Prepare for v0.37.0 #3507
knqyf263
started this conversation in
Development
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Draft to collaborate on v0.37.0 release announcement
🛫 Deprecation 🌆
--security-checks
flag--security-checks
was renamed to--scanners
.Before
After
🚀 What's new? 🚀
🧾 Complete CIS Kubernetes Benchmark (include host-level checks ) ⎈
This release provides the complete CIS Kubernetes Benchmark report including "host-level checks" which bring Trivy compatible with kube-bench.
🐳 Docker CIS Benchmark 📝
Trivy support Docker CIS Benchmark. You can enable it with
--compliance docker-cis
. It detects issues on container images.👮♂️ Misconfiguration scanning on container image config 🤘
Trivy now detects misconfigurations on image configuration.
--image-config-scanners config
enables this feature. Trivy tries to restore the original Dockerfile from the configuration and scan against it. You can see the history withdocker history [YOUR_IMAGE_NAME]
.The above example passes
--scanners none
so it can disable scanners on files inside the container image, but you can also enable it liketrivy --scanners vuln --image-config-scanners config [YOUR_IMAGE_NAME]
.🔒 Secret scanning on container image config ㊙️
Trivy now detects secrets on image configuration.
--image-config-scanners secret
enables this feature. It is especially useful to detect credentials in environmental variables. You can see the environmental variables of your image withdocker inspect [YOUR_IMAGE_NAME]
.The above example passes
--scanners none
so it can disable scanners on files inside the container image, but you can also enable it liketrivy --scanners vuln --image-config-scanners secret [YOUR_IMAGE_NAME]
.🐍 Conda support for SBOM
Trivy looks for conda packages in container images (and root fileysystem), extract the information and put them into SBOM.
🎯 Dart (pubspec.lock) support for vulnerability detection
Trivy now scans pubspec.lock for vulnerabilities.
☕ Java Index Database 🍡
Trivy used to rely on the third-party API for JAR (Java) scanning. It was leading to slow and unstable scanning. See here for the detail. The
--offline-scan
flag mitigates the issues, but the result accuracy gets worse.After all, we decided to build the Java index database and distribute it [on GitHub(https://github.com/aquasecurity/trivy-java-db). When Trivy finds JAR files while scanning, it downloads the Java database and use it for JAR scanning. It means you can get accurate results even under air-gapped environment if you download the Java database and put it in advance.
--download-java-db-only
and--skip-java-db-update
flags like the vulnerability database were also added.See here for more details.
📈 Automatic policy bundle fetching
This release adds the ability for Trivy to fetch newly published polices, when needed, automatically as a bundle.
Yo can disable this behaviour with the
--skip-policy-update
flag. In this case Trivy will use embedded polices as it does today.🧩 Improved schema support for policies
This release enables support for Rego schema input via Rego Metadata as per the OPA convention
Where the
foo
directory holds a custom user defined schemaBeta Was this translation helpful? Give feedback.
All reactions