What causes CapacityLimitExceeded?

The Fabric capacity SKU is too small for the combined workload running in the workspace. Other workloads (Spark jobs, dataflows, pipelines) are consuming all Compute Units and the capacity is throttled. A sudden spike in event volume overwhelmed the capacity allocation for the tenant

How do I fix CapacityLimitExceeded?

Contact your Fabric capacity administrator and ask them to review usage in the Capacity Metrics app.. The admin should identify which workloads are consuming the most CUs and either reduce them or upgrade the capacity SKU.. Once the capacity issue is resolved, manually reactivate any Activator rules that were paused — paused rules do not auto-resume.

High severitycapacityMicrosoft Fabric →

Microsoft Fabric Error:
CapacityLimitExceeded

What does this error mean?

A Fabric Activator rule stopped firing because the workspace's Fabric capacity has been overloaded for more than 24 hours. Throttling policies pause all Activator rule evaluations, background operations, and alert deliveries until capacity is restored.

Common causes

1The Fabric capacity SKU is too small for the combined workload running in the workspace
2Other workloads (Spark jobs, dataflows, pipelines) are consuming all Compute Units and the capacity is throttled
3A sudden spike in event volume overwhelmed the capacity allocation for the tenant

How to fix it

1Contact your Fabric capacity administrator and ask them to review usage in the Capacity Metrics app.
2The admin should identify which workloads are consuming the most CUs and either reduce them or upgrade the capacity SKU.
3Once the capacity issue is resolved, manually reactivate any Activator rules that were paused — paused rules do not auto-resume.

Beyond the docs

Common practitioner solutions not covered in the official documentation.

1Consider enabling Autoscale Billing for Spark workloads to offload CU consumption and protect alerting workloads from throttling

Frequently asked questions

Does this error mean I need to upgrade my capacity?

Not necessarily — capacity errors can be caused by query parallelism, large models, or concurrent refreshes. Optimize heavy datasets and schedule refreshes off-peak before upgrading.

Does this error affect all workspaces on the capacity?

Yes — Premium and Fabric capacities are shared. Heavy usage in one workspace can cause capacity errors in others on the same SKU.

Can I retry a failed refresh without waiting for the next schedule?

Yes — go to the dataset in Power BI Service and click 'Refresh now'. If the capacity issue was transient, the retry succeeds.

Does Power BI automatically retry capacity-related failures?

Power BI does not automatically retry failed refreshes. You need to trigger a manual refresh or wait for the next scheduled slot.

Source · learn.microsoft.com/en-us/fabric/real-time-intelligence/data-activator/activator-troubleshooting