diff --git a/faults/aws/experiments.yaml b/faults/aws/experiments.yaml index 720abdfc5..dbc10bb87 100644 --- a/faults/aws/experiments.yaml +++ b/faults/aws/experiments.yaml @@ -2,110 +2,12 @@ apiVersion: litmuschaos.io/v1alpha1 description: message: | - Detaching an EBS volume from EC2 instance for a certain chaos duration. -kind: ChaosExperiment -metadata: - name: ebs-loss-by-id - labels: - name: ebs-loss-by-id - app.kubernetes.io/part-of: litmus - app.kubernetes.io/component: chaosexperiment - app.kubernetes.io/version: 3.11.0 -spec: - definition: - scope: Cluster - permissions: - # Create and monitor the experiment & helper pods - - apiGroups: [""] - resources: ["pods"] - verbs: - [ - "create", - "delete", - "get", - "list", - "patch", - "update", - "deletecollection", - ] - # Performs CRUD operations on the events inside chaosengine and chaosresult - - apiGroups: [""] - resources: ["events"] - verbs: ["create", "get", "list", "patch", "update"] - # Fetch configmaps & secrets details and mount it to the experiment pod (if specified) - - apiGroups: [""] - resources: ["secrets", "configmaps"] - verbs: ["get", "list"] - # Track and get the runner, experiment, and helper pods log - - apiGroups: [""] - resources: ["pods/log"] - verbs: ["get", "list", "watch"] - # for creating and managing to execute commands inside target container - - apiGroups: [""] - resources: ["pods/exec"] - verbs: ["get", "list", "create"] - # for configuring and monitor the experiment job by the chaos-runner pod - - apiGroups: ["batch"] - resources: ["jobs"] - verbs: ["create", "list", "get", "delete", "deletecollection"] - # for creation, status polling and deletion of litmus chaos resources used within a chaos workflow - - apiGroups: ["litmuschaos.io"] - resources: ["chaosengines", "chaosexperiments", "chaosresults"] - verbs: ["create", "list", "get", "patch", "update", "delete"] - image: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" - imagePullPolicy: Always - args: - - -c - - ./experiments -name ebs-loss-by-id - command: - - /bin/bash - env: - - name: TOTAL_CHAOS_DURATION - value: "30" - - - name: CHAOS_INTERVAL - value: "30" - - # Period to wait before and after injection of chaos in sec - - name: RAMP_TIME - value: "" - - - name: EBS_VOLUME_ID - value: "" - - - name: REGION - value: "" - - - name: SEQUENCE - value: "parallel" - - - name: DEFAULT_HEALTH_CHECK - value: "false" - - # Provide the path of AWS credentials mounted from secret - - name: AWS_SHARED_CREDENTIALS_FILE - value: "/tmp/cloud_config.yml" - - labels: - name: ebs-loss-by-id - app.kubernetes.io/part-of: litmus - app.kubernetes.io/component: experiment-job - app.kubernetes.io/version: 3.11.0 - secrets: - - name: cloud-secret - mountPath: /tmp/ - ---- ---- -apiVersion: litmuschaos.io/v1alpha1 -description: - message: | - Execute AWS SSM Chaos on given EC2 instance Tag + Execute AWS SSM Chaos on given EC2 instance IDs kind: ChaosExperiment metadata: - name: aws-ssm-chaos-by-tag + name: aws-ssm-chaos-by-id labels: - name: aws-ssm-chaos-by-tag + name: aws-ssm-chaos-by-id app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 @@ -154,7 +56,7 @@ spec: imagePullPolicy: Always args: - -c - - ./experiments -name aws-ssm-chaos-by-tag + - ./experiments -name aws-ssm-chaos-by-id command: - /bin/bash env: @@ -168,9 +70,9 @@ spec: - name: RAMP_TIME value: "" - # provide tag of the target EC2 instances - # ex: team:devops (key:value) - - name: EC2_INSTANCE_TAG + # Instance ID of the target EC2 instance + # Multiple IDs can also be provided as comma separated values ex: id1,id2 + - name: EC2_INSTANCE_ID value: "" - name: REGION @@ -185,10 +87,6 @@ spec: - name: AWS_SHARED_CREDENTIALS_FILE value: "/tmp/cloud_config.yml" - # percentage of total instance to target - - name: INSTANCE_AFFECTED_PERC - value: "" - # provide the number of workers for memory stress - name: NUMBER_OF_WORKERS value: "1" @@ -197,7 +95,7 @@ spec: - name: MEMORY_PERCENTAGE value: "80" - # provide the CPU chores to comsumed + # provide the CPU chores to be consumed # 0 will consume all the available CPU cores - name: CPU_CORE value: "0" @@ -230,7 +128,7 @@ spec: value: "True" labels: - name: aws-ssm-chaos-by-tag + name: aws-ssm-chaos-by-id app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job app.kubernetes.io/version: 3.11.0 @@ -243,12 +141,12 @@ spec: apiVersion: litmuschaos.io/v1alpha1 description: message: | - Stopping an EC2 instance identified by tag. + Detaching an EBS volume from EC2 instance. kind: ChaosExperiment metadata: - name: ec2-stop-by-tag + name: ebs-loss-by-tag labels: - name: ec2-stop-by-tag + name: ebs-loss-by-tag app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 @@ -293,15 +191,11 @@ spec: - apiGroups: ["litmuschaos.io"] resources: ["chaosengines", "chaosexperiments", "chaosresults"] verbs: ["create", "list", "get", "patch", "update", "delete"] - # for experiment to perform node status checks - - apiGroups: [""] - resources: ["nodes"] - verbs: ["get", "list"] image: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" imagePullPolicy: Always args: - -c - - ./experiments -name ec2-terminate-by-tag + - ./experiments -name ebs-loss-by-tag command: - /bin/bash env: @@ -311,28 +205,21 @@ spec: - name: CHAOS_INTERVAL value: "30" - # Period to wait before and after injection of chaos in sec - name: RAMP_TIME value: "" - # Provide a common tag to target ec2 instances - - name: EC2_INSTANCE_TAG + - name: EBS_VOLUME_TAG value: "" - # enable it if the target instance is a part of self-managed nodegroup. - - name: MANAGED_NODEGROUP - value: "disable" - - name: REGION value: "" - # Target the percentage of instance filtered from tag - - name: INSTANCE_AFFECTED_PERC - value: "" - - name: SEQUENCE value: "parallel" + - name: VOLUME_AFFECTED_PERC + value: "" + - name: DEFAULT_HEALTH_CHECK value: "false" @@ -340,11 +227,8 @@ spec: - name: AWS_SHARED_CREDENTIALS_FILE value: "/tmp/cloud_config.yml" - - name: SEQUENCE - value: "parallel" - labels: - name: ec2-stop-by-tag + name: ebs-loss-by-tag app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job app.kubernetes.io/version: 3.11.0 @@ -357,12 +241,12 @@ spec: apiVersion: litmuschaos.io/v1alpha1 description: message: | - Execute AWS SSM Chaos on given EC2 instance IDs + Execute AWS SSM Chaos on given EC2 instance Tag kind: ChaosExperiment metadata: - name: aws-ssm-chaos-by-id + name: aws-ssm-chaos-by-tag labels: - name: aws-ssm-chaos-by-id + name: aws-ssm-chaos-by-tag app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 @@ -411,7 +295,7 @@ spec: imagePullPolicy: Always args: - -c - - ./experiments -name aws-ssm-chaos-by-id + - ./experiments -name aws-ssm-chaos-by-tag command: - /bin/bash env: @@ -425,9 +309,9 @@ spec: - name: RAMP_TIME value: "" - # Instance ID of the target EC2 instance - # Multiple IDs can also be provided as comma separated values ex: id1,id2 - - name: EC2_INSTANCE_ID + # provide tag of the target EC2 instances + # ex: team:devops (key:value) + - name: EC2_INSTANCE_TAG value: "" - name: REGION @@ -442,6 +326,10 @@ spec: - name: AWS_SHARED_CREDENTIALS_FILE value: "/tmp/cloud_config.yml" + # percentage of total instance to target + - name: INSTANCE_AFFECTED_PERC + value: "" + # provide the number of workers for memory stress - name: NUMBER_OF_WORKERS value: "1" @@ -450,7 +338,7 @@ spec: - name: MEMORY_PERCENTAGE value: "80" - # provide the CPU chores to be consumed + # provide the CPU chores to comsumed # 0 will consume all the available CPU cores - name: CPU_CORE value: "0" @@ -483,7 +371,7 @@ spec: value: "True" labels: - name: aws-ssm-chaos-by-id + name: aws-ssm-chaos-by-tag app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job app.kubernetes.io/version: 3.11.0 @@ -496,73 +384,92 @@ spec: apiVersion: litmuschaos.io/v1alpha1 description: message: | - It injects AZ chaos to detach the target zones from the load balancer. + Detaching an EBS volume from EC2 instance for a certain chaos duration. kind: ChaosExperiment metadata: - name: aws-az-chaos + name: ebs-loss-by-id labels: - name: aws-az-chaos + name: ebs-loss-by-id app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 spec: definition: - scope: Namespaced + scope: Cluster permissions: # Create and monitor the experiment & helper pods - apiGroups: [""] resources: ["pods"] - verbs: ["create","delete","get","list","patch","update", "deletecollection"] + verbs: + [ + "create", + "delete", + "get", + "list", + "patch", + "update", + "deletecollection", + ] # Performs CRUD operations on the events inside chaosengine and chaosresult - apiGroups: [""] resources: ["events"] - verbs: ["create","get","list","patch","update"] + verbs: ["create", "get", "list", "patch", "update"] # Fetch configmaps & secrets details and mount it to the experiment pod (if specified) - apiGroups: [""] - resources: ["secrets","configmaps"] - verbs: ["get","list",] + resources: ["secrets", "configmaps"] + verbs: ["get", "list"] # Track and get the runner, experiment, and helper pods log - apiGroups: [""] resources: ["pods/log"] - verbs: ["get","list","watch"] - # for creating and managing to execute comands inside target container + verbs: ["get", "list", "watch"] + # for creating and managing to execute commands inside target container - apiGroups: [""] resources: ["pods/exec"] - verbs: ["get","list","create"] + verbs: ["get", "list", "create"] # for configuring and monitor the experiment job by the chaos-runner pod - apiGroups: ["batch"] resources: ["jobs"] - verbs: ["create","list","get","delete","deletecollection"] + verbs: ["create", "list", "get", "delete", "deletecollection"] # for creation, status polling and deletion of litmus chaos resources used within a chaos workflow - apiGroups: ["litmuschaos.io"] - resources: ["chaosengines","chaosexperiments","chaosresults"] - verbs: ["create","list","get","patch","update","delete"] + resources: ["chaosengines", "chaosexperiments", "chaosresults"] + verbs: ["create", "list", "get", "patch", "update", "delete"] image: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" imagePullPolicy: Always args: - -c - - python3 -u experiment -name aws-az-chaos + - ./experiments -name ebs-loss-by-id command: - /bin/bash env: - name: TOTAL_CHAOS_DURATION - value: '30' + value: "30" + - name: CHAOS_INTERVAL - value: '30' - - name: LIB - value: 'litmus' - - name: LOAD_BALANCER_NAME - value: '' - - name: LOAD_BALANCER_ZONES - value: '' - - name: LOAD_BALANCERNAME_ARN - value: 'na' + value: "30" + + # Period to wait before and after injection of chaos in sec + - name: RAMP_TIME + value: "" + + - name: EBS_VOLUME_ID + value: "" + + - name: REGION + value: "" + + - name: SEQUENCE + value: "parallel" + + - name: DEFAULT_HEALTH_CHECK + value: "false" + + # Provide the path of AWS credentials mounted from secret - name: AWS_SHARED_CREDENTIALS_FILE value: "/tmp/cloud_config.yml" - - name: RAMP_TIME - value: '' + labels: - name: aws-az-chaos + name: ebs-loss-by-id app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job app.kubernetes.io/version: 3.11.0 @@ -575,12 +482,12 @@ spec: apiVersion: litmuschaos.io/v1alpha1 description: message: | - Detaching an EBS volume from EC2 instance. + Stopping an EC2 instance identified by tag. kind: ChaosExperiment metadata: - name: ebs-loss-by-tag + name: ec2-stop-by-tag labels: - name: ebs-loss-by-tag + name: ec2-stop-by-tag app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 @@ -625,11 +532,15 @@ spec: - apiGroups: ["litmuschaos.io"] resources: ["chaosengines", "chaosexperiments", "chaosresults"] verbs: ["create", "list", "get", "patch", "update", "delete"] + # for experiment to perform node status checks + - apiGroups: [""] + resources: ["nodes"] + verbs: ["get", "list"] image: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" imagePullPolicy: Always args: - -c - - ./experiments -name ebs-loss-by-tag + - ./experiments -name ec2-terminate-by-tag command: - /bin/bash env: @@ -639,21 +550,28 @@ spec: - name: CHAOS_INTERVAL value: "30" + # Period to wait before and after injection of chaos in sec - name: RAMP_TIME value: "" - - name: EBS_VOLUME_TAG + # Provide a common tag to target ec2 instances + - name: EC2_INSTANCE_TAG value: "" + # enable it if the target instance is a part of self-managed nodegroup. + - name: MANAGED_NODEGROUP + value: "disable" + - name: REGION value: "" + # Target the percentage of instance filtered from tag + - name: INSTANCE_AFFECTED_PERC + value: "" + - name: SEQUENCE value: "parallel" - - name: VOLUME_AFFECTED_PERC - value: "" - - name: DEFAULT_HEALTH_CHECK value: "false" @@ -661,8 +579,11 @@ spec: - name: AWS_SHARED_CREDENTIALS_FILE value: "/tmp/cloud_config.yml" + - name: SEQUENCE + value: "parallel" + labels: - name: ebs-loss-by-tag + name: ec2-stop-by-tag app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job app.kubernetes.io/version: 3.11.0 @@ -779,3 +700,82 @@ spec: mountPath: /tmp/ --- +--- +apiVersion: litmuschaos.io/v1alpha1 +description: + message: | + It injects AZ chaos to detach the target zones from the load balancer. +kind: ChaosExperiment +metadata: + name: aws-az-chaos + labels: + name: aws-az-chaos + app.kubernetes.io/part-of: litmus + app.kubernetes.io/component: chaosexperiment + app.kubernetes.io/version: 3.11.0 +spec: + definition: + scope: Namespaced + permissions: + # Create and monitor the experiment & helper pods + - apiGroups: [""] + resources: ["pods"] + verbs: ["create","delete","get","list","patch","update", "deletecollection"] + # Performs CRUD operations on the events inside chaosengine and chaosresult + - apiGroups: [""] + resources: ["events"] + verbs: ["create","get","list","patch","update"] + # Fetch configmaps & secrets details and mount it to the experiment pod (if specified) + - apiGroups: [""] + resources: ["secrets","configmaps"] + verbs: ["get","list",] + # Track and get the runner, experiment, and helper pods log + - apiGroups: [""] + resources: ["pods/log"] + verbs: ["get","list","watch"] + # for creating and managing to execute comands inside target container + - apiGroups: [""] + resources: ["pods/exec"] + verbs: ["get","list","create"] + # for configuring and monitor the experiment job by the chaos-runner pod + - apiGroups: ["batch"] + resources: ["jobs"] + verbs: ["create","list","get","delete","deletecollection"] + # for creation, status polling and deletion of litmus chaos resources used within a chaos workflow + - apiGroups: ["litmuschaos.io"] + resources: ["chaosengines","chaosexperiments","chaosresults"] + verbs: ["create","list","get","patch","update","delete"] + image: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" + imagePullPolicy: Always + args: + - -c + - python3 -u experiment -name aws-az-chaos + command: + - /bin/bash + env: + - name: TOTAL_CHAOS_DURATION + value: '30' + - name: CHAOS_INTERVAL + value: '30' + - name: LIB + value: 'litmus' + - name: LOAD_BALANCER_NAME + value: '' + - name: LOAD_BALANCER_ZONES + value: '' + - name: LOAD_BALANCERNAME_ARN + value: 'na' + - name: AWS_SHARED_CREDENTIALS_FILE + value: "/tmp/cloud_config.yml" + - name: RAMP_TIME + value: '' + labels: + name: aws-az-chaos + app.kubernetes.io/part-of: litmus + app.kubernetes.io/component: experiment-job + app.kubernetes.io/version: 3.11.0 + secrets: + - name: cloud-secret + mountPath: /tmp/ + +--- diff --git a/faults/gcp/experiments.yaml b/faults/gcp/experiments.yaml index f852941aa..15e241e91 100644 --- a/faults/gcp/experiments.yaml +++ b/faults/gcp/experiments.yaml @@ -2,12 +2,12 @@ apiVersion: litmuschaos.io/v1alpha1 description: message: | - Stops GCP VM instances and GKE nodes filtered by a label for a specified duration and later restarts them + Causes loss of a non-boot storage persistent disk from a GCP VM instance for a specified duration before attaching them back kind: ChaosExperiment metadata: - name: gcp-vm-instance-stop-by-label + name: gcp-vm-disk-loss labels: - name: gcp-vm-instance-stop-by-label + name: gcp-vm-disk-loss app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 @@ -48,56 +48,52 @@ spec: - apiGroups: ["litmuschaos.io"] resources: ["chaosengines", "chaosexperiments", "chaosresults"] verbs: ["create", "list", "get", "patch", "update", "delete"] - # for experiment to perform node status checks - - apiGroups: [""] - resources: ["nodes"] - verbs: ["get", "list"] image: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" imagePullPolicy: Always args: - -c - - ./experiments -name gcp-vm-instance-stop-by-label + - ./experiments -name gcp-vm-disk-loss command: - /bin/bash env: - # set chaos duration (in sec) as desired - name: TOTAL_CHAOS_DURATION value: "30" - # set chaos interval (in sec) as desired - name: CHAOS_INTERVAL value: "30" + # Period to wait before and after injection of chaos in sec + - name: RAMP_TIME + value: "" + + # parallel or serial; determines how chaos is injected - name: SEQUENCE value: "parallel" - # GCP project ID to which the VM instances belong + # set the GCP project id - name: GCP_PROJECT_ID value: "" - # Label of the target VM instance(s) - - name: INSTANCE_LABEL + # set the disk volume name(s) as comma seperated values + # eg. volume1,volume2,... + - name: DISK_VOLUME_NAMES value: "" - # Zone in which the target VM instance(s) filtered by the label exist - # all the instances should lie in a single zone + # set the disk zone(s) as comma seperated values in the corresponding + # order of DISK_VOLUME_NAME + # eg. zone1,zone2,... - name: ZONES value: "" - # enable it if the target instance is a part of a managed instance group - - name: MANAGED_INSTANCE_GROUP - value: "disable" - - # set the percentage value of the instances with the given label - # which should be targeted as part of the chaos injection - - name: INSTANCE_AFFECTED_PERC - value: "" - - name: DEFAULT_HEALTH_CHECK value: "false" + # parallel or serial; determines how chaos is injected + - name: SEQUENCE + value: "parallel" + labels: - name: gcp-vm-instance-stop-by-label + name: gcp-vm-disk-loss app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job app.kubernetes.io/version: 3.11.0 @@ -110,12 +106,12 @@ spec: apiVersion: litmuschaos.io/v1alpha1 description: message: | - Causes loss of a non-boot storage persistent disk from a GCP VM instance filtered by a label for a specified duration before attaching them back + Stops GCP VM instances and GKE nodes for a specified duration and later restarts them kind: ChaosExperiment metadata: - name: gcp-vm-disk-loss-by-label + name: gcp-vm-instance-stop labels: - name: gcp-vm-disk-loss-by-label + name: gcp-vm-instance-stop app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 @@ -156,48 +152,60 @@ spec: - apiGroups: ["litmuschaos.io"] resources: ["chaosengines", "chaosexperiments", "chaosresults"] verbs: ["create", "list", "get", "patch", "update", "delete"] + # for experiment to perform node status checks + - apiGroups: [""] + resources: ["nodes"] + verbs: ["get", "list"] image: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" imagePullPolicy: Always args: - -c - - ./experiments -name gcp-vm-disk-loss-by-label + - ./experiments -name gcp-vm-instance-stop command: - /bin/bash env: - # set chaos duration (in sec) as desired - name: TOTAL_CHAOS_DURATION value: "30" - # set chaos interval (in sec) as desired - name: CHAOS_INTERVAL value: "30" - # set the GCP project id - - name: GCP_PROJECT_ID + # parallel or serial; determines how the VM instances are terminated, all at once or one at a time + - name: SEQUENCE + value: "parallel" + + # period to wait before and after injection of chaos in sec + - name: RAMP_TIME value: "" - # set the zone in which all the disks are created - # all the disks must exist in the same zone - - name: ZONES + # enable or disable; shall be set to enable if the target instances are a part of a managed instance group + - name: MANAGED_INSTANCE_GROUP + value: "disable" + + # Instance name of the target VM instance(s) + # Multiple instance names can be provided as comma separated values ex: instance1,instance2 + - name: VM_INSTANCE_NAMES value: "" - # set the label of the target disk volumes - - name: DISK_VOLUME_LABEL + # GCP project ID to which the VM instances belong + - name: GCP_PROJECT_ID value: "" - # set the percentage value of the disks with the given label - # which should be targeted as part of the chaos injection - - name: DISK_AFFECTED_PERC + # Instance zone(s) of the target VM instance(s) + # If more than one instance is targetted, provide zone for each in the order of their + # respective instance name in VM_INSTANCE_NAME as comma separated values ex: zone1,zone2 + - name: ZONES value: "" - name: DEFAULT_HEALTH_CHECK value: "false" + # parallel or serial; determines how chaos is injected - name: SEQUENCE value: "parallel" labels: - name: gcp-vm-disk-loss-by-label + name: gcp-vm-instance-stop app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job app.kubernetes.io/version: 3.11.0 @@ -210,12 +218,12 @@ spec: apiVersion: litmuschaos.io/v1alpha1 description: message: | - Causes loss of a non-boot storage persistent disk from a GCP VM instance for a specified duration before attaching them back + Causes loss of a non-boot storage persistent disk from a GCP VM instance filtered by a label for a specified duration before attaching them back kind: ChaosExperiment metadata: - name: gcp-vm-disk-loss + name: gcp-vm-disk-loss-by-label labels: - name: gcp-vm-disk-loss + name: gcp-vm-disk-loss-by-label app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 @@ -260,48 +268,44 @@ spec: imagePullPolicy: Always args: - -c - - ./experiments -name gcp-vm-disk-loss + - ./experiments -name gcp-vm-disk-loss-by-label command: - /bin/bash env: + # set chaos duration (in sec) as desired - name: TOTAL_CHAOS_DURATION value: "30" + # set chaos interval (in sec) as desired - name: CHAOS_INTERVAL value: "30" - # Period to wait before and after injection of chaos in sec - - name: RAMP_TIME - value: "" - - # parallel or serial; determines how chaos is injected - - name: SEQUENCE - value: "parallel" - # set the GCP project id - name: GCP_PROJECT_ID value: "" - # set the disk volume name(s) as comma seperated values - # eg. volume1,volume2,... - - name: DISK_VOLUME_NAMES + # set the zone in which all the disks are created + # all the disks must exist in the same zone + - name: ZONES value: "" - # set the disk zone(s) as comma seperated values in the corresponding - # order of DISK_VOLUME_NAME - # eg. zone1,zone2,... - - name: ZONES + # set the label of the target disk volumes + - name: DISK_VOLUME_LABEL + value: "" + + # set the percentage value of the disks with the given label + # which should be targeted as part of the chaos injection + - name: DISK_AFFECTED_PERC value: "" - name: DEFAULT_HEALTH_CHECK value: "false" - # parallel or serial; determines how chaos is injected - name: SEQUENCE value: "parallel" labels: - name: gcp-vm-disk-loss + name: gcp-vm-disk-loss-by-label app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job app.kubernetes.io/version: 3.11.0 @@ -314,12 +318,12 @@ spec: apiVersion: litmuschaos.io/v1alpha1 description: message: | - Stops GCP VM instances and GKE nodes for a specified duration and later restarts them + Stops GCP VM instances and GKE nodes filtered by a label for a specified duration and later restarts them kind: ChaosExperiment metadata: - name: gcp-vm-instance-stop + name: gcp-vm-instance-stop-by-label labels: - name: gcp-vm-instance-stop + name: gcp-vm-instance-stop-by-label app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 @@ -368,52 +372,48 @@ spec: imagePullPolicy: Always args: - -c - - ./experiments -name gcp-vm-instance-stop + - ./experiments -name gcp-vm-instance-stop-by-label command: - /bin/bash env: + # set chaos duration (in sec) as desired - name: TOTAL_CHAOS_DURATION value: "30" + # set chaos interval (in sec) as desired - name: CHAOS_INTERVAL value: "30" - # parallel or serial; determines how the VM instances are terminated, all at once or one at a time - name: SEQUENCE value: "parallel" - # period to wait before and after injection of chaos in sec - - name: RAMP_TIME + # GCP project ID to which the VM instances belong + - name: GCP_PROJECT_ID value: "" - # enable or disable; shall be set to enable if the target instances are a part of a managed instance group - - name: MANAGED_INSTANCE_GROUP - value: "disable" - - # Instance name of the target VM instance(s) - # Multiple instance names can be provided as comma separated values ex: instance1,instance2 - - name: VM_INSTANCE_NAMES + # Label of the target VM instance(s) + - name: INSTANCE_LABEL value: "" - # GCP project ID to which the VM instances belong - - name: GCP_PROJECT_ID + # Zone in which the target VM instance(s) filtered by the label exist + # all the instances should lie in a single zone + - name: ZONES value: "" - # Instance zone(s) of the target VM instance(s) - # If more than one instance is targetted, provide zone for each in the order of their - # respective instance name in VM_INSTANCE_NAME as comma separated values ex: zone1,zone2 - - name: ZONES + # enable it if the target instance is a part of a managed instance group + - name: MANAGED_INSTANCE_GROUP + value: "disable" + + # set the percentage value of the instances with the given label + # which should be targeted as part of the chaos injection + - name: INSTANCE_AFFECTED_PERC value: "" - name: DEFAULT_HEALTH_CHECK value: "false" - # parallel or serial; determines how chaos is injected - - name: SEQUENCE - value: "parallel" - labels: - name: gcp-vm-instance-stop + name: gcp-vm-instance-stop-by-label app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job app.kubernetes.io/version: 3.11.0 diff --git a/faults/kubernetes/experiments.yaml b/faults/kubernetes/experiments.yaml index bb81c22e2..5bb3d379e 100644 --- a/faults/kubernetes/experiments.yaml +++ b/faults/kubernetes/experiments.yaml @@ -2,12 +2,12 @@ apiVersion: litmuschaos.io/v1alpha1 description: message: | - Pod DNS Spoof can spoof particular DNS requests in target pod container to desired target hostnames + Inject network packet corruption into application pod kind: ChaosExperiment metadata: - name: pod-dns-spoof + name: pod-network-corruption labels: - name: pod-dns-spoof + name: pod-network-corruption app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 @@ -69,9 +69,10 @@ spec: resources: ["chaosengines", "chaosexperiments", "chaosresults"] verbs: ["create", "list", "get", "patch", "update", "delete"] image: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" + imagePullPolicy: Always args: - -c - - ./experiments -name pod-dns-spoof + - ./experiments -name pod-network-corruption command: - /bin/bash env: @@ -82,8 +83,14 @@ spec: - name: LIB_IMAGE value: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" - - name: DEFAULT_HEALTH_CHECK - value: "false" + - 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 @@ -99,25 +106,40 @@ spec: - name: TARGET_PODS value: "" - # provide the name of container runtime, it supports docker, containerd, crio + # To select pods on specific node(s) + - name: NODE_LABEL + 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: "containerd" + # 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: "/run/containerd/containerd.sock" + - name: DEFAULT_HEALTH_CHECK + value: "false" + ## it defines the sequence of chaos execution for multiple target pods ## supported values: serial, parallel - name: SEQUENCE value: "parallel" - # map of the target hostnames eg. '{"abc.com":"spoofabc.com"}' . If empty no queries will be spoofed - - name: SPOOF_MAP - value: "" - labels: - experiment: pod-dns-spoof + name: pod-network-corruption app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job app.kubernetes.io/runtime-api-usage: "true" @@ -128,12 +150,12 @@ spec: apiVersion: litmuschaos.io/v1alpha1 description: message: | - Give a memory hog on a node belonging to a deployment + Kills the Docker service on the application node to check the resiliency. kind: ChaosExperiment metadata: - name: node-memory-hog + name: docker-service-kill labels: - name: node-memory-hog + name: docker-service-kill app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 @@ -186,57 +208,36 @@ spec: imagePullPolicy: Always args: - -c - - ./experiments -name node-memory-hog + - ./experiments -name docker-service-kill command: - /bin/bash env: - name: TOTAL_CHAOS_DURATION - value: "120" - - ## Specify the size as percent of total node capacity Ex: '30' - ## NOTE: for selecting this option keep MEMORY_CONSUMPTION_MEBIBYTES empty - - name: MEMORY_CONSUMPTION_PERCENTAGE - value: "" - - ## Specify the amount of memory to be consumed in mebibytes - ## NOTE: for selecting this option keep MEMORY_CONSUMPTION_PERCENTAGE empty - - name: MEMORY_CONSUMPTION_MEBIBYTES - value: "" - - - name: NUMBER_OF_WORKERS - value: "1" + value: "90" # in seconds - # ENTER THE COMMA SEPARATED TARGET NODES NAME - - name: TARGET_NODES + # Period to wait before injection of chaos in sec + - name: RAMP_TIME value: "" - name: NODE_LABEL value: "" - # Period to wait before and after injection of chaos in sec - - name: