The WiKID Blog

The WiKID Blog, musings on two-factor authentication, information security and some other stuff.

It always comes to this: why making the right security designs up front matters.

Posted by: root in Two-factor authentication 6 years, 1 month ago

When we started WiKID, we knew we had to be as secure as or more secure than the leading players at the time (RSA, Vasco, mostly, way back then). We decided that using asymmetric keys generated on users' devices was the best way to overcome objections to software-based tokens. After all, R,S & A had developed public key encryption to overcome the weaknesses of shared secret encryption.

Fast-forward and the dominant form of consumer-oriented two-factor authentication is "two-step" authentication using a shared secret-based protocol (even after hackers successfully stole the shared secretsof a major 2FA vendor) or worse, using SMS. Of course, we know the saying that marketing trumps technology. This seemed like a typical case of that. No one much cared about the increased security offered by asymmetric encryption.

But, security is a slightly different beast because: 1. Attackers are always getting better. 2. Regulationsand compliance can force a market to change despite marketing. The #PCI-DSS Council may be in the process of doing that with their most recent guidance on multi-factor authentication, stating that multi-step authentication leaks account information and should not be used. NIST has said that using SMS as an authentication mechanism is deprecated.

In a way, this will be easier for many systems administrators. Most VPNs and remote access services by default support OTP-based 2FA via RADIUS (which also allows authorization in AD/LDAP another recommended practice) and they do not support a multi-step authentication process. There is no way, for example, to do two-step authentication on a Cisco ASA. But, two-factor authentication is easy and can be added to ASA Admin accounts as well, a great idea and soon to be required for PCI's non-console admin access requirements.

PCI DSS disses multi-step authentication

Posted by: root in PCI-DSS 6 years, 1 month ago

The PCI Council has published an "Information Supplement" on multi-factor authentication (pdf).  The document that multi-step and mutl-factor authentication are not the same and that the former is not acceptable. 

New release, bug fixes and updates

Posted by: root 6 years, 3 months ago

We released a new version of the WiKID server today and it warrants a few notes.

Preventing pass-the-hash via RDP with two-factor authentication

Posted by: root 6 years, 8 months ago

In researching pass-the-hash attacks, we discovered that when Microsoft implemented "Restricted Admin" mode they inadvertantly enabled pass-the-hash attacks via RDP 8.1.  This attack tool is now included in Kali Linux and probably other tools. 

Users: before you use two-factor authentication, make sure the admins do!

Posted by: root 6 years, 9 months ago

Dropbox is the latest internet-based service to suffer a mega-breach.  

• Page 4 of 135
• ←
• 1
• 2
• 3
• 4
• 5
• 6
• 7
• 8
• 9
• 10
• 11
• 12
• 13
• 14
• 15
• →