Error Handling and Monitoring:
Implement mechanisms to promptly identify and handle timeout errors, such as automatic retries or escalation workflows for queued jobs.
Retain a copy of the input JSON for at least 5–10 days to support post-processing troubleshooting. Having the JSON available simplifies troubleshooting.
Logging and Reporting:
Each request generates a Generation Log which captures all of the details of the generation, including the time it took to generate the document. Capture and store these generation logs for each request, including the unique reference number and duration metrics. This data will help monitor system performance and diagnose anomalies.
Maintain these logs for a defined period, with optional scheduled cleanup to manage storage overhead.