We have restored phpMyAdmin startup for the affected MySQL services following our AWS migration.

Impact: Users attempting to open the browser-based database manager could see "The database manager could not start. Please try again." The confirmed impact was phpMyAdmin access. We did not establish an interruption to application-to-database connections, and no customer database contents were modified during recovery.

Cause: The manager workflow used the previous proxy verification endpoint, wrote routing files that the AWS non-root proxy could not read, and retained management addresses pointing to previous infrastructure.

Resolution: We corrected the verification endpoint and routing-file permissions, updated 10 AWS database-manager addresses, and applied and committed the code fixes. Authentication, isolation and HTTPS checks remain enforced.

Verification: Both reported services passed their automated database, authentication, proxy and HTTPS startup checks. A normal dashboard login to phpMyAdmin and a read-only SELECT 1 query succeeded on one affected service. All 13 existing phpMyAdmin runtime-policy tests passed. Customer browser login on the second service was not independently verified because the operator account did not have access to that workspace.

This is a retrospective incident report for 25 October 2026. The earliest confirmed failed launch we reviewed was at 07:40 UTC; the exact onset is unconfirmed. Recovery was verified during the investigation today. The publication timestamp is not the incident start time.

Our migration validation did not catch these management-access incompatibilities. Follow-up work is to add AWS-specific regression coverage and review related management features.

If you still encounter an error, reopen phpMyAdmin from your service's Connections page and contact support with the service name and approximate time of the attempt. We are sorry for the disruption and the time spent troubleshooting access.