Created by James Hurrell, last modified on Nov 03, 2021
This documentation is not maintained. Please refer to doc.castsoftware.com/technologies to find the latest updates.

Target audience:

CAST Administrators, Delivery Managers

Summary: This page explains how to onboard a PeopleSoft application.

Register your PeopleSoft applications in the CAST AIC Portal

In the CAST AIC Portal, register two new Applications as follows:

  • One for the Vanilla: the PeopleSoft Repository without any custom code
  • One for the Project: the PeopleSoft Repository that contains the custom code for one or multiple PeopleSoft applications
Note that only ONE CAST AIP application per PeopleSoft repository should be created in the CAST AIC Portal, even though the PeopleSoft repository may contain multiple custom "applications". Source code will be extracted "in one go" from each PeopleSoft repository.

Deliver the source code for Vanilla

Click here to expand...

To deliver the source code for the Vanilla, you must download and install the CAST Delivery Manager Tool (DMT) on a machine that has access to the server that hosts the PeopleSoft repository for Vanilla. CAST recommends installing the DMT on the same machine as the server is installed on, to take advantage of faster direct disk to disk performance for the code extraction. The DMT establishes a JDBC connection to the database and executes a set of queries. If the DMT is installed on another machine, the network will negatively impact the time required to extract the source code.

  • With regard to packages:
    • Create one package for the PeopleSoft Vanilla extraction
    • Ideally, you should also create a second package to deliver the database extraction (database structure). The resulting objects will be linked to PeopleCode embedded SQL in order to compute 7 quality rules (dead code on several types of records). This second package should be skipped in the following cases:
      • PeopleSoft application is FSCM : the volume to extract and analyze for the database part would be very large, so the effort is not worth the benefit.
      • The PeopleSoft repository is hosted on an unsupported RDBMS version, thus cannot be extracted by the database extractor.
  • Once the DMT has been downloaded and installed, you'll have access to the interface:

  • First create a new version:

  • The version editor is then displayed:

  • Create a package. In the wizard, select "PeopleSoft" under "Select a vendor specific repository":

  • In the wizard, select the appropriate option as described in the expandable sections below:


PeopleSoft on DB2 / PeopleSoft on Oracle
Click here to expand...

These two options will perform an extraction using the database extractor embedded into the CAST Delivery Manager Tool - connections are made using JDBC drivers. Use the table below to help fill in the required fields:

OptionDescription

Host name

Your Oracle / DB2 server host name: either its netbios name or its IP address

Port

  • Oracle instance port number, as per Tnsname.ora file - usually 1521
  • DB2 server port - usually 50000
Instance Identification (Oracle)

Choose either:

  • Service as per Tnsname.ora file
  • The SID (System ID) is the instance name as per Tnsname.ora file.
Database (DB2)The DB2 database name.
Credentials

Fill in the credentials for the target server:

User nameOracle or DB2 user.
PasswordPassword for the user selected in the User name field.
Remember password

This option enables you to force the CAST Delivery Manager Tool to save the database access credentials you have entered above.

Choosing an option or not has no impact on the extraction (i.e. the CAST Delivery Manager Tool can still access the required resources). However, if you are creating subsequent Versions of the same schemas on the same server, you can choose to store the password in which case you will not need to re-enter it.

  • Not stored > The credentials are not saved.
  • Local > The credentials are saved in the user's local workspace on the current machine. Choose this option if you do not want the password to be available to other Delivery Managers.
  • Server > The credentials are saved locally (as above) and are also synchronized back to the CAST AIC Portal (i.e. the Source Code Delivery Folder). Choose this option if you want the password to be available to other Delivery Managers.
JVM Memory Size

The CAST extractors are Java based and require the use of the JRE JVM (Java Virtual Machine). This option allows you to configure the Maximum Java Heap Size of your JVM for use during the database/schema browse and extraction processes. By default 1GB of memory is allocated and in most scenarios this value can be left as it is.

