In Chapter 9, Searching, Reporting, and Analysis, you looked at how you can search for data in JIRA. In order to support large instances with thousands of issues, JIRA creates and maintains search indexes on the filesystem. This index is added and modified whenever issues are created or updates are made, automatically in the background. Sometimes, however, JIRA will require you to perform a manual reindexing. This may occur when the index files are corrupt or missing, or changes are made directly in the database causing the index to be out of sync. You will notice this when your searches are constantly returning with the wrong set of issues.

When you reindex, you effectively recreate the search index on the filesystem based on the most current data in the database. This ensures that the values in both are in sync. It is important to note that as the amount of data grows in your database, so will the search index, so you will need to allocate enough disk space to the server where you install JIRA. Perform the following steps to reindex your JIRA:

Starting with JIRA 5.2, you can choose to reindex JIRA either in the background, or fully lock down JIRA to rebuild the index from scratch. When you choose the Lock JIRA and rebuild index option, you will not be able to use JIRA until the reindex process completes.

When you run reindexing in the background, you can continue using JIRA while the reindex process is running. This is very useful when you make changes, such as the search template for a custom field, and need to reindex JIRA to reflect the change during working hours.

Depending upon the size of your JIRA instance, the reindex process can take some time to complete (several hours for some large installations; for example, 200 thousand issues).