Zenith

Smart software crash detection system

Comment

ZENITH'S OBJECTIVE Addressing challenges related to crash detection, categorization, and resolution within software development cycles. CORE FOCUS Streamlining the entire process for efficient management of crash data and facilitating swift issue resolution. OVERALL GOAL Enhancing the software development cycle by providing a comprehensive solution to effectively manage crashes. CRASH MANAGEMENT

  • Implemented a robust platform for the organized storage and management of crash logs.

USER INTERFACE

  • Placed emphasis on creating a user-friendly interface to ensure seamless navigation and ease of use.

ERROR CATEGORIZATION

  • Classified errors into critical, fatal, and warning categories to prioritize attention and resolution.

FOCUS ON COMMON ERRORS

  • Developed a strategy to help developers concentrate on resolving the most frequently occurring errors, enhancing overall efficiency.

COMPREHENSIVE ERROR DETAILS

  • Ensured in-depth information on error details to provide a thorough understanding of each issue.

ALGORITHM OVERVIEW Backend Engine:

  • Utilized Python as the backend engine, implementing the entire algorithm in Python, including some shell scripting.
  • Employed MongoDB for efficient storage and retrieval of error data.

Error Classification Model:

  • Developed a custom error classification Support Vector Machine (SVM) model using scikit-learn (sklearn) for categorizing errors into fatal, critical, and warning types.

Comprehensive Crash Report:

  • Generated detailed crash reports in a log file, capturing error output alongside system details such as RAM usage, CPU core percentage, and data transfer during the process.

Similarity Clustering:

  • Implemented a similarity algorithm to compare log files, creating clusters of similar log files for efficient analysis.

Versatility in Software Development:

  • Designed the algorithm to be generic, enabling its usage at any stage of software development, whether it be testing, implementation, or deployment.

Frontend Technology:

  • Developed a web application using React for the user interface.

Database Integration:

  • Pulled crash data from MongoDB using a Node.js REST API for seamless data retrieval and presentation.

Hosting on AWS EC2:

  • Deployed both the React web application and Node.js API on an AWS EC2 Virtual Server for efficient hosting.

Desktop Application Development:

  • Leveraged Electron.js to transform the web application and API into a desktop application, enhancing versatility and accessibility.

Theme Options:

  • Implemented both light and dark themes to cater to user preferences.

Optimize Software Issue Resolution

  • Enhance the efficiency of resolving software issues by refining the error classification process.

Integration with Agile Process

  • Establish a connection between the application and the entire agile software development process, spanning from initial planning to deployment.

Algorithmic Improvement

  • Focus on developing an improved algorithm and incorporating automated log file clustering based on a document similarity score.

Streamlined Crash Resolution

  • Simplify and expedite software crash resolution for developer teams, aiming for a quick and straightforward process.
Share this project:

Updates