PDS Deployment

The server runs on any supported release of z/OS. For current information about supported releases:

1. Go to http://techsupport.informationbuilders.com.

   The Information Builders Technical Support home page opens.

2. In the Quick Links section on the right side of the page, click Supported Systems/Adapters.

   The Supported Systems and Adapters page opens.

3. Click the link for the release you want.

   The Supported Systems and Adapters page for that release opens.

4. Click the link for your platform.

   The support chart for that platform opens.

In general, the operating system should have the latest cumulative patch levels applied.

Confirm that your server installation software is labeled for your operating system level.

If JVM-based adapters, server-side graphics, XBRL, or user-written CALLJAVA applications are to be used, a Java Runtime Environment (JRE) JVM must be installed on the machine, and the server must be configured to use it

The minimum JVM release level is 1.8 or higher, due to required internal components of the server. The Java Listener will not start unless 1.8 (or higher) is in use. Prior 7.x releases would allow the listener to start with any release, and sub-components would fail if they required a higher Java Level. The primary reason for this change is that Java 1.5 (and prior releases) are past their End of Service Life (EOSL) dates, and, as such, are unsupportable, in addition to lacking newer functionality. The following URL has Java EOL and EOSL information:

http://www.oracle.com/technetwork/java/eol-135779.html

Installing maintenance updates to the EDAHOME of an existing server running releases prior to production 7.7.05 will also have the requirement of moving up all dependent configurations to use Java 1.8 (as instructed in this section).

You may install a Java JRE or a Java SDK. When you install a Java SDK, the JRE component (where the JVM lives) is included, so either is allowed. If using servlet, the Java SDK is required for the jar command, so it is generally preferred over the Java JRE. The SDK or JRE build type must also match the 32-bit or 64-bit bit type of the server. If an appropriate JVM from a JRE or SDK is not found on the library path or using variables as described below, or is not the appropriate bit type, a Failed to find JVM message will be displayed. Further Java Services debugging information about loading the JVM will be written to the server start log indicating JSCOM3 start failed as well as additional information that may be useful in resolving the problem. JSCOM3 is the actual process name for the Java Services Listener and the terms are often used interchangeably.

The JDK JRE bin and server (or client) subdirectories must be specified in the load library path environment variable. A server restart is required, plus the appropriate JVM must be on the path if switching JRE levels. The load library path will be prompted at install time if JVM-based adapters or features are required. Otherwise, it can be manually set by editing the EDAENV file using any of the following methods.

• For Java JDK, set JDK_HOME (to the install home location) in the server environment configuration file (EDAENV).
• For Java JRE, set JAVA_HOME (to the install home location) in the server environment configuration file (EDAENV).
• Use library path (LIBPATH) to set explicit pathing. Use of JDK_HOME or JAVA_HOME is preferred as they are less prone to error. The JRE bin and server (or client) subdirectories must be specified in a path-based environment variable and a server restart is required.

To change or add operating system environment variables, set and export the variable in a .profile or script that always gets called during a server start. It is very common to place variables in the server edastart script, but it is recommended that they be placed in a script that in turn calls edastart (so that the edastart script remains vanilla).

To change or add a variable in a server environment start up file (EDACONF bin\edaenv.cfg), either edit the file in a text editor before starting the server or:

1. Start the server (services like Java Listener may fail until configured and the server is restarted).
2. Open the Web Console and log on using an administrator ID.
3. Select Workspace from the main menu.
4. In the navigation pane, open the Configuration Files and Miscellaneous folders.
5. Right-click Environment - edaenv.cfg and select Edit.
6. Make the desired edit.
7. Save the file.
8. Restart the server (changes are not effective until server is restarted).

The format of edaenv.cfg variables is one per line in name=value pairs. Spaces before and after the equal sign are optional. Values with embedded spaces do not require quoting. Variables are always uppercase.

If JVM-based adapters or features are not required, and the JVM environment is not configured, the message Failed to find JVM is normal and can be ignored.

To add classes to the JVM class path for customer-written CALLJAVA applications, set and export the CLASSPATH variable to the operating system level before server start-up or use the Web Console to set the Java Listener IBI_CLASSPATH property.

The install process prompts for two IP port numbers: the TCP Listener and HTTP Listener. It also uses the next two consecutive ports after the supplied HTTP Listener port for FDS use. This results in a total of four IP ports.

The supplied IP port numbers must be above the IANA registered well-known reserve range (numbers under 1024) and not over the maximum legal number (65535). Do not use IP port numbers already used by other applications or products. Netstat, or netstat like commands, should reveal what actual ports are in use.

The Web Console requires one of the following web browsers:

• Microsoft Internet Explorer® 11 or higher.
• Mozilla Firefox® 59 or higher.
• Google Chrome® 65 or higher.

The server disk space requirements are:

Note: Deferred Execution Datasets are not created by the install procedure. They are created when the Scheduler/Deferred service starts. By default, the Scheduler/Deferred service auto starts at server startup. To disable the feature, enter dfm_autostart = n in the EDASERVE administration control file.

where:

high_level_qualifier

Is the high-level qualifier to be used for all output libraries. You specify the high-level qualifier during server installation, as described in Step 4. Run ISETUP, in Step 4.

product_type

Is one of the following:

FFS	for a Full-Function Server
DM	for a DataMigrator Server
WFS	for a WebFOCUS Reporting Server
WFM	for a Shared Application Server for WebFOCUS Maintain

approot

Is the default location for storing applications. You specify approot during server installation, as described in Step 4. Run ISETUP,

Memory usage of a configured environment consists of the following elements:

• Workspace Manager
• Listeners
• Concurrently running application agents

Actual memory usage depends on application features, and varies depending on the application. The SHRLIBRGNSIZE parameter (defined on SYS1.PARMLIB, member BPXPRMxx) can affect the amount of memory that the server address space will allocate. For SHRLIBRGNSIZE, we recommend the default MVS installation value of 64Mb:

SHRLIBRGNSIZE(67108864)

Server memory usage:

• The workspace (including Listeners) uses 200 megabytes.
• Each pre-started agent requires 4 megabytes.

The minimum amount of memory for a newly installed server with no workload is 250Mb. However, depending on usage, workload, and configuration options, 500Mb is recommended to start, to be adjusted as needed.

You need four TCP/IP ports for each server instance that you configure. Three of these ports must be consecutive. You specify these port numbers during installation. You may require additional ports depending on which options you configure later.

