dockle
Container Image Linter for Security, Helping build the Best-Practice Docker Image, Easy to start
Top Related Projects
Find vulnerabilities, misconfigurations, secrets, SBOM in containers, Kubernetes, code repositories, clouds and more
A vulnerability scanner for container images and filesystems
Vulnerability Static Analysis for Containers
The Docker Bench for Security is a script that checks for dozens of common best-practices around deploying Docker containers in production.
A tool for exploring each layer in a docker image
Dockerfile linter, validate inline bash, written in Haskell
Quick Overview
Dockle is a container image linter for security, helping developers and DevOps teams identify and fix potential vulnerabilities in Docker images. It performs comprehensive checks against best practices and security standards, providing actionable insights to improve container security.
Pros
- Easy to use with a simple CLI interface
- Integrates well with CI/CD pipelines for automated security checks
- Provides detailed explanations and remediation suggestions for identified issues
- Supports custom rule configuration to adapt to specific project needs
Cons
- Limited to Docker image analysis, not covering other container technologies
- May produce false positives in some cases, requiring manual verification
- Requires regular updates to stay current with the latest security best practices
- Performance can be slow for large images with many layers
Getting Started
To get started with Dockle, follow these steps:
- Install Dockle:
# For macOS
brew install goodwithtech/r/dockle
# For Linux
wget https://github.com/goodwithtech/dockle/releases/download/v0.4.5/dockle_0.4.5_Linux-64bit.tar.gz
tar zxvf dockle_0.4.5_Linux-64bit.tar.gz
sudo mv dockle /usr/local/bin
- Run Dockle on a Docker image:
dockle <image_name>
- To ignore specific checks, use the
--ignore
flag:
dockle --ignore CIS-DI-0001,CIS-DI-0005 <image_name>
- For CI/CD integration, use the exit code to determine success or failure:
dockle <image_name>
if [ $? -ne 0 ]; then
echo "Dockle found issues in the image"
exit 1
fi
For more advanced usage and configuration options, refer to the official documentation on the GitHub repository.
Competitor Comparisons
Find vulnerabilities, misconfigurations, secrets, SBOM in containers, Kubernetes, code repositories, clouds and more
Pros of Trivy
- Broader scanning capabilities: Covers container images, filesystems, git repositories, and Kubernetes clusters
- Faster scanning speed, especially for large images
- More comprehensive vulnerability database, including multiple sources
Cons of Trivy
- Higher resource consumption during scans
- More complex configuration for advanced use cases
- Steeper learning curve for new users
Code Comparison
Dockle usage:
dockle image_name
Trivy usage:
trivy image image_name
Both tools offer simple command-line interfaces for basic scanning. However, Trivy provides more options for customization:
trivy image --severity HIGH,CRITICAL --ignore-unfixed image_name
This command scans for only high and critical vulnerabilities, ignoring those without fixes.
Dockle focuses on Dockerfile and image best practices, while Trivy primarily targets vulnerability scanning. Trivy's broader scope and more frequent updates make it suitable for comprehensive security checks, whereas Dockle excels in Docker-specific linting and best practice enforcement.
A vulnerability scanner for container images and filesystems
Pros of Grype
- More comprehensive vulnerability scanning, covering a wider range of package ecosystems
- Faster scanning speed, especially for large container images
- Active development with frequent updates and community support
Cons of Grype
- Focused primarily on vulnerability scanning, lacking some container-specific checks
- May produce more verbose output, requiring additional filtering or processing
- Steeper learning curve for advanced features and customizations
Code Comparison
Dockle (shell command):
dockle --exit-code 1 --exit-level fatal myimage:latest
Grype (shell command):
grype myimage:latest
Both tools can be easily integrated into CI/CD pipelines, but Grype offers more flexibility in output formats and scanning options. Dockle provides a simpler interface for quick container checks, while Grype excels in comprehensive vulnerability analysis across various package types and ecosystems.
Dockle is more focused on Docker best practices and CIS benchmarks, making it ideal for container-specific security checks. Grype, on the other hand, offers broader vulnerability scanning capabilities, making it suitable for a wider range of security assessments beyond just containers.
