At approximately 12:25 AM UTC on Wednesday March 27th, some customers may have received alerts from the YugabyteDB Managed system indicating their clusters were offline. Most of these alerts were false positives due to an administrative issue in YugabyteDB Managed impacting management of database clusters running in AWS. Only a small number of clusters were directly impacted by the issues, but a larger number received false positive alerts due to missing data in our internal monitoring infrastructure. Once the administrative issue was resolved, all management of those clusters returned to normal.
For the small number of clusters which were severely impacted by the outage, all of them were free Sandbox clusters, which do not provide any redundancy against failure. All production customer clusters were non impacted by this event. All of the impacted clusters have been returned to normal operations as of 17:00 UTC.