What happened:
During a planned migration process, we hit a scaling limit on a core infrastructure component which impacted certain customer traffic.
Impact:
A portion of customer traffic was dropped causing a full outage for customers on Silos 7, 8, and 9.
Fix:
The migration was rolled back in all customer environments.
Prevention:
The migration has not been reapplied and is on hold, pending further investigation by the team to establish a safe migration path.