Installing InfoSphere Data Click

Follow this procedure to install IBM® InfoSphere® Data Click. Links are provided on some of the steps in the procedure to give you more information or detailed steps.

Before you begin

About this task

There are two general paths to take when installing InfoSphere Data Click. In both cases, the InfoSphere BigInsights database system is used for the InfoSphere Data Click repositories:
  • The first and recommended path is to install InfoSphere Data Click on a computer separate from InfoSphere BigInsights to achieve best performance and avoid resource conflicts. In this case the database system is remote to InfoSphere Data Click.
  • The second path, useful only in a proof-of-concept, would be to install InfoSphere Data Click on the same computer as InfoSphere BigInsights. In this case the database system is local to InfoSphere Data Click.

Procedure

  1. On the InfoSphere Data Click computer, log in as the root user.
  2. On the InfoSphere Data Click computer, open a command shell and prepare the installation environment.
    1. Prepare the shell environment.
      • Run these commands if you are using the sh or bash shell:
        export IBM_JAVA_OPTIONS=
        umask 022
      • Run these commands if you are using the csh or tcsh shell:
        unsetenv IBM_JAVA_OPTIONS=
        umask 022
    2. If the target directory for installation is an NFS file system, ensure that the lock daemon is running.
      ps ax | grep -i lockd
      Start the lock daemon if it is not running. For more information, refer to your operating system documentation.
    3. Ensure that the limit to the number of open file descriptors is no less than 10240. For more information, refer to your operating system documentation
    4. Ensure that the volumes that contain the following directory paths have the required disk space. Also, ensure that existing directories have the required read and write permissions. Learn more.
      File path Description Minimum disk space

      /tmp
      This directory must exist.

      Your computer's temporary directory. 1.5 GB

      Download_directory
      You create this directory.

      The directory where you choose to download and extract the installation image. 3 GB

      /opt/IBM/InformationServer
      The installation program
      creates this directory.
      The path specified is the
      default.

      The directory where InfoSphere Data Click is installed. 2.5 GB

      /opt/IBM/InformationServer/Repos/xmeta
      The installation program
      creates this directory if
      the database is installed
      locally.
      The path specified is the
      default.

      The directory where the database is installed. Allocate this amount of space on the computer that hosts the database. 2.7 GB initially, with no data.
  3. On the InfoSphere Data Click computer, extract the installation image.
    1. Create and change to a directory in which to download and extract the installation image. The directory and each of its parent directories must have global read and write privileges.
    2. Download the software into this directory. The installation image can be found on Passport Advantage. Use the Knowledge Collection support document or refer to the Quick Start Guide for information on where to obtain the image file.
    3. In the download directory, extract the installation image:
      tar -zxvf is-bi.tar.gz
  4. You might need to create users depending on your authentication method and where the databases will be created. The default user names are shown in parentheses. If possible, use the default names unless they conflict with existing names. The documentation sometimes refers to these users by name.
    • If your database will be created on the local computer that uses operating system authentication, then the installation program can create the users for you.
    • If your database will be created on a remote computer that uses operating system authentication, create the following users on the remote computer:
      • A user to own the metadata repository database (xmeta).
      • A user to own the repository staging area (xmetasr).
      • A user to own the operations database schema (dsodb).
    • If you authenticate users with LDAP, PAM, or NIS, create the following group and users in the appropriate registry before running the installation program:
      1. On the InfoSphere Data Click computer, create a unique group to be used by engine users (dstage). Also, create an engine administrator user (dsadm) in the engine group.
      2. On the computer where the database system is installed, whether local or remote, create the following users:
        • A user to own the metadata repository database (xmeta).
        • A user to own the repository staging area (xmetasr).
        • A user to own the operations database schema (dsodb).
  5. Prepare the DB2® database system.
    • If you plan to use a database system that is already installed on the local InfoSphere Data Click computer:
      1. It must be a supported version. See the system requirements document.
      2. The existing installation cannot be set up as a partitioned database environment. This type of environment and the fast communications manager (FCM) are not supported for use with InfoSphere Data Click.
      3. Ensure that DB2 Fault Monitoring components are enabled. Ensure that the /etc/init/db2fmcd.conf file exists.
      4. Ensure that the /etc/services file contains a port entry for the instance:
        db2c_<db2_instance_owner>    <port_number>/tcp

        For example:

        db2c_db2inst1    50000/tcp
    • If you plan to use a database system that is installed on a remote computer, such as the InfoSphere BigInsights computer, create the metadata repository database.
      1. On the InfoSphere Data Click computer, use scp, ftp, or a similar tool to copy the scripts that create the metadata repository to the DB2 instance home directory on the computer where the database system is installed. For example:
        cd /tmp/dcinstall/is-bi/DatabaseSupport/UNIX_Linux/MetadataRepository
        scp ./DB2_LUW_10_1/* db2inst1@remote_hostname:/home/db2inst1

        In this example, db2inst1 is the DB2 administrator user name. If not db2inst1, substitute the appropriate name for your setup.