RAMP_TIME - value: "" - # provide lib image - name: LIB_IMAGE - value: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" - - - name: DEFAULT_HEALTH_CHECK - value: "false" + value: "ubuntu:16.04" - ## percentage of total nodes to target - - name: NODES_AFFECTED_PERC + # provide the target node name + - name: TARGET_NODE value: "" - ## it defines the sequence of chaos execution for multiple target nodes - ## supported values: serial, parallel - - name: SEQUENCE - value: "parallel" + - name: DEFAULT_HEALTH_CHECK + value: "false" labels: - name: node-memory-hog + name: docker-service-kill app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job + app.kubernetes.io/service-kill: "true" app.kubernetes.io/version: 3.11.0 --- @@ -244,12 +245,12 @@ spec: apiVersion: litmuschaos.io/v1alpha1 description: message: | - Give a CPU spike on a node belonging to a deployment + Kills the Kubelet service on the application node to check the resiliency. kind: ChaosExperiment metadata: - name: node-cpu-hog + name: kubelet-service-kill labels: - name: node-cpu-hog + name: kubelet-service-kill app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 @@ -302,54 +303,36 @@ spec: imagePullPolicy: Always args: - -c - - ./experiments -name node-cpu-hog + - ./experiments -name kubelet-service-kill command: - /bin/bash env: - name: TOTAL_CHAOS_DURATION - value: "60" + value: "60" # in seconds # 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: "" - - ## LOAD CPU WITH GIVEN PERCENT LOADING FOR THE CPU STRESS WORKERS. - ## 0 IS EFFECTIVELY A SLEEP (NO LOAD) AND 100 IS FULL LOADING - - name: CPU_LOAD - value: "100" - - # ENTER THE COMMA SEPARATED TARGET NODES NAME - - name: TARGET_NODES - value: "" - - name: NODE_LABEL value: "" # provide lib image - name: LIB_IMAGE - value: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" + value: "ubuntu:16.04" - ## percentage of total nodes to target - - name: NODES_AFFECTED_PERC + # provide the target node name + - name: TARGET_NODE value: "" - name: DEFAULT_HEALTH_CHECK value: "false" - ## it defines the sequence of chaos execution for multiple target nodes - ## supported values: serial, parallel - - name: SEQUENCE - value: "parallel" - labels: - name: node-cpu-hog + name: kubelet-service-kill app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job + app.kubernetes.io/service-kill: "true" app.kubernetes.io/version: 3.11.0 --- @@ -357,18 +340,18 @@ spec: apiVersion: litmuschaos.io/v1alpha1 description: message: | - Injects memory consumption on pods belonging to an app deployment + Give a CPU spike on a node belonging to a deployment kind: ChaosExperiment metadata: - name: pod-memory-hog-exec + name: node-cpu-hog labels: - name: pod-memory-hog-exec + name: node-cpu-hog app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 spec: definition: - scope: Namespaced + scope: Cluster permissions: # Create and monitor the experiment & helper pods - apiGroups: [""] @@ -399,22 +382,6 @@ spec: - apiGroups: [""] resources: ["pods/exec"] verbs: ["get", "list", "create"] - # deriving the parent/owner details of the pod(if parent is anyof {deployment, statefulset, daemonsets}) - - apiGroups: ["apps"] - resources: ["deployments", "statefulsets", "replicasets", "daemonsets"] - verbs: ["list", "get"] - # deriving the parent/owner details of the pod(if parent is deploymentConfig) - - apiGroups: ["apps.openshift.io"] - resources: ["deploymentconfigs"] - verbs: ["list", "get"] - # deriving the parent/owner details of the pod(if parent is deploymentConfig) - - apiGroups: [""] - resources: ["replicationcontrollers"] - verbs: ["get", "list"] - # deriving the parent/owner details of the pod(if parent is argo-rollouts) - - apiGroups: ["argoproj.io"] - resources: ["rollouts"] - verbs: ["list", "get"] # for configuring and monitor the experiment job by the chaos-runner pod - apiGroups: ["batch"] resources: ["jobs"] @@ -423,49 +390,60 @@ spec: - apiGroups: ["litmuschaos.io"] resources: ["chaosengines", "chaosexperiments", "chaosresults"] verbs: ["create", "list", "get", "patch", "update", "delete"] + # for experiment to perform node status checks + - apiGroups: [""] + resources: ["nodes"] + verbs: ["get", "list"] image: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" imagePullPolicy: Always args: - -c - - ./experiments -name pod-memory-hog-exec + - ./experiments -name node-cpu-hog command: - /bin/bash env: - name: TOTAL_CHAOS_DURATION value: "60" - ## enter the amount of memory in megabytes to be consumed by the application pod - - name: MEMORY_CONSUMPTION - value: "500" + # Period to wait before and after injection of chaos in sec + - name: RAMP_TIME + value: "" - ## percentage of total pods to target - - name: PODS_AFFECTED_PERC + ## 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: "" - ## Period to wait before and after injection of chaos in sec - - name: RAMP_TIME + ## LOAD CPU WITH GIVEN PERCENT LOADING FOR THE CPU STRESS WORKERS. + ## 0 IS EFFECTIVELY A SLEEP (NO LOAD) AND 100 IS FULL LOADING + - name: CPU_LOAD + value: "100" + + # ENTER THE COMMA SEPARATED TARGET NODES NAME + - name: TARGET_NODES value: "" - # The command to kill the chaos process - - name: CHAOS_KILL_COMMAND - value: "kill $(find /proc -name exe -lname '*/dd' 2>&1 | grep -v 'Permission denied' | awk -F/ '{print $(NF-1)}' | head -n 1)" + - name: NODE_LABEL + value: "" - ## it defines the sequence of chaos execution for multiple target pods - ## supported values: serial, parallel - - name: SEQUENCE - value: "parallel" + # provide lib image + - name: LIB_IMAGE + value: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" + + ## percentage of total nodes to target + - name: NODES_AFFECTED_PERC + value: "" - name: DEFAULT_HEALTH_CHECK value: "false" - - name: TARGET_CONTAINER - value: "" - - - name: TARGET_PODS - value: "" + ## it defines the sequence of chaos execution for multiple target nodes + ## supported values: serial, parallel + - name: SEQUENCE + value: "parallel" labels: - name: pod-memory-hog-exec + name: node-cpu-hog app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job app.kubernetes.io/version: 3.11.0 @@ -475,12 +453,12 @@ spec: apiVersion: litmuschaos.io/v1alpha1 description: message: | - Injects HTTP request latency on pods belonging to an app deployment + Injects network latency on pods belonging to an app deployment kind: ChaosExperiment metadata: - name: pod-http-latency + name: pod-network-latency labels: - name: pod-http-latency + name: pod-network-latency app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 @@ -545,37 +523,25 @@ spec: imagePullPolicy: Always args: - -c - - ./experiments -name pod-http-latency + - ./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.docker.scarf.sh/litmuschaos/go-runner:3.11.0" - - name: LATENCY - value: "2000" #in ms - - # port of the target service - - name: TARGET_SERVICE_PORT - value: "80" - - # toxicity is the probability of the request to be affected - # provide the percentage value in the range of 0-100 - # 0 means no request will be affected and 100 means all request will be affected - - name: TOXICITY - value: "100" - - # port on which the proxy will listen - - name: PROXY_PORT - value: "20000" + - name: TC_IMAGE + value: "gaiadocker/iproute2" - # network interface on which the proxy will listen - - name: NETWORK_INTERFACE - value: "eth0" + - name: NETWORK_LATENCY + value: "2000" #in ms - name: TOTAL_CHAOS_DURATION value: "60" # in seconds @@ -584,10 +550,10 @@ spec: - name: RAMP_TIME value: "" - - name: DEFAULT_HEALTH_CHECK - value: "false" + - name: JITTER + value: "0" - # percentage of total pods to target + ## percentage of total pods to target - name: PODS_AFFECTED_PERC value: "" @@ -600,11 +566,24 @@ spec: - name: CONTAINER_RUNTIME value: "containerd" - # provide the socket file path - - name: SOCKET_PATH - value: "/run/containerd/containerd.sock" + - name: DEFAULT_HEALTH_CHECK + value: "false" - # To select pods on specific node(s) + # 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: "/run/containerd/containerd.sock" + + # To select pods on specific node(s) - name: NODE_LABEL value: "" @@ -614,7 +593,7 @@ spec: value: "parallel" labels: - name: pod-http-latency + name: pod-network-latency app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job app.kubernetes.io/runtime-api-usage: "true" @@ -625,12 +604,12 @@ spec: apiVersion: litmuschaos.io/v1alpha1 description: message: | - IO stress on a app pods belonging to an app deployment + Injects 100% network packet loss on pods belonging to an app deployment kind: ChaosExperiment metadata: - name: pod-io-stress + name: pod-network-partition labels: - name: pod-io-stress + name: pod-network-partition app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 @@ -667,26 +646,14 @@ spec: - apiGroups: [""] resources: ["pods/exec"] verbs: ["get", "list", "create"] - # deriving the parent/owner details of the pod(if parent is anyof {deployment, statefulset, daemonsets}) - - apiGroups: ["apps"] - resources: ["deployments", "statefulsets", "replicasets", "daemonsets"] - verbs: ["list", "get"] - # deriving the parent/owner details of the pod(if parent is deploymentConfig) - - apiGroups: ["apps.openshift.io"] - resources: ["deploymentconfigs"] - verbs: ["list", "get"] - # deriving the parent/owner details of the pod(if parent is deploymentConfig) - - apiGroups: [""] - resources: ["replicationcontrollers"] - verbs: ["get", "list"] - # deriving the parent/owner details of the pod(if parent is argo-rollouts) - - apiGroups: ["argoproj.io"] - resources: ["rollouts"] - verbs: ["list", "get"] # for configuring and monitor the experiment job by the chaos-runner pod - apiGroups: ["batch"] resources: ["jobs"] verbs: ["create", "list", "get", "delete", "deletecollection"] + # performs CRUD operations on the network policies + - apiGroups: ["networking.k8s.io"] + resources: ["networkpolicies"] + verbs: ["create", "delete", "list", "get"] # for creation, status polling and deletion of litmus chaos resources used within a chaos workflow - apiGroups: ["litmuschaos.io"] resources: ["chaosengines", "chaosexperiments", "chaosresults"] @@ -695,93 +662,70 @@ spec: imagePullPolicy: Always args: - -c - - ./experiments -name pod-io-stress + - ./experiments -name pod-network-partition command: - /bin/bash env: - name: TOTAL_CHAOS_DURATION - value: "120" - - ## specify the size as percentage of free space on the file system - ## default value 90 (in percentage) - - name: FILESYSTEM_UTILIZATION_PERCENTAGE - value: "10" + value: "60" # in seconds - ## we can specify the size in Gigabyte (Gb) also in place of percentage of free space - ## NOTE: for selecting this option FILESYSTEM_UTILIZATION_PERCENTAGE should be empty - - name: FILESYSTEM_UTILIZATION_BYTES + # time period to wait before and after injection of chaos in sec + - name: RAMP_TIME value: "" - ## Total number of workers default value is 4 - - name: NUMBER_OF_WORKERS - value: "4" + - name: DEFAULT_HEALTH_CHECK + value: "false" - ## Percentage of total pods to target - - name: PODS_AFFECTED_PERC + # provide the destination ips + # chaos injection will be triggered for these destination ips + - name: DESTINATION_IPS value: "" - # provide volume mount path - - name: VOLUME_MOUNT_PATH + # provide the destination hosts + # chaos injection will be triggered for these destination hosts + - name: DESTINATION_HOSTS value: "" - - name: TARGET_CONTAINER - value: "" + # provide network policy type + # support ingress, egress, all values + - name: POLICY_TYPES + value: "all" - ## specify the comma separated target pods - - name: TARGET_PODS + # provide labels of the destination pods + - name: POD_SELECTOR value: "" - # To select pods on specific node(s) - - name: NODE_LABEL + # provide labels the destination namespaces + - name: NAMESPACE_SELECTOR value: "" - # Period to wait before and after injection of chaos in sec - - name: RAMP_TIME + # provide comma separated ports + - name: PORTS value: "" - # provide lib image - - name: LIB_IMAGE - value: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" - - - name: DEFAULT_HEALTH_CHECK - value: "false" - - ## provide the cluster runtime - - name: CONTAINER_RUNTIME - value: "containerd" - - # provide the socket file path - - name: SOCKET_PATH - value: "/run/containerd/containerd.sock" - - ## it defines the sequence of chaos execution for multiple target pods - ## supported values: serial, parallel - - name: SEQUENCE - value: "parallel" - labels: - name: pod-io-stress + name: pod-network-partition app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job - app.kubernetes.io/runtime-api-usage: "true" app.kubernetes.io/version: 3.11.0 --- --- apiVersion: litmuschaos.io/v1alpha1 description: - message: "Kills a container belonging to an application pod \n" + message: | + Scale the application replicas and test the node autoscaling on cluster kind: ChaosExperiment metadata: - name: container-kill + name: pod-autoscaler labels: - name: container-kill + name: pod-autoscaler app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 spec: definition: - scope: Namespaced + scope: Cluster permissions: # Create and monitor the experiment & helper pods - apiGroups: [""] @@ -812,22 +756,10 @@ spec: - apiGroups: [""] resources: ["pods/exec"] verbs: ["get", "list", "create"] - # deriving the parent/owner details of the pod(if parent is anyof {deployment, statefulset, daemonsets}) + # performs CRUD operations on the deployments and statefulsets - apiGroups: ["apps"] - resources: ["deployments", "statefulsets", "replicasets", "daemonsets"] - verbs: ["list", "get"] - # deriving the parent/owner details of the pod(if parent is deploymentConfig) - - apiGroups: ["apps.openshift.io"] - resources: ["deploymentconfigs"] - verbs: ["list", "get"] - # deriving the parent/owner details of the pod(if parent is deploymentConfig) - - apiGroups: [""] - resources: ["replicationcontrollers"] - verbs: ["get", "list"] - # deriving the parent/owner details of the pod(if parent is argo-rollouts) - - apiGroups: ["argoproj.io"] - resources: ["rollouts"] - verbs: ["list", "get"] + resources: ["deployments", "statefulsets"] + verbs: ["list", "get", "patch", "update"] # for configuring and monitor the experiment job by the chaos-runner pod - apiGroups: ["batch"] resources: ["jobs"] @@ -840,64 +772,28 @@ spec: imagePullPolicy: Always args: - -c - - ./experiments -name container-kill + - ./experiments -name pod-autoscaler command: - /bin/bash env: - - name: TARGET_CONTAINER - value: "" + - name: TOTAL_CHAOS_DURATION + value: "60" # Period to wait before and after injection of chaos in sec - name: RAMP_TIME value: "" - - name: TARGET_PODS - value: "" - - # provide the chaos interval - - name: CHAOS_INTERVAL - value: "10" - - - name: SIGNAL - value: "SIGKILL" - - # provide the socket file path - - name: SOCKET_PATH - value: "/run/containerd/containerd.sock" - - # provide the name of container runtime - # it supports docker, containerd, crio - - name: CONTAINER_RUNTIME - value: "containerd" - - # provide the total chaos duration - - name: TOTAL_CHAOS_DURATION - value: "20" - - ## percentage of total pods to target - - name: PODS_AFFECTED_PERC - value: "" - - # To select pods on specific node(s) - - name: NODE_LABEL - value: "" + # Number of replicas to scale + - name: REPLICA_COUNT + value: "5" - name: DEFAULT_HEALTH_CHECK value: "false" - - name: LIB_IMAGE - value: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" - - ## it defines the sequence of chaos execution for multiple target pods - ## supported values: serial, parallel - - name: SEQUENCE - value: "parallel" - labels: - name: container-kill + name: pod-autoscaler app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job - app.kubernetes.io/runtime-api-usage: "true" app.kubernetes.io/version: 3.11.0 --- @@ -905,12 +801,12 @@ spec: apiVersion: litmuschaos.io/v1alpha1 description: message: | - It injects the chaos inside the pod which modifies the header of the request/response from the provided application server to the headers provided by the user and reverts after a specified duration + Injects HTTP request latency on pods belonging to an app deployment kind: ChaosExperiment metadata: - name: pod-http-modify-header + name: pod-http-latency labels: - name: pod-http-modify-header + name: pod-http-latency app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 @@ -975,7 +871,7 @@ spec: imagePullPolicy: Always args: - -c - - ./experiments -name pod-http-modify-header + - ./experiments -name pod-http-latency command: - /bin/bash env: @@ -986,14 +882,8 @@ spec: - name: LIB_IMAGE value: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" - # map of headers to modify/add; Eg: {"X-Litmus-Test-Header": "X-Litmus-Test-Value"} - # to remove a header, just set the value to ""; Eg: {"X-Litmus-Test-Header": ""} - - name: HEADERS_MAP - value: "{}" - - # whether to modify response headers or request headers. Accepted values: request, response - - name: HEADER_MODE - value: "response" + - name: LATENCY + value: "2000" #in ms # port of the target service - name: TARGET_SERVICE_PORT @@ -1050,7 +940,7 @@ spec: value: "parallel" labels: - name: pod-http-modify-header + name: pod-http-latency app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job app.kubernetes.io/runtime-api-usage: "true" @@ -1061,18 +951,18 @@ spec: apiVersion: litmuschaos.io/v1alpha1 description: message: | - Kills the Kubelet service on the application node to check the resiliency. + It injects the chaos inside the pod which modifies the header of the request/response from the provided application server to the headers provided by the user and reverts after a specified duration kind: ChaosExperiment metadata: - name: kubelet-service-kill + name: pod-http-modify-header labels: - name: kubelet-service-kill + name: pod-http-modify-header app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 spec: definition: - scope: Cluster + scope: Namespaced permissions: # Create and monitor the experiment & helper pods - apiGroups: [""] @@ -1103,6 +993,22 @@ spec: - apiGroups: [""] resources: ["pods/exec"] verbs: ["get", "list", "create"] + # deriving the parent/owner details of the pod(if parent is anyof {deployment, statefulset, daemonsets}) + - apiGroups: ["apps"] + resources: ["deployments", "statefulsets", "replicasets", "daemonsets"] + verbs: ["list", "get"] + # deriving the parent/owner details of the pod(if parent is deploymentConfig) + - apiGroups: ["apps.openshift.io"] + resources: ["deploymentconfigs"] + verbs: ["list", "get"] + # deriving the parent/owner details of the pod(if parent is deploymentConfig) + - apiGroups: [""] + resources: ["replicationcontrollers"] + verbs: ["get", "list"] + # deriving the parent/owner details of the pod(if parent is argo-rollouts) + - apiGroups: ["argoproj.io"] + resources: ["rollouts"] + verbs: ["list", "get"] # for configuring and monitor the experiment job by the chaos-runner pod - apiGroups: ["batch"] resources: ["jobs"] @@ -1111,44 +1017,89 @@ spec: - apiGroups: ["litmuschaos.io"] resources: ["chaosengines", "chaosexperiments", "chaosresults"] verbs: ["create", "list", "get", "patch", "update", "delete"] - # for experiment to perform node status checks - - apiGroups: [""] - resources: ["nodes"] - verbs: ["get", "list"] image: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" imagePullPolicy: Always args: - -c - - ./experiments -name kubelet-service-kill + - ./experiments -name pod-http-modify-header command: - /bin/bash env: - - name: TOTAL_CHAOS_DURATION - value: "60" # in seconds - - # Period to wait before and after injection of chaos in sec - - name: RAMP_TIME - value: "" - - - name: NODE_LABEL + - name: TARGET_CONTAINER value: "" # provide lib image - name: LIB_IMAGE - value: "ubuntu:16.04" - - # provide the target node name - - name: TARGET_NODE - value: "" + value: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" - - name: DEFAULT_HEALTH_CHECK + # map of headers to modify/add; Eg: {"X-Litmus-Test-Header": "X-Litmus-Test-Value"} + # to remove a header, just set the value to ""; Eg: {"X-Litmus-Test-Header": ""} + - name: HEADERS_MAP + value: "{}" + + # whether to modify response headers or request headers. Accepted values: request, response + - name: HEADER_MODE + value: "response" + + # port of the target service + - name: TARGET_SERVICE_PORT + value: "80" + + # toxicity is the probability of the request to be affected + # provide the percentage value in the range of 0-100 + # 0 means no request will be affected and 100 means all request will be affected + - name: TOXICITY + value: "100" + + # port on which the proxy will listen + - name: PROXY_PORT + value: "20000" + + # network interface on which the proxy will listen + - name: NETWORK_INTERFACE + value: "eth0" + + - name: TOTAL_CHAOS_DURATION + value: "60" # in seconds + + # Time period to wait before and after injection of chaos in sec + - name: RAMP_TIME + value: "" + + - name: DEFAULT_HEALTH_CHECK value: "false" + # 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: "containerd" + + # provide the socket file path + - name: SOCKET_PATH + value: "/run/containerd/containerd.sock" + + # To select pods on specific node(s) + - name: NODE_LABEL + value: "" + + ## it defines the sequence of chaos execution for multiple target pods + ## supported values: serial, parallel + - name: SEQUENCE + value: "parallel" + labels: - name: kubelet-service-kill + name: pod-http-modify-header app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job - app.kubernetes.io/service-kill: "true" + app.kubernetes.io/runtime-api-usage: "true" app.kubernetes.io/version: 3.11.0 --- @@ -1156,12 +1107,12 @@ spec: apiVersion: litmuschaos.io/v1alpha1 description: message: | - Injects network latency on pods belonging to an app deployment + Injects memory consumption on pods belonging to an app deployment kind: ChaosExperiment metadata: - name: pod-network-latency + name: pod-memory-hog labels: - name: pod-network-latency + name: pod-memory-hog app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 @@ -1226,77 +1177,65 @@ spec: imagePullPolicy: Always args: - -c - - ./experiments -name pod-network-latency + - ./experiments -name pod-memory-hog command: - /bin/bash env: - - name: TARGET_CONTAINER - value: "" - - - name: NETWORK_INTERFACE - value: "eth0" - - # provide lib image - - name: LIB_IMAGE - value: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" - - - name: TC_IMAGE - value: "gaiadocker/iproute2" - - - name: NETWORK_LATENCY - value: "2000" #in ms - - name: TOTAL_CHAOS_DURATION - value: "60" # in seconds + value: "60" - # Time period to wait before and after injection of chaos in sec - - name: RAMP_TIME - value: "" + ## enter the amount of memory in megabytes to be consumed by the application pod + - name: MEMORY_CONSUMPTION + value: "500" - - name: JITTER - value: "0" + ## Number of workers to perform stress + - name: NUMBER_OF_WORKERS + value: "1" ## percentage of total pods to target - name: PODS_AFFECTED_PERC value: "" - - name: TARGET_PODS + ## Period to wait before and after injection of chaos in sec + - name: RAMP_TIME value: "" - # provide the name of container runtime - # for litmus LIB, it supports docker, containerd, crio - # for pumba LIB, it supports docker only + ## It is used in pumba lib only + - name: LIB_IMAGE + value: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" + + ## It is used in pumba lib only + - name: STRESS_IMAGE + value: "alexeiled/stress-ng:latest-ubuntu" + + ## provide the cluster runtime - name: CONTAINER_RUNTIME value: "containerd" + # provide the socket file path + - name: SOCKET_PATH + value: "/run/containerd/containerd.sock" + + ## it defines the sequence of chaos execution for multiple target pods + ## supported values: serial, parallel + - name: SEQUENCE + value: "parallel" + - name: DEFAULT_HEALTH_CHECK value: "false" - # provide the destination ips - # chaos injection will be triggered for these destination ips - - name: DESTINATION_IPS + - name: TARGET_CONTAINER value: "" - # provide the destination hosts - # chaos injection will be triggered for these destination hosts - - name: DESTINATION_HOSTS + - name: TARGET_PODS value: "" - # provide the socket file path - - name: SOCKET_PATH - value: "/run/containerd/containerd.sock" - # To select pods on specific node(s) - name: NODE_LABEL value: "" - ## it defines the sequence of chaos execution for multiple target pods - ## supported values: serial, parallel - - name: SEQUENCE - value: "parallel" - labels: - name: pod-network-latency + name: pod-memory-hog app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job app.kubernetes.io/runtime-api-usage: "true" @@ -1307,12 +1246,12 @@ spec: apiVersion: litmuschaos.io/v1alpha1 description: message: | - Inject network packet corruption into application pod + Pod DNS Spoof can spoof particular DNS requests in target pod container to desired target hostnames kind: ChaosExperiment metadata: - name: pod-network-corruption + name: pod-dns-spoof labels: - name: pod-network-corruption + name: pod-dns-spoof app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 @@ -1374,10 +1313,9 @@ spec: resources: ["chaosengines", "chaosexperiments", "chaosresults"] verbs: ["create", "list", "get", "patch", "update", "delete"] image: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" - imagePullPolicy: Always args: - -c - - ./experiments -name pod-network-corruption + - ./experiments -name pod-dns-spoof command: - /bin/bash env: @@ -1388,14 +1326,8 @@ spec: - name: LIB_IMAGE value: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" - - name: NETWORK_INTERFACE - value: "eth0" - - - name: TC_IMAGE - value: "gaiadocker/iproute2" - - - name: NETWORK_PACKET_CORRUPTION_PERCENTAGE - value: "100" #in PERCENTAGE + - name: DEFAULT_HEALTH_CHECK + value: "false" - name: TOTAL_CHAOS_DURATION value: "60" # in seconds @@ -1411,40 +1343,25 @@ spec: - name: TARGET_PODS value: "" - # To select pods on specific node(s) - - name: NODE_LABEL - value: "" - - # provide the name of container runtime - # for litmus LIB, it supports docker, containerd, crio - # for pumba LIB, it supports docker only + # provide the name of container runtime, it supports docker, containerd, crio - name: CONTAINER_RUNTIME value: "containerd" - # 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: "/run/containerd/containerd.sock" - - name: DEFAULT_HEALTH_CHECK - value: "false" - ## it defines the sequence of chaos execution for multiple target pods ## supported values: serial, parallel - name: SEQUENCE value: "parallel" + # map of the target hostnames eg. '{"abc.com":"spoofabc.com"}' . If empty no queries will be spoofed + - name: SPOOF_MAP + value: "" + labels: - name: pod-network-corruption + experiment: pod-dns-spoof app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job app.kubernetes.io/runtime-api-usage: "true" @@ -1455,18 +1372,18 @@ spec: apiVersion: litmuschaos.io/v1alpha1 description: message: | - Injects network packet duplication on pods belonging to an app deployment + Taint the node where application pod is scheduled kind: ChaosExperiment metadata: - name: pod-network-duplication + name: node-taint labels: - name: pod-network-duplication + name: node-taint app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 spec: definition: - scope: Namespaced + scope: Cluster permissions: # Create and monitor the experiment & helper pods - apiGroups: [""] @@ -1495,24 +1412,12 @@ spec: verbs: ["get", "list", "watch"] # for creating and managing to execute commands inside target container - apiGroups: [""] - resources: ["pods/exec"] + resources: ["pods/exec", "pods/eviction"] verbs: ["get", "list", "create"] - # deriving the parent/owner details of the pod(if parent is anyof {deployment, statefulset, daemonsets}) + # ignore daemonsets while draining the node - apiGroups: ["apps"] - resources: ["deployments", "statefulsets", "replicasets", "daemonsets"] - verbs: ["list", "get"] - # deriving the parent/owner details of the pod(if parent is deploymentConfig) - - apiGroups: ["apps.openshift.io"] - resources: ["deploymentconfigs"] - verbs: ["list", "get"] - # deriving the parent/owner details of the pod(if parent is deploymentConfig) - - apiGroups: [""] - resources: ["replicationcontrollers"] - verbs: ["get", "list"] - # deriving the parent/owner details of the pod(if parent is argo-rollouts) - - apiGroups: ["argoproj.io"] - resources: ["rollouts"] - verbs: ["list", "get"] + resources: ["daemonsets"] + verbs: ["list", "get", "delete"] # for configuring and monitor the experiment job by the chaos-runner pod - apiGroups: ["batch"] resources: ["jobs"] @@ -1521,44 +1426,154 @@ spec: - apiGroups: ["litmuschaos.io"] resources: ["chaosengines", "chaosexperiments", "chaosresults"] verbs: ["create", "list", "get", "patch", "update", "delete"] + # for experiment to perform node status checks + - apiGroups: [""] + resources: ["nodes"] + verbs: ["get", "list", "patch", "update"] image: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" imagePullPolicy: Always args: - -c - - ./experiments -name pod-network-duplication + - ./experiments -name node-taint command: - /bin/bash env: - - name: TOTAL_CHAOS_DURATION - value: "60" - - - name: RAMP_TIME + - name: TARGET_NODE value: "" - - name: TARGET_CONTAINER + - name: NODE_LABEL value: "" - - name: TC_IMAGE - value: "gaiadocker/iproute2" - - - name: NETWORK_INTERFACE - value: "eth0" - - - name: NETWORK_PACKET_DUPLICATION_PERCENTAGE - value: "100" # in percentage + - name: TOTAL_CHAOS_DURATION + value: "60" - - name: TARGET_PODS + # Period to wait before and after injection of chaos in sec + - name: RAMP_TIME value: "" - # To select pods on specific node(s) - - name: NODE_LABEL + # set taint label & effect + # key=value:effect or key:effect + - name: TAINTS value: "" - ## percentage of total pods to target - - name: PODS_AFFECTED_PERC - value: "" + - name: DEFAULT_HEALTH_CHECK + value: "false" - - name: LIB_IMAGE + labels: + name: node-taint + app.kubernetes.io/part-of: litmus + app.kubernetes.io/component: experiment-job + app.kubernetes.io/version: 3.11.0 + +--- +--- +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: 3.11.0 +spec: + definition: + scope: Namespaced + permissions: + # Create and monitor the experiment & helper pods + - apiGroups: [""] + resources: ["pods"] + verbs: + [ + "create", + "delete", + "get", + "list", + "patch", + "update", + "deletecollection", + ] + # Performs CRUD operations on the events inside chaosengine and chaosresult + - apiGroups: [""] + resources: ["events"] + verbs: ["create", "get", "list", "patch", "update"] + # Fetch configmaps details and mount it to the experiment pod (if specified) + - apiGroups: [""] + resources: ["configmaps"] + verbs: ["get", "list"] + # Track and get the runner, experiment, and helper pods log + - apiGroups: [""] + resources: ["pods/log"] + verbs: ["get", "list", "watch"] + # for creating and managing to execute commands inside target container + - apiGroups: [""] + resources: ["pods/exec"] + verbs: ["get", "list", "create"] + # deriving the parent/owner details of the pod(if parent is anyof {deployment, statefulset, daemonsets}) + - apiGroups: ["apps"] + resources: ["deployments", "statefulsets", "replicasets", "daemonsets"] + verbs: ["list", "get"] + # deriving the parent/owner details of the pod(if parent is deploymentConfig) + - apiGroups: ["apps.openshift.io"] + resources: ["deploymentconfigs"] + verbs: ["list", "get"] + # deriving the parent/owner details of the pod(if parent is deploymentConfig) + - apiGroups: [""] + resources: ["replicationcontrollers"] + verbs: ["get", "list"] + # deriving the parent/owner details of the pod(if parent is argo-rollouts) + - apiGroups: ["argoproj.io"] + resources: ["rollouts"] + verbs: ["list", "get"] + # for configuring and monitor the experiment job by the chaos-runner pod + - apiGroups: ["batch"] + resources: ["jobs"] + verbs: ["create", "list", "get", "delete", "deletecollection"] + # for creation, status polling and deletion of litmus chaos resources used within a chaos workflow + - apiGroups: ["litmuschaos.io"] + resources: ["chaosengines", "chaosexperiments", "chaosresults"] + verbs: ["create", "list", "get", "patch", "update", "delete"] + image: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" + imagePullPolicy: Always + 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 + + - name: TARGET_PODS + value: "" + + # To select pods on specific node(s) + - name: NODE_LABEL + value: "" + + ## percentage of total pods to target + - name: PODS_AFFECTED_PERC + value: "" + + - name: LIB_IMAGE value: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" # provide the name of container runtime @@ -1601,18 +1616,18 @@ spec: apiVersion: litmuschaos.io/v1alpha1 description: message: | - Poweroff a node in the cluster + Fillup Ephemeral Storage of a Resource kind: ChaosExperiment metadata: - name: node-poweroff + name: disk-fill labels: - name: node-poweroff + name: disk-fill app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 spec: definition: - scope: Cluster + scope: Namespaced permissions: # Create and monitor the experiment & helper pods - apiGroups: [""] @@ -1631,18 +1646,34 @@ spec: - apiGroups: [""] resources: ["events"] verbs: ["create", "get", "list", "patch", "update"] - # Fetch configmaps & secrets details and mount it to the experiment pod (if specified) + # Fetch configmaps details and mount it to the experiment pod (if specified) - apiGroups: [""] - resources: ["configmaps", "secrets"] + resources: ["configmaps"] verbs: ["get", "list"] # Track and get the runner, experiment, and helper pods log - apiGroups: [""] resources: ["pods/log"] verbs: ["get", "list", "watch"] - # for creating and managing to execute comands inside target container + # for creating and managing to execute commands inside target container - apiGroups: [""] resources: ["pods/exec"] verbs: ["get", "list", "create"] + # deriving the parent/owner details of the pod(if parent is anyof {deployment, statefulset, daemonsets}) + - apiGroups: ["apps"] + resources: ["deployments", "statefulsets", "replicasets", "daemonsets"] + verbs: ["list", "get"] + # deriving the parent/owner details of the pod(if parent is deploymentConfig) + - apiGroups: ["apps.openshift.io"] + resources: ["deploymentconfigs"] + verbs: ["list", "get"] + # deriving the parent/owner details of the pod(if parent is deploymentConfig) + - apiGroups: [""] + resources: ["replicationcontrollers"] + verbs: ["get", "list"] + # deriving the parent/owner details of the pod(if parent is argo-rollouts) + - apiGroups: ["argoproj.io"] + resources: ["rollouts"] + verbs: ["list", "get"] # for configuring and monitor the experiment job by the chaos-runner pod - apiGroups: ["batch"] resources: ["jobs"] @@ -1651,66 +1682,83 @@ spec: - apiGroups: ["litmuschaos.io"] resources: ["chaosengines", "chaosexperiments", "chaosresults"] verbs: ["create", "list", "get", "patch", "update", "delete"] - # for experiment to perform node status checks - - apiGroups: [""] - resources: ["nodes"] - verbs: ["get", "list"] image: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" imagePullPolicy: Always args: - -c - - ./experiments -name node-restart + - ./experiments -name disk-fill command: - /bin/bash env: - - name: SSH_USER - value: "root" + - name: TARGET_CONTAINER + value: "" + + - name: FILL_PERCENTAGE + value: "80" - name: TOTAL_CHAOS_DURATION value: "60" - - name: REBOOT_COMMAND - value: '-o ServerAliveInterval=1 -o ServerAliveCountMax=1 "sudo systemctl poweroff --force --force" ; true' - # Period to wait before and after injection of chaos in sec - name: RAMP_TIME value: "" - # provide lib image - - name: LIB_IMAGE - value: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" + # provide the data block size + # supported unit is KB + - name: DATA_BLOCK_SIZE + value: "256" - # ENTER THE TARGET NODE NAME - - name: TARGET_NODE + - name: TARGET_PODS value: "" + - name: EPHEMERAL_STORAGE_MEBIBYTES + value: "" + + # To select pods on specific node(s) - name: NODE_LABEL value: "" - # ENTER THE TARGET NODE IP - - name: TARGET_NODE_IP + ## percentage of total pods to target + - name: PODS_AFFECTED_PERC value: "" + - name: DEFAULT_HEALTH_CHECK + value: "false" + + - name: LIB_IMAGE + value: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" + + # provide the socket file path + - name: SOCKET_PATH + value: "/run/containerd/containerd.sock" + + # provide the name of container runtime + # it supports docker, containerd, crio + - name: CONTAINER_RUNTIME + value: "containerd" + + ## it defines the sequence of chaos execution for multiple target pods + ## supported values: serial, parallel + - name: SEQUENCE + value: "parallel" + labels: - name: node-poweroff + name: disk-fill app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job + app.kubernetes.io/host-path-usage: "true" app.kubernetes.io/version: 3.11.0 - secrets: - - name: id-rsa - mountPath: /mnt/ --- --- apiVersion: litmuschaos.io/v1alpha1 description: - message: | - Pod DNS Error injects DNS failure/error in target pod containers + message: "Kills a container belonging to an application pod \n" kind: ChaosExperiment metadata: - name: pod-dns-error + name: container-kill labels: - name: pod-dns-error + name: container-kill app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 @@ -1772,59 +1820,64 @@ spec: resources: ["chaosengines", "chaosexperiments", "chaosresults"] verbs: ["create", "list", "get", "patch", "update", "delete"] image: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" + imagePullPolicy: Always args: - -c - - ./experiments -name pod-dns-error + - ./experiments -name container-kill command: - /bin/bash env: - name: TARGET_CONTAINER value: "" - # provide lib image - - name: LIB_IMAGE - value: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" + # Period to wait before and after injection of chaos in sec + - name: RAMP_TIME + value: "" - - name: DEFAULT_HEALTH_CHECK - value: "false" + - name: TARGET_PODS + value: "" - - name: TOTAL_CHAOS_DURATION - value: "60" # in seconds + # provide the chaos interval + - name: CHAOS_INTERVAL + value: "10" - # Time period to wait before and after injection of chaos in sec - - name: RAMP_TIME - value: "" + - name: SIGNAL + value: "SIGKILL" + + # provide the socket file path + - name: SOCKET_PATH + value: "/run/containerd/containerd.sock" + + # provide the name of container runtime + # it supports docker, containerd, crio + - name: CONTAINER_RUNTIME + value: "containerd" + + # provide the total chaos duration + - name: TOTAL_CHAOS_DURATION + value: "20" ## percentage of total pods to target - name: PODS_AFFECTED_PERC value: "" - - name: TARGET_PODS + # To select pods on specific node(s) + - name: NODE_LABEL value: "" - # provide the name of container runtime, it supports docker, containerd, crio - - name: CONTAINER_RUNTIME - value: "containerd" + - name: DEFAULT_HEALTH_CHECK + value: "false" - # provide the socket file path - - name: SOCKET_PATH - value: "/run/containerd/containerd.sock" + - name: LIB_IMAGE + value: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" ## it defines the sequence of chaos execution for multiple target pods ## supported values: serial, parallel - name: SEQUENCE value: "parallel" - # list of the target hostnames or kewywords eg. '["litmuschaos","chaosnative.io"]' . If empty all hostnames are targets - - name: TARGET_HOSTNAMES - value: "" - - # can be either exact or substring, determines whether the DNS query has to match exactly with one of the targets or can have any of the targets as substring - - name: MATCH_SCHEME - value: "exact" - labels: - experiment: pod-dns-error + name: container-kill app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job app.kubernetes.io/runtime-api-usage: "true" @@ -1835,18 +1888,18 @@ spec: apiVersion: litmuschaos.io/v1alpha1 description: message: | - Scale the application replicas and test the node autoscaling on cluster + Deletes a pod belonging to a deployment/statefulset/daemonset kind: ChaosExperiment metadata: - name: pod-autoscaler + name: pod-delete labels: - name: pod-autoscaler + name: pod-delete app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 spec: definition: - scope: Cluster + scope: Namespaced permissions: # Create and monitor the experiment & helper pods - apiGroups: [""] @@ -1877,10 +1930,22 @@ spec: - apiGroups: [""] resources: ["pods/exec"] verbs: ["get", "list", "create"] - # performs CRUD operations on the deployments and statefulsets + # deriving the parent/owner details of the pod(if parent is anyof {deployment, statefulset, daemonsets}) - apiGroups: ["apps"] - resources: ["deployments", "statefulsets"] - verbs: ["list", "get", "patch", "update"] + resources: ["deployments", "statefulsets", "replicasets", "daemonsets"] + verbs: ["list", "get"] + # deriving the parent/owner details of the pod(if parent is deploymentConfig) + - apiGroups: ["apps.openshift.io"] + resources: ["deploymentconfigs"] + verbs: ["list", "get"] + # deriving the parent/owner details of the pod(if parent is deploymentConfig) + - apiGroups: [""] + resources: ["replicationcontrollers"] + verbs: ["get", "list"] + # deriving the parent/owner details of the pod(if parent is argo-rollouts) + - apiGroups: ["argoproj.io"] + resources: ["rollouts"] + verbs: ["list", "get"] # for configuring and monitor the experiment job by the chaos-runner pod - apiGroups: ["batch"] resources: ["jobs"] @@ -1893,26 +1958,47 @@ spec: imagePullPolicy: Always args: - -c - - ./experiments -name pod-autoscaler + - ./experiments -name pod-delete command: - /bin/bash env: - name: TOTAL_CHAOS_DURATION - value: "60" + value: "15" # Period to wait before and after injection of chaos in sec - name: RAMP_TIME value: "" - # Number of replicas to scale - - name: REPLICA_COUNT + - name: FORCE + value: "true" + + - name: CHAOS_INTERVAL value: "5" + ## percentage of total pods to target + - name: PODS_AFFECTED_PERC + value: "" + + - name: TARGET_CONTAINER + value: "" + + - name: TARGET_PODS + value: "" + - name: DEFAULT_HEALTH_CHECK value: "false" + # To select pods on specific node(s) + - name: NODE_LABEL + value: "" + + ## it defines the sequence of chaos execution for multiple target pods + ## supported values: serial, parallel + - name: SEQUENCE + value: "parallel" + labels: - name: pod-autoscaler + name: pod-delete app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job app.kubernetes.io/version: 3.11.0 @@ -1922,12 +2008,12 @@ spec: apiVersion: litmuschaos.io/v1alpha1 description: message: | - Kills the Docker service on the application node to check the resiliency. + Drain the node where application pod is scheduled kind: ChaosExperiment metadata: - name: docker-service-kill + name: node-drain labels: - name: docker-service-kill + name: node-drain app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 @@ -1962,8 +2048,12 @@ spec: verbs: ["get", "list", "watch"] # for creating and managing to execute commands inside target container - apiGroups: [""] - resources: ["pods/exec"] + resources: ["pods/exec", "pods/eviction"] verbs: ["get", "list", "create"] + # ignore daemonsets while draining the node + - apiGroups: ["apps"] + resources: ["daemonsets"] + verbs: ["list", "get", "delete"] # for configuring and monitor the experiment job by the chaos-runner pod - apiGroups: ["batch"] resources: ["jobs"] @@ -1975,41 +2065,35 @@ spec: # for experiment to perform node status checks - apiGroups: [""] resources: ["nodes"] - verbs: ["get", "list"] + verbs: ["get", "list", "patch"] image: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" imagePullPolicy: Always args: - -c - - ./experiments -name docker-service-kill + - ./experiments -name node-drain command: - /bin/bash env: - - name: TOTAL_CHAOS_DURATION - value: "90" # in seconds - - # Period to wait before injection of chaos in sec - - name: RAMP_TIME + - name: TARGET_NODE value: "" - name: NODE_LABEL value: "" - # provide lib image - - name: LIB_IMAGE - value: "ubuntu:16.04" + - name: TOTAL_CHAOS_DURATION + value: "60" - # provide the target node name - - name: TARGET_NODE + # Period to wait before and after injection of chaos in sec + - name: RAMP_TIME value: "" - name: DEFAULT_HEALTH_CHECK value: "false" labels: - name: docker-service-kill + name: node-drain app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job - app.kubernetes.io/service-kill: "true" app.kubernetes.io/version: 3.11.0 --- @@ -2188,12 +2272,12 @@ spec: apiVersion: litmuschaos.io/v1alpha1 description: message: | - Deletes a pod belonging to a deployment/statefulset/daemonset + Pod DNS Error injects DNS failure/error in target pod containers kind: ChaosExperiment metadata: - name: pod-delete + name: pod-dns-error labels: - name: pod-delete + name: pod-dns-error app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 @@ -2255,145 +2339,62 @@ spec: resources: ["chaosengines", "chaosexperiments", "chaosresults"] verbs: ["create", "list", "get", "patch", "update", "delete"] image: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" - imagePullPolicy: Always args: - -c - - ./experiments -name pod-delete + - ./experiments -name pod-dns-error command: - /bin/bash env: + - name: TARGET_CONTAINER + value: "" + + # provide lib image + - name: LIB_IMAGE + value: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" + + - name: DEFAULT_HEALTH_CHECK + value: "false" + - name: TOTAL_CHAOS_DURATION - value: "15" + value: "60" # in seconds - # Period to wait before and after injection of chaos in sec + # Time 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: TARGET_CONTAINER - value: "" - - name: TARGET_PODS value: "" - - name: DEFAULT_HEALTH_CHECK - value: "false" + # provide the name of container runtime, it supports docker, containerd, crio + - name: CONTAINER_RUNTIME + value: "containerd" - # To select pods on specific node(s) - - name: NODE_LABEL - value: "" + # provide the socket file path + - name: SOCKET_PATH + value: "/run/containerd/containerd.sock" ## 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: 3.11.0 - ---- ---- -apiVersion: litmuschaos.io/v1alpha1 -description: - message: | - Drain the node where application pod is scheduled -kind: ChaosExperiment -metadata: - name: node-drain - labels: - name: node-drain - app.kubernetes.io/part-of: litmus - app.kubernetes.io/component: chaosexperiment - app.kubernetes.io/version: 3.11.0 -spec: - definition: - scope: Cluster - permissions: - # Create and monitor the experiment & helper pods - - apiGroups: [""] - resources: ["pods"] - verbs: - [ - "create", - "delete", - "get", - "list", - "patch", - "update", - "deletecollection", - ] - # Performs CRUD operations on the events inside chaosengine and chaosresult - - apiGroups: [""] - resources: ["events"] - verbs: ["create", "get", "list", "patch", "update"] - # Fetch configmaps details and mount it to the experiment pod (if specified) - - apiGroups: [""] - resources: ["configmaps"] - verbs: ["get", "list"] - # Track and get the runner, experiment, and helper pods log - - apiGroups: [""] - resources: ["pods/log"] - verbs: ["get", "list", "watch"] - # for creating and managing to execute commands inside target container - - apiGroups: [""] - resources: ["pods/exec", "pods/eviction"] - verbs: ["get", "list", "create"] - # ignore daemonsets while draining the node - - apiGroups: ["apps"] - resources: ["daemonsets"] - verbs: ["list", "get", "delete"] - # for configuring and monitor the experiment job by the chaos-runner pod - - apiGroups: ["batch"] - resources: ["jobs"] - verbs: ["create", "list", "get", "delete", "deletecollection"] - # for creation, status polling and deletion of litmus chaos resources used within a chaos workflow - - apiGroups: ["litmuschaos.io"] - resources: ["chaosengines", "chaosexperiments", "chaosresults"] - verbs: ["create", "list", "get", "patch", "update", "delete"] - # for experiment to perform node status checks - - apiGroups: [""] - resources: ["nodes"] - verbs: ["get", "list", "patch"] - image: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" - imagePullPolicy: Always - args: - - -c - - ./experiments -name node-drain - command: - - /bin/bash - env: - - name: TARGET_NODE - value: "" - - - name: NODE_LABEL - value: "" - - - name: TOTAL_CHAOS_DURATION - value: "60" - - # Period to wait before and after injection of chaos in sec - - name: RAMP_TIME + # list of the target hostnames or kewywords eg. '["litmuschaos","chaosnative.io"]' . If empty all hostnames are targets + - name: TARGET_HOSTNAMES value: "" - - name: DEFAULT_HEALTH_CHECK - value: "false" + # can be either exact or substring, determines whether the DNS query has to match exactly with one of the targets or can have any of the targets as substring + - name: MATCH_SCHEME + value: "exact" labels: - name: node-drain + experiment: pod-dns-error app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job + app.kubernetes.io/runtime-api-usage: "true" app.kubernetes.io/version: 3.11.0 --- @@ -2401,12 +2402,12 @@ spec: apiVersion: litmuschaos.io/v1alpha1 description: message: | - Injects memory consumption on pods belonging to an app deployment + Injects network packet loss on pods belonging to an app deployment kind: ChaosExperiment metadata: - name: pod-memory-hog + name: pod-network-loss labels: - name: pod-memory-hog + name: pod-network-loss app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 @@ -2471,41 +2472,63 @@ spec: imagePullPolicy: Always args: - -c - - ./experiments -name pod-memory-hog + - ./experiments -name pod-network-loss command: - /bin/bash env: - - name: TOTAL_CHAOS_DURATION - value: "60" + - name: TARGET_CONTAINER + value: "" - ## enter the amount of memory in megabytes to be consumed by the application pod - - name: MEMORY_CONSUMPTION - value: "500" + # provide lib image + - name: LIB_IMAGE + value: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" - ## Number of workers to perform stress - - name: NUMBER_OF_WORKERS - value: "1" + - name: NETWORK_INTERFACE + value: "eth0" + + - name: TC_IMAGE + value: "gaiadocker/iproute2" + + - name: NETWORK_PACKET_LOSS_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: "" ## percentage of total pods to target - name: PODS_AFFECTED_PERC value: "" - ## Period to wait before and after injection of chaos in sec - - name: RAMP_TIME + - name: DEFAULT_HEALTH_CHECK + value: "false" + + - name: TARGET_PODS value: "" - ## It is used in pumba lib only - - name: LIB_IMAGE - value: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" - - ## It is used in pumba lib only - - name: STRESS_IMAGE - value: "alexeiled/stress-ng:latest-ubuntu" + # To select pods on specific node(s) + - name: NODE_LABEL + value: "" - ## provide the cluster runtime + # 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: "containerd" + # 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: "/run/containerd/containerd.sock" @@ -2515,21 +2538,8 @@ spec: - name: SEQUENCE value: "parallel" - - name: DEFAULT_HEALTH_CHECK - value: "false" - - - name: TARGET_CONTAINER - value: "" - - - name: TARGET_PODS - value: "" - - # To select pods on specific node(s) - - name: NODE_LABEL - value: "" - labels: - name: pod-memory-hog + name: pod-network-loss app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job app.kubernetes.io/runtime-api-usage: "true" @@ -2540,18 +2550,18 @@ spec: apiVersion: litmuschaos.io/v1alpha1 description: message: | - Taint the node where application pod is scheduled + Injects CPU consumption on pods belonging to an app deployment kind: ChaosExperiment metadata: - name: node-taint + name: pod-cpu-hog-exec labels: - name: node-taint + name: pod-cpu-hog-exec app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 spec: definition: - scope: Cluster + scope: Namespaced permissions: # Create and monitor the experiment & helper pods - apiGroups: [""] @@ -2580,12 +2590,24 @@ spec: verbs: ["get", "list", "watch"] # for creating and managing to execute commands inside target container - apiGroups: [""] - resources: ["pods/exec", "pods/eviction"] + resources: ["pods/exec"] verbs: ["get", "list", "create"] - # ignore daemonsets while draining the node + # deriving the parent/owner details of the pod(if parent is anyof {deployment, statefulset, daemonsets}) - apiGroups: ["apps"] - resources: ["daemonsets"] - verbs: ["list", "get", "delete"] + resources: ["deployments", "statefulsets", "replicasets", "daemonsets"] + verbs: ["list", "get"] + # deriving the parent/owner details of the pod(if parent is deploymentConfig) + - apiGroups: ["apps.openshift.io"] + resources: ["deploymentconfigs"] + verbs: ["list", "get"] + # deriving the parent/owner details of the pod(if parent is deploymentConfig) + - apiGroups: [""] + resources: ["replicationcontrollers"] + verbs: ["get", "list"] + # deriving the parent/owner details of the pod(if parent is argo-rollouts) + - apiGroups: ["argoproj.io"] + resources: ["rollouts"] + verbs: ["list", "get"] # for configuring and monitor the experiment job by the chaos-runner pod - apiGroups: ["batch"] resources: ["jobs"] @@ -2594,41 +2616,49 @@ spec: - apiGroups: ["litmuschaos.io"] resources: ["chaosengines", "chaosexperiments", "chaosresults"] verbs: ["create", "list", "get", "patch", "update", "delete"] - # for experiment to perform node status checks - - apiGroups: [""] - resources: ["nodes"] - verbs: ["get", "list", "patch", "update"] image: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" imagePullPolicy: Always args: - -c - - ./experiments -name node-taint + - ./experiments -name pod-cpu-hog-exec command: - /bin/bash env: - - name: TARGET_NODE + - name: TOTAL_CHAOS_DURATION + value: "60" + + ## Number of CPU cores to stress + - name: CPU_CORES + value: "1" + + ## Percentage of total pods to target + - name: PODS_AFFECTED_PERC value: "" - - name: NODE_LABEL + ## Period to wait before and after injection of chaos in sec + - name: RAMP_TIME value: "" - - name: TOTAL_CHAOS_DURATION - value: "60" + # The command to kill the chaos process + - name: CHAOS_KILL_COMMAND + value: "kill $(find /proc -name exe -lname '*/md5sum' 2>&1 | grep -v 'Permission denied' | awk -F/ '{print $(NF-1)}')" - # Period to wait before and after injection of chaos in sec - - name: RAMP_TIME + - name: TARGET_CONTAINER value: "" - # set taint label & effect - # key=value:effect or key:effect - - name: TAINTS + - name: TARGET_PODS value: "" - name: DEFAULT_HEALTH_CHECK value: "false" + ## it defines the sequence of chaos execution for multiple target pods + ## supported values: serial, parallel + - name: SEQUENCE + value: "parallel" + labels: - name: node-taint + name: pod-cpu-hog-exec app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job app.kubernetes.io/version: 3.11.0 @@ -2638,12 +2668,12 @@ spec: apiVersion: litmuschaos.io/v1alpha1 description: message: | - Injects CPU consumption on pods belonging to an app deployment + Injects memory consumption on pods belonging to an app deployment kind: ChaosExperiment metadata: - name: pod-cpu-hog + name: pod-memory-hog-exec labels: - name: pod-cpu-hog + name: pod-memory-hog-exec app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 @@ -2708,23 +2738,18 @@ spec: imagePullPolicy: Always args: - -c - - ./experiments -name pod-cpu-hog + - ./experiments -name pod-memory-hog-exec command: - /bin/bash env: - name: TOTAL_CHAOS_DURATION value: "60" - ## Number of CPU cores to stress - - name: CPU_CORES - value: "1" - - ## LOAD CPU WITH GIVEN PERCENT LOADING FOR THE CPU STRESS WORKERS. - ## 0 IS EFFECTIVELY A SLEEP (NO LOAD) AND 100 IS FULL LOADING - - name: CPU_LOAD - value: "100" + ## enter the amount of memory in megabytes to be consumed by the application pod + - name: MEMORY_CONSUMPTION + value: "500" - ## Percentage of total pods to target + ## percentage of total pods to target - name: PODS_AFFECTED_PERC value: "" @@ -2732,21 +2757,17 @@ spec: - name: RAMP_TIME value: "" - ## It is used in pumba lib only - - name: LIB_IMAGE - value: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" - - ## It is used in pumba lib only - - name: STRESS_IMAGE - value: "alexeiled/stress-ng:latest-ubuntu" + # The command to kill the chaos process + - name: CHAOS_KILL_COMMAND + value: "kill $(find /proc -name exe -lname '*/dd' 2>&1 | grep -v 'Permission denied' | awk -F/ '{print $(NF-1)}' | head -n 1)" - ## provide the cluster runtime - - name: CONTAINER_RUNTIME - value: "containerd" + ## it defines the sequence of chaos execution for multiple target pods + ## supported values: serial, parallel + - name: SEQUENCE + value: "parallel" - # provide the socket file path - - name: SOCKET_PATH - value: "/run/containerd/containerd.sock" + - name: DEFAULT_HEALTH_CHECK + value: "false" - name: TARGET_CONTAINER value: "" @@ -2754,23 +2775,10 @@ spec: - name: TARGET_PODS value: "" - - name: DEFAULT_HEALTH_CHECK - value: "false" - - # To select pods on specific node(s) - - name: NODE_LABEL - value: "" - - ## it defines the sequence of chaos execution for multiple target pods - ## supported values: serial, parallel - - name: SEQUENCE - value: "parallel" - labels: - name: pod-cpu-hog + name: pod-memory-hog-exec app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job - app.kubernetes.io/runtime-api-usage: "true" app.kubernetes.io/version: 3.11.0 --- @@ -2778,12 +2786,12 @@ spec: apiVersion: litmuschaos.io/v1alpha1 description: message: | - Injects 100% network packet loss on pods belonging to an app deployment + IO stress on a app pods belonging to an app deployment kind: ChaosExperiment metadata: - name: pod-network-partition + name: pod-io-stress labels: - name: pod-network-partition + name: pod-io-stress app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 @@ -2820,14 +2828,26 @@ spec: - apiGroups: [""] resources: ["pods/exec"] verbs: ["get", "list", "create"] + # deriving the parent/owner details of the pod(if parent is anyof {deployment, statefulset, daemonsets}) + - apiGroups: ["apps"] + resources: ["deployments", "statefulsets", "replicasets", "daemonsets"] + verbs: ["list", "get"] + # deriving the parent/owner details of the pod(if parent is