Vulnerability Static Analysis for Containers
Pros of Clair
- More comprehensive vulnerability scanning, covering a wider range of operating systems and package managers
- Better suited for enterprise-level deployments with scalable architecture
- Supports integration with various container registries and CI/CD pipelines
Cons of Clair
- More complex setup and configuration compared to Dockle
- Requires additional infrastructure components (database, API server)
- Steeper learning curve for new users
Code Comparison
Dockle can be run with a simple command:
dockle image:tag
Clair requires more setup, including database configuration:
clair:
database:
type: pgsql
options:
source: host=postgres port=5432 user=clair dbname=clair sslmode=disable
Both tools provide JSON output for integration with other tools, but Clair's output is more detailed:
Dockle:
{
"summary": {
"fatal": 0,
"warn": 2,
"info": 1,
"pass": 10
}
}
Clair:
{
"Layer": {
"Name": "sha256:5f70bf18a086007016e948b04aed3b82103a36bea41755b6cddfaf10ace3c6ef",
"NamespaceName": "alpine:3.10",
"Features": [
{
"Name": "musl",
"NamespaceName": "alpine:3.10",
"Version": "1.1.22-r3",
"Vulnerabilities": [
{
"Name": "CVE-2019-14697",
"NamespaceName": "alpine:3.10",
"Description": "In musl libc through 1.1.23, wcsnrtombs...",
"Severity": "High"
}
]
}
]
}
}
The Docker Bench for Security is a script that checks for dozens of common best-practices around deploying Docker containers in production.
Pros of docker-bench-security
- More comprehensive security checks, covering a wider range of Docker security best practices
- Actively maintained by Docker, ensuring up-to-date security recommendations
- Provides detailed explanations and remediation steps for each security issue
Cons of docker-bench-security
- Primarily focused on Docker host and daemon configuration, less emphasis on container image analysis
- Can be more complex to interpret results, especially for beginners
- Requires root access to run some checks, which may not be feasible in all environments
Code Comparison
docker-bench-security:
#!/bin/sh
# Docker Bench for Security v1.3.4
# https://github.com/docker/docker-bench-security
# Check for containers with excessive capabilities
check_4_5() {
id_4_5="4.5"
desc_4_5="Ensure Content trust for Docker is Enabled"
check_4_5="$id_4_5 - $desc_4_5"
starttestjson "$id_4_5" "$desc_4_5"
dockle:
func (s *Scanner) ScanImage(ctx context.Context, imageName string, opts types.ScanOptions) (types.Results, error) {
log.Logger.Debugf("Start scanning %s", imageName)
results := types.Results{}
err := s.initializeDockerClient(ctx)
if err != nil {
return nil, xerrors.Errorf("unable to initialize Docker client: %w", err)
}
A tool for exploring each layer in a docker image
Pros of dive
- Provides a detailed layer-by-layer analysis of Docker images
- Offers an interactive CLI interface for exploring image contents
- Helps identify large layers and potential areas for image optimization
Cons of dive
- Focused solely on image analysis, lacking security scanning features
- Does not provide compliance checks or best practice recommendations
- Limited to local image analysis, without integration for CI/CD pipelines
Code comparison
dive:
func analyzeImageLayers(image *image.Image) ([]layer.Layer, error) {
layers := make([]layer.Layer, len(image.RootFS.DiffIDs))
for idx, diffID := range image.RootFS.DiffIDs {
layers[idx] = layer.NewLayer(diffID)
}
return layers, nil
}
dockle:
func (a *Assessor) Assess() ([]types.Assessment, error) {
var assessments []types.Assessment
for _, checker := range a.Checkers {
assessment, err := checker.Check()
if err != nil {
return nil, err
}
assessments = append(assessments, assessment)
}
return assessments, nil
}
While dive focuses on analyzing image layers, dockle emphasizes security and compliance checks. The code snippets reflect their different purposes, with dive handling layer analysis and dockle performing assessments using various checkers.
Dockerfile linter, validate inline bash, written in Haskell
Pros of Hadolint
- More comprehensive Dockerfile linting rules, including shell script analysis
- Integrates well with CI/CD pipelines and popular code editors
- Supports custom rule creation using Haskell
Cons of Hadolint
- Focused solely on Dockerfile linting, lacking container image scanning capabilities
- May produce false positives in some cases, requiring careful configuration
- Steeper learning curve for custom rule development due to Haskell
Code Comparison
Hadolint example:
FROM ubuntu:18.04
RUN apt-get update && apt-get install -y python
COPY . /app
CMD ["python", "app.py"]
Dockle example:
FROM alpine:3.12
RUN apk add --no-cache python3
COPY . /app
USER nobody
CMD ["python3", "app.py"]
Hadolint would flag issues like using the latest tag and potential outdated packages, while Dockle would additionally check for security best practices like running as a non-root user.
Both tools aim to improve Dockerfile quality and security, but Hadolint focuses on linting while Dockle offers broader container image analysis. Hadolint provides more detailed Dockerfile checks, while Dockle includes additional security scans for the resulting container image.
Convert designs to code with AI
Introducing Visual Copilot: A new AI model to turn Figma designs to high quality code using your components.
