Register a SA Forums Account here!
JOINING THE SA FORUMS WILL REMOVE THIS BIG AD, THE ANNOYING UNDERLINED ADS, AND STUPID INTERSTITIAL ADS!!!

You can: log in, read the tech support FAQ, or request your lost password. This dumb message (and those ads) will appear on every screen until you register! Get rid of this crap by registering your own SA Forums Account and joining roughly 150,000 Goons, for the one-time price of $9.95! We charge money because it costs us money per month for bills, and since we don't believe in showing ads to our users, we try to make the money back through forum registrations.
 
  • Post
  • Reply
Dixie Cretin Seaman
Jan 22, 2008

all hat and one catte
Hot Rope Guy
There's a lot about cellphone OS security that I don't know. Is this a reasonable place to ask some basic questions?

For example, a lot of sites do 2fa through apps like Google Authenticator or Duo Mobile in lieu of hardware token like an RSA hardware token. How secure is 2fa through a phone in comparison, and are there distinct weak points to be aware of? Let's assume an unrooted phone with encryption enabled, using a strong passcode. There's always a chance of sec fuckups in general computing environment compared to a simple security token, but are there known weaknesses, say, if a malicious app is accidentally installed, or your phone connects to a hostile network?

If anyone happens to know about iPhone/TouchID security in particular, I'm curious how that ties in to device encryption. When the phone is shut off completely it requires the passcode for access (and again every few days for good measure). Otherwise it can be unlocked more conveniently with TouchID. Consider the following three iPhone states: 1) iPhone just turned on and is asking for device passcode to continue, 2) iPhone is locked and can be unlocked with TouchID, and 3) iPhone on for >2 days and is now locked and requiring password reentry to unlock. Is there a difference between these 3 states relating to the security of the full device encryption? For example does a phone being unlockable via TouchID (or generally being on with background processes running) indicate that a decryption key could be recovered from RAM more easily? If this is a dumb question, feel free to let me know; I have very little knowledge of this security architecture.

Adbot
ADBOT LOVES YOU

Dixie Cretin Seaman
Jan 22, 2008

all hat and one catte
Hot Rope Guy

apseudonym posted:

Between 2 and 3 as far as I know of iOS's encryption model the keys are still in RAM, so all those hardware attackers are equally applicable. The first boot requirements and passcode logic isn't so much about the keys as it is a bunch of other things.

Thanks for the info. Practically speaking, how hard is that kind of attack for an unrooted phone? Assuming you're not a target of a TLA is it reasonable to ignore?

Dixie Cretin Seaman
Jan 22, 2008

all hat and one catte
Hot Rope Guy

apseudonym posted:

Depends the attacker, the device, and you.

Are you on a newish version of the OS and avoid downloading and installing stuff outside the official stores or going out of your way to disable security features? If so you're in a very good spot, you're more likely to get your 2 factor phished than your phone owned, the person is way easier to exploit than the device.

Use 2FA, it makes it way harder for hackers in unspecifiedistan to get into your poo poo. Your phone is a good device for this since you have it on you and you are almost certainly not interesting enough for a targeted attack.

If your adversary is the government where those services are hosted or where their officers are you're already hosed anyways :)

Sorry, I guess I was unclear. I meant how hard is it to, for example, grab whole device encryption keys from RAM on an unrooted iPhone. My understanding is that this kind of thing is mostly theoretical and it would be beyond the capabilities of non-government adversaries. Is this about right? e.g. if there was some grey-market Chinese kit for owning smartphone encryption then there probably wouldn't be those anti-encryption political rants floating around recently...

Dixie Cretin Seaman
Jan 22, 2008

all hat and one catte
Hot Rope Guy
E: Apparently Lastpass security is already a discussion in the Infosec thread, so I'll avoid cluttering up this thread with the same arguments.

Dixie Cretin Seaman fucked around with this message at 20:03 on Dec 29, 2015

  • 1
  • 2
  • 3
  • 4
  • 5
  • Post
  • Reply