Recently, we introduced Insights support for Azure Logic Apps Standard, offering curated visualizations as a foundation for monitoring Azure services. These visualizations enable you to monitor Logic Apps Standard applications and workflows effectively.
In integration scenarios, where multiple systems and services interconnect, monitoring is crucial. Azure Logic Apps often serve as the backbone of these integrations, orchestrating complex workflows.
Whether you’re a developer, IT operations team member, or product owner, ensuring that systems run smoothly is essential. This release helps you gain a clear view of data flows, quickly detect and resolve issues, and maintain seamless operations.
Scope
The monitoring dashboards are designed specifically for Logic Apps Standard. The scope of the dashboard is the workflows inside the Logic App. The compute-based dashboard leverages the underlying App Service plan for that Logic App.
Data source
Most charts are based on Azure Metrics, requiring no user configuration. However, run details are not available in metrics, so we use Application Insights as the data source for these.
Visualization framework
The dashboard is built using Azure Workbooks, offering a variety of visualizations. Workbooks are easily extendable, allowing for customization to meet specific needs.
Pre-requisites to use the feature
To effectively use the dashboards, your Logic Apps must be linked to Application Insights to enable telemetry collection. Additionally, it's crucial to ensure that your application uses the updated V2 schema for telemetry. Here are the steps to update the host.json file to implement the new schema.
Capabilities
The dashboard is organized into four distinct tabs. The workflow starts with an aggregated view of all runs, followed by a summary at the workflow level, and then drills down into the details of specific workflow runs. The final tab is dedicated to performance analysis, focusing on compute utilization.
Overview
This section provides a bird’s eye view of your application’s health by displaying the success rates of runs, actions, and triggers. The accompanying table offers detailed counts of runs that have failed, succeeded, or been skipped.
Additionally, trend charts illustrate the patterns of runs, actions, and triggers over the selected time range. This overview helps you quickly assess the status and timeline of your runs. If failures are detected, you can proceed to the workflows tab to pinpoint and analyze them within specific workflows.
Workflows
The workflow and run information is organized into two tables: one focused on successful runs and the other on failed runs. Each table provides detailed, workflow-specific information, including the total number of runs, their statuses, and a breakdown of actions and triggers.
The timeline offers a visual representation of traffic, especially useful in the failure table to track transitions between failed and successful states. At this stage, you can drill deeper into specific workflows and their runs, with detailed information available in the Runs tab.
Runs
The run details are not available in metrics and are pulled from AppInsights associated with your Logic App. For this table to work properly, you need to ensure –
- Your Logic Apps is associated with AppInsights resource to push telemetry
- The telemetry is based on V2 schema. The below screenshot shows the flag that needs to be added to the Settings in the host.json. Learn more about the v2 schema.
The runs page offers filters to refine runs by workflow name, run status, and client tracking ID. You can select a specific run by clicking on a row, which will open a detailed table showing the actions and their statuses for that run. This helps you identify the exact action that caused a run to fail. Additionally, you can click through the rows to explore more about related dependencies.
Another useful feature in the runs table is the ability to resubmit runs. This is particularly valuable when workflow failures are due to intermittent issues or downstream failures that resolve over time.
Compute
Logic Apps Standard operates on the customer’s compute, providing greater control over the scaling and performance of applications. This section offers insights into compute utilization for your workloads. It’s important to note that compute resources can be shared across multiple Logic Apps, so if you’re sharing an App Service Plan or Workflow Standard Plan, this should be considered.
The first two charts display job count and execution delay in the underlying runtime engine. These metrics help diagnose unexpected performance issues by indicating whether load has increased (more jobs) or whether compute is temporarily unavailable / constrained. You can learn more about the relevance of these concepts here.
The next two charts provide number of nodes (VMs) and the average memory utilization across them. The trend charts should help you see how your application scales up and down and CPU and memory utilization based on those traffic patterns.
These charts represent the memory and CPU utilization trends for the entire App Service plan. If the plan is shared across applications, this will give you the utilization across applications. You should be able to use this in combination with above charts to get a better view into the compute utilization of this Logic App v/s others that use the same compute.
What’s Next
We’re excited to bring you out-of-the-box monitoring capabilities for Logic Apps Standard. We have many enhancements planned for future iterations, including:
- Bulk resubmission of runs
- Easy one-click transitions between charts (e.g., click on a workflow name to see its runs)
- Customizable workbooks to tailor the monitoring experience to your business needs
If this feature is important to you, we’d love to hear your feedback, feature requests, or input on our roadmap.
Microsoft
