Installation instructions for IBM SPSS Statistics Server

The following instructions are for installing IBM® SPSS® Statistics Server version 30.

System requirements for IBM SPSS Statistics Server

To view the system requirements for IBM SPSS Statistics Server, go to https://www.ibm.com/software/reports/compatibility/clarity/softwareReqsForProduct.html.

Additional requirements for UNIX only:
Client software

Beginning with version 20.0.1, the client software does not have to be at the same release level as the server software to which it is connecting. For example, the 20.0.1 client can connect to the 21 server software, and the 21 client can connect to a 20.0.1 server. You can also run multiple versions of the server software on a server computer.

Mixing release levels is allowed only to simplify upgrades. Release levels can be staggered during the upgrade period, and clients do not need to be upgraded simultaneously. However, it is not recommended to maintain this configuration for an extended time. If the server is newer than the client, the server might create output that cannot be read by the client. If the client is newer than the server, syntax that is submitted that is by the client might not be recognized by the server. Therefore, you need to upgrade the client or server software as soon as possible, depending on which one lags the other.

When you distribute connection information to users, keep in mind which version of the client software they are running and be sure that they have the connection information for a matching server version.

Locales for UNIX servers
The IBM SPSS Statistics Server locale is automatically synchronized with the client locale according to mappings defined in the loclmap.xml file, which is located in the config subdirectory of the IBM SPSS Statistics Server installation directory. If you need to add locales specific to your system or you need to change the order in which IBM SPSS Statistics Server looks for an appropriate locale, you can modify loclmap.xml. You can also issue the SHOW LOCALE command to display the locale that the IBM SPSS Statistics Server is using. For more information about locales, see the IBM SPSS Statistics Server Administrator's Guide (on the CD-ROM in /Documentation/<language>/Manuals).

Upgrading IBM SPSS Statistics Server

If you currently have IBM SPSS Statistics Server on the target computer, skip to either Installing IBM SPSS Statistics Server for Windows or Installing IBM SPSS Statistics Server for UNIX. If you do have an installation on the target computer, you can either:

  • Replace the current version. Replace the old version if the new version is a minor release compared to the old version. You can check whether the release is minor by looking at the first two parts of the version number. If these are the same as the installed version, the new version is a minor version compared to the installed version. For example, if the new version is 28.0.1 and the old version is 28.0, the new version is a minor release and you must replace the old version.
  • Install the new version in addition to your current version. You can install the new version in addition to the old version if the new version is a major release compared to the old version. For example, if the new version is 30.0 and the old version is 28.0.1, you can install the new version in addition to the current version.
To Replace the Current Version
  1. Back up your existing configuration file (statisticsd.conf) to another location. You can use the administration console to check the location of your configuration file.
  2. Back up your JRE keystore certification files (back up the contents of the ../JRE/security/cacerts directory).
  3. Remove the old version. Follow the steps in either Uninstalling IBM SPSS Statistics Server for Windows or Uninstalling IBM SPSS Statistics Server for UNIX.
  4. Install the new version. Follow the steps in either Installing IBM SPSS Statistics Server for Windows or Installing IBM SPSS Statistics Server for UNIX.
  5. Compare your old configuration file with the new configuration file and transfer any options settings that you want to preserve in the new file. Do not replace the new file with the old one.
  6. Replace the JRE keystore certification files in ../JRE/security/cacerts directory with the backed-up files from step 2.
To Add the New Version
Follow the steps in either Installing IBM SPSS Statistics Server for Windows or Installing IBM SPSS Statistics Server for UNIX, with the following comments:
  • Destination directory. Install the new version in the different location from the old version.
  • For Windows only: IP address. During installation you will be prompted to enter an IP address for an IBM SPSS Statistics Server to use. You can use the same address for both versions.
  • Port number. Use a different port number for the new version. By default, a different port number is set for the new version.
  • Temporary files. Use the same temporary file location for both versions. By default, the same location is used for both versions.

The port number and temporary file location are specified in the administration console. Refer to the Administration Consoles section of the Deployment Manager User's Guide (included in the help for ) for more information.

Whenever you install a new version, be sure to distribute the IBM SPSS Statistics Server product's host name and port number to the users.

Installing IBM SPSS Statistics Server for Windows

The Setup program installs the following components:

  • A Windows service that manages user requests.
  • Software that handles data access and statistical processing.

Note that IBM SPSS Statistics Server must be installed on a hard disk drive on the computer on which the Setup program is running.

Destination

You can install in a different destination folder, but you must install on the computer from which the setup is being run (you cannot install to a network location).

