Harness Outage History

The AI Summary Feature was unavailable across all Insights in Harness for 7.2 hours. The issue was identified and a fix was implemented to restore full functionality. The incident has been resolved.

Harness experienced a service incident where delegate upgrades were failing for some customers. The issue was identified and a fix was implemented within 31 minutes. The incident has been resolved and the service is being monitored.

The Harness FME SDK experienced elevated error rates and delayed response times for 1.9 hours. The issue was classified as minor and affected SDK functionality. A fix was implemented and the incident was resolved after monitoring confirmed the solution was effective.

Harness experienced a major service degradation lasting 4.9 hours that affected multiple components including Continuous Delivery, CI/CD pipelines, Cloud Cost Management, and other platform services due to elevated queue depth and throughput saturation on the MongoDB layer. The root cause was identified as database processing latency, which caused slowness in deployment executions and UI updates, though executions continued to function. The issue was resolved by implementing optimizations to restore database stability and normal processing speeds.

Harness experienced deployment degradation with failures across production environments Prod1, Prod2, and Prod3, affecting all major services including Continuous Delivery, Continuous Integration, Cloud Cost Management, Feature Flags, and other platform components. The incident lasted 3.5 hours while the team investigated the widespread issues across their infrastructure. A fix was implemented and monitored before the incident was fully resolved.

The AI Summary feature in Harness SEI Insights became unavailable across all modules due to a service issue. This affected users' ability to access AI-generated summaries within the SEI platform. The engineering team investigated and resolved the functionality within 17 minutes.

The Harness Infrastructure as Code Management (IaCM) service experienced degraded infracost integration functionality in pipelines for 41 minutes. The issue was investigated and fully resolved on May 5th, 2026.

Harness experienced an issue where Kubernetes customer billing data became stale, affecting the Cloud Cost Management (CCM) component. The incident lasted 21 minutes and has been resolved.

Harness Prod3 environment experienced pipeline slowness affecting Continuous Delivery and Continuous Integration services including self-hosted runners and cloud builds across Mac, Windows, and Linux platforms. The incident lasted 3.7 hours, during which teams investigated the root cause and implemented a fix. The issue was resolved with pipelines returning to normal operation after monitoring confirmed the mitigation was successful.