The server supports only IBM TCP/IP. It does not support Interlink or any other third-party TCP/IP.

PDS deployment requires each user of the server to be identified to USS by means of a default segment definition. This default OMVS segment is defined when USS is installed as a part of z/OS. Please refer to your IBM UNIX System Services documentation for more information about this subject.

Libraries and the APIs supporting them must reside in the HFS file system to enable the following features:

• Server-side graphics.

They are also required for the following Java-based and SAP-based adapters:

• Call Java.
• EJB.
• JDBC.
• Microsoft SQL Server.
• SQL SAP.
• SAP BWB.

At installation time, a panel with a list of adapters to be configured will be displayed. If any of the above adapters are selected, the installation will require the path for two HFS locations as follows:

• edahome_dir. Provide the edahome path for the dll modules that interface with Java/SAP and must reside in HFS. The directory will be created with 755 permissions, if it does not exist.
• edaconf_dir. Provide the edaconf path to be used for both configuration files (such as codepage files), as well as the root location for temporary files (such as traces and logs), that must reside in HFS. If it does not exist, the directory will be created with 755 permissions.

If no data adapters were selected, the next panel will allow an optional configuration for the JSCOM3 listener. If server-side graphics support is desired, the listener must be configured and all three paths are required (edahome_dir, edaconf_dir, plus the path to Java passed to either JDK_HOME or JAVA_HOME).

If edahome_dir is not defined at installation time, it will not be possible to configure it later using the Web Console. The server will have to be re-installed, configuring the JSCOM3 listener.

You can use any user ID to install and run the server. Whichever ID you use becomes the first server administrator ID (sometimes referred to as iadmin).

For current information about which adapters are supported:

1. Go to http://techsupport.informationbuilders.com.

   The Information Builders Technical Support home page opens.

2. In the Quick Links section on the right side of the page, click Supported Systems/Adapters.

   The Supported Systems and Adapters page opens.

3. Click the link for the release you want.

   The Supported Systems and Adapters page for that release opens.

4. Click the link for your platform.

   The support chart for that platform opens.

You must provide information to configure the adapters that you are licensed to install. The installation procedure automatically prompts you for this information. When you are prompted for an optional steplib, ddname, or environment variable, the installation procedure will indicate this with an OPT> prompt.

After you have installed and configured the server, you will be able to further configure your adapters using a web-based server configuration tool called the Web Console.

The following table describes what information you need to provide for each adapter that you have. (If an adapter is not listed, no information needs to be provided for it.) Note that the table refers to:

• IRUNJCL. This procedure contains the JCL procedure for the server, and is a member of the configuration library

  high_level_qualifier.PDS.product_type.DATA

  where:

  high_level_qualifier

  Is the high-level qualifier to be used for all output libraries. You specify the high-level qualifier during server installation, as described in Step 4. Run ISETUP, in Step 6.

  product_type

  Is one of the following:

  FFS	for a Full-Function Server
  DM	for a DataMigrator Server
  WFS	for a WebFOCUS Reporting Server
  WFM	for a Shared Application Server for WebFOCUS Maintain

You can choose to access the server installation software using either:

• Tape. The software is provided on a 3490 or 3590 cartridge.

  You must unload the installation data set from the tape before you can run the installation. This is how most installations are performed.

• FTP. You download the installation software from the Information Builders download site.

  Downloading the installation software involves:

  1. Registering at the Information Builders download site.
  2. Downloading the installation data set from the site.
  3. Running the isetup procedure to complete the download process and install the software.

Server installation consists of a series of ISPF panels, which gather the required information. After the panel dialog is complete, JCL is created and submitted to install the server on z/OS. This JCL job retrieves the remainder of the data sets from the media and configures a basic working server.

1. Execute the ISETUP member of your high_level_qualifier.HOME.DATA using ISPF option 6.

   The first Installation and Configuration panel opens.

   
2. Type 2 and press Enter to continue to the next panel.

   The following panel opens.

   

3. Complete the panel as directed in the following table.

   Field	Instructions
   Enter selection	Accept the default value 1, Install and Configure, for a new installation. For option 2, Add Additional Configuration Instance, see Adding a Configuration Instance for PDS. For option 3, Refresh Installation, see Upgrading Your Server Release for PDS.
   Enter License Key	Enter the license key that was provided with the software. Be sure to store this key in a safe place for future reference.
   Input source	Enter the input source: T for Tape - If you received your software on tape media. D for Disk - If you selected manual download from the download instructions. F for FTP - If you selected automatic download from the download instructions.
   Installation Userid	Shows the current logon ID. It cannot be changed.
   PTH Administrator Userid	An ID is required to administer the server immediately after initial installation. This ID is defined and maintained solely in the realm of the server. It defaults to srvadmin, and it can be changed here. For more information about running the server in secure mode, see Step 6. Configure Server Security.
   PTH Administrator Password	Password for the PTH Administrator ID. It cannot be left blank and must be matched at Retype field.
   Enter Job Card information	To provide JOB card information for submitting jobs to the JES queue, provide a valid job name (a maximum of seven characters following the // on the first JCL line), which defaults to the user ID that you are currently using. This job name is used for multiple submissions (for example, jobnameA, jobnameB, jobnameC, and so on) in the JCL generated by the installation procedure.
   Override JOB name checking	To provide your own JOB card information, including JOB name, enter Y and provide valid JOB card information in the Enter Job Card information field. The JOB card information that you enter will be used for each job that is submitted.

4. Press Enter to continue to the next panel.

   This following panel only appears if FTP was previously selected. Otherwise, skip to Step 6.

   
5. Complete the panel as follows.

   Field	Instructions
   HLQ for downloaded files	This value defaults to the high level qualifier of the HOME.DATA file that ISETUP is running from. If necessary, change the value to any other high level qualifier that is to be used to create the uncompressed version of the FTP files.
   FTP Userid	Cut and paste from the download instructions.
   FTP Password	Cut and paste from the download instructions.

6. Press Enter to continue to the next panel.

   Note that in the current panel (and some later panels), if you are running ISETUP from:

   • high_level_qualifier.HOME.DATA, the panel will display default values for some fields.
   • Any other library, the panel will not display any default values.

   In this and some later panels, you can see a field's default value (if one exists) by blanking out the field and pressing Enter.

   

   Complete the panel as follows.

