Automated Backup Provisioning – Banking Infrastructure & Data Protection

As Backup Product Owner at La Banque Postale, I led the development and integration of an automated backup provisioning solution designed to improve data protection across the bank's production infrastructure.

Previously, newly deployed servers were not automatically protected by a backup policy. Application owners had to request and configure backups manually, which frequently resulted in missing protection due to a lack of time, technical knowledge, or awareness.

To address this, I worked with the Server Automation team to integrate Veritas NetBackup provisioning directly into the server deployment process, using Ansible playbooks developed by my team. Every newly provisioned server could automatically receive a backup configuration based on predefined policies aligned with the application's Recovery Time Objective (RTO) and Recovery Point Objective (RPO).

The automation also performed an initial backup and restoration test to verify that data could actually be recovered.

This project was part of a broader strategy to achieve nearly 100% backup coverage across production infrastructure, strengthen regulatory compliance, and reduce the risk of critical data loss.

Project Context - Closing the Backup Coverage Gap

At La Banque Postale, ensuring that production servers are properly backed up is a critical requirement for business continuity and regulatory compliance. However, backup configuration was not automatically included in the server provisioning process.

After receiving a new server, application owners had to request backup activation separately. Many forgot, lacked the technical expertise, or postponed the task due to other priorities. As a result, some production servers remained unprotected.

As Backup Product Owner, I worked on a broader initiative to improve backup coverage and strengthen data protection practices across the bank. This involved regular interactions with internal stakeholders and representatives from regulatory and oversight organizations, including the Inspection Générale, ANSSI, and the European Central Bank (ECB).

One of our most important initiatives was to make backup protection a standard part of server provisioning, removing the need for users to configure it manually.

Designing Automated Backup Provisioning with Ansible

I worked closely with application owners and infrastructure teams to define a set of standardized backup policies adapted to different business requirements. These policies were designed around application criticality, recovery expectations, and RTO/RPO requirements.

We deliberately kept the selection process simple: application owners only needed to choose the appropriate backup profile when requesting a server. My team developed Ansible playbooks implementing the complete backup configuration workflow:

1. Install the NetBackup agent on the newly provisioned server.
2. Register the server with the appropriate NetBackup master server and backup policies.
3. Establish and validate connectivity between the backup agent and the master.
4. Apply the selected backup profile based on the options provided in the server creation form.
5. Execute an initial backup and restoration test to verify that the configuration was operational and data could be recovered.

The restoration test was particularly important. Successfully completing a backup doesn't necessarily mean the data can be restored when needed. We therefore included recovery validation as part of the automated provisioning process, helping detect configuration or connectivity problems from the beginning.

Coordinating Delivery Across Two Engineering Teams

One of my main responsibilities was coordinating delivery between the Backup Automation team and the Server Automation team. My team was responsible for developing the backup automation components, while the Server Automation team integrated them into the existing server provisioning workflows.

As Product Owner, I:

• Defined the product requirements and prioritized the development backlog.
• Worked with application owners to establish standardized backup policies aligned with RTO and RPO expectations.
• Coordinated the technical integration of Ansible playbooks into the server provisioning process.
• Managed dependencies and delivery planning with the Server Automation Product Owner.
• Established a release train to coordinate development and integration between both teams.
• Led technical discussions, user demonstrations, and communication sessions to support adoption.

Having previously worked as an engineer within the Server Automation team was a significant advantage. I understood their architecture, automation processes, and technical constraints, which made discussions more efficient and helped us resolve integration challenges faster.

The combination of my technical expertise and product ownership allowed me to bridge the gap between business requirements and engineering implementation. This allowed me to :

• Design and validate with real users a backup solution target : simple profiles to apply during server creation,
• Led discussion between backup engineering and Server automation teams,
• Coordinated all the technical tasks and the release,
• Share widely the new backup way of working in the company.

• Automated backup provisioning for newly created servers: Eliminated the separate manual backup activation request by integrating NetBackup directly into the standard server creation workflow.
• Introduced 5 automated provisioning and validation steps: From agent installation to an initial backup and restoration test, ensuring that servers were configured and their recoverability was verified.
• Standardized backup configuration across business use cases: Introduced reusable RTO/RPO-based profiles selectable directly from the server provisioning form, reducing configuration complexity and dependency on backup specialists.
• Contributed to the bank's near-100% production backup coverage objective: Removed a major source of unprotected servers—forgotten or delayed manual backup requests—while coordinating delivery across two engineering teams.