sysbox VS refpolicy

You are probably referring to Sysbox (https://github.com/nestybox/sysbox), which I believe will meet your requirements (systemd, inner containers, security, etc).

Btw, Sysbox is already supported in Docker-Desktop (business tier only), so you can easily do what you want with this instruction:

$ docker run -it --rm -e SYSBOX_SYSCONT_MODE=TRUE nestybox/ubuntu-focal-systemd-docker:latest bash

Disclaimer: I'm Sysbox's co-creator and currently working for Docker.

One project in this space that looked quite promising to me is sysbox[0]. I've used them once for a gitlab runner set-up similar to what is described in their blog[1].

It's currently working great and I have not had any major crashes/incidents for at least the past 8 months.

[0]: https://github.com/nestybox/sysbox

[1]: https://blog.nestybox.com/2020/10/21/gitlab-dind.html

Today, things are very different. Docker-in-Docker has a more secure and safe approach with rootless containers and freemium tools like sysbox. Tools like sysbox let you run Docker-in-Docker without the -privileged flag and optimizes specific scenarios, like running multiple nodes of a Kubernetes cluster as ordinary containers.

Right now I am going with sysbox rootless containers. https://github.com/nestybox/sysbox

We’ve been using Sysbox (https://github.com/nestybox/sysbox) for our Buildkite based CI/CD setup, allows docker-in-docker without privileged containers. Paired with careful IAM/STS design we’ve ended up with isolated job containers with their own IAM roles limited to least-privilege.

A good alternative to the VM approach is to use Kubernetes + Sysbox (a next-gen "runc", free, open-source).

But nowadays there is an option to run such software in containers securely. It's called Sysbox, and it's a new "runc" (the piece of software that creates the containers). I am one of the developers, so I am biased, but I think you'll find it helpful.

Distributions often start with the SELinux reference policy rather than starting from scratch.

really? I don't mean understand how to apply a new label. I mean understand what the policies are and how they work, be able to create new ones that apply to you, and verify that the ones given to you by the distro are correct for your use. You're saying this is not hard to understand: https://github.com/SELinuxProject/refpolicy/blob/master/poli... ?

Otherwise you are blindly applying some black box.

selinux-refpolicy-src pulls from the original SELinux Project refpolicy repo and just install-src these, which just places all policies into appropriate directory and does nothing futher, i. e. doesn't compile them to make usable

https://github.com/SELinuxProject/refpolicy/blob/4412ad507c5880d9ff52fd376c23183cc9ae10b7/policy/support/misc_patterns.spt#L55

Of those distributions, only Fedora sets SELinux to enforcing by default. Moreover, AFAIK Fedora (+ RHEL and Android) are the only distributions that had wide testing of the reference policy [1] [2]. So, if you enable SELinux with the reference policy on the other distributions that you mention, it is likely that you will run into all kinds of issues.

[1] https://github.com/SELinuxProject/refpolicy

[2] https://github.com/fedora-selinux/selinux-policy

No default or reference policy is provided in openSUSE Leap. SELinux will not operate without a policy, so you must build and install one. The SELinux Reference Policy Project (https://github.com/SELinuxProject/refpolicy/wiki) should be helpful in providing examples and detailed information on creating your own policies, and this chapter also provides guidance on managing your SELinux policy.

Interesting. I don't have a red hat based system handy at the moment to compare since I'm on holiday. Maybe that's just how they set up shm? If you compare to Reference Policy, it appears that the getattr permission you asked about is granted there: https://github.com/SELinuxProject/refpolicy/blob/8a1bc98a31a9f2396f3c1389f43ffb10df31157f/policy/modules/system/systemd.te#L600