Icon

You are viewing the documentation of TeamCity 10.x and 2017.x, which is not the most recently released version of TeamCity.
View this page in the latest documentation or refer to the listing to choose the documentation corresponding to your TeamCity version.

 

Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

  • Regularly review reported Server Health reports (including hidden ones)
  • Use a separate reverse proxy server (e.g. NGINX) to handle HTTPS
  • Use a separate server for the external database and monitor the database performance
  • Monitor the server's CPU and IO performance, increase hardware resources as necessary (see also hardware notes)
  • Make sure clean-up is configured for all the projects with a due retention policy, make sure clean-up completely finishes regularly (check Administration / Clean-Up page)
  • Consider ensuring good IO performance for the <TeamCity Data Directory>/system/caches directory, e.g. by moving it to a separate local drive (or storing on a local drive you choose to store the TeamCity Data Directory on a network storage)
  • Regularly archive obsolete projects
  • Regularly review the installed not bundled plugins and remove those not essential for the server functioning
  • Consider using agent-side checkout whenever possible
  • Make sure the build logs are not huge (tens megabytes at most, better less than 10 Mb)
  • If lots VCS roots are configured on the server, consider configuring repository commit hooks instead of using polling for changes or at least increase VCS polling interval to 300 seconds or more)
  • If the server is often used by large number of users (e.g. more than 1000), consider reducing increasing UI refresh intervals
  • When regularly exceeding 500 concurrently running builds which log a lot of data, consider using Several Nodes Setup

...

Allocating 1 GB for the redo log (see the table below) and undo files is sufficient in most cases.

...

We've seen patterns of having an agent per each 20 build configurations (types of builds). Or a build agent per 1-2 developers.

See also notes on maximum supported number of agents.

...

TeamCity agents farm can be reused between the main and the failover servers. Agents will automatically connect to the new server if you make the failover server to be resolved via the old server DNS name and agents connect to the server using the DNS name. See also information on switching from one server to another.
If appropriate, the agents can be replicated just as the server. However, there is no need to replicate any TeamCity-specific data on the agents except for the conf\buildAgent.properties file as all the rest of the data can typically be renewed from the server. In case of replicated agents farm, the replica agents just need to be connected to the failover server.

...

Consider adding the "teamcity.installation.completed=true" line into the <TeamCity Home Directory>\conf\teamcity-startup.properties file - this will prevent the server from creating an administrator user if no such user is found

TeamCity has no built-in protection against DoS attack: high rate of requests can overload the server and make it not responsive. If your TeamCity instance is deployed in the environment which allows such service abuse, implement the protection on the reverse proxy level.

Version-specific settings

...

  • You are running the latest released TeamCity version and are ready to upgrade to the newly released versions within weeks
  • Access to the TeamCity web interface is secured using HTTPS (e.g. with the help a proxying server like NGINX). Best practices for securing web application are employed for the TeamCity web interface
  • The TeamCity server machine does not run agents (at least under the user permitted to read the TeamCity server's home directory and TeamCity Data Directory)
  • TeamCity server and agents processes are run under limited users with minimal required permissions. Installation directories are readable and writable only by a limited set of OS users. The conf\buildAgent.properties file and server logs as well as the Data Directory are only readable by OS users who represent administrators of the services, because reading those locations may allow taking over the agent or server respectively.
  • Guest user and user registration is disabled or roles are reviewed for guest and the All Users group
  • TeamCity users with administrative permissions have non-trivial passwords
  • If you have external authentication configured (such as LDAP), the built-in authentication module is disabled
  • Passwords are not printed into the build log, not stored in build artifacts, nor are they stored in non-password parameters

Anchor
security-rolesRelated
security-rolesRelated
Security-related Risks Evaluation

Here are some notes on different security-related aspects:

  • man-in-the middle concerns
    • between the TeamCity server and the user's web browser: It is advised to use HTTPS for the TeamCity server. During login, TeamCity transmits the user login password in an encrypted form with a moderate encryption level.
    • between a TeamCity agent and the TeamCity server: see the section.
    • between the TeamCity server and other external servers (version control, issue tracker, etc.): the general rules apply as for a client (the TeamCity server in the case) connecting to the external server, see the guidelines for the server in question.
  • users that have access to the TeamCity web UI: the specific information accessible to the user is defined via TeamCity user roles.
  • users who can change the code that is used in the builds run by TeamCity (including committers in any branches/pull requests if they are built on TeamCity):
    • can do everything what can do the system user under whom the TeamCity agent is running, have access to OS resources and other applications installed on the agent machines where their builds can run.
    • can access and change source code of other projects built on the same agent, modify the TeamCity agent code, publish any files as artifacts for the builds run on the agent (which means the files can be then displayed in the TeamCity web UI and expose web vulnerabilities or can be used in other builds), etc.
    • can impersonate a TeamCity agent (run a new agent looking the same to the TeamCity server).
    • can do everything that users with the "View build configuration settings" permission for all the projects on the server can do (see below).
    • can retrieve settings of the build configurations where the builds are run, including the values of the password fields.
    • can download artifacts from any build on the server.
      Hense, it is advised to run TeamCity agents under users with only necessary set of permissions and use the agent pools feature to ensure that projects requiring a different set of access are not built on the same agents.
  • users with the "View build configuration settings" permission (the "Project developer" TeamCity role by default) can view all the projects on the server, but since TeamCity 9.0 there is a way to restrict this, see details in the corresponding issue TW-24904.

  • users with the "Edit project" permission (the "Project Administrator" TeamCity role by default) in one project, by changing settings can retrieve artifacts and trigger builds from any build configuration they have only the view permission for (TW-39209). The users might also be able to make the TeamCity server run any executable located on the server.
  • users with the "Change server settings" permission (the "System Administrator" TeamCity role by default): It is assumed that the users also have access to the computer on which the TeamCity server is running under the user account used to run the server process. Thus, the users can get full access to the machine under that OS user account: browse file system, change files, run arbitrary commands, etc.
  • TeamCity server computer administrators: have full access to TeamCity stored data and can affect TeamCity executed processes. Passwords that are necessary to authenticate in external systems (like VCS, issue trackers, etc.) are stored in a scrambled form in TeamCity Data Directory and can also be stored in the database. However, the values are only scrambled, which means they can be retrieved by the users who have access to the server file system or database.
  • Users who have read access for TeamCity server logs (TeamCity server home directory) can escalate their access to TeamCity server administrator
  • Users who have read access to TeamCity Data Directory can access all the settings on the server, including configured passwords
  • Users who have read access to the build artifacts in TeamCity Data Directory (<TeamCity Data Directory>\system\artifacts) get the same permissions as users with the "View build runtime parameters and data" permission (in particular, with access to the values of all the password parameters used in the build)
  • TeamCity agent computer administrators: same as "users who can change code that is used in the builds run by TeamCity".
  • When storing settings in VCS is enabled:
    • any user who can access the settings repository (including users with "View file content" permission for the build configurations using the same VCS root) can see the settings and retrieve the actual passwords based on their stored scrambled form
    • any user who can modify settings in VCS for a single build configuration built on the server, via changing settings can retrieve artifacts and trigger builds from any build configuration they have only view permission for (TW-39192).
    • users who can customize build configuration settings on a per-build basis (e.g. one who can run personal builds when versioned settings are set to "use settings from VCS") via changing settings in a build can retrieve artifacts and trigger builds from any build configuration they have only view permission for (TW-46065).
  • Other:
    • TeamCity web application vulnerabilities: the TeamCity development team makes a reasonable effort to fix any significant vulnerabilities (like cross-site scripting possibilities) once they are uncovered. Please note that any user that can affect build files ("users who can change code that is used in the builds run by TeamCity" or "TeamCity agent computer administrators") can make a malicious file available as a build artifact that will then exploit cross-site scripting vulnerability. (TW-27206)
    • TeamCity agent is fully controlled by the TeamCity server: since TeamCity agents support automatic updates download from the server, agents should only connect to a trusted server. An administrator of the server computer can force execution of arbitrary code on a connected agent.
    • Binaries of the agent plugins installed on the server are available to anyone who can access the server URL

...

