Microsoft Azure Logic Apps

Quick starts

With this understanding of Azure Logic Apps, lets straight away jump into some of the key components of Logic Apps and understand how it works?

Workflows

Workflow in an Azure Logic App is a sequence of actions or processes defined in steps. The actions will perform the business operations when input is provided through the managed connectors. Every workflow will be initiated by the trigger being fired. By defining the Workflow in Azure Logic Apps, Business processes can be automated and visualized in a hierarchy. Below is the image of the Azure Logic App workflow:

Managed Connectors

What are Logic App connectors?

Connectors in Azure Logic Apps are used to perform certain actions or processes. These connectors are designed to connect and work with user data. There are numerous connectors available for Azure Logic Apps including Enterprise connectors. Users can use pre-defined connectors as well as create their own custom connectors. It is also possible to define connectors using ARM templates. Connectors in Azure Logic Apps may act either as actions or as triggers.

How do I add connectors to Logic App?

1. Move to the Azure portal and click on Create a resource.
2. Enter Logic Apps Custom Connector in the search box and select it from the drop-down.
3. In Logic Apps Custom Connector, select Create.
4. Register the connector by providing the details and choose Review + create.
5. Review the details entered and click on Create.

After creating the custom connector resource, the menu opens automatically. If it doesn’t, you can go to the subscription and resource group you selected and open it directly. With the custom connector in hand, you can define the connector’s behaviour.

Actions

Actions are the steps those run once initiated by the trigger. Whenever a Azure Logic App gets triggered, there will be a succession of actions those run to complete the workflow. Actions are processes those will perform the designated business task based on the data provided by the user. Users can select the required action from the extensive set of actions available in the connector repository.

Triggers

Triggers are the starting point for a Azure Logic App workflow that will fire when new data or event that meets the trigger condition occur. Connectors in Logic Apps itself provide various triggers. Custom triggers can also be created using custom connectors.

How do you trigger a Logic App?

1. Create a blank Logic App in the Azure Portal
2. Select or search the trigger from the list given in the Logic App Designer
3. Run a trigger in a saved Logic App by using the Run Trigger option in the overview section

Can a Logic App have multiple triggers?

Yes, multiple triggers can be defined in a Logic App. Multiple triggers can be accommodated in a Logic App workflow. As mentioned by Microsoft Azure, a single Logic App can have up to 10 triggers in it. In general, triggers are of two types:

• Built-in triggers
• Managed API triggers

Built-in triggers are the predefined triggers using Http protocol and schedule for triggering the Azure Logic Apps. There are several Built-in trigger types like HTTP, HTTP Endpoint, Recurrence, and Request.

Managed API triggers are based on the API connections / API Webhook connections which check an endpoint using custom connectors.

How to create API connection in Azure Logic App?

1. Select All services > Web > API Connections from the Azure portal menu.
2. Select All resources from the Azure portal menu. The Type filter should be set to API Connection.

Let us further discuss the creation and working part of Azure Logic Apps.

How does Logic App work?

Azure Logic App works based on the workflow defined. Consider the Logic App in the Fly Wheel Cab Booking Application, defined with a trigger, Service Bus Trigger – “When a message is received in a queue”, followed by a series of actions. Now whenever the queue receives the message on a new booking, it will trigger the series of actions to run and workflow to complete.

The Values from one action will be passed on to the subsequent action through the dynamic values. According to the Azure Logic App pricing model, the cost for a Logic App is based on the Number of Actions, triggers and runs executed.

Here are the basic operations of a Logic App:

How to create a Logic App?

The first step to start with a Logic App is to create one. Following are the steps to create a Logic App:

1. In the Azure Portal, click on create a resource and select Logic Apps from the marketplace
2. Click create to proceed
3. Provide the basic credentials and click Review + Create
4. Finally, click on Go to resource to navigate to the Logic App created.

How do I authenticate my Azure Logic App?

Based on where your logic app processes execute, Azure Logic Apps provides the system-assigned managed identity, which you can use with only one logic app resource, and the user-assigned managed identity, which you may share across a collection of logic app resources.