Try Visual CopilotREADME
Dockle - Container Image Linter for Security, Helping build the Best-Practice Docker Image, Easy to start
Dockle
helps you:
- Build Best Practice Docker images
- Build secure Docker images
- Checkpoints includes CIS Benchmarks
$ brew untap goodwithtech/dockle # who use 0.1.16 or older version
$ brew install goodwithtech/r/dockle
$ dockle [YOUR_IMAGE_NAME]
See Installation and Common Examples
Checkpoints Comparison
TOC
- Features
- Comparison
- Installation
- Quick Start
- Checkpoint Summary
- Common Examples
- Continuous Integration
- Checkpoint Details
- CIS's Docker Image Checkpoints
- Dockle Checkpoints for Docker
- Dockle Checkpoints for Linux
- Credits
- Roadmap
Features
- Detect container's vulnerabilities
- Helping build best-practice Dockerfile
- Simple usage
- Specify only the image name
- See Quick Start and Common Examples
- CIS Benchmarks Support
- High accuracy
- DevSecOps
- Suitable for CI such as Travis CI, CircleCI, Jenkins, etc.
- See CI Example
Comparison
 | Dockle | Hadolint | Docker Bench for Security | Clair |
---|---|---|---|---|
Target | Image | Dockerfile | Host Docker Daemon Image Container Runtime | Image |
How to run | Binary | Binary | ShellScript | Binary |
Dependency | No | No | Some dependencies | No |
CI Suitable | â | â | x | x |
Purpose | Security Audit Dockerfile Lint | Dockerfile Lint | Security Audit Dockerfile Lint | Scan Vulnerabilities |
Installation
Homebrew (Mac OS X / Linux and WSL)
You can use Homebrew on Mac OS X or Linux and WSL (Windows Subsystem for Linux).
$ brew install goodwithtech/r/dockle
RHEL/CentOS
VERSION=$(
curl --silent "https://api.github.com/repos/goodwithtech/dockle/releases/latest" | \
grep '"tag_name":' | \
sed -E 's/.*"v([^"]+)".*/\1/' \
) && rpm -ivh https://github.com/goodwithtech/dockle/releases/download/v${VERSION}/dockle_${VERSION}_Linux-64bit.rpm
Debian/Ubuntu
VERSION=$(
curl --silent "https://api.github.com/repos/goodwithtech/dockle/releases/latest" | \
grep '"tag_name":' | \
sed -E 's/.*"v([^"]+)".*/\1/' \
) && curl -L -o dockle.deb https://github.com/goodwithtech/dockle/releases/download/v${VERSION}/dockle_${VERSION}_Linux-64bit.deb
$ sudo dpkg -i dockle.deb && rm dockle.deb
Arch Linux
dockle can be installed from the Arch User Repository using dockle
or dockle-bin
package.
git clone https://aur.archlinux.org/dockle-bin.git
cd dockle-bin
makepkg -sri
Windows
VERSION=$(
curl --silent "https://api.github.com/repos/goodwithtech/dockle/releases/latest" | \
grep '"tag_name":' | \
sed -E 's/.*"v([^"]+)".*/\1/' \
) && curl -L -o dockle.zip https://github.com/goodwithtech/dockle/releases/download/v${VERSION}/dockle_${VERSION}_Windows-64bit.zip
$ unzip dockle.zip && rm dockle.zip
$ ./dockle.exe [IMAGE_NAME]
Microsoft PowerShell 7
if (((Invoke-WebRequest "https://api.github.com/repos/goodwithtech/dockle/releases/latest").Content) -match '"tag_name":"v(?<ver>[^"]+)"') {
$VERSION=$Matches.ver &&
Invoke-WebRequest "https://github.com/goodwithtech/dockle/releases/download/v${VERSION}/dockle_${VERSION}_Windows-64bit.zip" -OutFile dockle.zip &&
Expand-Archive dockle.zip && Remove-Item dockle.zip }
Binary
You can get the latest version binary from releases page.
Download the archive file for your operating system/architecture. Unpack the archive, and put the binary somewhere in your $PATH
(on UNIX-y systems, /usr/local/bin
or the like).
- NOTE: Make sure that it's execution bits turned on. (
chmod +x dockle
)
asdf
You can install dockle with the asdf version manager with this plugin, which automates the process of installing (and switching between) various versions of github release binaries. With asdf already installed, run these commands to install dockle:
# Add dockle plugin
asdf plugin add dockle
# Show all installable versions
asdf list-all dockle
# Install specific version
asdf install dockle latest
# Set a version globally (on your ~/.tool-versions file)
asdf global dockle latest
# Now dockle commands are available
dockle --version
From source
$ GO111MODULE=off go get github.com/goodwithtech/dockle/cmd/dockle
$ cd $GOPATH/src/github.com/goodwithtech/dockle && GO111MODULE=on go build -o $GOPATH/bin/dockle cmd/dockle/main.go
Use Docker
There's a Dockle
image on Docker Hub also. You can try dockle
before installing the command.