   Field	Instructions
   Input Media (installing from tape only)
   Volume serial number	Provide the volume serial number of the server media. The number is located on the tape supplied in you server package.
   Unit type	Review the default value and change if necessary.
   Work unit type	Review the default value and change if necessary. You can specify a UNIT= type value (for example, SYSDA), or you can direct work files to a specific volume serial number by specifying, in single quotation marks ('), 'SYSDA,VOL=SER=volume'.
   Input Media (installing from disk/FTP only)
   Input Libraries HLQ (EDAHOME)	This is the high-level qualifier that you had specified when you manually downloaded the installation software from the download site. This is an input field if Disk input source was previously selected. Otherwise, it is a protected field.
   Copy to runtime libraries	If you want to use the downloaded installation software as a backup, and create a new copy from which to run, enter Y. If the Input Libraries HLQ and the Output Libraries HLQ are the same, this value will be ignored and no copy will take place. Otherwise, accept the default N to run from the downloaded software.
   General Installation Parameters
   Output Libraries HLQ	This is the high-level qualifier that the installation procedure will use to allocate output libraries.
   Unit/Type	These show the values that the installation process will use to allocate the output libraries. If necessary, you can change these to site-specific values. Type can be VOL=SER (default), DATACLAS, MGMTCLAS, or STORCLAS.
   Approot value	This is where application components will reside. Note that this high-level qualifier must differ from the output libraries high-level qualifier (EDACONF) that you entered at the top of the panel. To specify a different qualifier for application components, change the value for this field. It can be up to 21 characters.
   HTTP Listener Port	This is the port number that the server will use for HTTP. It is the first of three connection ports that must be available to the server. For example, if you choose port 8101, then ports 8101, 8102, and 8103 are used by the server. Ensure that you choose ports that are not currently being used.
   TCP Listener Port	This is the port number of the TCP Listener. The default is one less than the port specified for the HTTP Listener, but it can be any port number other than the three reserved for HTTP.

7. Press Enter to continue to the next panel.

   Depending on your license key, the Data Adapter panel may open. If the Data Adapter panel opens, continue with Step 8; otherwise, skip to Step 9.

8. The Data Adapter panel lists adapters that require the allocation of MVS libraries in IRUNJCL or environment variables in the EDAENV member. To select specific adapters:
   1. Type Y next to the required adapters and press Enter.
   2. Supply the requested information, which is described in Step 2. Collect Required Information for Adapters.

      After you have finished installing and configuring the server, you can use the Web Console to finish configuring these adapters, and to configure adapters that do not have MVS JCL requirements.

9. Press Enter to continue to the next panel.

   The JSCOM3 Listener configuration panel opens.

   1. The panel will prompt for the path to the Java environment to be passed to either JDK_HOME or JAVA_HOME, as described in JVM Requirements for Java Services, and it will also prompt for edahome_dir and edaconf_dir, as described in HFS Home and Configuration Directory Requirements.
   2. Configuration of the JSCOM3 Listener is either optional or mandatory depending on which adapters were selected. If any Java-based adapters were selected (EJB, Call Java, JDBC, MS SQL Server), the configuration of all three paths listed above is mandatory. If SAP-based adapters were selected (SAP or SAP BW), only edahome_dir and edaconf_dir are required.
   3. If no Java-based or SAP based adapters were select, this configuration might still be desirable to enable the server-side graphics feature. To skip the configuration, leave the path blank.
10. Press Enter to continue to the next panel.

    The confirmation panel opens.

    

11. If you wish to review a list of the data sets to be allocated, type Y in the Review output allocations field and press Enter.

    A panel opens listing the data sets. You may need to page down to see the entire list. Press Enter when you are done to return to the confirmation panel.

12. Ensure that all values on the Confirmation panel are correct, then select one of the following options
    • N to return to the initial panel so that you can change installation values.
    • C to create JCL which you can submit at a later time. The JCL is placed in your high_level_qualifier.PDS.product_type.DATA configuration library.
    • S to create JCL in high_level_qualifier.PDS.product_type.DATA, and submit the job immediately.

    Note: If FTP was selected, JCL will be created to download the software and run the install and configuration process.

    where:

    high_level_qualifier

    Is the high-level qualifier to be used for all output libraries. You specified the high-level qualifier during installation, as described in Step 4. Run ISETUP, in Step 4.

    product_type

    Is one of the following:

    FFS	for a Full-Function Server
    DM	for a DataMigrator Server
    WFS	for a WebFOCUS Reporting Server
    WFM	for a Shared Application Server for WebFOCUS Maintain

13. As the job is processed, in SDSF, check JESLOG for errors and return codes.

Following is a table of the jobs created. All members are created in the configuration library (as described in Step 11).

The following members all call procedure IRUNJCL, which is the main server JCL. If you need to change the server JCL, change member IRUNJCL.

The following members contain batch JCL for auxiliary functions, and are also created in the configuration library.

The following members contain sample started task JCL, and are also created in the configuration library.

To test the server installation:

1. Log on to TSO as iadmin.
2. Submit the ISTART JCL from the configuration library to start the server. This executes the IRUNJCL proc. The configuration library is

   high_level_qualifier.PDS.product_type.DATA

   where:

   high_level_qualifier

   Is the high-level qualifier to be used for all output libraries. You specified the high-level qualifier during server installation, as described in Step 4. Run ISETUP, in Step 6.

   product_type

   Is one of the following:

   FFS	for a Full-Function Server
   DM	for a DataMigrator Server
   WFS	for a WebFOCUS Reporting Server
   WFM	for a Shared Application Server for WebFOCUS Maintain

3. Check the job output for errors. Look for the EDAPRINT message:

   (EDA13023) ALL INITIAL SERVERS STARTED

4. Start the Web Console by opening a browser pointed at the listener port of the server. The URL format is

   http://host:port

   where:

   host

   Is the name of the machine on which the product is installed.

   port

   Is one port higher than the port specified when installing the server. For example, if you specified port 8100 during installation, then use port 8101 to access the Web Console.

   The Web Console opens.

5. If the Web Console opens and displays application tree folders in the left pane, the server is working because it uses its own underlying data access and reporting technologies to visualize the application tree.
6. Continue with adapter configuration, as described in the Adapter Administration manual.

When you are finished using the server, you can use the Web Console to stop the server by going to the Web Console menu bar, selecting Workspace, and then Stop.