1. Open your logic app resource in the Azure portal.
2. Select Identity from the logic app’s Settings menu.
3. Select On > Save in the Identity window under System assigned. When Azure asks if you want to confirm, say yes.
4. The system-assigned identity, which is registered with Azure AD and represented by an object ID, can now be used by your logic app resource.

How do you call Azure Logic Apps?

Below are the steps to call a Logic App from an existing Logic App

1. Select New step > Add an action under the step where you want to invoke another logic app.
2. Select Built-in from the drop-down menu under Choose an action. Enter logic apps as a filter in the search box. Choose a Logic Apps workflow from the activities list.
3. Inside the existing logic app, nest a logic app.
4. The designer displays the logic apps that are eligible for selection.
5. Choose a logic app to call from the current logic app.

Next steps:

• Azure Logic Apps Custom Connectors by Kent Weare
• Azure Logic App Life Cycle – The Big Picture by Steef-Jan Wiggers
• How to Call Deprecated Connector Actions on Logic Apps by Sandro Pereira
• Azure Logic App Expressions

Concepts

With an understanding of basic parameters of Azure Logic App, let’s jump on to the concepts those need to be considered when dealing with Azure Logic Apps.

Logic App state

The Azure Logic App can be in Enabled/Disabled state. User can modify the Logic App state from portal itself.

How do I turn off Azure Logic App?

1. In the Azure portal, open your Logic App.
2. Under Logic App’s menu select Overview and click on Disable option from the toolbar.
3. To disable multiple Logic Apps, move into the Logic App section, bulk select and disable the Logic Apps.

When it comes to Logic App trigger and run, there are various states possible. Below are the basic states of the Logic App run:

• Succeeded – When the Logic Apps run/trigger is succeeded
• Running – When the Logic App run/trigger is Running
• Failed – When the Logic App run/trigger is Failed
• Skipped – When the Logic App run/trigger is Skipped
• Timed Out – When the Logic App run/trigger is Timed Out

With the various state of the Logic App, users can check the health of the Logic App.

How long can a logic app run?

The default run limit for synchronous actions in a multi-tenant Logic App is 120 seconds that is 2 minutes.

How do you troubleshoot Logic App?

Logic app generates data that can help you in diagnosing and debugging errors in the app. In Azure portal, you can diagnose a logic app by analysing each stage of the workflow.

1. Check trigger history
2. Check runs history
3. Perform runtime debugging

Define workflow to meet business needs

Let us consider the Flywheel cab booking application, to define the Cab booking Scenario as a workflow using Azure Logic Apps. The client application will send the Booking message consisting of business data like User ID, Driver ID, Location, Price and Destination to a Service bus a queue. Now the requirement is to check whether the User ID and Driver ID is valid. Here comes the Logic App with the service bus queue trigger – “When a message is received in queue”. Now the Action can be the invocation of Function apps which will check whether the User ID and Driver ID in the Service bus message is valid or not. If Both User ID and Driver ID are valid, the Label of the Message will be set to true and the message will be sent to Azure Service bus topic through Topic connector. Now whenever the Service bus queue receives the booking message, the message will be validated and moved to Azure Service Bus Topic Subscription.

From the above Scenario, we can understand how the Logic App works with various triggers and actions.

Understand Logic App metrics

Azure metrics give information on the status of the resources in the Azure Subscription. With the extensive set of metrics, user can detect the health of the Azure Logic Apps. Metrics can also help troubleshoot root-cause of issues without needing to contact Azure support.

