The issue has been fixed on all tenants. We are still actively working on preventing any further occurence and improving our detection capabilities.
Monitoring · UTC
The mitigation is in progress and should take a few minutes to take effect on all the impacted tenants.
Identified · UTC
The cause of the issue has been identified. It comes from the concurrent occurence of several bugs in the third-party tool we're using in our deployment pipelines.
Investigating · UTC
Our temporal workers responsible for running the source refresh, as well as some housecleaning tasks, are failing to start on a few of our tenants. Both manual and scheduled refresh operations are impacted. The user experience on our web application and the data quality monitors are not impacted.