$ VERSION=$(
curl --silent "https://api.github.com/repos/goodwithtech/dockle/releases/latest" | \
grep '"tag_name":' | \
sed -E 's/.*"v([^"]+)".*/\1/' \
) && docker run --rm -v /var/run/docker.sock:/var/run/docker.sock \
goodwithtech/dockle:v${VERSION} [YOUR_IMAGE_NAME]
You only need -v /var/run/docker.sock:/var/run/docker.sock
when you'd like to scan the image on your host machine.
Quick Start
Basic
Simply specify an image name (and a tag).
$ dockle [YOUR_IMAGE_NAME]
Result
FATAL - CIS-DI-0009: Use COPY instead of ADD in Dockerfile
* Use COPY : /bin/sh -c #(nop) ADD file:81c0a803075715d1a6b4f75a29f8a01b21cc170cfc1bff6702317d1be2fe71a3 in /app/credentials.json
FATAL - CIS-DI-0010: Do not store credential in ENVIRONMENT vars/files
* Suspicious filename found : app/credentials.json
FATAL - DKL-DI-0005: Clear apt-get caches
* Use 'rm -rf /var/lib/apt/lists' after 'apt-get install' : /bin/sh -c apt-get update && apt-get install -y git
FATAL - DKL-LI-0001: Avoid empty password
* No password user found! username : nopasswd
WARN - CIS-DI-0001: Create a user for the container
* Last user should not be root
INFO - CIS-DI-0005: Enable Content trust for Docker
* export DOCKER_CONTENT_TRUST=1 before docker pull/build
INFO - CIS-DI-0008: Confirm safety of setuid/setgid files
* setuid file: app/suid.txt urw-r--r--
* setgid file: app/gid.txt grw-r--r--
* setuid file: usr/bin/gpasswd urwxr-xr-x
* setgid file: usr/bin/wall grwxr-xr-x
* setuid file: bin/su urwxr-xr-x
* setuid file: bin/umount urwxr-xr-x
* setuid file: bin/mount urwxr-xr-x
* setgid file: usr/bin/ssh-agent grwxr-xr-x
* setuid file: etc/shadow urw-r-----
* setuid file: usr/bin/chsh urwxr-xr-x
* setuid file: usr/bin/chfn urwxr-xr-x
* setuid file: usr/lib/openssh/ssh-keysign urwxr-xr-x
* setgid file: etc/passwd grw-r--r--
* setgid file: sbin/unix_chkpwd grwxr-xr-x
* setgid file: usr/bin/chage grwxr-xr-x
* setuid file: usr/bin/passwd urwxr-xr-x
* setgid file: usr/bin/expiry grwxr-xr-x
* setuid file: usr/bin/newgrp urwxr-xr-x
IGNORE - CIS-DI-0006: Add HEALTHCHECK instruction to the container image
Docker
Also, you can use Docker to use dockle
command as follow.
$ export DOCKLE_LATEST=$(
curl --silent "https://api.github.com/repos/goodwithtech/dockle/releases/latest" | \
grep '"tag_name":' | \
sed -E 's/.*"v([^"]+)".*/\1/' \
)
$ docker run --rm goodwithtech/dockle:v${DOCKLE_LATEST} [YOUR_IMAGE_NAME]
-
If you'd like to scan the image on your host machine, you need to mount
docker.sock
.$ docker run --rm -v /var/run/docker.sock:/var/run/docker.sock ...
Checkpoint Summary
- Details of each checkpoint see CHECKPOINT.md
CODE | DESCRIPTION | LEVELâ» |
---|---|---|
CIS's Docker Image Checkpoints | ||
CIS-DI-0001 | Create a user for the container | WARN |
CIS-DI-0002 | Use trusted base images for containers | FATAL |
CIS-DI-0003 | Do not install unnecessary packages in the container | FATAL |
CIS-DI-0004 | Scan and rebuild the images to include security patches | FATAL |
CIS-DI-0005 | Enable Content trust for Docker | INFO |
CIS-DI-0006 | Add HEALTHCHECK instruction to the container image | WARN |
CIS-DI-0007 | Do not use update instructions alone in the Dockerfile | FATAL |
CIS-DI-0008 | Confirm safety of setuid and setgid files | INFO |
CIS-DI-0009 | Use COPY instead of ADD in Dockerfile | FATAL |
CIS-DI-0010 | Do not store secrets in Dockerfiles | FATAL |
CIS-DI-0011 | Install verified packages only | INFO |
Dockle Checkpoints for Docker | ||
DKL-DI-0001 | Avoid sudo command | FATAL |
DKL-DI-0002 | Avoid sensitive directory mounting | FATAL |
DKL-DI-0003 | Avoid apt-get dist-upgrade | WARN |
DKL-DI-0004 | Use apk add with --no-cache | FATAL |
DKL-DI-0005 | Clear apt-get caches | FATAL |
DKL-DI-0006 | Avoid latest tag | WARN |
Dockle Checkpoints for Linux | ||
DKL-LI-0001 | Avoid empty password | FATAL |
DKL-LI-0002 | Be unique UID/GROUPs | FATAL |
DKL-LI-0003 | Only put necessary files | INFO |
Level
Dockle
has 5 check levels.
