Skip to content

Creating a Kubernetes connection

Learn how to create a Kubernetes connection for accessing Kubernetes cluster logs.

  • Access to a Kubernetes cluster
  • global_create_connection permission (Global Admin role)

Navigate to ConnectionsNew.

Select Kubernetes as the connection kind and provide your kubeconfig:

  • Option 1: Paste the kubeconfig file content (as YAML) directly in the configuration field
  • Option 2: Enable “Use local file path instead of content” and provide the path to your kubeconfig file (e.g., ~/.kube/config)

Optional settings:

  • Context FlyQL filter – Filter which contexts from your kubeconfig should be available
    • Use FlyQL expressions to filter contexts by name (e.g., name ~ "prod.*" to include only production contexts)
    • Leave empty to include all contexts from the kubeconfig
  • Max concurrent requests – Control parallelism when fetching logs from multiple pods (default: 20)
    • This limit applies per-context
    • Higher values provide faster results but increase cluster API load
  • Name – Connection name (e.g., “Production K8s”, “Staging Kubernetes”)
  • Description – Optional description explaining which clusters this connection accesses

Review your configuration and click “Create” to save the connection.

  1. Filter contexts with FlyQL when file contains multiple environments
  2. Tune concurrency based on cluster size and API capacity
  3. Use descriptive names indicating which cluster/environment
  • Use separate connections for different environments (prod/staging/dev)
  • Ensure kubeconfig has minimal required permissions (read-only access to pods/logs)

Connection test fails:

  • Verify kubeconfig is valid YAML
  • Ensure contexts have necessary credentials
  • Check cluster API is accessible from Telescope
  • Verify RBAC permissions allow listing pods and reading logs

No contexts available:

  • Check Context FlyQL filter isn’t too restrictive
  • Verify kubeconfig contains valid contexts
  • Ensure contexts have active clusters configured