| name | trace-message-flow |
| description | Trace a message's flow through an Interoperability production using iris_production_messages, iris_message_diagram, and iris_production_logs for any erroring items. |
Trace Message Flow
Session or header ID to trace: <sessionOrHeaderId>
No namespace specified — omit namespace to use the server's configured default on every call below.
- Call
iris_production_messages with sessionId: <sessionOrHeaderId> (traces every message in that session). If the result is empty or the ID does not look like a session ID, retry with headerId: <sessionOrHeaderId> instead — the ID may be a specific message header ID rather than a session ID.
- Review the returned message steps: source item, target item, message class, timestamp, and status for each hop.
- Call
iris_message_diagram with sessionIds: [<sessionOrHeaderId>] (use the session ID discovered in step 1, even if you started from a header ID) to render a Mermaid sequence diagram of the same flow — participants, sync vs async calls, request/response pairing, and [ERROR] flags on failed messages.
- For every step flagged as an error (in either the raw trace or the diagram's
[ERROR] markers), call iris_production_logs filtered by itemName (the erroring config item) and/or type: "Error" to pull the detailed error log entries for that item around the same timeframe.
- Summarize the failure point: which item failed, what the error log says, and where in the flow (which hop) it occurred. Recommend next steps (e.g.
iris_production_control action 'recover', or a data/config fix) but do NOT execute a recovery action without the user's confirmation.
