Public | Automated Build

Last pushed: 16 days ago
Short Description
chaoskube periodically kills random pods in your Kubernetes cluster.
Full Description


chaoskube periodically kills random pods in your Kubernetes cluster.


Test how your system behaves under arbitrary pod failures.


Running it will kill a pod in any namespace every 10 minutes by default.

$ chaoskube
INFO[0000] Targeting cluster at
INFO[0001] Killing pod kube-system/kube-dns-v20-6ikos
INFO[0601] Killing pod chaoskube/nginx-701339712-u4fr3
INFO[1201] Killing pod kube-system/kube-proxy-gke-earthcoin-pool-3-5ee87f80-n72s
INFO[1802] Killing pod chaoskube/nginx-701339712-bfh2y
INFO[2402] Killing pod kube-system/heapster-v1.2.0-1107848163-bhtcw
INFO[3003] Killing pod kube-system/l7-default-backend-v1.0-o2hc9
INFO[3603] Killing pod kube-system/heapster-v1.2.0-1107848163-jlfcd
INFO[4203] Killing pod chaoskube/nginx-701339712-bfh2y
INFO[4804] Killing pod chaoskube/nginx-701339712-51nt8

chaoskube allows to filter target pods by namespaces, labels and annotations.
See below for details.


Get chaoskube via go get, make sure your current context points to your target cluster and use the --deploy flag.

$ go get -u
$ chaoskube --deploy
INFO[0000] Dry run enabled. I won't kill anything. Use --no-dry-run when you're ready.
INFO[0000] Targeting cluster at
INFO[0000] Deployed

By default chaoskube will be friendly and not kill anything. When you validated your target cluster you may disable dry-run mode. You can also specify a more aggressive interval and other supported flags for your deployment.

$ chaoskube --interval=1m --no-dry-run --debug --deploy
DEBU[0000] Using current context from kubeconfig at /Users/you/.kube/config.
INFO[0000] Targeting cluster at
DEBU[0000] Deploying
INFO[0000] Deployed

You can also install chaoskube with Helm. Follow Helm's Quickstart Guide and then install the chaoskube chart.

$ helm install stable/chaoskube --version 0.5.0 --set interval=1m,dryRun=false

Refer to chaoskube on to learn how to configure it and to find other useful Helm charts.

Otherwise use the following equivalent manifest file or let it serve as an inspiration.

apiVersion: extensions/v1beta1
kind: Deployment
  name: chaoskube
  replicas: 1
        app: chaoskube
      - name: chaoskube
        - --interval=1m
        - --no-dry-run

If you're running in a Kubernetes cluster and want to target the same cluster then this is all you need to do.

If you want to target a different cluster or want to run it locally specify your cluster via the --master flag or provide a valid kubeconfig via the --kubeconfig flag. By default, it uses your standard kubeconfig path in your home. That means, whatever is the current context in there will be targeted.

If you want to increase or decrease the amount of chaos change the interval between killings with the --interval flag. Alternatively, you can increase the number of replicas of your chaoskube deployment.

Remember that chaoskube by default kills any pod in all your namespaces, including system pods and itself.

Filtering targets

However, you can limit the search space of chaoskube by providing label, annotation and namespace selectors.

$ chaoskube --labels 'app=mate,chaos,stage!=production'
INFO[0000] Filtering pods by labels: app=mate,chaos,stage!=production

This selects all pods that have the label app set to mate, the label chaos set to anything and the label stage not set to production or unset.

You can filter target pods by namespace selector as well.

$ chaoskube --namespaces 'default,testing,staging'
INFO[0000] Filtering pods by namespaces: default,staging,testing

This will filter for pods in the three namespaces default, staging and testing.

You can also exclude namespaces and mix and match with the label and annotation selectors.

$ chaoskube \
    --labels 'app=mate,chaos,stage!=production' \
    --annotations '!' \
    --namespaces '!kube-system,!production'
INFO[0000] Filtering pods by labels: app=mate,chaos,stage!=production
INFO[0000] Filtering pods by annotations: !
INFO[0000] Filtering pods by namespaces: !kube-system,!production

This further limits the search space of the above label selector by also excluding any pods in the kube-system and production namespaces as well as ignore all pods that are marked as critical.

The annotation selector can also be used to run chaoskube as a cluster addon and allow pods to opt-in to being terminated as you see fit. For example, you could run chaoskube like this:

$ chaoskube --annotations ''
INFO[0000] Filtering pods by annotations:
INFO[0000] No victim could be found. If that's surprising double-check your selectors.

Unless you already use that annotation somewhere, this will initially ignore all of your pods. You could then selectively opt-in individual deployments to chaos mode by annotating their pods with

apiVersion: extensions/v1beta1
kind: Deployment
  name: my-app
  replicas: 3
      annotations: "true"


Feel free to create issues or submit pull requests.

Docker Pull Command
Source Repository

Comments (0)