Two directories called 'nullcomments' and 'nullissues' are created in the Working directory when setting up JIRA for the first time.

1. Start JIRA on blank database.
2. Move onto the second step on the setup wizard.

There will be a 'nullcomments' and 'nullissues' directory in JIRA's working directory.

Its caused by the JiraSystemInfo.obtainDatabaseStatistics method. This method getss the comment Searcher before the Indexing system has been initialised (thus the null*). I find it strange that we allow people to use the indexing system even though it has not been initialised yet. I would expect and exception to be thrown. It also appears that people would be able to get a searcher even when indexing is disabled.