LATEST VERSION: 3.2 - CHANGELOG
Pivotal tc Server v3.2

Installing Pivotal tc Server

Installation options vary according to whether your operating system is Linux or another supported platform. Installation also varies according to whether you are installing tc Server for the first time or are upgrading from a previous version.

Install tc Server Standard Edition From a ZIP or TAR File

Prerequisites

  • Verify that your system meets the supported configurations and installation requirements. See Supported Configurations and System Requirements.
  • Review information about tc Server Standard edition and its *.zip or *.tar.gz distribution files. See tc Server Editions.
  • If you are installing from a *.tar.gz on a Solaris, make sure to use GNUtar to unpack the archive.
  • Determine the user that is going to create and run the tc Runtime instances, and create it if necessary. Consider creating a user dedicated to tc Server tasks, putting the user in a separate group from regular users, and disabling its interactive login for security purposes. See Setting Up Unix Users for tc Server and Hyperic.

    Attention: On Unix, never run tc Runtime instances as the root user.

    For clarity, it is assumed in this topic that you will install and run tc Server as the tcserver user.

Procedure

  1. From the Pivotal tc Server download page.
  2. Download the Standard Edition package distribution in ZIP or compressed TAR format to a directory on your computer, for example /home/Downloads.

    • pivotal-tc-server-standard-version.RELEASE.zip
    • pivotal-tc-server-standard-version.RELEASE.tar.gz
  3. Log in to the computer on which you are installing tc Server as the appropriate user, such as tcserver. On Unix, if you have disabled interactive login, login as the root user and use su - tcserver to become the user.

  4. Open a terminal (Unix) or command window (Windows) and create the main tc Server installation directory, such as /opt/pivotal.

    For example, on Unix:

    prompt$ mkdir /opt/pivotal 
    
  5. Extract the tc Server distribution file into the new directory.

    This action installs tc Runtime; there is no installer program.

    For example, if you created a directory called /opt/pivotal in the preceding step, and downloaded the Standard Edition ZIP file in the /home/Downloads directory:

    prompt$ cd /opt/pivotal
    prompt$ unzip /home/Downloads/pivotal-tc-server-standard-3.0.2.RELEASE.zip 
    

    This action creates a directory called pivotal-tc-server-standard-version in the main tc Server installation directory that contains the tc Runtime utility scripts, the templates directory, the tomcat-version directory, and so on.

What to do next

Mac OS X - Install Pivotal tc Server Developer Edition Using Homebrew

You can install Pivotal tc Server Developer Edition on Mac OS X computers using Homebrew.

Note: You may not be able to install the tc Server brew from inside a firewall.

Install Pivotal tc Server Developer Edition

Prerequisites

Procedure

  1. Log in to the Mac OS X computer on which you will install Pivotal tc Server.
  2. Execute the following brew commands:

    brew tap pivotal/tap
    
    brew install tcserver
    

What to do next

  • For details about the directories you installed, see Overview of tc Server Directories, Variables, and Configuration Files.
  • For typical post-installation procedures such as creating tc Runtime instances and starting tc Server components, see Creating and Managing tc Runtime Instances.
  • Run the following command to create a new tc Server instance in the current directory:

    tcruntime-instance.sh create <instance_name>

  • Run the following command to create a new tc Server instance with Spring Insight monitoring:

    tcruntime-instance.sh create -t insight <instance_name>

  • Run the following command to start a tc Server instance in current directory:

    tcruntime-ctl.sh <instance_name> start

RHEL - Install Pivotal tc Server Standard Edition from an RPM

Pivotal recommends that you install Pivotal tc Server (Standard Edition) on a Red Hat Linux Enterprise (RHEL) computer by first installing the Pivotal RPM repository and then using yum to perform the actual installation. See Install Pivotal tc Server from the Pivotal RPM Repository.

You can also download the RPM from the Pivotal download page and install it on your RHEL computer using the rpm command, as described in Install Pivotal tc Server from a Downloaded RPM.

Install Pivotal tc Server from the Pivotal RPM Repository

Pivotal recommends that you install tc Server on RHEL computers using the Pivotal RPM repository.

Prerequisites

  • Set the JAVA_HOME environment variable in the root user’s environment. For example, you could add the following line to /etc/profile and then open a new terminal window:

    export JAVA_HOME=/usr/java/latest

    Replace/usr/java/latest with the base directory of your JVM installation.

