Track runs

Each run is one execution of a workflow at a point in time. History lets you audit, retry, and understand failures.

🔁 Where to find runs

1. Knot → Workflows → pick a workflow.
2. Runs or History tab.
3. Filter by status, date, or trigger source.

Typical columns: run ID, status, duration, trigger, user.

🔁 Run statuses

Status	Meaning
Running	In progress
Success	Finished without blocking error
Failed	Node error or timeout
Cancelled	Manual stop or policy

Open details to see the failing node and message (HTTP, Dolibarr, SMTP, AI).

🔁 Logs and context

Run detail keeps:

• per-node context snapshot;
• masked HTTP requests (secrets not shown);
• timestamps to correlate with Knot audit log.

This user doc focuses on the Runs UI, not raw llx_knot_audit_log schema.

🔁 Replay and retry

• Retry: rerun a failed run or resume after the error node (version-dependent).
• Manual replay: same input payload after a config fix.

Before bulk retry, check idempotency: a second HTTP POST or email may duplicate effects.

🔁 Effective debugging

1. Reproduce in Simulate with the same payload.
2. Compare failed vs successful run context.
3. Check permissions and entity.
4. See Troubleshooting overview when detailed pages ship.

Next steps

• Test with Simulate
• Credentials