If you experience problems at start up, examine the job output for more information.

If you will be configuring security provider OPSYS, you must perform the instructions in How to Configure Security With All Security Products, regardless of which security product you use. (For security providers PTH, DBMS, and LDAP, skip these topics.)

For a full description of all security providers:

1. From the Web Console menu bar, select Help, then Contents and Search.

   The Web Console Help window opens.

2. In the left pane expand Server Administration.

Alternatively, see the Server Administration manual.

To start the server, submit the ISTART member of the MVS configuration library (high_level_qualifier.product_type.DATA) for your server.

ISETUP creates started task JCL to start the server. This started task in a member of the MVS configuration library is IWAYS.

In order to execute the started task, you must:

• Copy it into SYS1.PROCLIB or any other JES2 Proclib data set.
• Satisfy security requirements. All external security-related permissions must exist for both the data sets and the started tasks. In order to issue the started tasks, the user must satisfy both of the following requirements:
  • Have at least OPERATOR authority defined within the Web Console.
  • Be in the same security group, or associated with the same security group, as the owner of the server directory structure (for example, as iadmin).

To submit the started task from the MVS console, issue the following command:

S IWAYS

You can add the started task to any automation product that you run.

You can stop the server using any of the following methods:

• Web Console. From the Web Console menu bar select Workspace and then Stop.
• MVS operator command. On the MVS Console or SDSF, issue the following operator MODIFY command:

  F jobname, -stop

  where:

  jobname

  Is the job under which the server is running.

• Cancel the server job. In SDSF, cancel the job.

To add a configuration instance, perform the following steps.

1. Execute ISETUP again. You should have a high_level_qualifier.HOME.DATA PDS unloaded from the installation tape. Use option 6 in ISPF to execute the ISETUP member of this PDS.

   Note: If this PDS is not available, run an IEBCOPY job to allocate and unload it from the installation tape.

   The first Installation and Configuration panel opens.

   
2. Type 2 and press Enter to continue to the next panel.

   The following panel opens.

   

3. Complete the first Installation and Configuration panel as follows.

   Field	Instructions
   Enter selection	Choose option 2, Add Additional Configuration Instance.
   Enter License Key	Enter the license key that was provided with the software for the type of server instance that you want to configure (for example, the license key for a WebFOCUS Reporting Server or for a Full-Function Server).
   Input source	This is ignored for option 2, adding a configuration instance.
   Installation Userid	Shows the current logon ID. It cannot be changed.
   PTH Administrator Userid	An ID is required to administer the server immediately after initial installation. This ID is defined and maintained solely in the realm of the server. It defaults to srvadmin, and it can be changed here. For more information about running the server in secure mode, see Step 6. Configure Server Security.
   PTH Administrator Password	Password for the PTH Administrator ID. It cannot be left blank and must be matched at Retype field.
   Enter Job Card information	To provide JOB card information for submitting jobs to the JES queue, provide a valid job name (a maximum of seven characters following the // on the first JCL line), which defaults to the user ID that you are currently using. This job name is used for multiple submissions (for example, jobnameA, jobnameB, jobnameC, and so on) in the JCL generated by the installation procedure.
   Override JOB name checking	To provide your own JOB card information, including JOB name, enter Y and provide valid JOB card information in the Enter Job Card information field. The JOB card information that you enter will be used for each job that is submitted.

4. Press Enter to continue to the next panel.

   The following panel opens.

   
5. Enter the current base high-level qualifier used for EDAHOME.

   This indicates where to install the configuration (EDACONF) and where the binaries (EDAHOME) are installed. The installation procedure checks whether the required set of EDAHOME data sets exist. If the test fails, you receive a message indicating the failure and available options.

6. Press Enter to continue to the next panel.

   If you are configuring the first instance of a given product type, the following panel opens.

   

   Otherwise, if you are configuring an additional instance of a given product type, the following panel opens. In this sample panel, two Full-Function Server configuration instances (FFS and FFS1) already exist, and a third (FFS2) is being added.

   

7. Complete the panel as follows.

   Field	Instructions
   Configuration Parameters
   Approot value	This indicates where application components will reside for this configuration. The default value is based on the value specified for Current Base HLQ (EDAHOME) on the previous panel. To specify a different location for application components, change the value of this field. Different configurations which use the same base HLQ (high level qualifier) libraries (EDAHOME) can share the same approot value (that is, the same application files). Alternatively, they can use different approot values so that they have different sets of application files. If you specify the approot value of an existing server configuration, the installation process will recreate the server supplementary data sets and sample files (see Disk Space Requirements). If you do not want them to be recreated, provide a different value for approot.
   Output Libraries HLQ (EDACONF)	You are prompted for this information only if you are configuring the first instance of a product type (for example, if you are configuring the first instance of a WebFOCUS Reporting Server). This is the high-level qualifier that the installation procedure will use to allocate output libraries, that is, to allocate the configuration libraries for this server instance. This high-level qualifier is also known as EDACONF.
   Unit/Type	You are prompted for this information only if you are configuring the first instance of a product type (for example, if you are configuring the first instance of a WebFOCUS Reporting Server). These show the values that the installation process will use to allocate the output libraries. If necessary, you can change these to site-specific values. Type can be VOL=SER (default), DATACLAS, MGMTCLAS, or STORCLAS.
   Field	Instructions
   EDACONF suffix	You are prompted for this information only if you are configuring an additional instance of a product type (for example, if you are configuring a second instance of a WebFOCUS Reporting Server). Each instance must have its own set of configuration libraries. To guarantee this, and to prevent a new set of configuration libraries from overwriting an existing set, the suffix that you specify here will be appended to the name of the product type qualifier. For example, if you are configuring the second instance of a WebFOCUS Reporting Server, you could specify that the suffix "1" be added, so that the EDACONF high-level qualifier would be: IADMIN.SRV77.PDS.WFS1.DATA You can add a new configuration as a numeric or string suffix to the base product type. If you supply a string, the installation procedure ignores any numeric suffix. For a: Numeric suffix, enter a digit between 1 and 9. This will be added to the product type in the directory name and library name to distinguish it from other configuration instances. String suffix, enter a string of between 1 and 5 characters (for example, TEST, PROD, or DEV). The string cannot contain embedded spaces. You can also use the string suffix to extend the numeric numbering past 9. Just supply a number greater than 9. If you change the suffix value, when you press Enter, the panel refreshes with a new value for EDACONF Library.
   HTTP Listener Port	This indicates the port number that the server will use for HTTP. It is the first of three connection ports that must be available to the server. For example, if you choose port 8101, then ports 8101, 8102, and 8103 are used by the server. Ensure that you choose ports that are not currently being used.
   TCP Listener Port	This is the port number of the TCP Listener. The default is one less than the port specified for the HTTP Listener, but it can be any port number other than the three reserved for HTTP.