If you are installing on the same computer with other Server products, install in a separate directory. Do not install multiple Server products in the same directory.

For doing an Interactive install of 29.0. 1 IBM SPSS Statistics Server,

  1. Use the self-extracting, Windows installation .exe file to install the IBM SPSS Statistics Server.
  2. Go to the directory where you have downloaded the executable file. Right-click on that file and select ‘Run as administrator’.
  3. After installation is completed, Windows prompts you to click ‘Finish’.

For doing a Silent install of 29.0. 1 IBM SPSS Statistics Server.

Steps before you do a push installation:

  1. Download the software.
  2. Create a directory where you want to extract the .msi file.
Steps for push installation:
Note: Modify the commands based on where you created the installation directory.
  1. Go to the command line. Change to the installation directory.
  2. Run the following command to extract the installer *.msi file from the downloaded .exe file. 
    SPSSSS_29.0.1_MP_ML______.exe /b"C:\shared\media\install"

    - Replace the file path with the path, which points to the folder that you created to extract the .msi file.

    - Cancel the installer after the files are extracted. (Usually after the first step in the GUI).

  3. Copy and run the following msi executable command after modifying the command line by adding the appropriate properties. Make sure to specify the correct .msi file in the command line. 
    MsiExec.exe /i "IBM SPSS Statistics Server.msi" /qn /L*v logfile.txt INSTALLDIR="C:\StatisticsServer\2901"
  4. Share the IP address and port number with the users.

Installing IBM SPSS Statistics Server for UNIX

You can install IBM SPSS Statistics Server as root or as a non-root user. If your site restricts the use of the root password, use an authentication method that supports running as non-root (see the IBM SPSS Statistics Server Administrator's Guide). Then, install the product as the user who will run the daemon.

Important: The file system on which you install IBM SPSS Statistics Server must be mounted with the suid option. The product will not work correctly if the file system is mounted with the nosuid option.
  1. Check hard drive space. In addition to the permanent hard drive space, you need temporary disk space for the installer files. The installer files are extracted to your system's temporary folder. If there is not enough space in the temporary folder, the installer files are extracted to your home folder. If neither location has enough space, the installer cannot continue. In this case, you can temporarily set the IATEMPDIR environment variable to a location with adequate space.
  2. Check the destination directory. By default, IBM SPSS Statistics Server is installed to /opt/IBM/SPSS/StatisticsServer. If desired, you can change this path in the graphical installation wizard. If you are going to run the silent installer, you can set the value for USER_INSTALL_DIR in installer.properties. Regardless, you need read and write permissions to the installation directory, so log on with an account that has sufficient permissions. Note that if you are upgrading by adding a new version of the product, install the new version in a separate directory.
  3. Change execute permissions of installer. Be sure that the installer is executable by the user who will run the installer.
  4. Run the installer. You can run the installer in a graphical user interface or silently. If you plan to run IBM SPSS Statistics Server as a user other than root, run the installer as the same user who will run the product. The following instructions explain each method.
Graphical installation wizard

The graphical installation wizard displays a graphical interface that will ask you about installation parameters. You will need an X Window System.

  1. At the UNIX prompt, change to the directory where the installer file was copied or extracted.
  2. Run the installer by using the following command: 
    ./<installer_name>
  3. After the installation wizard is launched, follow the instructions that appear on the screen.
Silent installation
Silent mode enables the installation without any user interaction. Installation parameters are specified as a properties file.
  1. Copy installer.properties to the same location where you copied the installer.
  2. Create the installer.properties file by coping the following text.
    ################################################################################
#It is important that this file is placed in the same folder as the installer file
#Now launch the installer as usual and it will detect the file and run in silent mode
################################################################################
#
# IBM SPSS Statistics server installler properties File
#
#
# This file can be used to configure Install with the options specified below
# when the wizard is run with the "-options" command line option. Read each
# setting's documentation for information on how to change its value.
# 
# A common use of an options file is to run the wizard in silent mode. This lets
# the options file author specify wizard settings without having to run the
# wizard in graphical or console mode. To use this options file for silent mode
# execution, use the following command line arguments when running the wizard:
# 
# LINUX   ./setup.bin -f installer.properties
#
################################################################################

# IBM SPSS Statistics 27 silent parameter
INSTALLER_UI=silent

################################################################################
#
# IBM SPSS Statistics 27 Install Directory
#
# 
#

USER_INSTALL_DIR=<value>

###############################################################################

# IBM SPSS Statistics 27   License Acceptance
#
# license acceptance for silent installers
LICENSE_ACCEPTED=true
  3. Go to the location where you copied or extracted the installer files, and open installer.properties in a text editor.
  4. Set the value for USER_INSTALL_DIR to the desired installation directory. The directory path cannot contain spaces. If you want to install to the default installation directory, comment out this line; do not leave it set to <value>.
  5. Save the file.
  6. Run the installer by executing the following command:
    ./<installer_name> -f installer.properties
    Where <installer_name> is the installer .bin file.

