Scheduling Partial Indexing

After a baseline indexing, you can run a partial indexing process periodically to keep the CloudView server up to date with changes made to your database.

Before you begin: First run a baseline indexing process of the database.
Important: Before configuring the partial indexing script, make sure to:
  • Put it on the same machine that runs partial indexing processes. This machine must also be on the same time zone as your database server and 3DSpace Index.
  • Run the script at an interval that reflects your database size and activity. The interval determines how quickly changes to your database are available for searching. For most environments, an interval of every 5 minutes is enough.
See Also
Running Indexing
Clearing the Index
Clearing Checkpoint Records
  1. Go to <INSTALLDIR>\<OS>\code\command
  2. Edit the startXLPartialSearchIndex.bat[.sh] script to include:

    • The authorized system Administrator user ID and password, used to run the start searchindex command.
    • The list of vaults in which the partial indexing process needs to run. You define vaults in the start searchindex command. Default is '*'.

  3. Automate the script according to your platform.

    • On Windows, use the Windows Scheduler to automate the startXLPartialSearchIndex.bat script.
    • On UNIX/Linux, use a cron job to automate the startXLPartialSearchIndex.sh script.
      Warning: The startXLPartialSearchIndex.sh script refers to ./mql. However, the MQL startup script is located under the server installation, not the 3DSpace Index installation. Before executing the startXLPartialSearchIndex.sh script, modify the path to MQL.

Important: If latency issues occur for event-based indexing, you can run the startXLPartialSearchIndexLowLatency.[bat|.sh] script instead. As it is more aggressive in terms of incremental indexing frequency, it might not be appropriate for implementations that have large impact analysis.