8. Press Enter to continue to the next panel.

   Depending on your license key, the Data Adapter panel may open before the confirmation panel. If the Data Adapter panel opens, continue with Step 9; otherwise, skip to Step 10.

9. The Data Adapter panel lists adapters that require the allocation of libraries in IRUNJCL or environment variables in the EDAENV member. To select specific adapters:
   1. Type Y next to the required adapters and press Enter.
   2. Supply the requested information, which is described in Step 2. Collect Required Information for Adapters.

      After you have finished installing and configuring the server, you can use the Web Console to finish configuring these adapters, and to configure adapters that do not have JCL requirements.

10. Press Enter to continue to the next panel.

    The JSCOM3 Listener configuration panel opens.

    1. The panel will prompt for the path to the Java environment to be passed to either JDK_HOME or JAVA_HOME, as described in JVM Requirements for Java Services, and it will also prompt for edahome_dir and edaconf_dir, as described in HFS Home and Configuration Directory Requirements.
    2. Configuration of the JSCOM3 Listener is either optional or mandatory depending on which adapters were selected. If any Java-based adapters were selected (EJB, Call Java, JDBC, MS SQL Server), the configuration of all three paths listed above is mandatory. If SAP-based adapters were selected (SAP or SAP BW), only edahome_dir and edaconf_dir are required.
    3. If no Java-based or SAP based adapters were select, this configuration might still be desirable to enable the server-side graphics feature. To skip the configuration, leave the path blank.
11. Press Enter to continue to the next panel.

    The confirmation panel opens.

12. Ensure that all values on the Confirmation panel are correct, then select one of the following options:
    • N to return to the initial panel so that you can change installation values.
    • C to create JCL which you can submit at a later time. The JCL is placed in your configuration library.
    • S to create JCL and submit the job immediately.
13. As the job is processed, validate the installation as described in Step 2. Test the New Configuration Instance.

To test the configuration instance that you just added:

1. Log on to TSO as iadmin.
2. Submit the ISTART JCL from the configuration library to start the server. This executes the IRUNJCL proc. The configuration library is

   high_level_qualifier.PDS.product_type[suffix].DATA

   where:

   high_level_qualifier

   Is the high-level qualifier to be used for all output libraries. You specified the high-level qualifier during server installation, as described in Step 4. Run ISETUP, in Step 6.

   product_type

   Is one of the following:

   FFS	for a Full-Function Server
   DM	for a DataMigrator Server
   WFS	for a WebFOCUS Reporting Server
   WFM	for a Shared Application Server for WebFOCUS Maintain

   suffix

   If you are testing an additional instance of a product type for which an earlier configuration exists, the new configuration library product type qualifier will have a suffix (for example, FFS1 or FFSDEV). The suffix distinguishes the new configuration library from the original one.

3. Check the job output for errors. Look for the EDAPRINT message:

   (EDA13023) ALL INITIAL SERVERS STARTED

4. Start the Web Console by opening a browser pointed at the listener port of the server. The URL format is

   http://host:port

   where:

   host

   Is the name of the machine on which the server is installed.

   port

   Is one port higher than the port specified when installing the server. For example, if you specified port 8100 during installation, then use port 8101 to access the Web Console.

   The Web Console opens.

5. If the Web Console opens and displays application tree folders in the left pane, the server is working because it uses its own underlying data access and reporting technologies to visualize the application tree.
6. Continue with adapter configuration, as described in the Adapter Administration manual.

You can choose to access the server installation software using either:

• Tape. The software is provided on a 3490 or 3590 cartridge.

  You must unload the installation data set from the tape before you can run the installation. This is how most installations are performed.

• FTP. You download the installation software from the Information Builders download site.

  Downloading the installation software involves:

  1. Registering at the Information Builders download site.
  2. Downloading the installation data set from the site.
  3. Running the isetup procedure to complete the download process and install the software.

Note: The above FTP steps are for the AUTOMATIC download instructions to MVS. If you downloaded the software to another platform, then transferred the files to MVS (following the manual FTP instructions), a new set of EDAHOME Libraries will already exist on MVS, therefore there will be no need to run ISETUP. The only process that needs to occur is the copying of the DB2VPR members from the new HOME.DATA to the current HOME.DATA of the release to be upgraded. Complete the upgrade by following steps 3-5 in Step 2. Run ISETUP.

Software upgrade consists of a series of ISPF panels, which gather information for the upgrade. After the panel dialog is complete, JCL is created and submitted (if required) to create a new set of EDAHOME libraries.

1. Execute the ISETUP member of your high_level_qualifier.HOME.DATA using ISPF option 6.

   The Installation and Configuration panel opens.

2. Select 2 for PDS deployment and press Enter to continue to the next panel.
3. Complete the panel as follows.

   Field	Instructions
   Enter selection	Choose option 3, Refresh Installation.
   Enter License Key	Enter the 10-digit license key that was provided with the software.
   Input source	Choose the Input source, T for tape or F for automatic FTP download direct to MVS.
   Installation Userid	Shows the current logon ID. It cannot be changed.
   PTH Administrator Userid	An ID is required to administer the server immediately after initial installation. This ID is defined and maintained solely in the realm of the server. It defaults to srvadmin. Note: For a Refresh Installation, this parameter is ignored, as no configuration files are updated. ISETUP must be run by the OPSYS Administration userid.
   PTH Administrator Password	Password for the PTH Administrator ID. Note: For a Refresh Installation, this parameter is ignored, as no configuration files are updated. ISETUP must be run by the OPSYS Administration userid.
   Enter Job Card information	To provide JOB card information for submitting jobs to the JES queue, provide a valid job name (a maximum of seven characters following the // on the first JCL line), which defaults to the user ID that you are currently using. This job name is used for multiple submissions (for example, jobnameA, jobnameB, jobnameC, and so on) in the JCL generated by the installation procedure.
   Override JOB name checking	To provide your own JOB card information, including JOB name, enter Y and provide valid JOB card information in the Enter Job Card information field. The JOB card information that you enter will be used for each job that is submitted.

