Introducing the Power Automate Monitoring Dashboard for Azure Data Explorer

Shaf Mahar

Like(1)

Report

The Microsoft FastTrack for Dynamics 365 team has published a number of assets that give you a kick start on predefined dashboards, queries, and how to consume App Insights. These resources are available on GitHub and can help you get started, refer GitHub - microsoft/Dynamics-365-FastTrack-Implementation-Assets: Dynamics 365 FastTrack Implementation guides.

We have recently made the Power Automate Monitoring Dashboard available in GitHub that provides a comprehensive view of all Power Automate telemetry available in Azure App Insights data. This dashboard consists of two integrated pages. The first page, "Performance Monitoring," and “Errors and Exceptions” providing Power Automate runs, triggers, actions and any underline errors and exceptions.

Let's dig into details of these Indvidual pages.

Performance Monitoring Page

The "Performance Monitoring" page displays detailed information about Power Automate executions, types of flows, weekly and monthly executions, failures, and exceptions in your environment. This allows you to keep track of your Power Automate activities and identify any issues that may arise.

Using this dashboard, you can identify:

Flow executions details

This view provides the list of distinct flows which are executing on the environment in specific timeframe selected in the dashboard. You can use this FlowId in the other queries to gain insights about the specific flow.

Source Query:
requests
| where customDimensions['resourceProvider'] == 'Cloud Flow'
| extend cd = parse_json(tostring(customDimensions["Data"]))
| distinct FlowDisplayName = tostring(cd.FlowDisplayName), FlowId=tostring(customDimensions['resourceId'])

Types of flows Tile

This tile shows the types of flows which are executing in the environment within selected time range. If you are expecting for a certain type of flow to be executing in a time period and don’t notice it here, you can use the query below to identify the flow and investigate trigger rules.

Source Query:
dependencies
| where type == "Cloud Flow/Cloud flow triggers"
| summarize count() by name

Flow execution status Tile

This tile shows number of times specific flow has executed within selected time period. This is useful query which shows flow volumes. You can use this query to create alerts if you want to be notified on presence/absence of expected volume of flows.

Source Query:
requests
| extend cd = parse_json(customDimensions)
| where cd.resourceProvider has "Cloud Flow"
| extend data = parse_json(tostring(cd.Data))
| extend FlowName = tostring(data.FlowDisplayName)
|summarize nRun = count(), nFailed = countif(success==false) by FlowName
| extend nSuccess = nRun-toint(nFailed)
| project FlowName, nRun,nSuccess,nFailed
| order by toint(nRun) desc

Weekly & Monthly Executions Tiles

These tiles shows the selected flow’s statistics based on Month and Week executions’ durations.

Source Query:
requests
| where customDimensions.resourceProvider == 'Cloud Flow'
| extend data = todynamic(tostring(customDimensions.Data))
| summarize count() by resultCode, bin(timestamp, 7d)
| render columnchart

And
requests
| where customDimensions.resourceProvider == 'Cloud Flow'
| extend data = todynamic(tostring(customDimensions.Data))
| summarize count() by resultCode, bin(timestamp, 30d)
| render columnchart

Flow Run Details Tiles

After identifying the flow you want to monitor and investigate further, you will need to find the Flow Run ID from previous steps and apply it in the filters, for example:

These tiles should display the details of a specific selected Flow Run ID, such as:

Flow Runs Tile: Shows the Flow Runs and their specific times.
Flow Action of Specific Run: Displays the Flow Actions on that specific run.
Flow Triggers: Provides details of the Flow triggers for that specific run, including the duration of each trigger (there could be multiple triggers).
Number of Actions of the Selected Run: Shows the details of the actions within the flow for that specific run, including the time taken for each action. This view helps identify any performance bottlenecks or opportunities to pinpoint slow-performing actions.
Action Statistics: Displays the statistics in terms of success codes, such as the number of successful or failed actions. (Hint: The query ignores skipped actions, but you can edit the query to include them.)

Error and Exception Monitoring Page

The second page of the dashboard focuses on errors and exceptions in Power Automate runs, triggers, and actions. By using this page, you can identify problematic flows, failed triggers, and most problematic actions, helping you troubleshoot and resolve issues quickly.

For instance, you can identify the action that has failed within a specific time range or for a particular flow run. As shown in the screenshot, there is an action named Call_GetExpiredCompleteStepActionEvents that failed. For further investigation, you can view the customDimensions column, which will provide hints about the root cause of the failure. Once you identify the root cause, you can fix it and continue monitoring future executions.

How to connect Azure Data Explorer with App Insights instance

To connect Azure Data Explorer to your App Insights instance, you will need to use a specific URL and add your Subscription ID. Here are the key components you will need to update the URL:

TENANT_ID: ID of your Azure AD Tenant where your Azure Application Insights instance lives.
SUBSCRIPTION_ID: ID of the Azure Subscription that contains the Application Insights instance.
RESOURCE_GROUP: Name of the Resource Group where the Application Insights instance is deployed.
APPINSIGHTS_INSTANCE_NAME: Name of the Application Insights instance.
ENCODED_KQL_QUERY: Encoded KQL query to execute.
TIMESPAN: Time filter for the query (optional)

Simply, go to https://dataexplorer.azure.com/ and then connect Dashboard’s to data sources with and https://ade.applicationinsights.io/subscriptions/[SUBSCRIPTIONID]
And for running direct queries connect complete URL à https://ade.applicationinsights.io/subscriptions/[SUBSCRIPTIONID]/resourcegroups/[InstanceNAME]/providers/microsoft.insights/components/[RESOURCEGROUP]

How to Import this dashboard in Azure Data Explorer

Steps to import the sample dashboard:

Import the file "dashboard-Power Automate Monitoring.json". Refer https://github.com/microsoft/Dynamics-365-FastTrack-Implementation-Assets/blob/master/MonitoringAndTelemetry/Power%20Automate/dashboard-Power%20Automate%20Monitoring.json

Name the dashboard appropriately and then click to select datasources

In the Datasource selection pane you have to put your Azure Application Insights subscriptionID in the placeholder .

After updating the correct subscriptionID. click on connect.
You will get a list of databases. Select your ApplicationInsights name from that list and save changes.
your dashboard should have data now. Feel free to edit the queries to suit your needs.

Conclusion

Azure Data Explorer based Power Automate Monitoring Dashboard combined with Power Platform Admin Center (PPAC) based reports, provides a powerful solution for real-time data exploration and analysis on telemetry. By leveraging these tools, you can gain valuable insights, monitor your Power Automate activities, and troubleshoot issues effectively. Whether you are dealing with log and telemetry data or time series analysis on executions, Power Automate Monitoring Dashboard have got you covered.

Community site session details