Existing database upgrade to use full-text search

The tasks described in this section are necessary only if you have an existing user database that you are upgrading to use the ClearQuest full-text search feature. Go to the next section of the help if you are creating a new user database.

If you are starting with an existing user database, you must index the submittable records that you selected to index when you created the ClearQuest search properties file to ensure that these records are included in the Lucene index. You use the command-line record extractor in batch mode to index existing user database records.

When you run the command-line record extractor in batch mode, the record extractor reads all record types and their fields that are configured for indexing in the ClearQuest search properties file, CQ-db_name-dbset_name.xml, and sends the data to Solr for Lucene to index. This process can be time consuming because it is processor, I/O, database, and network intensive. The time it will take to index your user database is bound to the number of records and size of each record to index, and the performance of the ClearQuest database server. Because you will be running the record extractor in batch mode on your production database, schedule the job at a time of low use to limit the performance impact.


Feedback