Azure pipeline docker image not found

CrowdStrike Container Image Scan

This script will scan a container and return response codes indicating pass/fail status.

Specifically, this script:

Tags your image using docker tag or podman tag
Authenticates to CrowdStrike using your OAuth2 API keys
Pushes your image to CrowdStrike for evaluation using docker push, after which CrowdStrike performs an Image Scan
Parses returned scan report, generating return error codes as needed

All output is sent to stdout/stderr.

Prerequisites

This sample/demo script requires the Docker Engine API python library or the Bindings for Podman RESTful API and the FalconPy SDK. These can be installed via pip:

OAuth2 API Key Prerequisites

A CrowdStrike OAuth2 API keys with the following permissions is required:

Permission	Needed scopes
Falcon Container Image	`read` and `write`

Docker Python Prerequisites

$ pip3 install docker crowdstrike-falconpy

Podman Python Prerequisites

$ pip3 install podman crowdstrike-falconpy

Once the Podman python dependencies are installed, configure the URI path for the service.

$ export CONTAINER_HOST="unix:///var/run/podman/podman.sock"

Install all Python Prerequisites

$ pip3 install -r requirements.txt

Usage

$ python3 cs_scanimage.py --help
usage: cs_scanimage.py [-h] -u CLIENT_ID -r REPO [-t TAG]
                       [-c {us-1,us-2,eu-1,us-gov-1}] [-s SCORE]
                       [--json-report REPORT]
                       [--log-level {DEBUG,INFO,WARNING,ERROR,CRITICAL}]
                       [-R RETRY_COUNT] [--plugin] [--user-agent USERAGENT]

optional arguments:
  -h, --help            show this help message and exit
  --json-report REPORT  Export JSON report to specified file
  --log-level {DEBUG,INFO,WARNING,ERROR,CRITICAL}
                        Set the logging level
  --plugin              Prints the report as json to stdout
  --user-agent USERAGENT
                        HTTP User agent to use for API calls

required arguments:
  -u CLIENT_ID, --clientid CLIENT_ID
                        Falcon OAuth2 API ClientID
  -r REPO, --repo REPO  Container image repository
  -t TAG, --tag TAG     Container image tag
  -c {us-1,us-2,eu-1,us-gov-1}, --cloud-region {us-1,us-2,eu-1,us-gov-1}
                        CrowdStrike cloud region
  -s SCORE, --score_threshold SCORE
                        Vulnerability score threshold
  -R RETRY_COUNT, --retry_count RETRY_COUNT
                        Scan report retry count

Note that CrowdStrike Falcon OAuth2 credentials may be supplied also by the means of environment variables: FALCON_CLIENT_ID, FALCON_CLIENT_SECRET, and FALCON_CLOUD_REGION. Establishing and retrieving OAuth2 API credentials can be performed at https://falcon.crowdstrike.com/support/api-clients-and-keys.

FALCON_CLIENT_ID and FALCON_CLIENT_SECRET can be set via environment variables for automation.

Example Scans

Example 1:

$ python cs_scanimage.py --clientid FALCON_CLIENT_ID --repo <repo> --tag <tag> --cloud-region <cloud_region>

please enter password to login
Password:

The command above will return output similar to:

INFO    Downloading Image Scan Report
INFO    Searching for vulnerabilities in scan report...
INFO    Searching for leaked secrets in scan report...
INFO    Searching for malware in scan report...
INFO    Searching for misconfigurations in scan report...
WARNING Alert: Misconfiguration found
INFO    Vulnerability score threshold not met: '0' out of '500'

Example 2:

The script provided was built to score vulnerabilities on a scale show below.

critical_score = 2000
high_score = 500
medium_score = 100
low_score = 20

The default value to return a non-zero error code for vulnerabilties is one high vulnerabilty. This can be overridden by providing the -s parameters to the script.

The example below will accomodate vulnerabilities with a sum of 1500.

$ python cs_scanimage.py --clientid FALCON_CLIENT_ID --repo <repo> --tag <tag> \
    --cloud-region <cloud_region> -s 1500

The echo $? command can be utilized to review the return code, e.g:

echo $?
1

The echo $? above displays the returned code with the following mappings:

VulnerabilityScoreExceeded = 1
Malware = 2
Secrets = 3
Success = 0
Misconfig = 0
ScriptFailure = 10

Running the Scan using CICD

You can use the container-image-scan GitHub Action in your GitHub workflows. Checkout the action at https://github.com/marketplace/actions/crowdstrike-container-image-scan
Pipeline examples, including the GitHub Action, can be found at the CrowdStrike image-scan-example repository.

Using the `container-image-scan` Container

export FALCON_CLIENT_ID=<client_id>
export FALCON_CLIENT_SECRET=<client_secret>

docker run -it --rm -e FALCON_CLIENT_ID -e FALCON_CLIENT_SECRET \
    -v /var/run/docker.sock:/var/run/docker.sock \
    quay.io/crowdstrike/container-image-scan:latest --repo <repo> --tag <tag>

You can use any argument that is supported by the script. For example, to set the cloud region:

docker run -it --rm -e FALCON_CLIENT_ID -e FALCON_CLIENT_SECRET \
    -v /var/run/docker.sock:/var/run/docker.sock \
    quay.io/crowdstrike/container-image-scan:latest --repo <repo> --tag <tag> -c us-2

crowdstrike / container-image-scan Goto Github PK

container-image-scan's Introduction

CrowdStrike Container Image Scan

Prerequisites

OAuth2 API Key Prerequisites

Docker Python Prerequisites

Podman Python Prerequisites

Install all Python Prerequisites

Usage

Example Scans

Example 1:

Example 2:

Running the Scan using CICD

Using the container-image-scan Container

container-image-scan's People

Contributors

Stargazers

Watchers

Forkers

container-image-scan's Issues

Recommend Projects

Recommend Topics

Recommend Org

Using the `container-image-scan` Container