Modification-Based Multipathing Code for Regression-Safe Deployment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing code patching methods overwrite the original code, leading to potential regression errors and resource hogging during exhaustive retry logic, causing operational failures and outages.
Innovation Solution
Implementing multipathing code that maintains alternate code paths for updated and original versions, allowing seamless switching between them based on error detection and state restoration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If code patching overwrites original code, then updated code is deployed, but regression errors occur and resource hogging happens during exhaustive retry logic
Solution Approach 1:
The code is segmented into multiple paths: an updated code path and an original code path. The system divides the single code execution flow into separate segments that can be independently executed based on error conditions, allowing retry logic to switch between segments rather than exhausting resources on a single overwritten path.
Solution Approach 2:
The system changes the execution parameter from a single fixed code path to a selectable path parameter. By introducing path selection capability, the system can dynamically change which code version executes based on error detection, preventing regression errors from causing system failure while maintaining deployment efficiency.
2Reliability
If exhaustive retry logic is implemented, then system attempts to recover from errors, but resources are hogged causing operational failures
Solution Approach 1:
The system extracts the retry logic from a resource-intensive exhaustive loop and replaces it with a path-switching mechanism. Instead of repeatedly executing the same failing code, the system takes out the problematic retry loop and substitutes it with an efficient path selection approach that switches to alternative code versions, reducing resource consumption while maintaining error recovery capability.
Solution Approach 2:
The system creates a copy of the original code path alongside the updated code path. This copy serves as a fallback option when the updated path fails, enabling error recovery without requiring exhaustive retries of the same failing code, thereby reducing resource consumption while maintaining reliability.
3Adaptability or versatility
If original code is overwritten during patching, then updated functionality is achieved, but fallback to original code becomes impossible
Solution Approach 1:
The system merges the updated code path and original code path into a unified multipathing structure. Both code versions coexist within the same system framework, allowing the system to combine the benefits of updated functionality with the safety net of original code, enabling both adaptability and ease of repair simultaneously.
Solution Approach 2:
The system performs preliminary action by preserving the original code before overwriting it with updated code. By proactively maintaining a copy of the original code path before deployment, the system ensures rollback capability is already in place, making repair easy while maintaining full update capability.
Data Source
AI summary
Modification-based multipathing code, includes receiving a modification to a portion of code of an application; modifying the application to include a portion of multipathing code comprising a first path corresponding to the portion of code reflecting the modification and a second path corresponding to the portion of code independent of the modification; and configuring the application to execute either the first path or the second path as a default execution path.