Unraid icon

Unraid

Interact with the Unraid API — manage Docker containers, VMs, array health, disks, notifications, and system metrics

Actions36

VM → Restart

AI-generated

Summary

Restarts an Unraid virtual machine by sending a graceful stop, waiting 3 seconds, then sending a start for the given VM ID.

Inputs

  • VM ID (required) — Prefixed VM ID (e.g. vm:abc123). Use the VM Get Many operation to find IDs.
  • Maximum Control Level — Optional cap on what this node may do. It can only narrow the credential's limit, never raise it. Restart is a 'control' operation, so the effective level must be Control or Full. At Read, the node throws an error before contacting the server.

Output shape

One item per input item: {id, operation: "restart", success: true}.

The stop and start calls are two separate GraphQL mutations. The node doesn't check the VM's state between them, and a fixed 3-second delay separates them. A graceful stop can take longer than that, in which case the start may run before the VM has fully stopped. The node doesn't check this, and success: true only means both API calls returned without an error. If a call fails, the node throws, or returns {error} when Continue On Fail is enabled.

Examples

Example 1: Restart a VM nightly on a schedule.

Resource: VM, Operation: Restart, VM ID: vm:abc123. The credential's Maximum Control Level is Control or Full.

Example 2: Restart a VM that an AI agent has flagged as unresponsive, without letting the agent force-stop it.

Resource: VM, Operation: Restart, VM ID set from an expression. Maximum Control Level: Control, which blocks Force Stop.

Links

Discussion