Quick guides
Quick guide: Start an operational investigation
Use a short checklist to follow a problem from the visible symptom to operations, logs, and audit records.
Before you use a printed copy. Check the documentation version and verification date at the top of the page. Use the online guide if this copy is older.
When to use this guide
Use it when work is waiting, failed, missing, or giving an unexpected result.
Checklist
- Record what the person sees and the time it happened.
- Record the tenant, product area, and safe item ID.
- Keep the request ID when one is shown.
- Check System operations or Tenant operations.
- Find the related service, queue, or job.
- Open logs for the same time and service.
- Search by request ID when one is available.
- Open the matching audit record when the action changes important data.
- Write one evidence based explanation.
- Apply only the safe action that matches the cause.
- Repeat the original check once.
Stop and get help when
- Data versions do not match.
- A database or file store is unavailable.
- Several tenants or service groups are affected.
- The next action could lose or expose data.
Record
Record times, service group, state, queue or job ID, request ID, error code, action taken, and result. Remove personal information and secrets.
Full guide
Was this page helpful?
Your answer helps us improve the documentation.
Do not include personal information, customer information, passwords, or keys.