Normal data collection resumed at 8:18 AM EDT.

The affected window was approximately 5 hours and 27 minutes, from 2:51 AM to 8:18 AM EDT. Data from that period is unavailable.

The service startup issue has been corrected and additional buffering has been added to reduce the impact of similar interruptions.

An automatic operating-system update restarted several backend services. One internal data pipeline service did not recover correctly, interrupting data collection.