  • Verify that your system meets the supported configurations and installation requirements. See Supported Configurations and System Requirements.

  • Install the Pivotal repository RPM, which makes it easier for you to browse the Pivotal RPMs, including the Pivotal tc Server RPM. You install the Pivotal repository RPM on each RHEL computer on which you want to install one or more Pivotal products, such as Pivotal tc Server.

    1. On the RHEL computer, start a terminal either as the root user or as an unprivileged user who has sudo privileges.
    2. Install the Pivotal repository RPM using the following wget command, passing it the appropriate URL.

      Important: Run the entire wget command on a single line. Be sure you include the | sh at the end, or the RPM installation fails.

      prompt# wget -q -O - http://packages.pivotal.io | sh 
      

      Use sudo to run the preceding commands if you are not logged in as the root user. For example:

      prompt$ wget -q -O - http://packages.pivotal.io | sudo sh
      

      The command performs the following tasks:

      • Imports the Pivotal GNU Privacy Guard (GPG) key.
      • Installs the Pivotal repository RPM.
      • Launches the Pivotal End User License Agreement (EULA) acceptance and repository configuration script.
      • Outputs the EULA for you to read; you must answer yes to accept the terms and continue.
    3. Use the yum search pivotal command to view the list of Pivotal components that you can install from the Pivotal repository. For example (output truncated for clarity):

      prompt# yum search pivotal
      ...
      ======================================== Matched: pivotal ========================================
      pivotal-rabbitmq-java-client-bin.noarch : The RabbitMQ Java Client Library
      pivotal-rabbitmq-server.x86_64 : The RabbitMQ server
      pivotal-tc-server-standard.noarch : Pivotal tc Server Standard
      pivotal-web-server.x86_64 : Pivotal Web Server
      ...
      

      The Pivotal tc Server RPM is called pivotal-tc-server-standard.

Procedure

  1. From the RHEL computer on which you will install Pivotal tc Server, log in as the root user (or as an unprivileged user who has sudo privileges) and start a terminal.
  2. Execute the following yum command:

    prompt# yum install pivotal-tc-server-standard 
    

    The yum command begins the install process, resolves dependencies, and displays the packages it will install.

    If necessary, use sudo to run the preceding command if you are not logged in as the root user. For example:

    prompt$ sudo yum install pivotal-tc-server-standard 
    
  3. Enter y at the prompt to begin the actual installation.

    If the installation is successful, you will see a Complete! message at the end.

What the yum install command does

The yum install command:

  • Installs Pivotal tc Server into the /opt/pivotal/pivotal-tc-server-standard directory and sets the owner of the directory, along with all child directories and files, to root:pivotal.
  • If the user does not already exist, adds a tcserver user (in the group pivotal). Pivotal recommends that you create and run tc Server instances as this user.

    You cannot log in directly as the tcserver user. Rather, you must log in as the root user or as a privileged user using sudo , and then su - tcserver .

  • Creates an empty directory called /var/opt/pivotal/pivotal-tc-server-standard and sets the owner to tcserver:pivotal.

    Pivotal recommends that you create new tc Server instances in this directory rather than the installation directory. You do this by specifying the -i option of the tcruntime-instance command.

What to do next

Install Pivotal tc Server From a Downloaded RPM

You can install Pivotal tc Server on RHEL by downloading the RPM from the Pivotal download center and executing the rpm command.

Prerequisites

Procedure

  1. Log in to the RHEL computer on which you will install Pivotal tc Server as the root user (or as an unprivileged user who has sudo privileges).
  2. From the Pivotal tc Server download page.
  3. Download the tc Server Standard Edition RPM file to a directory on your computer. The RPM file is called pivotal-tc-server-standard-version-RELEASE.noarch.rpm.
  4. Start a terminal and change to the directory in which you downloaded the RPM.
  5. Execute the following rpm command to install tc Server:

    prompt# rpm -ivhf pivotal-tc-server-standard-version-RELEASE.noarch.rpm 
    

    If necessary, use sudo to run the preceding command if you are not logged in as the root user. For example:

    prompt$ sudo rpm -ivhf pivotal-tc-server-standard-version-RELEASE.noarch.rpm 
    
  6. In the previous section, see What the yum install command does for post-installation information, such as the installation directory and the user that is automatically created by the RPM installation. (The yum install command corresponds to the rpm command in this procedure.)

