Specify jobs in your pipeline

TFS 2017


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.

You can organize your release pipeline into jobs. Every release pipeline has at least one job. Jobs are not supported in a build pipeline in this version of TFS.


You must install Update 2 to use jobs in a release pipeline in TFS 2017. Jobs in build pipelines are available in Azure Pipelines, TFS 2018.2, and newer versions.

Define a single job

YAML is not supported in TFS 2017.

Types of jobs

Jobs can be of different types, depending on where they run.

  • Agent pool jobs run on an agent in the agent pool. These jobs are only available release pipelines.

Agent pool jobs

These are the most common type of jobs and they run on an agent in an agent pool.

  • When using Microsoft-hosted agents, each job in a pipeline gets a fresh agent.
  • Use demands with self-hosted agents to specify what capabilities an agent must have to run your job. You may get the same agent for consecutive jobs, depending on whether there is more than one agent in your agent pool that matches your pipeline's demands. If there is only one agent in your pool that matches the pipeline's demands, the pipeline will wait until this agent is available.


Demands and capabilities are designed for use with self-hosted agents so that jobs can be matched with an agent that meets the requirements of the job. When using Microsoft-hosted agents, you select an image for the agent that matches the requirements of the job, so although it is possible to add capabilities to a Microsoft-hosted agent, you don't need to use capabilities with Microsoft-hosted agents.

YAML is not supported in TFS 2017.

Learn more about agent capabilities.

Server jobs

Tasks in a server job are orchestrated by and executed on the server (Azure Pipelines or TFS). A server job does not require an agent or any target computers. Only a few tasks are supported in a server job at present.

Agentless jobs supported tasks

Currently, only the following tasks are supported out of the box for agentless jobs:

Because tasks are extensible, you can add more agentless tasks by using extensions. The default timeout for agentless jobs is 60 minutes.

YAML is not supported in TFS 2017.


When you define multiple jobs in a single stage, you can specify dependencies between them. Pipelines must contain at least one job with no dependencies.


Each agent can run only one job at a time. To run multiple jobs in parallel you must configure multiple agents. You also need sufficient parallel jobs.

YAML builds are not yet available on TFS.


You can specify the conditions under which each job runs. By default, a job runs if it does not depend on any other job, or if all of the jobs that it depends on have completed and succeeded. You can customize this behavior by forcing a job to run even if a previous job fails or by specifying a custom condition.

YAML builds are not available on TFS 2017.


To avoid taking up resources when your job is unresponsive or waiting too long, it's a good idea to set a limit on how long your job is allowed to run. Use the job timeout setting to specify the limit in minutes for running the job. Setting the value to zero means that the job can run:

  • Forever on self-hosted agents
  • For 360 minutes (6 hours) on Microsoft-hosted agents with a public project and public repository
  • For 60 minutes on Microsoft-hosted agents with a private project or private repository (unless additional capacity is paid for)

The timeout period begins when the job starts running. It does not include the time the job is queued or is waiting for an agent.

YAML builds are not available on TFS 2017.

You can also set the timeout for each task individually - see task control options.

Multi-job configuration

From a single job you author, you can run multiple jobs on multiple agents in parallel. Some examples include:

  • Multi-configuration builds: You can build multiple configurations in parallel. For example, you could build a Visual C++ app for both debug and release configurations on both x86 and x64 platforms. To learn more, see Visual Studio Build - multiple configurations for multiple platforms.

  • Multi-configuration deployments: You can run multiple deployments in parallel, for example, to different geographic regions.

  • Multi-configuration testing: You can run test multiple configurations in parallel.

  • Multi-configuration will always generate at least one job, even if a multi-configuration variable is empty.

YAML builds are not available on TFS 2017.


An agent job can be used to run a suite of tests in parallel. For example, you can run a large suite of 1000 tests on a single agent. Or, you can use two agents and run 500 tests on each one in parallel.

To leverage slicing, the tasks in the job should be smart enough to understand the slice they belong to.

The Visual Studio Test task is one such task that supports test slicing. If you have installed multiple agents, you can specify how the Visual Studio Test task will run in parallel on these agents.

YAML builds are not available on TFS 2017.

Job variables

If you are using YAML, variables can be specified on the job. The variables can be passed to task inputs using the macro syntax $(variableName), or accessed within a script using the stage variable.

YAML builds are not available on TFS 2017.

For information about using a condition, see Specify conditions.


When you run an agent pool job, it creates a workspace on the agent. The workspace is a directory in which it downloads the source, runs steps, and produces outputs. The workspace directory can be referenced in your job using Agent.BuildDirectory variable. Under this, various subdirectories are created:

  • Build.SourcesDirectory is where tasks download the application's source code.
  • Build.ArtifactStagingDirectory is where tasks download artifacts needed for the pipeline or upload artifacts before they are published.
  • Build.BinariesDirectory is where tasks write their outputs.
  • Common.TestResultsDirectory is where tasks upload their test results.

YAML builds are not available on TFS 2017.

Artifact download

This example YAML file publishes the artifact WebSite and then downloads the artifact to $(Pipeline.Workspace). The Deploy job only runs if the Build job is successful.

YAML builds are not available on TFS 2017.

For information about using dependsOn and condition, see Specify conditions.

Access to OAuth token

You can allow scripts running in a job to access the current Azure Pipelines or TFS OAuth security token. The token can be use to authenticate to the Azure Pipelines REST API.

YAML builds are not available on TFS 2017.

What's next