4. If you selected input source T, skip to Step 5.

   If you selected F for automatic FTP only, complete the panel as follows.

   Field	Instructions
   HLQ for downloaded files	High level qualifier to be used as the target for the FTP files. If this is the same value as "Current Base HLQ" (see below) then files will be overwritten. Two warning messages will be given and the enter key must be pressed to continue.
   FTP Userid	User ID provided on the download instructions.
   FTP Password	Password provided on the download instructions.

5. Press Enter to continue to the next panel, and complete the panel as follows.

   Field	Instructions
   Current Base HLQ (EDAHOME)	High level qualifier of the current server that is to be refreshed. This HLQ will be used to check if a full set of EDAHOME libraries exist. From this value, the current installation library name is obtained and this will be the location used to create the refresh JCL.

6. Press Enter to continue to the next panel, and complete the panel as follows.

   Field	Instructions
   Input Media (installing from tape)
   Volume serial number	Provide the volume serial number of the server media. The number is located on the tape supplied in you server package.
   Volume unit type	Review the default value and change it if necessary.
   Work unit type	Review the default value and change if necessary. You can specify a UNIT= type value (for example, SYSDA), or you can direct work files to a specific volume serial number by specifying, in single quotation marks ('), 'SYSDA,VOL=SER=volume'.
   Output Libraries
   Output Libraries HLQ	For tape input only: High level qualifier that will be used to allocate and load a full set of EDAHOME libraries. Change the value as necessary. If this is the same value as "Current Base HLQ" (see previous page) then files will be overwritten. Two warning messages will be given and the enter key must be pressed to continue.
   HLQ for downloaded files	If FTP is selected, this "HLQ for downloaded files" was provided on the previous panel.
   Unit/Type	Values that the installation process will use to allocate the EDAHOME libraries on MVS. If necessary, you can change these to site-specific values. Type can be VOL=SER (default), DATACLAS, MGMTCLAS, or STORCLAS.
   Refresh edahome_dir?	Select Y/N to refresh the contents of the edahome_dir directory.
   edahome_dir	If Y was selected above, enter the path to edahome_dir.
   RA Active?	If you are refreshing a 7.7, 7.7.01, or 7.7.02 installation, and Resource Management is configured, select Y to recreate RMLDATxx datasets, otherwise select N. The default value is N.

   Note: The installation JCL library name is where the refresh JCL will be created. This library is the current software installation library. The value cannot be changed.

7. Ensure that all values on the panel are correct, then select one of the following options:
   • N to return to the initial panel so that you can change installation values.
   • C to create JCL which you can submit at a later time. The JCL is placed in your configuration library.
   • S to create JCL and submit the job immediately.
8. As the job is processed, in SDSF, check JESLOG for errors and return codes.

   The following jobs are added to the current server configuration library:

   Job	Description
   ISETUPJ3 ISOPTS3	Main JCL Job stream that is used to install the software.

To test the installation:

1. Log on to TSO as iadmin.
2. Using a test server, replace all the EDAHOME libraries referenced in IRUNJCL with the new set.
3. Submit the ISTART JCL to start the server.
4. Check the job output for errors. Look for the EDAPRINT message:

   (EDA13023) ALL INITIAL SERVERS STARTED

5. Start the Web Console by opening a browser pointed at the listener port of the server. The URL format is

   http://host:port              

   where:

   host

   Is the name of the machine on which the server is installed.

   port

   Is one port higher than the port specified when installing the server. For example, if you specified port 8100 during installation, then use port 8101 to access the Web Console.

   The Web Console opens.

6. If the Web Console opens and displays application tree folders in the left pane, the server is working because it uses its own underlying data access and reporting technologies to visualize the application tree.

When you are finished using the server, you can use the Web Console to stop the server by going to the Web Console menu bar, selecting Workspace, and then Stop.

If you experience problems at start up, examine the job output for more information.

For information about configuring server security, see Step 6. Configure Server Security.

To reconfigure server security to OPSYS provider only:

1. Log on to TSO.
2. The libraries allocated to STEPLIB in IRUNJCL must be APF-authorized. Any non APF-authorized libraries must be allocated the TASKLIB DDNAME.
3. Test server security by repeating the process described in Step 3. Test the Server Installation.

While most adapters do not require additional steps after updating binary files, the following table notes the adapters that do require some consideration.

Though physically identical to a Central Processor, the zIIP engine is microcoded at installation time to run specific types of workloads. The Central Processor continues to handle the operating system, I/O interrupts and timer interrupts, job initiations, and user interactions with the operating system. The zIIP concentrates on CPU intensive workloads, leaving the Central Processor more time to absorb otherwise queued workloads, thereby achieving some overall performance improvement across all mainframe activity.

This section describes steps and requirements for the server use of the zIIP processor.

The steps to server zIIP enablement are:

1. If your site has purchased the zIIP feature, obtain a license code that licenses you to use the zIIP feature.
2. Obtain APF authorization for the server load library.
3. Activate the zIIP feature using the SET ZIIP=ON or SET ZIIP=ON/SIMMAXZIIP command. For instructions, see Activating a zIIP Environment or Projecting zIIP Usage.

The last step in zIIP enablement is to activate the use of the zIIP processor in the server. zIIP enablement is activated by the SET ZIIP command.

The SET ZIIP command has three options:

• OFF. This setting prevents the server from offloading its processing to a zIIP.
• ON. This setting causes the server to offload processing to a zIIP engine if you have a zIIP processor and the environment is properly APF-authorized.
• ON/SIMMAXZIIP. This setting enables you to project zIIP processing in two different environments:
  • You do not have a zIIP processor. Using this setting along with the PROJECTCPU parameter, you can project how much server workload would have been offloaded to a zIIP.
  • You do have a zIIP processor. Using this setting you can project how much advantage you would achieve by offloading 100% of eligible server processing to the zIIP.

The server diverts eligible workload to the zIIP engine by switching from TCB (Task Control Block) mode for workloads that can run only on a Central Processor to SRB (Service Request Block) mode for execution of enabled workloads on the zIIP engine.

Types of server processing that are offloaded to the zIIP engine include:

• Computations.
• Aggregation.
• Screening.
• Sorting.
• Report formatting and styling.
• Transaction Processing.

