Taintube: Taint Tracking for Security Analysis of Kubernetes Manifests

Development Environment

We use Conda to manage the virtual environment for KubeSec. To see the content of the environment, see environment.yml.

To use the environment, use the following commands.

# Create
conda env create -f environment.yml

# Activate
conda activate KUBESEC

# Deactivate
conda deactivate

# Export (if you modify the environment)
conda env export --from-history > environment.yml

Running the tool

The tool is available as a Docker image: https://hub.docker.com/repository/docker/akondrahman/sli-kube

Instruction to run the tool from Docker Hub:

• docker rm $(docker ps -a -f status=exited -f status=created -q)
• docker rmi -f $(docker images -a -q)
• docker pull akondrahman/sli-kube
• docker images -a
• docker run --rm -it akondrahman/sli-kube bash
• cd SLI-KUBE-WORK/KubeSec-master/
• python3 main.py

Build and run locally

You can also build the docker container locally.

After running the tool, you will find a CSV file called 'slikube_results.csv' in the scanned directory.

# Build the image
docker build -t slikube .

# Run the container 
# Replace '/Users/phu/Desktop/tf-open-source/aws-eks-base' with your local path
docker run --rm -v /Users/phu/Desktop/tf-open-source/aws-eks-base:/iac --name slikube slikube /iac

Collaborators

Akond Rahman (Lead), Rahul Pandita, and Shazibul Islam Shamim