Invoke a function task

Azure DevOps Services | Azure DevOps Server 2020 | Azure DevOps Server 2019 | TFS 2018

Use this task in an agentless job of a release pipeline to invoke an HTTP triggered function in a function app that is created and hosted in Azure Functions and parse the response.


In Microsoft Team Foundation Server (TFS) 2018 and previous versions, build and release pipelines are called definitions, runs are called builds, service connections are called service endpoints, stages are called environments, and jobs are called phases.


Can be used in only an agentless job of a release pipeline.

YAML snippet

# Invoke Azure Function
# Invoke an Azure Function
- task: AzureFunction@1
    function:             # Required. Azure function URL. Example: ""
    key:                  # Required. Function key
    #method:              # Required. 'POST' # Options: OPTIONS, GET, HEAD, POST, PUT, DELETE, TRACE, PATCH
    #headers:             # Optional. Default value: '{Content-Type:application/json, PlanUrl: $(system.CollectionUri), ProjectId: $(system.TeamProjectId), HubName: $(system.HostType), PlanId: $(system.PlanId), JobId: $(system.JobId), TimelineId: $(system.TimelineId), TaskInstanceId: $(system.TaskInstanceId), AuthToken: $(system.AccessToken)}' 
    #queryParameters:     # Optional
    #body:                # Required when method != GET && Method != HEAD
    #waitForCompletion:   # Required Default value:'ApiResponse' # Options: Callback, ApiResponse
    #successCriteria:     # Optional


Parameter Comments
Azure function URL
Required. The URL of the function to be invoked. Example: ""
Function key
Required. Function or host key to access and invoke the function. Using a secret pipeline variable to store the function key is recommended.
Required. The HTTP method with which the function will be invoked.
Optional. The header in JSON format to be attached to the request sent to the function.
Query parameters
Optional. the string query to append to the function URL. Must not start with "?" or "&".
Optional. The request body in JSON format.
Completion event
Required. How the task reports completion. Options: "ApiResponse", "Callback."
API response (default) - the function returns success and success criteria evaluates to true.
Callback - the function makes a callback to update the timeline record.
Success criteria
Optional. The criteria to consider the task successful. By default, the task returns 200 OK status when successful.

Open source

This task is open source on GitHub. Feedback and contributions are welcome.


Where should a task signal completion when Callback is chosen as the completion event?

To signal completion, the function should POST completion data to the following pipelines REST endpoint.


**Request Body**
{ "name": "TaskCompleted", "taskId": "taskInstanceId", "jobId": "jobId", "result": "succeeded" }

See this simple cmdline application for specifics. In addition, a C# helper library is available to enable live logging and managing task status for agentless tasks. Learn more

Why does the task failed within 1 minute when the timeout is longer?

If the function executes for more than 1 minute, use the Callback completion event. The API Response completion option is supported for requests that complete within 60 seconds.

Is there an example of an Azure Function that uses the callback completion mode?

#r "Newtonsoft.Json"

using System;
using System.Net;
using System.Net.Http;
using System.Net.Http.Headers;
using System.Text;
using System.Threading;
using Microsoft.AspNetCore.Mvc;
using Microsoft.Extensions.Primitives;
using Newtonsoft.Json;

public static async Task<IActionResult> Run(HttpRequest req, ILogger log)
    var url = req.Headers["PlanUrl"];
    var projectId = req.Headers["ProjectId"];
    var hubName = req.Headers["HubName"];
    var planId = req.Headers["PlanId"];
    var jobId = req.Headers["JobId"];
    var timelineId = req.Headers["TimelineId"];
    var taskInstanceId = req.Headers["TaskinstanceId"];
    var authToken = req.Headers["AuthToken"];

    var callbackUrl = $"{url}/{projectId}/_apis/distributedtask/hubs/{hubName}/plans/{planId}/events?api-version=2.0-preview.1";
    var successBody = JsonConvert.SerializeObject(new {
        name = "TaskCompleted",
        taskId = taskInstanceId.ToString(),
        jobId = jobId.ToString(),
        result = "succeeded"

    // the following call does not block
    Task.Run(() =>
        Thread.Sleep(70000); // simulate long running work
        PostEvent(callbackUrl, successBody, authToken, log);
    return new OkObjectResult("Long-running job succesfully scheduled!");
public static void PostEvent(String callbackUrl, String body, String authToken, ILogger log)
        var client = new HttpClient();
        client.DefaultRequestHeaders.Authorization = new AuthenticationHeaderValue("Bearer", authToken);
        var requestContent = new StringContent(body, Encoding.UTF8, "application/json");
        var response = client.PostAsync(new Uri(callbackUrl), requestContent).Result;
        var responseContent = response.Content.ReadAsStringAsync().Result;
    catch (Exception ex)