pages/txt/simplicity-security.txt (view raw)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 |
--- template: url: simplicity-security title: Simplicity (mostly) guarantees security subtitle: This is why I meme mnmlsm so much date: 2020-05-07 --- Although it is a very comfy one, it's not just an aesthetic. Simplicity and minimalism, in technology, is great for security too. I say "mostly" in the title because human error cannot be discounted, and nothing is perfect. However, the simpler your tech stack is, it is inherentely more secure than complex monstrosities. Let's look at systemd, for example. It's got over 1.2 million lines of code. "Hurr durr but LoC doesn't mean anything!" Sure ok, but can you _imagine_ auditing this? How many times has it even been audited? I couldn't find any audit reports. No, the developers are not security engineers and a trustworthy audit must be done by a third-party. What's scarier, is this thing runs on a huge percentage of the world's critical infrastructure and contains privileged core subsystems. "B-but Linux is much bigger!" Indeed, it is, but it has a thousand times (if not more) the number of eyes looking at the code, and there have been multiple third-party audits. There are hundreds of independent orgs and multiple security teams looking at it. That's not the case with systemd -- it's probably just RedHat. Compare this to a bunch of shell scripts. Agreed, writing safe shell can be hard and there are a ton of weird edge-cases depending on your shell implementation, but the distinction here is _you_ wrote it. Which means, you can identify what went wrong -- things are predictable. systemd, however, is a large blackbox, and its state at runtime is largely unprovable and unpredictable. I am certain even the developers don't know. And this is why I whine about complexity so much. A complex, unpredictable system is nothing more than a large attack surface. Drew DeVault, head of [sourcehut](https://sourcehut.org) wrote something similar (yes that's the link, yes it has a typo).: https://sourcehut.org/blog/2020-04-20-prioritizing-simplitity/ He manually provisions all sourcehut infrastructure, because tools like Salt, Kubernetes etc. are just like systemd in our example -- large monstrosities which can get you RCE'd. Don't believe me? See [this](https://threatpost.com/salt-bugs-full-rce-root-cloud-servers/155383/). *This was day 3 of the #100DaysToOffload challenge. It came out like a systemd-hate post, but really, I couldn't think of a better example.* |