On Aug 15, 2008, at 11:11 , Jeff Johnson wrote:
Jason, See the following threads for some discussion of these issues: http://lists.apple.com/archives/Macnetworkprog/2008/Apr/msg00033.html http://lists.apple.com/archives/Xcode-users/2008/Jul/msg00283.html
Interesting... thanks, Jeff. So I guess the answer is for speed/non- sensitive cache data, maybe confstr(_CS_DARWIN_USER_CACHE_DIR, path, length) is the appropriate call... and maybe for data that may need to actually reside in the filevault, regardless of speed, the return value from the Cocoa call is more appropriate (~/ Library/Caches)?
I would like to point out a couple of interesting things, though...1) ~/Library/Caches is world writable too... so as long as you're logged in, even if you have your filevault armed, you're still gonna be somewhat vulnerable to cache attacks. 2) The new temporary directory (returned the same by confstr(_CS_DARWIN_USER_TEMP_DIR,...) and NSTemporaryDirectory(...) is also outside the sphere of filevault /and/ your files there are not necessarily erased on log-out. I think it's cleaned up with the computer boots (although it may be deleted on shutdown, but I don't think so)... so if any sensitive information were written to the temp dir and the application relied on it being cleaned by the OS, that could be an issue too if
your physical drive were compromised...
Too bad these aren't sysctl variables that could be set if security
were more important to the user
than performance... I checked the darwin source and the directories
returned by confstr(...) are
hard-coded into libc... /Jason
smime.p7s
Description: S/MIME cryptographic signature
_______________________________________________ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