LEVEL | DESCRIPTION |
---|---|
FATAL | Be practical and prudent |
WARN | Be practical and prudent, but limited uses (even if official images) |
INFO | May negatively inhibit the utility or performance |
SKIP | Not found target files |
PASS | Not found any problems |
Common Examples
Scan an image
Simply specify an image name (and a tag).
$ dockle goodwithtech/test-image:v1
Result
FATAL - CIS-DI-0001: Create a user for the container
* Last user should not be root
WARN - CIS-DI-0005: Enable Content trust for Docker
* export DOCKER_CONTENT_TRUST=1 before docker pull/build
FATAL - CIS-DI-0006: Add HEALTHCHECK instruction to the container image
* not found HEALTHCHECK statement
FATAL - CIS-DI-0007: Do not use update instructions alone in the Dockerfile
* Use 'Always combine RUN 'apt-get update' with 'apt-get install' : /bin/sh -c apt-get update && apt-get install -y git
FATAL - CIS-DI-0008: Remove setuid and setgid permissions in the images
* Found setuid file: etc/passwd grw-r--r--
* Found setuid file: usr/lib/openssh/ssh-keysign urwxr-xr-x
* Found setuid file: app/hoge.txt ugrw-r--r--
* Found setuid file: app/hoge.txt ugrw-r--r--
* Found setuid file: etc/shadow urw-r-----
FATAL - CIS-DI-0009: Use COPY instead of ADD in Dockerfile
* Use COPY : /bin/sh -c #(nop) ADD file:81c0a803075715d1a6b4f75a29f8a01b21cc170cfc1bff6702317d1be2fe71a3 in /app/credentials.json
FATAL - CIS-DI-0010: Do not store secrets in ENVIRONMENT variables
* Suspicious ENV key found : MYSQL_PASSWD
FATAL - CIS-DI-0010: Do not store secret files
* Suspicious filename found : app/credentials.json
PASS - DKL-DI-0001: Avoid sudo command
FATAL - DKL-DI-0002: Avoid sensitive directory mounting
* Avoid mounting sensitive dirs : /usr
PASS - DKL-DI-0003: Avoid apt-get/apk/dist-upgrade
PASS - DKL-DI-0004: Use apk add with --no-cache
FATAL - DKL-DI-0005: Clear apt-get caches
* Use 'apt-get clean && rm -rf /var/lib/apt/lists/*' : /bin/sh -c apt-get update && apt-get install -y git
PASS - DKL-DI-0006: Avoid latest tag
FATAL - DKL-LI-0001: Avoid empty password
* No password user found! username : nopasswd
PASS - DKL-LI-0002: Be unique UID
PASS - DKL-LI-0002: Be unique GROUP
Scan an image file
$ docker save alpine:latest -o alpine.tar
$ dockle --input alpine.tar
Get or Save the results as JSON
$ dockle -f json goodwithtech/test-image:v1
$ dockle -f json -o results.json goodwithtech/test-image:v1
Result
{
"summary": {
"fatal": 6,
"warn": 2,
"info": 2,
"pass": 7
},
"details": [
{
"code": "CIS-DI-0001",
"title": "Create a user for the container",
"level": "WARN",
"alerts": [
"Last user should not be root"
]
},
{
"code": "CIS-DI-0005",
"title": "Enable Content trust for Docker",
"level": "INFO",
"alerts": [
"export DOCKER_CONTENT_TRUST=1 before docker pull/build"
]
},
{
"code": "CIS-DI-0006",
"title": "Add HEALTHCHECK instruction to the container image",
"level": "WARN",
"alerts": [
"not found HEALTHCHECK statement"
]
},
{
"code": "CIS-DI-0008",
"title": "Remove setuid and setgid permissions in the images",
"level": "INFO",
"alerts": [
"Found setuid file: usr/lib/openssh/ssh-keysign urwxr-xr-x"
]
},
{
"code": "CIS-DI-0009",
"title": "Use COPY instead of ADD in Dockerfile",
"level": "FATAL",
"alerts": [
"Use COPY : /bin/sh -c #(nop) ADD file:81c0a803075715d1a6b4f75a29f8a01b21cc170cfc1bff6702317d1be2fe71a3 in /app/credentials.json "
]
},
{
"code": "CIS-DI-0010",
"title": "Do not store secrets in ENVIRONMENT variables",
"level": "FATAL",
"alerts": [
"Suspicious ENV key found : MYSQL_PASSWD"
]
},
{
"code": "CIS-DI-0010",
"title": "Do not store secret files",
"level": "FATAL",
"alerts": [
"Suspicious filename found : app/credentials.json "
]
},
{
"code": "DKL-DI-0002",
"title": "Avoid sensitive directory mounting",
"level": "FATAL",
"alerts": [
"Avoid mounting sensitive dirs : /usr"
]
},
{
"code": "DKL-DI-0005",
"title": "Clear apt-get caches",
"level": "FATAL",
"alerts": [
"Use 'rm -rf /var/lib/apt/lists' after 'apt-get install' : /bin/sh -c apt-get update \u0026\u0026 apt-get install -y git"
