Fabric → Workers

Workers are the distributed agents that execute tasks (health checks, snapshot collection, session proxying, automation actions) against your resources. Every MisterShell instance ships with a built-in default worker; larger deployments run additional workers close to the resources they manage — typically one per site, region, or network segment.

Workers are viewed and managed on the Fabric page’s Estate tab, alongside your Sensors and Proxies. Filter the Estate to Workers to focus on them.

How work reaches a worker

A task targets a resource, and MisterShell routes it by the resource’s location — walking up to parent locations if no worker is online there. The built-in worker sits at the root location, so the walk-up always has a catch-all to land on:

flowchart TB
  task["Task targets a resource"] --> loc["The resource's location"]
  loc --> walk["Find an online worker in that location,<br/>then walk up to parent locations if none there"]
  walk --> pick["Pick the most specific match<br/>ties broken by fewest active tasks"]
  pick -->|found| dispatch["Dispatch → worker runs the task → result returned"]
  pick -->|none online| waits["No worker available · task waits"]

What you can do

• See every registered worker and its live status.
• Register a new worker and obtain its deployment token.
• Edit a worker’s configuration.
• Regenerate a worker’s authentication token.
• Delete a worker.

Table columns

On the Estate tab, worker rows show:

Column	Notes
Name	Worker display name. Disabled workers carry a disabled chip.
Type	Worker.
Status	pending (orange — awaiting connection admission), online (green), offline (grey), or error (red).
Location	The location the worker is assigned to serve. The default worker sits at the root location.
Version	The worker agent version.
Activity	Live count of running tasks versus the concurrency limit (active / max). The limit is workspace-wide, not per worker — see Editing.
Extra	A compatibility badge — green RDK compatible, red RDK incompatible, grey until the worker first reports. A worker marked incompatible is running a version too far from the core’s and is not assigned any work until upgraded.
Last Heartbeat	Timestamp of the most recent check-in.
Actions	View details / Edit / Regenerate Token / Delete.

Open View details to see a worker’s capacity bar and its recent tasks. A Live badge above the table indicates that status and activity counters refresh as workers check in.

Licensing

Creating, editing, or regenerating tokens for remote workers requires the Pro edition. Without it these actions stay visible but disabled, with a padlock naming the edition you need; the default worker is system-managed and keeps working regardless. Existing remote workers keep checking in and running tasks, and deleting them remains available so you can retire them.

Common tasks

Register a new worker

1. On the Fabric Estate tab, click Create and choose Worker.
2. Fill the form:
   • Name — a descriptive label (for example, Worker — EMEA Site 1).
   • Description — optional notes.
   • Location — pick the location this worker serves. Tasks targeting resources in this location will be routed to this worker first.
   • Enabled — leave on unless you want to pre-register a worker in a disabled state.
   • Manual Worker Presence IP — optional; set the address other components should use to reach this worker for throughput probes, when it can’t be discovered automatically.
   • Collector — an optional per-worker syslog collector; see Collectors.
3. Click Create.
4. A dialog shows the deployment token. Copy it and paste it into the worker’s configuration when you install the worker on the target machine. The full token is only shown once.

Regenerate a worker’s token

Use this when a token has been exposed or on a rotation schedule.

1. Click the Regenerate Token action on the row.
2. Confirm in the dialog.
3. A dialog shows the new token. Update the worker’s configuration on the host with the new value — the old token stops working immediately.

Edit a worker

1. Click the Edit action on the row.
2. Update name, description, location, enabled state, presence IP, or the collector settings. The Enabled toggle lives here — a disabled worker receives no tasks.
3. Click Update.

Task concurrency is not set per worker: the limit is the workspace-wide worker_max_concurrent_tasks setting, adjusted on Advanced Settings or the Config wizard’s Performance Tuning step.

The default worker can be edited for configuration only, and has neither a regenerable token nor a delete action.

Delete a worker

1. Click the Delete action on the row.
2. Confirm.

The worker’s entry is removed; if the worker is still running on its host, it can no longer authenticate. Stop and uninstall the worker process on the host to complete decommissioning.

Permissions

• Read: app.fabric.read.
• Create / edit / regenerate token: app.fabric.write.
• Delete: app.fabric.delete.