Security Basics mailing list archives
[Fwd: Re: Security procedure question]
From: "Mario A. Spinthiras" <mario () netway com cy>
Date: Mon, 25 Sep 2006 17:10:00 +0300
--- Begin Message --- From: "Henry Troup" <HenryT () watchfire com>
Date: Mon, 25 Sep 2006 09:52:28 -0400
Note that I did not suggest that the pre-boot was weak; I suggest that the human factors side of it needed to be carefully considered. Pre-boot with full-disk encryption is potentially very strong indeed. As Saqib Ali says in his later email, it can be "too strong" or at least pose significant difficulties in recovering legitimate access. Corporate key escrow is a necessary requirement, and can be hard to fulfill without the potential for compromise. Probably the major need is to make sure that corporate information always has a primary source that's not on a laptop, so that the worst case is the need to re-clone the laptop and lose some number of unmerged updates. Henry Troup Watchfire Corporation Suite 300, 1 Hines Rd. Kanata, ON K2K 3C7 Canada 613-599-3888 x4048 -----Original Message----- From: Mario A. Spinthiras [mailto:mario () netway com cy] Sent: Saturday, September 23, 2006 04:25 To: Saqib Ali Cc: Henry Troup; Brown, Sam; security-basics () securityfocus com; lists () hwf cc Subject: Re: Security procedure question Saqib Ali wrote:There is a misconception that bio-metric somehow increases the security of the mobile device. IT DOES NOT! All it is does is make the logon process easier and simpler. The bio-metric logon process can, in most cases, be circumvented by escaping out of the logon sequence, and logging in using the regular username and password. However if the user set a really complex password, a dictionary attack would be impossible, while a brute-forceattack would take a very very long time. The user can use a really complex password like '3mb55y53curity', without actually having to type in this password upon each logon. So indirectly biometrics improves the security of the mobile devices when used in conjunction with a complex strong password. See: http://www.full-disc-encryption.com/biometrics_and_encryption.htm#mozT ocId415582 for more info on biometric readers that come with Dell and HP laptops. As for the issue with the "residual skin oils left on the device surface", swipe-through biometric scanners are designed to address those issues. With swipe through scanner the attacks with shining light and breathing on the scanner are not possible. On 9/21/06, Henry Troup <HenryT () watchfire com> wrote:Mario A. Spinthiras describes a three-factor authentication system:- What you know - What you have - Who you arewhich is excellent, but there are a couple of caveats. To maintain the independence of the factors requires end-user best practices, specifically not keeping the USB device conveniently at hand in the laptop bag. This requires training and a continual awareness campaign. In the case where the USB fingerprint reader is stolen with the laptop, there is some degradation of security, possibly a lot: I haven't found an authoritative update to show that today's fingerprint readers are any more secure than the ones that Tsutomu Matsumoto spoofed in 2002 - details at http://cryptome.org/gummy.htm and http://cryptome.org/fake-prints.htm At that time, some fingerprint readers could be spoofed as easily as breathing on them, or with a flashlight at just the correct angle. Both of these techniques leverage the residual skin oils left on the device surface. So, a careless user could take it down to single-factorauthentication.To manage this, you need to use the principle of "make the right thing an easy thing"; somehow make it in the user's interest to keep the parts separated. (As an aside, remember that male and female users may have significantly different preferred styles of device; ingeneral males have pockets, females may have no pockets and prefer a purse.) Strangely enough, that would push in the direction of Bluetooth over USB; even though normally we feel that wireless devices don't add security. BMW has gone this route with some recentautomobiles, preferring a proximity card over a physical key. Also, you need to ensure that authorized service people can work on the laptop without compromise of the confidential information; that is, you still need user-level encryption inside the boot-levelprotection.Henry Troup Watchfire Corporation Suite 300, 1 Hines Rd. Kanata, ON K2K 3C7 Canada 613-599-3888 x4048 --------------------------------------------------------------------- ------ This list is sponsored by: Norwich University EARN A MASTER OF SCIENCE IN INFORMATION ASSURANCE - ONLINE The NSA has designated Norwich University a center of Academic Excellence in Information Security. Our program offers unparalleled Infosec management education and the case study affords you unmatched consulting experience. Using interactive e-Learning technology, you can earn this esteemed degree, without disrupting your career or home life. http://www.msia.norwich.edu/secfocus --------------------------------------------------------------------- ------I would like to begin with stating that an opinion and a misconception have two different meanings. Even if a group of people believe in a certain opinion , the majority does not make it A FACT or even a MISCONCEPTION! Biometric authentication is can be fooled yes. but it is all based on HOW you set it up. This is why i said to boot from the USB STICK and not the actual computer. This eliminates a method of brute forcing the kernel. If you were to boot from your computer , and enter in a usb stick crafter for bruteforcing , then eventually it would give you "something" to go step-by-step to proceed with authenticating. But in this case you should have looked at the facts before tagging my contribution as "misconception". Facts: 1. You dont boot from the computer originally for authentication. You boot from the usb stick. 2. The key can be 256bit if you like , it can be stored on the usb stick , not the computer. It simply has to match the required key by the computer. 3. I didnt mention biometric authentication for making things "easier" or "simpler" - rather than an addition to security. 4. A password is required anyway. In conclusion I would like people to research more on technologies before posting since there are methods that have been in implementation for quite a while now. Even the computer Im talking to has the above implemented and has had quite a few try to break it. that gives birth to another so called "majority" that definately believes the opposing opinion to the misconcepted reply earlier in this post. Good luck on this, Mario A. Spinthiras Netway Ltd Nicosia , Cyprus
--- End Message ---
--------------------------------------------------------------------------- This list is sponsored by: Norwich University EARN A MASTER OF SCIENCE IN INFORMATION ASSURANCE - ONLINE The NSA has designated Norwich University a center of Academic Excellence in Information Security. Our program offers unparalleled Infosec management education and the case study affords you unmatched consulting experience. Using interactive e-Learning technology, you can earn this esteemed degree, without disrupting your career or home life. http://www.msia.norwich.edu/secfocus ---------------------------------------------------------------------------
Current thread:
- [Fwd: Re: Security procedure question] Mario A. Spinthiras (Sep 25)