Do you always feel like somebody's watching you? Well, Rockwell certainly did, and you should, too. I don't mean to make you paranoid, but if you've got valuable data on your System i, then you're part of a crowd that is falling within the crosshairs of organized crime. You may believe your security is rock solid, and in a traditional sense, you may be right. But as the sophistication of attackers increases and vectors multiply, risk can pop up in the most unexpected places. A look at some emerging trends can help you batten down the hatches before someone starts playing tricks on you.

Got My Mind Set on You

With each passing year—or more accurately month, week, or even day—the purpose behind viruses and malicious code increasingly is to facilitate the goals of organized crime. The target: corporate information, customer data, and other sensitive materials that can be exploited for profit. And the bigger the fish, the bigger the potential payoff.

In recent years, there has been an increase in attacks targeting the banking and healthcare sectors, to highlight just two examples of the targeting of vertical industry. Both are excellent examples of organizations that store vast amounts of sensitive information. With such information under their control, criminals can rake in huge bucks. At the same time, the companies that are their victims suffer enormous costs in cleanup and loss of customer confidence.

It's easy to view such attacks as the scourge of those running Windows servers, but that would be failing to see the forest for the trees. Regardless of the platform on which you run your operations, if you've got the goods, then at some point they're coming after you. With System i at the heart of your network, you're starting off from a very defensible position. But it is part of a network, after all, and with that reality comes a wide range of vectors that attackers may exploit. It is not necessary to directly attack System i or execute code under i5/OS to compromise the data stored on the system or bring down the network that it serves.

The Tide Is High

Long gone are the days when a virus arrived via email as an .exe file with a flashing wink-wink-nudge-nudge "click me now" sign on its back. Today, malicious code accosts us at every turn as a wolf in sheep's clothing. Flash drives, iPods, file sharing services, IM, Web sites, employee laptops, and connections to file servers are all examples of the rapidly growing number of roads that lead to the heart of your network.

Today's attacks employ stealthier and more sophisticated approaches aimed at infecting large numbers of systems or penetrating and compromising critical servers. New types of threats are emerging all the time, and "blended threats"—multiple types of malicious code packaged together—are challenging the ability of administrators and software to keep up.

New threats are coming through sources that are considered legitimate business tools, including IM services and even Skype. It is becoming increasingly difficult to balance security with the needs of the company. Many necessary tools are being discarded in the name of safety, so there is a price to pay for this protection against perceived threats.

Another example of such a situation is the use of the popular ZIP archiving format to hide viruses from scanners. Many companies now block email that includes ZIP attachments despite the fact that most commercial anti-virus solutions can scan inside these archives. Given the need to share large files with colleagues and third-party vendors, such blocking can seriously hinder productivity and profitability.

In addition to the appearance of IM and Skype threats, the emergence of "bots"—small programs designed to perform automated (and malicious) tasks—and the reemergence of parasitic malware are adding to the minefield. Quiet automation is a key theme in today's malware, as cyberspace teems with worms constantly searching for open ports, drive-by downloads drop Trojans onto systems without user interaction, and IM worms install rogue browsers that use familiar icons to dupe users.

Anti-virus firm McAfee estimates that approximately 150 to 200 viruses, Trojans, and other threats emerge every day. Indeed, the floodgates are open.

Every Rose Has Its Thorn

As users become wise to obvious tactics such as email attachments, attackers are forced to find new methods of using social engineering against a more sophisticated public; and our IM-, MP3-, MySpace-, YouTube-obsessed society is playing right into their hands. All of these entertainment and communications channels offer wonderful abilities that we could not even imagine a decade ago, but they also bring with them perfect mediums for the distribution of malicious code. And like it or not, unless you have definitive measures in place, employees are accessing these things while at work and exposing your network and your System i.

Attackers can turn the things that make our lives easier and more enjoyable against us by spreading worms via IM, injecting malicious code into audio or video files, dropping malware onto legitimate Web sites and delivering it via drive-by downloads or using meta tags in formats such as MP3, JPEG, and GIF to deliver malicious code that can run when the file is opened. An example of the latter was the 2004 Mac OS X Trojan MP3Concept, which embedded code in the ID3 tag of an MP3 file. MP3Concept was just a proof of concept and did no real damage, but the technique is an example of the craftiness against which administrators must guard.

