A regression in datastore configuration during the addition of additional checksum/error detection resulted in a temporary loss of resiliency.The regression has been identified and forward-fixed to retain additional error detection.

The API is responding normally, and we are continuing to investigate the root causes.

Data is fully replicated.The API is responding normally, and we will continue to monitor through the weekend.

Restoration of full replication level is proceeding, and expected to complete with the next 24 hours.The API is responding normally, and we continue to monitor.

We have identified a faulty storage node, and are migrating data away from it. Until migration is complete, data is below normal replication levels.The API is functioning normally again.

The cluster error rates have returned to normal levels, but we continue to investigate the source of the problem.

Error rates are declining, but we are continuing to investigate.

We are continuing to investigate the cause of the issue and recover alternatives.We are working to make the most recent backup available as a fallback.

The timeseries and sequences datastore replication process began experiencing errors at 9:07 UTC, and API error rates for timeseries and sequences increased significantly.

This will be visible in services and applications reading and writing to timeseries and sequences, including Fusion.

We are continuing to investigate this issue.