Harness Outage History

Harness experienced degraded Autostopping service in AWS Middle East South 1 region due to elevated latency from AWS, affecting Cloud Cost Management operations including warm-up, cool-down, schedule execution, and traffic detection. The issue also impacted CCM Asset Governance functionality for resources in the affected region. The incident was resolved by isolating the problematic region to restore functionality for other customers, while resources in me-south-1 continued experiencing issues until the region fully recovered.

Harness experienced slow authentication operations in their Feature Flag SDK on the Prod2 environment, affecting the Feature Flags component for 2.4 hours. The issue was identified and a fix was implemented to restore normal performance. The incident was resolved after monitoring confirmed the fix was successful.

Harness experienced degraded performance affecting SCIM users during login for 47 minutes. The issue was identified and a fix was implemented to restore normal login functionality. The incident was resolved after monitoring confirmed the fix was effective.

The Harness Test Intelligence service experienced an outage that impacted Continuous Integration Enterprise cloud builds across Mac, Windows, and Linux platforms. The issue was identified and resolved with a fix implementation after 39 minutes of downtime. All affected CI/CD build services were restored to normal operation.

The Harness FME SDK experienced elevated error rates affecting impressions and events starting around 6:45AM PT on March 17, 2026. This minor incident impacted the SDK's ability to properly process impressions and event data for approximately 7.1 hours. The issue was resolved after investigation and monitoring, with full resolution confirmed at 4:38PM PT the same day.

Harness experienced intermittent degraded performance in CI Steps across Prod 2 and Prod 3 environments, affecting all Continuous Integration Enterprise cloud builds including Mac, Windows, and Linux platforms. The issue was identified and resolved with a fix implementation, with the incident lasting 1.4 hours total.

Harness experienced a UI display issue where some pipeline executions from March 11 incorrectly showed as "running" even though they had actually completed, aborted, or failed due to a previous incident. This was purely a visual artifact that did not impact customers' ability to start new pipeline executions, and the issue was resolved after 3 hours by clearing the problematic display data.

Harness experienced a major incident affecting pipelines and dashboards in their Prod2 environment, impacting Continuous Delivery, Cloud Cost Management, Continuous Integration builds, and Custom Dashboards. While pipeline executions continued to run normally, users experienced delays in viewing pipeline status and dashboard information through the UI. The incident was resolved after 3.2 hours once fixes were implemented and monitoring confirmed full recovery.

Harness experienced delays in GCP billing data export starting around March 8, causing incomplete cost data for GCP resources in their Cloud Cost Management service. The issue was caused by problems with Google Cloud's billing export system and affected the CCM component for 68.5 hours. The incident was resolved on March 12 after Google Cloud fixed the underlying billing export delays.

Harness Prod4 experienced login issues that prevented users from accessing multiple services including Continuous Delivery, Cloud Cost Management, Continuous Integration, Feature Flags, and other core platform components. The incident lasted 1.7 hours and was resolved after implementing a fix and monitoring the results.