EU1 – Slowness in API Requests

Incident: EU1 - Slowness in API Requests
Region: EU1
Time: Wed, July 23 – 14:45 to 20:03 PDT

Summary
On July 23 and July 24, the EU1 region experienced slowness in API requests due to an edge-case scenario involving a combination of specific customer-side configuration and insufficient handling of that edge case within our platform. This led to suboptimal distribution of print jobs within the rendering cluster. After addressing the situation, system performance returned to normal.

What Happened
A particular combination of parameters in a customer’s configuration triggered a scenario where print jobs were not evenly distributed across the rendering cluster. Although the platform remained functional, the affected rendering logic did not optimally balance the load, resulting in slowness during periods of high demand. The behavior exposed an opportunity to improve how the platform handles unusual or edge-case job distribution patterns.

Resolution
The engineering team diagnosed the problem and implemented an adjustment that improved the handling of such edge-case scenarios. This update ensured more efficient job distribution across the rendering cluster. After the fix was applied, performance normalized and the issue did not reoccur.

Next Steps
To prevent similar issues in the future, we will:

• Enhance exception and edge-case handling in job distribution logic
• Improve monitoring and alerting for resource underutilization in the rendering cluster
• Expand detection logic to proactively identify unusual job distribution patterns