Another example of useful little gems that could be turned against the user are the handy little programs known for years to Mac users as "widgets" and now, with the release of Windows Vista, to Windows users as "gadgets." Such programs are also available from Google and Yahoo, as well as through other third-party software.

These widgets are typically written in languages such as C++, C#, JavaScript, or Visual Basic. Through APIs, widgets may bypass normal authentication requirements to gain access to system-level services. And because many widgets are written in script languages, an attacker could easily inject malicious code and use a widget's access to services to gain control of and manipulate the system. Lastly, because widgets support JavaScript, an infection would have the ability to jump from platform to platform. This includes potential jumps to OS/400 and i5/OS, which have traditionally been viewed as immune to virus attacks.

Careless Whisper

Another area of concern is employees' use of their own home computers or laptops to perform company work. Since home users are less diligent than corporate IT departments when it comes to installing and maintaining anti-virus software, the use of home systems for company work represents both an unchecked source of infection to your network and an opportunity for data theft at the hand of viruses.

The number of incidents involving laptops or discs containing sensitive customer data being left behind on airplanes seems too numerous to count. This is obvious carelessness by the employee—should've known better—but it also highlights how much control you lose once data and equipment leave the company premises.

But in addition to this obvious danger, there is also the risk that a virus on a user's home system could leak confidential information to the public without the user's knowledge. A prime example is a 2005 incident in Japan involving a Mitsubishi Electric Plant Engineering employee whose home computer was infected with a virus that copied secret information about Japanese nuclear power plants to the popular file sharing service Winny. The data leaked included photos of the inside of nuclear plants and other critical information concerning staffing and inspections. In all, some 40MB of data was leaked.

If that story weren't bad enough, a similar incident plagued Japan's power industry again in 2006 when an employee at the Chubu Electric Power Company installed software for the popular "Share" file sharing service on his computer and sensitive information was leaked to the Internet. And this was a follow-up to another incident that occurred at the beginning of that year.

Shattered Dreams

Compromised data at Japanese nuclear facilities isn't the problem making news. From lookalike online banking sites to DoS attacks to domain hijacking and online blackmail, nary a day passes that there's not a news report or discussion of a security breach involving a trusted company. It's not always the case that the company's IT staff simply dropped the ball. There are so many vectors to guard and the threat is evolving so quickly that sometimes they are just taken by surprise. But then again, sometimes they really do get caught with their pants down. Here are a few more incidents that have made news in recent years:

• In February 2007, the Iowa Department of Education had information—including names, addresses, birthdates, and social security numbers—of nearly 600 G.E.D. recipients compromised when the online database was hacked.
• In January 2007, TJ stores (TJX), which own such stores as TJMaxx, Marshalls, Winners, HomeSense, and AJWright, exposed the credit card information of an undisclosed number of customers (believed to be in the thousands) when the company's systems were compromised.
  
• In November 2006, TransUnion Credit Bureau was hit by four scam companies that downloaded customer credit information after obtaining the password to a TransUnion account held by the Kingman, Arizona, court office. The data stolen included the credit histories and social security numbers of more than 1,700 customers.
  
• In October 2006, VISA/FirstBank had the Visa Check Card numbers of an undisclosed number of customers compromised when a merchant card processor's transaction database was compromised.
  
• In August 2005, Australian automaker Holden was hit by the Zotob worm and was forced to shut down its manufacturing facilities in Adelaide. As a result of the attack and forced shutdown, Holden lost $6 million (Australian) in auto production. At the same time, the Zotob worm also wreaked havoc on U.S. and Canadian businesses, including DaimlerChrysler, ABC News, CNN, The New York Times, and General Electric.
• In May 2005, four major banks—Commerce Bank, PNC Bank, Wachovia, and Bank of America—had the financial records of some 700,000 customers stolen by bank employees and sold to collection agencies.
• In October 2004, police in Brazil arrested 53 people suspected of stealing $30 million through Internet fraud by delivering emails containing hidden Trojans to people over the Internet to record bank account access information. This type of theft can end up dealing banks serious losses when they must reimburse customers as a result of fraud protection.
• In May 2004, in an attack that spanned industries, the Sasser worm caused major disruption of operations at British Airways and the British Coast Guard, among others, by causing systems to repeatedly quit or reboot without explanation. Sasser is significant in that it represented a new, insidious way of delivering malicious code by simply scanning systems from the Internet for a known security flaw and entering the system without any initiation by a user.

