Docker lets security bug sniffer dogs off the leash at container images

Libraries and other code scanned for known programming blunders

Docker has hit upon an idea that perhaps other platforms could potentially incorporate: scanning software components for publicly known vulnerabilities prior to deployment.

Today, the software container biz will announce Docker Security Scanning, which scours private repositories in the Docker Cloud for recognized security bugs and alerts developers if any are found. The feature will be expanded to Docker Datacenter customers later this year.

Specifically, this opt-in service scans a Docker image when it is pushed to the Docker cloud, and programmatically builds a bill-of-materials (BOM) of the image's software components.

It then runs the BOM against a set of security vulnerability databases, including the US government-backed National Vulnerability Database (NVD). If a library or some other dependency in the image is known to harbor exploitable programming blunders, an alarm is raised, and its developers can step in to fix the problem – hopefully by selecting a corrected version to include in the image.

Right now, Docker Security Scanning studies operating system components, application-level libraries, programming-language modules and the like. It can't find bugs in developers' own code, but it can stop them from pulling in insecure dependencies when patched versions exist.

Nathan McCauley, director of security at Docker, told The Register he wanted to make patching as easy as possible for all Docker users, even in large organizations that usually have to leap through hoops and hoops of change processes to update a dependency in an application even if it's an important or critical security patch.

With the vulnerability scanner enabled, it's hoped known vulnerable code can be blocked before it's deployed, thus minimizing headaches down the line. It's actually not a bad idea that could be extended to other platforms: your humble hack can imagine, say, running apt-get install on a GNU/Linux Debian box to bring in a package, only to get a warning or an alert that the package, or one of its dependencies, has a known bug in public vulnerability databases from a real-time on-demand lookup.

"I want the entire Docker ecosystem to have the tooling to ensure its users are secure," said McCauley.

"One of the most important things in a secure software supply chain is making sure you are on top of patching. In some organizations, updating is a hard process and really difficult. I saw an opportunity with Docker to improve this process; Docker presents a window of opportunity in which we can make things dramatically better, and I felt it was worth doing."

The idea is to scan signed images as they are pushed by developers to a registry. If they pass the bug scans, IT ops can deploy the images to production systems as containers.

This tech was previously known as Project Nautilus, which showed its head at the DockerCon Europe event in Spain in November. In a statement on Tuesday, Docker explained:

Docker Security Scanning is available today to Docker Cloud users with a private repo plan, expanding to include all Docker Cloud repo users by the end of Q3 [2016]. Pricing begins at $2 per repo as an add-on service for private repo plans. Docker Security Scanning will also be available as an integrated feature in Docker Datacenter during the second half of 2016.

You can find more about the service here in an official blog post. ®

Similar topics

Other stories you might like

Biting the hand that feeds IT © 1998–2022