Action Metrics

These are the metrics related to Actions in the Logic App workflow

Metric Name	Description	Monitoring Perspective
Action Latency	The average time taken between when an action starts and ends Unit: Seconds Aggregation: Total Dimension: Entity Name	Performance
Actions Succeeded	The number of Actions succeeded in the Logic App runs in the given time Unit: Count Aggregation: Total Dimension: Entity Name	Availability
Actions Failed	The number of Actions failed in the Logic App runs in the given time Unit: Count Aggregation: Total Dimensions: Entity Name	Reliability
Action Skipped	The number of Actions skipped in the Logic App runs in the given time Unit: Count Aggregation: Total Dimensions: Entity Name	Frequency
Billable Action Executions	The number of Actions that cost for the Execution Unit: Count Aggregation: Total Dimensions: Entity Name	Consumption

Run Metrics

These metrics are related to the runs of the workflow:

Metric Name	Description	Monitoring Perspective
Run Latency	The average time taken between when a run starts and ends Unit: Seconds Aggregation: Total Dimension: Entity Name	Performance
Run Succeeded	The number of runs succeeded in the given time Unit: Count Aggregation: Total Dimension: Entity Name	Availability
Run Failed	The number of runs failed in the given time Unit: Count Aggregation: Total Dimensions: Entity Name	Reliability
Run Skipped	The number of runs skipped in the given time Unit: Count Aggregation: Total Dimensions: Entity Name	Frequency
Run Success Percentage	The number of successful runs divided by the total Unit: Count Aggregation: Total Dimensions: Entity Name	Reliability

Trigger Metrics

These metrics are related to the triggers of the workflow:

Metric	Description	Monitoring Perspective
Trigger Fire Latency	The time taken for firing the trigger of the Logic app in the given time Unit: Seconds Aggregation: Total Dimensions: Entity Name	Performance
Trigger Succeeded	The number of triggers succeeded in the given time Unit: Count Aggregation: Total Dimension: Entity Name	Availability
Trigger Failed	The number of triggers failed in the given time Unit: Count Aggregation: Total Dimensions: Entity Name	Reliability
Trigger Skipped	The number of triggers skipped in the given time Unit: Count Aggregation: Total Dimensions: Entity Name	Frequency
Billable Trigger Execution	The number of Triggers that cost for the Execution Unit: Count Aggregation: Total Dimensions: Entity Name	Consumption

How do I debug Logic App?

When it comes to debugging the Azure Logic Apps, there will be a need for triggers and run history to check if every action has performed well. When there is any failure in at least one of the run actions, the whole run will be shown as the failure. Users can get into the failed run and debug the failure. Following are the steps to debug a Logic App:

1. On the Logic App overview section, click on see trigger history to check if any triggers have failed.
2. Next, check for failed runs in the run history listed in the overview section.
3. Click on the failed run and perform runtime debugging.

Azure Logic App pricing

Azure Logic App Pricing is completely based on executions of triggers, actions, and connectors.

• Actions – $0.000025/ Exec
• Standard Connector – $0.000125/ Exec
• Enterprise Connector – $0.001/ Exec

For Integration Service Environment (ISE), the pricing counts on the base unit and scale unit. For Integration account with Logic App B2B, the pricing will vary based on the tiers

• Basic – $0.42/ Hr
• Standard – $1.37/ Hr

What are the benefits of using Azure Logic Apps?

Azure Logic Apps have remarkable advantages which can solve various business challenges.

Below are some of the advantages of Azure Logic Apps

• Easy to define the workflow – User can easily define the workflow with triggers and actions with user-intuitive GUI
• Support for Enterprise applications – Logic Apps support integration with Office 365, Twitter, Dropbox and more
• Easy debugging – User can easily debug the Logic App using Trigger and Run history
• Automation of workflow – By defining the workflow with triggers, it will automatically fire the Logic App when the trigger condition is succeeded
• Integration with on-premise solutions – It is also possible to integrate the Logic App with on-premise solutions like BizTalk server, SAP and more

Logic App Integration Account

Integration account with Logic Apps will be needed for Enterprise Integration and B2B solutions. By Integrating the Integration account with Logic Apps, user can store all the artefacts at one place, build B2B workflows, integrate with Third-party SaaS applications and more.

How do I secure my Azure Logic App?