      2. On the computer where the database system is installed, ensure that the /etc/services file contains a port entry for the instance:
        db2c_<db2_instance_owner>    <port_number>/tcp

        For example:

        db2c_db2inst1    50000/tcp
      3. On the computer where the database system is installed, create the metadata repository database:
        1. Log in as the DB2 instance owner and navigate to the DB2 instance home directory.
        2. Edit the create_xmeta_db.sql and configure_xmeta_db_64bit.sql SQL scripts, replacing the indicated template values with the values for your desired setup. Use the default values unless they conflict with existing database object names.
          Template variables Description Example value

          @DATABASE_NAME@
          @DATABASE_ALIAS@

          The name of the metadata repository database. xmeta
          @INSTALL_ROOT@ The directory path where the database will be created. Specify a path appropriate to your file system setup. /home/db2inst1/db/xmeta
          @DATABASE_USERNAME@ The name of the operating system user that you want to have access to the database. Use the same name that you specified in step 4. xmeta
          Important: IBM certifies and validates InfoSphere Data Click using the provided database creation scripts. If you modify the scripts beyond the changes that are specified here, IBM will not support software failures that might result from the modifications. Modifications to the these scripts that do not have a functional impact might require that you make similar changes in the future to scripts that are generated by other administrative tools that depend on the information in the scripts here. You are responsible for making those changes.
        3. Run the following shell script to create the metadata repository database:
          ./create_xmeta_db.sh [logfile]
          Note: The scripts run the following commands when configuring the DB2 database system. These changes affect all databases in the DB2 instance:
          db2set DB2_INLIST_TO_NLJN=YES
          db2set DB2_USE_ALTERNATE_PAGE_CLEANING=ON
          db2set DB2_REDUCED_OPTIMIZATION=YES
        4. Edit the db2_grant_Permissions.sql SQL script, replacing the indicated template values with the values for your desired setup. Use the default values unless they conflict with existing database object names.
          Template variable Description Example value
          @DATABASE_NAME@ The name of the metadata repository database. Use the same value specified previously. xmeta
          @DATABASE_USERNAME@ The name of the operating system user that you want to have access to the staging repository schema. Use the same name that you specified in step 4. xmetasr
        5. Run the following shell script to create the staging area schema:
          ./configure_staging_area.sh [logfile]
  6. On the InfoSphere BigInsights computer, start InfoSphere BigInsights if you have not already done so. The InfoSphere Data Click installation program integrates with InfoSphere BigInsights during installation.
    cd $BIGINSIGHTS_HOME/bin
    ./start-all.sh
  7. On the InfoSphere Data Click computer, back up critical data and the following directories: /etc/services, /etc/inittab, /etc/passwd, and /etc/group.
  8. Disable firewall and antivirus software.
  9. Check if the following port numbers are available: 8443, 9443, 31531, and 31538. Learn more. If these ports are not available, the installation program will select the next available port by incrementing each port value by one until it finds one available. After installation, the assigned port numbers are listed in the resulting response file. If necessary, you can refer to the response file and configure your firewall accordingly after installation is complete.
  10. Run the installation program:
    • To run the installation program as a wizard:
      1. Start the installation program:
        cd Download_directory/is-bi
        ./setup -verbose
      2. Several messages are displayed. The program then displays a message about how to access the installation wizard:

        ======> Enter the following URL to your web browser to begin the
        installation process:
        https://hostname:8443/ISInstall

        Once you start the installation program, you access the wizard from a supported web browser.

      3. Ensure that your browser is configured to run the installation program.
      4. From a computer with a supported web browser, access the installation program by entering the URL specified above.
    • To run the installation program in silent mode with a response file follow these instructions. If you choose this method, skip to step 14 upon completion.
    Remember: Do not delete the installation log files that are written to the temporary directory. Preserve these files for troubleshooting and verification purposes.
  11. After logging into the wizard, after the Firewall Requirement page, the wizard will verify your system requirements. Correct any requirements that fail. Learn more.
  12. After the initial requirements check succeeds, proceed through the wizard to install the product. When specifying user names and passwords, follow these rules. Refer to the following information for more details on some of the wizard pages:
    IBM InfoSphere Data Click Installation Directory page
    The default is /opt/IBM/InformationServer. Subsequent documentation refers to this directory by either the default path or IS_install_path.
    WebSphere Application Server Liberty Core Installation page
    Because of convention, such as in support documents, it is recommended to accept the default port numbers unless they conflict with existing ports on your computer. The SSL information that you provide is used to create an SSL certificate for secure connections to the application server. The installation program creates the certificate for you. For the SSL subject distinguished name, you can provide information that is meaningful to your organization. The certificate information is used to provide secure communication with the application and application server. The distinguished name has the format:
    CN=Common_Name,OU=Org_Unit,O=Org_Name,C=Country_name

