DocumentationExperience Cloud KCS

Resolving high disk utilization from large log files in Adobe Experience Manager publisher nodes

Last update: August 31, 2026

This article explains how to troubleshoot high disk utilization on Adobe Experience Manager (AEM) Publish instances when large log files cause a Critical storage status. To fix this, identify the largest log files, remove or archive old logs and thread dumps, review the logging configuration, investigate recurring application errors, and monitor disk utilization.

Description description

Environment

  • Adobe Experience Manager 6.5 LTS
  • Self-managed, on-premises deployment
  • Publish instances with file-system access

Symptoms

  • In the AEM System Monitoring dashboard, one or more Publish instances show a Critical status under Storage.
  • Disk utilization is abnormally high on the affected instances.
  • Large log files are present, especially error.log files, with individual files exceeding 1 GB in size.

Cause

Excessive log file growth results from application errors, recurring stack traces, and verbose logging configurations. Enabling DEBUG-level logging for certain packages also contributes to rapid log growth. Cleaning up old logs and thread dumps can restore disk capacity, but you must address the underlying application and logging issues to prevent recurrence.

Resolution resolution

Follow these steps to reduce disk utilization and prevent excessive log growth on the affected Publish instances:

  1. Connect to each affected Publish instance using SSH or a file browser.
  2. List the log files by size to identify the files consuming the most disk space. On Linux systems, use ls -lhS /path/to/aem/logs/.
  3. Delete or archive old log files and thread dumps that are no longer needed.
  4. Before deleting any logs, retain logs required for compliance or an ongoing investigation in another location.
  5. Review the logging configuration for the affected Publish instance.
  6. Check whether DEBUG logging is enabled for packages related to custom code or high-volume components.
  7. If DEBUG logging is not required for active troubleshooting, change the package to an appropriate production logging level, such as INFO or WARN.
  8. Review the other loggers and confirm that only necessary information is logged at higher verbosity levels.
  9. Analyze the error logs for recurring exceptions, such as NullPointerException, and address the underlying code issues that generate excessive stack traces.
  10. Review exception handling in the application code to ensure that expected or recoverable errors are handled appropriately and do not generate unnecessary stack traces.
  11. Use the AEM System Monitoring dashboard and external monitoring tools to track disk usage and log growth trends.
  12. Configure alerts to identify abnormal disk or log growth before it affects the service.
  13. After cleanup, confirm that disk utilization returns to normal and that the affected Publish instance shows a healthy status in the monitoring dashboard.

Notes:

  • This procedure applies to self-managed, on-premises Adobe Experience Manager 6.5 LTS deployments where you have access to the Publish instance file system.
  • Do not apply the direct file-system cleanup steps to Adobe Experience Manager as a Cloud Service or Adobe Managed Services environments. Those deployment models use different log-access and infrastructure-management procedures.
recommendation-more-help
experience-cloud-kcs-help-kbarticles