - apiGroups: - - "" - resources: - - "nodes" - verbs : - - "get" - - "list" - labels: - name: k8-service-kill - app.kubernetes.io/part-of: litmus - image: "litmuschaos/chaostoolkit:latest" - args: - - -c - - python /app/chaos/chaostest/kubernetes/k8_wrapper.py; exit 0 - command: - - /bin/bash - env: - - name: CHAOSTOOLKIT_IN_POD - value: 'true' - - - name: FILE - value: 'service-app-kill-health.json' - - - name: NAME_SPACE - value: '' - - - name: LABEL_NAME - value: '' - - - name: APP_ENDPOINT - value: '' - - - name: PERCENTAGE - value: '50' - - - name: REPORT - value: 'true' - - - name: REPORT_ENDPOINT - value: 'none' - - - name: TEST_NAMESPACE - value: 'default' - - ---- ---- -apiVersion: litmuschaos.io/v1alpha1 -description: - message: | - Taint the node where application pod is scheduled -kind: ChaosExperiment -metadata: - name: node-taint - labels: - name: node-taint + name: pod-autoscaler app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: latest @@ -223,11 +22,11 @@ spec: resources: - "jobs" - "pods" - - "events" - "pods/log" - "pods/exec" - - "daemonsets" - - "pods/eviction" + - "deployments" + - "statefulsets" + - "events" - "chaosengines" - "chaosexperiments" - "chaosresults" @@ -246,39 +45,505 @@ spec: verbs: - "get" - "list" + - "create" - "patch" - "update" image: "litmuschaos/go-runner:latest" imagePullPolicy: Always args: - -c - - ./experiments -name node-taint + - ./experiments -name pod-autoscaler command: - /bin/bash env: - - name: TARGET_NODE + - name: TOTAL_CHAOS_DURATION + value: '60' + + # Period to wait before and after injection of chaos in sec + - name: RAMP_TIME value: '' + # Number of replicas to scale + - name: REPLICA_COUNT + value: '5' + + # PROVIDE THE LIB HERE + # ONLY LITMUS SUPPORTED + - name: LIB + value: 'litmus' + + labels: + name: pod-autoscaler + app.kubernetes.io/part-of: litmus + app.kubernetes.io/component: experiment-job + app.kubernetes.io/version: latest + +--- +apiVersion: litmuschaos.io/v1alpha1 +description: + message: "Kills a container belonging to an application pod \n" +kind: ChaosExperiment +metadata: + name: container-kill + labels: + name: container-kill + app.kubernetes.io/part-of: litmus + app.kubernetes.io/component: chaosexperiment + app.kubernetes.io/version: latest +spec: + definition: + scope: Namespaced + permissions: + - apiGroups: + - "" + - "apps" + - "batch" + - "apps.openshift.io" + - "argoproj.io" + - "litmuschaos.io" + resources: + - "jobs" + - "pods" + - "pods/log" + - "events" + - "replicationcontrollers" + - "deployments" + - "statefulsets" + - "daemonsets" + - "replicasets" + - "deploymentconfigs" + - "rollouts" + - "pods/exec" + - "chaosengines" + - "chaosexperiments" + - "chaosresults" + verbs: + - "create" + - "list" + - "get" + - "update" + - "patch" + - "delete" + - "deletecollection" + image: "litmuschaos/go-runner:latest" + imagePullPolicy: Always + args: + - -c + - ./experiments -name container-kill + command: + - /bin/bash + env: + + - name: TARGET_CONTAINER + value: '' + + # Period to wait before and after injection of chaos in sec + - name: RAMP_TIME + value: '' + + # lib can be litmus or pumba + - name: LIB + value: 'litmus' + + - name: TARGET_PODS + value: '' + + # provide the chaos interval + - name: CHAOS_INTERVAL + value: '10' + + # provide the socket file path + - name: SOCKET_PATH + value: '/var/run/docker.sock' + + # provide the name of container runtime + # for litmus LIB, it supports docker, containerd, crio + # for pumba LIB, it supports docker only + - name: CONTAINER_RUNTIME + value: 'docker' + + # provide the total chaos duration + - name: TOTAL_CHAOS_DURATION + value: '20' + + ## percentage of total pods to target + - name: PODS_AFFECTED_PERC + value: '' + + - name: LIB_IMAGE + value: 'litmuschaos/go-runner:latest' + + ## it defines the sequence of chaos execution for multiple target pods + ## supported values: serial, parallel + - name: SEQUENCE + value: 'parallel' + + labels: + name: container-kill + app.kubernetes.io/part-of: litmus + app.kubernetes.io/component: experiment-job + app.kubernetes.io/version: latest + +--- +apiVersion: litmuschaos.io/v1alpha1 +description: + message: | + Detaching a persistent disk from a node/instance. Supports only for AWS and GCP +kind: ChaosExperiment +metadata: + name: disk-loss + labels: + name: disk-loss + app.kubernetes.io/part-of: litmus + app.kubernetes.io/component: chaosexperiment + app.kubernetes.io/version: latest +spec: + definition: + scope: Cluster + permissions: + - apiGroups: + - "" + - "batch" + - "litmuschaos.io" + resources: + - "jobs" + - "pods" + - "events" + - "pods/log" + - "secrets" + - "chaosengines" + - "chaosexperiments" + - "chaosresults" + verbs: + - "create" + - "list" + - "get" + - "patch" + - "update" + - "delete" + image: "litmuschaos/ansible-runner:latest" + imagePullPolicy: Always + args: + - -c + - ansible-playbook ./experiments/generic/disk_loss/disk_loss_ansible_logic.yml -i /etc/ansible/hosts -vv; exit 0 + command: + - /bin/bash + env: + - name: ANSIBLE_STDOUT_CALLBACK + value: 'default' + + - name: TOTAL_CHAOS_DURATION + value: '15' + + # Period to wait before and after injection of chaos in sec + - name: RAMP_TIME + value: '' + + - name: APP_CHECK + value: 'true' + + # GKE and AWS supported + - name: CLOUD_PLATFORM + value: 'GKE' + + - name: PROJECT_ID + value: '' + + - name: NODE_NAME + value: '' + + - name: DISK_NAME + value: '' + + # provide the LIB + # only litmus supported + - name: LIB + value: 'litmus' + + - name: ZONE_NAME + value: '' + + - name: DEVICE_NAME + value: '' + + labels: + name: disk-loss + app.kubernetes.io/part-of: litmus + app.kubernetes.io/component: experiment-job + app.kubernetes.io/version: latest + secrets: + - name: cloud-secret + mountPath: /tmp/ + +--- +apiVersion: litmuschaos.io/v1alpha1 +description: + message: | + Fillup Ephemeral Storage of a Resource +kind: ChaosExperiment +metadata: + name: disk-fill + labels: + name: disk-fill + app.kubernetes.io/part-of: litmus + app.kubernetes.io/component: chaosexperiment + app.kubernetes.io/version: latest +spec: + definition: + scope: Cluster + permissions: + - apiGroups: + - "" + - "apps" + - "batch" + - "apps.openshift.io" + - "argoproj.io" + - "litmuschaos.io" + resources: + - "jobs" + - "pods" + - "pods/exec" + - "pods/log" + - "replicationcontrollers" + - "deployments" + - "statefulsets" + - "daemonsets" + - "replicasets" + - "deploymentconfigs" + - "rollouts" + - "events" + - "chaosengines" + - "chaosexperiments" + - "chaosresults" + verbs: + - "create" + - "list" + - "get" + - "patch" + - "update" + - "delete" + - "deletecollection" + image: "litmuschaos/go-runner:latest" + imagePullPolicy: Always + args: + - -c + - ./experiments -name disk-fill + command: + - /bin/bash + env: + + - name: TARGET_CONTAINER + value: '' + + - name: FILL_PERCENTAGE + value: '80' + - name: TOTAL_CHAOS_DURATION value: '60' + # Period to wait before and after injection of chaos in sec + - name: RAMP_TIME + value: '' + # Provide the LIB here # Only litmus supported - name: LIB value: 'litmus' + - name: TARGET_PODS + value: '' + + ## percentage of total pods to target + - name: PODS_AFFECTED_PERC + value: '' + + - name: LIB_IMAGE + value: 'litmuschaos/go-runner:latest' + + # Provide the container runtime path + # Default set to docker container path + - name: CONTAINER_PATH + value: '/var/lib/docker/containers' + + ## it defines the sequence of chaos execution for multiple target pods + ## supported values: serial, parallel + - name: SEQUENCE + value: 'parallel' + + labels: + name: disk-fill + app.kubernetes.io/part-of: litmus + app.kubernetes.io/component: experiment-job + app.kubernetes.io/version: latest + +--- +apiVersion: litmuschaos.io/v1alpha1 +description: + message: | + Kills the kubelet service on the application node to check the resiliency. +kind: ChaosExperiment +metadata: + name: kubelet-service-kill + labels: + name: kubelet-service-kill + app.kubernetes.io/part-of: litmus + app.kubernetes.io/component: chaosexperiment + app.kubernetes.io/version: latest +spec: + definition: + scope: Cluster + permissions: + - apiGroups: + - "" + - "batch" + - "litmuschaos.io" + resources: + - "jobs" + - "pods" + - "pods/log" + - "pods/exec" + - "events" + - "chaosengines" + - "chaosexperiments" + - "chaosresults" + verbs: + - "create" + - "list" + - "get" + - "patch" + - "update" + - "delete" + - "deletecollection" + - apiGroups: + - "" + resources: + - "nodes" + verbs: + - "get" + - "list" + image: "litmuschaos/go-runner:latest" + imagePullPolicy: Always + args: + - -c + - ./experiments -name kubelet-service-kill + command: + - /bin/bash + env: + + - name: TOTAL_CHAOS_DURATION + value: '90' # in seconds + # Period to wait before and after injection of chaos in sec - name: RAMP_TIME value: '' - # set taint label & effect - # key=value:effect or key:effect - - name: TAINTS - value: '' + - name: LIB + value: 'litmus' + + # provide lib image + - name: LIB_IMAGE + value: 'ubuntu:16.04' + + # provide the target node name + - name: TARGET_NODE + value: '' labels: - name: node-taint + name: kubelet-service-kill + app.kubernetes.io/part-of: litmus + app.kubernetes.io/component: experiment-job + app.kubernetes.io/version: latest + +--- +apiVersion: litmuschaos.io/v1alpha1 +description: + message: | + Deletes a pod belonging to a deployment/statefulset/daemonset +kind: ChaosExperiment +metadata: + name: pod-delete + labels: + name: pod-delete + app.kubernetes.io/part-of: litmus + app.kubernetes.io/component: chaosexperiment + app.kubernetes.io/version: latest + version: 0.1.19 +spec: + definition: + scope: Namespaced + permissions: + - apiGroups: + - "" + - "apps" + - "apps.openshift.io" + - "argoproj.io" + - "batch" + - "litmuschaos.io" + resources: + - "deployments" + - "jobs" + - "pods" + - "pods/log" + - "replicationcontrollers" + - "deployments" + - "statefulsets" + - "daemonsets" + - "replicasets" + - "deploymentconfigs" + - "rollouts" + - "pods/exec" + - "events" + - "chaosengines" + - "chaosexperiments" + - "chaosresults" + verbs: + - "create" + - "list" + - "get" + - "patch" + - "update" + - "delete" + - "deletecollection" + image: "litmuschaos/go-runner:latest" + imagePullPolicy: Always + args: + - -c + - ./experiments -name pod-delete + command: + - /bin/bash + env: + + - name: TOTAL_CHAOS_DURATION + value: '15' + + # Period to wait before and after injection of chaos in sec + - name: RAMP_TIME + value: '' + + - name: FORCE + value: 'true' + + - name: CHAOS_INTERVAL + value: '5' + + ## percentage of total pods to target + - name: PODS_AFFECTED_PERC + value: '' + + - name: LIB + value: 'litmus' + + - name: TARGET_PODS + value: '' + + ## it defines the sequence of chaos execution for multiple target pods + ## supported values: serial, parallel + - name: SEQUENCE + value: 'parallel' + + labels: + name: pod-delete app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job app.kubernetes.io/version: latest @@ -385,6 +650,128 @@ spec: app.kubernetes.io/component: experiment-job app.kubernetes.io/version: latest +--- +apiVersion: litmuschaos.io/v1alpha1 +description: + message: | + Injects network latency on pods belonging to an app deployment +kind: ChaosExperiment +metadata: + name: pod-network-latency + labels: + name: pod-network-latency + app.kubernetes.io/part-of: litmus + app.kubernetes.io/component: chaosexperiment + app.kubernetes.io/version: latest +spec: + definition: + scope: Namespaced + permissions: + - apiGroups: + - "" + - "batch" + - "apps" + - "apps.openshift.io" + - "argoproj.io" + - "litmuschaos.io" + resources: + - "jobs" + - "pods" + - "pods/log" + - "replicationcontrollers" + - "deployments" + - "statefulsets" + - "daemonsets" + - "replicasets" + - "deploymentconfigs" + - "rollouts" + - "pods/exec" + - "events" + - "chaosengines" + - "chaosexperiments" + - "chaosresults" + verbs: + - "create" + - "list" + - "get" + - "patch" + - "update" + - "delete" + - "deletecollection" + image: "litmuschaos/go-runner:latest" + imagePullPolicy: Always + args: + - -c + - ./experiments -name pod-network-latency + command: + - /bin/bash + env: + + - name: TARGET_CONTAINER + value: '' + + - name: NETWORK_INTERFACE + value: 'eth0' + + # provide lib image + - name: LIB_IMAGE + value: 'litmuschaos/go-runner:latest' + + - name: TC_IMAGE + value: 'gaiadocker/iproute2' + + - name: NETWORK_LATENCY + value: '2000' #in ms + + - name: TOTAL_CHAOS_DURATION + value: '60' # in seconds + + # Time period to wait before and after injection of chaos in sec + - name: RAMP_TIME + value: '' + + # lib can be litmus or pumba + - name: LIB + value: 'litmus' + + ## percentage of total pods to target + - name: PODS_AFFECTED_PERC + value: '' + + - name: TARGET_PODS + value: '' + + # provide the name of container runtime + # for litmus LIB, it supports docker, containerd, crio + # for pumba LIB, it supports docker only + - name: CONTAINER_RUNTIME + value: 'docker' + + # provide the destination ips + # chaos injection will be triggered for these destination ips + - name: DESTINATION_IPS + value: '' + + # provide the destination hosts + # chaos injection will be triggered for these destination hosts + - name: DESTINATION_HOSTS + value: '' + + # provide the socket file path + - name: SOCKET_PATH + value: '/var/run/docker.sock' + + ## it defines the sequence of chaos execution for multiple target pods + ## supported values: serial, parallel + - name: SEQUENCE + value: 'parallel' + + labels: + name: pod-network-latency + app.kubernetes.io/part-of: litmus + app.kubernetes.io/component: experiment-job + app.kubernetes.io/version: latest + --- apiVersion: litmuschaos.io/v1alpha1 description: @@ -476,315 +863,6 @@ spec: - name: id-rsa mountPath: /mnt/ ---- -apiVersion: litmuschaos.io/v1alpha1 -description: - message: | - Give a cpu spike on a node belonging to a deployment -kind: ChaosExperiment -metadata: - name: node-cpu-hog - labels: - name: node-cpu-hog - app.kubernetes.io/part-of: litmus - app.kubernetes.io/component: chaosexperiment - app.kubernetes.io/version: latest -spec: - definition: - scope: Cluster - permissions: - - apiGroups: - - "" - - "batch" - - "litmuschaos.io" - resources: - - "jobs" - - "pods" - - "pods/log" - - "pods/exec" - - "events" - - "chaosengines" - - "chaosexperiments" - - "chaosresults" - verbs: - - "create" - - "list" - - "get" - - "patch" - - "update" - - "delete" - - "deletecollection" - - apiGroups: - - "" - resources: - - "nodes" - verbs: - - "get" - - "list" - image: "litmuschaos/go-runner:latest" - imagePullPolicy: Always - args: - - -c - - ./experiments -name node-cpu-hog - command: - - /bin/bash - env: - - - name: TOTAL_CHAOS_DURATION - value: '60' - - # Period to wait before and after injection of chaos in sec - - name: RAMP_TIME - value: '' - - ## ENTER THE NUMBER OF CORES OF CPU FOR CPU HOGGING - ## OPTIONAL VALUE IN CASE OF EMPTY VALUE IT WILL TAKE NODE CPU CAPACITY - - name: NODE_CPU_CORE - value: '' - - # ENTER THE COMMA SEPARATED TARGET NODES NAME - - name: TARGET_NODES - value: '' - - # PROVIDE THE LIB HERE - # ONLY LITMUS SUPPORTED - - name: LIB - value: 'litmus' - - # provide lib image - - name: LIB_IMAGE - value: 'litmuschaos/go-runner:latest' - - ## percentage of total nodes to target - - name: NODES_AFFECTED_PERC - value: '' - - ## it defines the sequence of chaos execution for multiple target nodes - ## supported values: serial, parallel - - name: SEQUENCE - value: 'parallel' - - labels: - name: node-cpu-hog - app.kubernetes.io/part-of: litmus - app.kubernetes.io/component: experiment-job - app.kubernetes.io/version: latest - ---- -apiVersion: litmuschaos.io/v1alpha1 -description: - message: | - Inject network packet corruption into application pod -kind: ChaosExperiment -metadata: - name: pod-network-corruption - labels: - name: pod-network-corruption - app.kubernetes.io/part-of: litmus - app.kubernetes.io/component: chaosexperiment - app.kubernetes.io/version: latest -spec: - definition: - scope: Namespaced - permissions: - - apiGroups: - - "" - - "batch" - - "apps" - - "apps.openshift.io" - - "argoproj.io" - - "litmuschaos.io" - resources: - - "jobs" - - "pods" - - "pods/log" - - "replicationcontrollers" - - "deployments" - - "statefulsets" - - "daemonsets" - - "replicasets" - - "deploymentconfigs" - - "rollouts" - - "pods/exec" - - "events" - - "chaosengines" - - "chaosexperiments" - - "chaosresults" - verbs: - - "create" - - "delete" - - "list" - - "patch" - - "update" - - "get" - - "deletecollection" - image: "litmuschaos/go-runner:latest" - imagePullPolicy: Always - args: - - -c - - ./experiments -name pod-network-corruption - command: - - /bin/bash - env: - - - name: TARGET_CONTAINER - value: '' - - # provide lib image - - name: LIB_IMAGE - value: 'litmuschaos/go-runner:latest' - - - name: NETWORK_INTERFACE - value: 'eth0' - - - name: TC_IMAGE - value: 'gaiadocker/iproute2' - - - name: NETWORK_PACKET_CORRUPTION_PERCENTAGE - value: '100' #in PERCENTAGE - - - name: TOTAL_CHAOS_DURATION - value: '60' # in seconds - - # Time period to wait before and after injection of chaos in sec - - name: RAMP_TIME - value: '' - - # lib can be litmus or pumba - - name: LIB - value: 'litmus' - - ## percentage of total pods to target - - name: PODS_AFFECTED_PERC - value: '' - - - name: TARGET_PODS - value: '' - - # provide the name of container runtime - # for litmus LIB, it supports docker, containerd, crio - # for pumba LIB, it supports docker only - - name: CONTAINER_RUNTIME - value: 'docker' - - # provide the destination ips - # chaos injection will be triggered for these destination ips - - name: DESTINATION_IPS - value: '' - - # provide the destination hosts - # chaos injection will be triggered for these destination hosts - - name: DESTINATION_HOSTS - value: '' - - # provide the socket file path - - name: SOCKET_PATH - value: '/var/run/docker.sock' - - ## it defines the sequence of chaos execution for multiple target pods - ## supported values: serial, parallel - - name: SEQUENCE - value: 'parallel' - - labels: - name: pod-network-corruption - app.kubernetes.io/part-of: litmus - app.kubernetes.io/component: experiment-job - app.kubernetes.io/version: latest - ---- -apiVersion: litmuschaos.io/v1alpha1 -description: - message: | - Deletes a pod belonging to a deployment/statefulset/daemonset -kind: ChaosExperiment -metadata: - name: pod-delete - labels: - name: pod-delete - app.kubernetes.io/part-of: litmus - app.kubernetes.io/component: chaosexperiment - app.kubernetes.io/version: latest - version: 0.1.19 -spec: - definition: - scope: Namespaced - permissions: - - apiGroups: - - "" - - "apps" - - "apps.openshift.io" - - "argoproj.io" - - "batch" - - "litmuschaos.io" - resources: - - "deployments" - - "jobs" - - "pods" - - "pods/log" - - "replicationcontrollers" - - "deployments" - - "statefulsets" - - "daemonsets" - - "replicasets" - - "deploymentconfigs" - - "rollouts" - - "pods/exec" - - "events" - - "chaosengines" - - "chaosexperiments" - - "chaosresults" - verbs: - - "create" - - "list" - - "get" - - "patch" - - "update" - - "delete" - - "deletecollection" - image: "litmuschaos/go-runner:latest" - imagePullPolicy: Always - args: - - -c - - ./experiments -name pod-delete - command: - - /bin/bash - env: - - - name: TOTAL_CHAOS_DURATION - value: '15' - - # Period to wait before and after injection of chaos in sec - - name: RAMP_TIME - value: '' - - - name: FORCE - value: 'true' - - - name: CHAOS_INTERVAL - value: '5' - - ## percentage of total pods to target - - name: PODS_AFFECTED_PERC - value: '' - - - name: LIB - value: 'litmus' - - - name: TARGET_PODS - value: '' - - ## it defines the sequence of chaos execution for multiple target pods - ## supported values: serial, parallel - - name: SEQUENCE - value: 'parallel' - - labels: - name: pod-delete - app.kubernetes.io/part-of: litmus - app.kubernetes.io/component: experiment-job - app.kubernetes.io/version: latest - --- --- apiVersion: litmuschaos.io/v1alpha1 @@ -1006,32 +1084,154 @@ spec: apiVersion: litmuschaos.io/v1alpha1 description: message: | - Give a memory hog on a node belonging to a deployment + Inject network packet corruption into application pod kind: ChaosExperiment metadata: - name: node-memory-hog + name: pod-network-corruption labels: - name: node-memory-hog + name: pod-network-corruption app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: latest spec: definition: - scope: Cluster + scope: Namespaced permissions: - apiGroups: - "" - "batch" + - "apps" + - "apps.openshift.io" + - "argoproj.io" - "litmuschaos.io" resources: - "jobs" - "pods" - "pods/log" + - "replicationcontrollers" + - "deployments" + - "statefulsets" + - "daemonsets" + - "replicasets" + - "deploymentconfigs" + - "rollouts" - "pods/exec" - "events" - "chaosengines" - "chaosexperiments" - "chaosresults" + verbs: + - "create" + - "delete" + - "list" + - "patch" + - "update" + - "get" + - "deletecollection" + image: "litmuschaos/go-runner:latest" + imagePullPolicy: Always + args: + - -c + - ./experiments -name pod-network-corruption + command: + - /bin/bash + env: + + - name: TARGET_CONTAINER + value: '' + + # provide lib image + - name: LIB_IMAGE + value: 'litmuschaos/go-runner:latest' + + - name: NETWORK_INTERFACE + value: 'eth0' + + - name: TC_IMAGE + value: 'gaiadocker/iproute2' + + - name: NETWORK_PACKET_CORRUPTION_PERCENTAGE + value: '100' #in PERCENTAGE + + - name: TOTAL_CHAOS_DURATION + value: '60' # in seconds + + # Time period to wait before and after injection of chaos in sec + - name: RAMP_TIME + value: '' + + # lib can be litmus or pumba + - name: LIB + value: 'litmus' + + ## percentage of total pods to target + - name: PODS_AFFECTED_PERC + value: '' + + - name: TARGET_PODS + value: '' + + # provide the name of container runtime + # for litmus LIB, it supports docker, containerd, crio + # for pumba LIB, it supports docker only + - name: CONTAINER_RUNTIME + value: 'docker' + + # provide the destination ips + # chaos injection will be triggered for these destination ips + - name: DESTINATION_IPS + value: '' + + # provide the destination hosts + # chaos injection will be triggered for these destination hosts + - name: DESTINATION_HOSTS + value: '' + + # provide the socket file path + - name: SOCKET_PATH + value: '/var/run/docker.sock' + + ## it defines the sequence of chaos execution for multiple target pods + ## supported values: serial, parallel + - name: SEQUENCE + value: 'parallel' + + labels: + name: pod-network-corruption + app.kubernetes.io/part-of: litmus + app.kubernetes.io/component: experiment-job + app.kubernetes.io/version: latest + +--- +apiVersion: litmuschaos.io/v1alpha1 +description: + message: | + Deletes a pod belonging to a deployment/statefulset/daemonset +kind: ChaosExperiment +metadata: + name: k8-service-kill + labels: + name: k8-service-kill + app.kubernetes.io/part-of: litmus + app.kubernetes.io/component: chaosexperiment + app.kubernetes.io/version: latest +spec: + definition: + scope: Namespaced + permissions: + - apiGroups: + - "" + - "apps" + - "batch" + - "litmuschaos.io" + resources: + - "deployments" + - "jobs" + - "pods" + - "configmaps" + - "chaosengines" + - "chaosexperiments" + - "chaosresults" verbs: - "create" - "list" @@ -1039,62 +1239,50 @@ spec: - "patch" - "update" - "delete" - - "deletecollection" - apiGroups: - "" - resources: + resources: - "nodes" - verbs: + verbs : - "get" - "list" - image: "litmuschaos/go-runner:latest" - imagePullPolicy: Always + labels: + name: k8-service-kill + app.kubernetes.io/part-of: litmus + image: "litmuschaos/chaostoolkit:latest" args: - -c - - ./experiments -name node-memory-hog + - python /app/chaos/chaostest/kubernetes/k8_wrapper.py; exit 0 command: - /bin/bash env: + - name: CHAOSTOOLKIT_IN_POD + value: 'true' - - name: TOTAL_CHAOS_DURATION - value: '120' + - name: FILE + value: 'service-app-kill-health.json' - ## specify the size as percent of total available memory (in percentage) - ## Default value "90" - - name: MEMORY_PERCENTAGE - value: '90' - - # ENTER THE COMMA SEPARATED TARGET NODES NAME - - name: TARGET_NODES - value: '' - - # Period to wait before and after injection of chaos in sec - - name: RAMP_TIME + - name: NAME_SPACE value: '' - # Provide the LIB here - # Only litmus supported - - name: LIB - value: 'litmus' - - # provide lib image - - name: LIB_IMAGE - value: 'litmuschaos/go-runner:latest' - - ## percentage of total nodes to target - - name: NODES_AFFECTED_PERC + - name: LABEL_NAME value: '' - ## it defines the sequence of chaos execution for multiple target nodes - ## supported values: serial, parallel - - name: SEQUENCE - value: 'parallel' - - labels: - name: node-memory-hog - app.kubernetes.io/part-of: litmus - app.kubernetes.io/component: experiment-job - app.kubernetes.io/version: latest + - name: APP_ENDPOINT + value: '' + + - name: PERCENTAGE + value: '50' + + - name: REPORT + value: 'true' + + - name: REPORT_ENDPOINT + value: 'none' + + - name: TEST_NAMESPACE + value: 'default' + --- apiVersion: litmuschaos.io/v1alpha1 @@ -1185,12 +1373,12 @@ spec: apiVersion: litmuschaos.io/v1alpha1 description: message: | - Kills the kubelet service on the application node to check the resiliency. + Kills the docker service on the application node to check the resiliency. kind: ChaosExperiment metadata: - name: kubelet-service-kill + name: docker-service-kill labels: - name: kubelet-service-kill + name: docker-service-kill app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: latest @@ -1201,12 +1389,12 @@ spec: - apiGroups: - "" - "batch" + - "apps" - "litmuschaos.io" resources: - "jobs" - "pods" - "pods/log" - - "pods/exec" - "events" - "chaosengines" - "chaosexperiments" @@ -1218,47 +1406,40 @@ spec: - "patch" - "update" - "delete" - - "deletecollection" - apiGroups: - "" - resources: + resources: - "nodes" verbs: - "get" - "list" - image: "litmuschaos/go-runner:latest" + image: "litmuschaos/ansible-runner:ci" imagePullPolicy: