File System User Modification Tracking via Extended Attributes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current computer storage systems lack efficient mechanisms for tracking and managing user modifications to files, which hinders auditing, antivirus scanning, and data tiering processes.
Innovation Solution
Implementing a user file modification tracking system that utilizes extended attributes to maintain a FIFO queue of users who have edited a file, allowing for the tracking of the last N non-blacklisted users who have modified a particular file, with options to configure global settings for tracking and blacklist specifications.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If user file modification tracking is implemented using extended attributes and FIFO queues, then auditing capability and security monitoring are improved, but device complexity and storage overhead increase
Solution Approach 1:
The patent embeds the user tracking queue directly within the file's extended attributes, creating a nested structure where the queue is contained inside the file metadata. This eliminates the need for separate external tracking structures and reduces system complexity while maintaining auditing capabilities
Solution Approach 2:
Instead of maintaining complex original tracking structures, the patent creates simplified copies of user identification data (user IDs) and stores them in the FIFO queue within extended attributes. This copying approach reduces the complexity of tracking while preserving the essential auditing information needed for security monitoring
2Measurement precision
If tracking queue stores detailed user modification information, then measurement precision of user activity is improved, but loss of time for processing and retrieving data increases
Solution Approach 1:
The patent extracts only the essential user identification information (user IDs) from complete user activity records and stores only these extracted identifiers in the FIFO queue. This selective extraction maintains sufficient precision for auditing purposes while significantly reducing the amount of data that needs to be processed and retrieved, thereby reducing time loss
Solution Approach 2:
The patent implements partial tracking by storing only the necessary portion of user activity data (user IDs) rather than complete activity records. This partial action approach provides adequate measurement precision for security auditing while minimizing the processing and retrieval time required
3Productivity
If FIFO queue limits are set to track only last N users, then productivity of auditing processes is improved, but loss of information about earlier users increases
Solution Approach 1:
The patent changes the parameter of information retention by implementing a configurable limit N for the FIFO queue size. This parameter change allows the system to balance auditing efficiency (by limiting queue size) against information retention (by adjusting N), enabling flexible optimization based on specific auditing requirements
Solution Approach 2:
The patent makes the tracking queue dynamic by implementing automatic removal of oldest user entries when the queue reaches capacity N. This dynamic behavior allows the system to maintain a manageable size for efficient auditing while automatically adapting to varying levels of user activity, balancing productivity gains with information retention
Data Source
AI summary
Techniques are provided for tracking users modifying, writing, or editing a file. In an example, a file system maintains a first-in-first-out queue that logs a finite set of users that have most-recently modified a file. This queue can be maintained in an extended attribute of an Mode that corresponds to a file. Where a user modifies a file, and the user is currently identified in the queue, the user can be removed from the queue. Where the user modifies a file, is not currently identified in the queue, and the queue is full, an oldest user in the queue can be removed from the queue. Then, the user can be added to the back of the queue.