]
},
{
"code": "DKL-LI-0001",
"title": "Avoid empty password",
"level": "FATAL",
"alerts": [
"No password user found! username : nopasswd"
]
}
]
}
Get or Save the results as SARIF
$ dockle -f sarif goodwithtech/test-image:v1
$ dockle -f sarif -o results.json goodwithtech/test-image:v1
Result
{
"version": "2.1.0",
"$schema": "https://raw.githubusercontent.com/oasis-tcs/sarif-spec/master/Schemata/sarif-schema-2.1.0.json",
"runs": [
{
"tool": {
"driver": {
"name": "Dockle",
"informationUri": "https://github.com/goodwithtech/dockle",
"rules": [
{
"id": "CIS-DI-0009",
"shortDescription": {
"text": "Use COPY instead of ADD in Dockerfile"
},
"help": {
"text": "https://github.com/goodwithtech/dockle/blob/master/CHECKPOINT.md#CIS-DI-0009"
}
},
{
"id": "CIS-DI-0010",
"shortDescription": {
"text": "Do not store credential in ENVIRONMENT vars/files"
},
"help": {
"text": "https://github.com/goodwithtech/dockle/blob/master/CHECKPOINT.md#CIS-DI-0010"
}
},
{
"id": "DKL-DI-0005",
"shortDescription": {
"text": "Clear apt-get caches"
},
"help": {
"text": "https://github.com/goodwithtech/dockle/blob/master/CHECKPOINT.md#DKL-DI-0005"
}
},
{
"id": "DKL-LI-0001",
"shortDescription": {
"text": "Avoid empty password"
},
"help": {
"text": "https://github.com/goodwithtech/dockle/blob/master/CHECKPOINT.md#DKL-LI-0001"
}
},
{
"id": "CIS-DI-0005",
"shortDescription": {
"text": "Enable Content trust for Docker"
},
"help": {
"text": "https://github.com/goodwithtech/dockle/blob/master/CHECKPOINT.md#CIS-DI-0005"
}
},
{
"id": "CIS-DI-0008",
"shortDescription": {
"text": "Confirm safety of setuid/setgid files"
},
"help": {
"text": "https://github.com/goodwithtech/dockle/blob/master/CHECKPOINT.md#CIS-DI-0008"
}
},
{
"id": "CIS-DI-0001",
"shortDescription": {
"text": "Create a user for the container"
},
"help": {
"text": "https://github.com/goodwithtech/dockle/blob/master/CHECKPOINT.md#CIS-DI-0001"
}
},
{
"id": "CIS-DI-0006",
"shortDescription": {
"text": "Add HEALTHCHECK instruction to the container image"
},
"help": {
"text": "https://github.com/goodwithtech/dockle/blob/master/CHECKPOINT.md#CIS-DI-0006"
}
}
]
}
},
"results": [
{
"ruleId": "CIS-DI-0009",
"level": "error",
"message": {
"text": "Use COPY : /bin/sh -c #(nop) ADD file:81c0a803075715d1a6b4f75a29f8a01b21cc170cfc1bff6702317d1be2fe71a3 in /app/credentials.json "
}
},
{
"ruleId": "CIS-DI-0010",
"level": "error",
"message": {
"text": "Suspicious filename found : app/credentials.json , Suspicious ENV key found : MYSQL_PASSWD"
}
},
{
"ruleId": "DKL-DI-0005",
"level": "error",
"message": {
"text": "Use 'rm -rf /var/lib/apt/lists' after 'apt-get install' : /bin/sh -c apt-get update \u0026\u0026 apt-get install -y git"
}
},
{
"ruleId": "DKL-LI-0001",
"level": "error",
"message": {
"text": "No password user found! username : nopasswd"
}
},
{
"ruleId": "CIS-DI-0005",
"level": "note",
"message": {
"text": "export DOCKER_CONTENT_TRUST=1 before docker pull/build"
}
},
{
"ruleId": "CIS-DI-0008",
"level": "note",
"message": {
"text": "setuid file: urwxr-xr-x usr/bin/newgrp, setgid file: grwxr-xr-x usr/bin/ssh-agent, setgid file: grwxr-xr-x usr/bin/expiry, setuid file: urwxr-xr-x usr/lib/openssh/ssh-keysign, setuid file: urwxr-xr-x bin/umount, setgid file: grwxr-xr-x usr/bin/chage, setuid file: urwxr-xr-x usr/bin/passwd, setgid file: grwxr-xr-x sbin/unix_chkpwd, setuid file: urwxr-xr-x usr/bin/chsh, setgid file: grwxr-xr-x usr/bin/wall, setuid file: urwxr-xr-x bin/ping, setuid file: urwxr-xr-x bin/su, setuid file: urwxr-xr-x usr/bin/chfn, setuid file: urwxr-xr-x usr/bin/gpasswd, setuid file: urwxr-xr-x bin/mount"
}
},
{
"ruleId": "CIS-DI-0001",
"level": "none",
"message": {
"text": "Last user should not be root"
}
},
{
"ruleId": "CIS-DI-0006",
"level": "none",
"message": {
"text": "not found HEALTHCHECK statement"
}
}
]
}
]
}
Specify exit code
By default, Dockle
exits with code 0
even if there are some problems.
