Taking the fight to end-user password insecurity
What is today's biggest IT security threat? IT itself, according to recent reports from IDC and Carnegie Mellon/US Department of Defense. To begin with, IDC research finds that enterprise companies rank insider sources as their top security threat (Source: "Privileged Password Management," Sally Hudson, IDC).
In addition, research from Carnegie Mellon University for the Department of Defense (DoD) finds that when it comes to insider attacks, 86 percent of perpetrators held technical positions. Of these, 57 percent performed the attack after termination. (Source: Management and Education of the Risk of Insider Threat (MERIT), CERT3 Program, Software Engineering Institute and CyLab at Carnegie Mellon University.)
Both reports found that insider attacks result in costly outages, lost business, legal liability and, inevitably, failed audits. In one case study, it took 115 employees 1,800 hours to restore data deleted by a disgruntled insider. At the time of the attack, the perpetrator was an ex-employee of the IT department who was able to remotely access key systems. According to these reports, IT insiders commonly acquire and maintain powerful system access using privileged accounts and passwords even after termination.
Every day I work with auditors and enterprises in order to minimize the threat from privileged passwords. Here are six of the best practices I've found to battle this menace.
1: Create an inventory of privileged passwords
Privileged passwords are the non-personal, shared passwords that exist in virtually every device or software application in an enterprise, such as root on a UNIX server, Administrator on a Windows workstation, and an Application ID used by a script to connect two databases. Many companies begin the process of securing their privileged passwords by taking an inventory of how many exist and how often they're updated.
In this effort, it is important to note that privileged passwords exist in many places within your enterprise, such as:
- Administrative accounts that are shared by multiple IT professionals and come predefined by the manufacturer. These include UNIX root, Cisco enable, DBA accounts, Windows domain and so on.
- General shared administrative accounts, such as help-desk, fire-call, operations and emergency accounts.
- Hard-coded and embedded application accounts, including resource DB IDs, Generic IDs, batch jobs, testing scripts and application IDs.
- Service accounts such as Windows service accounts and scheduled tasks.
- Personal computer accounts, including the Windows local administrator on laptops and desktops.
Today many organizations still manually update these passwords, if they change them at all. For example, a recent study showed that 42 percent of application passwords are never changed (Source: Cyber-Ark Enterprise Privileged Password Survey.)
2: Define the role of identity and access management (IAM)
When it comes to managing privileged passwords, a common first misstep is to import all administrator or shared IDs into a system built for managing human identities. The benefit of this approach is that you can quickly start to automatically update your organization's privileged passwords. The negative? Your organization still has no way of assigning individual responsibility. For example, the reports will show that the "administrator" identity downloaded your database of top clients at 1:47 a.m. Sunday morning. You won't be able to tie that action — or its consequences — to a particular individual.
To deliver true accountability, your system for privileged password management (PPM) must tie individual identities to shared accounts. This is incredibly sensitive data — a hacker's dream list of all your privileged passwords — so this information must be stored in an exceptionally secure place. IAM solutions are not designed to store sensitive data and typically partner with a PPM solution for the privileged accounts/passwords.
3: Apply change policies to privileged passwords
This may sound obvious, but you'd be surprised how often policies for privileged passwords are not as explicit as those for their human counterparts. For instance, you may now change the password on your laptop every 30 days; however, surveys show that a workstation has a 20 percent chance of never having had the administrator ID changed from its default (Source: Cyber-Ark Enterprise Privileged Password Survey.) In other words, if you lost your laptop, the finder may not know who you are or what company you work for… but they can search the web to find the default administrator password that ships with a Dell Latitude D600. Within seconds, your laptop's new owner will have more access to your systems than you do.
We suggest having an explicit policy that names all the password types uncovered during your privileged password internal survey spelling out update policies for each. Best practices dictate that these policies are at least as stringent as those for individual employees.
4. Make sure privileged passwords are stored securely
Again, this may seem obvious, but it is imperative that organizations store their privileged passwords in the most secure vaulting system available. Placing the passwords in sealed envelopes, locked binders, within an encrypted file or on wallet-sized cards are not acceptable alternatives (and yes, I have seen all of these in use at real-world enterprises).
5. Create a staged approach to deployment
Privileged passwords are literally the keys to your kingdom and must be controlled properly. One common stumbling block for projects around privileged passwords is that once the password inventory is created, the sheer volume and prevalence of these codes is overwhelming. Personnel can throw up their hands, saying: "We never secured these before so why bother now?" In these situations, the most successful auditors take a deep breath, drink a tall latte and start putting together a stepped plan with reasonable deadlines, deliverables and consequences.
6: Remember computers are people, too
While 99 percent of enterprises change passwords for employees, up to 42 percent never change hard-coded and embedded passwords for application IDs, testing scripts and batch jobs. (Source: Cyber-Ark Enterprise Privileged Password Survey.) According to Mark Diodati of the Burton Group, this creates an App2App password problem that is exponential.
"For example: 300 hosts times 2 applications per host times 5 scripts per application = 3,000 stored passwords." Often, these passwords are in clear text and readily available to every developer or database administrator in an organization.
All in all, no privileged password management system is complete without an App2App component. However, since application passwords are stored in scripts that must be re-coded, tested and deployed, most organizations I work with break out fixing past code from making future mistakes. Once again, a stepped plan can be your best friend.
One final note: no policy for managing privileged passwords would be complete without related reporting structures. Audit reports for privileged passwords often cover such topics as when passwords are updated, any update failures and which individual identities performed tasks under a shared account.
So there you have it: the greatest threat now posed to IT security is due to the smallest of things, a tiny code embedded in virtually every piece of hardware and software. However, armed with a strong plan and the knowledge that you're protecting your organization, any auditor can become a successful warrior against today's top IT security menace.
Calum MacLeod is European director for Cyber-Ark Software