You need to set up a reverse proxy (see Proxy Server Setup below) and also configure TeamCity's bundled Tomcat server (see TeamCity Tomcat Configuration below) to make sure TeamCity "knows" the actual absolute URL used by the client to access the resources. These URLs are then used to generate absolute URLs in client redirects and other responses.

...

When using Apache, make sure to use the Dedicated "Connector" node approach for configuring TeamCity server.

...

For the NGINX configuration below, use the "RemoteIpValve" Approach for configuring TeamCity server.

...

To point TeamCity to your proxy server:
Since TeamCity 2017.1.5 the following server internal properties are available (see the section below for previous version):

Code Block
languagetext
# For HTTP protocol
## The domain name or the IP address of the proxy host and the port: 
teamcity.http.proxyHost=proxy.domain.com
teamcity.http.proxyPort=8080
 
## The hosts that should be accessed without going through the proxy, usually internal hosts. Provide a list of hosts, separated by the '|' character. The wildcard '*' can be used:
teamcity.http.nonProxyHosts=localhost|*.mydomain.com)
 
## For an authenticated proxy add the following properties:
### Authentication type. "basic" and "ntlm" values are supported. The default is basic.
teamcity.http.proxyAuthenticationType=basic
### Login and Password for the proxy:
teamcity.http.proxyLogin=username
teamcity.http.proxyPassword=password
 
# For HTTPS protocol
## The domain name or the IP address of the proxy host and the port: 
teamcity.https.proxyHost=proxy.domain.com
teamcity.https.proxyPort=8080

## The hosts that should be accessed without going through the proxy, usually internal hosts. Provide a list of hosts, separated by the '|' character. The wildcard '*' can be used:
teamcity.https.nonProxyHosts=localhost|*.mydomain.com

## For an authenticated proxy add the following properties:
### Authentication type. "basic" and "ntlm" values are supported. The default is basic.
teamcity.https.proxyAuthenticationType=basic
### Login and Password for the proxy:
teamcity.https.proxyLogin=login
teamcity.https.proxyPassword=password

...

It's advised to try a new TeamCity version before upgrading your production server. The usual procedure is to create a copy of your production TeamCity installation, then upgrade it, try the things out, and, when everything is checked, drop the test server and upgrade the main one.
When you start the test server, remember to change the Server URL, disable Email and Jabber notifiers as well as other features on the new server.

Anchor
copy_server
copy_server

...

In case you want to preserve the original server as well as the copy, make sure to check the lincensing considerations.

Create a Server Copy

You can create a copy of the server using TeamCity backup functionality or manually.

...

  1. Create a backup.
  2. Install a new TeamCity server of the same version that you are already runnning. Ensure that:
    - the appropriate environment is configured (see the notes below) 
    - the server uses its own TeamCity Data Directory and its own database (check config/database.properties in the Data Directory)
  3. Restore the backup.
  4. Perform the necessary environment transfer.
Note

The new server won't get build artifacts and some other data. If you need them, you will need to copy appropriate directories (e.g. the entire "artifacts" directory) from .BuildServer/system from the original to the copied server. Make sure to finish copying over the artifacts before starting the new server.

...