deploymentConfig) + - apiGroups: ["apps.openshift.io"] + resources: ["deploymentconfigs"] + verbs: ["list", "get"] + # deriving the parent/owner details of the pod(if parent is deploymentConfig) + - apiGroups: [""] + resources: ["replicationcontrollers"] + verbs: ["get", "list"] + # deriving the parent/owner details of the pod(if parent is argo-rollouts) + - apiGroups: ["argoproj.io"] + resources: ["rollouts"] + verbs: ["list", "get"] # for configuring and monitor the experiment job by the chaos-runner pod - apiGroups: ["batch"] resources: ["jobs"] verbs: ["create", "list", "get", "delete", "deletecollection"] - # performs CRUD operations on the network policies - - apiGroups: ["networking.k8s.io"] - resources: ["networkpolicies"] - verbs: ["create", "delete", "list", "get"] # for creation, status polling and deletion of litmus chaos resources used within a chaos workflow - apiGroups: ["litmuschaos.io"] resources: ["chaosengines", "chaosexperiments", "chaosresults"] @@ -2836,51 +2856,75 @@ spec: imagePullPolicy: Always args: - -c - - ./experiments -name pod-network-partition + - ./experiments -name pod-io-stress command: - /bin/bash env: - name: TOTAL_CHAOS_DURATION - value: "60" # in seconds + value: "120" - # time period to wait before and after injection of chaos in sec - - name: RAMP_TIME + ## specify the size as percentage of free space on the file system + ## default value 90 (in percentage) + - name: FILESYSTEM_UTILIZATION_PERCENTAGE + value: "10" + + ## we can specify the size in Gigabyte (Gb) also in place of percentage of free space + ## NOTE: for selecting this option FILESYSTEM_UTILIZATION_PERCENTAGE should be empty + - name: FILESYSTEM_UTILIZATION_BYTES value: "" - - name: DEFAULT_HEALTH_CHECK - value: "false" + ## Total number of workers default value is 4 + - name: NUMBER_OF_WORKERS + value: "4" - # provide the destination ips - # chaos injection will be triggered for these destination ips - - name: DESTINATION_IPS + ## Percentage of total pods to target + - name: PODS_AFFECTED_PERC value: "" - # provide the destination hosts - # chaos injection will be triggered for these destination hosts - - name: DESTINATION_HOSTS + # provide volume mount path + - name: VOLUME_MOUNT_PATH value: "" - # provide network policy type - # support ingress, egress, all values - - name: POLICY_TYPES - value: "all" + - name: TARGET_CONTAINER + value: "" - # provide labels of the destination pods - - name: POD_SELECTOR + ## specify the comma separated target pods + - name: TARGET_PODS value: "" - # provide labels the destination namespaces - - name: NAMESPACE_SELECTOR + # To select pods on specific node(s) + - name: NODE_LABEL value: "" - # provide comma separated ports - - name: PORTS + # Period to wait before and after injection of chaos in sec + - name: RAMP_TIME value: "" + # provide lib image + - name: LIB_IMAGE + value: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" + + - name: DEFAULT_HEALTH_CHECK + value: "false" + + ## provide the cluster runtime + - name: CONTAINER_RUNTIME + value: "containerd" + + # provide the socket file path + - name: SOCKET_PATH + value: "/run/containerd/containerd.sock" + + ## it defines the sequence of chaos execution for multiple target pods + ## supported values: serial, parallel + - name: SEQUENCE + value: "parallel" + labels: - name: pod-network-partition + name: pod-io-stress app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job + app.kubernetes.io/runtime-api-usage: "true" app.kubernetes.io/version: 3.11.0 --- @@ -2888,12 +2932,12 @@ spec: apiVersion: litmuschaos.io/v1alpha1 description: message: | - Restart node + Poweroff a node in the cluster kind: ChaosExperiment metadata: - name: node-restart + name: node-poweroff labels: - name: node-restart + name: node-poweroff app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 @@ -2926,7 +2970,7 @@ spec: - apiGroups: [""] resources: ["pods/log"] verbs: ["get", "list", "watch"] - # for creating and managing to execute commands inside target container + # for creating and managing to execute comands inside target container - apiGroups: [""] resources: ["pods/exec"] verbs: ["get", "list", "create"] @@ -2952,12 +2996,13 @@ spec: env: - name: SSH_USER value: "root" - - name: REBOOT_COMMAND - value: "sudo systemctl reboot; true" - name: TOTAL_CHAOS_DURATION value: "60" + - name: REBOOT_COMMAND + value: '-o ServerAliveInterval=1 -o ServerAliveCountMax=1 "sudo systemctl poweroff --force --force" ; true' + # Period to wait before and after injection of chaos in sec - name: RAMP_TIME value: "" @@ -2977,11 +3022,8 @@ spec: - name: TARGET_NODE_IP value: "" - - name: DEFAULT_HEALTH_CHECK - value: "false" - labels: - name: node-restart + name: node-poweroff app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job app.kubernetes.io/version: 3.11.0 @@ -2994,12 +3036,12 @@ spec: apiVersion: litmuschaos.io/v1alpha1 description: message: | - Injects network packet loss on pods belonging to an app deployment + It injects the chaos inside the pod which modifies the body of the response from the provided application server to the body string provided by the user and reverts after a specified duration kind: ChaosExperiment metadata: - name: pod-network-loss + name: pod-http-modify-body labels: - name: pod-network-loss + name: pod-http-modify-body app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 @@ -3064,7 +3106,7 @@ spec: imagePullPolicy: Always args: - -c - - ./experiments -name pod-network-loss + - ./experiments -name pod-http-modify-body command: - /bin/bash env: @@ -3075,34 +3117,54 @@ spec: - name: LIB_IMAGE value: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" - - name: NETWORK_INTERFACE - value: "eth0" + # provide the body string to overwrite the response body + # if no value is provided, response will be an empty body. + - name: RESPONSE_BODY + value: "" - - name: TC_IMAGE - value: "gaiadocker/iproute2" + # provide the encoding type for the response body + # currently supported value are gzip, deflate + # if empty no encoding will be applied + - name: CONTENT_ENCODING + value: "" - - name: NETWORK_PACKET_LOSS_PERCENTAGE - value: "100" #in PERCENTAGE + # provide the content type for the response body + - name: CONTENT_TYPE + value: "text/plain" + + # port of the target service + - name: TARGET_SERVICE_PORT + value: "80" + + # toxicity is the probability of the request to be affected + # provide the percentage value in the range of 0-100 + # 0 means no request will be affected and 100 means all request will be affected + - name: TOXICITY + value: "100" + + # port on which the proxy will listen + - name: PROXY_PORT + value: "20000" + + # network interface on which the proxy will listen + - name: NETWORK_INTERFACE + value: "eth0" - name: TOTAL_CHAOS_DURATION value: "60" # in seconds - # time period to wait before and after injection of chaos in sec + # Time period to wait before and after injection of chaos in sec - name: RAMP_TIME value: "" - ## percentage of total pods to target - - name: PODS_AFFECTED_PERC - value: "" - - name: DEFAULT_HEALTH_CHECK value: "false" - - name: TARGET_PODS + # percentage of total pods to target + - name: PODS_AFFECTED_PERC value: "" - # To select pods on specific node(s) - - name: NODE_LABEL + - name: TARGET_PODS value: "" # provide the name of container runtime @@ -3111,30 +3173,23 @@ spec: - name: CONTAINER_RUNTIME value: "containerd" - # 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: "/run/containerd/containerd.sock" + # To select pods on specific node(s) + - name: NODE_LABEL + value: "" + ## it defines the sequence of chaos execution for multiple target pods ## supported values: serial, parallel - name: SEQUENCE value: "parallel" labels: - name: pod-network-loss + name: pod-http-modify-body app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job - app.kubernetes.io/runtime-api-usage: "true" app.kubernetes.io/version: 3.11.0 --- @@ -3267,18 +3322,18 @@ spec: apiVersion: litmuschaos.io/v1alpha1 description: message: | - Fillup Ephemeral Storage of a Resource + Give a memory hog on a node belonging to a deployment kind: ChaosExperiment metadata: - name: disk-fill + name: node-memory-hog labels: - name: disk-fill + name: node-memory-hog app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 spec: definition: - scope: Namespaced + scope: Cluster permissions: # Create and monitor the experiment & helper pods - apiGroups: [""] @@ -3309,22 +3364,6 @@ spec: - apiGroups: [""] resources: ["pods/exec"] verbs: ["get", "list", "create"] - # deriving the parent/owner details of the pod(if parent is anyof {deployment, statefulset, daemonsets}) - - apiGroups: ["apps"] - resources: ["deployments", "statefulsets", "replicasets", "daemonsets"] - verbs: ["list", "get"] - # deriving the parent/owner details of the pod(if parent is deploymentConfig) - - apiGroups: ["apps.openshift.io"] - resources: ["deploymentconfigs"] - verbs: ["list", "get"] - # deriving the parent/owner details of the pod(if parent is deploymentConfig) - - apiGroups: [""] - resources: ["replicationcontrollers"] - verbs: ["get", "list"] - # deriving the parent/owner details of the pod(if parent is argo-rollouts) - - apiGroups: ["argoproj.io"] - resources: ["rollouts"] - verbs: ["list", "get"] # for configuring and monitor the experiment job by the chaos-runner pod - apiGroups: ["batch"] resources: ["jobs"] @@ -3333,84 +3372,184 @@ spec: - apiGroups: ["litmuschaos.io"] resources: ["chaosengines", "chaosexperiments", "chaosresults"] verbs: ["create", "list", "get", "patch", "update", "delete"] + # for experiment to perform node status checks + - apiGroups: [""] + resources: ["nodes"] + verbs: ["get", "list"] image: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" imagePullPolicy: Always args: - -c - - ./experiments -name disk-fill + - ./experiments -name node-memory-hog command: - /bin/bash env: - - name: TARGET_CONTAINER + - name: TOTAL_CHAOS_DURATION + value: "120" + + ## Specify the size as percent of total node capacity Ex: '30' + ## NOTE: for selecting this option keep MEMORY_CONSUMPTION_MEBIBYTES empty + - name: MEMORY_CONSUMPTION_PERCENTAGE value: "" - - name: FILL_PERCENTAGE - value: "80" + ## Specify the amount of memory to be consumed in mebibytes + ## NOTE: for selecting this option keep MEMORY_CONSUMPTION_PERCENTAGE empty + - name: MEMORY_CONSUMPTION_MEBIBYTES + value: "" - - name: TOTAL_CHAOS_DURATION - value: "60" + - name: NUMBER_OF_WORKERS + value: "1" + + # ENTER THE COMMA SEPARATED TARGET NODES NAME + - name: TARGET_NODES + value: "" + + - name: NODE_LABEL + value: "" # Period to wait before and after injection of chaos in sec - name: RAMP_TIME value: "" - # provide the data block size - # supported unit is KB - - name: DATA_BLOCK_SIZE - value: "256" + # provide lib image + - name: LIB_IMAGE + value: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" - - name: TARGET_PODS + - name: DEFAULT_HEALTH_CHECK + value: "false" + + ## 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: 3.11.0 + +--- +--- +apiVersion: litmuschaos.io/v1alpha1 +description: + message: | + Restart node +kind: ChaosExperiment +metadata: + name: node-restart + labels: + name: node-restart + app.kubernetes.io/part-of: litmus + app.kubernetes.io/component: chaosexperiment + app.kubernetes.io/version: 3.11.0 +spec: + definition: + scope: Cluster + permissions: + # Create and monitor the experiment & helper pods + - apiGroups: [""] + resources: ["pods"] + verbs: + [ + "create", + "delete", + "get", + "list", + "patch", + "update", + "deletecollection", + ] + # Performs CRUD operations on the events inside chaosengine and chaosresult + - apiGroups: [""] + resources: ["events"] + verbs: ["create", "get", "list", "patch", "update"] + # Fetch configmaps & secrets details and mount it to the experiment pod (if specified) + - apiGroups: [""] + resources: ["configmaps", "secrets"] + verbs: ["get", "list"] + # Track and get the runner, experiment, and helper pods log + - apiGroups: [""] + resources: ["pods/log"] + verbs: ["get", "list", "watch"] + # for creating and managing to execute commands inside target container + - apiGroups: [""] + resources: ["pods/exec"] + verbs: ["get", "list", "create"] + # for configuring and monitor the experiment job by the chaos-runner pod + - apiGroups: ["batch"] + resources: ["jobs"] + verbs: ["create", "list", "get", "delete", "deletecollection"] + # for creation, status polling and deletion of litmus chaos resources used within a chaos workflow + - apiGroups: ["litmuschaos.io"] + resources: ["chaosengines", "chaosexperiments", "chaosresults"] + verbs: ["create", "list", "get", "patch", "update", "delete"] + # for experiment to perform node status checks + - apiGroups: [""] + resources: ["nodes"] + verbs: ["get", "list"] + image: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" + imagePullPolicy: Always + args: + - -c + - ./experiments -name node-restart + command: + - /bin/bash + env: + - name: SSH_USER + value: "root" + - name: REBOOT_COMMAND + value: "sudo systemctl reboot; true" + + - name: TOTAL_CHAOS_DURATION + value: "60" + + # Period to wait before and after injection of chaos in sec + - name: RAMP_TIME value: "" - - name: EPHEMERAL_STORAGE_MEBIBYTES + # provide lib image + - name: LIB_IMAGE + value: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" + + # ENTER THE TARGET NODE NAME + - name: TARGET_NODE value: "" - # To select pods on specific node(s) - name: NODE_LABEL value: "" - ## percentage of total pods to target - - name: PODS_AFFECTED_PERC + # ENTER THE TARGET NODE IP + - name: TARGET_NODE_IP value: "" - name: DEFAULT_HEALTH_CHECK value: "false" - - name: LIB_IMAGE - value: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" - - # provide the socket file path - - name: SOCKET_PATH - value: "/run/containerd/containerd.sock" - - # provide the name of container runtime - # it supports docker, containerd, crio - - name: CONTAINER_RUNTIME - value: "containerd" - - ## it defines the sequence of chaos execution for multiple target pods - ## supported values: serial, parallel - - name: SEQUENCE - value: "parallel" - labels: - name: disk-fill + name: node-restart app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job - app.kubernetes.io/host-path-usage: "true" app.kubernetes.io/version: 3.11.0 + secrets: + - name: id-rsa + mountPath: /mnt/ --- --- apiVersion: litmuschaos.io/v1alpha1 description: message: | - it injects chaos into the pod which stops outgoing HTTP requests by resetting the TCP connection and then reverts