Post-migration preparation tasks

After migrating, you will need to perform additional post-migration tasks.

About this task

Perform the following preparation tasks:

Procedure

  1. By default, the migration process secures the application in the following ways:
    • adds authentication to the interop servlet
    • adds authenticated access to web pages
    • enables XAPI security, authorization, and permissioning
      Note: The property, api.security.mode, is set to Strict during migration. However, you can specify other modes, such as Lax and Debug, for deploying, debugging, customizing, or testing of the application.

      To run the application in backward-compatible mode, use the customer_overrides.properties file to set the following properties:

    • interopservlet.security.enabled to false to disable authentication
    • userauthfilter.enabled to false to remove authenticated access to web pages
    • api.security.enabled to N and api.security.token.enabled to N to disable XAPI security, authorization, and permissioning.
  2. To migrate index and CHAR columns to VARCHAR columns on Db2, ensure that you complete the Db2 migration process.
  3. To migrate the existing LONG columns to CLOB, perform the Oracle Online Redefinition process, as described in the Oracle® Database PL/SQL Packages and Types Reference and the Oracle® Database Administrator's Guide. After completing the redefinition process, ensure that you run the dbverify script. To prevent dbverify from generating "IndexAdds" SQLs for dropped indexes, ensure that the dropped indexes are listed in the <INSTALL_DIR>/extensions/schemagenerator/indexes_not_created.txt file. Or, if the INDEXES_NOT_CREATED property in the <INSTALL_DIR>/properties/sandbox.cfg file is configured, add the list of dropped indexes to the file referenced by the property. Run all the index changes that the dbverify script generates.

    Table 1 lists the LONG columns that can be migrated to CLOB.

    Note: If you are upgrading to 9.5 from a fresh installation of 8.5 (or later), you can skip this step. However, if either of the following conditions is true, you may want to migrate existing LONG columns to CLOB:
    • You are upgrading to 9.5 from 8.2 (or earlier).
    • You upgraded to 8.5 (or later) without migrating existing LONG columns to CLOB and are now upgrading to 9.5.
    Table 1. Columns that can be migrated to CLOB
    Table Column
    YFS_SUB_FLOW CONFIG_XML
    YFS_REPROCESS_ERROR MESSAGE
    INTEROP_ERROR_DISPATCH MESSAGE
    YFS_GRAPH_UI GRAPH_XML
    YFS_INBOX INBOX_ADDNL_DATA
    YFS_AGENT_CRITERIA_TEMPLATE MESSAGE_XML
    YFS_AGENT_CRITERIA MESSAGE_XML
    YFS_BASE_DOCUMENT_TEMPLATE TEMPLATE_XML
    YFS_DOCUMENT_PARAMS DOCUMENT_PARAMS
    YFS_HM_THRESHOLDS ADDITIONAL_INFO
    YFS_QUERY QUERY_DEFINITION
    YFS_EXPORT MESSAGE
    YFS_IMPORT MESSAGE
    YFS_SAVED_SEARCH SEARCH_DATA
    YFS_AUDIT AUDIT_XML
    YFS_TRANSACTION_DEPENDENCY TEMPLATE_XML
    YFS_ORDER_AUDIT_DETAIL AUDIT_XML
    YFS_DEVICE DEVICE_PARAMS_XML
    YFS_DEVICE_TYPE_TEMPLATE DEVICE_TYPE_PARAMS_XML
    YFS_BATCH REQ_BATCH_COMPLETION_XML
    YCS_MANIFEST_UPS_DTL INPUT_XML
    YFS_ASYNC_REQ MESSAGE
    YFS_BASE_DOCUMENT_PARAMS DOCUMENT_PARAMS
    YFS_DOCUMENT_TEMPLATE TEMPLATE_XML
    Note: The columns listed in Table 1 are also applicable for the corresponding history tables.
  4. If you upgraded in nonsharded mode, you can set up a sharded deployment after completing the nonsharded migration process. To set up a sharded deployment, move from a nonsharded environment to a sharded environment.
  5. The case insensitive search feature allows you to make search operations in the application case insensitive. You can enable case insensitive search on existing column data.
  6. If the YFS_ITEM entity is extended to have custom columns in releases prior to Release 8.5 and these extended columns are used to define ItemAttribute records, you can manually add the extended columns of the YFS_ITEM entity as the YFS_ATTRIBUTE record and the YFS_ATTRIBUTE_REFERENCE record. These columns can also be added through the Sterling Business Center user interface.
  7. If you had any extensions in your classpath in your old runtime, ensure that you copy the same extensions into the classpath for your new runtime.
  8. When upgrading to 9.5, run <INSTALL_DIR>/bin/install3rdParty.sh (for UNIX and Linux® operating systems) or <INSTALL_DIR>\bin\install3rdParty.cmd (for Windows operating system) script for all the custom-installed JARs.

