April 30, 2013, 12:27 PM — I've spent the last few weeks evaluating some technologies to improve my company's security posture. These are replacements for products we already have, as opposed to brand-new technologies for us. That's because some of our products are getting past their support lifetime, and they're not performing up to my expectations. This is an interesting situation, because I think it's the first time in my career that I've had the luxury of focusing on technology improvement instead of closing a gap.
The first area I'm looking at involves our patching capabilities. I've written recently about Java vulnerabilities and their associated zero-day exploits and our efforts to keep up with all the software updates. Until now, the way our IT department has been dealing with that was to manually push out and install the updates. The same was true for Adobe products. But the newest versions of these products have built-in capabilities to check for and install updates and new versions.
Auto updates can cause problems. Updates that self-install without any in-house testing may not work properly with all of an organization's software, breaking functionality. You then have to spend a lot of time troubleshooting and resolving that issue. Another problem is that end users are bombarded by update prompts, sometimes several times a day. Users usually don't know how to deal with installation failures, which can happen because of connectivity or software issues. And they don't know how to determine whether an update is legitimate. With so many fake update scams going around that try to trick users into installing malware, it's hard for them to know when it's OK to install an update and when it's not. In fact, I'd like to tell them never to install any updates, and let IT handle it.
In short, automatic updates are really not the best way to keep enterprise software up to date. Yet we still need to install the security fixes as soon as possible.