Logic App ISE

Integration Service Environment (ISE) is where users can run their Logic App which can interact with the services as VMs secured inside the Azure Virtual Network. The Logic App and its storage will be isolated from public Logic App services. Hence isolating the Azure Logic Apps will improve its performance too.

Sentiment Analysis with Azure Logic Apps

Till now we got the light on Logic App Integration, Pricing, and Metrics. Now the focus will be on some use cases like Sentiment Analysis with Azure Logic Apps. In this Sentimental Analysis, we will rank the tweets with the hashtag by using Logic Apps and other services. Whenever there is a tweet with some hashtag, the Azure Logic App will get triggered and with the help of Cognitive Service, it will score the tweet and store it into the blob storage and some of the score data will be sent to Power BI for analytics. Read more in detail on Sentimental Analysis with Azure Logic Apps

Azure Logic Apps with Mock Data

Another use case we can see is Apartment Customer Service built with Azure Logic Apps. In this scenario, the customer wants to place a ticket to the customer support of the apartment about the maintenance work. Logic App will watch the ticket if the request by the customer is addressed. Once the request is addressed, it will send the mail to the customer saying the work is completed. Read more on Testing Azure Logic Apps with Mock Data for the implementation of this Scenario.

Next steps:

• Connecting Azure Logic Apps with Integration Account
• Azure Logic Apps – Integration Service Environment (ISE)
• Sentimental Analysis with Azure Logic Apps by Steef-Jan Wiggers
• Testing Azure Logic Apps with Mock Data by Kent Weare

Native monitoring tools for Logic Apps

By now it is very clear how powerful Azure Logic Apps are. The ease of addressing the business use cases using Logic App connectors makes them the best choice while building complex business orchestrations. However, managing and monitoring them in the Azure portal is not so simple and straight forward. This opens scope for tools to better manage and monitor Azure Logic Apps.

Log Analytics

Log Analytics is a tool to execute the Azure monitor Log Queries. By turning on the Log Analytics while creating the Logic App, user can perform monitoring and diagnostics logging for Azure Logic Apps. User can also query the log for efficient debugging. By configuring the Log Analytics, user can get the complete view. E.g., number of actions failed, number of triggers failed and more. Below are the detailed steps to enable Log Analytics for Azure Logic App:

How do I enable Log Analytics in Azure Logic App?

You can enable log analytics once you create your Logic App. Follow these steps on the Azure portal, where you submit the details to create your logic app:

1. Select On under Log Analytics.
2. Select the workspace where you want the data from your logic app to be sent from the Log Analytics workspace list.
3. After you complete this step, Azure will create your logic app, which will be linked to your Log Analytics workspace. This step also installs the Logic Apps Management solution in your workplace automatically.
4. Select Create when you’re finished.

How do you check logs in Azure Logic App?

You may view the data from your logic app’s runs in your Log Analytics workspace after it has completed.

1. Find and open your Log Analytics workspace via the Azure portal. 
2. Select Workspace summary > Logic Apps Management from the workspace menu. 
3. Select the row for a logic app or status to see all the runs for that logic app or status. You can also view tracked properties by clicking View in the Tracked Properties column for actions where you put them up. Use the column filter to find the properties that have been tracked. 
4. Client-side and server-side filtering are both options for filtering your results. 
• Client-side filter: Choose the filters you want for each column. 
• Server-side filter: Use the scope control at the top of the page to select a certain time window or to limit the number of runs that display. By default, only 1,000 records are displayed at once. 
5. Select the row for a logic app run to see all the actions and their information for that run. 

Next Steps:

• Insights using Log Analytics Author: Kent Weare

App Insights

As mentioned, App Insights is available only for the Logic App Standard resource type. When you create a new Azure Logic Apps (Standard) instance, you can directly associate it with App Insights, with which you can determine the overall health of the service nodes that are hosted within your Azure Logic Apps (Standard) instance.

Azure Monitor

Get to set up monitoring and alerting on the Azure Logic Apps. With this, you get to monitor the run status, Logic App triggers and failures. For instance, you can configure to get notified whenever a Logic app run fails more than a certain number of times.

