Read health, activity, and traffic

Understand current runtime health, follow recovery, and use observations to investigate real customer journeys.

Read current bot health

Deploy and Overview show the runtime, heartbeat, platform health, and recovery. Running describes the process. Check the connection and refresh a stale report before concluding there is a new failure.

States such as verifying, restarting, repairing, connection issue, and check overdue describe different situations. If automatic recovery is working, follow its progress. CookMyBots verifies renewed health before marking recovery complete. A disconnected account may need reauthorization rather than a code edit.

Use Activity to understand events

Activity shows publishing results, connection changes, operational warnings, and recovery attempts. Open the event’s suggested action to reach the relevant report, configuration, billing, or BOTS SI request. Historical events remain useful context, but a previous success does not override a current warning.

  1. Note when the problem began and which platform or feature is affected.
  2. Open the latest relevant Activity event and current Deploy status.
  3. Compare the published version with the most recent edit.
  4. Use Retry recovery when offered for a runtime problem, or ask BOTS SI to check saved diagnostics.
  5. After the indicated action finishes, repeat the original user journey.

Interpret observations carefully

Analytics and Bot traffic show the observations available for the project. Use their time range and platform filters where offered to look for an incoming interaction and its result. An incoming count is not a count of unique customers, and the absence of a recorded event is not proof that every upstream service was reachable.

Inspect each Hybrid platform independently. A service response provides evidence for that request; the customer journey may still need a saved record, button action, or platform reply. Test the complete journey.