You may need to change the amount of memory allocated to the JVM if you are getting out of memory exceptions for the Java Heap Space when you either browse to select a database/schema or when you use the Package action to invoke the extraction - this can occur more specifically when attempting to extract a large single database/schema or multiple databases/schemas. Please increase the memory allocation until you no longer receive errors, however, please bear the following in mind:

  • 32bit JRE: If you are running the CAST Delivery Manager Tool via the CAST AIC Portal and have installed a 32bit JRE for this purpose, or when the CAST Delivery Manager Tool is launched from the CAST Management Studio (when it will use a 32bit JRE that is provided with the CAST installation), CAST recommends that you do not increase the JVM Memory Size value to above 2GB, otherwise you risk running out of system memory.
  • If your database/schema extraction consistently runs out of memory despite having allocated the maximum possible with a 32bit JRE, you may need to ensure you are using a 64bit JRE to take advantage of increased memory availability. To do so, you must launch the CAST Delivery Manager Tool from the CAST AIC Portal, rather than directly from the CAST Management Studio (when it will use a 32bit JRE that is provided with the CAST installation). And before you do so, you must manually modify the shortcut to the CAST Delivery Manager Tool (available in the Windows Start menu) to replace the existing path to the 32bit JRE (C:\Windows\SysWOW64\javaws.exe) with your 64bit JRE path ("C:\Program Files\java\jre1.7.0_51\bin\javaws.exe" for example). You can add the -Xmx parameter (maximum memory allocation), that will allow the process to use the memory defined. The shortcut should now look similar to this:
"C:\Program Files\java\jre1.7.0_51\bin\javaws.exe" -Xmx4096M -localfile
"C:\Users\<user_name>\AppData\LocalLow\Sun\Java\Deployment\cache\6.0\6\1c274146-29c73860"
Repository TypeVanilla or Project
SchemaSchema in which the PeopleSoft tables are stored:
  • Oracle USER name
  • DB2 schema name
PeopleTools versionChoose the appropriate PeopleTools version from the drop down list.
Reuse existing PeopleSoft extractor output

Click here to expand...

This option allows you to perform an offline extraction using the CAST Database Extractor (which can be downloaded from https://extend.castsoftware.com/#/extension?id=com.castsoftware.aip.extractor.sqldatabase&version=latest) and then package the output for delivery. A tailor made batch file provided in the downloaded PeopleSoft extension is available for running the offline extraction with the CAST Database Extractor - this should be used where possible. The batch file is available here as shown below. Replace all instances of <PARAM> in the batch file with the appropriate values as explained in the batch file comments. Please note all values are mandatory.

%PROGRAMDATA%\CAST\CAST\Extensions\com.castsoftware.peoplesoft.<version>\TOOLS\OfflineExtraction\PeopleSoft-Extract-CLI.bat

The following errors may occur when running the batch file:

ErrorDescription / Action
Error while extracting PeopleSoft Project : Unable to establish a connection. Check the log file <path>\Vanilla.log.

The extractor was not able to connect to the target server, for example, server/schema name incorrect, port number incorrect, server offline, incorrect credentials etc.:

  • Please check the logs, fix the configuration and retry the extraction.
  • If the issue is not a configuration issue, please contact the CAST Support instead.
Error while extracting PeopleSoft Project : Bad Java Location. Check the parameter JAVA_HOME.

There is an issue with the JAVA_HOME environment variable on the machine on which the batch file is being run: JAVA_HOME may not be defined (JRE not installed for example), incorrect path etc.:

  • Please check the logs, fix the configuration and retry the extraction.
  • If the issue is not a configuration issue, please contact the CAST Support instead.
Error while extracting PeopleSoft Project : Config file is missing: <path>\<name>.config. Copy it from E:\DMT\Peoplesoftoracleextractor-1.0.0.This error should not occur in normal circumstances. Please contact the CAST Support if the issue occurs.
Error 1000 / 2000 while extracting PeopleSoft Project.

This error may occur when extracting from an Oracle instance and you have used the SERVICE name instead of the SID name for the following PARAM on line 31 of PeopleSoft-Extract-CLI.bat:

SET DATABASE_NAME=<PARAM>

To solve this issue, change the following on line 112 of PeopleSoft-Extract-CLI.bat (change ":" to "/"):

:%DATABASE_NAME%

to:

/%DATABASE_NAME%

  • To extract the source code, click Package...

  • When the extraction has completed, the content state for the Package should display Packaging successful and the action required is Package ready to be delivered. 

Post extraction check

Validation error #-1: Schema EFORCE_G5R3R is not a PeopleSoft repository: table PSRECDEFN is missing

If errors similar to "Validation error #-1: Schema EFORCE_G5R3R is not a PeopleSoft repository: table PSRECDEFN is missing" are reported in extactor's log, it means either the schema has no PSRECDEFN table, or USER rights are insufficient (SELECT has not been granted on this table).

Resolution error type in / Invalid identification group value

The extraction log may contain the following messages:

  • Resolution error type in
  • Invalid identification group value

These messages have a limited impact if the ratio of "Number of information messages" / "Number of rows extracted" is kept low (i.e. less the 1 or 2%). Some Quality Rule violations may be missed if these messages appear, however, the overall impact is low.

