bubblewrap VS multipass

Nothing nix specific but you may be interested in https://github.com/containers/bubblewrap

Recently, I came across Chainguard and wrote the article How to build Docker Images with Melange and Apko. As a fervent supporter of Kubernetes and GitLab CI, I was eager to experiment with building images using Melange in this particular setup. GitLab's shared Runners work seamlessly with Bubblewrap, eliminating the need for additional configurations. This post is intended for enthusiasts like myself, interested in hosting their own Kubernetes Runners and leveraging the Kubernetes Runner Type of Melange.

```

This is basically manually invoking what Flatpak does:

https://github.com/containers/bubblewrap

This is also useful for more than just security. E.G., you can test how your app would behave on a fresh install by masking your user configuration files. I personally also have a tool that uses it to basically bundle all dependencies from an entire Linux distribution in order to make highly portable AppImages— Been meaning to post that, will get around to it eventually maybe.

The flags above should hide your user data (`--tmpfs`), disable network access (`--unshare-all`), hide/virtualize devices and OS state (`--dev` and `--proc`), and make the rest of the root filesystem read-only (`--ro-bind`­— Including the insecure X11 socket in `/tmp`, which you might want to expose for GUI apps).

Check them against `bwrap --help`; I might have omitted one or two more things you'd need.

While trying to find out more comparison information, found this light on details issue:

https://github.com/containers/bubblewrap/issues/81

It mentions nsjail and minijail.

Install Multipass from https://multipass.run

Multipass orchestrates virtual Ubuntu instances

Multipass I love Multipass for quick Ubuntu instances spun up for testing or as a playground. Wish I would have known and used of it sooner.

If you just need Ubuntu then you can try "Multipass" from Canonical (https://multipass.run/). Works quite well on my M2 Air. I haven't tried using Linux GUI with it though as I need only terminal based VMs.

Multipass is pretty clutch for trivial VMs on MacOs for sure. I use it for a bunch of ssh jump boxes running vpns to different sites. The macOS build does not support custom images (lest not without [some truly insane hacks](https://github.com/canonical/multipass/issues/1260#issuecomm...) , which doesn’t really matter for what I use it for but it is kind of a bummer. If you need something with a little more grunt but don’t want to go full blown with writing your own QEMU tooling or fussing with something like UTM or Parallels, [quickemu](https://github.com/quickemu-project/quickemu) is a really nice qemu wrapper with sane defaults that can expose a whole lot of power if you need it.

I would be cautious or even distrustful of using anything from Oracle. VirtualBox components come under three different licenses - GPLv2, personal use & evaluation license, and an enterprise license. Their VirtualBox license FAQ [1] gives them enough leeway to change future licenses at will. If an exploit is discovered in your old VirtualBox and they've changed the license, you're out of luck.

We've moved our development to KVM and Virtual Machine Manager on Linux [3] and UTM on Mac [4]. There are other options to run your VM, such as Multipass [5] or VirtualBuddy [6].

On a digressive topic - it was fun migrating our legacy application server stack from Oracle Java (old & poorly considered decision) to OpenJDK, thanks to their license [2].

[1] https://www.virtualbox.org/wiki/Licensing_FAQ

[2] https://www.oracle.com/java/technologies/javase/jdk-faqs.htm...

[3] https://ubuntu.com/blog/kvm-hyphervisor

[4] https://mac.getutm.app/

[5] https://multipass.run/

[6] https://github.com/insidegui/VirtualBuddy

How does it compare to https://multipass.run/?

If you have a reasonably beefy computer, you can always try setting up Multipass and set up 2-3 nodes for a k8s cluster, it's how I'm doing my own certification training. I do have a k3s Raspberry Pi cluster, but with Pi prices being what they are still it'd almost be cheaper to do a cloud setup. ☹️