MySQL Flexible Server point-in-time restore failing with InternalServerError — West US 3

InternalServerError during restore is not explained in the provided material, but the context does cover how restores are expected to work and what options are available. Given that both point-in-time restore and geo-restore are consistently failing despite valid backup configuration, this is likely a platform-side issue that requires Azure support to investigate using the tracking IDs.

From the available information, the following checks and options are supported:

1. Validate that restore points exist and are within retention
   • In the Azure portal, open the MySQL Flexible Server instance.
   • Go to Backup and Restore and confirm that:
     • Full backups are listed within the configured retention period.
     • The restore point timestamps being used match entries in this list.
   • Restores in Flexible Server always create a new server from the original server’s backups; ensure the restore operation is being invoked in that mode (not in-place).
2. Use supported restore types and regions
   • Point-in-time restore:
     • Creates a new server in the same region as the original server.
   • Geo-restore:
     • Requires geo-redundant backup storage to be enabled (already confirmed).
     • Restores to a geo-paired or any supported Azure region where Flexible Server is available.
   • If point-in-time restore in West US 3 and geo-restore to eastus2 both fail with InternalServerError, this indicates an issue beyond configuration that must be checked by the service team.
3. Try a PowerShell-based point-in-time restore (alternative client)
   • If not already attempted, use the Az.MySql PowerShell cmdlet for a point-in-time restore to rule out client-side CLI issues:

      $restorePointInTime = (Get-Date).AddMinutes(-10)  # or a specific valid time within retention
      Get-AzMySqlFlexibleServer -ResourceGroupName <ResourceGroup> -ServerName <SourceServerName> |
        Restore-AzMySqlFlexibleServer -Name <NewServerName> -ResourceGroupName <ResourceGroup> -RestorePointInTime $restorePointInTime
   

   • This uses the documented Restore-AzMySqlFlexibleServer PointInTimeRestore flow and should succeed if the platform and backups are healthy.
4. Consider backup/restore characteristics and expectations
   • Restores can take longer depending on:
     • Database size.
     • Number of transaction logs to replay.
     • Network bandwidth (especially for geo-restore).
     • Number of concurrent restore requests in the target region.
     • Presence of primary keys in tables (recommended for faster recovery).
   • However, these factors affect duration, not the presence of an immediate InternalServerError.
5. Engage Azure support with tracking IDs
   • Because InternalServerError is a generic platform error and both restore paths fail despite valid configuration and retention, the next step is to open a support ticket with Azure.
   • Provide:
     • The server name and region (West US 3, Burstable tier Standard_B2s).
     • The exact restore type attempted (point-in-time vs geo-restore) and target region (for geo-restore).
     • The restore timestamps used.
     • The tracking IDs listed.
   • The support team can inspect backend logs for those tracking IDs, verify backup integrity, and identify any regional or service-side issues impacting restore.

Until the platform-side cause is identified and resolved, repeated attempts with the same parameters are unlikely to succeed.

References:

• Backup and restore in Azure Database for MySQL
• Restore-AzMySqlFlexibleServer
• Restore an Azure Database for MySQL flexible server by using Azure Backup (preview)