    Example:

    CN=analytics.example.com,OU=Sales Division,O=Example Co.,C=US
    Important: Set the Common Name (CN) field to the value of the InfoSphere Data Click host name, to be used by remote clients to access the server. As part of the SSL handshake, clients verify that the host name that is used to access the server matches the certificate CN value (or one of the values if there are multiple values).
    Specify any length of days for the certificate to be valid from the time that it is created (for example, 365). When the certificate expires, clients will no longer be able to connect until you update the certificate. Learn more.
    Database Management System Selection page
    • If you already created the metadata repository in step 5, select Use an existing repository (do not create the databases).
    • If you want the installation program to create the metadata repository on the local computer, select Create the repository in an existing DB2 instance on this computer.
    Metadata Repository Configuration page
    If you already created the metadata repository in step 5, the information that you enter on this page must match the information that you used to create the metadata repository. If you did not already create the metadata repository, the information that you enter is used to create the database owner and database.
    Staging Area Configuration page
    If you already created the staging area repository in step 5, the information that you enter on this page must match the information that you used to create the staging area repository. If you did not already create the staging area, the information that you enter is used to create the staging area owner and schema.
    IBM InfoSphere DataStage Administrator page
    The installation program uses this information to create the engine administrator user. InfoSphere Data Click makes use of this user name to access engine operations through the application server. The installation program automatically maps the InfoSphere Data Click administrator user credentials with this user.
    Operations Database Configuration page
    • If, in the Database Management System Selection page, you selected to install the metadata repository in a local database instance on this computer, then the installation program can also install the operations database in the same instance. In this case, the information you enter here will be used to create the database schema. You can choose to specify the database schema in the metadata repository database or its own database.
    • If you already created the staging area repository in step 5, such as in a database instance on a remote computer, then the information that you enter here will be registered in the metadata repository. After installation, you use the registered information to generate scripts that you run to create the operations database.
    • Whether you created the metadata repository before installation or not, you can choose to skip this page. If you do this, you are required to register the operations database and generate and run the scripts after InfoSphere Data Click is installed.
    IBM InfoSphere BigInsights Integration page
    Specify the connection information for InfoSphere BigInsights. Your InfoSphere BigInsights installation contains configuration information and files that are required by InfoSphere Data Click. When the installation program runs, it will connect to InfoSphere BigInsights to obtain the configuration information and files.

    If you connect with SSL, the hostname must be specified exactly as it is specified in the InfoSphere BigInsights certificate. If you have a multiple node InfoSphere BigInsights configuration, the certificate hostname might contain a wildcard so that it's valid for a domain. In this case, specify a hostname in the domain.

    Note: To integrate with InfoSphere BigInsights during installation, ensure InfoSphere BigInsights is started before proceeding.

    Although not recommended, you can skip this step and integrate with InfoSphere BigInsights after installation.

  13. Select Install IBM InfoSphere Data Click and save my settings in a response file and click Next to start the installation. Consider these conditions while the installation program is running:
    • Leave the shell window open until the installation is complete.
    • After you have started the installation process, keep the web browser open to monitor progress and to respond if there is an error. If you close the web browser, the installation continues to the end or until it encounters an error. You can reconnect to the session by opening a new web browser to the same URL. You can also monitor the installation process by reading the log files.
    • If the system encounters an error, it requests a response from you. Check periodically to make sure that the system is not waiting for you to respond. If the installation pauses for a long time, it might fail. Learn more.
  14. Launch the ISALite tool to verify the success of the installation.
    /opt/IBM/InformationServer/ISALite/runISALite.sh

    Once ISALite is started, run the General Diagnostic Health Checker available in the main menu. Learn more.

  15. If you skipped the InfoSphere BigInsights Integration page during installation, refer to the techdoc at http://www.ibm.com/support/docview.wss?uid=swg27040051 to integrate now. You must integrate before you can use InfoSphere Data Click.

What to do next

InfoSphere Data Click is now installed and ready to be configured.