Skyfall and Solace

Discussion in 'other security issues & news' started by Trooper, Jan 18, 2018.

  1. Trooper

    Trooper Registered Member

    Joined:
    Jan 26, 2005
    Posts:
    5,509
  2. deBoetie

    deBoetie Registered Member

    Joined:
    Aug 7, 2013
    Posts:
    1,832
    Location:
    UK
    Well, we'll see. But, most likely to be Spectre-class vulnerabilities, there will be a stream of those.

    All using side-channels, whose risks have been known forever but conveniently forgotten in the rush for speed and profit. Instead of giving us extra cores (harming their high end/data centre profits), we've had incremental improvements based on slapdash implementations of dodgy optimisations.
     
  3. hawki

    hawki Registered Member

    Joined:
    Dec 17, 2008
    Posts:
    6,080
    Location:
    DC Metro Area
    "...A website that began going viral today, Skyfallattacks, suggests more potential attack vectors are imminent. Currently, the site is basically just some text that alludes to two allegedly potential new attacks dubbed Skyfall and Solace. (Someone out there really likes Daniel Craig’s James Bond—marketing!) Little is known about this new pair of alleged exploits, and it’s entirely possible they’re entirely bull<snip>..."

    https://gizmodo.com/intel-claims-90-percent-of-affected-cpus-have-live-patc-1822192075
     
  4. Fly

    Fly Registered Member

    Joined:
    Nov 1, 2007
    Posts:
    2,201
    As someone who is not a techie: how likely will one be vulnerable in the real world against these attacks, asssuming one uses a browser/ email client (interface with the internet) that is fully patched ?
     
  5. Minimalist

    Minimalist Registered Member

    Joined:
    Jan 6, 2014
    Posts:
    14,885
    Location:
    Slovenia, EU
    IMO, at beginning not very likely if you are not high value target. Most members of this forum will probably be "high hanging fruit" even if vulnerabilities get exploited in wild. If it gets exploited and affects most of regular users it will be a problem of whole tech industry and not just individuals.
     
  6. itman

    itman Registered Member

    Joined:
    Jun 22, 2010
    Posts:
    8,593
    Location:
    U.S.A.
    Meltdown and Spectre have opened up the whole "firmware vulnerability can of worms" that has existed for years for full public disclosure. And frankly, it's high time this was done. Like I keep saying, this isn't one isolated hardware issue. The whole small computer architecture is flawed security-wise and has been since "day one."
     
  7. Peter2150

    Peter2150 Global Moderator

    Joined:
    Sep 20, 2003
    Posts:
    20,590
    Am I correct in what i saw that implies the only way this exploits can run is if they somehow get on your machine.
     
  8. deBoetie

    deBoetie Registered Member

    Joined:
    Aug 7, 2013
    Posts:
    1,832
    Location:
    UK
    Exactly - they have to execute instructions (including high-resolution timing measures). So for clients, that is most likely delivered by browsers as scripts. Conventional sandboxing, process isolation, rings etc. avail you naught in controlling this. But fuzzing the accuracy of the timing measures is pretty helpful, which is what the latest FF patch does, for example.

    As @itman notes of course, these side channel attacks have been around forever, and well-known to competent hardware/chip designers. Perhaps what's different is that Intel in particular has maximised profit and milked the cash cow for many years, based on dodgy and dangerous optimisation rather than letting people have more real cores for reasonable money. So you have the absurd situation where a smartphone has more cores than your desktop and some at least have not gone down the Spectre-vulnerable optimisation route on the Smartphones. As Snowden said: clients are "terrifically weak".

    I'd encourage people to also think about the cloud and their indirect exposure to systems and data running there. While hopefully your main providers will not be running on shared systems, and will control their software extremely tightly, that won't always be the case. For instance, if you're running a VPS or something, your certificates on the machine might possibly be vulnerable - depending on your hoster.
     
  9. Peter2150

    Peter2150 Global Moderator

    Joined:
    Sep 20, 2003
    Posts:
    20,590
    Thanks deBoetie. I've approached this by being strict about email practices and also tight control of what can run on the system via MZwritescanner, Appguard and ExeRadarPro.. Also using Firefox 57 and Noscript
     
  10. deBoetie

    deBoetie Registered Member

    Joined:
    Aug 7, 2013
    Posts:
    1,832
    Location:
    UK
    I think that's pretty much all we can do at this stage. There's also ad-blocking and malicious site blocking (which I do on the router these days), and also some merit in using Firejail or Sandboxie to block internet access for those programs that don't need it, so that they can't phone home (as well as for most userspace programs). I also think that VMs continue to offer good protection, even though they are likely more permeable than desired, the risk is small.
     
  11. Peter2150

    Peter2150 Global Moderator

    Joined:
    Sep 20, 2003
    Posts:
    20,590
    All my high risk stuff is done in a VM machine that is big enough to replicate my desktop.
     
  12. hawki

    hawki Registered Member

    Joined:
    Dec 17, 2008
    Posts:
    6,080
    Location:
    DC Metro Area
  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.