Install Hyperic Agent Plugin

Two versions of the Hyperic Agent plugin are available:

tc Runtime 8 Plugin

  1. Download the Pivotal tc Server plugin from the Pivotal Network.
  2. Rename the plugin file pivotal-tcserver-plugin.jar.
  3. Install the plugin using the Hyperic Plugin Manager.

    For instructions, see “Deploying and Managing Plug-ins” in the VMware vCenter Hyperic documentation.

tc Runtime 7 Plugin

  1. Download the Pivotal tc Server plug-in 2.9.x for VMware Hyperic from Pivotal Network.
  2. Complete the installation using the instructions found in the README file. The README is located inside the plugin zip archive.

Install tc Server Developer Edition From a ZIP or TAR File

When you install the Developer Edition of tc Server, you also typically create a tc Runtime instance that contains Spring Insight Developer. The procedure covers Unix and Windows installation, although most instructions are specific to Unix. If you install on Windows, change the forward slashes (/) to back slashes (\); other differences in the installation are called out.

Prerequisites

  • Verify that your system meets the supported configurations and installation requirements. See Supported Configurations and System Requirements.
  • Review information about tc Server Developer edition and its *.zip or *.tar.gz distribution files. See tc Server Editions.
  • If you are installing from a *.tar.gz on a Solaris, make sure to use GNU tar to unpack the archive.
  • Determine the user that is going to create and run the tc Runtime instances, and create it if necessary. Consider creating a user dedicated to tc Server tasks, putting the user in a separate group from regular users, and disabling its interactive login for security purposes. See Setting Up Unix Users for tc Server and Hyperic.

    Attention: On Unix, never run tc Runtime instances as the root user.

    For clarity, it is assumed in this topic that you will install and run tc Server as the tcserver user.

Procedure

  1. From the Pivotal tc Server download page.
  2. Download the Developer Edition distribution in ZIP or compressed TAR file format.

    • pivotal-tc-server-developer-version.RELEASE.zip
    • pivotal-tc-server-developer-version.RELEASE.tar.gz
  3. Login to the computer on which you are installing tc Server as the appropriate user, such as tcserver. On Unix, if you have disabled interactive login, login as the root user and use su - tcserver to become the user.

  4. Open a terminal (Unix) or command window (Windows) and create the main tc Server installation directory, such as /opt/pivotal.

    For example, on Unix:

    prompt$ mkdir /opt/pivotal                  
    
  5. Extract the tc Server distribution file into the new directory.

    This action creates a directory called pivotal-tc-server-developer-version in the main tc Server installation directory that contains the tc Runtime utility scripts, the templates directory, the tomcat-version directory, and so on.

    The templates directory contains a template called insight that contains the Spring Insight application.

  6. Create a tc Runtime instance that contains Spring Insight by specifying the insight template.

    Unix: Change to the /opt/pivotal/pivotal-tc-server-developer-version directory and execute the tcruntime-instance.sh script to create an instance.

    Pivotal recommends that you use the -i option to specify the full pathname of a directory in which the new instance will be created, and that this directory be different from the installation directory. Be sure the tcserver user can write to this directory and that the directory already exists.

    For example:

    prompt$ cd /opt/pivotal/pivotal-tc-server-developer-3.0.2.RELEASE
    prompt$ ./tcruntime-instance.sh create -t insight 
            -i /var/opt/pivotal/pivotal-tc-server-developer insight-instance 
    

    Windows: Change to the \opt\pivotal\pivotal-tc-server-developer-version directory and execute the tcruntime-instance.bat script to create an instance:

    prompt> cd \opt\pivotal\pivotal-tc-server-developer-3.0.2.RELEASE
    prompt> tcruntime-instance.bat create -t insight 
            -i \var\opt\pivotal\pivotal-tc-server-developer insight-instance 
    
  7. Start the new tc Runtime instance.

    Unix: Execute the tcruntime-ctl.sh script to start the instance; use the -i option to specify the directory in which the instance is located. For example:

    prompt$ ./tcruntime-ctl.sh insight-instance start -i /var/opt/pivotal/pivotal-tc-server-developer 
    

