Hash: SHA512

On 2016-10-15 14:09, QubesOS User wrote:
> Hello everyone,
> I could imagine that this question has been discussed before already, and if 
> this should be the case, then I'm very sorry for posting this (I'd be 
> thankful for an according link if so though).
> I think that I've gained quite much knowledge about possible attack surfaces 
> provided on hardware and software level during the last 15 years, trying to 
> keep up-to-date and often doing research on new approaches in this field. 
> First of all, I'd like to stress that the 'objection' (which I don't mean as 
> such) I may raise by this post does not have any intention of criticizing the 
> great work and effort done by the QubesOS developers and the community (it's 
> not meant as an unhelpful 'critique' at all). Much rather I have a huge 
> respect for the commitment shown by everyone involved in the development of 
> QubesOS.
> Having compared various approaches in this field (e. g. OpenBSD, Linux using 
> a hardened security kernel, GNU Hurd), I'd basically come to the conclusion 
> that QubesOS is the most promising approach, especially if VT-d isolation is 
> available.
> However, the main points I'd like to address are:
> 1) XEN is developed by people working for a company based in the U.S. (I know 
> the difference between open-source and proprietary software, but still they 
> belong to the same team/company). If even developers of TrueCrypt received 
> one of those 'blue letters' - What is the reason to assume that the XEN 
> developers didn't receive one of those as well? Seen from the perspective of 
> the NSA it looks totally odd and irrational to me if they would not to so, 
> since they can do so, and it's their task to thwart any efforts which might 
> hinder them from collecting data. I don't regard those people as being 'evil' 
>  or anything like that (nor do I regard this as being positive, which should 
> go without saying), I just look at things in a rational way: If QubesOS is a 
> great approach to ensure security, then one must be naive to assume that this 
> won't automatically lead to classifiying this as a 'high priority target' - 
> With all the consequences.
> 1.2) Since this looks so obvious to me: Why isn't it a top priority for 
> QubesOS developers to make use of a supervisor (or develop an independent 
> one, which would surely need endless efforts, but wouldn't it be worth it?), 
> which is not subjected to the objections I tried to express?
> 2) QubesOS totally relies on 2.1) trusting XEN developers to completely 
> understand the more than just complex x64 architecture being used today and 
> 2.2) on trusting Intel's VT technology.
> Regarding 2.2): Just assuming Intel would have received some kind of 'advice' 
> (they may even find motivation without getting such - I certainly don't think 
> that Intel is an 'NSA subcontractor', but they are simply a big and 
> profit-orientated company, not an idealistic open-source community like the 
> QubesOS developers etc.) - Then how realistic is it that an absolutely 
> professionally designed and implemented backdoor etc. as the result of sheer 
> endless human, technological and financial ressources gets discovered by 
> people like the QubesOS community, no matter how enthusiastic, intelligent, 
> cautious and sceptical those are?
> Referring once again to 2.1) I'd like to point to and quote from a highly 
> interesting Qubes Security Bulletin 
> (https://github.com/QubesOS/qubes-secpack/blob/master/QSBs/qsb-018-2015.txt):
> "2) We are not entirely convinced if the way Xen Security Team decided to 
> address this vulnerability is really optimal, security wise. It seems like a 
> more defensive approach would be to get rid of this
> dangerous construct of reusing the same memory for both an internal pointer 
> and VM-provided data. Apparently Xen developers believe that they can fully 
> understand the code, with all its execution paths, for decoding x86 operands. 
> This optimistic attitude seems surprising, given the very bug we're 
> discussing today."
> [One should read the whole bulletin to know the context, but I didn't want 
> this to become too long.]
> One might also like to take a look at this bulletin, which gives me, among 
> other XEN-related informations and facts, the strong impression that seeking 
> an alternative hyperadvisor should have higest priority for the QubesOS 
> development (believe me, I'd more than like to contribute to doing so by 
> myself, too, and if I shold be able to aquire the necessary skills, I'll 
> definitely try to do so):
> https://github.com/QubesOS/qubes-secpack/blob/master/QSBs/qsb-024-2016.txt
> "A more radical reader might be of the opinion that we should completely 
> replace Xen with some other hypervisor. Such an opinion is surely not 
> unfounded, as we have previously expressed our disappointment in the Xen 
> security process [5]. Sadly, not much has improved over the past several 
> months. Moreover, even though Qubes is now based on a hypervisor-abstracting 
> architecture ("Odyssey"), which should make switching to a different VMM a 
> relatively easy task, the primary problem that remains is the lack of a good 
> alternative hypervisor to which we could move [6]."
> Hopefully my post won't get misunderstood or even discourage people (if so, 
> I'd really regret having written this) - I'm just worried regarding those 
> points, and I believe that nothing is more dangerous than thinking one would 
> have ensured security and therefore feeling even more motivated to rely on 
> this (imaginary) security with all the consequences (to express this using 
> quite harsh words: In the worst case users of QubesOS might be enticed to 
> step into a huge honeypot, and I'd be pretty sure that - if this should be 
> true - won't become known for many years, allowing the NSA etc. to collect 
> petabytes of data in the meantime).
> Finally, I don't think (this might sound even more provocative to many 
> people) that organisations like the NSA etc. are 'useless' or 'to be 
> abolished by all means' - This is similar to a simple but fundamental problem 
> like "Well, complete disarment would surely be a great thing, but we'll run 
> into big trouble if other nations won't think so", and of course I don't feel 
> sorry at all for any terrorist etc. who gets prevented from killing other 
> people because of getting caught before he can do that (or for criminals 
> being caught). But nonetheless everyone knows where mass surveilance can lead 
> to, and I think governments should generally respect people's privacy unless 
> they have good reasons to watch them (and this should not be determined by 
> computer algorithms) - But that's not what I wanted to address here, and I 
> think it also doesn't belong here at all.
> I'd be very curious to hear other people's opinions regarding the issues I 
> addressed above.
> Kind regards and all the best to everyone

If I've understood you correctly, your main thesis is that the security of 
Qubes depends directly on several factors that are outside of our control, in 
particular on Xen and Intel not having been strong-armed by the US government. 
I think it's safe to say that the Qubes team fundamentally agrees with you. 
Joanna has written about this topic in several of her blog posts and papers, so 
please take a look if you haven't already:


On Intel x86 specifically, make sure to read this:


Finally, note that the future isn't necessarily hopeless:


- -- 
Andrew David Wong (Axon)
Community Manager, Qubes OS


You received this message because you are subscribed to the Google Groups 
"qubes-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to qubes-users+unsubscr...@googlegroups.com.
To post to this group, send email to qubes-users@googlegroups.com.
To view this discussion on the web visit 
For more options, visit https://groups.google.com/d/optout.

Reply via email to