Always args: - -c - - ./experiments -name kubelet-service-kill + - ansible-playbook ./experiments/generic/docker_service_kill/docker_service_kill_ansible_logic.yml -i /etc/ansible/hosts -vv; exit 0 command: - /bin/bash env: - + - name: ANSIBLE_STDOUT_CALLBACK + value: 'default' + - name: TOTAL_CHAOS_DURATION value: '90' # in seconds - # Period to wait before and after injection of chaos in sec + # Period to wait before injection of chaos in sec - name: RAMP_TIME value: '' - name: LIB value: 'litmus' - - # provide lib image - - name: LIB_IMAGE - value: 'ubuntu:16.04' - - # provide the target node name - - name: TARGET_NODE - value: '' - labels: - name: kubelet-service-kill + name: docker-service-kill app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job app.kubernetes.io/version: latest + --- --- apiVersion: litmuschaos.io/v1alpha1 @@ -1339,6 +1520,124 @@ spec: app.kubernetes.io/component: experiment-job app.kubernetes.io/version: latest +--- +apiVersion: litmuschaos.io/v1alpha1 +description: + message: | + Injects network packet duplication on pods belonging to an app deployment +kind: ChaosExperiment +metadata: + name: pod-network-duplication + labels: + name: pod-network-duplication + app.kubernetes.io/part-of: litmus + app.kubernetes.io/component: chaosexperiment + app.kubernetes.io/version: latest +spec: + definition: + scope: Namespaced + permissions: + - apiGroups: + - "" + - "batch" + - "apps" + - "apps.openshift.io" + - "argoproj.io" + - "litmuschaos.io" + resources: + - "jobs" + - "pods" + - "pods/log" + - "replicationcontrollers" + - "deployments" + - "statefulsets" + - "daemonsets" + - "replicasets" + - "deploymentconfigs" + - "rollouts" + - "pods/exec" + - "events" + - "chaosengines" + - "chaosexperiments" + - "chaosresults" + verbs: + - "get" + - "list" + - "patch" + - "create" + - "update" + - "delete" + - "deletecollection" + image: "litmuschaos/go-runner:latest" + args: + - -c + - ./experiments -name pod-network-duplication + command: + - /bin/bash + env: + - name: TOTAL_CHAOS_DURATION + value: '60' + + - name: RAMP_TIME + value: '' + + - name: TARGET_CONTAINER + value: '' + + - name: TC_IMAGE + value: 'gaiadocker/iproute2' + + - name: NETWORK_INTERFACE + value: 'eth0' + + - name: NETWORK_PACKET_DUPLICATION_PERCENTAGE + value: '100' # in percentage + + # lib can be litmus or pumba + - name: LIB + value: 'litmus' + + - name: TARGET_PODS + value: '' + + ## percentage of total pods to target + - name: PODS_AFFECTED_PERC + value: '' + + - name: LIB_IMAGE + value: 'litmuschaos/go-runner:latest' + + # provide the name of container runtime + # for litmus LIB, it supports docker, containerd, crio + # for pumba LIB, it supports docker only + - name: CONTAINER_RUNTIME + value: 'docker' + + # provide the destination ips + # chaos injection will be triggered for these destination ips + - name: DESTINATION_IPS + value: '' + + # provide the destination hosts + # chaos injection will be triggered for these destination hosts + - name: DESTINATION_HOSTS + value: '' + + # provide the socket file path + - name: SOCKET_PATH + value: '/var/run/docker.sock' + + ## it defines the sequence of chaos execution for multiple target pods + ## supported values: serial, parallel + - name: SEQUENCE + value: 'parallel' + + labels: + name: pod-network-duplication + app.kubernetes.io/part-of: litmus + app.kubernetes.io/component: experiment-job + app.kubernetes.io/version: latest + --- apiVersion: litmuschaos.io/v1alpha1 description: @@ -1436,198 +1735,6 @@ spec: - name: id-rsa mountPath: /mnt/ ---- -apiVersion: litmuschaos.io/v1alpha1 -description: - message: | - Fillup Ephemeral Storage of a Resource -kind: ChaosExperiment -metadata: - name: disk-fill - labels: - name: disk-fill - app.kubernetes.io/part-of: litmus - app.kubernetes.io/component: chaosexperiment - app.kubernetes.io/version: latest -spec: - definition: - scope: Cluster - permissions: - - apiGroups: - - "" - - "apps" - - "batch" - - "apps.openshift.io" - - "argoproj.io" - - "litmuschaos.io" - resources: - - "jobs" - - "pods" - - "pods/exec" - - "pods/log" - - "replicationcontrollers" - - "deployments" - - "statefulsets" - - "daemonsets" - - "replicasets" - - "deploymentconfigs" - - "rollouts" - - "events" - - "chaosengines" - - "chaosexperiments" - - "chaosresults" - verbs: - - "create" - - "list" - - "get" - - "patch" - - "update" - - "delete" - - "deletecollection" - image: "litmuschaos/go-runner:latest" - imagePullPolicy: Always - args: - - -c - - ./experiments -name disk-fill - command: - - /bin/bash - env: - - - name: TARGET_CONTAINER - value: '' - - - name: FILL_PERCENTAGE - value: '80' - - - name: TOTAL_CHAOS_DURATION - value: '60' - - # Period to wait before and after injection of chaos in sec - - name: RAMP_TIME - value: '' - - # Provide the LIB here - # Only litmus supported - - name: LIB - value: 'litmus' - - - name: TARGET_PODS - value: '' - - ## percentage of total pods to target - - name: PODS_AFFECTED_PERC - value: '' - - - name: LIB_IMAGE - value: 'litmuschaos/go-runner:latest' - - # Provide the container runtime path - # Default set to docker container path - - name: CONTAINER_PATH - value: '/var/lib/docker/containers' - - ## it defines the sequence of chaos execution for multiple target pods - ## supported values: serial, parallel - - name: SEQUENCE - value: 'parallel' - - labels: - name: disk-fill - app.kubernetes.io/part-of: litmus - app.kubernetes.io/component: experiment-job - app.kubernetes.io/version: latest - ---- -apiVersion: litmuschaos.io/v1alpha1 -description: - message: | - Detaching a persistent disk from a node/instance. Supports only for AWS and GCP -kind: ChaosExperiment -metadata: - name: disk-loss - labels: - name: disk-loss - app.kubernetes.io/part-of: litmus - app.kubernetes.io/component: chaosexperiment - app.kubernetes.io/version: latest -spec: - definition: - scope: Cluster - permissions: - - apiGroups: - - "" - - "batch" - - "litmuschaos.io" - resources: - - "jobs" - - "pods" - - "events" - - "pods/log" - - "secrets" - - "chaosengines" - - "chaosexperiments" - - "chaosresults" - verbs: - - "create" - - "list" - - "get" - - "patch" - - "update" - - "delete" - image: "litmuschaos/ansible-runner:latest" - imagePullPolicy: Always - args: - - -c - - ansible-playbook ./experiments/generic/disk_loss/disk_loss_ansible_logic.yml -i /etc/ansible/hosts -vv; exit 0 - command: - - /bin/bash - env: - - name: ANSIBLE_STDOUT_CALLBACK - value: 'default' - - - name: TOTAL_CHAOS_DURATION - value: '15' - - # Period to wait before and after injection of chaos in sec - - name: RAMP_TIME - value: '' - - - name: APP_CHECK - value: 'true' - - # GKE and AWS supported - - name: CLOUD_PLATFORM - value: 'GKE' - - - name: PROJECT_ID - value: '' - - - name: NODE_NAME - value: '' - - - name: DISK_NAME - value: '' - - # provide the LIB - # only litmus supported - - name: LIB - value: 'litmus' - - - name: ZONE_NAME - value: '' - - - name: DEVICE_NAME - value: '' - - labels: - name: disk-loss - app.kubernetes.io/part-of: litmus - app.kubernetes.io/component: experiment-job - app.kubernetes.io/version: latest - secrets: - - name: cloud-secret - mountPath: /tmp/ - --- --- apiVersion: litmuschaos.io/v1alpha1 @@ -1734,12 +1841,12 @@ spec: apiVersion: litmuschaos.io/v1alpha1 description: message: | - Kills the docker service on the application node to check the resiliency. + Give a cpu spike on a node belonging to a deployment kind: ChaosExperiment metadata: - name: docker-service-kill + name: node-cpu-hog labels: - name: docker-service-kill + name: node-cpu-hog app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: latest @@ -1750,92 +1857,11 @@ spec: - apiGroups: - "" - "batch" - - "apps" - "litmuschaos.io" resources: - "jobs" - "pods" - "pods/log" - - "events" - - "chaosengines" - - "chaosexperiments" - - "chaosresults" - verbs: - - "create" - - "list" - - "get" - - "patch" - - "update" - - "delete" - - apiGroups: - - "" - resources: - - "nodes" - verbs: - - "get" - - "list" - image: "litmuschaos/ansible-runner:ci" - imagePullPolicy: Always - args: - - -c - - ansible-playbook ./experiments/generic/docker_service_kill/docker_service_kill_ansible_logic.yml -i /etc/ansible/hosts -vv; exit 0 - command: - - /bin/bash - env: - - name: ANSIBLE_STDOUT_CALLBACK - value: 'default' - - - name: TOTAL_CHAOS_DURATION - value: '90' # in seconds - - # Period to wait before injection of chaos in sec - - name: RAMP_TIME - value: '' - - - name: LIB - value: 'litmus' - labels: - name: docker-service-kill - app.kubernetes.io/part-of: litmus - app.kubernetes.io/component: experiment-job - app.kubernetes.io/version: latest - - ---- -apiVersion: