Example: Retrying SQL Queries for Applications

If the primary DN is faulty and cannot be restored within 40s, its standby is automatically promoted to primary to ensure the normal running of the cluster. Jobs running during the failover will fail and those started after the failover will not be affected. To protect upper-layer services from being affected by the failover, refer to the following example to construct a SQL retry mechanism at the service layer.