Use the --exit-code, -c
option to exit with a non-zero exit code if WARN
or FATAL
alert were found.
$ dockle --exit-code 1 [IMAGE_NAME]
Specify exit level
By default, --exit-code
run when there are WARN
or FATAL
level alerts.
Use the --exit-level, -l
option to change alert level. You can set info
, warn
or fatal
.
$ dockle --exit-code 1 --exit-level info [IMAGE_NAME]
$ dockle --exit-code 1 --exit-level fatal [IMAGE_NAME]
Ignore the specified checkpoints
The --ignore, -i
option can ignore specified checkpoints.
$ dockle -i CIS-DI-0001 -i DKL-DI-0006 [IMAGE_NAME]
Or, use DOCKLE_IGNORES
:
export DOCKLE_IGNORES=CIS-DI-0001,DKL-DI-0006
dockle [IMAGE_NAME]
Or, use .dockleignore
file:
$ cat .dockleignore
# set root to default user because we want to run nginx
CIS-DI-0001
# Use latest tag because to check the image inside only
DKL-DI-0006
Accept suspicious environment variables
/ files
/ file extensions
# --accept-key value, --ak value You can add acceptable keywords.
dockle -ak GPG_KEY -ak KEYCLOAK_VERSION [IMAGE_NAME]
or DOCKLE_ACCEPT_KEYS=GPG_KEY,KEYCLOAK_VERSION dockle [IMAGE_NAME]
# --accept-file value, --af value You can add acceptable file names.
dockle -af id_rsa -af id_dsa [IMAGE_NAME]
or DOCKLE_ACCEPT_FILES=id_rsa,id_dsa dockle [IMAGE_NAME]
# --accept-file-extension value, --ae value You can add acceptable file extensions.
dockle -ae pem -ae log [IMAGE_NAME]
or DOCKLE_ACCEPT_FILE_EXTENSIONS=pem,log dockle [IMAGE_NAME]
Continuous Integration (CI)
You can scan your built image with Dockle
in Travis CI/CircleCI.
In these examples, the test will fail with if any warnings were found.
Though, you can ignore the specified target checkpoints by using .dockleignore
file.
Or, if you just want the results to display and not let the test fail for this, specify --exit-code
to 0
in dockle
command.
GitHub Action
We provide goodwithtech/dockle-action.
- uses: goodwithtech/dockle-action@main
with:
image: 'target'
format: 'list'
exit-code: '1'
exit-level: 'warn'
ignore: 'CIS-DI-0001,DKL-DI-0006'
Travis CI
.travis.yml
services:
- docker
env:
global:
- COMMIT=${TRAVIS_COMMIT::8}
before_install:
- docker build -t dockle-ci-test:${COMMIT} .
- export VERSION=$(curl --silent "https://api.github.com/repos/goodwithtech/dockle/releases/latest" | grep '"tag_name":' | sed -E 's/.*"v([^"]+)".*/\1/')
- wget https://github.com/goodwithtech/dockle/releases/download/v${VERSION}/dockle_${VERSION}_Linux-64bit.tar.gz
- tar zxvf dockle_${VERSION}_Linux-64bit.tar.gz
script:
- ./dockle dockle-ci-test:${COMMIT}
- ./dockle --exit-code 1 dockle-ci-test:${COMMIT}
- Example: https://travis-ci.org/goodwithtech/dockle-ci-test
- Repository: https://github.com/goodwithtech/dockle-ci-test
CircleCI
.circleci/config.yml
jobs:
build:
docker:
- image: docker:18.09-git
steps:
- checkout
- setup_remote_docker
- run:
name: Build image
command: docker build -t dockle-ci-test:${CIRCLE_SHA1} .
