Time to talk a bit about Spectre and Meltdown. I know, I touched on these two last week, but there is more to discuss. There are things afoot with these two that have given me some thoughts. No, I do not think the sky is falling.
I am going to start with a little tweet that I saw:
I can finally efficiently (fast) and reliably (no errors) read paged pool/non-L1 data. Time for MeltiKatz/MimiDown. I’ll sit on this a few weeks before setting the world on fire and watching it burn. Or probably someone will do it first ? pic.twitter.com/iLQOezrPV7
— Alex Ionescu (@aionescu) January 11, 2018
Worrisome? Yes. Sky is falling? No. Outside of a POC with JavaScript, I have heard nothing that does not show these bugs are LOCAL which is even mentioned in the CVEs. Add on they are Data Leak and not RCE (Remote Code Execution for those unfamiliar with the term). This demo shows though that there is code to take advantage of Meltdown/Spectre. Seeing something like this makes me believe that there is code used in the wild that we do not know about yet. So, what we need to do is update ourselves. Keeping an eye on processes through things like CarbonBlack Response or similar types of tools might be able to give us some insight into this sort of exploit happening. Once Alex releases his code, it will be easier to create alerts and watchlists for such activity.
Next up on my parade with Spectre/Meltdown is IoT. We all know that IoT can be difficult at best to update. So much hard coded passwords, or no security really at all in the devices. You might think, so what if my fridge is leaking data? OSINT, passwords for Google or Amazon, what apps do the devices use? There are so man possibilities. Smart TVs, think about that. There people have passwords for Hulu, Amazon, Netflix, etc… let alone viewing history and other data. How fast are patches going to be put out for those items, and will those patches be worse than the potential exploit? Which brings me to the final thought for this post…
Ever heard of the cure being worse than the disease? This statement was a fact with the Microsoft patches. AV could cause blue screening and bricking of systems as well as just having an AMD chip. It has been said that companies like Microsoft had known about Spectre/Meltdown for a couple of months prior to the disclosure. You would think they would have been building and testing patches for it for a while if they did. Instead, it looks like the patches were rushed out. So Microsoft has stopped sending out patches in certain instances. I keep hearing conflicting reports that the key that AV vendors are supposed to put in is required not just for the Spectre/Meltdown patch but without it, all patches will stop (if you have automatic patching set up). That could affect home users big time. Mind you I heard about that from Smashing Security’s podcast on 1/11/18.
Still I maintain that more is being made out of this in the mainstream media in the wrong way. Especially as far as IoT goes, this could be a great tool to start forcing those device makers to do a better job with security overall. Once again though, I think being vigilant is the best solution at this time. Keep our heads up, and watch for the signs, test the fixes, and go about our daily business. Interestingly enough a major security issue with Dell EMC happened and was not mentioned while we have been freaking out about Spectre/Meltdown. Time for us to stop melting down about this one I think.