    Run this script for all JARs in the <INSTALL_DIR>/extensions/${extension.root}/jars directory, too.

  9. If you have installed an extension package for customization, run the <INSTALL_DIR>\bin\InstallService.cmd script (for Windows) or the <INSTALL_DIR>/bin/InstallService.sh script (for Linux).
  10. Ensure that you are able to restore the database on another server.
  11. To deploy the extensions in EAR, the resource JARs must be rebuilt. To rebuild the resource JARs, execute the ./deployer.sh -t resourcejar command from the <INSTALL_DIR>/bin directory (for UNIX and Linux operating systems) or the deployer.cmd -t resourcejar command from the <INSTALL_DIR>\bin directory (for Windows operating system).
  12. Modify the value of the following properties in the sandbox.cfg file located in the <INSTALL_DIR>/properties file:
    • REINIT_DB=true
    • LOAD_FACTORY_SETUP=true
    • NO_DBVERIFY=false
    • Rerun setupfiles.sh from the <INSTALL_DIR>/bin folder (for UNIX and Linux operating systems) or setupfiles.cmd from the <INSTALL_DIR>\bin folder (for Windows operating system).
  13. Execute the following tasks, which generate the upgrade_diff.xml:

    For Windows:<INSTALL_DIR>\bin\tmp.cmd

    • set CLASSPATH=<DBDRIVER>;<INSTALL_DIR>
    • \jar\install_foundation.jar;<INSTALL_DIR>
    • \jar\platform_afc\10_0\platform_dv.jar;<INSTALL_DIR>
    • \jar\platform_afc\10_0\platform_afc.jar;<INSTALL_DIR>
    • \jar\platform_baseutils.jar;<INSTALL_DIR>\jar\xml4j\4_5_16\xercesImpl.jar;<INSTALL_DIR>\jar\log4j\1_2_17\log4j-1.2.17.jar;<INSTALL_DIR>\icu4j\55_1\icu4j-55_1.jar;%CLASSPATH%

    • Here, <DBDRIVER> points to the dbdriver JARs located in the <INSTALL_DIR>\dbjar\jdbc\<DB> directory.

    For Linux: <INSTALL_DIR>\bin\tmp.sh

    • export CLASSPATH=<DBDRIVER>:<INSTALL_DIR>
    • \jar\install_foundation.jar:<INSTALL_DIR>
    • \jar\platform_afc\10_0\platform_dv.jar:<INSTALL_DIR>
    • \jar\platform_afc\10_0\platform_afc.jar:<INSTALL_DIR>
    • \jar\platform_baseutils.jar:<INSTALL_DIR>\jar\xml4j\4_5_16\xercesImpl.jar:<INSTALL_DIR>
    • \jar\log4j\1_2_17\log4j-1.2.17.jar:<INSTALL_DIR>\icu4j\55_1\icu4j-55_1.jar:%CLASSPATH%

    • Here, <DBDRIVER> points to the dbdriver JARs located in the <INSTALL_DIR>\dbjar\jdbc\<DB> directory.

    • java com.yantra.tools.entityguru.reports.UpgradeXMLGenerator -ED <INSTALL_DIR>/repository/entity -DT <INSTALL_DIR>/repository/datatypes/datatypes.xml -U <User> -P <passwd> -D <jdbc driver> -URL <jdbc url>

      Refer to the jdbc.properties file for the values of <jdbc url>, <user> and <passwd>.

      This provides a list of all the columns whose lengths must be taken into consideration when building dbclasses for the prepared statements to function correctly. This requires all the CHAR fields to be padded to their column length.

  14. If you extended the resource XML file, orderApprovalGreexTemplate.xml, and want to enable the customer rating feature, add the following element under the Order element in the orderApprovalGreexTemplate.xml file: 
    
   <Customer CustomerKey="">
      <CustomerAnalytics GradeCode=""/> 
   </Customer>
  15. If you were using property encryption in your old runtime and want to use property encryption in Release 9.5, use the <INSTALL_DIR>/properties/customer_overrides.properties file to enable property encryption.