- run:
name: Install dockle
command: |
apk add --update curl
VERSION=$(
curl --silent "https://api.github.com/repos/goodwithtech/dockle/releases/latest" | \
grep '"tag_name":' | \
sed -E 's/.*"v([^"]+)".*/\1/'
)
wget https://github.com/goodwithtech/dockle/releases/download/v${VERSION}/dockle_${VERSION}_Linux-64bit.tar.gz
tar zxvf dockle_${VERSION}_Linux-64bit.tar.gz
mv dockle /usr/local/bin
- run:
name: Scan the local image with dockle
command: dockle --exit-code 1 dockle-ci-test:${CIRCLE_SHA1}
workflows:
version: 2
release:
jobs:
- build
- Example: https://circleci.com/gh/goodwithtech/dockle-ci-test
- Repository: https://github.com/goodwithtech/dockle-ci-test
GitLab CI
.gitlab-ci.yml
image: docker:stable
stages:
- test
variables:
DOCKER_HOST: tcp://docker:2375/
DOCKER_DRIVER: overlay2
services:
- docker:dind
unit_test:
stage: test
before_script:
- apk -Uuv add bash git curl tar sed grep
script:
- docker build -t dockle-ci-test:${CI_COMMIT_SHORT_SHA} .
- |
VERSION=$(
curl --silent "https://api.github.com/repos/goodwithtech/dockle/releases/latest" | \
grep '"tag_name":' | \
sed -E 's/.*"v([^"]+)".*/\1/' \
) && curl -L -o dockle.tar.gz https://github.com/goodwithtech/dockle/releases/download/v${VERSION}/dockle_${VERSION}_Linux-64bit.tar.gz && \
tar zxvf dockle.tar.gz
- ./dockle --exit-code 1 dockle-ci-test:${CI_COMMIT_SHORT_SHA}
- Example: https://gitlab.com/tomoyamachi/dockle-ci-test/-/jobs/238215077
- Repository: https://github.com/goodwithtech/dockle-ci-test
Authorization for Private Docker Registry
Dockle
can download images from a private registry, without installing Docker
or any other 3rd party tools. It's designed so for ease of use in a CI process.
All you have to do is: install Dockle
and set ENVIRONMENT variables.
- NOTE: I don't recommend using ENV vars in your local machine.
Docker Hub
To download the private repository from Docker Hub, you need to set DOCKLE_AUTH_URL
, DOCKLE_USERNAME
and DOCKLE_PASSWORD
ENV vars.
export DOCKLE_AUTH_URL=https://registry.hub.docker.com
export DOCKLE_USERNAME={DOCKERHUB_USERNAME}
export DOCKLE_PASSWORD={DOCKERHUB_PASSWORD}
- NOTE: You don't need to set ENV vars when downloading from the public repository.
Amazon ECR (Elastic Container Registry)
Dockle
uses the AWS SDK. You don't need to install aws
CLI tool.
Use AWS CLI's ENVIRONMENT variables.
export AWS_ACCESS_KEY_ID={AWS ACCESS KEY}
export AWS_SECRET_ACCESS_KEY={SECRET KEY}
export AWS_DEFAULT_REGION={AWS REGION}
GCR (Google Container Registry)
Dockle
uses the Google Cloud SDK. So, you don't need to install gcloud
command.
If you'd like to use the target project's repository, you can settle via GOOGLE_APPLICATION_CREDENTIAL
.
# must set DOCKLE_USERNAME empty char
export GOOGLE_APPLICATION_CREDENTIALS=/path/to/credential.json
Self Hosted Registry (BasicAuth)
BasicAuth server needs DOCKLE_USERNAME
and DOCKLE_PASSWORD
.
export DOCKLE_USERNAME={USERNAME}
export DOCKLE_PASSWORD={PASSWORD}
# if you'd like to use 80 port, use NonSSL
export DOCKLE_NON_SSL=true
Contributors
Code Contributors
This project exists thanks to all the people who contribute. [Contribute].
Financial Contributors
Become a financial contributor and help us sustain our community. [Contribute]
Individuals
Organizations
Support this project with your organization. Your logo will show up here with a link to your website. [Contribute]
License
- Apache License 2.0
Author
@tomoyamachi (Tomoya Amachi)
Top Related Projects
Find vulnerabilities, misconfigurations, secrets, SBOM in containers, Kubernetes, code repositories, clouds and more
A vulnerability scanner for container images and filesystems
Vulnerability Static Analysis for Containers
The Docker Bench for Security is a script that checks for dozens of common best-practices around deploying Docker containers in production.
A tool for exploring each layer in a docker image
Dockerfile linter, validate inline bash, written in Haskell
Convert designs to code with AI
Introducing Visual Copilot: A new AI model to turn Figma designs to high quality code using your components.
Try Visual Copilot