Why the IE-Google Incident Should Worry You

Most of the time, alarming reports about zero-day flaws and new security patches make me skeptical. But not when it comes to the attack against Google through a zero-day flaw in Internet Explorer (IE).

Hackers used a security hole in a browser famous for vulnerabilities to compromise the biggest search giant on the planet. In the process, a lot of other companies suffered. Here’s how my colleague, Robert McMillan, described recent events in his article:

Microsoft is scrambling to patch an Internet Explorer flaw that was used to hack into Google’s corporate networks last month. The attack was used to hack into networks at 34 companies, including Adobe, security experts say. Typically such hacks involve several such attacks, but the IE bug is the only one definitively linked to the hacking incident, which security experts say originated in China.

In a security advisory released Thursday, Microsoft said IE 6 users on Windows XP are most at risk from the flaw, but that other users could be affected by modified versions of the attack. Microsoft said it is developing a fix, but it did not say when it expects to patch the issue. The company is slated to release its next set of security updates on Feb. 9. A Google spokesman confirmed Thursday that the Internet Explorer attack was used against Google and that the company then reported the issue to Microsoft.

Google learned of the issue in December and, after discovering the server used to control the hacked computers, notified other companies affected by the hack. Apparently convinced that the infiltration was sanctioned by the Chinese government, Google has threatened to effectively pull its business out of China.

This column isn’t about bashing Microsoft or Google. My opinion is that both companies have been fairly aggressive about security, especially Microsoft. Despite the best effort of companies, the bad guys still score big with increased frequency.

I do, however, see this as a big wake-up call for all of us.

In past columns, I’ve taken security vendors and PR agencies to task for putting out alarmist alerts for every vulnerability that rolls around the pike. My reasoning has been based on history. When I was reporting on Patch Tuesday each month in my previous job, my e-mail inbox would start clogging by 10 a.m. with messages from PR reps eager to get me on the phone with one of their clients to discuss the latest cause for alarm. Then I’d get on the phone with the vendors and hear pretty much the same grim scenario — regardless of the flaw — that I heard the month before. Most of the time, the warnings were not followed by the big attack. Then I’d get on the phone with IT security administrators who wouldn’t be feeling the alarm coming from the outside.

That’s because most of them have a patching process that stretches across several days. The first couple days are for running the patch on test systems to see if any compatibility problems would result from a full deployment. The answer is often yes, requiring the IT staff to make the right network adjustments so the patch will play properly with business-critical applications. A full week often passes between Patch Tuesday and when IT shops deploy all the patches across the network.

In these cases, attacks are kept to a minimum and business rolls along without incident.

But there’s a flip side to the equation.

For every IT shop that has its act together, there’s a company that’s constantly behind in patching systems. I remember when the Zotob attack hit in the summer of 2005 and the victims were primarily organizations running Windows 2000 without patches that had been available for a year or more. In these cases, the victims really have only themselves to blame.

But then there are companies that are on top of the security picture and have strong policies in place to harden defenses. In this class of company, there’s still the potential for human failure, specifically in the case of patch management. It only takes one or two forgotten patches to cause the security breach.

I’m not knowledgeable enough about Google’s internal procedures to know if that sort of thing is happening, though in the incident described above, it’s pretty clear that the attack vector was an unpatched flaw.

And the victim is a company just about all of us have come to rely on to conduct business — a company so dominating that its name has become a verb.

I’m not sure what the right answers are here. It seems as though Google and Microsoft are doing what they should. Microsoft was quick to confirm the flaw and suggest an emergency patch is around the corner. Google’s security concerns have led to a showdown with China that’s attracted the attention of the U.S. State Department, and the company deserves credit for standing tall.

The only advice I can offer is vigilance. Be aware that despite your ironclad patching procedures you’re still potentially vulnerable. It doesn’t matter if you use Internet Explorer, Firefox, Chrome or Opera. All it takes is one weak link in the network and the bad guys will get through.

Awareness is key — making all the computer users in your company aware of the dangers lurking behind the programs they use all day so they will hopefully be careful of the sites they visit and the applications they choose to download (if you allow them to do that!).

But as part of that vigilance, it’s more important than ever for companies to assume that they someday WILL be attacked. When that day comes, the quality of your incident response plan will make the difference. If everyone knows who is responsible for everything from managing an investigation to alerting customers and relevant regulatory bodies, then the company will survive.

Live under the illusion that it’ll never happen to you and the outcome will be sadly different.