Skip to content
RakkrRakkr

Track recording jobs

Where the Recordings page shows finished results, the Jobs page shows the work that produces them. Use it when something is stuck or has failed.

Who can do this: viewing needs recording:read; stopping and retrying need recording:control (or the room’s operate capability).

See what’s happening

  1. Open Jobs in the left nav (titled “Recording Jobs”).
  2. The status tiles summarize active / queued / completed / failed jobs.
  3. Filter by status, capture backend, node, interface, and created date to focus on what matters.
  4. Each job row shows its capture settings, a claimed-by/lease badge, a created / started / completed timeline, and — if it failed — the failure reason.

Retry or stop a job

  • Stop an active job (needs recording:control). The partial is finalized as a completed recording.
  • Retry a failed or cancelled job to run it again.
  • Do either individually or in bulk, and export the filtered/selected jobs as CSV.

Understand the lifecycle

A job moves through these states:

  1. queued — created, waiting for a node.
  2. running — a node has leased it (its claimed-by badge shows which) and is capturing, heartbeating to the controller.
  3. stop_requested — a stop was requested while it was running; the node is wrapping up.
  4. completed / failed / cancelled — terminal.

claimed is not a status — it’s a lease phase, tracked via the job’s claimedBy field as a job starts running. A controller safety net automatically fails orphaned “running” jobs whose lease expired, so a crashed agent never leaves a recording stranded. The full sequence is in the Recording guide.

See also