PHP-Based Platform for Business Monitoring & IT Availability Management

As a PHP Software Engineer at La Banque Postale, I designed and developed a suite of three interconnected applications for business process monitoring, alert management, and IT service availability reporting.

Unlike traditional infrastructure monitoring, which focuses on CPU, memory, or storage usage, our solution monitored whether critical business processes were functioning correctly. The platform allowed application owners to configure custom monitoring checks in seconds, such as verifying file generation, command execution, storage accessibility, or business-specific thresholds. When an anomaly was detected, the system could automatically create ServiceNow incidents, send email notifications, or trigger both actions.

We also developed maintenance filtering capabilities to prevent unnecessary alerts and an availability reporting application providing annual service availability statistics for IT management and internal auditors.

Built with PHP, a custom CakePHP-based framework, Angular, and REST APIs, the solution reused our existing distributed monitoring infrastructure and integrated with the company's major incident management applications.

First Application - Building Self-Service Business Process Monitoring

At Groupe La Poste, traditional infrastructure monitoring wasn't sufficient to determine whether business applications were operating correctly. A server could have perfectly normal CPU and memory usage while a critical application process was failing. We therefore developed a monitoring platform allowing application owners to configure custom functional checks without requiring specialized monitoring expertise.

The application supported several monitoring scenarios:

• File Generation: Verify that an expected file has been created.
• Command Execution: Execute commands and validate their results.
• Storage Accessibility: Check access to required storage resources.
• Business Thresholds: Monitor application-specific values and detect threshold violations.

The interface was designed to be simple enough for users to configure a new monitoring check in just a few seconds. When a check detected an anomaly, the application could automatically send email notifications, create ServiceNow incidents, or perform both actions.

Technical Decision - Reusing and Extending Distributed Monitoring Infrastructure

Rather than developing a new execution infrastructure, we reused the distributed monitoring architecture previously built for our performance analytics platform. This infrastructure included satellite collectors deployed across different network zones and a custom job scheduler.

The scheduler combined:

• A database storing job configurations and execution information.
• Cron-based scheduling to execute monitoring tasks.
• Distributed collectors running the scheduled checks on target servers.
• Centralized storage for monitoring results and execution history.

This allowed monitoring jobs to run at configurable frequencies across different infrastructure environments. Reusing the existing collection infrastructure significantly simplified the development of the new monitoring capabilities.

Second Application - Developing Maintenance-Aware Alert Filtering

One recurring problem was the generation of unnecessary alerts during planned infrastructure maintenance. Some monitoring configurations targeted entire groups of servers. When only a few servers were undergoing maintenance, their expected failures could generate notifications affecting a much larger audience.

To address this, we developed a dedicated alert-filtering application. Users could define and activate filters targeting specific servers or monitoring scopes.

When a filter was active:

• Monitoring execution continued normally.
• Detected anomalies were still recorded in the database.
• Events were explicitly marked as filtered.
• Email notifications and ServiceNow incident creation were suppressed for matching events.

This preserved monitoring history while preventing unnecessary notifications during maintenance.

Integration with other developed Applications - Major Incident Management

We also integrated the platform with our existing major incident management applications through REST APIs. When an anomaly required further investigation, users could review the monitoring results and create a significant incident directly from the application.

That incident could then be used by other tools to organize crisis management meetings, coordinate technical experts, and communicate with affected users. This connected automated monitoring with the broader incident management process.

Third Application - Building IT Availability Reporting for Audit and Management

In a banking environment, certain monitoring checks are particularly important for demonstrating service availability and operational compliance. We introduced a distinction between standard monitoring checks and auditable monitoring checks.

Unlike regular checks, auditable monitoring could not be suppressed by maintenance filters. We developed a third application dedicated to availability reporting and audit requirements.

It provided dashboards with interactive charts and statistics showing application availability over time, including annual availability percentages. These reports allowed IT management and internal auditors to evaluate service reliability, identify recurring availability issues, and define improvement objectives.

• Full-Stack Application Development: Developed PHP backend and Angular frontend features across the monitoring, alert-filtering, and availability reporting applications using our custom CakePHP-based framework.
• Monitoring Orchestration: Integrated configurable functional checks with the existing database-driven scheduler and distributed cron-based collector infrastructure.
• Alert Management & API Integration: Developed monitoring workflows supporting maintenance filters, ServiceNow incident creation, email notifications, and integration with major incident management applications.
• Availability Reporting & Audit Requirements: Contributed to auditable monitoring rules and dashboards providing historical service availability statistics for IT management and internal auditors.

• Delivered 3 complementary monitoring applications: Provided self-service business process monitoring, maintenance-aware alert filtering, and IT availability reporting through an integrated platform.
• Enabled monitoring configuration in seconds: Allowed application owners to create functional checks for file generation, command execution, storage access, and business thresholds without specialized monitoring expertise.
• Reduced unnecessary maintenance alerts without losing monitoring history: Suppressed notifications for filtered servers while retaining every detected anomaly in the database with an explicit filtered status.
• Introduced auditable annual service availability reporting: Implemented non-filterable monitoring checks and dashboards calculating application availability percentages over the year, supporting internal audits and management decisions.