Krkn injects deliberate failures into Kubernetes clusters so teams can assess how applications and clusters respond to disruption. Its more than 20 scenarios cover pods, containers, nodes, networks, applications, resources, storage, and time. The recommended krknctl tool discovers, runs, and orchestrates scenarios; other options include pre-built containers, a browser dashboard, and a standalone Python program. Krkn can evaluate Prometheus metrics against SLO pass/fail criteria and review metrics during and after chaos runs. The dashboard supports Elasticsearch and Grafana, and telemetry can be sent to S3 or Elasticsearch. Cerberus monitors Kubernetes and OpenShift health and provides a go/no-go signal, reports, and a metrics API; it can also report failures in Slack. Krkn-AI uses evolutionary algorithms with user-defined objectives to identify impactful experiments. The project says it saves recovery state and can restore a cluster automatically, but warns some disruptive scenarios may leave a target unrecoverable. Requirements include Kubernetes or OpenShift 1.21+, cluster-admin kubeconfig access, and Docker 20.10+ or Podman 4.0+. Krkn is for resilience testing, not security provision, and does not currently adhere to compliance standards.
Who it is for
Krkn suits first-time chaos engineers as well as teams building resilience-testing workflows for Kubernetes or OpenShift clusters. It is intended to measure application or cluster resilience, not provide security assurance.
What is good
- More than 20 chaos scenarios
- Validates SLO criteria with Prometheus metrics
- Offers CLI, containers, dashboard, and Python options
- Can monitor cluster health through Cerberus
What to know first
- Requires cluster-admin kubeconfig access
- Requires Kubernetes or OpenShift 1.21+
- Some scenarios may leave a system unrecoverable
- Does not currently adhere to compliance standards
Verdict
Krkn offers several ways to run a wide range of cluster disruption scenarios and evaluate their impact against SLOs. Check its access and runtime requirements, and account for the possibility that some disruptions cannot be automatically recovered.
Krkn plans and pricing
All plansCompared on chaos engineering platforms
- Free plan
- Yes
- Kubernetes support
- Yes
- Cloud fault injection
- Yes
- Network fault injection
- Yes
- Deployment model
- self_hosted
- Blast-radius controls
- Yes