back to the original state after a specified duration + Injects CPU consumption on pods belonging to an app deployment kind: ChaosExperiment metadata: - name: pod-http-reset-peer + name: pod-cpu-hog labels: - name: pod-http-reset-peer + name: pod-cpu-hog app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 @@ -3475,59 +3614,39 @@ spec: imagePullPolicy: Always args: - -c - - ./experiments -name pod-http-reset-peer + - ./experiments -name pod-cpu-hog command: - /bin/bash env: - - name: TARGET_CONTAINER - value: "" - - # provide lib image - - name: LIB_IMAGE - value: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" - - # reset timeout specifies after how much duration to reset the connection - - name: RESET_TIMEOUT - value: "0" #in ms + - name: TOTAL_CHAOS_DURATION + value: "60" - # port of the target service - - name: TARGET_SERVICE_PORT - value: "80" + ## Number of CPU cores to stress + - name: CPU_CORES + value: "1" - # toxicity is the probability of the request to be affected - # provide the percentage value in the range of 0-100 - # 0 means no request will be affected and 100 means all request will be affected - - name: TOXICITY + ## LOAD CPU WITH GIVEN PERCENT LOADING FOR THE CPU STRESS WORKERS. + ## 0 IS EFFECTIVELY A SLEEP (NO LOAD) AND 100 IS FULL LOADING + - name: CPU_LOAD value: "100" - # port on which the proxy will listen - - name: PROXY_PORT - value: "20000" - - # network interface on which the proxy will listen - - name: NETWORK_INTERFACE - value: "eth0" - - - name: TOTAL_CHAOS_DURATION - value: "60" # in seconds + ## Percentage of total pods to target + - name: PODS_AFFECTED_PERC + value: "" - # Time period to wait before and after injection of chaos in sec + ## Period to wait before and after injection of chaos in sec - name: RAMP_TIME value: "" - - name: DEFAULT_HEALTH_CHECK - value: "false" - - # percentage of total pods to target - - name: PODS_AFFECTED_PERC - value: "" + ## It is used in pumba lib only + - name: LIB_IMAGE + value: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" - - name: TARGET_PODS - value: "" + ## It is used in pumba lib only + - name: STRESS_IMAGE + value: "alexeiled/stress-ng:latest-ubuntu" - # provide the name of container runtime - # for litmus LIB, it supports docker, containerd, crio - # for pumba LIB, it supports docker only + ## provide the cluster runtime - name: CONTAINER_RUNTIME value: "containerd" @@ -3535,6 +3654,15 @@ spec: - name: SOCKET_PATH value: "/run/containerd/containerd.sock" + - name: TARGET_CONTAINER + value: "" + + - name: TARGET_PODS + value: "" + + - name: DEFAULT_HEALTH_CHECK + value: "false" + # To select pods on specific node(s) - name: NODE_LABEL value: "" @@ -3545,9 +3673,10 @@ spec: value: "parallel" labels: - name: pod-http-reset-peer + name: pod-cpu-hog app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job + app.kubernetes.io/runtime-api-usage: "true" app.kubernetes.io/version: 3.11.0 --- @@ -3555,12 +3684,12 @@ spec: apiVersion: litmuschaos.io/v1alpha1 description: message: | - It injects the chaos inside the pod which modifies the body of the response from the provided application server to the body string provided by the user and reverts after a specified duration + it injects chaos into the pod which stops outgoing HTTP requests by resetting the TCP connection and then reverts back to the original state after a specified duration kind: ChaosExperiment metadata: - name: pod-http-modify-body + name: pod-http-reset-peer labels: - name: pod-http-modify-body + name: pod-http-reset-peer app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 @@ -3625,7 +3754,7 @@ spec: imagePullPolicy: Always args: - -c - - ./experiments -name pod-http-modify-body + - ./experiments -name pod-http-reset-peer command: - /bin/bash env: @@ -3636,20 +3765,9 @@ spec: - name: LIB_IMAGE value: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" - # provide the body string to overwrite the response body - # if no value is provided, response will be an empty body. - - name: RESPONSE_BODY - value: "" - - # provide the encoding type for the response body - # currently supported value are gzip, deflate - # if empty no encoding will be applied - - name: CONTENT_ENCODING - value: "" - - # provide the content type for the response body - - name: CONTENT_TYPE - value: "text/plain" + # reset timeout specifies after how much duration to reset the connection + - name: RESET_TIMEOUT + value: "0" #in ms # port of the target service - name: TARGET_SERVICE_PORT @@ -3706,125 +3824,7 @@ spec: value: "parallel" labels: - name: pod-http-modify-body - app.kubernetes.io/part-of: litmus - app.kubernetes.io/component: experiment-job - app.kubernetes.io/version: 3.11.0 - ---- ---- -apiVersion: litmuschaos.io/v1alpha1 -description: - message: | - Injects CPU consumption on pods belonging to an app deployment -kind: ChaosExperiment -metadata: - name: pod-cpu-hog-exec - labels: - name: pod-cpu-hog-exec - app.kubernetes.io/part-of: litmus - app.kubernetes.io/component: chaosexperiment - app.kubernetes.io/version: 3.11.0 -spec: - definition: - scope: Namespaced - permissions: - # Create and monitor the experiment & helper pods - - apiGroups: [""] - resources: ["pods"] - verbs: - [ - "create", - "delete", - "get", - "list", - "patch", - "update", - "deletecollection", - ] - # Performs CRUD operations on the events inside chaosengine and chaosresult - - apiGroups: [""] - resources: ["events"] - verbs: ["create", "get", "list", "patch", "update"] - # Fetch configmaps details and mount it to the experiment pod (if specified) - - apiGroups: [""] - resources: ["configmaps"] - verbs: ["get", "list"] - # Track and get the runner, experiment, and helper pods log - - apiGroups: [""] - resources: ["pods/log"] - verbs: ["get", "list", "watch"] - # for creating and managing to execute commands inside target container - - apiGroups: [""] - resources: ["pods/exec"] - verbs: ["get", "list", "create"] - # deriving the parent/owner details of the pod(if parent is anyof {deployment, statefulset, daemonsets}) - - apiGroups: ["apps"] - resources: ["deployments", "statefulsets", "replicasets", "daemonsets"] - verbs: ["list", "get"] - # deriving the parent/owner details of the pod(if parent is deploymentConfig) - - apiGroups: ["apps.openshift.io"] - resources: ["deploymentconfigs"] - verbs: ["list", "get"] - # deriving the parent/owner details of the pod(if parent is deploymentConfig) - - apiGroups: [""] - resources: ["replicationcontrollers"] - verbs: ["get", "list"] - # deriving the parent/owner details of the pod(if parent is argo-rollouts) - - apiGroups: ["argoproj.io"] - resources: ["rollouts"] - verbs: ["list", "get"] - # for configuring and monitor the experiment job by the chaos-runner pod - - apiGroups: ["batch"] - resources: ["jobs"] - verbs: ["create", "list", "get", "delete", "deletecollection"] - # for creation, status polling and deletion of litmus chaos resources used within a chaos workflow - - apiGroups: ["litmuschaos.io"] - resources: ["chaosengines", "chaosexperiments", "chaosresults"] - verbs: ["create", "list", "get", "patch", "update", "delete"] - image: "litmuschaos.docker.scarf.sh/litmuschaos/go-runner:3.11.0" - imagePullPolicy: Always - args: - - -c - - ./experiments -name pod-cpu-hog-exec - command: - - /bin/bash - env: - - name: TOTAL_CHAOS_DURATION - value: "60" - - ## Number of CPU cores to stress - - name: CPU_CORES - value: "1" - - ## Percentage of total pods to target - - name: PODS_AFFECTED_PERC - value: "" - - ## Period to wait before and after injection of chaos in sec - - name: RAMP_TIME - value: "" - - # The command to kill the chaos process - - name: CHAOS_KILL_COMMAND - value: "kill $(find /proc -name exe -lname '*/md5sum' 2>&1 | grep -v 'Permission denied' | awk -F/ '{print $(NF-1)}')" - - - name: TARGET_CONTAINER - value: "" - - - name: TARGET_PODS - value: "" - - - name: DEFAULT_HEALTH_CHECK - value: "false" - - ## it defines the sequence of chaos execution for multiple target pods - ## supported values: serial, parallel - - name: SEQUENCE - value: "parallel" - - labels: - name: pod-cpu-hog-exec + name: pod-http-reset-peer app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job app.kubernetes.io/version: 3.11.0 diff --git a/faults/spring-boot/experiments.yaml b/faults/spring-boot/experiments.yaml index d0bf7c145..240fe6d3d 100644 --- a/faults/spring-boot/experiments.yaml +++ b/faults/spring-boot/experiments.yaml @@ -105,12 +105,12 @@ spec: apiVersion: litmuschaos.io/v1alpha1 description: message: | - It can target random pods with a Spring Boot application and allows configuring the assaults to inject cpu-stress + It can target random pods with a Spring Boot application and allows configuring the assaults to inject network latency kind: ChaosExperiment metadata: - name: spring-boot-faults + name: spring-boot-exceptions labels: - name: spring-boot-faults + name: spring-boot-exceptions app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 @@ -155,54 +155,15 @@ spec: imagePullPolicy: Always args: - -c - - ./experiments -name spring-boot-faults + - ./experiments -name spring-boot-exceptions command: - /bin/bash env: - # it enables spring app-kill fault - - name: CM_KILL_APPLICATION_ACTIVE - value: "" - - # it enables spring-boot latency fault - - name: CM_LATENCY_ACTIVE - value: "" - - # provide the latency (ms) - # it is applicable when latency is active - - name: LATENCY - value: "2000" - - # it enables spring-boot memory stress fault - - name: CM_MEMORY_ACTIVE - value: "" - - # it contains fraction of memory to be stressed(0.70 equals 70%) - # it supports value in range [0.01,0.95] - # it is applicable when memory is active - - name: MEMORY_FILL_FRACTION - value: "0.70" - - # it enables spring-boot cpu stress fault - - name: CM_CPU_ACTIVE - value: "" - - # it contains fraction of cpu to be stressed(0.95 equals 95%) - # it supports value in range [0.1,1.0] - # it is applicable when cpu is active - - name: CPU_LOAD_FRACTION - value: "0.9" - - # it enables spring-boot exceptions fault - - name: CM_EXCEPTIONS_ACTIVE - value: "" - # Type of raised exception - # it is applicable when exceptions is active - name: CM_EXCEPTIONS_TYPE value: "java.lang.IllegalArgumentException" # Argument of raised exception - # it is applicable when exceptions is active - name: CM_EXCEPTIONS_ARGUMENTS value: "java.lang.String:custom illegal argument exception" @@ -241,7 +202,7 @@ spec: value: "parallel" labels: - name: spring-boot-faults + name: spring-boot-exceptions app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job app.kubernetes.io/version: 3.11.0 @@ -355,12 +316,12 @@ spec: apiVersion: litmuschaos.io/v1alpha1 description: message: | - It can target random pods with a Spring Boot application and allows configuring the assaults to inject network latency + It can target random pods with a Spring Boot application and allows configuring the assaults to inject app-kill kind: ChaosExperiment metadata: - name: spring-boot-exceptions + name: spring-boot-app-kill labels: - name: spring-boot-exceptions + name: spring-boot-app-kill app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 @@ -405,18 +366,10 @@ spec: imagePullPolicy: Always args: - -c - - ./experiments -name spring-boot-exceptions + - ./experiments -name spring-boot-app-kill command: - /bin/bash env: - # Type of raised exception - - name: CM_EXCEPTIONS_TYPE - value: "java.lang.IllegalArgumentException" - - # Argument of raised exception - - name: CM_EXCEPTIONS_ARGUMENTS - value: "java.lang.String:custom illegal argument exception" - # port of the spring boot application - name: CM_PORT value: "" @@ -435,9 +388,6 @@ spec: - name: CM_WATCHERS value: "restController" - - name: TOTAL_CHAOS_DURATION - value: "30" - ## percentage of total pods to target - name: PODS_AFFECTED_PERC value: "" @@ -452,7 +402,7 @@ spec: value: "parallel" labels: - name: spring-boot-exceptions + name: spring-boot-app-kill app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job app.kubernetes.io/version: 3.11.0 @@ -462,12 +412,12 @@ spec: apiVersion: litmuschaos.io/v1alpha1 description: message: | - It can target random pods with a Spring Boot application and allows configuring the assaults to inject app-kill + It can target random pods with a Spring Boot application and allows configuring the assaults to inject cpu-stress kind: ChaosExperiment metadata: - name: spring-boot-app-kill + name: spring-boot-faults labels: - name: spring-boot-app-kill + name: spring-boot-faults app.kubernetes.io/part-of: litmus app.kubernetes.io/component: chaosexperiment app.kubernetes.io/version: 3.11.0 @@ -512,10 +462,57 @@ spec: imagePullPolicy: Always args: - -c - - ./experiments -name spring-boot-app-kill + - ./experiments -name spring-boot-faults command: - /bin/bash env: + # it enables spring app-kill fault + - name: CM_KILL_APPLICATION_ACTIVE + value: "" + + # it enables spring-boot latency fault + - name: CM_LATENCY_ACTIVE + value: "" + + # provide the latency (ms) + # it is applicable when latency is active + - name: LATENCY + value: "2000" + + # it enables spring-boot memory stress fault + - name: CM_MEMORY_ACTIVE + value: "" + + # it contains fraction of memory to be stressed(0.70 equals 70%) + # it supports value in range [0.01,0.95] + # it is applicable when memory is active + - name: MEMORY_FILL_FRACTION + value: "0.70" + + # it enables spring-boot cpu stress fault + - name: CM_CPU_ACTIVE + value: "" + + # it contains fraction of cpu to be stressed(0.95 equals 95%) + # it supports value in range [0.1,1.0] + # it is applicable when cpu is active + - name: CPU_LOAD_FRACTION + value: "0.9" + + # it enables spring-boot exceptions fault + - name: CM_EXCEPTIONS_ACTIVE + value: "" + + # Type of raised exception + # it is applicable when exceptions is active + - name: CM_EXCEPTIONS_TYPE + value: "java.lang.IllegalArgumentException" + + # Argument of raised exception + # it is applicable when exceptions is active + - name: CM_EXCEPTIONS_ARGUMENTS + value: "java.lang.String:custom illegal argument exception" + # port of the spring boot application - name: CM_PORT value: "" @@ -534,6 +531,9 @@ spec: - name: CM_WATCHERS value: "restController" + - name: TOTAL_CHAOS_DURATION + value: "30" + ## percentage of total pods to target - name: PODS_AFFECTED_PERC value: "" @@ -548,7 +548,7 @@ spec: value: "parallel" labels: - name: spring-boot-app-kill + name: spring-boot-faults app.kubernetes.io/part-of: litmus app.kubernetes.io/component: experiment-job app.kubernetes.io/version: 3.11.0