The server zIIP Monitor detects situations in which the overhead cost of zIIP usage is exceeding the CPU benefits gained. When this threshold is reached, the server may decide to suspend use of the zIIP for the duration of a logical phase of the server request. When it does so, it places a message to that effect in the JES log. It then resets to make the zIIP processor accessible to the next logical phase of the server request.

TABLE, MATCH, MODIFY, and MORE requests may suspend and resume more than once as they progress through logical phases of execution.

In every case, the server attempts to optimize the use of the zIIP and minimize chargeable CPU costs.

Applications that perform significant database I/O, high-volume sorting, or the use of third party tools or user functions during processing require switching out of SRB (zIIP) mode into TCB (non-zIIP) mode to communicate, and then back again to continue processing. Although each switch is miniscule, the cumulative effect can absorb measurable amounts of CPU time on both the zIIP engine and the Central Processor.

In order to diminish this effect, the server buffers the collection of records passed to the system sort utility and some adapters rather than passing one record at a time, thus reducing the number of switches between TCB and SRB modes.

These third party products may themselves be zIIP enabled and may offload some or all of their processing to the zIIP engine independent of the server. The server always calls these products from the Central Processor because it cannot know whether they will perform any processing that is prohibited on the zIIP.

Even though zIIP usage occurs more frequently on non-optimized requests to a relational data source, optimized requests are still inherently more efficient and, therefore, may incur less CPU time. Being zIIP enabled, Db2 may also take advantage of the zIIP processor for server requests based on the local configuration of Db2 relative to the server.

Requests against some types of data sources whose I/O can be buffered gain a lot of advantage from zIIP enablement. Data sources that gain the most benefit from zIIP processing due to buffered I/O include:

• Blocked flat files.
• FOCUS.
• XFOCUS.
• VSAM.
• Db2.

In order to evaluate server zIIP processing in a session, you can query three Dialogue Manager variables that accumulate statistics about CPU processing:

• &FOCCPU accumulates the time spent on a Central Processor. This is an existing variable that precedes zIIP enablement.
• &FOCZIIPCPU accumulates the time actually spent on the zIIP processor (in SRB mode). This is the normalized CPU value using the same scale as &FOCCPU.
• &FOCZIIPONCP accumulates the time that processing could have been offloaded to the zIIP processor but was diverted to the Central Processor by the system.

  Note:

  • &FOCCPU includes the value of &FOCZIIPONCP.
  • The sum of &FOCZIIPCPU and &FOCCPU represents the total CPU utilized by the server agent.
  • If you set ZIIP=OFF, the zIIP variables do not accumulate further but are not reset to zero. If you later set ZIIP=ON, they resume accumulating statistics.

The RM (Resource Manager) that monitors server usage also captures zIIP statistics.

It is possible to specify DYNAM commands in member SRVINIT of the data set referenced by //EDACCFG DD in IRUNJCL. These commands will be executed during server startup and will be in effect until the server is shut down. You can execute the following DYNAM commands from SRVINIT:

• DYNAM SET APP FOR  filetype [SKIP}CREATE] [POSTFIX a.b] [ parms]

  Specify the types of component files that are skipped or created for the application when an APP CREATE command is issued. By default, all component file types are generated.

  where:

  filetype

  Are the component types that may be affected by this command: ACCESS, DTD, ETG, FOCCOMP, FOCEXEC, FOCSTYLE, GIF, HTML, MAINTAIN, MASTER, SQL, WINFORMS, XML, XSD. You must issue a separate command for each component type you wish to affect.

  SKIP

  Indicates that the designated file type should not be created when the APP CREATE command is issued.

  CREATE

  Creates the designated file type when the APP CREATE command is issued. This is the default setting.

  POSTFIX

  Specifies the lower-level qualifier of the DSN (data set name) for the component type. The APPROOT value is used to complete the full DSN, which is expressed as

  approotvalue.appname.component_type 

  The default value for component_type is

  filetype.DATA

  parms

  Are the allocation parameters you can set. The default parameter values are:

  Filetype	Parms
  ACCESS	RECFM FB TRKS LRECL 80 BLKSIZE 22000 SPACE 50 50 DIR 50
  DTD	RECFM VB TRKS LRECL 4096 BLKSIZE 27998 SPACE 50 50 DIR 50
  ETG	RECFM FB TRKS LRECL 80 BLKSIZE 22000 SPACE 50 50 DIR 50
  FOCCOMP	RECFM VB TRKS LRECL 32756 BLKSIZE 32760 SPACE 50 50 DIR 50
  FOCEXEC	RECFM VB TRKS LRECL 4096 BLKSIZE 27998 SPACE 50 50 DIR 50
  FOCSTYLE	RECFM FB TRKS LRECL 1024 BLKSIZE 27648 SPACE 50 50 DIR 50
  GIF	RECFM VB TRKS LRECL 1028 BLKSIZE 27998 SPACE 50 50 DIR 50 GIF type creates libraries for GIF and JPG files.
  HTML	RECFM VB TRKS LRECL 4096 BLKSIZE 27998 SPACE 50 50 DIR 50
  MAINTAIN	RECFM VB TRKS LRECL 4096 BLKSIZE 27998 SPACE 50 50 DIR 50
  MASTER	RECFM FB TRKS LRECL 80 BLKSIZE 22000 SPACE 50 50 DIR 50
  SQL	RECFM VB TRKS LRECL 32756 BLKSIZE 32760 SPACE 50 50 DIR 50
  WINFORM	RECFM VB TRKS LRECL 4096 BLKSIZE 27998 SPACE 50 50 DIR 50
  XML	RECFM VB TRKS LRECL 4096 BLKSIZE 27998 SPACE 50 50 DIR 50
  XSD	RECFM VB TRKS LRECL 4096 BLKSIZE 27998 SPACE 50 50 DIR 50

• DYNAM APP app1 [app2 ...]

  Enable application libraries to be allocated during the server startup, improving performance. This command is not applicable to sequential data sets in the application (for example, FOCUS, FTM) which will only be allocated when they are referenced. For example:

  DYNAM APP IBISAMP BASEAPP (default command at installation time)

• DYNAM ALLOC commands

  For sequential data sets in the application (for example, FOCUS, FTM) to be allocated at server startup (equivalent to adding a JCL allocation for these files in IRUNJCL).

