Reprovision

Reprovisioning destroys the current VPS and creates a new one with the latest configuration from the Tulsk database.

When to Reprovision

You should reprovision your cluster in the following scenarios:

Persona Edits: After changing SOUL.md, IDENTITY.md, or other persona files.
Agent Membership: After adding or removing agents from the cluster.
System Recovery: If the cluster enters an ERROR state.
System Updates: When Tulsk releases bridge or runtime updates.

The Reprovisioning Process

Database Collection

Tulsk gathers the latest configurations for all agents in the cluster.

Server Destruction

The current VPS instance is securely deleted.

Fresh Provisioning

A new VPS is created with the updated AGENT_SLOT_MAP and per-agent workspace files.

Synchronization

The new instance is linked back to the workspace and heartbeats begin.

Ready

Status transitions: RUNNING → DESTROYING → PROVISIONING → RUNNING.

Downtime Warning

⚠️

During reprovisioning, all agents on the cluster are offline for 3-5 minutes. Agent runs triggered during this window will fail until the new instance is healthy.

Safety & Fallbacks

Error Handling: If provisioning fails, a fallback instance is created in ERROR state so agents remain grouped and manageable.
Clean Slate: The old server is always deleted before the new one is created to prevent conflicts.

How to Reprovision

Cluster Settings

Go to the Cluster Settings page via the gear icon on the Agents dashboard.

Execute

Click the “Reprovision” button.

Monitor

You will be redirected back to the Agents page where you can monitor the PROVISIONING status in real-time.