Chapter 2

This chapter details Apex and Summit configuration, performed after installation. You must complete these steps before using Apex or Summit.

The following topics are covered in this section:

• "Configuring Lock Recovery"
• "Configuring Apex/ClearCase Integration"
• "Configuring Apex or Summit with a Java Compiler"
• "Configuring a SPARC C++ Compiler"
• "Configuring the Summit TM Default Domain"
• "Configuring the Web User Interface"
• "Using Apex or Summit"

In order for Apex or Summit to automatically recover file locks that were abandoned by a terminated process from another machine, they need to remotely run a command to check process existence on that other machine. There are several options available:

Lock-Recovery Option	Description
No_Remote	Do not attempt such lock recoveries
Rsh	Use a remote shell (default)
On	Use the on command (default for SunOS & AIX)

• No_Remote

  If you specify the No_Remote option, it is assumed that all remote processes exist without making any attempt to verify that assumption. This is the least intrusive and most secure option. However, with No_Remote, only locks that were acquired on the same machine can be automatically recovered.

• Rsh or On

  If you specify the Rsh or On option, a remote shell or the on command is used to run a process status command on a remote machine to determine whether a process on that machine still exists. One difference between these two methods is performance; the remote shell is usually significantly slower.

  You can alter the behavior by rerunning the apexinit_setup command in the install directory.

If you install the Apex/Clearcase integration component and have ClearCase installed, it will be configured during the post-install operation. You need to know the location of the ClearCase root directory. This is typically in /usr/atria. You can configure it later using the clearcase_setup command in the install directory. you can use this command to disable the Apex/Clearcase integration or re-enable it.

If you have a Java compiler that you would like to use with Apex or Summit, it can be configured during the post-install operation. You need to know the location of your Java compiler root directory. You can configure it later or for a different Java compiler using the java_setup command in the install directory. This command will prompt you for the location of your Java compiler directory.

If you have a Sparc C++ compiler that you would like to use with Apex or Summit (other than the Rational Apex C++ compiler), it can be configured during the post-install operation. You need to know the location of the Sparc C++ compiler directory. You can configure it later or for different Sparc C++ compiler using the cpp_setup command in the install directory.

If you have installed Summit TM, you will be asked to create a task directory, a task domain subsystem and if you want to configure it as the default domain during the post-install operation. You can configure this later using the summit_tm_setup command in the install directory.

The web user interface can be configured during the post-install operation.You need to know the location of your web server root directory. You need to know the root password so the setup script can set the Apex or Summit login program to run setuid mode. The source of the setuid login program is provided in the $APEX_HOME/web/src directory.

You can configure the web user interface later or for another web server using the web_ui_setup command in the install directory. This needs to be configured on the machine where the web server files are stored.

If you do not have root privileges on the web server machine you can get your system administrator to run the web_ui_setup command.

You have completed the Apex/Summit installation.

To start Apex, follow the instructions for Apex users in the Rational Apex Getting Started Manual.

To start Summit, follow the instructions for Summit users in the Summit Getting Started Manual.