Enterprise-grade monitoring tool: Serverless360

Azure Logic Apps solve huge business challenges with easy workflow design and automation but managing and monitoring the Azure Logic Apps in application context is not possible with Azure Portal. To solve this Challenge user must go ahead with Serverless360. Serverless360 is the solution for managing and monitoring Azure Serverless Applications. With the help of Serverless360, user can manage and monitor the Azure Logic Apps in the efficient way.

Serverless360 vs Azure Monitor

Azure Monitor restricts monitoring Logic Apps only on a few metrics per alert rule and if you want to monitor it on more metrics, additional alerts have to be configured. This results in increased cost, whereas a platform like Serverless360 allows to monitor any number of metrics under a single alert.

Capability	Azure Monitor	Serverless360
Monitoring Logic Apps on multiple metrics at no extra cost	✗	✓
Unified monitoring at the application level	✗	✓
Intuitive Dashboards	✓	✓
Alerts on critical problems	✓	✓
Consolidated alerting	✗	✓
Auto-correct Logic App status	✗	✓
Root-cause analysis	✗	✓
Repair & resubmit failed Logic App runs	✗	✓
End-to-end tracing	✗	✓

Serverless360 vs Insights using Log Analytics

Since Application Insights is not available for Azure Logic Apps, user can monitor the status of the runs with a small work around using Log analytics and Logic App Management. While creating the Azure Logic Apps, user need to enable the Log Analytics and workspace. Once the Logic App is created, user need to enable the Send to Log Analytics and configure the appropriate workspace in Diagnostics Settings. Now the user can monitor the runs in Logic App Management workspace. With this work around, user can only track the status of the runs, actions and connectors.

Whereas with Serverless360, user can manage and monitor Azure Serverless resources as well as with BAM in Serverless360, user can perform end-to-end distributed tracing on the Business Process and the message flow through the orchestration. With Serverless360, user can monitor the Azure resources in various perspectives and it also has integration with Various Notification channels like

• Microsoft Teams
• Slack
• Pager Duty
• SMTP
• OMS
• Web Hook
• Service Now

Visit Serverless360 vs Application Insights page for more information.

Serverless360 vs Log Analytics

With Log Analytics, User can view the monitor logs on whether the run is succeeded or not, but it is not possible to monitor the state of every Actions and triggers. For this Challenge, user need to go with Serverless360 for monitoring Azure Logic Apps. In Serverless360, user can monitor the Performance, latency, Billable executions and more.

Next Steps:

• Serverless360 vs Application Insights

Manage Azure Logic App in Serverless360

With Serverless360, user can perform various remote actions and operations on Azure Logic Apps with all actions Governed and Audited and possible to monitor the Azure Logic Apps with extensive set of metrics

Modify state

In Serverless360, the user can modify the state of the entity like Enable/Disable or run the triggers if there is a need to test the Azure Logic Apps. User can also access the properties like Resource Group Name, Region Name and Subscription Name of the Logic Apps.

Access Trigger and run details

With the help of Serverless360, the user can access the trigger and run history. User can also access the Input and Output bound of the Logic Apps triggers and Run Actions. This would be of great help for DevOps engineers to investigate why a run action failed?

Search run histories

In Serverless360, User can search the run history either by Identifier or with some time bound and status. It is also possible to access the state of a Logic App run and Input/ Output bounds of the Run Actions. Fetching the failed run on a specific date and time and further looking into its input and output bound will be a key feature to have to support effective debugging.

Resubmit runs

There would be a scenario like Logic App runs fail due to the wrong configuration. After changing the Configuration, user would need to test Logic App with the same input. With Resubmission in Serverless360, user can resubmit the Logic App runs and achieve correlation.

Correlate run resubmission

In Azure portal, user cannot achieve correlation between resubmitted run and parent run but in real-world business scenario, user would need to achieve correlation when there are multiple run records which are interconnected. To overcome this challenge, Serverless360 offers the capability to achieve a correlation between the Resubmitted run and the Parent run. The Resubmitted run will be added with Tag with ID of the parent run and user can investigate the status of the Parent run by clicking on that tag. It is also possible to add the Ignore tag with custom message to inform other users dealing with the Logic App not to do any operations.

