deployment-admin（部署管理员）

Deployment and Administration

Overview

Power BI deployment spans from development to production across workspaces, capacities, and environments. This skill covers deployment pipelines, CI/CD automation, security configuration, capacity management, and governance best practices.

Deployment Pipelines

Built-in Power BI feature for promoting content through environments:

Requirements: Premium, PPU, or Fabric capacity on all stage workspaces.

Deployment rules:

• Content types deployed: reports, semantic models, dataflows, paginated reports
• Parameterization rules handle environment-specific values (server names, databases)
• Backward deployment (prod to dev) is supported but use with caution
• Auto-bind connects deployed reports to the correct semantic model in each stage

Pipeline Automation via REST API

# Get deployment pipelines
GET https://api.powerbi.com/v1.0/myorg/pipelines

# Deploy all content from stage 0 (Dev) to stage 1 (Test)
POST https://api.powerbi.com/v1.0/myorg/pipelines/{pipelineId}/deployAll
{
  "sourceStageOrder": 0,
  "options": {
    "allowOverwriteArtifact": true,
    "allowCreateArtifact": true,
    "allowOverwriteTargetArtifactLabel": true
  },
  "note": "Automated deployment from CI/CD"
}

CI/CD with GitHub Actions

PBIP-Based Deployment

name: Power BI Deploy
on:
  push:
    branches: [main]
    paths: ['reports/**']

permissions:
  id-token: write
  contents: read

jobs:
  deploy:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4

      - name: Azure Login (Service Principal)
        uses: azure/login@v2
        with:
          client-id: ${{ secrets.AZURE_CLIENT_ID }}
          tenant-id: ${{ secrets.AZURE_TENANT_ID }}
          subscription-id: ${{ secrets.AZURE_SUBSCRIPTION_ID }}

      - name: Get Access Token
        id: token
        run: |
          TOKEN=$(az account get-access-token \
            --resource https://analysis.windows.net/powerbi/api \
            --query accessToken -o tsv)
          echo "::add-mask::$TOKEN"
          echo "token=$TOKEN" >> $GITHUB_OUTPUT

      - name: Deploy via Fabric REST API
        run: |
          # Import PBIP to workspace using Fabric Git integration
          # or use deployment pipeline API
          curl -X POST \
            "https://api.powerbi.com/v1.0/myorg/pipelines/${{ secrets.PIPELINE_ID }}/deployAll" \
            -H "Authorization: Bearer ${{ steps.token.outputs.token }}" \
            -H "Content-Type: application/json" \
            -d '{
              "sourceStageOrder": 0,
              "options": {
                "allowOverwriteArtifact": true,
                "allowCreateArtifact": true
              }
            }'

Validation Step (Best Practice Analyzer)

      - name: Run Tabular Editor BPA
        run: |
          # Install Tabular Editor CLI
          dotnet tool install -g TabularEditor.TOMWrapper
          # Run Best Practice Analyzer
          tabulareditor model.bim -A BPARules.json -V

CI/CD with Azure DevOps

trigger:
  branches:
    include: [main]
  paths:
    include: ['reports/*']

pool:
  vmImage: 'ubuntu-latest'

steps:
  - task: PowerBIActions@5
    displayName: 'Deploy to Test'
    inputs:
      PowerBIServiceConnection: 'PowerBI-ServicePrincipal'
      Action: 'Publish'
      WorkspaceName: 'Sales-Test'
      PbixFile: '$(Build.SourcesDirectory)/reports/*.pbix'
      OverWrite: true

  - task: PowerBIActions@5
    displayName: 'Refresh Dataset'
    inputs:
      PowerBIServiceConnection: 'PowerBI-ServicePrincipal'
      Action: 'DatasetRefresh'
      WorkspaceName: 'Sales-Test'
      DatasetName: 'SalesModel'

Azure DevOps extension: Install "Power BI Actions" from the Visual Studio Marketplace for native pipeline tasks.

Fabric Git Integration

Native source control integration (GA since late 2023):

1. Connect workspace to repo: Workspace settings > Git integration > Connect
2. Supported repos: Azure DevOps Repos, GitHub (2025+)
3. Sync direction: Workspace to repo (commit) or repo to workspace (update)
4. Supported items: Reports, semantic models, notebooks, pipelines, lakehouses

Branching strategy:

• Main branch connected to Production workspace
• Feature branches for development
• PR-based review and merge
• Auto-sync on merge to main

Security

Row-Level Security (RLS)

Define in Power BI Desktop (Modeling > Manage Roles):

// Static RLS - filter by specific values
[Region] = "West"

// Dynamic RLS - filter by logged-in user
[Email] = USERPRINCIPALNAME()

// Dynamic RLS with lookup table
[ManagerEmail] = USERPRINCIPALNAME()
|| PATHCONTAINS([ManagerPath], LOOKUPVALUE(
    Employees[EmployeeID],
    Employees[Email], USERPRINCIPALNAME()
))

Testing:

• Desktop: Modeling > View as > select role (does not support USERPRINCIPALNAME)
• Service: semantic model settings > Security > test role with specific user
• Embedded: generate embed token with effective identity

Rules:

• RLS applies to Viewers only; Admins/Members/Contributors bypass RLS
• Must add users/groups to roles in the Power BI Service
• Service Principal can test RLS by generating tokens with EffectiveIdentity
• RLS filters propagate through relationships (single-direction)
• With bidirectional filtering, test carefully for security leaks

Object-Level Security (OLS)

Restrict access to specific tables/columns for certain roles. Configured via:

• Tabular Editor (recommended)
• TOM/.NET SDK
• XMLA endpoint with TMSL

// TMSL to add OLS
{
  "createOrReplace": {
    "object": { "database": "model", "role": "RestrictedUser" },
    "role": {
      "name": "RestrictedUser",
      "tablePermissions": [{
        "name": "Employees",
        "columnPermissions": [{
          "name": "Salary",
          "metadataPermission": "none"
        }]
      }]
    }
  }
}

Capacity Management

Fabric F-SKU equivalency (2026):

Key rule: F64 is the minimum Fabric SKU that includes Power BI content viewing rights for users without Pro/PPU licenses.

Workspace Management

Best practices:

• Use security groups for role assignment (not individual users)
• Separate workspaces per environment (Dev/Test/Prod)
• Use apps for end-user content distribution
• Limit Admin role to 2-3 people per workspace
• Enable audit logging for governance

Power BI Report Server (On-Premises)

For comprehensive Report Server guidance, see references/report-server-detail.md. Key facts: requires SQL Server Enterprise SA or Premium license; uses PBIX only (no PBIR); release cycle January/May/September; latest version January 2026.

Fabric Git Integration Updates (2025-2026)

Git integration enhancements since initial GA:

• GitHub support added alongside Azure DevOps Repos (2025)
• Real-time intelligence items (Eventstream, KQL DB, Data Activator) now support Git integration
• Dataflow Gen2 supports CI/CD and Git integration
• Deployment pipelines support Git-triggered automation
• Improved conflict resolution for PBIR-format reports (per-visual file granularity)

Additional Resources

Reference Files

• references/governance-checklist.md -- Tenant settings, audit logging, sensitivity labels, data loss prevention, and compliance checklist
• references/report-server-detail.md -- Power BI Report Server on-premises: versions, feature comparison, REST API, security, deployment architecture