🚀 Microsoft Fabric metadata created by E2E tutorials

This repository demonstrates how resources are created in a Git repository when Git integration is activated with a Fabric workspace created by following the E2E tutorials.

Tip

A workspace admin can connect a workspace to a Git repository. Using these resources can simplify resource creation in the end-to-end tutorial by minimizing manual steps.

📚 Documentation and Tutorial

• Microsoft Fabric portal
• Documentation
• End to end tutorials > Docs
• Fabric samples: GitHub repository for Fabric samples
• Microsoft Fabric Git integration?

📋 Resource Description

1. 🎯 Hands-on by end to end tutorials

• Part 1: Lakehouse 🌊
  • part 1 prefixed assets: Dataflow -> Semantic model -> Power BI
• Part 2: Lakehouse 🌊
  • part 2 prefixed assets: Data Pipeline -> Notebook -> Semantic model -> Power BI
  • tutorial_lakehouse
• Part 3: Data science 💙
  • part 3 prefixed assets
  • Random Forest (Bagging (parallel ensemble)), LightGBM (Gradient Boosting (sequential ensemble))
  • bank-churn-experiment
  • lgbm_sm
  • rfc1_sm
  • rfc2_sm
• Part 4: Real-Time Intelligence ⚡
  • part 4 prefixed assets: Eventhouse -> Eventstream -> KQL -> Realtime Dashboard
• Part 5: Data warehouse 🪣
  • part 5 prefixed assets: Data warehouse -> Data Pipeline -> SQL (Incl. Visual Query) & Notebook -> Semantic model -> Power BI Report
• Dataset 🗂️
  • part 1: dimension_customer.csv: Copy permalink from the official sample repository or see the dataset directory.
  • part 2: Import parquet files from https://fabrictutorialdata.blob.core.windows.net/sampledata/ using Data Pipeline. Check the details in the tutorial.
  • part 3: bankcustomer_churn.csv: see the dataset directory
  • part 4: Use the sample data provided by Fabric by default (Bicycle rentals)
  • part 5: Import parquet files from https://fabrictutorialdata.blob.core.windows.net/sampledata/ using Data Pipeline. Check the details in the tutorial.
    • https://fabrictutorialdata.blob.core.windows.net/sampledata/WideWorldImportersDW/tables/fact_sale.parquet
    • https://fabrictutorialdata.blob.core.windows.net/sampledata/WideWorldImportersDW/tables/dimension_city.parquet

2. 📝 Supplement

• Semantic Link
  • Semantic Link all-in-one demo.ipynb

3. ⚙️ Environment setup

• Fabric stores metadata only. Actual data (OneLake files, table contents) and connection secrets/credentials are not stored in Git. After a fresh sync/clone into another workspace, you'll typically need to recreate or rebind connections and re-enter credentials.
• Manually fix a failed update
• Re-deploy the semantic model:
  1. Copy the SQL connection string from the SQL analytics endpoint settings.
  2. Copy the Lakehouse analytics GUID from the Fabric URL. Lakehouse analytics GUID: Copy the second GUID from the Fabric URL
     (e.g. https://app.fabric.microsoft.com/groups/<group-id>/lakehouses/<lakehouse-id>?experience=fabric-developer)
  3. Update these values in Fabric_artifacts/agentic_semantic_model.SemanticModel/definition/expressions.tmdl, commit, and push.
  4. Trigger a Source Control update in Fabric and wait for completion.
• Re-deploy the metadata
  1. Modify the datawarehouse endpoint in part5_Load Customer Data.DataPipeline\pipeline-content.json
  2. Modify the kusto endpoint in part4_eventhouse_tutorial.KQLDatabase\EmbeddedRealTimeQueryset.json

Trouble shooting

Troubleshooting GitHub sync issues, especially when syncing data to another tenant.

1. Set the value to null to resolve the “connection not found” issue during synchronization with the GitHub repository.

   • part2_IngestDataFromSourceToLakehouse.DataPipeline/pipeline-content.json
   • part5_Load Customer Data.DataPipeline/pipeline-content.json

   "externalReferences": {
       "connection": ""
   },

2. The email address in the Fabric Event Alert Tutorial should be replaced with your own email address. The placeholder is set to [email protected].

   • part4_WorkspaceItemChange.Reflex/ReflexEntities.json

   "definition": {
   "type": "Rule",
   "instance": "[email protected].."

   Set your tenant id and workspace id in the json.

   "tenantId": "...", // Get this from Azure Portal
   "workspaceId": "..." // Get from your Microsoft Fabric URL. https://app.fabric.microsoft.com/groups/YOUR_WORKSPACE_ID

3. Recreate connection for DataPipeline

   • part2_IngestDataFromSourceToLakehouse.DataPipeline: see this tutorial
   • part5_Load Customer Data.DataPipeline: see this tutorial

Resource list

Reference

• End-to-end tutorials in Microsoft Fabric