We believe all errata at this point to be resolved. There are some clear causes and room for improvement, mostly in graceful degradation in some metrics components. Please write in immediately if anything is out of place. Our apologies.
Update
UTC
Update
UTC
The problem was a fault in monitoring. The reporting is improving, we're watching for any systems that have been negatively affected. There may be some extra false positive restarts in this time.
Investigating
UTC
Investigating
UTC
We're currently investigating. Provisioning affected, as is responsiveness to failures. False positives may also cause additional downtime. We will keep you posted.