    Windows: Execute the tcruntime-ctl.bat script to first install the tc Runtime instance as a Windows service and then start it; for both commands, use the -i option to specify the directory in which the instance is located:

    prompt> tcruntime-ctl.bat insight-instance install -i \var\opt\pivotal\pivotal-tc-server-developer
    prompt> tcruntime-ctl.bat insight-instance start -i \var\opt\pivotal\pivotal-tc-server-developer 
    

    Note: On Windows, Pivotal recommends that you subsequently start and stop the tc Runtime instance through the Windows Services console. The tc Runtime instance is displayed in the console with the name Pivotal tc Runtime instance - unique-name, where unique-name is a unique combination of server name and server directory.

  8. After the tc Runtime instance starts, invoke Spring Insight in your browser:

    http://host:8080/insight 
    

    where host refers to the computer on which Spring Insight is running. If you are on the same computer, you can use localhost:

    http://localhost:8080/insight 
    

What to do next

Overview of tc Server Directories, Variables, and Configuration Files

When you install the tc Runtime component, you simply unpack the appropriate *.zip or *.tar.gz file into the main installation directory. This action creates a pivotal-tc-server-edition-version subdirectory, where edition-version refers to the edition of tc Server that you are using (standard or developer) and the version of tc Server. This subdirectory in turn contains the following tc Server-related files and directories:

  • tomcat-*version* Where version is the version of the core Apache Tomcat on which this version of the tc Runtime is based, such as tomcat-7.0.42.A.RELEASE or tomcat-8.0.33.A.RELEASE. These directories are the basic Apache Tomcat CATALINA_HOME directory. Standard Apache Tomcat users recognize its contents.
  • templates Out-of-the-box templates for creating customized tc Runtime instances, such as cluster-node enabled or SSL-ready. You can specify one or more of these templates when you run the tcruntime-instance.sh|bat script to create a new tc Runtime instance. This is also the default location for custom template or template retrieved from the template repository. See Templates Provided by tc Runtime for the full list.
  • lib JAR files that implement the templating mechanism and are used by the tcruntime-instance script.
  • tcruntime-instance.sh|bat Scripts for creating new tc Runtime instances. When you create a new tc Runtime instance with this script, the script creates the instance directory specified with the -i option or, by default, a subdirectory of the pivotal-tc-server-edition-version directory with the same name as the new tc Runtime instance. This new directory is the CATALINA_BASE of the tc Runtime instance. The new directory contains the instance-specific configuration files, its own Web application deployment directory, log files, and so on.
  • tcruntime-ctl.sh|bat Scripts for controlling tc Runtime instances, such as start and stop scripts. The bin directories of individual tc Runtime instances include their own versions of these scripts that in turn call these main scripts. You can also call the top-level scripts if you specify the name of the tc Runtime instance.
  • tcruntime-admin.sh|bat Scripts used to handle adminstrative functions like encoding value for properties, retrieving the latest or a specific tc Runtime version, and retrieving templates from the tc Server Template Repository.
  • bash_completion. The bash_completion directory contains scripts to enable the bash completion capabilities for tcruntime-instance.sh, tcruntime-ctl.sh, and tcruntime-admin.sh. If you use a bash shell on a Unix-like system and you have the bash-completion package installed, you can use the Tab key to complete command arguments and suggest alternatives when using these tc Server scripts. See Enabling Bash Completion for tc Server Scripts for instructions on setting up this feature.

tc Server Variables

tc Server uses the following variables:

  • CATALINA_HOME . Root directory of your tc Runtime installation.

    The CATALINA_HOME variable points to the directory INSTALL_DIR/pivotal-tc-server-edition-version/tomcat-version, where INSTALL_DIR is the directory in which you installed tc Server (such as /opt/pivotal); edition-version refers to the version and edition of tc Server you are using (developer-3.0.2.RELEASE or standard-3.0.2.RELEASE); and version is the version of the underlying Tomcat, such as 7.0.47.A.RELEASE.

  • CATALINA_BASE . Root directory of a particular tc Runtime instance.

    This directory contains the instance-specific files, such as the conf/server.xml file that configures this particular instance. If you created a tc Runtime instance called myserver and you are using the Standard Edition, then the CATALINA_BASE of the instance is INSTALL_DIR/pivotal-tc-server-standard-version/myserver by default.

