Diskover 2.4.3 Release Notes

Overview

Diskover 2.4.3 is a maintenance release focused on platform stability and performance improvements. This release addresses a couple of minor issues with task scheduling in multi-worker environments and delivers significant login performance optimizations for large-scale deployments

Key highlights include fixes for task duplication when using "ANY" worker configurations, enhanced error handling for scheduled task execution, and substantial login time improvements for environments with thousands of Elasticsearch indices.

Enhancements

Scanner Improvements

Task ID in Index Metadata
[DEV-648]

Scan operations now write the task_id field to indexinfo documents in Elasticsearch indices, enabling users to identify which scheduled task built a specific index. This enhancement improves traceability and simplifies troubleshooting by directly linking indices to their originating scan tasks.

Web Interface

Login Performance Optimization
[DEV-462]

Significant performance improvements for login times, particularly on systems with large numbers of Elasticsearch indices and LDAP groups. Environments with thousands of indices (e.g., ~5,000) previously experienced login delays of up to two minutes, especially for non-admin users.

The optimization reduces Elasticsearch round trips during permission checking and improves index information retrieval efficiency. Login times have been reduced from minutes to seconds in large-scale deployments.

Bug Fixes

Scanner Fixes

Task Duplication with Multiple Workers
[DEV-615]

Fixed a race condition that caused tasks to duplicate across workers when configured with the "ANY" worker setting in multi-worker environments. Symptoms included tasks executing multiple times simultaneously, Elasticsearch resource_already_exists_exception errors during index creation, and exponentially longer scan times due to duplicate processing.

The fix enhances task claim/release locking in diskoverd, corrects the is_task_idle() function to properly include 'waiting' status, and adds [TASK_CLAIM] logging for improved troubleshooting.

Task Thread Crash Causing Tasks to Hang
[DEV-654]

Fixed an issue where the get_tasks thread could silently crash when the tasks API returned an unexpected null response, causing scheduled tasks to fail to execute. The fix adds proper null response handling with explicit error logging, a 60-second timeout to API requests, and exception handling for socket timeout errors.

Technical Notes

Compatibility

All fixes are backward compatible with existing Diskover 2.4.x deployments
No configuration changes required for upgrade

Performance

Multi-worker task scheduling stability significantly improved
Login performance enhanced for large-scale environments with thousands of indices
Reduced API timeouts through improved error handling

Upgrade Considerations

Recommended for all users running Diskover 2.4.x, particularly those with multi-worker configurations
Users experiencing task duplication or slow login times should upgrade immediately

For detailed information on any feature or fix, please contact Diskover support.