netshoot VS docker-bench-security

Now, let's use netshoot, a Docker image equipped with a set of networking troubleshooting tools that can be used for Docker networking.

nicolaka/netshoot for sure. having a preloaded array of networking troubleshooting tools in an image for kubernetes debugging is a godsend.

First off, it seems like getting tcpdump or Wireshark to capture from the container network and dumping to a file will take some extra steps. (A lot of blog posts seem to recommend this Github project - https://github.com/nicolaka/netshoot)

Some hints how to debug such problems in the future: 1. ufw drops/rejections are logged (use dmesg to view them) 2. wireshark shows all connection attempts, even those that are blocked by firewall 3. ss or netstat shows if the server is actually listening for connections 4. netshoot allows to enter docker container network namespace and do pings, curl or telnet from there

Scanning your container images for vulnerabilities is a good approach. But this scanning is not one time job, it should be done regularly (weekly, monthly, etc.) You need to follow vulnerability reports and fix all of the vulnerabilities as soon as possible. I recommend some open-source tools that could be useful: Trivy, Docker-Bench, Grype.

For Docker configuration I have used this in the past (it utilizes the CIS Docker Benchmark): https://github.com/docker/docker-bench-security

So the main tool to scan against the CIS Docker benchmark (I'm presuming that's the one you're interested in) is https://github.com/docker/docker-bench-security .

git clone https://github.com/docker/docker-bench-security.git cd docker-bench-security sudo sh docker-bench-security.sh

when deploying images on cloud, I always run it thru "docker bench security" It helps finding potential security holes in my images.

Use Docker Bench for Security to audit your container images

Other tools you can use are linux-bench, docker-bench, kube-bench, kube-hunter, kube-striker, Cloud Custodian, OVAL, and OS Query.

I run https://github.com/docker/docker-bench-security against my environment. I would determine what was non-applicable/not scored and then start with scored. Then I would do not scored. My team had made their own Dockerfiles when I started and just grabbed whatever image/version and getting things baselined was not fun. I had to do this for docker-compose and stay on version 2 yml as otherwise I had to go to swarm.