Connecting a Canvas App to an Azure Function via Custom Connector

PL-400 By islever Editorial Last updated

Create custom connectors

Short answer

BE Create an OpenAPI API definition for the Azure Function and then create a custom connector from that API so the canvas app can call it.

Question

You are creating a canvas app to retrieve user sign in information from Microsoft Azure Active Directory (Azure AD) when someone searches for information about an end user.
You create an Azure Function to retrieve the required information by using JSON.
You need to ensure that the application functions correctly.
Which two actions should you perform? Each correct answer presents part of the solution.
NOTE: Each correct selection is worth one point.
  1. Create a Power Automate flow to import data.
  2. Create a custom connector by using the Azure Function API. Correct Answer
  3. Use app designer in the Power Platform admin center.
  4. Use Azure Service Bus.
  5. Create an API definition for the Azure Function. Correct Answer

Key Takeaways

The test is whether you know the two-step custom-connector workflow — publish an API definition for the Azure Function, then register a custom connector from that definition — with the trap being to reach for Power Automate or Azure Service Bus instead of a direct connector.

When a canvas app must call an Azure Function that returns Azure AD sign-in data as JSON, the function's HTTP endpoint has to be described with an API definition (OpenAPI/Swagger) and then wrapped in a Power Platform custom connector. This question establishes that the correct pair of actions is creating the API definition (E) and then creating a custom connector from it (B).

Common mistake: The most common wrong pick is option A, creating a Power Automate flow to import the data; a flow can proxy a call but it is not required and adds an unnecessary layer, while options C and D (app designer in the admin center, Azure Service Bus) address neither authentication nor schema for the app-to-function call.

Expert Analysis

Why the Answer Is Correct

Option E is required first: the Azure Function exposes an HTTP endpoint, and Power Platform needs a machine-readable API definition (OpenAPI/Swagger, often auto-generated from the function's function.json and OpenAPI extension) that describes the operations, parameters, and JSON response schema. Option B then consumes that definition — the custom connector registers the Azure Function API as a callable connector with policy templates and (optionally) Azure AD or API-key authentication, so the canvas app can call it directly with ConnectorName.OperationName. Together, E and B form Microsoft's documented custom connector workflow: define the API, then create the connector from that definition.

Why the Other Options Are Wrong

Option A (a Power Automate flow to import data) is an indirect workaround: a cloud flow calling the function and returning its output can work, but it is not the mechanism that makes the canvas app "function correctly" and it is not needed for a live search on an end user. Option C (the app designer in the Power Platform admin center) is irrelevant — the admin center does not configure data connectivity for a canvas app; adding a connector happens in the app designer inside Power Apps Studio. Option D (Azure Service Bus) is an asynchronous messaging service for decoupled, queue-based integration and cannot let a canvas app synchronously query sign-in information for a search box. Note also that only a custom connector secures and standardizes the JSON contract between the app and the function.

Community Comment Notes

The comments converge on BE with strong agreement: as itmaxuser explained, "The custom connector enables the app to call your Azure Function securely and handle the communication between the canvas app and the Azure Function." A_A_C made the pragmatic observation that "Lets say, we all agree with E. So the only logical answer left that go with E, is B", and alevalli9 acknowledged that "Creating a Power Automate can be a possible solution but not the best practice I think," which is exactly why A is a distractor rather than part of the required pair.

Exam Strategy

When a question pairs an Azure API/hosting piece with a Power Platform piece, look for the two-step "describe the API, then register the connector" pattern — an OpenAPI/Swagger definition feeding a custom connector. Eliminate options that add an unnecessary middle tier (flows, Service Bus) for what is meant to be a direct app-to-function call.

Frequently Asked Questions

Why is creating an API definition for the Azure Function necessary before the custom connector?

The API definition (OpenAPI/Swagger) documents the function's operations, parameters, and JSON response schema. The Power Platform import wizard uses it to generate the connector's actions and parameter bindings automatically.

Why not just use a Power Automate flow to call the Azure Function from the canvas app?

A flow can act as a proxy, but it adds latency, licensing, and maintenance overhead for what is a synchronous lookup. A custom connector calling the function directly is the intended, best-practice pattern.

More PL-400 FAQ →

Discussion 10 contributions

Compiled from public community threads and reader submissions, lightly edited for clarity. Handles and vote tallies are shown as recorded, and comments reflect the views of individual contributors.

From public community threads 10

  1. DimpleG 👍 35
    A. Create a Power Automate flow to import data is not necessary for the application to function correctly, as the application is already retrieving the user sign-in information from Azure AD by using JSON through an Azure Function. C. Using app designer in the Power Platform admin center would not be necessary for this task as it is not used to retrieve user sign-in information from Azure AD. D. Use Azure Service Bus is not necessary for this task as it is not used to retrieve user sign-in information from Azure AD. To ensure that the application functions correctly, you should create a custom connector by using the Azure Function API. This will allow you to connect your canvas app to the Azure Function to retrieve the user sign-in information. Additionally, you need to create an API definition for the Azure Function, this will allow you to define the specific parameters and methods that your app will use to communicate with the Azure Function.
  2. Fyrus 👍 8
    DimpleG said everything.
  3. A_A_C 👍 4
    Lets say, we all agree with E. So the only logical answer left that go with E, is B So BE
  4. At09 👍 2
    Answer is BE
  5. jkaur 👍 2
    Answer:- BE
Show 5 more comments
  1. itmaxuser 👍 1
    1-B. Create a custom connector by using the Azure Function API The custom connector enables the app to call your Azure Function securely and handle the communication between the canvas app and the Azure Function. You’ll need to create a custom connector in Power Apps that connects to the Azure Function API you created, allowing the app to retrieve data from Azure AD through the Azure Function. 2-E. Create an API definition for the Azure Function To create a custom connector, you'll first need to define the API for your Azure Function, which includes specifying the endpoints, methods, and data formats. The API definition ensures that the app knows how to interact with the Azure Function, especially when using JSON for data retrieval.
  2. Tarnk 👍 1
    These two actions are crucial for integrating the Azure Function with the Power Apps canvas app and ensuring that it functions correctly by allowing the app to communicate with the function and properly handle the data returned.
  3. alevalli9 👍 1
    Creating a Power Automate can be a possible solution but not the best practice I think. So I'd go with BE
  4. SSS_S_S 👍 1
    BE is correct
  5. Fardji 👍 1
    seems good

Reader comments & corrections

No reader corrections yet — spotted an error or have a note? Share it below.

Log in to comment, report an error, or add a note about this question.

Diagnosing the 'remote name could not be resolved' 404 for the Northwind Web API connector

S-grade

Solution for Long-Running Transactional Dataverse Operations

S-grade

Why exporting a custom connector directly from the maker portal fails to deploy it

S-grade

Preventing plug-in timeouts when calling a slow external web service

S-grade

Practice All PL-400 Questions

Access 85 questions with complete answers and detailed explanations.

View Full PL-400 Practice Test →

← Back to PL-400 Study Guide