Deliver

  • You can now deliver the source code. To do so, click Deliver…. In the wizard, answer Yes when prompted whether the delivery is complete:

  • When the delivery has completed, the content state for the Package should display Packaging successful and the action required is empty (i.e. nothing further to do).

Deliver the source code for Project

Click here to expand...

To deliver the source code for the Vanilla, you must download and install the CAST Delivery Manager Tool (DMT) on a machine that has access to the server that hosts the PeopleSoft repository for Project. CAST recommends installing the DMT on the same machine as the server is installed on, to take advantage of faster direct disk to disk performance for the code extraction. The DMT establishes a JDBC connection to the database and executes a set of queries. If the DMT is installed on another machine, the network will negatively impact the time required to extract the source code.

  • If your Project and Vanilla repositories are on the same server, you can re-use the DMT you used to extract the Vanilla.
  • With regard to packages:
    • Create one package for the PeopleSoft Project extraction (as was required for the Vanilla repository)
    • Ideally, you should also create a second package to deliver the database extraction (database structure). The resulting objects will be linked to PeopleCode embedded SQL in order to compute 7 quality rules (dead code on several types of records). This second package should be skipped in the following cases:
      • PeopleSoft application is FSCM : the volume to extract and analyze for the database part would be very large, so the effort is not worth the benefit.
      • The PeopleSoft repository is hosted on an unsupported RDBMS version, thus cannot be extracted by the database extractor.

  • Once the DMT has been downloaded and installed, you'll have access to the interface:

  • First create a new version:

  • The version editor is then displayed:

  • Create a package. In the wizard, select "PeopleSoft" under "Select a vendor specific repository":

  • In the wizard, select the appropriate option as described in the expandable sections below:


PeopleSoft on DB2 / PeopleSoft on Oracle
Click here to expand...

These two options will perform an extraction using the database extractor embedded into the CAST Delivery Manager Tool - connections are made using JDBC drivers. Use the table below to help fill in the required fields:

OptionDescription

Host name

Your Oracle / DB2 server host name: either its netbios name or its IP address

Port

  • Oracle instance port number, as per Tnsname.ora file - usually 1521
  • DB2 server port - usually 50000
Instance Identification (Oracle)

Choose either:

  • Service as per Tnsname.ora file
  • The SID (System ID) is the instance name as per Tnsname.ora file.
Database (DB2)The DB2 database name.
Credentials

Fill in the credentials for the target server:

User nameOracle or DB2 user.
PasswordPassword for the user selected in the User name field.
Remember password

This option enables you to force the CAST Delivery Manager Tool to save the database access credentials you have entered above.

Choosing an option or not has no impact on the extraction (i.e. the CAST Delivery Manager Tool can still access the required resources). However, if you are creating subsequent Versions of the same schemas on the same server, you can choose to store the password in which case you will not need to re-enter it.

  • Not stored > The credentials are not saved.
  • Local > The credentials are saved in the user's local workspace on the current machine. Choose this option if you do not want the password to be available to other Delivery Managers.
  • Server > The credentials are saved locally (as above) and are also synchronized back to the CAST AIC Portal (i.e. the Source Code Delivery Folder). Choose this option if you want the password to be available to other Delivery Managers.
JVM Memory Size

The CAST extractors are Java based and require the use of the JRE JVM (Java Virtual Machine). This option allows you to configure the Maximum Java Heap Size of your JVM for use during the database/schema browse and extraction processes. By default 1GB of memory is allocated and in most scenarios this value can be left as it is.

You may need to change the amount of memory allocated to the JVM if you are getting out of memory exceptions for the Java Heap Space when you either browse to select a database/schema or when you use the Package action to invoke the extraction - this can occur more specifically when attempting to extract a large single database/schema or multiple databases/schemas. Please increase the memory allocation until you no longer receive errors, however, please bear the following in mind:

  • 32bit JRE: If you are running the CAST Delivery Manager Tool via the CAST AIC Portal and have installed a 32bit JRE for this purpose, or when the CAST Delivery Manager Tool is launched from the CAST Management Studio (when it will use a 32bit JRE that is provided with the CAST installation), CAST recommends that you do not increase the JVM Memory Size value to above 2GB, otherwise you risk running out of system memory.
  • If your database/schema extraction consistently runs out of memory despite having allocated the maximum possible with a 32bit JRE, you may need to ensure you are using a 64bit JRE to take advantage of increased memory availability. To do so, you must launch the CAST Delivery Manager Tool from the CAST AIC Portal, rather than directly from the CAST Management Studio (when it will use a 32bit JRE that is provided with the CAST installation). And before you do so, you must manually modify the shortcut to the CAST Delivery Manager Tool (available in the Windows Start menu) to replace the existing path to the 32bit JRE (C:\Windows\SysWOW64\javaws.exe) with your 64bit JRE path ("C:\Program Files\java\jre1.7.0_51\bin\javaws.exe" for example). You can add the -Xmx parameter (maximum memory allocation), that will allow the process to use the memory defined. The shortcut should now look similar to this:
"C:\Program Files\java\jre1.7.0_51\bin\javaws.exe" -Xmx4096M -localfile
"C:\Users\<user_name>\AppData\LocalLow\Sun\Java\Deployment\cache\6.0\6\1c274146-29c73860"
Repository TypeVanilla or Project
SchemaSchema in which the PeopleSoft tables are stored:
  • Oracle USER name
  • DB2 schema name
PeopleTools versionChoose the appropriate PeopleTools version from the drop down list.
Reuse existing PeopleSoft extractor output

Click here to expand...

This option allows you to perform an offline extraction using the CAST Database Extractor (which can be downloaded from https://extend.castsoftware.com/#/extension?id=com.castsoftware.aip.extractor.sqldatabase&version=latest) and then package the output for delivery. A tailor made batch file provided in the downloaded PeopleSoft extension is available for running the offline extraction with the CAST Database Extractor - this should be used where possible. The batch file is available here as shown below. Replace all instances of <PARAM> in the batch file with the appropriate values as explained in the batch file comments. Please note all values are mandatory.

%PROGRAMDATA%\CAST\CAST\Extensions\com.castsoftware.peoplesoft.<version>\TOOLS\OfflineExtraction\PeopleSoft-Extract-CLI.bat

The following errors may occur when running the batch file:

ErrorDescription / Action
Error while extracting PeopleSoft Project : Unable to establish a connection. Check the log file <path>\Vanilla.log.

The extractor was not able to connect to the target server, for example, server/schema name incorrect, port number incorrect, server offline, incorrect credentials etc.:

  • Please check the logs, fix the configuration and retry the extraction.
  • If the issue is not a configuration issue, please contact the CAST Support instead.
Error while extracting PeopleSoft Project : Bad Java Location. Check the parameter JAVA_HOME.

There is an issue with the JAVA_HOME environment variable on the machine on which the batch file is being run: JAVA_HOME may not be defined (JRE not installed for example), incorrect path etc.:

  • Please check the logs, fix the configuration and retry the extraction.
  • If the issue is not a configuration issue, please contact the CAST Support instead.
Error while extracting PeopleSoft Project : Config file is missing: <path>\<name>.config. Copy it from E:\DMT\Peoplesoftoracleextractor-1.0.0.This error should not occur in normal circumstances. Please contact the CAST Support if the issue occurs.
Error 1000 / 2000 while extracting PeopleSoft Project.

This error may occur when extracting from an Oracle instance and you have used the SERVICE name instead of the SID name for the following PARAM on line 31 of PeopleSoft-Extract-CLI.bat:

SET DATABASE_NAME=<PARAM>

To solve this issue, change the following on line 112 of PeopleSoft-Extract-CLI.bat (change ":" to "/"):

:%DATABASE_NAME%

to:

/%DATABASE_NAME%

  • To extract the source code, click Package...


  • When the extraction has completed, the content state for the Package should display Packaging successful and the action required is Package ready to be delivered. 

Post extraction check

Validation error #-1: Schema EFORCE_G5R3R is not a PeopleSoft repository: table PSRECDEFN is missing

If errors similar to "Validation error #-1: Schema EFORCE_G5R3R is not a PeopleSoft repository: table PSRECDEFN is missing" are reported in extactor's log, it means either the schema has no PSRECDEFN table, or USER rights are insufficient (SELECT has not been granted on this table).

Resolution error type in / Invalid identification group value

The extraction log may contain the following messages:

  • Resolution error type in
  • Invalid identification group value

These messages have a limited impact if the ratio of "Number of information messages" / "Number of rows extracted" is kept low (i.e. less the 1 or 2%). Some Quality Rule violations may be missed if these messages appear, however, the overall impact is low.

Deliver

  • You can now deliver the source code. To do so, click Deliver…. In the wizard, answer Yes when prompted whether the delivery is complete:

  • When the delivery has completed, the content state for the Package should display Packaging successful and the action required is empty (i.e. nothing further to do).