Function execution model in Azure - Commands & Configuration

Introduction

Azure Functions lets you run small pieces of code without managing servers. It automatically runs your code when triggered by events, so you only pay for the time your code runs.

When you want to run code in response to a file upload without setting up a server.

When you need to process messages from a queue automatically and quickly.

When you want to run scheduled tasks like cleaning up data every night.

When you want to build a simple API without managing web servers.

When you want to scale your code automatically based on demand.

Config File - function.json

function.json

{
  "bindings": [
    {
      "name": "req",
      "type": "httpTrigger",
      "direction": "in",
      "authLevel": "function",
      "methods": ["get", "post"]
    },
    {
      "name": "$return",
      "type": "http",
      "direction": "out"
    }
  ]
}

This file defines how the Azure Function is triggered and what it returns.

• bindings: Lists triggers and outputs.
• httpTrigger: Runs the function when an HTTP GET or POST request arrives.
• authLevel: Controls access; 'function' means a key is needed.
• http: Defines the HTTP response output.

Commands

Create a new Azure Functions app project using the .NET runtime.

Terminal

func init MyFunctionApp --worker-runtime dotnet

Expected Output

Creating new Azure Functions project in C:\Users\user\MyFunctionApp

Project created successfully.

→

--worker-runtime - Specifies the language runtime for the function app.

Add a new function named HttpExample triggered by HTTP requests with function-level authorization.

Terminal

func new --name HttpExample --template "HTTP trigger" --authlevel function

Expected Output

Created new function 'HttpExample' with HTTP trigger.

→

--template - Specifies the function trigger template.

→

--authlevel - Sets the authorization level for the function.

Run the function app locally to test the function execution model.

Terminal

func start

Expected Output

Functions:

HttpExample: [GET,POST] http://localhost:7071/api/HttpExample

Host started (http://localhost:7071, Ctrl+C to shut down)

Send an HTTP GET request to trigger the function and see the response.

Terminal

curl http://localhost:7071/api/HttpExample?name=Azure

Expected Output

"Hello, Azure"

Key Concept

Azure Functions run your code only when triggered by events, automatically managing resources and scaling for you.

Common Mistakes

Not setting the correct authLevel in function.json

The function may reject requests or be open to unauthorized access.

Set authLevel to 'function' or 'anonymous' based on your security needs.

Running func start without creating a function first

The app will start but no functions will be available to trigger.

Create at least one function using 'func new' before starting.

Using the wrong HTTP method when triggering the function

The function will not respond if the method is not allowed.

Use only the HTTP methods defined in the function.json bindings.

Summary

Initialize an Azure Functions app with 'func init' specifying the runtime.

Create a function with an HTTP trigger using 'func new' and set authorization.

Run the function app locally with 'func start' to test the execution.

Trigger the function by sending an HTTP request and receive the response.

Practice

1. What triggers an Azure Function to run?

easy

A. Only when the server restarts

B. Manual start from the Azure portal only

C. An event like an HTTP request or timer

D. When a user logs into Azure

Solution

1. Step 1: Understand Azure Function triggers

   Azure Functions start running when an event happens, such as an HTTP request or a timer firing.

2. Step 2: Eliminate incorrect options

   Manual starts, server restarts, or user logins are not triggers for Azure Functions.

3. Final Answer:

   An event like an HTTP request or timer -> Option C

4. Quick Check:

   Azure Functions run on events = A [OK]

Hint: Azure Functions run on events, not manual or login actions [OK]

Common Mistakes:

• Thinking functions run only manually
• Confusing server restart with trigger
• Assuming user login triggers function

2. Which of the following is the correct way to define an HTTP trigger in an Azure Function's function.json file?

easy

A. {"bindings": [{"type": "httpTrigger", "direction": "in", "authLevel": "function", "methods": ["get"]}]}

B. {"bindings": [{"type": "timerTrigger", "direction": "out", "schedule": "0 */5 * * * *"}]}

C. {"bindings": [{"type": "queueTrigger", "direction": "in", "queueName": "myqueue"}]}

D. {"bindings": [{"type": "httpOutput", "direction": "in"}]}

Solution

1. Step 1: Identify HTTP trigger binding

   The correct binding for an HTTP trigger uses "type": "httpTrigger", "direction": "in", and specifies methods like "get".

