On April 29, 2025, at 06:20 UTC, we detected degraded performance across all regions. During the incident, customers attempting to load workspaces or models may have encountered intermittent time-out errors. In addition, API or CloudWorks™ integration jobs may have failed.
Our alerting identified that a component was intermittently failing to respond. We took manual steps to improve service and engaged our vendor for support. By 08:03 UTC, errors had subsided, and full service was restored.
We continued to work with our vendor to understand the issue. At 09:15 UTC, we started to see a recurrence of time-out errors. Through further analysis, we identified constraints affecting the cache of the impacted component. The size of the cache was increased, and performance started to improve shortly after 10:00 UTC. By 10:22 UTC, workspaces and models were reliably opening, but loading time was extended. Full service was restored at 10:49 UTC.
Our vendor completed a thorough analysis of the logs. They advised that the issue stemmed from exhaustion in the cache resource due to suboptimal configuration settings. The settings were modified during the incident to restore full service. Our vendor will be implementing additional configuration and capacity monitoring. We are also updating our monitoring to provide additional granularity into the functionality of the component to ensure action before customer impact.
We apologize for any disruption this may have caused to your business. We are committed to improving the reliability of our services. If you have any further questions or concerns, please visit our Support website. Thank you for your patience during this incident and thank you for being an Anaplan customer.