litmuschaos.io/v1alpha1 -description: - message: | - Injects network latency on pods belonging to an app deployment -kind: ChaosExperiment -metadata: - name: pod-network-latency - labels: - name: pod-network-latency - app.kubernetes.io/part-of: litmus - app.kubernetes.io/component: chaosexperiment - app.kubernetes.io/version: latest -spec: - definition: - scope: Namespaced - permissions: - - apiGroups: - - "" - - "batch" - - "apps" - - "apps.openshift.io" - - "argoproj.io" - - "litmuschaos.io" - resources: - - "jobs" - - "pods" - - "pods/log" - - "replicationcontrollers" - - "deployments" - - "statefulsets" - - "daemonsets" - - "replicasets" - - "deploymentconfigs" - - "rollouts" - "pods/exec" - "events" - "chaosengines" @@ -1849,76 +1875,152 @@ spec: - "update" - "delete" - "deletecollection" + - apiGroups: + - "" + resources: + - "nodes" + verbs: + - "get" + - "list" image: "litmuschaos/go-runner:latest" imagePullPolicy: Always args: - -c - - ./experiments -name pod-network-latency + - ./experiments -name node-cpu-hog command: - /bin/bash env: - - - name: TARGET_CONTAINER + + - name: TOTAL_CHAOS_DURATION + value: '60' + + # Period to wait before and after injection of chaos in sec + - name: RAMP_TIME value: '' - - name: NETWORK_INTERFACE - value: 'eth0' + ## ENTER THE NUMBER OF CORES OF CPU FOR CPU HOGGING + ## OPTIONAL VALUE IN CASE OF EMPTY VALUE IT WILL TAKE NODE CPU CAPACITY + - name: NODE_CPU_CORE + value: '' + + # ENTER THE COMMA SEPARATED TARGET NODES NAME + - name: TARGET_NODES + value: '' + + # PROVIDE THE LIB HERE + # ONLY LITMUS SUPPORTED + - name: LIB + value: 'litmus' # provide lib image - name: LIB_IMAGE value: 'litmuschaos/go-runner:latest' - - name: TC_IMAGE - value: 'gaiadocker/iproute2' - - - name: NETWORK_LATENCY - value: '2000' #in ms - - - name: TOTAL_CHAOS_DURATION - value: '60' # in seconds - - # Time period to wait before and after injection of chaos in sec - - name: RAMP_TIME + ## percentage of total nodes to target + - name: NODES_AFFECTED_PERC value: '' - # lib can be litmus or pumba - - name: LIB - value: 'litmus' - - ## percentage of total pods to target - - name: PODS_AFFECTED_PERC - value: '' - - - name: TARGET_PODS - value: '' - - # provide the name of container runtime - # for litmus LIB, it supports docker, containerd, crio - # for pumba LIB, it supports docker only - - name: CONTAINER_RUNTIME - value: 'docker' - - # provide the destination ips - # chaos injection will be triggered for these destination ips - - name: DESTINATION_IPS - value: '' - - # provide the destination hosts - # chaos injection will be triggered for these destination hosts - - name: DESTINATION_HOSTS - value: '' - - # provide the socket file path - - name: SOCKET_PATH - value: '/var/run/docker.sock' - - ## it defines the sequence of chaos execution for multiple target pods + ## it defines the sequence of chaos execution for multiple target nodes ## supported values: serial, parallel - name: SEQUENCE value: 'parallel' - + labels: - name: pod-network-latency + name: node-cpu-hog + app.kubernetes.io/part-of: litmus + app.kubernetes.io/component: experiment-job + app.kubernetes.io/version: latest + +--- +apiVersion: litmuschaos.io/v1alpha1 +description: + message: | + Give a memory hog on a node belonging to a deployment +kind: ChaosExperiment +metadata: + name: node-memory-hog + labels: + name: node-memory-hog + app.kubernetes.io/part-of: litmus + app.kubernetes.io/component: chaosexperiment + app.kubernetes.io/version: latest +spec: + definition: + scope: Cluster + permissions: + - apiGroups: + - "" + - "batch" + - "litmuschaos.io" + resources: + - "jobs" + - "pods" + - "pods/log" + - "pods/exec" + - "events" + - "chaosengines" + - "chaosexperiments" + - "chaosresults" + verbs: + - "create" + - "list" + - "get" + - "patch" + - "update" + - "delete" + - "deletecollection" + - apiGroups: + - "" + resources: + - "nodes" + verbs: + - "get" + - "list" + image: "litmuschaos/go-runner:latest" + imagePullPolicy: Always + args: + - -c + - ./experiments -name node-memory-hog + command: + - /bin/bash + env: + + - name: TOTAL_CHAOS_DURATION + value: '120' + + ## specify the size as percent of total available memory (in percentage) + ## Default value "90" + - name: MEMORY_PERCENTAGE + value: '90' + + # ENTER THE COMMA SEPARATED TARGET NODES NAME + - name: TARGET_NODES + value: '' + + # Period to wait before and after injection of chaos in sec + - name: RAMP_TIME + value: '' + + # Provide the LIB here + # Only litmus supported + - name: LIB + value: 'litmus' + + # provide lib image + - name: LIB_IMAGE + value: 'litmuschaos/go-runner:latest' + + ## percentage of total nodes to target + - name: NODES_AFFECTED_PERC + value: '' + + ## it defines the sequence of chaos execution for multiple target nodes + ## supported values: serial, parallel + - name: SEQUENCE + value: 'parallel' + + labels: + name: node-memory-hog app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job app.kubernetes.io/version: latest @@ -2045,16 +2147,17 @@ spec: app.kubernetes.io/component: experiment-job app.kubernetes.io/version: latest +--- --- apiVersion: litmuschaos.io/v1alpha1 description: message: | - Scale the application replicas and test the node autoscaling on cluster + Taint the node where application pod is scheduled kind: ChaosExperiment metadata: - name: pod-autoscaler + name: node-taint labels: - name: pod-autoscaler + name: node-taint app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: latest @@ -2070,11 +2173,11 @@ spec: resources: - "jobs" - "pods" + - "events" - "pods/log" - "pods/exec" - - "deployments" - - "statefulsets" - - "events" + - "daemonsets" + - "pods/eviction" - "chaosengines" - "chaosexperiments" - "chaosresults" @@ -2093,142 +2196,39 @@ spec: verbs: - "get" - "list" - - "create" - "patch" - "update" image: "litmuschaos/go-runner:latest" imagePullPolicy: Always args: - -c - - ./experiments -name pod-autoscaler + - ./experiments -name node-taint command: - /bin/bash env: + - name: TARGET_NODE + value: '' + - name: TOTAL_CHAOS_DURATION value: '60' - # Period to wait before and after injection of chaos in sec - - name: RAMP_TIME - value: '' - - # Number of replicas to scale - - name: REPLICA_COUNT - value: '5' - - # PROVIDE THE LIB HERE - # ONLY LITMUS SUPPORTED + # Provide the LIB here + # Only litmus supported - name: LIB value: 'litmus' - - labels: - name: pod-autoscaler - app.kubernetes.io/part-of: litmus - app.kubernetes.io/component: experiment-job - app.kubernetes.io/version: latest - ---- -apiVersion: litmuschaos.io/v1alpha1 -description: - message: "Kills a container belonging to an application pod \n" -kind: ChaosExperiment -metadata: - name: container-kill - labels: - name: container-kill - app.kubernetes.io/part-of: litmus - app.kubernetes.io/component: chaosexperiment - app.kubernetes.io/version: latest -spec: - definition: - scope: Namespaced - permissions: - - apiGroups: - - "" - - "apps" - - "batch" - - "apps.openshift.io" - - "argoproj.io" - - "litmuschaos.io" - resources: - - "jobs" - - "pods" - - "pods/log" - - "events" - - "replicationcontrollers" - - "deployments" - - "statefulsets" - - "daemonsets" - - "replicasets" - - "deploymentconfigs" - - "rollouts" - - "pods/exec" - - "chaosengines" - - "chaosexperiments" - - "chaosresults" - verbs: - - "create" - - "list" - - "get" - - "update" - - "patch" - - "delete" - - "deletecollection" - image: "litmuschaos/go-runner:latest" - imagePullPolicy: Always - args: - - -c - - ./experiments -name container-kill - command: - - /bin/bash - env: - - - name: TARGET_CONTAINER - value: '' # Period to wait before and after injection of chaos in sec - name: RAMP_TIME value: '' - # lib can be litmus or pumba - - name: LIB - value: 'litmus' - - - name: TARGET_PODS - value: '' - - # provide the chaos interval - - name: CHAOS_INTERVAL - value: '10' - - # provide the socket file path - - name: SOCKET_PATH - value: '/var/run/docker.sock' - - # provide the name of container runtime - # for litmus LIB, it supports docker, containerd, crio - # for pumba LIB, it supports docker only - - name: CONTAINER_RUNTIME - value: 'docker' - - # provide the total chaos duration - - name: TOTAL_CHAOS_DURATION - value: '20' - - ## percentage of total pods to target - - name: PODS_AFFECTED_PERC - value: '' - - - name: LIB_IMAGE - value: 'litmuschaos/go-runner:latest' - - ## it defines the sequence of chaos execution for multiple target pods - ## supported values: serial, parallel - - name: SEQUENCE - value: 'parallel' + # set taint label & effect + # key=value:effect or key:effect + - name: TAINTS + value: '' labels: - name: container-kill + name: node-taint app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job app.kubernetes.io/version: latest diff --git a/charts/openebs/experiments.yaml b/charts/openebs/experiments.yaml index c736598..a6a50c7 100644 --- a/charts/openebs/experiments.yaml +++ b/charts/openebs/experiments.yaml @@ -2,16 +2,115 @@ apiVersion: litmuschaos.io/v1alpha1 description: message: | - Kill the OpenEBS NFS provisioner container and check if pods consuming the NFS PVs continue to be available and volumes are writable (RWM mode) + Kill the pool pod and check if gets scheduled again kind: ChaosExperiment metadata: labels: litmuschaos.io/name: openebs - name: openebs-nfs-provisioner-kill + name: openebs-pool-pod-failure app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: latest - name: openebs-nfs-provisioner-kill + name: openebs-pool-pod-failure +spec: + definition: + scope: Cluster + permissions: + - apiGroups: + - "" + - "extensions" + - "apps" + - "batch" + - "litmuschaos.io" + - "openebs.io" + - "storage.k8s.io" + resources: + - "deployments" + - "replicasets" + - "jobs" + - "pods/log" + - "events" + - "pods" + - "configmaps" + - "secrets" + - "storageclasses" + - "persistentvolumeclaims" + - "cstorvolumereplicas" + - "chaosengines" + - "chaosexperiments" + - "chaosresults" + verbs: + - "create" + - "get" + - "delete" + - "list" + - "patch" + - "update" + - apiGroups: + - "" + resources: + - "nodes" + verbs: + - "get" + - "list" + image: "litmuschaos/ansible-runner:latest" + imagePullPolicy: Always + args: + - -c + - ansible-playbook ./experiments/openebs/openebs-pool-pod-failure/openebs_pool_pod_failure_ansible_logic.yml -i /etc/ansible/hosts -vv; exit 0 + command: + - /bin/bash + env: + - name: ANSIBLE_STDOUT_CALLBACK + value: default + + - name: OPENEBS_NS + value: 'openebs' + + - name: APP_PVC + value: '' + + - name: LIVENESS_APP_LABEL + value: '' + + - name: LIVENESS_APP_NAMESPACE + value: '' + + - name: CHAOS_ITERATIONS + value: '2' + + # provide the kill count + - name: KILL_COUNT + value: '' + + - name: DATA_PERSISTENCE + value: '' + + labels: + name: openebs-pool-pod-failure + app.kubernetes.io/part-of: litmus + app.kubernetes.io/component: experiment-job + app.kubernetes.io/version: latest + #configmaps: + #- name: openebs-pool-pod-failure + # mountPath: /mnt + + +--- +--- +apiVersion: litmuschaos.io/v1alpha1 +description: + message: | + OpenEBS Pool Disk Loss contains chaos to disrupt state of infra resources. Experiments can inject disk loss against openEBS pool. +kind: ChaosExperiment +metadata: + labels: + litmuschaos.io/name: openebs + name: openebs-pool-disk-loss + app.kubernetes.io/part-of: litmus + app.kubernetes.io/component: chaosexperiment + app.kubernetes.io/version: latest + name: openebs-pool-disk-loss spec: definition: scope: Cluster @@ -23,21 +122,127 @@ spec: - "batch" - "extensions" - "storage.k8s.io" + - "openebs.io" resources: - "pods" + - "jobs" + - "pods/log" + - "events" - "pods/exec" + - "cstorpools" + - "configmaps" + - "secrets" + - "storageclasses" + - "persistentvolumes" + - "persistentvolumeclaims" + - "cstorvolumereplicas" + - "chaosexperiments" + - "chaosresults" + - "chaosengines" + verbs: + - "create" + - "list" + - "get" + - "patch" + - "update" + - "delete" + image: "litmuschaos/ansible-runner:latest" + imagePullPolicy: Always + args: + - -c + - ansible-playbook ./experiments/openebs/openebs-pool-disk-loss/openebs_pool_disk_loss_ansible_logic.yml -i /etc/ansible/hosts -vv; exit 0 + command: + - /bin/bash + env: + - name: ANSIBLE_STDOUT_CALLBACK + value: 'default' + + - name: TOTAL_CHAOS_DURATION + value: '60' + + # Period to wait before injection of chaos in sec + - name: RAMP_TIME + value: '' + + # GKE and AWS supported + - name: CLOUD_PLATFORM + value: 'GKE' + + - name: PROJECT_ID + value: '' + + - name: NODE_NAME + value: '' + + - name: DISK_NAME + value: '' + + - name: ZONE_NAME + value: '' + + - name: DEVICE_NAME + value: '' + + - name: LIVENESS_APP_LABEL + value: '' + + - name: LIVENESS_APP_NAMESPACE + value: '' + + - name: DATA_PERSISTENCE + value: '' + + - name: OPENEBS_NAMESPACE + value: 'openebs' + + labels: + name: openebs-pool-disk-loss + app.kubernetes.io/part-of: litmus + app.kubernetes.io/component: experiment-job + app.kubernetes.io/version: latest + + configmaps: + - name: openebs-pool-disk-loss + mountPath: /mnt/ + + secrets: + - name: cloud-secret + mountPath: /tmp/ + +--- +--- +apiVersion: litmuschaos.io/v1alpha1 +description: + message: | + Kill all openebs control plane pod and check if gets scheduled again +kind: ChaosExperiment +metadata: + labels: + litmuschaos.io/name: openebs + name: openebs-control-plane-chaos + app.kubernetes.io/part-of: litmus + app.kubernetes.io/component: chaosexperiment + app.kubernetes.io/version: latest + name: openebs-control-plane-chaos +spec: + definition: + scope: Namespaced + permissions: + - apiGroups: + - "" + - "litmuschaos.io" + - "batch" + - "apps" + resources: + - "pods" - "pods/log" - "deployments" - "events" - "jobs" - "configmaps" - - "services" - - "persistentvolumeclaims" - - "storageclasses" - - "persistentvolumes" + - "chaosengines" - "chaosexperiments" - "chaosresults" - - "chaosengines" verbs: - "create" - "list" @@ -57,104 +262,7 @@ spec: imagePullPolicy: Always args: - -c - - ansible-playbook ./experiments/openebs/openebs-nfs-provisioner-kill/openebs_nfs_provisioner_kill_ansible_logic.yml -i /etc/ansible/hosts -vv; exit 0 - command: - - /bin/bash - env: - - name: ANSIBLE_STDOUT_CALLBACK - value: 'default' - - # NFS default container - - name: TARGET_CONTAINER - value: 'nfs-provisioner' - - # Period to wait before injection of chaos in sec - - name: RAMP_TIME - value: '' - - # It supports pumba and containerd - - name: LIB - value: 'pumba' - - # LIB_IMAGE can be - gaiaadm/pumba:0.6.5, gprasath/crictl:ci - # For pumba image use: gaiaadm/pumba:0.6.5 - # For containerd image use: gprasath/crictl:ci - - name: LIB_IMAGE - value: 'gaiaadm/pumba:0.6.5' - - # provide the chaos interval - - name: CHAOS_INTERVAL - value: '10' - - # provide the total chaos duration - - name: TOTAL_CHAOS_DURATION - value: '20' - - labels: - name: openebs-nfs-provisioner-kill - app.kubernetes.io/part-of: litmus - app.kubernetes.io/component: experiment-job - app.kubernetes.io/version: latest - configmaps: - - name: openebs-nfs-provisioner-kill - mountPath: /mnt/ - ---- -apiVersion: litmuschaos.io/v1alpha1 -description: - message: | - Network delay to pool pod belonging to a OpenEBS PVC - This experiment is using pumba lib for network chaos -kind: ChaosExperiment -metadata: - labels: - litmuschaos.io/name: openebs - name: openebs-pool-network-delay - app.kubernetes.io/part-of: litmus - app.kubernetes.io/component: chaosexperiment - app.kubernetes.io/version: latest - name: openebs-pool-network-delay -spec: - definition: - scope: Cluster - permissions: - - apiGroups: - - "" - - "apps" - - "litmuschaos.io" - - "batch" - - "extensions" - - "storage.k8s.io" - - "openebs.io" - resources: - - "pods" - - "pods/exec" - - "jobs" - - "pods/log" - - "events" - - "configmaps" - - "services" - - "persistentvolumeclaims" - - "storageclasses" - - "persistentvolumes" - - "chaosengines" - - "chaosexperiments" - - "chaosresults" - - "cstorpools" - - "cstorvolumereplicas" - - "replicasets" - verbs: - - "create" - - "get" - - "list" - - "patch" - - "update" - - "delete" - image: "litmuschaos/ansible-runner:latest" - imagePullPolicy: Always - args: - - -c - - ansible-playbook ./experiments/openebs/openebs-pool-network-delay/openebs_pool_network_delay_ansible_logic.yml -i /etc/ansible/hosts -vv; exit 0 + - ansible-playbook ./experiments/openebs/openebs-control-plane-chaos/openebs_control_plane_chaos_ansible_logic.yml -i /etc/ansible/hosts -vv; exit 0 command: - /bin/bash env: @@ -164,35 +272,122 @@ spec: - name: OPENEBS_NAMESPACE value: 'openebs' + ## Period to wait before injection of chaos + - name: RAMP_TIME + value: '' + + - name: FORCE + value: '' + + ## env var that describes the library used to execute the chaos + ## default: litmus. Supported values: litmus, powerfulseal + - name: LIB + value: '' + + labels: + name: openebs-control-plane-chaos + app.kubernetes.io/part-of: litmus + app.kubernetes.io/component: experiment-job + app.kubernetes.io/version: latest + +--- +--- +apiVersion: litmuschaos.io/v1alpha1 +description: + message: | + Kill the pool container and check if gets scheduled again +kind: ChaosExperiment +metadata: + labels: + litmuschaos.io/name: openebs + name: openebs-pool-container-failure + app.kubernetes.io/part-of: litmus + app.kubernetes.io/component: chaosexperiment + app.kubernetes.io/version: latest + name: openebs-pool-container-failure +spec: + definition: + scope: Cluster + permissions: + - apiGroups: + - "" + - "extensions" + - "apps" + - "batch" + - "litmuschaos.io" + - "openebs.io" + - "storage.k8s.io" + resources: + - "replicasets" + - "events" + - "jobs" + - "pods" + - "pods/log" + - "pods/exec" + - "configmaps" + - "secrets" + - "persistentvolumeclaims" + - "cstorvolumereplicas" + - "chaosengines" + - "chaosexperiments" + - "chaosresults" + verbs: + - "create" + - "get" + - "delete" + - "list" + - "patch" + - "update" + image: "litmuschaos/ansible-runner:latest" + imagePullPolicy: Always + args: + - -c + - ansible-playbook ./experiments/openebs/openebs-pool-container-failure/openebs_pool_container_failure_ansible_logic.yml -i /etc/ansible/hosts -vv; exit 0 + command: + - /bin/bash + env: + - name: ANSIBLE_STDOUT_CALLBACK + value: 'default' + + - name: OPENEBS_NS + value: 'openebs' + + - name: APP_PVC + value: '' + + - name: LIVENESS_APP_LABEL + value: '' + # only pumba supported # For pumba image use : gaiaadm/pumba:0.6.5 - name: LIB_IMAGE value: 'gaiaadm/pumba:0.6.5' - # in milliseconds - - name: NETWORK_DELAY - value: '60000' - - - name: TC_IMAGE - value: 'gaiadocker/iproute2' - - - name: TOTAL_CHAOS_DURATION - value: '60' # in seconds - - - name: LIVENESS_APP_LABEL - value: '' - - name: LIVENESS_APP_NAMESPACE - value: '' + value: '' + + # provide the chaos interval + - name: CHAOS_INTERVAL + value: '10' + + # provide the total chaos duration + - name: TOTAL_CHAOS_DURATION + value: '20' - name: DATA_PERSISTENCE value: '' + - name: CHAOS_ITERATIONS + value: '2' + labels: - name: openebs-pool-network-delay - app.kubernetes.io/part-of: litmus + name: openebs-pool-container-failure + app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job app.kubernetes.io/version: latest + #configmaps: + #- name: openebs-pool-container-failure + # mountPath: /mnt --- --- @@ -314,50 +509,49 @@ spec: apiVersion: litmuschaos.io/v1alpha1 description: message: | - Kill the cstor target/Jiva controller pod and check if gets created again + Kill the OpenEBS NFS provisioner container and check if pods consuming the NFS PVs continue to be available and volumes are writable (RWM mode) kind: ChaosExperiment metadata: labels: litmuschaos.io/name: openebs - name: openebs-target-pod-failure + name: openebs-nfs-provisioner-kill app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: latest - name: openebs-target-pod-failure + name: openebs-nfs-provisioner-kill spec: definition: scope: Cluster permissions: - apiGroups: - "" - - "extensions" - "apps" - - "batch" - "litmuschaos.io" + - "batch" + - "extensions" - "storage.k8s.io" resources: - - "deployments" - - "jobs" - - "events" - "pods" - - "pods/log" - "pods/exec" + - "pods/log" + - "deployments" + - "events" + - "jobs" - "configmaps" - - "secrets" - "services" - - "chaosengines" - - "chaosexperiments" - - "chaosresults" - "persistentvolumeclaims" - "storageclasses" - "persistentvolumes" + - "chaosexperiments" + - "chaosresults" + - "chaosengines" verbs: - "create" - - "get" - - "delete" - "list" + - "get" - "patch" - "update" + - "delete" - apiGroups: - "" resources: @@ -365,133 +559,36 @@ spec: verbs: - "get" - "list" - image: "litmuschaos/ansible-runner:latest" - imagePullPolicy: Always - args: - - -c - - ansible-playbook ./experiments/openebs/openebs-target-pod-failure/openebs_target_pod_failure_ansible_logic.yml -i /etc/ansible/hosts -vv; exit 0 - command: - - /bin/bash - env: - - name: ANSIBLE_STDOUT_CALLBACK - value: 'default' - - - name: OPENEBS_NAMESPACE - value: 'openebs' - - name: APP_PVC - value: '' - - - name: FORCE - value: 'true' - - - name: LIVENESS_APP_LABEL - value: '' - - - name: LIVENESS_APP_NAMESPACE - value: '' - - - name: DATA_PERSISTENCE - value: '' - - - name: TOTAL_CHAOS_DURATION - value: '60' - - # provide the kill count - - name: KILL_COUNT - value: '' - - - name: CHAOS_INTERVAL - value: '15' - - - name: DEPLOY_TYPE - value: 'deployment' - - labels: - name: openebs-target-pod-failure - app.kubernetes.io/part-of: litmus - app.kubernetes.io/component: experiment-job - app.kubernetes.io/version: latest - #configmaps: - #- name: openebs-target-pod-failure - # mountPath: /mnt - ---- ---- -apiVersion: litmuschaos.io/v1alpha1 -description: - message: | - Kill the pool container and check if gets scheduled again -kind: ChaosExperiment -metadata: - labels: - litmuschaos.io/name: openebs - name: openebs-pool-container-failure - app.kubernetes.io/part-of: litmus - app.kubernetes.io/component: chaosexperiment - app.kubernetes.io/version: latest - name: openebs-pool-container-failure -spec: - definition: - scope: Cluster - permissions: - - apiGroups: - - "" - - "extensions" - - "apps" - - "batch" - - "litmuschaos.io" - - "openebs.io" - - "storage.k8s.io" - resources: - - "replicasets" - - "events" - - "jobs" - - "pods" - - "pods/log" - - "pods/exec" - - "configmaps" - - "secrets" - - "persistentvolumeclaims" - - "cstorvolumereplicas" - - "chaosengines" - - "chaosexperiments" - - "chaosresults" - verbs: - - "create" - - "get" - - "delete" - - "list" - - "patch" - - "update" image: "litmuschaos/ansible-runner:latest" imagePullPolicy: Always args: - -c - - ansible-playbook ./experiments/openebs/openebs-pool-container-failure/openebs_pool_container_failure_ansible_logic.yml -i /etc/ansible/hosts -vv; exit 0 + - ansible-playbook ./experiments/openebs/openebs-nfs-provisioner-kill/openebs_nfs_provisioner_kill_ansible_logic.yml -i /etc/ansible/hosts -vv; exit 0 command: - /bin/bash env: - name: ANSIBLE_STDOUT_CALLBACK value: 'default' - - name: OPENEBS_NS - value: 'openebs' + # NFS default container + - name: TARGET_CONTAINER + value: 'nfs-provisioner' - - name: APP_PVC + # Period to wait before injection of chaos in sec + - name: RAMP_TIME value: '' - - name: LIVENESS_APP_LABEL - value: '' + # It supports pumba and containerd + - name: LIB + value: 'pumba' - # only pumba supported - # For pumba image use : gaiaadm/pumba:0.6.5 + # LIB_IMAGE can be - gaiaadm/pumba:0.6.5, gprasath/crictl:ci + # For pumba image use: gaiaadm/pumba:0.6.5 + # For containerd image use: gprasath/crictl:ci - name: LIB_IMAGE value: 'gaiaadm/pumba:0.6.5' - - name: LIVENESS_APP_NAMESPACE - value: '' - # provide the chaos interval - name: CHAOS_INTERVAL value: '10' @@ -500,63 +597,59 @@ spec: - name: TOTAL_CHAOS_DURATION value: '20' - - name: DATA_PERSISTENCE - value: '' - - - name: CHAOS_ITERATIONS - value: '2' - labels: - name: openebs-pool-container-failure + name: openebs-nfs-provisioner-kill app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job app.kubernetes.io/version: latest - #configmaps: - #- name: openebs-pool-container-failure - # mountPath: /mnt + configmaps: + - name: openebs-nfs-provisioner-kill + mountPath: /mnt/ ---- --- apiVersion: litmuschaos.io/v1alpha1 description: message: | - Kill the pool pod and check if gets scheduled again + Network loss to pool pod belonging to a OpenEBS PVC kind: ChaosExperiment metadata: labels: litmuschaos.io/name: openebs - name: openebs-pool-pod-failure + name: openebs-pool-network-loss app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: latest - name: openebs-pool-pod-failure + name: openebs-pool-network-loss spec: definition: scope: Cluster permissions: - apiGroups: - "" - - "extensions" - "apps" - - "batch" - "litmuschaos.io" - - "openebs.io" + - "batch" + - "extensions" - "storage.k8s.io" + - "openebs.io" resources: - - "deployments" - - "replicasets" - - "jobs" - - "pods/log" - - "events" - "pods" + - "pods/log" + - "pods/exec" + - "events" + - "jobs" - "configmaps" - - "secrets" + - "services" + - "persistentvolumeclaims" - "storageclasses" - "persistentvolumeclaims" - - "cstorvolumereplicas" + - "persistentvolumes" - "chaosengines" - "chaosexperiments" - "chaosresults" + - "cstorpools" + - "cstorvolumereplicas" + - "replicasets" verbs: - "create" - "get" @@ -564,55 +657,48 @@ spec: - "list" - "patch" - "update" - - apiGroups: - - "" - resources: - - "nodes" - verbs: - - "get" - - "list" image: "litmuschaos/ansible-runner:latest" imagePullPolicy: Always args: - -c - - ansible-playbook ./experiments/openebs/openebs-pool-pod-failure/openebs_pool_pod_failure_ansible_logic.yml -i /etc/ansible/hosts -vv; exit 0 + - ansible-playbook ./experiments/openebs/openebs-pool-network-loss/openebs_pool_network_loss_ansible_logic.yml -i /etc/ansible/hosts -vv; exit 0 command: - /bin/bash env: - name: ANSIBLE_STDOUT_CALLBACK - value: default + value: 'default' - - name: OPENEBS_NS + - name: OPENEBS_NAMESPACE value: 'openebs' - - name: APP_PVC - value: '' + # only pumba supported + # For pumba