How to throw an exception in Logic Apps - made simple

One of the features that are missing in Logic Apps (there is an idea posted since 2017) is the ability to throw an exception. What is currently available is the ability to Termine the flow, but that won't work if you are executing a parallel branch as it will finish it too. In order to workaround this, I have thought a very simple way to intentionally produce an exception. For this you can pick whatever connector and call it in a wrongly manner. For example: SQL connector executing an store procedure that doesn't exist. The above example I have used in the past and work very well, however, it makes you configure an connection to an existing SQL DB. Hence, to bringe this approach to a general escenario I have test it with an HTTP request node, which doesn't require connections at all. In order to our approach to work we need to meet two properties: 1. Whatever we use the connector the exception needs to be thrown in the 100% of the times. 2. We cannot introduce a delay on the flow. Fow example: An exception thrown because of a timeout is not desireble. To meet the above, I have configured the HTTP request node using the URI http://127.0.0.1 which is the loopback and with 100% certainty that socket will be closed - if you are more paranoic just randomize the URI-. Also the connector has the restriction to not use that address.
To really test this scenario, let's enclose it on a scope:
After running the flow we see that the scope that manages the exception is triggered.

Comments

Post a Comment

Popular posts from this blog

Telemetry and Monitory with Applications Insights and Dynamics 365 F&O

Image
Telemetry and Monitory with Applications Insights and Dynamics 365 F&O With version 10.0.39 Applications Insights integrations is GA. Hence, here is a very simple example how to start using this great feature and start monitoring the performance of your application.  For more about D365FO and Application Insights integration:  Monitoring and telemetry using Application Insights - Finance & Operations | Dynamics 365 | Microsoft Learn Dynamics 365 Finance and Operations - Monitoring and Telemetry using Application Insights -TechTalk (youtube.com) In this example we will monitor the execution time of a batch job. In particular, we will use a metric to analyze the execution over time.  // Create a class file and paste this code class SlowBatchJobClass extends RunBaseBatch implements BatchRetryable { [SysEntryPointAttribute] public void run() { int maxWaitInSeconds = 20; // Maximum wait time in seconds           SysApplicationInsightsTelemetryLogger logger =
Read more
Image
Distributed Printing Architecture with Power Platform  Going back to posting after a long period. Now I have started a new blog where I'll be posting content related to my work with Microsoft technology, and how you can combine Dynamics 365 + Power Platform + Azure to solve virtually any problem in a reliable and robust architecture. In this first post I'll share how to build an architecture for unattended distributed printing from Power Apps. Although there are other options (look at PCF controls ) I found this architecture simple and reliable. This solution will serve to customers looking to: Use network shared printers across their organization. Serve multiple users allowing them to distribute their printing task according to specific business rules (ex. some group will only print on 'Printer 01') Rely on the SO to print as you can use the print command Access your local resources (printers) in a secure way To connect from the canvas app to the A
Read more

Creating a DMF data package from files stored in an Azure blob storage

Image
Dynamics 365 for Finance and Operations support different scenarios for data integration.  Each scenario is listed on the following documentation: https://docs.microsoft.com/en-us/dynamics365/fin-ops-core/dev-itpro/data-entities/integration-overview One of the most common scenarios is when a large number of records need to be imported in D365FO. These record normally are coming from a third-party application as batches. If we follow the documentation guidelines for this scenario, the recommended pattern to be applied is Batch Data API For this kind of pattern we have the following API available:  Recurring Integration API  Data Management API  Many people uses the Recurring Integration API, however, I found it lacks of features (like asking for the task status) and also I haven't seen any further investment to improve it. For these reasons I choose to use the Data Management API.   Data Management API is a robust API and their specification can be found here: Data management pac
Read more