The following variables are “exposed” by tc Runtime, which means that you can set them or use them in your environment (or in the bin/setenv.sh file of your tc Runtime instance) to achieve the specified results:

  • CATALINA_OUT . Unix only. Use this environment variable to specify a file to which a tc Runtime instance writes stdout and stderr messages. If you do not set this environment variable explicitly, the tc Runtime instance writes stdout and stderr messages to the file CATALINA_BASE/logs/catalina.out.

    For example, to specify that the tc Runtime instance write its stdout and stderr messages to /opt/pivotal/tcserver/tcruntime-instance-6.log, set the variable in your environment or setenv.sh as follows:

    CATALINA_OUT=/opt/pivotal/tcserver/tcruntime-instance-6.log
    
  • INSTANCE_NAME . Name of the tc Runtime instance. You can use this variable to create other unique variables within configuration scripts.

    For example, on Unix platforms you can update the bin/setenv.sh file to use the name of the tc Runtime instance when defining the CATALINA_OPTS variable as follows:

    CATALINA_OPTS="-Dinstance.name=$INSTANCE_NAME"
    

    On Windows, the equivalent change would be to the conf/wrapper.conf file as follows:

    set CATALINA_OPTS=-Dinstance.name=%INSTANCE_NAME%
    
  • INSTANCE_BASE . Specifies the parent directory of the tc Runtime instance. The full pathname of the tc Runtime instance directory would be $INSTANCE_BASE/$INSTANCE_NAME.

    You can use the INSTANCE_BASE variable in the same way as the INSTANCE_NAME variable, as described in the preceding bullet.

tc Runtime Instance Directory Structure

After you create a new tc Runtime instance, its CATALINA_BASE directory contains the following subdirectories:

  • bin . Contains the tcruntime-ctl.* scripts to start and stop tc Runtime instances, as well as the setenv.* scripts. The *.sh Unix files are functional duplicates of the *.bat Windows files.
  • conf . Contains the configuration files for the tc Runtime instance, such as server.xml, catalina.properties, web.xml, context.xml, and so on.
  • lib . Contains resources shared by all Web applications deployed to the tc Runtime instance.
  • logs . Location of the logs files.
  • webapps . Deployment directory for the Web applications deployed to the tc Runtime instance.
  • work . Temporary work directory for all deployed Web applications.
  • temp . Directory used by the JVM for temporary files.

tc Runtime Instance Configuration Files

You configure a particular tc Runtime instance by changing its configuration files. Other topics in this documentation describe how to do this. All the configuration files for a tc Runtime instance are located in its CATALINA_BASE/conf directory. The most important configuration files are as follows:

  • server.xml . Main configuration file for a tc Runtime instance. It configures the behavior of the servlet/JSP container.

    By default, the server.xml file for a tc Runtime instance uses variable substitution for configuration properties that must be unique across multiple tc Runtime instances on the computer, such as HTTP and JMX port numbers. These variables take the form ${var}. For example, the variable for the HTTP port that the tc Runtime instance listens to is ${http.port}. The specific values for these variables for a particular tc Runtime instance are stored in the catalina.properties file, in the same directory as the server.xml file.

  • catalina.properties . Properties file that contains the tc Runtime instance-specific values for variables in the server.xml file.

  • context.xml . Configures the context that is loaded by all Web applications deployed to the tc Runtime instance.

  • web.xml . Default web.xml file that is loaded by all deployed Web applications, in addition to their individual web.xml files.

  • wrapper.conf . Windows only. Configures the Java Service Wrapper from Tanuki Software used to install the tc Runtime instance as a Windows service. The Wrapper correctly handles user log outs under Windows, service dependencies, and the ability to run services that interact with the desktop.

  • jmxremote.access and jmxremote.password . Configures the JMX users and passwords. The default JMX user, added at instance creation time unless you specify something different, is called admin with a password made up of a list of random characters.

  • logging.properties . Configures the logging system of the tc Runtime instance.

Enabling Bash Completion for tc Server Scripts

If you use the bash shell on a Unix-like system and you have the bash-completion package installed, you can enable completion support for the tc Server tcruntime-instance.sh, tcruntime-ctl.sh, tcruntime-admin.sh scripts. When enabled, you can press the Tab key after entering a few letters of a command argument and either the argument is completed for you or possible alternatives are suggested.