Service Map

Consider a Sales person would need to visualize the orchestration and want to know the status of every entity involved in the orchestration. Here comes the Service Map in Serverless360 with which user can create the visual representation of the orchestration and get to view the status of the entities involved in the orchestration as well. The Status of the entities is completely based on the Monitoring configuration done in the monitoring section.

Next Steps:

• Search Logic Apps Run and Trigger History
• Azure Logic Apps better managed and monitored using Serverless360

Handle failed Logic App runs efficiently

With Serverless360, user can automate the complex and bulk processes for several entities.

Resubmit Failed runs

Consider the Fly wheel Cab booking Application’s cab booking scenario, where there were multiple failed bookings which need to be resubmitted to complete the transaction. With the Activity in Serverless360, user can automate the resubmission of Failed Logic App runs along with Scheduling. It is possible to set the required context by adding tag. Activity can also be configured to resubmit the failed runs of the selected trigger only.

Business Activity Monitoring

Track the business process and message flow in the orchestration using Business Activity Monitoring in Serverless360

Track Logic App with BAM

With BAM in Serverless360, user can perform end to end tracking of the message flow through the business orchestration built using Azure Logic Apps. With the BAM connectors in Logic App, user can track the message flowing through the stages and transactions of the Logic Apps in a Graphical view. With Exception logging, user can log the custom exception in the business process.

It is also possible to reprocess the failed business process to restore the business. Below is the good read on Business Activity Monitoring in Serverless360.

Next Steps:

• End to End Tracking with Business Activity Monitoring

Monitor Azure Logic App in Serverless360

There would be a scenario like user would need to get the status report of the entities for every 2 hours or at the end of the day.

The Product manager of the Cab Booking Application needs to know if the status of the Logic App is Enabled as expected. The need here is a consolidated report on status of all the entities those participate in the application at specified interval say every 4 hours once.

To meet this requirement, user can choose the Status monitor in Serverless360. It monitors the entities based on their status and basic properties. The consolidated report can be received on the configured Notification Channels that Serverless360 supports.

Monitor metrics

The monitoring need of an Architect of the Fly Wheel Cab Booking Application would be different. The Logic App should be monitored in multiple perspectives which would include performance, availability, consumption and so on.

Data monitor in Serverless360 is the best solution that can facilitate monitoring Logic Apps in various perspectives on the extensive set of available Azure metrics. The metrics can be selected based on the monitoring need. Please refer the section 3.3 for quick reference on the available metrics and the monitoring perspective that can be achieved choosing the metric.

Violation Monitor

Consider a DevOps engineer of the Fly Wheel Cab Booking Application, would need to monitor the entities for any violation and needs to be informed as and when the violation occur, here comes the Threshold monitor in Serverless360 to meet this need. In a Threshold monitor, user will configure the violation persist time and number of alerts per violation while creating the monitor. If the violation exists beyond the violation persist time, user will get the alerts.

How to automatically monitor your Azure Logic App?

In Serverless360 user can configure the auto correct option on the entity status. Users can configure the expected state and number of retries. Based on the configuration, monitor will automatically correct the state of the entity. Serverless360 also automatically generates system monitoring profiles once you sign up for the tool. This feature reduces the time taken to configure the monitoring profiles and is automatically available in the tool to monitor your resources.

Detect Failures

Consider a developer would need to detect any failure in Logic App trigger or run. With Watch monitor in Serverless360, developers can detect the failure in trigger or run history of Logic Apps within the specified duration which can be as low as 5 mins. With Watch Monitor, User can also monitor based on Error reason of a Logic App run.

Next Steps:

• Azure Logic Apps Monitoring using Serverless360
• Azure Logic Apps better managed and monitored using Serverless360
• Monitoring Azure Logic Apps using Serverless360 by Steef-Jan Wiggers