Skip to main content

Basic install

Last updated on

Use this install when you can grant the Delegate the default cluster-admin role. The chaos runner inherits that role and can target any namespace in the cluster.

Install the Harness Delegate​

Follow the platform install guide at Install Delegate. The recommended path is the Helm chart with the standard Delegate image (it ships kubectl and go-template, both of which chaos needs).

Using the minimal Delegate image

The minimal Delegate image does not include kubectl or go-template, so chaos and discovery do not work on it out of the box. Go to Using the minimal Delegate image for the two install options (INIT_SCRIPT or custom image).

Create a Kubernetes infrastructure​

  1. Go to Resilience Testing → Project Settings → Resilience Testing Infrastructures.

  2. Select the Kubernetes (Harness Infrastructure) tab.

  3. Click + New Infrastructure. The Create a Chaos Infrastructure with a Harness Delegate dialog opens.

  4. Pick the environment the infrastructure belongs to. The environment picker lists project, organization, and account-scoped environments. Click + New Environment if you need one. Click Continue.

  5. In the Create New Infrastructure form:

    • Name, Description, Tags. Standard Harness metadata.
    • Storage: choose Inline (definition stored in Harness) or Remote (definition stored in a Git repository).
    • Deployment Type: Kubernetes.
    • Infrastructure Type: Direct Connection (Kubernetes) or Via Cloud Provider.
    • Cluster Details:
      • Connector: the Kubernetes connector that reaches the target cluster.
      • Namespace: where the chaos runner and fault pods are created.
      • Release name (Advanced): default release-<+INFRA_KEY_SHORT_ID>. Leave unchanged unless you have a naming convention.
    • Map Dynamically Provisioned Infrastructure: enable when the target cluster is provisioned through an upstream pipeline.
    • Scope to Specific Services: enable to limit chaos to a subset of services within the namespace.
    • Allow simultaneous deployments on the same infrastructure: leave off unless you knowingly need parallel runs.

  6. Click Save. The new infrastructure appears in the list with status Inactive until you enable chaos on it.

YAML mode

The Visual form has a YAML toggle. Switching to YAML shows the infrastructureDefinition.yaml produced by the form, which you can edit directly or commit to Git.

Terraform alternative

To create a Kubernetes infrastructure through Terraform instead, use the harness_chaos_infrastructure_v2 resource in the Harness Terraform provider.

Create chaos experiments on the new infrastructure​

After you save the infrastructure, the Create Chaos Experiments on your Infrastructure wizard opens with Step 1 (Select Infrastructure) already complete.

  1. In Step 2, pick the creation mode:
    • Beginner: Harness creates a small set of Application Maps and recommended chaos experiments for you. You can edit them later.
    • Expert: multi-step guided creation. Pick fault types per workload.
  2. Click Go! to continue with the defaults. To override defaults for the chaos runner and Discovery Agent, click Configure Advanced Settings (Optional) first.

The infrastructure status flips to Active when the Delegate registers the chaos runner and the discovery agent finishes its first sweep.

Last updated on