CDK VS kata-containers

It does show how incompetent the attacker was, I report below what Retr0id wrote in the issue:

"tl;dr: The "leak" seems real, but doesn't prove any of the claims made in the readme.

This statement from Peiyuan Liao, the rabbit CTO, is consistent with what I'm seeing here: https://twitter.com/liaopeiyuan/status/ 1782922595199033662

So the "leak" is a bit of a nothingburger, containing partial code for the relatively boring process of letting users authenticate with online services through a sandboxed browser session, from which auth tokens etc. can be extracted. You can't infer anything about how LAM does or doesn't work from this.

They likely used "kiosk escape" tricks to get code exec within the box that runs the browser. Assuming their sandboxing is all set up correctly, this isn't particularly concerning, but it does expose the code that runs within the sandbox for analysis. That's what we appear to have here.

The attacker left behind a file named cdk.log, which is an artifact of https://github.com/cdk-team/CDK/, a container pentesting tool. They were clearly trying to escape the sandbox and pivot to somewhere more interesting, but I don't think they managed it. I think "part 2" is a bluff, this is all they have (feel free to prove me wrong, lol).

But that doesn't mean there's nothing here. Lets look at what we do have.

The most interesting detail to me is a package name list in repo/ typescript/common/base-tsconfig.json

[...]

The only code actually present is for q-web-minion-

What follows is my speculation based on the names alone:

"q" seems like a codename for the rabbit device (so q-hole rabbit hole). Q might stand for "quantum".

The problem with trying to log into and interface with consumer-facing services from 'the cloud" is that you'll get IP rate limited, blocked as a bot, etc. It would make sense to proxy traffic back out through the user's device, and that's what I'd hope q-proxy is about. The big downside with this is that it ~doubles latency and halves available bandwidth, magnifying any deficiencies of a flaky 4G connection. This is perhaps partly why their doordash demo chugged so hard. (protip to the team; use a caching proxy, with SSL, MitM. Detect CDN URLs and don't proxy those.)

This is a total stab in the dark but my guess is that bunny-host is where the LAM action happens, and bunny-builder is for LAM training.

cm-quantum-peripheral-common might be the wrist-mounted device teased in the launch event.

Addendum:

It's also possible there were some juicy credentials accessible within the container. But if there were, they aren't in this leak. In particular, it looks like they're using GCP "service account keys' (/credentials/ cm-gcp-service-account-quantum-workload/gcp-service-account- quantum-workload.json), which according to google's docs "create a security risk and are not recommended. Unlike the other credential file types, compromised service account keys can be used by a bad actor without any additional information".

There isn't enough information here (and/or my analysis isn't deep enough - "cloud" is not my forte) to determine if that'll cause any issues in practice, but if there really is a "part 2" leak, I'd guess this is how they got it."

I OCR two screenshots that I did so there could be errors.

3. Tools for network actions, probe, tunnel and K8s cluster management (7 tools).

See more in https://github.com/cdk-team/CDK

Seems like Fly.io Machines are trying reimplement Kata Containers with the Firecracker backend [0].

Kata has a guest image and guest agent to run multiple isolated containers [1].

[0] https://katacontainers.io/

[1] https://github.com/kata-containers/kata-containers/blob/main...

> Last time I looked (a few months ago), the documentation was pretty sparse or outdated.

It still is, though it works somewhat seamlessly when installing with https://github.com/kata-containers/kata-containers/blob/main...

Though only one of the hypervisors works well.

Better to use a secure VM, can even get container-like VMs with kata-containers

As I understand,Kata Containers

Kubernetes using Kata containers as a containerd backend

https://github.com/kata-containers/kata-containers/blob/main...

My home k8s cluster is now "locked down" using micro-vms (kata-containers[0]), pod level firewalling (cilium[1]), permission-limited container users, and mostly immutable environments. Given how quickly I rolled this out; the tools to enhance cluster environment security seem more accessible now than my previous research a few years ago.

I know it's not exactly a production setup, but I really do feel that it's the most secure runtime environment I've ever had accessible at home. Probably more so than my desktops, which you could argue undermines most of my effort, but I like to think I'm pretty careful.

In the beginning I was very skeptical, but being able to just build a docker/OCI image and then manage its relationships with other services with "one pane of glass" that I can commit to git is so much simpler to me than my previous workflows. My previous setup involved messing with a bunch of tools like packer, cloud-init, terraform, ansible, libvirt, whatever firewall frontend was on the OS, and occasionally sshing in for anything not covered.

[0] https://github.com/kata-containers/kata-containers

I'm really impressed by fly.io, and the candidness with which they share some of their really awesome technology. Being container-first is the next step for PaaS IMO and they are ahead of the pack.

I aim to build a platform like theirs someday (probably not any time soon) but I don't think I'd do any of what they're doing -- it feels unnecessary. Bear with me as I recently learned that they use nomad[0] and some of these suggestions are kubernetes projects but I'd love to hear why the following technologies were decided against (if they were):

- kata-containers[1] (it does the whole container -> VM flow for you, automatically, nemu, firecracker) with multiple VMM options[2]

- linuxkit[3] (let's say you didn't go with kata-containers, this is another container->VM path)

- firecracker-containerd[4] (very minimal keep-your-container-but-run-it-as-a-VM)

- kubevirt[5] (if you just want to actually run VMs, regardless of how you built them)

- Ceph[6] for storage -- make LVM pools and just give them to Ceph, you'll get blocks, distributed filesystems (CephFS), and object gateways (S3/Swift) out of it (in the k8s space Rook manages this)

As an aside to all this, there's also LXD, which supports running "system" (user namespace isolated) containers, VMs (somewhat recent[7][8]), live migration via criu[9], management/migration of underlying filesystems, runs on LVM or zfs[10], it's basically all-in-one, but does fall behind in terms of ecosystem since everyone else is aboard the "cloud native"/"works-with-kubernetes" train.

I've basically how I plan to run a service like fly.io if I ever did -- so maybe my secret is out, but I sure would like to know just how much of this fly.io got built on (if any of it), and/or what was turned down.

[0]: https://news.ycombinator.com/item?id=26745514

[1]: https://github.com/kata-containers/kata-containers

[2]: https://github.com/kata-containers/kata-containers/blob/2fc7...

[3]: https://github.com/linuxkit/linuxkit

[4]: https://github.com/firecracker-microvm/firecracker-container...

[5]: https://github.com/kubevirt/kubevirt

[6]: https://docs.ceph.com/

[7]: https://discuss.linuxcontainers.org/t/running-virtual-machin...

[8]: https://github.com/lxc/lxd/issues/6205

[9]: https://criu.org/Main_Page

[10]: https://linuxcontainers.org/lxd/docs/master/storage

Kata Containers https://github.com/kata-containers/kata-containers