Continuous Integration with TeamCity
What is Continuous Integration?
Continuous Integration is a software development practice in which developers commit code changes into a shared repository several times a day. Each commit is followed by an automated build to ensure that new changes integrate well into the existing code base and to detect problems early.
What is TeamCity?
JetBrains TeamCity is a user-friendly and easy to set up continuous integration (CI) server for developers and build engineers.
This video gives a general overview of the main TeamCity features:
What can you do with TeamCity?
Run parallel builds simultaneously on different platforms and environments
Optimize the code integration cycle and be sure you never get broken code in the repository
Review on-the-fly test results reporting with intelligent tests reordering
Run code coverage and duplicates finder for Java and .NET
Customize statistics on build duration, success rate, code quality, and custom metrics
and much more
To learn more about major TeamCity features, refer to the official JetBrains website.
TeamCity works great not only for admins and build engineers, but also for developers. Watch this video to see some of the signature features it provides:
The complete list of supported platforms and environments can be found here.
Basic TeamCity concepts
This section describes the main TeamCity concepts. You can find a more comprehensive explanation in the Introduction to TeamCity terminology.
The TeamCity build system comprises a server and build agents.
A piece of software that actually executes a build process. It is installed and configured separately from the TeamCity server. You get access to Cloud agents with your TeamCity Cloud subscription, but you can also host agents on a physical machine.
The server itself does not run either builds or tests: the server's job is to monitor all the connected build agents, distribute queued builds to the agents based on compatibility requirements, and report the results. All information on the build results (build history and all the build-associated data except for artifacts and build logs), VCS changes, agents, build queue, user accounts and user permissions, and so on, are stored in a database.
A TeamCity project corresponds to a software project or a specific version/release of a software project. A project is a collection of build configurations.
A combination of settings defining a build procedure. The settings include VCS roots, build steps, build triggers described below.
A collection of version control settings (paths to sources, username, password, checkout mode and other settings) that defines how TeamCity communicates with a version control (SCM) system to monitor changes and get sources for a build.
A task to be executed. Each build step is represented by a build runner providing integration with a specific build tool (like Ant, Gradle, MSBuild, and so on), a testing framework (for example, NUnit), or a code analysis engine. Thus, in a single build you can have several steps and sequentially invoke test tools, code coverage, and, for instance, compile your project.
Any modification of the source code which you introduce. If a change has been committed to the version control system, but not yet included in a build, it is considered pending for a certain build configuration.
Refers to both: the actual process of creating an application version and the version itself. After the build process is triggered, it is put into the build queue and is started when there are agents available to run it. After the build is finished, the build agent sends build artifacts to the server.
A list of builds that were triggered and are waiting to be started. TeamCity will distribute them to compatible build agents as soon as the agents become idle. A queued build is assigned to an agent at the moment when it is started on the agent; no preassignment is made while the build is waiting in the build queue.
Files produced by a build, for example, installers, WAR files, reports, log files, when they become available for download.
Basic CI Workflow in TeamCity
To understand the data flow between the server and the agents, what is passed to the agents, how and when TeamCity gets the results, let's take a look at a simple build lifecycle.
The TeamCity server detects a change in your VCS root (repository).
The server stores this change in the database.
The trigger, attached to the build configuration, detects the relevant change in the database and initiates the build.
The triggered build gets to the build queue.
The build is assigned to a free and compatible build agent.
The agent executes build steps, described in the build configuration. While executing the steps, the agent reports the build progress to the TeamCity server. It is sending all the log messages, test reports, code coverage results on the fly, so you can monitor the build process in real time.
After finishing the build, the agent sends build artifacts to the server.