Bash completion is enabled by linking the bash completion scripts supplied with tc Server into the bash_completion.d directory on your system.

See Bash Completion for usage instructions.

Prerequisites

  • You must be using the bash shell on a Unix-like system and have the bash-completion package installed and enabled.
  • Know the location of the bash_completion.d directory on your system, usually /etc/bash_completion.d.
  • You need superuser access to install the scripts. Either log in as root, or use su or sudo to temporarily become root.

Procedure

  1. As the superuser, create symbolic links for the tc Server bash completion scripts in the bash_completion.d directory, using a command like the following:

    prompt# ln -s /opt/pivotal/tcserver/pivotal-tc-server-standard-3.0.2.RELEASE/bash_completion/* /etc/bash_completion.d
    

    Replace /etc/bash_completion.d with the path to the bash_completion.ddirectory on your system, if it differs.

  2. With your regular user login, start a new bash shell or re-source your environment to allow the bash_completion script to recognize the new tc Server scripts.

Setting Up Unix Users for tc Server and VMware vCenter Hyperic

On Unix-like systems, the interaction between VMware vCenter Hyperic and tc Server is straightforward as long as tc Runtime instances and the Hyperic Agent run as the same user.

You can run Hyperic Agent and tc Runtime instances with different user IDs. You might do this for increased security, or because the Hyperic Agent needs to run as a privileged user to manage some other resource on the computer, or perhaps you want to run different tc Runtime instances as different users to take advantage of process accounting.

The Hyperic tc Server plug-in detects the user and group running the tc Server process and records them in parameters in the Hyperic Server resource created for the instance. If the user is different from the user running Hyperic Agent, the plug-in uses su or sudo to set the user whenever you start, restart, or stop a tc Runtime instance or change the tc Runtime instance’s configuration through Hyperic.

Both Hyperic Agent and tc Runtime instances should run as regular, non-root users. Never run a tc Server instance as root.

If you use different non-root users to run tc Server instances and Hyperic Agent, you must create them in the same primary group. This is necessary to allow Hyperic Agent to read files written by the tc Runtime instance.

Subtopics

Creating Users and Groups for Hyperic and tc Server

Setting the tc Server User in Hyperic

Enabling Hyperic Agent Access to su or sudo

Creating Users and Groups for Hyperic Agent and tc Server

When you run Hyperic Agent and tc Runtime instances with different users, they must be in the same primary group to allow them to share files. For better security, you can create a separate group for them.

The following procedure shows how to create a group and add users to it for tc Server and Hyperic Agent on Red Hat Linux. The exact commands may be different on other operating systems.

Procedure

  1. Log in as root and start a terminal session.
  2. Use the groupadd command to create a new group. The following example creates a pivotal group:

    prompt$ groupadd pivotal 
    

    Note that if you installed from RPM on RHEL, the pivotal group may already exist.

  3. Use the useradd command to create a user for Hyperic Agent in the group you created in the previous step. The following example creates a hyperic user in the pivotal group:

    prompt$ useradd hyperic -g pivotal 
    

    You can include the -M option to prevent creating a home directory for the user and the -s /sbin/nologin option to prevent anyone from logging in as the hyperic user.

    Install and run Hyperic Agent as this user.

  4. Use the useradd command to create a user to run tc Server instances. The following example creates a tcserver user in the pivotal group:

    prompt$ useradd tcserver -g pivotal 
    

    You can include the -M option to prevent creating a home directory for the user and the -s /sbin/nologin option to prevent anyone from logging in as the tcserver user.

    Create the tc Server instance and run it as this user.

  5. If you want to run multiple tc Runtime instances under separate user accounts on the same computer, repeat the previous step to create additional tc Server users.

Setting the tc Server User in Hyperic

Hyperic uses auto-discovery to detect tc Runtime instances. The first time it discovers an instance, it records the user and group running the process. Therefore, the usual method to set the tc Server user is to create the instance and run it as the desired user, allowing Hyperic Agent to discover the instance.