image use : gaiaadm/pumba:0.6.5 + - name: LIB_IMAGE + value: 'gaiaadm/pumba:0.6.5' + + - name: TC_IMAGE + value: 'gaiadocker/iproute2' + + - name: NETWORK_PACKET_LOSS_PERCENTAGE + value: '100' # in percentage + + - name: TOTAL_CHAOS_DURATION + value: '120' # in seconds - name: LIVENESS_APP_LABEL value: '' - name: LIVENESS_APP_NAMESPACE - value: '' - - - name: CHAOS_ITERATIONS - value: '2' - - # provide the kill count - - name: KILL_COUNT value: '' - name: DATA_PERSISTENCE value: '' labels: - name: openebs-pool-pod-failure + name: openebs-pool-network-loss app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job app.kubernetes.io/version: latest - #configmaps: - #- name: openebs-pool-pod-failure - # mountPath: /mnt - --- apiVersion: litmuschaos.io/v1alpha1 @@ -806,102 +892,21 @@ spec: #- name: openebs-target-network-loss # mountPath: /mnt ---- --- apiVersion: litmuschaos.io/v1alpha1 description: message: | - Kill all openebs control plane pod and check if gets scheduled again + Network delay to pool pod belonging to a OpenEBS PVC + This experiment is using pumba lib for network chaos kind: ChaosExperiment metadata: labels: litmuschaos.io/name: openebs - name: openebs-control-plane-chaos + name: openebs-pool-network-delay app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: latest - name: openebs-control-plane-chaos -spec: - definition: - scope: Namespaced - permissions: - - apiGroups: - - "" - - "litmuschaos.io" - - "batch" - - "apps" - resources: - - "pods" - - "pods/log" - - "deployments" - - "events" - - "jobs" - - "configmaps" - - "chaosengines" - - "chaosexperiments" - - "chaosresults" - verbs: - - "create" - - "list" - - "get" - - "patch" - - "update" - - "delete" - - apiGroups: - - "" - resources: - - "nodes" - verbs: - - "get" - - "list" - - image: "litmuschaos/ansible-runner:latest" - imagePullPolicy: Always - args: - - -c - - ansible-playbook ./experiments/openebs/openebs-control-plane-chaos/openebs_control_plane_chaos_ansible_logic.yml -i /etc/ansible/hosts -vv; exit 0 - command: - - /bin/bash - env: - - name: ANSIBLE_STDOUT_CALLBACK - value: 'default' - - - name: OPENEBS_NAMESPACE - value: 'openebs' - - ## Period to wait before injection of chaos - - name: RAMP_TIME - value: '' - - - name: FORCE - value: '' - - ## env var that describes the library used to execute the chaos - ## default: litmus. Supported values: litmus, powerfulseal - - name: LIB - value: '' - - labels: - name: openebs-control-plane-chaos - app.kubernetes.io/part-of: litmus - app.kubernetes.io/component: experiment-job - app.kubernetes.io/version: latest - ---- ---- -apiVersion: litmuschaos.io/v1alpha1 -description: - message: | - OpenEBS Pool Disk Loss contains chaos to disrupt state of infra resources. Experiments can inject disk loss against openEBS pool. -kind: ChaosExperiment -metadata: - labels: - litmuschaos.io/name: openebs - name: openebs-pool-disk-loss - app.kubernetes.io/part-of: litmus - app.kubernetes.io/component: chaosexperiment - app.kubernetes.io/version: latest - name: openebs-pool-disk-loss + name: openebs-pool-network-delay spec: definition: scope: Cluster @@ -916,127 +921,14 @@ spec: - "openebs.io" resources: - "pods" + - "pods/exec" - "jobs" - "pods/log" - "events" - - "pods/exec" - - "cstorpools" - - "configmaps" - - "secrets" - - "storageclasses" - - "persistentvolumes" - - "persistentvolumeclaims" - - "cstorvolumereplicas" - - "chaosexperiments" - - "chaosresults" - - "chaosengines" - verbs: - - "create" - - "list" - - "get" - - "patch" - - "update" - - "delete" - image: "litmuschaos/ansible-runner:latest" - imagePullPolicy: Always - args: - - -c - - ansible-playbook ./experiments/openebs/openebs-pool-disk-loss/openebs_pool_disk_loss_ansible_logic.yml -i /etc/ansible/hosts -vv; exit 0 - command: - - /bin/bash - env: - - name: ANSIBLE_STDOUT_CALLBACK - value: 'default' - - - name: TOTAL_CHAOS_DURATION - value: '60' - - # Period to wait before injection of chaos in sec - - name: RAMP_TIME - value: '' - - # GKE and AWS supported - - name: CLOUD_PLATFORM - value: 'GKE' - - - name: PROJECT_ID - value: '' - - - name: NODE_NAME - value: '' - - - name: DISK_NAME - value: '' - - - name: ZONE_NAME - value: '' - - - name: DEVICE_NAME - value: '' - - - name: LIVENESS_APP_LABEL - value: '' - - - name: LIVENESS_APP_NAMESPACE - value: '' - - - name: DATA_PERSISTENCE - value: '' - - - name: OPENEBS_NAMESPACE - value: 'openebs' - - labels: - name: openebs-pool-disk-loss - app.kubernetes.io/part-of: litmus - app.kubernetes.io/component: experiment-job - app.kubernetes.io/version: latest - - configmaps: - - name: openebs-pool-disk-loss - mountPath: /mnt/ - - secrets: - - name: cloud-secret - mountPath: /tmp/ - ---- -apiVersion: litmuschaos.io/v1alpha1 -description: - message: | - Network loss to pool pod belonging to a OpenEBS PVC -kind: ChaosExperiment -metadata: - labels: - litmuschaos.io/name: openebs - name: openebs-pool-network-loss - app.kubernetes.io/part-of: litmus - app.kubernetes.io/component: chaosexperiment - app.kubernetes.io/version: latest - name: openebs-pool-network-loss -spec: - definition: - scope: Cluster - permissions: - - apiGroups: - - "" - - "apps" - - "litmuschaos.io" - - "batch" - - "extensions" - - "storage.k8s.io" - - "openebs.io" - resources: - - "pods" - - "pods/log" - - "pods/exec" - - "events" - - "jobs" - "configmaps" - "services" - "persistentvolumeclaims" - "storageclasses" - - "persistentvolumeclaims" - "persistentvolumes" - "chaosengines" - "chaosexperiments" @@ -1047,15 +939,15 @@ spec: verbs: - "create" - "get" - - "delete" - "list" - "patch" - "update" + - "delete" image: "litmuschaos/ansible-runner:latest" imagePullPolicy: Always args: - -c - - ansible-playbook ./experiments/openebs/openebs-pool-network-loss/openebs_pool_network_loss_ansible_logic.yml -i /etc/ansible/hosts -vv; exit 0 + - ansible-playbook ./experiments/openebs/openebs-pool-network-delay/openebs_pool_network_delay_ansible_logic.yml -i /etc/ansible/hosts -vv; exit 0 command: - /bin/bash env: @@ -1070,14 +962,15 @@ spec: - name: LIB_IMAGE value: 'gaiaadm/pumba:0.6.5' + # in milliseconds + - name: NETWORK_DELAY + value: '60000' + - name: TC_IMAGE value: 'gaiadocker/iproute2' - - name: NETWORK_PACKET_LOSS_PERCENTAGE - value: '100' # in percentage - - name: TOTAL_CHAOS_DURATION - value: '120' # in seconds + value: '60' # in seconds - name: LIVENESS_APP_LABEL value: '' @@ -1089,9 +982,116 @@ spec: value: '' labels: - name: openebs-pool-network-loss - app.kubernetes.io/part-of: litmus + name: openebs-pool-network-delay + app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job app.kubernetes.io/version: latest --- +--- +apiVersion: litmuschaos.io/v1alpha1 +description: + message: | + Kill the cstor target/Jiva controller pod and check if gets created again +kind: ChaosExperiment +metadata: + labels: + litmuschaos.io/name: openebs + name: openebs-target-pod-failure + app.kubernetes.io/part-of: litmus + app.kubernetes.io/component: chaosexperiment + app.kubernetes.io/version: latest + name: openebs-target-pod-failure +spec: + definition: + scope: Cluster + permissions: + - apiGroups: + - "" + - "extensions" + - "apps" + - "batch" + - "litmuschaos.io" + - "storage.k8s.io" + resources: + - "deployments" + - "jobs" + - "events" + - "pods" + - "pods/log" + - "pods/exec" + - "configmaps" + - "secrets" + - "services" + - "chaosengines" + - "chaosexperiments" + - "chaosresults" + - "persistentvolumeclaims" + - "storageclasses" + - "persistentvolumes" + verbs: + - "create" + - "get" + - "delete" + - "list" + - "patch" + - "update" + - apiGroups: + - "" + resources: + - "nodes" + verbs: + - "get" + - "list" + image: "litmuschaos/ansible-runner:latest" + imagePullPolicy: Always + args: + - -c + - ansible-playbook ./experiments/openebs/openebs-target-pod-failure/openebs_target_pod_failure_ansible_logic.yml -i /etc/ansible/hosts -vv; exit 0 + command: + - /bin/bash + env: + - name: ANSIBLE_STDOUT_CALLBACK + value: 'default' + + - name: OPENEBS_NAMESPACE + value: 'openebs' + + - name: APP_PVC + value: '' + + - name: FORCE + value: 'true' + + - name: LIVENESS_APP_LABEL + value: '' + + - name: LIVENESS_APP_NAMESPACE + value: '' + + - name: DATA_PERSISTENCE + value: '' + + - name: TOTAL_CHAOS_DURATION + value: '60' + + # provide the kill count + - name: KILL_COUNT + value: '' + + - name: CHAOS_INTERVAL + value: '15' + + - name: DEPLOY_TYPE + value: 'deployment' + + labels: + name: openebs-target-pod-failure + app.kubernetes.io/part-of: litmus + app.kubernetes.io/component: experiment-job + app.kubernetes.io/version: latest + #configmaps: + #- name: openebs-target-pod-failure + # mountPath: /mnt + +---