Configuration Management System Using Push-Based Event Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional configuration management systems face challenges in scaling to track changes efficiently in large IT environments, leading to inconsistencies due to limitations in polling intervals and the inability to record out-of-band changes, which can affect dependent entities, resulting in ripple effects and inconsistencies.
Innovation Solution
A method that learns configuration dependencies in a production environment, builds a model, subscribes to events, and correlates them to generate an enhanced model for predicting configuration errors, using a test environment to mimic production setups and introducing a push-based component to detect changes and update the Configuration Management Database (CMDB) quickly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional discovery tools perform periodic scan-and-gather sweeps to track configuration changes, then changes can be detected systematically, but the polling interval cannot be decreased to update changes more quickly due to significantly increased overhead
Solution Approach 1:
Instead of having the discovery tool actively scan and gather configuration data periodically (pull-based approach), the invention inverts the approach by having agents on monitored systems push configuration change events to the discovery tool (push-based approach). This eliminates the need for frequent polling while ensuring changes are detected promptly when they occur.
Solution Approach 2:
The invention replaces the mechanical periodic scanning mechanism with an event-driven notification mechanism. Configuration changes are detected through system events (file system events, process events, registry events, etc.) that trigger automatic updates, substituting the heavy mechanical polling system with a lightweight event subscription system.
2Loss of information
If the entire scan sweep of a large IT environment is performed periodically, then configuration data can be collected comprehensively, but the process takes several hours thereby inviting measure inconsistency
Solution Approach 1:
The invention extracts and monitors only the specific configuration parameters that have changed, rather than performing comprehensive periodic scans of the entire IT environment. Configuration change events are captured at the source systems and pushed immediately to the CMDB, eliminating the need for lengthy periodic resynchronization scans.
Solution Approach 2:
The invention performs preliminary configuration monitoring at the source systems through deployed agents that watch for changes before they propagate. By detecting changes at the origin and immediately notifying the CMDB, the system maintains consistency without waiting for periodic comprehensive scans to complete.
3Productivity
If out-of-band changes are executed using stand-alone scripts or directly through product GUIs in critical situations, then critical changes can be implemented quickly, but the system state changes without creating a record of the change
Solution Approach 1:
The invention implements feedback by deploying agents on monitored systems that continuously observe configuration state and automatically generate change records. When out-of-band changes occur through scripts or GUIs, the agents detect the resulting configuration events and push records to the CMDB, ensuring visibility of all changes regardless of the method used.
Solution Approach 2:
The system performs self-service configuration monitoring through agents that autonomously detect changes on monitored systems and automatically generate and push change records to the CMDB without requiring manual intervention or special handling for different change methods.
4Device complexity
If configuration dependencies are not tracked, then the configuration management system is simpler to implement, but changes to one software entity affect several other dependent entities creating a ripple of inconsistencies
Solution Approach 1:
The invention segments the configuration management approach by deploying distributed agents on individual systems that independently monitor local configuration events. Each agent handles its own system's changes and pushes relevant events to the CMDB, avoiding the need for complex centralized dependency tracking while maintaining consistency through localized event capture.
Data Source
AI summary
Methods and arrangements for tracking changes in an information technology environment. Configuration dependencies are learned in a production environment, and a model of the production environment is built. Events on a target system in the production environment are subscribed to. Events are received and correlated with the model to generate an enhanced model, and the enhanced model is thereupon employed to predict configuration errors.