04 - Logging and Monitoring (Flex Consumption)¶

Set up observability for your Flex Consumption app so you can verify deployments, inspect failures, and monitor scale behavior.

Prerequisites¶

What You'll Build¶

You will generate real requests and exception telemetry from a Flex Consumption app and validate the data path in Application Insights.

flowchart TD
    INET[Internet] -->|HTTPS| FA[Function App\nFlex Consumption FC1\nLinux Python 3.11]

    subgraph VNET["VNet 10.0.0.0/16"]
        subgraph INT_SUB["Integration Subnet 10.0.1.0/24\nDelegation: Microsoft.App/environments"]
            FA
        end
        subgraph PE_SUB["Private Endpoint Subnet 10.0.2.0/24"]
            PE_BLOB[PE: blob]
            PE_QUEUE[PE: queue]
            PE_TABLE[PE: table]
            PE_FILE[PE: file]
        end
    end

    PE_BLOB --> ST["Storage Account\nallowPublicAccess: false\nallowSharedKeyAccess: false"]
    PE_QUEUE --> ST
    PE_TABLE --> ST
    PE_FILE --> ST

    subgraph DNS[Private DNS Zones]
        DNS_BLOB[privatelink.blob.core.windows.net]
        DNS_QUEUE[privatelink.queue.core.windows.net]
        DNS_TABLE[privatelink.table.core.windows.net]
        DNS_FILE[privatelink.file.core.windows.net]
    end

    PE_BLOB -.-> DNS_BLOB
    PE_QUEUE -.-> DNS_QUEUE
    PE_TABLE -.-> DNS_TABLE
    PE_FILE -.-> DNS_FILE

    FA -.->|User-Assigned MI| UAMI[Managed Identity]
    UAMI -->|RBAC| ST
    FA --> AI[Application Insights]

    subgraph DEPLOY[Deployment]
        BLOB_CTR[Blob Container\ndeployment-packages]
    end
    ST --- BLOB_CTR

    style FA fill:#107c10,color:#fff
    style VNET fill:#E8F5E9,stroke:#4CAF50
    style ST fill:#FFF3E0
    style DNS fill:#E3F2FD

flowchart LR
    Client[HTTP requests] --> Func[Flex Function App]
    Func --> AI[Application Insights]
    AI --> KQL[KQL queries via Azure CLI]

Steps¶

Step 1: Set Variables¶

export BASE_NAME="flexdemo"
export RG="rg-flexdemo"
export APP_NAME="flexdemo-func"
export PLAN_NAME="flexdemo-plan"
export STORAGE_NAME="flexdemostorage"
export APPINSIGHTS_NAME="flexdemo-insights"
export LOCATION="koreacentral"

Expected output:

Step 2: Confirm Application Insights Wiring¶

az monitor app-insights component show --app "$APPINSIGHTS_NAME" --resource-group "$RG" --output json
az functionapp config appsettings list --name "$APP_NAME" --resource-group "$RG" --query "[?name=='APPLICATIONINSIGHTS_CONNECTION_STRING']" --output json

Expected output:

{
  "appId": "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx",
  "id": "/subscriptions/<subscription-id>/resourceGroups/rg-flexdemo/providers/microsoft.insights/components/flexdemo-insights",
  "name": "flexdemo-insights",
  "type": "microsoft.insights/components"
}

[
  {
    "name": "APPLICATIONINSIGHTS_CONNECTION_STRING",
    "slotSetting": false,
    "value": null
  }
]

Step 3: Generate Traffic¶

curl --request GET "https://$APP_NAME.azurewebsites.net/api/health"
curl --request GET "https://$APP_NAME.azurewebsites.net/api/info"
curl --request GET "https://$APP_NAME.azurewebsites.net/api/exceptions/test-error"

Expected output:

{"status":"healthy","timestamp":"2026-01-01T00:00:00Z","version":"1.0.0"}

{"error":"Handled exception","type":"ValueError","message":"Simulated error for testing"}

Step 4: Query Request and Exception Telemetry¶

az monitor app-insights query --app "$APPINSIGHTS_NAME" --analytics-query "requests | where timestamp > ago(30m) | project timestamp, name, resultCode, duration | order by timestamp desc | take 20" --output json
az monitor app-insights query --app "$APPINSIGHTS_NAME" --analytics-query "exceptions | where timestamp > ago(30m) | project timestamp, type, outerMessage | order by timestamp desc | take 20" --output json

APPINSIGHTS_ID=$(az monitor app-insights component show \
  --app "$APPINSIGHTS_NAME" --resource-group "$RG" \
  --query "appId" --output tsv)
az monitor app-insights query --apps "$APPINSIGHTS_ID" --analytics-query "..."

Expected output:

{
  "tables": [
    {
      "name": "PrimaryResult",
      "rows": [
        [
          "2026-01-01T00:00:00Z",
          "GET /api/health",
          "200",
          "00:00:00.0430000"
        ]
      ]
    }
  ]
}

Step 5: Check Scale-Sensitive Telemetry¶

Flex can scale to zero and out to 1000 instances, so monitor cold starts, errors, and dependency latency.

az monitor app-insights query --app "$APPINSIGHTS_NAME" --analytics-query "traces | where timestamp > ago(30m) | where message contains 'Function started' or message contains 'Host lock lease acquired' | project timestamp, severityLevel, message | order by timestamp desc | take 50" --output json

Expected output:

{
  "tables": [
    {
      "name": "PrimaryResult",
      "rows": [
        [
          "2026-01-01T00:01:12Z",
          1,
          "Function started (Id=xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx)"
        ]
      ]
    }
  ]
}

Step 6: Logging Guidance for Flex¶

• Keep structured logs in Python (logging.info(..., extra={...})) for KQL filtering.
• Treat deployment verification as: workflow logs + health checks + App Insights (not Kudu).
• Track trigger-specific behavior separately; queue-triggered functions scale per function.

Verification¶

• GET /api/health and GET /api/info appear in the requests table with resultCode 200.
• GET /api/exceptions/test-error appears in traces/requests and returns a handled exception payload.
• KQL output confirms recent telemetry ingestion within the last 15-30 minutes.

Next Steps¶

Next: 05 - Infrastructure as Code

See Also¶

• Tutorial Overview & Plan Chooser
• Python Language Guide
• Platform: Hosting Plans
• Operations: Deployment
• Recipes Index

Sources¶

• Monitor Azure Functions
• Azure Monitor Logs query overview
• Application Insights overview