This Root Cause Analysis (RCA) summarizes the service disruption related to the Pillar web application's security certificate.

Incident Summary
On October 8, 2026, the security certificate (SSL) for the Pillar web application (app.pillar.hr) and its associated customer subdomains expired. This caused users to encounter a "Your connection is not private" error message when attempting to access the platform or log in. The issue affected all customers using the Pillar web app and its API.

• Impact Window: October 8, 2026, 23:59 UTC – October 9, 2026, 18:15 UTC (~18 hours).
• Affected Services: Pillar Web Application, Tenant Subdomains, and Pillar API.

Detection
The incident was identified following a customer report to our Support team at 16:55 UTC on October 9, 2026. Because the expired certificate was part of a legacy manual process, no automated internal alerts were triggered at the time of expiry.

Root Cause
The disruption was caused by the expiration of a manually managed security certificate. Unlike modern automated systems, this specific certificate required a manual renewal and upload process once a year. The renewal was not initiated due to a lack of automated reminders and unclear internal ownership of the manual task.

Resolution
Service was restored through the following steps:

• Temporary Mitigation: A temporary security certificate was issued and applied to the service at 18:15 UTC on October 9, restoring immediate access for all users.
• Permanent Fix: The system was migrated to a managed certificate service. This new setup automatically handles renewals, removing the risk of manual oversight in the future.
• Verification: Engineering confirmed that all services were reachable and that security chains were valid across multiple subdomains.

Preventative Measures
To prevent a recurrence of this issue, the following actions have been completed or are underway:

• Automation: Successfully migrated to auto-renewing certificates managed by our cloud infrastructure provider, eliminating manual renewal steps.
• Infrastructure as Code: The certificate and load balancer configurations have been integrated into our automated infrastructure management tools to ensure visibility and consistency.
• Enhanced Monitoring: We are implementing new alerts that will notify the engineering team 30 days and 14 days before any certificate is set to expire, as well as alerts for any failures in the automated renewal process.
• Documentation: Updated internal documentation to clearly define ownership and access for domain and certificate management.

We sincerely apologize for the disruption this caused to your operations. We are committed to maintaining the reliability of our services and believe these automated improvements will prevent similar issues moving forward.