CheckOps home

Tools directory

C

CheckOps

Chaos Mesh

Chaos Mesh is a CNCF incubating, Apache-2.0 chaos engineering platform for Kubernetes that defines pod, network, DNS, HTTP, I/O, time, stress, kernel, JVM and cloud faults as custom resources.

Visit website

Latest public signals

Not refreshed yet. Refresh reads the public site and, when the entry has a GitHub repository, the latest release.

Visit Website

Introduction

Chaos Mesh is an open-source, cloud-native chaos engineering platform for Kubernetes. It uses Kubernetes custom resources to define, orchestrate and observe controlled fault injection against workloads, infrastructure, cloud services and applications. It is a CNCF incubating project, Apache-2.0, with about 8,000 GitHub stars.

The README lists broad fault coverage: pod, network, DNS, HTTP, I/O, time, stress, kernel, block device, JVM, physical machine, AWS, Azure and GCP faults.

Because experiments are custom resources, they are managed through the Kubernetes API with standard tooling and RBAC, and can be versioned in Git like any other manifest. A dashboard and workflow features help schedule and combine experiments.

In the quality loop, chaos experiments verify resilience before and after release: does the service degrade gracefully when a dependency is slow, a pod dies or the network partitions?

Features

  1. Wide fault library: Pod, network, DNS, HTTP, I/O, time, stress, kernel and JVM faults.
  2. Cloud faults: AWS, Azure and GCP fault types.
  3. Kubernetes-native: Experiments as custom resources with RBAC.
  4. Workflows: Orchestrate multiple experiments in sequence or parallel.
  5. Scheduling: Run experiments on a schedule.
  6. Dashboard: Web UI to create and observe experiments.
  7. Physical machine faults: Fault types for physical machines, per the README.
  8. GitOps friendly: Version experiments with manifests.

Pros

  1. Broad coverage: Many fault types.
  2. CNCF project: Community governance.
  3. Free: Apache-2.0.
  4. Native to Kubernetes: Fits existing tooling.
  5. Active: Commits in November 2026.

Cons

  1. Kubernetes focus: Best suited to Kubernetes environments.
  2. Privileged components: Some faults need elevated permissions.
  3. Blast radius risk: Experiments need careful scoping.
  4. Observability required: You need metrics to judge outcomes.
  5. Learning curve: Many CRD types.

Install with Helm (see docs for the current chart and runtime settings), then apply a simple experiment:

apiVersion: chaos-mesh.org/v1alpha1
kind: PodChaos
metadata:
  name: pod-kill-example
spec:
  action: pod-kill
  mode: one
  selector:
    namespaces: [staging]
    labelSelectors:
      app: checkout

Tips: start in staging with a clear hypothesis and abort criteria, watch dashboards during the run, and graduate to production only with tight selectors.

Resources:

Pricing

Chaos Mesh is free and open source under Apache-2.0.

  1. Kubernetes platform teams practising chaos engineering.
  2. SREs validating failover and graceful degradation.
  3. Teams wanting GitOps-managed experiments.
  4. Not a fit for teams without Kubernetes or observability.

Conclusion

Chaos Mesh is a broad, Kubernetes-native chaos platform with CNCF backing.

Start small with a hypothesis, observe carefully, and expand gradually.

Alternatives

Other tools on CheckOps that cover similar ground. Open a page, or compare one with Chaos Mesh side by side.