After You Install IBM SPSS Statistics Server on Windows

This section describes some required and optional steps that you can perform after installation. It does not describe all possible configuration options. You can find information about all the configuration options in the IBM SPSS Statistics Server Administrator's Guide.

Checking the Server Status
  1. On the computer where you installed IBM SPSS Statistics Server, select Services from Administrative Tools on the Control Panel.
  2. Locate IBM SPSS Statistics Server 30.x in the list. If the service is not started, double-click its name and start it on the dialog box that appears. Note that if the service startup is configured to be Automatic, the service will start automatically whenever the computer is restarted.
  3. Click OK to close the dialog box.

IBM SPSS Statistics Server is now ready to accept connections from end users when they have been authorized. See the following "Connecting End Users" instructions for more information.

Connecting users

Users connect to IBM SPSS Statistics Server by logging in from the client software. See the IBM SPSS Statistics Server Administrator's Guide for a description of how the software works and what you need to do to administer it. Give the users the information that they need to connect, including the IP address or host name of the server machine.

Example files: Example files are installed in the IBM SPSS Statistics Server installation directory. They are available for users who connect to the Server and want to test the software.
Administration Console Application
The administration console application is IBM SPSS Statistics Administration Console, which is installed as part of IBM SPSS Deployment Manager. The IBM SPSS Statistics Administration Console collects and displays information about IBM Corp. server products. It also allows you to control and configure server software, including IBM SPSS Statistics ServerIBM SPSS Statistics Server. Complete documentation for the IBM SPSS Statistics Administration Console is included in the Administration Consoles section of the IBM SPSS Statistics User's Guide (included in the help for IBM SPSS Collaboration and Deployment Services).
IBM SPSS Data Access Pack Technology

IBM SPSS Statistics Server has a distributed architecture that uses ODBC to access data. You can use IBM SPSS Data Access Pack for accessing data. IBM SPSS Data Access Pack is available for download or from the DVD.

For information about the data access technologies, including information on choosing, installing, and deploying a technology, refer to the IBM SPSS Data Access Pack Installation Instructions.

There is additional information in the DataDirect online documentation, which includes Adobe Acrobat versions of detailed installation instructions, administration information, and ODBC reference guides. These documents are highly recommended if you plan to use IBM SPSS Data Access Pack technology.

DataDirect's product documentation for Connect ODBC is included, by default, as part of the IBM SPSS Data Access Pack installation. The installer creates the entry IBM SPSS OEM Connect and ConnectXE for ODBC along with the entries for your other programs on the Start menu. The DataDirect product documentation is accessed from this menu item.

Note that the documentation can also be accessed from the DataDirect home page at http://www.datadirect.com.

After You Install IBM SPSS Statistics Server on UNIX

This section describes some required and optional steps that you can perform after installation. It does not describe all possible configuration options. You can find information about all the configuration options in the IBM SPSS Statistics Server Administrator's Guide.

Starting the Process

IBM SPSS Statistics Server runs as a daemon process and has root privileges by default. IBM SPSS Statistics Server can be configured to run without root privileges. Refer to the IBM SPSS Statistics Server Administrator's Guide for more information. You need to choose an authentication method that does not require that the daemon runs as root. IBM SPSS Statistics Server must be started by the root user or the user who owns the analytic server software daemon. Start the application by running a startup script, start_statistics_server, which is included in the /bin subdirectory of the installation directory. The startup script configures the environment for and executes the software.

  1. Log in as root or the user who owns the analytic server software daemon.
  2. Change to the /bin subdirectory of the IBM SPSS Statistics Server installation directory. The startup script must be run from this location.
  3. Run the startup script. For example, at the UNIX prompt type: 
    ./start_statistics_server -d

    The -d parameter starts the application as a daemon process, which is the recommended way to run the software. You can also add the -g <groupname> parameter to run a separate instance of the server software for the specified group. The instance must have already been created. Refer to the IBM SPSS Statistics Server Administrator's Guide for more information about creating multiple instances for group authorization.

