Skip to content

Scan8 is a Kubernetes-based rapid URL/File scan system that allows to submit a list of URLs/files and take out the scan results.

License

Notifications You must be signed in to change notification settings

najeebweeranbangsa/Scan8

 
 

Repository files navigation

Scan8 (Open for GSoC and Hacktoberfest)

Scan8 is a distributed scanning system for detecting trojans, viruses, malware, and other malicious threats embedded in files. The system will allow one to submit a list of URLs or files and get the scan results in return.

The project is divided into various modules namely Dashboard, Coordinator Node, Worker Node, and Testing.
The Dashboard provides a responsive web interface for uploading files for new scans and tracking the status of all the submitted scans.
The Coordinator Node listens to updates for new scans, subsequently creating and adding scan jobs to the Redis Queue.
The Worker Node listens to updates for new jobs in Redis Queue and executes them.
The Testing module helps in maintaining the application and facilitating the CI/CD process for the same.

Application Architecture

Scan8 application architecture

Dependencies

  • Language: Python 3.8.10
  • Database: MongoDB
  • Tools: redis-server clamav clamav-daemon

Specific dependencies for the Dashboard, Coordinator and Worker can be found in the respective directories in requirements.txt file.

Local Setup Guide

  1. Clone the current repository to your local machine using git clone.
  2. Install the dependencies as specified in Dependencies section.
  3. Make sure the mongod and clamav-daemon services are running in the background.
  4. Check the .env file to have the appropriate MongoDB and Redis host and port (variables are set to defaults).
  5. Access the terminal and move to the Dashboard directory.
  6. Run the flask application using export FLASK_APP=app.py followed by flask run.
  7. Access another terminal and move to the Coordinator directory.
  8. Run the coordinator node application using python3 app.py.
  9. Access another terminal and move to the Worker directory.
  10. Run the worker node application using python3 app.py.
  11. Create Uploads and Results directories in the project directory.

Usage

  1. After following the Local setup guide, use any web browser to access the IP address mentioned in the terminal after running flask run (by default it is http://127.0.0.1:5000/).
  2. Submit new scans using the New Scan button and track their progress on the dashboard.
  3. The results for the submitted scans can be found in the Results directory as <scan id>_<file_name>.json.

Testing

The application comes with a test suite to help users ensure correct installation and help developers verify any updates.

  1. Ensure the Results and Uploads directories are empty.
  2. Ensure the MongoDB collections are empty.
  3. Ensure the scan8 application is up and running.
  4. Access a terminal and move to the Testing directory.
  5. Run the test suite using python3 app.py -v.
  6. Run a single scan using the Scan8 dashboard and wait till completion.
  7. Run the test suite again using python3 app.py -v.

Demo videos

About

Scan8 is a Kubernetes-based rapid URL/File scan system that allows to submit a list of URLs/files and take out the scan results.

Resources

License

Code of conduct

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 37.9%
  • HTML 34.4%
  • Shell 13.0%
  • Dockerfile 12.1%
  • CSS 2.6%