  1. Create a backup so that you can restore it if anything goes wrong,
  2. Ensure the server is not running,
  3. Either perform clean installation or copy the TeamCity binaries (TeamCity Home Directory) into a new place (the temp and work subdirectories can be omitted during copying). (warning) Use exactly the same TeamCity version. If you plan to upgrade after copying, perform the upgrade only after you have the existing version up and running.
  4. Copy TeamCity Data Directory. If you do not need the full copy, refer to the items below for options.
    • .BuildServer/config to preserve projects and build configurations settings
    • .BuildServer/lib and .BuildServer/plugins if you have them
    • files from the root of .BuildServer/system if you use internal database and you do not want to perform database move.
    • .BuildServer/system/artifacts (optional) if you want build artifacts and build logs (including tests failure details) preserved on the new server
    • .BuildServer/system/changes (optional) if you want personal changes preserved on the new server
    • .BuildServer/system/pluginData (optional) if you want to preserve state of various plugins, build triggers and settings audit diff
    • .BuildServer/system/caches and .BuildServer/system/caches (optional) are not necessary to copy to the new server, they will be recreated on startup, but can take some time to be rebuilt (expect some slow down).
  5. Artifacts directory is usually large and if you need to minimize the downtime of the server in case of the server move, you can use the generic approach for copying the data: use rsync, robocopy or alike tool to copy the data while the original server is running. Repeat the sync run several times until the amount of data synced reduces. Run the final sync after the original server shut down. Alternative solution for the server move is to make the old data artifacts directory accessible to the new server and configure it as second location of artifacts. Then copy the files over from this second location to the main one while the server is running, restart the server after copying completion.
  6. Create copy of the database that your TeamCity installation is using in new schema or new database server. This can be done with database-specific tools or with the bundled maintainDB tool by backing up database data and then restoring it.
  7. Configure new TeamCity installation to use proper TeamCity Data Directory and database (.BuildServer/config/database.properties points to a copy of the database)
  8. Perform the necessary environment transfer.
Info

If you want to do a quick check and do not want to preserve builds history on the new server you can skip step 6 (cloning database) and all items of the step 5 marked as optional.

...

  • ensure the new server is configured to use another data directory and another database than the original server; check also "Artifact directories" setting on server's Global Settings;
  • change server unique id by removing "uuid" attribute from XML of <TeamCity Data Directory>\config\main-config.xml file before the first start;
  • ensure the same license keys are not used on several servers (more on licensing);
  • update Server URL on Administration | Global Settings page to the actual URL of the server;
  • check that you can successfully authenticate on the new server, use super user access if necessary;
  • check that VCS servers, issue tracker servers, email and Jabber server and other server-accessed systems are accessible;
  • check that any systems configured to push events to TeamCity server (like VCS hooks, automated build triggering, monitors, etc.) are updated to know about the new server;
  • review the list of installed plugins to determine if their settings need changes;
  • install new agents (or select some from the existing ones) and configure them to connect to the new server (using the new server URL);
  • check that clients reading from the server (downloading artifact, using server's REST API, NuGet feed, etc.) are reconfigured, if necessary.

...

See also the section below on moving the server from one machine to another.

...

If you need to move data to a fresh server without existing data, it is recommended to move the server or copy it and then delete the data which is not necessary on the new server.

...

If you need to move the existing TeamCity installation to a new hardware or clean OS, it is recommended to follow the instructions on copying the server from one machine to another and then switch from the old server to a new one. If you plan to use the same database instance on the new server, just skip the items on copying the database. If you are sure you do not need to preserve the old server, you can perform move operations instead of copying in those instructions.

...

  1. Write a build script that will perform the deployment task for the binary files available on the disk. (e.g. use Ant or MSBuild for this. For typical deployment transports use Deployer runners). See also #Integrate with Build and Reporting Tools. You can use Meta-Runner to reuse a script with convenient UI.
  2. Create a build configuration in TeamCity that will execute the build script and perform the actual deployment. If the deployment is to be visible or startable only by the limited set of users, place the build configuration in a separate TeamCity project and make sure the users have appropriate permissions in the project.
  3. In this build configuration configure artifact dependency on a build configuration that produces binaries that need to be deployed.
  4. Configure one of the available triggers in the deploying build configuration if you need the deployment to be triggered automatically (e.g. to deploy last successful of last pinned build), or use "Promote" action in the build that produced the binaries to be deployed.
  5. Consider using snapshot dependencies in addition to artifact ones and check Build Chains tab to get the overview of the builds. In this case artifact dependency should use "Build from the same chain" option.
  6. If you need to parametrize the deployment (e.g. specify different target machines in different runs), pass parameters to the build script using custom build run dialog. Consider using Typed Parameters to make the custom run dialog easier to use or handle passwords.
  7. If the deploying build is triggered manually consider also adding commands in the build script to pin and tag the build being deployed (via sending a REST API request).
    You can also use a build number from the build that generated the artifact.

