next

Edit

Node Memory Hog Experiment Details

Experiment Metadata

Type Description Tested K8s Platform
Generic Exhaust Memory resources on the Kubernetes Node GKE, EKS, AKS

Prerequisites

  • Ensure that Kubernetes Version > 1.16
  • Ensure that the Litmus Chaos Operator is running by executing kubectl get pods in operator namespace (typically, litmus). If not, install from here
  • Ensure that the node-memory-hog experiment resource is available in the cluster by executing kubectl get chaosexperiments in the desired namespace. If not, install from here
  • There should be administrative access to the platform on which the Kubernetes cluster is hosted, as the recovery of the affected node could be manual. For example, gcloud access to the GKE project

Entry Criteria

  • Application pods are healthy on the respective Nodes before chaos injection

Exit Criteria

  • Application pods may or may not be healthy post chaos injection

Details

  • This experiment causes Memory resource exhaustion on the Kubernetes node. The experiment aims to verify resiliency of applications whose replicas may be evicted on account on nodes turning unschedulable (Not Ready) due to lack of Memory resources.
  • The Memory chaos is injected using a job running the linux stress-ng tool (a workload generator). The chaos is effected for a period equalling the TOTAL_CHAOS_DURATION and upto MEMORY_CONSUMPTION_PERCENTAGE(out of 100) or MEMORY_CONSUMPTION_MEBIBYTES(in Mebibytes out of total available memory).
  • Application implies services. Can be reframed as: Tests application resiliency upon replica evictions caused due to lack of Memory resources

Integrations

  • Node Memory Hog can be effected using the chaos library: litmus
  • The desired chaos library can be selected by setting litmus as value for the env variable LIB

Steps to Execute the Chaos Experiment

  • This Chaos Experiment can be triggered by creating a ChaosEngine resource on the cluster. To understand the values to provide in a ChaosEngine specification, refer Getting Started

  • Follow the steps in the sections below to create the chaosServiceAccount, prepare the ChaosEngine & execute the experiment.

Prepare chaosServiceAccount

  • Use this sample RBAC manifest to create a chaosServiceAccount in the desired (app) namespace. This example consists of the minimum necessary role permissions to execute the experiment.

Sample Rbac Manifest

---
apiVersion: v1
kind: ServiceAccount
metadata:
  name: node-memory-hog-sa
  namespace: default
  labels:
    name: node-memory-hog-sa
    app.kubernetes.io/part-of: litmus
---
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRole
metadata:
  name: node-memory-hog-sa
  labels:
    name: node-memory-hog-sa
    app.kubernetes.io/part-of: litmus
rules:
- apiGroups: [""]
  resources: ["pods","events"]
  verbs: ["create","list","get","patch","update","delete","deletecollection"]
- apiGroups: [""]
  resources: ["pods/exec","pods/log"]
  verbs: ["create","list","get"]
- apiGroups: ["batch"]
  resources: ["jobs"]
  verbs: ["create","list","get","delete","deletecollection"]
- apiGroups: ["litmuschaos.io"]
  resources: ["chaosengines","chaosexperiments","chaosresults"]
  verbs: ["create","list","get","patch","update"]
- apiGroups: [""]
  resources: ["nodes"]
  verbs: ["get","list"]
---
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
metadata:
  name: node-memory-hog-sa
  labels:
    name: node-memory-hog-sa
    app.kubernetes.io/part-of: litmus
roleRef:
  apiGroup: rbac.authorization.k8s.io
  kind: ClusterRole
  name: node-memory-hog-sa
subjects:
- kind: ServiceAccount
  name: node-memory-hog-sa
  namespace: default

Note: In case of restricted systems/setup, create a PodSecurityPolicy(psp) with the required permissions. The chaosServiceAccount can subscribe to work around the respective limitations. An example of a standard psp that can be used for litmus chaos experiments can be found here.

