Before you upgrade to the current version of Data Archive, perform the following tasks:
Read the release notes for information about known and fixed limitations for the release.
Verify that you have read and write privileges on the directory where the
ILM_HOME
folder exists. If you do not have read and write privileges on this directory, the upgrade fails.
Create a backup of the ILM repository database and the ILM_HOME folder. If the upgrade fails, you must restore the backup before you try to upgrade again.
Stop the ILM application server and end all ILM services.
Optionally, delete the
work
folder from the
ILM_HOME
directory.
Extract the installer files from the compressed installation file:
Informatica_ILM_<release><hotfix>.zip
The installer files include the Data Archive installer.
If you use the Application Retirement for Healthcare accelerator, verify that the catalog that exists in
ILM_HEALTHCARE\<ARCHIVE_FOLDER_NAME>
and the catalog deployed on the JReport Server are the same. If not, download the latest catalog from the JReport Server through JReport Designer and place the catalog inside
ILM_HEALTHCARE\<ARCHIVE_FOLDER_NAME>
before you upgrade.
Smart Partitioning Users
If you use Smart Partitioning, perform the following steps before upgrading to Data Archive 6.5.
Uncompress the compressed partitions.
Disable access to the policy jobs.
Turn the segments into read-write mode.
Perform one of the following steps:
If the original tables exist in the database
If the original tables exist in the database, run the Replace Segmented Tables with Original Tables job. The Replace Segmented Tables with Original Tables job mends the tables that you have run a segmentation policy on to return the tables to their original state. Run this job if you want to undo the smart partitioning process in your environment.
If the original tables were dropped in the database
If the original tables were dropped in the database, run the Unpartition Tables job. The Unpartition Tables job replaces all of the segmented tables in a segmentation group with tables that are identical to the original, unsegmented tables. If you have run a segmentation policy and dropped the original tables, you can run the Unpartition Tables job to revert the segmentation group to its original state. After you run the job, the segmentation group status changes to