Back to status
MajorResolved

Intermittent errors accessing Light Platform

Some users experienced intermittent errors when accessing parts of Light Platform

Summary: Some users experienced intermittent errors when accessing parts of Light Platform.

Root cause: A connectivity issue affected traffic being routed to one of several application instances. Requests routed to that instance could fail, while requests handled by the other instances continued normally.

Resolution: We identified and replaced the affected application instance. Service has remained stable following the change.

Customer impact: Some users experienced intermittent errors when accessing parts of Light Platform.

Follow-ups: Close the identified gap in automated health monitoring so this type of connectivity and routing failure can be detected and alerted on earlier.

Affected components
Platform API
Duration
1 hr 15 min
Resolved
Opened
Timeline
Post-mortem

Summary: Some users experienced intermittent errors when accessing parts of Light Platform.

Root cause: A connectivity issue affected traffic being routed to one of several application instances. Requests routed to that instance could fail, while requests handled by the other instances continued normally.

Resolution: We identified and replaced the affected application instance. Service has remained stable following the change.

Customer impact: Some users experienced intermittent errors when accessing parts of Light Platform.

Follow-ups: Close the identified gap in automated health monitoring so this type of connectivity and routing failure can be detected and alerted on earlier.

Resolved

This incident has been resolved. We identified and replaced the affected application instance, and Light Platform has remained stable following the change.

Investigating

Some users experienced intermittent errors when accessing parts of Light Platform.

A connectivity issue affected traffic being routed to one of several application instances. Requests routed to that instance could fail, while requests handled by the other instances continued normally.

We identified and replaced the affected application instance. Service has remained stable following the change.

We are closing the identified gap in automated health monitoring so this type of connectivity and routing failure can be detected and alerted on earlier.