Note:
  • If you are running another IBM SPSS Statistics Server product on the same computer, be sure to start each product from its installation directory, as shown in the above steps. Also, be sure to start IBM SPSS Statistics Server with a script that defines its environment; do not define the environment in an external script (for example, do not define the IBM SPSS Statistics Server environment in .profile).
  • If the server computer does not support the localhost IP address (127.0.0.1/::1), then you must set the environment variable STATS_LH_OVERRIDE to YES before you start the server. STATS_LH_OVERRIDE is set from statsenv.sh, which is included in the /bin subdirectory of the installation directory. For information on starting and stopping the server, see the topic "Starting and Stopping the Server Software" in the IBM SPSS Statistics Server Administrator's Guide.

Additional options are recognized on the script's command line. For more information, see the "Configuration, Monitoring Usage, and Maintenance" chapter in the IBM SPSS Statistics Server Administrator's Guide.

Checking the Server Status
  1. At the UNIX prompt, type:
    ps –ef | grep statisticsd
  2. Look at the output. If the server is running, you will see it as the first process in the list.

IBM SPSS Statistics Server is now ready to accept connections from end users when they have been authorized. See the following "Connecting End Users" instructions for more information.

Connecting End Users

End users connect to IBM SPSS Statistics Server by logging in from the client software. See the IBM SPSS Statistics Server Administrator's Guide for a description of how the software works and what you need to do to administer it. You must give end users the information that they need to connect, including the IP address or host name of the server machine.

Example Files:

Example files are installed in the IBM SPSS Statistics Server installation directory. They are available for end users who connect to the Server and want to test the software.

Administration Console Application

The administration console application is IBM SPSS Statistics Administration Console, which is installed as part of IBM SPSS Deployment Manager. The IBM SPSS Statistics Administration Console collects and displays information about IBM Corp. server products. It also allows you to control and configure server software, including IBM SPSS Statistics Server. Complete documentation for the IBM SPSS Statistics Administration Console is included in the "Administration Consoles" section of the Deployment Manager User's Guide (included in the help for ).

IBM SPSS Data Access Pack Technology

IBM SPSS Statistics Server has a distributed architecture that uses ODBC to access data. You can use IBM SPSS Data Access Pack for accessing data. IBM SPSS Data Access Pack is available for download or from the DVD.

For information about the data access technologies, including information on choosing, installing, and deploying a technology, refer to the IBM SPSS Data Access Pack Installation Instructions for UNIX. There is additional information in the DataDirect online documentation, which includes Adobe Acrobat versions of detailed installation instructions, administration information, and ODBC reference guides. These documents are highly recommended if you plan to use IBM SPSS Data Access Pack technology.

Configuring IBM SPSS Statistics Server for Data Access:

If you want to use the IBM SPSS Data Access Pack with IBM SPSS Statistics Server, you will need to configure the startup scripts. This is a critical step because data access will not work otherwise. For instructions, refer to the IBM SPSS Statistics Server Administrator's Guide.

IBM SPSS Statistics Batch Facility

The IBM SPSS Statistics Batch Facility is a batch-processing utility included with IBM SPSS Statistics Server. The IBM SPSS Statistics Batch Facility is intended for automatic production of statistical reports. Automated production provides the ability to run IBM SPSS Statistics analyses without user intervention.

The IBM SPSS Statistics Batch Facility is automatically installed with IBM SPSS Statistics Server.

For UNIX only: The IBM SPSS Statistics Batch Facility is distributed with a manual page, statisticsb.1, which is installed in the /bin subdirectory of the installation directory. Copy it to the location in which you keep your manual pages.

For information about administering the IBM SPSS Statistics Batch Facility, see the IBM SPSS Statistics Server Administrator's Guide. For information about using the IBM SPSS Statistics Batch Facility, see the IBM SPSS Statistics Batch Facility User's Guide.

Uninstalling IBM SPSS Statistics Server for Windows

  1. From the Windows Start menu, click Settings > Control Panel.
  2. From the Control Panel, click Add/Remove Programs.
  3. Click the Change or Remove Programs button on the left, select IBM SPSS Statistics Server from the list, and click Change/Remove.

If you have more than one version of IBM SPSS Statistics Server installed on the computer, be sure to choose the version that you want to remove.

A message will be displayed when uninstallation is complete. This may take several minutes.

Uninstalling IBM SPSS Statistics Server for UNIX

Uninstall IBM SPSS Statistics Server by removing the program files:

  1. Stop the server process. Instructions for stopping the server process are in the IBM SPSS Statistics Server Administrator's Guide.
  2. Remove the installation directory.