Confluence Home Directory

Often in the documentation, you'll see a reference to the 'Confluence Home directory'.

What is the Confluence Home Directory?

The Confluence Home directory is the folder where Confluence stores its configuration information, search indexes and page attachments. If you are using the embedded HSQLDB database supplied for evaluation purposes, the database files are also stored in this directory.

Tip: Another term for 'Home directory' would be 'data directory'.

The information on this page does not apply to Confluence OnDemand.

Finding the Confluence Home Directory

The location of the Confluence Home directory is defined when you install Confluence. This location is stored in a configuration file called confluence-init.properties, which is located inside the confluence/WEB-INF/classes directory in your Confluence Installation directory.

When Confluence first starts up, it reads the confluence-init.properties file to determine where to look for the Home directory.

Once Confluence is running you can find the Confluence Home directory via the Administration console, under Administration > System Information > Confluence Information - Confluence Home.

Content of the Confluence Home Directory

The Confluence home directory contains some of the configuration data used by Confluence. Other data is stored in the database. This section outlines the purpose of the files and directories in the Confluence home directory.

`confluence.cfg.xml`

This file contains all of the information necessary for Confluence to start up, such as:

Product license
Context path
Database details, such as location and connection pool settings
Paths to important directories

`attachments`

This directory contains every version of each attachment stored in Confluence. This directory is not used when Confluence is configured to store attachments in the database. Attachments are always stored in the database in clustered instances of Confluence.

Paths within this directory have the following structure:

/attachments/PAGE_ID/ATTACHMENT_ID/VERSION

You can specify an alternative directory for attachment storage by setting the attachments.dir property in confluence.cfg.xml.

`backups`

Confluence will place its daily backup archives in this directory, as well as any manually generated backups. Backup files in this directory take the following form:

daily-backup-YYYY_MM_DD.zip

You can specify an alternative directory for backups by setting the daily.backup.dir property in confluence.cfg.xml.

`bundled-plugins`

Confluence ships with a set of bundled plugins. These are plugins written by the Atlassian and the Confluence community that we think provide useful and broadly applicable functionality in Confluence. The {{bundled-plugins)) directory is where Confluence will unpack its bundled plugins when it starts up. This directory is refreshed on every restart, so removing a plugin from this directory will not uninstall the plugin. It will simply be replaced the next time Confluence starts up.

`database`

This is where Confluence stores its database when configured to run with the HSQL embedded database. In such cases this directory contains all Confluence runtime data. Installations configured to run using an external database such as MySQL will not use this directory.

`index`

This is where Confluence stores its indexes for rapid retrieval of often used data. The Confluence index is used heavily by the application for content searching and recently updated lists and as such is critical for a running Confluence instance. It is important to note however that should the data in this directory be lost or corrupted, it can be restored by running a full reindex from within Confluence. This can take a long time depending on how much data is stored Confluence's database.

An alternative directory may be specified for the index by setting the lucene.index.dir property in confluence.cfg.xml. As this is the most heavily accessed directory in the Confluence home directory you might want to consider hosting it on the fastest disk available. It would also be useful if the disk holding the Confluence index was not heavily used by any other application to reduce access contention.

`plugin-cache`

All Confluence plugins are stored in the Confluence database. To allow for quicker access to classes contained within the plugin JARs, Confluence will cache these plugins in the plugin-cache directory. This directory is updated as plugins are installed and uninstalled from the system and is completely repopulated from the database every time Confluence is restarted. Removing plugins from this directory does not uninstall them.

`resources`

The resources directory stores any space logos used in your Confluence instance. For each space with a space logo, there is a directory within resources named after the space's key. That directory contains the space's logo.

`temp`

The temp directory is used for various runtime functions such as exporting, importing, file upload and indexing. As the name suggests, and file in this directory is of temporary importance and is only used during runtime. This directory can be safely emptied when Confluence is offline.

An alternative directory may be specified for temporary data by setting the webwork.multipart.saveDir property in confluence.cfg.xml.

`thumbnails`

When Confluence generates a thumbnail of an image (for example when the gallery macro is used), the resulting thumbnail is stored in this directory for quicker retrieval on subsequent accesses. This directory is essentially a thumbnail cache, and deleting files from this directory simply means the thumbnail will have to be regenerated on the next access.

Page tree