Skip to main content

baseline-monitoring

Monitoring for Microsoft Fabric​

Monitoring for Microsoft Fabric

Monitoring Microsoft Fabric landing zone components is essential to ensure their availability, reliability, security, and scalability. Effective monitoring allows your organization to detect issues early, optimize performance, and ensure business continuity for your data workloads.

Why Monitoring Matters for Fabric​

Microsoft Fabric provides a unified SaaS platform for data analytics. With multiple integrated components like Data Factory, Synapse Data Engineering, Data Warehousing, and Power BI, a holistic monitoring strategy is required:

  • Availability: Ensure services like Lakehouses, Pipelines, Eventstreams, and Power BI reports are functioning properly.
  • Performance: Monitor data refresh latency, query durations, and resource consumption (e.g., capacity usage).
  • Security: Track access control changes, DLP violations, and sharing activities.
  • Scalability: Measure workspace growth, dataset sizes, and workload throughput.

Baseline Monitoring for Fabric​

Microsoft Fabric integrates with Azure Monitor and Microsoft Purview. The recommended components for baseline monitoring include:

  • Microsoft Fabric Admin Portal Logs: Provides audit trails for workspace, item, and permission changes.
  • Microsoft Purview: Offers data classification, access policy violations, and compliance insights.
  • Log Analytics Workspaces: Collect logs and metrics across Fabric and supporting Azure services.
  • Power BI Metrics App: Visualize and report on usage, performance, and refresh failures.

Policies and Alerts​

Use Azure Policy and the Azure Monitor Baseline Alerts (AMBA) framework to configure deployIfNotExists alert rules for supporting services such as:

  • Azure SQL (for linked endpoints)
  • Azure Storage (used in Dataflows Gen2 and Eventstreams)
  • Azure Key Vault (for secrets and identities)
  • Azure Monitor Action Groups (for alert routing)

Specific alert recommendations:

  • Dataset refresh failures
  • Capacity overload (CPU/memory limits)
  • Pipeline run failures
  • Fabric item permission changes
  • Power BI gateway failures

Integration Scenarios​

  • Visualize Alerts: Connect Fabric telemetry to a centralized Azure Monitor workspace and visualize in Power BI.
  • Notify Operations Teams: Configure alert Action Groups with email, webhook, or Teams channels.
  • Automated Remediation: Use Logic Apps or Azure Automation to restart failed pipelines or notify stakeholders.

Advanced Monitoring Considerations​

For enterprise-grade implementations:

  • Use Azure Lighthouse for cross-tenant monitoring.
  • Leverage Microsoft Sentinel for security event correlation and detection.
  • Adopt Azure Monitor Workbooks to build custom dashboards spanning Fabric, Azure, and hybrid services.

Resources​

Contributors