Amperity is experiencing failing Predictive workflows due to a Databricks outage

We have successfully restarted all failed workflows and the incident has been resolved.

Databricks has confirmed that their incident is resolved. Amperity is unpausing all Predictive tasks and will be restarting the workflows that failed due to this issue.

Amperity is tracking a Databricks outage that might cause Predictive workflows to fail. We are monitoring the status of resolution on the Databricks side. In the meantime, we are pausing all Predictive tasks and will re-enable them once the Databricks incident is resolved. During this time, the workflows will remain in the Running state, even if all the other tasks succeed.

The Databricks incident status can be tracked here: https://status.databricks.com/

If the Databricks incident is resolved within 24 hours, most customers should not see any observable impact to their operations that rely on Predictive inference.