KB-1255 How to address high disk usage in self-managed Appian environments

This article details corrective actions to be taken in situations where an Appian environment is running low on disk space. In addition, some ideas to optimize disk usage and prevent potential issues caused due to insufficient disk space are presented.

Managing high disk usage

The standard approach when addressing high disk usage is isolating the particular objects (files or folders) that are consuming the most space and then following the appropriate steps to either clean up or relocate them. A third option of adding additional capacity to existing disks may also be considered if sufficient space cannot be reclaimed from cleanup or relocation activities.

Review the current disk usage metrics using the File Explorer (Windows) or running df -h (UNIX). Once the top disk/partition is identified, review the contents and their individual usage within. For example, running du -h --max-depth=1 in UNIX or WindirStat on Windows can display the usage per file or subfolder within a particular folder. Repeat the process until the top consumers have been identified.

The following sections detail the appropriate cleanup steps for each component of the Appian environment

Application Server

For application servers the objects that typically grow and need to be cleaned up are:

  • Log files
  • Application server cache

Application log files described in the documentation can be compressed or moved to another location to save disk space. Old log files can also be cleaned up periodically using the cleanup script. If log files are observed to be growing at a faster than normal rate, this may be due to enabling DEBUG for one or more classes. Check for DEBUG messages in the log files and make sure to disable any loggers that are not needed. Note: Do not touch files within the <APPIAN_HOME>/services/data/kafka-logs directory. For steps on dealing with Kafka logs, see below. 

Appian Technical Support may also be able to assist in identifying possible causes for excessive growth in log file size.

Kafka Logs

If the Kafka transaction logs are using large amounts of disk space, refer to KB-1455.

Search Server

The search server index data stored in <APPIAN_HOME>/_admin/search_local/ can be cleared if required to release some disk space, following the steps in KB-1082. In addition, the search server log files stored in <APPIAN_HOME>/logs/search-server/ can also be periodically deleted.

Engines

The cleanup script can be used to manage engine files (*.kdb) and archived processes. KDB files should not be modified manually.  

Documents

Business data such as files and folders within Appian’s document management are also stored as physical files on disk within the <APPIAN_HOME>/_admin/accdocs*/ directories. By following the steps below, It is possible to pinpoint the exact Knowledge Center in Appian that is consuming the most disk space:

  1. Navigate to <APPIAN_HOME>/_admin/.
  2. Execute du -h --max-depth=1 or use WinDirStat and navigate into the particular accdocs folder that records the most disk usage.
  3. The folder names correspond to Appian knowledge centers. With this in mind, run du -h --max-depth=1 or use WinDirStat within the accdocs folder to see top knowledge centres by disk usage and note the name/number of the folder
  4. Login to Appian and navigate to /suite/kc/<FOLDER_NUMBER>. This allows you to browse the top consumers by name and make it easier to review and cleanup unwanted documents or folders.

Note: For System Knowledge Centers except for the Temporary Documents Knowledge Center, contact Appian Technical Support to coordinate the necessary cleanup operations.

Alternately, this command can be used in UNIX to know from the OS level who the top 10 consumers are within a particular folder:

find -type f -exec du -Sh {} + | sort -rh | head -n 10 

This returns the internal names for each content object, which can be passed as an argument into getContentDetailsByInternalName() to get more information.

Recommendations to optimize Disk usage

  • Perform regular Data Maintenance
  • Set file upload and Knowledge centre limits such as conf.content.max.temporary.uploaded.files.age, resources.appian.ac.collaboration.MAX_UPLOAD_FILE_SIZE, or server.conf.collaboration.DEFAULTSPACE.

Affected Versions

This article applies to all versions of Appian.

Last Reviewed: November 2019

Related
Recommended