If you are migrating to a new Hyperic release and you have existing tc Runtime instances detected by an earlier version of Hyperic, the user and group parameters are blank. The first time auto-discovery runs, the instances will show up as modified in the auto-discovery queue. When you accept the modified resources, the user and group are recorded.

If you decide to change the tc Server user for an instance previously created with a different user, be sure to chown all the files in the tc Runtime instance directory and ensure they are readable and writable by the new user. Then start the instance as the new user and trigger auto-detect in Hyperic to record the new user in the Hyperic resource record.

Enabling Hyperic Agent Access to su or sudo

Hyperic Agent uses the su or sudo command to execute tasks as the tc Server user. Specifically, if Hyperic Agent is running as root, it uses /bin/su to change to the desired user to perform the task. If running as a non-root user, Hyperic Agent instead uses /bin/sudo to do the work as the tc Server user. There are some prerequisites you must verify to ensure that Hyperic can use su or sudo, described below.

If Hyperic Agent is running as root

If Hyperic Agent is running as root, it will use su to execute tasks as the tc Server user. You must ensure that /bin/su exists. If not, create a link to it.

For example, if su is in /sbin, but not /bin, create a link as follows:

prompt$ sudo ln -s /sbin/su /bin/su

If Hyperic Agent is running as a non-root user

If Hyperic Agent is running as a non-root user, it will use sudo to execute tasks as the tc Server user. You must ensure that /usr/bin/sudo exists and also grant required permissions to the tc Server user in the /etc/sudoers file.

For example, if sudo is in /usr/sbin/, but not /bin, create a link as follows:

prompt$ sudo ln -s /usr/sbin/sudo /bin/sudo

The user running Hyperic Agent needs permission to run the tcruntime-ctl.sh script as the tc Server user without having to enter a password. This is accomplished by editing the /etc/sudoers file as root and adding an entry. For example, if Hyperic Agent is running as user hyperic, tc Server runtime instances are running as user tcserver, and the tcruntime-ctl.sh script is in /opt/pivotal/pivotal-tc-server-standard-3.0.2.RELEASE/tcruntime-ctl.sh, you would add the following entry to /etc/sudoers:

hyperic ALL=(tcserver) NOPASSWD: /opt/pivotal/pivotal-tc-server-standard-3.0.2.RELEASE/tcruntime-ctl.sh

Uninstalling tc Server - Typical Steps

You can uninstall one or more of the following components:

Uninstallation of tc Server mostly entails removing the directories that contain the component files, although a few extra steps might be required, as described below.

Each section covers both Unix and Windows commands. The documentation uses Unix-like forward slashes (/) for directories; if you are on a Windows platform, change these to back slashes (\).

Warning: The procedures in this section describe how to completely remove the components of tc Server from your computer.

Uninstalling tc Runtime

The following procedure describes how to uninstall the tc Runtime and all its associated instances.

  1. If currently running, stop all tc Runtime instances. See Starting and Stopping tc Runtime Instances.
  2. Start a terminal window (Unix) or Command Prompt (Windows).
  3. Windows only. If you installed any tc Runtime instances as Windows services, change to the CATALINA_BASE\bin directory of each instance (such as \var\opt\pivotal\pivotal-tc-server-standard\myserver\bin) and uninstall the service using the following command:

    prompt> \var\opt\pivotal\pivotal-tc-server-standard\myserver\bin
    prompt> tcruntime-ctl.bat uninstall 
    
  4. Remove the main tc Server installation directory. For example, if you installed Standard Edition, the delete command might look something like the following:

    prompt$ rm -rf /opt/pivotal/pivotal-tc-server-standard 
    

    By default, the home directory of all tc Runtime instances is under the main tc Server installation directory; if you used this default location when you created the tc Runtime instances with the tcruntime-instance script, then the preceding delete command also deleted all tc Runtime instances.

  5. If you created any tc Runtime instances in locations other than the default tc Server installation directory, remove their corresponding home directories.

Uninstalling Hyperic Agent

To uninstall the Hyperic Agent component of tc Server:

  1. If the agent itself is managed by Hyperic, remove the platform for the agent using the Hyperic user interface.
  2. Start a terminal window (Unix) or Command Prompt (Windows).
  3. Remove the directory in which you installed the Hyperic Agent. For example:

    prompt$ cd /opt/vmware/hyperic
    prompt$ rm -rf <version>