Skip to main content

Manage clusters and claws

Once a clawster exists, the day-to-day work is mostly operational clarity.

Clusters

Clusters group related claws and their provider relationships.

Use a cluster to:

  • keep one project or team isolated
  • understand which provider lane is primary
  • track what is active, inactive, or misconfigured

Claws

A claw is one running unit in the clawster. Depending on your model, that may be a VPS, cloud runtime, or Kubernetes-backed path.

Typical actions:

  • review status
  • inspect activation context
  • confirm region and service selection
  • reprovision or retire when needed

Single vs hybrid

Use single-provider clawsters when you want simplicity. Use hybrid clawsters when different claw roles need different provider paths.

Hybrid clawster

Signals that matter

  • provider binding exists and is active
  • integration health is green
  • selected characteristics still map to valid provider-native parameters
  • the claw is attached to the expected region and lane