Another Brick in the Wall

So what can you do about these threats, and how can you guard your network and System i when an attack can come from so many directions? Certainly anti-virus software is a necessity—and not just on those Windows desktops but on all systems and servers in your environment, be they Windows, UNIX, Linux, or i5/OS. But it's going take a lot more than just anti-virus software to become truly secure.

Architects of physical security have long understood the benefit of layers in achieving their protection goals. The same concept applies to system security and is your best bet when it comes to combating viruses, malicious code, and the actions that can open doors to them.

Layer #1: The Firewall

Firewalls have long been an important element of corporate computing security, and they are now even commonplace in home computing environments. While firewalls do provide important protection, there is a common misconception that they provide an impenetrable wall around the computing environment and therefore are the first and the last step to security. A firewall will keep out casual intruders and provide protection against unwanted incoming traffic. It will not, however, prevent viruses, malicious code, or hacks that enter the environment via email, portable devices, secured connections, or insiders. Firewalls also will not prevent downloads from the Internet, which could introduce malicious code to the system. For that, additional layers are required.

Layer #2: User IDs and Passwords

Strong user IDs and passwords are the next layer of defense against attacks on systems security. If a hacker, a virus, or other malicious code gains access to the system despite the firewall, these access controls can eliminate a large percentage of the risk now facing the system.

Layer #3: Access Control

User IDs and passwords can be compromised, so to thwart those who manage to get past that protective layer, access control to data on the user level is critical. Network security applications can give you control over which files and folders a given user can access, thereby reducing the chance that someone who compromises a given user ID and password will gain access to the company's most sensitive and critical data. This layer also provides protection against threats originating from employees. Access control aspects of security can be cumbersome and difficult to maintain. You should look for a solution that offers easy management and is smart enough to do a lot of the decision-making on its own. For example, you don't want variations in the syntax of a request—such as "Select * from Myfile" and "select * from myfile"—to throw a wrench into the system and require you to train for every possible option. Object-based designs are best at this.

Layer #4: Scanning on All Systems

If a virus or malicious code makes it through the upper layers of your security and finds a home on the system, it is critical to immediately detect and eradicate it on all systems that could host the code. This means that simply scanning PCs is not enough. If the virus is sitting on System i, for example, but is surfacing on the PC, then cleaning only the PC means that you are treating the symptom but not curing the disease. If you have Windows PCs in your network, they all must have anti-virus software running. If you have a System i server, it must have anti-virus software running. If you have Linux PCs, they must have anti-virus software running. Leaving any area unprotected can be counterproductive to the implementation of your other layers and could potentially defeat the overall goal of your security policy.

Layer #5: The Safety Net

Despite your best efforts, security breaches will occur. With a solid, layered approach to security, these breaches should be limited, but nevertheless you must have a last-ditch recovery plan in place. Daily backups are key and supplemental software that can catch deleted files throughout the day can provide a thorough safety net that will allow you to easily recover any deleted data. This layer should also include an auditing process that examines the system and maintains a reasonable log that can be used to pinpoint the causes and results of a security breach to help prevent similar future occurrences.

The Winner Takes It All

The battle for security and data protection is not an easy one, and the rules of this war are changing all the time. There's no comfort in the truth, but knowledge of it will put you in a stronger position to win the fight. The consequences of losing can mean much more than the loss of financial assets. Perhaps more important than the money is the damage to customer confidence that can result, a tarnished image that can linger long after the computers are cleaned up and stronger security is put in place. First and foremost, treating System i as you would any other member of your network—protecting it with anti-virus software and exit-point and access control security—will pay dividends when it comes to shoring up the levees before the next storm.

Christopher Jones is a member of the Bytware team that brings to System i users the tools and knowledge needed to ensure the highest level of security for today’s highly connected, vulnerable computing environments. Bytware’s StandGuard Security Suite includes StandGuard Network Security, StandGuard Recycle Bin, StandGuardAudit, and StandGuard Anti-Virus, which is available for System i partitions running OS/400, i5/OS, Linux, and AIX, as well as for x86-based Linux PCs.