Platform Alerts

On June 17, 2026, at 06:50 UTC, we became aware of an issue affecting a subset of workspaces in our eu4: Cloud - Europe region. Customers with affected workspaces were unable to load models, seeing persistent loading screens or errors when opening their work. CloudWorks™ integrations in the region also experienced a period of degradation, from approximately 05:20 UTC to 06:26 UTC, because of the same underlying issue. The impact was limited to a specific subset of workspaces — all other regions, and many workspaces in eu4: Cloud - Europe, continued to operate normally. Full service for the affected workspaces was restored at 10:00 UTC. 

 Root cause 

The affected workspaces ran on a previous storage configuration in the eu4 region. Over time, a storage directory used by these workspaces accumulated a large number of small temporary files. The directory reached its operational limit and could no longer accept new file operations. This prevented models from loading for customers whose workspaces were running on this configuration. 

Recovery 

Our engineering team cleared the accumulated temporary files, which provided immediate relief. We then moved all affected workspaces from the old configuration to the new configuration through an automated process. We checked that model loading was restored through testing. By 10:00 UTC, the issue was fully resolved. On the same day, we implemented a change across the eu4 region to make sure no workspace can be directed to the old configuration. 

Corrective and preventative actions 

We've taken the following actions to address this incident and prevent it from happening again: 

1. We've deployed an update across all regions that prevents any workspace from being directed to the older storage configuration. This eliminates the specific configuration condition that caused this incident. 
2. We're decommissioning the older storage configuration. This permanently removes the configuration that caused this incident. 
3. We’re strengthening our automated validation checks to confirm that workspaces are running on the correct configuration. This catches any configuration mismatch before it can cause customer impact. 
4. We're implementing automated cleanup rules for the temporary files involved in this incident. This prevents storage directories from reaching operational limits and supports faster automatic recovery. 

Closing 

We apologize for the impact this issue has had on your operations. We're committed to the improvements outlined above to prevent similar disruptions. If you have questions or concerns, please contact Support.