2. Step 2: Check other options for errors

   {"bindings": [{"type": "timerTrigger", "direction": "out", "schedule": "0 */5 * * * *"}]} is a timer trigger with wrong direction; {"bindings": [{"type": "queueTrigger", "direction": "in", "queueName": "myqueue"}]} is a queue trigger; {"bindings": [{"type": "httpOutput", "direction": "in"}]} uses invalid "httpOutput" type with wrong direction.

3. Final Answer:

   {"bindings": [{"type": "httpTrigger", "direction": "in", "authLevel": "function", "methods": ["get"]}]} -> Option A

4. Quick Check:

   HTTP trigger binding = {"bindings": [{"type": "httpTrigger", "direction": "in", "authLevel": "function", "methods": ["get"]}]} [OK]

Hint: HTTP triggers use "httpTrigger" type with "in" direction [OK]

Common Mistakes:

• Confusing trigger type with output binding
• Using wrong direction for triggers
• Mixing trigger types like queue or timer

3. Given this Azure Function code snippet triggered by an HTTP request:

module.exports = async function (context, req) {
  context.log('Function started');
  context.res = { status: 200, body: 'Hello ' + (req.query.name || 'world') };
};

What will be the HTTP response body if the request URL is http://example.com/api?name=Alice?

medium

A. "Error: name not found"

B. "Hello world"

C. "Hello undefined"

D. "Hello Alice"

Solution

1. Step 1: Understand the code logic

   The function reads the query parameter "name" from the request. If "name" exists, it uses it; otherwise, it defaults to "world".

2. Step 2: Apply the input URL query

   The URL has "name=Alice", so the response body becomes "Hello Alice".

3. Final Answer:

   "Hello Alice" -> Option D

4. Quick Check:

   Query name present = "Hello Alice" [OK]

Hint: Check if query parameter exists; else use default [OK]

Common Mistakes:

• Ignoring query parameters
• Assuming default always used
• Confusing response body with error

4. You deployed an Azure Function with a timer trigger but it never runs. Which of these is the most likely cause?

medium

A. The schedule expression in function.json is incorrect

B. The function app is running and healthy

C. The HTTP trigger is missing

D. The function code has no output binding

Solution

1. Step 1: Check timer trigger configuration

   If the schedule expression is wrong, the timer never fires, so the function won't run.

2. Step 2: Evaluate other options

   Running app is good; missing HTTP trigger is irrelevant for timer trigger; output binding absence doesn't stop trigger execution.

3. Final Answer:

   The schedule expression in function.json is incorrect -> Option A

4. Quick Check:

   Wrong schedule = no timer run [OK]

Hint: Check cron schedule syntax for timer triggers [OK]

Common Mistakes:

• Confusing trigger types
• Ignoring schedule format errors
• Assuming output binding affects trigger

5. You want an Azure Function to process messages from a queue and then send results to a database without writing extra code to connect to the database. Which feature should you use to achieve this?

hard

A. Manually call the database API from the function

B. Use an output binding for the database

C. Use a timer trigger to poll the database

D. Write database connection code inside the function

Solution

1. Step 1: Understand output bindings

   Output bindings let functions send data to other services automatically without extra connection code.

2. Step 2: Match requirement to feature

   Using an output binding for the database allows sending results directly after processing queue messages.

3. Final Answer:

   Use an output binding for the database -> Option B

4. Quick Check:

   Output binding sends data without extra code [OK]

Hint: Output bindings connect services without manual code [OK]

Common Mistakes:

• Thinking manual code is always needed
• Confusing triggers with output bindings
• Using timer triggers incorrectly