What the Fabric Monitoring Hub covers
The Fabric Monitoring Hub is Microsoft's built-in operational view for Fabric workspaces — available to all workspace members at no additional cost and requiring zero configuration. It shows the run history and status of all Fabric items — Data Pipelines, semantic models, notebooks, Dataflows Gen2, Lakehouses, and Eventstreams — in a single, filterable list.
For teams that have fully migrated to Fabric, the Monitoring Hub can be the only operational monitoring tool they need. It provides deep execution details — Spark job metrics, notebook run-level output, and Lakehouse operation history — that purpose-built external tools do not capture. For Spark notebooks and jobs in particular, the Monitoring Hub is the primary and most detailed native monitoring surface.
The Monitoring Hub can send email failure notifications for most scheduled item types via its 'Schedule failures' feature, currently in preview (semantic models are not yet supported in this feature). It does not group failures into incidents or correlate activity across different item types.
Where MetricSign extends coverage
Most enterprise data stacks are not purely Fabric-native. A common pattern: raw data lands via Azure Data Factory or Databricks, gets transformed by dbt, and feeds Fabric semantic models consumed by Power BI reports. The Monitoring Hub sees only the Fabric portion of this chain.
MetricSign covers the full chain. ADF pipeline failures, Databricks job errors, and dbt run failures — including both dbt Cloud accounts and dbt Core CLI pipelines, not only Fabric-native dbt jobs — are all surfaced alongside Fabric semantic model refresh failures in a single incident feed. When an upstream ADF pipeline fails, the affected downstream Fabric semantic models appear in the same incident — so the root cause is immediately clear without navigating across tool dashboards.
MetricSign also adds the alerting layer the Monitoring Hub lacks. When a semantic model fails at midnight, MetricSign sends an alert to the configured channel. The on-call engineer does not need to check the Fabric workspace to know something is wrong.
Complementary roles in a hybrid stack
The Fabric Monitoring Hub and MetricSign are not direct substitutes. The Monitoring Hub is the deepest native view of Fabric-internal activity, including notebook run details and Spark execution metrics that MetricSign does not capture. For debugging within Fabric, the Monitoring Hub is the right place to go.
MetricSign provides the cross-tool operational layer: know that something failed, know which downstream systems are affected, and get the alert before your users notice. The two tools complement each other in any stack that uses both Fabric and non-Fabric sources.