Before maintenance

• Identify the systems and customers affected.
• Create a suitable backup.
• Validate available disk space.
• Record current service state.
• Plan a rollback path for high-impact changes.

Recovery sequence

1. Identify the failed layer.
2. Stop unnecessary further mutation.
3. Preserve logs and diagnostic evidence.
4. Restore only the required component from a known-good source.
5. Validate websites, databases and dependent services.
6. Return customer traffic only after checks pass.

Post-recovery

Document the cause, corrective action and any preventive change required to reduce recurrence.