2 Poll Source Status

Step 2: Poll Source Status

Re-read the discovery task until it reaches a terminal state. While the grounded sweep is still running the task sits in pending or processing; when it finishes it flips to completed (results ready) or failed (with an error message).

What this step does

It reports how far the background discovery has progressed. The handler folds the upstream task and its sub-tasks into one normalised status, so you poll a single field: pending โ†’ processing โ†’ completed | failed. On failure, attributes.error carries the reason.

Why it's needed

The results endpoint has nothing to return until the sweep has finished โ€” fetching it mid-flight returns a 409 (still running). This step is the gate that confirms the candidates have actually been discovered before Step 3 reads them. A typical run takes anywhere from tens of seconds to a few minutes depending on scope, so poll every few seconds until the status is terminal.

When you can skip it

You always need at least one read here, because discovery is asynchronous by design. In a scripted client, loop this call on an interval until attributes.status is completed (then go to Step 3) or failed (then inspect attributes.error).

Visualization

Renders a status panel colour-coded by lifecycle state โ€” amber while processing, green on completed, red on failed (with the error) โ€” so the poll loop is readable without inspecting the JSON.

Next Step

โ†’ Step 3: Get Discovery Results (GET /discovery/source/:id/results)