...

Data collection
The easiest way for a start is to modify your build scripts to make use of the selected tool and collect all the required data.
If you can run the tool from a command line console, then you can run it in TeamCity with a command line runner. This will give you detection of the messages printed into standard error output. The build can be marked as failed is the exit code is not zero or there is output to standard error via build failure condition.
If the tool has launchers for any of the supported build scripting engines like Ant, Maven or MSBuild, then you can use corresponding runner in TeamCity to start the tool.
See also #Use an External Tool that My Build Relies on for the recommendations on how to run an external tool.

...

If the tool reports code-attributing information like Inspections or Duplicates, TeamCity-bundled report can be used to display the results. A custom plugin will be necessary to process the tool-specific report into TeamCity-specific data model. Example of this can be found in XML Test Reporting plugin and FXCop plugin (see a link on Open-source Bundled Plugins).

See also #Import coverage results in TeamCity.

For advanced integration, a custom plugin will be necessary to store and present the data as required. See Developing TeamCity Plugins for more information on plugin development.

...

  • publish coverage HTML report as TeamCity artifact: most of the tools produce coverage report in HTML format, you can publish it as artifact and configure report tab to show it in TeamCity. If artifact is published in the root artifact directory and its name is coverage.zip and there is index.html file in it, report tab will be shown automatically. As to running an external tool, check #Integrate with Build and Reporting Tools.
  • extract coverage statistics from coverage report and publish statistics values to TeamCity with help of service message: if you do so, you'll see coverage chart on build configuration Statistics tab and also you'll be able to fail a build with the help of a build failure condition on a metric change (for example, you can fail build if the coverage drops).

...

CVE-2017-5638 affects Jakarta Multipart parser in Apache Struts. Since no versions of TeamCity use struts, TeamCity is not affected.CVE-2016-1181 also affects multipart requests processing in some older versions of Struts.

TeamCity bundles IntelliJ IDEA which contains jars from both: Apache Struts 1.x and Apache Struts 2.x. These jars are only used by IntelliJ IDEA Struts plugin when IntelliJ IDEA collects inspections for a project on a TeamCity agent.

But under no circumstances these versions of Apache Struts are used to handle any HTTP requests. Thus neither TeamCity server, not TeamCity agent are affected by these vulnerabilities.

 

Tomcat Under Windows

Based on the wording of the description of CVE-2017-12615, CVE-2017-12616 and CVE-2017-12617 TeamCity server installed under Windows is a potential subject for the attack. However, our analysis of the vulnerabilities indicates that these potential vulnerabilities cannot be exploited in the default TeamCity installation as the related configuration of Tomcat is inactive in all the TeamCity versions.
If necessary, Tomcat bundled with TeamCity can be upgraded to the version 7.0.82 which also removes the vulnerability form the Tomcat code.

hidden-data
related issues:
https://youtrack.jetbrains.com/issue/TW-51625
https://youtrack.jetbrains.com/issue/TW-51905

...

"major" release below means any release with a change in the first or second version number (e.g. X in X.X.Z)
"bugfix" release means releases with a change in the third version number (e.g. Z in X.X.Z)

Release stages that we generally have are:
Available under EAP (Early Access Program) - usually available only for major releases, starts several months after previous major release and usually months before the next major release. Typically, new EAP releases are published on a monthly or bi-monthly basis.
General Availability - : as a rule, there is a major release each 8 months. There are multiple bugfix releases following the major release. Bugfix The bugfix releases and support patches for critical issues (if applicable) are provided until "End of Sale" of the release.
End of Sale - occurs  occurs with the release of a new major version. After this time, no bugfix updates or patches are usually provided (Exceptions are critical issues without a workaround which at the same time allow for relatively simple fix and inability for the customer to upgrade for an important reason). Only limited support is provided for these versions.
End of Support - occurs with the release of  occurs when two newer major versions have been released. At this point we stop providing regular technical support for the release.

...