How Do I Handle Growing CVE Backlogs in My Container Images?

0
7
Asked By TechWhiz83 On

I've been using Prisma Cloud for container security for about 8 months now, and while it does find vulnerabilities, our backlog of CVEs in Jira keeps growing. Just last week, I tested a fresh node:18 image from Docker Hub with Trivy and it flagged 340 CVEs before I've even added any of our app's code. The base image seems to be the main culprit; it includes components we don't even call, like Curl and half of libc. With our engineering team of 60 and just two dedicated security folks, we can patch maybe 30-40 CVEs per sprint, but then Docker Hub releases a new image and we're back to over 300. I'm wondering if switching to distroless or scratch images would help, and what the best practices are for managing this issue.

4 Answers

Answered By ImageMaster123 On

The number of vulnerabilities increases with the contents in your image. Streamlining your image is a good idea. Try using wolfi images; at my company, they helped cut the CVEs in half. Definitely look into that as part of your strategy!

Answered By SecuRanger99 On

It's crucial to evaluate the actual CVEs you're facing. Many could be irrelevant to your application's security. A straightforward approach is to document those you deem non-threatening and ignore them, especially if you have mitigation strategies in place. Also, consider using tools like Chainguard to minimize your images for better security. Remember, being on Node 18 is pretty out of date; you might want to move to the latest LTS version for better support.

Answered By DevGuru22 On

Node 18 is actually EOL, so it’s not surprising you're facing these issues. An outdated base image definitely increases your vulnerability risk. Have you considered upgrading to a more recent version? It might alleviate some of the backlog.

Answered By CodeNinja77 On

Sure, switching to distroless or minimal images is a common suggestion, but it won't completely eliminate your backlog. The real challenge is how vulnerability scanners like Prisma treat all CVEs equally, regardless of whether they're actually relevant. You should determine which vulnerabilities are truly exploitable in your context rather than simply triaging by severity. Focusing on reachable vulnerabilities can significantly reduce the workload. Upgrading from Node 18 is essential, but it's not the only solution you need to consider.

Related Questions

LEAVE A REPLY

Please enter your comment!
Please enter your name here

This site uses Akismet to reduce spam. Learn how your comment data is processed.