The following page may contain information related to upcoming products, features and functionality. It is important to note that the information presented is for informational purposes only, so please do not rely on the information for purchasing or planning purposes. Just like with all projects, the items mentioned on the page are subject to change or delay, and the development, release, and timing of any products, features or functionality remain at the sole discretion of GitLab Inc.
|Content Last Reviewed||
Thanks for visiting this category direction page on Container Scanning in GitLab. This page belongs to the Container Security group of the Protect stage and is maintained by Sam White (firstname.lastname@example.org).
This direction page is a work in progress, and everyone can contribute. We welcome feedback, bug reports, feature requests, and community contributions.
/label ~"devops::protect" ~"Category:Container Scanning" ~"group::container security".
Our best practices are to package applications into containers, so they can be deployed to Kubernetes.
Container Scanning checks your Docker images against known vulnerabilities that may affect software that is contained in the image. Users often use existing images as the base for their containers. It means that they rely on the security of those images and their preinstalled software. Unfortunately, as this software is subject to vulnerabilities, this may affect the security of the entire project.
Our goal is to provide Container Scanning as part of the standard development process. This means that Container Scanning is executed every time a new commit is pushed to a branch, and only vulnerabilities introduced within the merge request are shown. We also include Container Scanning as part of Auto DevOps.
In the future, another place where Container Scanning results would be useful is in the GitLab Container Registry. Images built during pipelines are stored in the registry, and then used for deployments. Integrating Container Scanning into GitLab Container Registry will help to monitor if it is safe to deploy a specific version of the app.
Primary: Sasha (Software Developer wants to know when adding a container if it has known vulnerabilities so alternate versions or containers can be considered.
Secondary: Sam (Security Analyst) wants to know what containers have known vulnerabilities (to reduce the OWASP A9 risk - Using Components with Known Vulnerabilities), to be alerted if a new vulnerability is published for an existing component, and how behind current version the components are.
Our vision for container security is to provide the ability to scan container images regardless of where they may reside and to shift those results as far left as possible.
To reach the Complete Maturity level, at a minimum we will need to implement the following features. We will likely need to implement additional features as well and this is currently being researched.
Some of the key long-term themes for our pipeline-scanning functionality include the following:
In an upcoming milestone, we plan to allow users to scan container images that are actively running in a Kubernetes instance for vulnerabilities and to report those vulnerabilities back to the Security Center.
Additionally, in the short-term, we are working on the existing container pipeline scanning functionality to accomplish the following:
To accomplish these goals, we plan to replace our current Container Scanning engine Clair with Trivy. Although the work to integrate Trivy was primarily completed in the %13.11 release, we do not plan to change the default scanning engine until the %14.0 release. To minimize our on-going maintenance work, we have formally deprecated our integration with Clair.
For additional details and context behind the change, or to provide feedback, please reference our deprecation issue.
We will be researching current user challenges in this issue. Please feel free to comment!
Currently we notify developers when they add containers with known vulnerabilities in a merge request, if security approvals are configured, we will require an approval for critical, high or unknown findings. A summary of all findings for a project can be found in the Security Dashboard where Security Teams can quickly check the security status of projects. In some cases we are able to offer automatic remediation for the findings.
Our primary success metric is the number of unique users who run a container security scan each month.
In addition to being A9 Using Components with Known Vulnerabilities in the OWASP top 10, keeping components up to date is code quality issue, and finally as the need for software bill of materials (SBoM) grows being able to list your dependencies will become a needed feature for all application developers.
We continue to engage analysts so they remain aware that we offer Container Scanning, which is sometimes considered stand-alone and other times it is considered part of Application Security Testing (AST) or Software Composition Analysis (SCA) bundles as defined in our Solutions, since vulnerabilities for base images can be considered very similar to vulnerabilities for software dependencies.
We can get valuable feedback from analysts, and use it to drive our vision.
Top Epics and Issues can be viewed in this list
If you don't see the
customer success label on an issue yet, and you are a customer success team-member, feel free to add it!
If you don't see the
customer label on an issue yet, feel free to add it if you are the first customer!
If you don't see the
internal customer label on an issue yet, and you are a team-member, feel free to add it!
To be determined.