We recommend that you run the server in a non-swappable address space. In order to make the server address space permanently non-swappable, add the following entry to SYS1.PARMLIB(SCHEDxx):

PPT PGMNAME(TSCOM300)         /* PROGRAM NAME */
NOSWAP                        /* NON-SWAPPABLE */
CANCEL                        /* CAN BE CANCELLED */

Do not use the KEY 0 parameter, or any other parameter (such as NOPASS), unless the system programmer completely understands the consequences of adding the parameter.

All local spawn transactions will perform in the mode of the server. For example, if the server address space is non-swappable, all local spawn execute as non-swappable.

The server executes limited non-local spawn, such as when the user executes a UNIX system command. Non-local spawn execute as swappable.

The server never executes a fork subroutine. (A fork subroutine creates a new process. The new process, called the child process, is an almost exact copy of the calling process, which is called the parent process.)

Although the server may run in a specific performance group, transactions submitted by server agents may perform differently than the server by adding the following keyword to edaserve.cfg:

wlm_enclave_trname = WLM_transaction_name          

where:

WLM_transaction_name

Can be up to 8 characters.

This is a service-level keyword.

Using this setting, the task will join a Workload Manager (WLM) enclave when a request starts, and leave the enclave when the request finishes. This gives WLM control of the dispatching priority of the task. The transaction rules defined on WLM will determine the default service class assigned to this transaction, and that service class will determine how the request runs.

This feature helps to balance a workload so that a long request will not affect a short request. This can be achieved through WLM rules designed to lower the priority of a long request after a certain period of time. Without this feature, all requests share the region priority.

The transaction name passed in this keyword must match one defined in the WLM Classification Rules for the Job Entry Subsystem (JES). A corresponding WLM Service Class pointed to by this rule will then be associated with this service.

The classification rules for JES must be used even if the server is started as a started task. The subtasks are always run under JES.

For example, you would include the following in edaserve.cfg:

SERVICE = DEFAULT
 
BEGIN
wlm_enclave_trname = IWAYFAST
.
.
.
END

The WLM definition is:

Subsystem Type JES - Batch Jobs
Classification:
 
Default service class is PRDBATLO
There is no default report class.
 
  Qualifier  Qualifier      Starting       Service   Report
# type       name           position       Class     Class
- ---------- -------------- ---------      --------- --------
1 TN         IWAYFAST                      EDAQRYHI

WLM sub-rules (levels 2 and above) are supported. For a server request to join an enclave in a particular service class, the names of all rule qualifiers below our transaction name are checked. For example, consider the following WLM definition:

Subsystem Type JES - Batch Jobs
Classification:
 
Default service class is PRDBATLO
There is no default report class.
 
  Qualifier Qualifier     Starting Service  Report
# type      name          position Class    Class
- --------- ------------- -------- -------- --------
1 SSC       PRDMVS                 PRDDFLT
2 TN        IWAYFAST               EDAQRYHI

In this particular case, the qualifier 1 type is SSC (Subsystem Collection), and a server request will only join the enclave IWAYFAST if it is running on a particular LPAR in the SYSPLEX. This qualifier (PRDMVS) must match the XCF group definition: issue $DMASDEF (for JES2) and check the value of XCFGRPNM field.

You can handle WLM scheduling environments by defining them to WLM and then adding the JOB statement parameter SCHENV=xxxxx to the ISTART JCL.

Releases prior to 7.7.06 pre-load various samples into the IBISAMP application. As of 7.7.06, on a new installation, the IBISAMP application is created, but is not pre-loaded. The server Web Console has a new feature on the ribbon and on the application tree (under new), Tutorials (the Create Tutorial Framework page), which has a pull-down for various samples. The DMC also has this feature on the application tree.

There are currently about 10 different tutorial/sample selections available on the pull-down select list to match various customer needs. The bulk of the prior IBISAMP sample objects can be generated by selecting the Create Legacy Sample Tables and Files tutorial. Other prior IBISAMP DataMigrator sample objects (usually starting with the characters dm*) are now loaded by choosing their respective DataMigrator tutorials. Under the new method, the tutorials/samples may be loaded to any application, not just IBISAMP.

If you are doing just a software refresh, the prior IBISAMP objects will be unchanged (because a refresh does not touch app directories).

Q: Why might someone want to use the PDS deployment?

A: PDS deployment provides the same rich level of features as the HFS-deployed server, including the Web Console, but removes the requirement for interaction with Unix System Services at installation and run time. It deploys the server software in partitioned data sets. Configuration and user-created source files, such as procedures and metadata, are also stored in PDS libraries.

Administration of the server, from a systems perspective, has been streamlined to match that of the classic MVS version of the server (also known as the SSCTL server). There are fewer user ID requirements for installing and operating the PDS-deployed server than the HFS-deployed version, and security management has been simplified.

Q: Does this replace the older MVS server (also known as SSCTL)?

A: The z/OS server with PDS deployment is a migration path from the older MVS server.

Q: Can one refresh a server's installation software that had been deployed one way with software using other type of deployment?

A: No. Each deployment type is independent of the other with regards to installation.

Q: Can both deployments of the server coexist on one z/OS system?

A: Yes, if your license agreement allows for this.

Q: Can one configure two server instances of the same server, one instance an HFS/USS deployment, and the other a PDS deployment?

A: No. Although the media and installation are unified, once the base server software is installed, the two deployment types run separately.

As with the HFS/USS deployment, the PDS deployment can have many instances running from the same EDAHOME set of libraries.

Q: Can I monitor server startup by checking the MVS SYSLOG?

A: Yes.

The following messages are written to the SYSLOG when

• The server starts successfully:

  (EDA13023) ALL INITIAL SERVERS STARTED

• The Server does not start:

  (EDA13171) UNABLE TO START IWAY SERVER

Q: What, if anything, does the PDS deployment not support? In what installation implementation?

A: The PDS deployment of the server currently does not support the following functions:

• The Web Console Run Stress option.
• Displaying server logs and traces in the Web Console.
• Formats XLSX and PPTX.

  Note: As a workaround, you can issue the SET EXCELSERVEURL command to point to another reporting server that supports these formats.

• Adobe Flex.
• RACF TEMPDSN class—Supported except for FOCCACHE application datasets.

OpenFlex SDK is included by Information Builders for use with its HOLD FORMAT FLEX feature. This distribution is subject to the terms and conditions of the Mozilla Public License Version 1.1.

For more information, see Zip Archiver License or visit our website http://www.informationbuilders.com.