Prepare ChaosEngine

  • Provide the application info in spec.appinfo. It is an optional parameter for infra level experiment.
  • Provide the auxiliary applications info (ns & labels) in spec.auxiliaryAppInfo
  • Override the experiment tunables if desired in experiments.spec.components.env
  • To understand the values to provided in a ChaosEngine specification, refer ChaosEngine Concepts

Supported Experiment Tunables

Variables Description Specify In ChaosEngine Notes
TARGET_NODES Comma separated list of nodes, subjected to node memory hog Mandatory
NODE_LABEL It contains node label, which will be used to filter the target nodes if TARGET_NODES ENV is not set Optional
TOTAL_CHAOS_DURATION The time duration for chaos insertion (in seconds) Optional Defaults to 120
LIB The chaos lib used to inject the chaos Optional Defaults to `litmus`
LIB_IMAGE Image used to run the stress command Optional Defaults to litmuschaos/go-runner:latest
MEMORY_CONSUMPTION_PERCENTAGE Percent of the total node memory capacity Optional Defaults to 30
MEMORY_CONSUMPTION_MEBIBYTES The size in Mebibytes of total available memory. When using this we need to keep MEMORY_CONSUMPTION_PERCENTAGE empty as the percentage have more precedence Optional
NUMBER_OF_WORKERS It is the number of VM workers involved in IO disk stress Optional Default to 1
RAMP_TIME Period to wait before and after injection of chaos in sec Optional
NODES_AFFECTED_PERC The Percentage of total nodes to target Optional Defaults to 0 (corresponds to 1 node), provide numeric value only
SEQUENCE It defines sequence of chaos execution for multiple target nodes Optional Default value: parallel. Supported: serial, parallel
INSTANCE_ID A user-defined string that holds metadata/info about current run/instance of chaos. Ex: 04-05-2020-9-00. This string is appended as suffix in the chaosresult CR name. Optional Ensure that the overall length of the chaosresult CR is still < 64 characters

Sample ChaosEngine Manifest

apiVersion: litmuschaos.io/v1alpha1
kind: ChaosEngine
metadata:
  name: nginx-chaos
  namespace: default
spec:
  # It can be active/stop
  engineState: 'active'
  #ex. values: ns1:name=percona,ns2:run=nginx 
  auxiliaryAppInfo: ''
  chaosServiceAccount: node-memory-hog-sa
  experiments:
    - name: node-memory-hog
      spec:
        components:
          env:
            # set chaos duration (in sec) as desired
            - name: TOTAL_CHAOS_DURATION
              value: '120'

            ## Specify the size as percent of total node capacity Ex: '30'
            ## Note: For consuming memory in mebibytes change the variable to MEMORY_CONSUMPTION_MEBIBYTES
            - name: MEMORY_CONSUMPTION_PERCENTAGE
              value: '30'
            
            ## percentage of total nodes to target
            - name: NODES_AFFECTED_PERC
              value: ''

            # provide the comma separated target node names
            - name: TARGET_NODES
              value: ''

Create the ChaosEngine Resource

  • Create the ChaosEngine manifest prepared in the previous step to trigger the Chaos.

    kubectl apply -f chaosengine.yml

  • If the chaos experiment is not executed, refer to the troubleshooting section to identify the root cause and fix the issues.

Watch Chaos progress

  • Setting up a watch of the Memory consumed by nodes in the Kubernetes Cluster

    watch kubectl top nodes

OR

  • Setting up a monitoring pod on the target node.

myhtop.yml

---
apiVersion: v1
kind: Pod
metadata:
  name: myhtop
spec:
  containers:
  - name: myhtop
    image: litmuschaos/go-runner:latest
    imagePullPolicy: Always
    command: ['sh', '-c', 'sleep 3600']
  nodeName: kube-01 ## Replace this with the target node name...

Create the htop monitoring pod on Target node:

kubectl apply -f myhtop.yml

Now Exec into the pod and run htop command

kubectl exec -it myhtop -- /bin/sh
/litmus $ htop

The output will be similar to:

Screenshot from 2021-01-22 16-01-30

It is showing:

  • The node size is 3.79G out of which 517M is in use. Now run the experiment and keep checking the available memory.

Abort/Restart the Chaos Experiment

  • To stop the pod-delete experiment immediately, either delete the ChaosEngine resource or execute the following command:

    kubectl patch chaosengine <chaosengine-name> -n <namespace> --type merge --patch '{"spec":{"engineState":"stop"}}'

  • To restart the experiment, either re-apply the ChaosEngine YAML or execute the following command:

    kubectl patch chaosengine <chaosengine-name> -n <namespace> --type merge --patch '{"spec":{"engineState":"active"}}'

Check Chaos Experiment Result

  • Check whether the application is resilient to the memory hog, once the experiment (job) is completed. The ChaosResult resource name is derived like this: <ChaosEngine-Name>-<ChaosExperiment-Name>.

    kubectl describe chaosresult nginx-chaos-node-memory-hog -n <application-namespace>

Node Memory Hog Experiment Demo

  • A sample recording of this experiment execution is provided here.
← Node CPU HogNode Drain →