Object index maintenance

To ensure that your rich search results are accurate, you can use the index regeneration script, indexRegenerator.sh, to regenerate object indexes.

The indexRegenerator.sh script is not run automatically; however, to ensure that your search results are accurate, run the script after you change an attribute from non-indexed to indexed.

You cannot run the indexRegenerator.sh script on workflows or collaboration areas.

Use the following guidelines to determine when to run the indexRegenerator.sh script:

  • If there are newly marked attributes as indexed for existing items, those are not updated in the relational table automatically and thus would result in incomplete result set. Running the indexRegenerator.sh tool parses all existing items and update the relational table with all attribute values that are marked as indexed.
  • If a new version of the container is generated after you regenerate indexes for an object, you do not need to run the indexRegenerator.sh script.
  • If you roll back to a version of an object that includes regenerated indexes, you do not need to run the indexRegenerator.sh script.
  • If the object that you roll back to does not include regenerated indexes, you need to run the indexRegenerator.sh script.
    Restriction: If you roll back a catalog or hierarchy object concurrently while the indexRegenerator.sh script is processing one of those objects, the indexRegenerator.sh script exits and logs an error in the log file.