Dear Azure Support Team, We would like to report an incident involving our Azure Database for PostgreSQL instance where no transactions were recorded on 12 January, despite the application operating normally. Issue summary Date of incident: 12 January

Question

Dear Azure Support Team,

We would like to report an incident involving our Azure Database for PostgreSQL instance where no transactions were recorded on 12 January, despite the application operating normally.

Issue summary

• Date of incident: 12 January
• Database service: Azure Database for PostgreSQL
• Impact: Database shows no transaction activity for the entire day
• Observed behavior:
  • Application services were running and accessible
    • No database errors were logged at the application or database level
      • No service outage or alerts were triggered
        • Transactions before and after 12 January were recorded normally

What we have checked

• Application logs: No errors or failed write attempts observed
• Database logs: No errors, restarts, or failover events detected
• Connection pooling and credentials: Unchanged and working before/after the incident
• Monitoring metrics: No obvious anomalies detected from our side

This raises concern that transactions may not have been persisted or were silently dropped during that period.

Request We would appreciate your assistance to:

1. Investigate whether there were any underlying platform issues, maintenance activities, or internal errors affecting transaction recording on 12 January.
2. Confirm if any writes could have been lost, delayed, or routed incorrectly.
3. Advise on additional diagnostics or safeguards we should implement to prevent or detect similar issues in the future.

Please let us know if you require:

• Database server name
• Resource ID
• Subscription ID
• Region
• Relevant timestamps (UTC)Dear Azure Support Team, We would like to report an incident involving our Azure Database for PostgreSQL instance where no transactions were recorded on 12 January, despite the application operating normally. Issue summary
  • Date of incident: 12 January
  • Database service: Azure Database for PostgreSQL
  • Impact: Database shows no transaction activity for the entire day
  • Observed behavior:
    • Application services were running and accessible
    • No database errors were logged at the application or database level
    • No service outage or alerts were triggered
    • Transactions before and after 12 January were recorded normally
  What we have checked
  • Application logs: No errors or failed write attempts observed
  • Database logs: No errors, restarts, or failover events detected
  • Connection pooling and credentials: Unchanged and working before/after the incident
  • Monitoring metrics: No obvious anomalies detected from our side
  This raises concern that transactions may not have been persisted or were silently dropped during that period. Request
  We would appreciate your assistance to:
  1. Investigate whether there were any underlying platform issues, maintenance activities, or internal errors affecting transaction recording on 12 January.
  2. Confirm if any writes could have been lost, delayed, or routed incorrectly.
  3. Advise on additional diagnostics or safeguards we should implement to prevent or detect similar issues in the future.
  Please let us know if you require:
  • Database server name
  • Resource ID
  • Subscription ID
  • Region
  • Relevant timestamps (UTC)

Answer 1

It appears that you are experiencing an issue with your Azure Database for PostgreSQL instance where no transactions were recorded on 12 January, despite the application functioning normally. Here are some points to consider regarding your situation:

1. Unplanned Downtime: Although your application was accessible, it is possible that there were underlying issues or maintenance activities that affected transaction recording. Azure Database for PostgreSQL has mechanisms to handle unplanned downtimes, but they may not always be visible at the application level.
2. Monitoring and Alerts: Since you mentioned that no service outages or alerts were triggered, it would be beneficial to review the service health notifications in the Azure portal. This can help determine if there were any known issues on that date that could have impacted your database. You can check the Service Health page in the Azure portal for any active events or issues.
3. Transaction Logs: If you suspect that transactions may not have been persisted, you might want to investigate the transaction logs and metrics available for your database. Azure Database for PostgreSQL provides various metrics that can help you analyze transaction activity and identify any anomalies.
4. Point-in-Time Recovery: If you are concerned about data loss, consider implementing point-in-time recovery (PITR) strategies. This allows you to restore your database to a specific point in time, which can be useful if you need to recover from user errors or other issues.
5. Future Safeguards: To prevent similar issues in the future, consider setting up alerts for transaction metrics and implementing more robust monitoring solutions. This can help you detect issues in real-time and take action before they impact your application.

If you need further assistance, it may be helpful to provide the specific details you mentioned, such as the database server name, resource ID, subscription ID, region, and relevant timestamps (UTC) to Azure Support for a more thorough investigation.

---

References:

• Overview of business continuity with Azure Database for PostgreSQL
• Monitor metrics in Azure Database for PostgreSQL