Tech Time Warp: How human error leads to July 1997 email outage
This week’s tech time warp looks at how, in a world that’s so utterly reliant on the internet, service outages are hardly uncommon anymore, whether they’re the result of distributed denial of service (DDoS) attacks, like that which caused Microsoft service outages in early June, or can be attributed to human error, like the grounding of flights Jan. 11 because contractors deleted some critical files. All you can is sit back, use another service and—it’s a Zoom outage—enjoy a break from meetings.
Such service disruptions weren’t routine back in July 1997, when a small human error resulted in a major email outage. A domain name server (DNS) at Network Solutions became corrupted during its daily middle-of-the-night regeneration of .com and .net domains, making it impossible for computers to match user-entered URLs, such as www.nytimes.com, with IP addresses. This also stopped delivery of email messages sent to addresses using affected domains. The problem was resolved within four hours, but not before thwarting the delivery of millions of messages.
Such outages are inevitable, and, while some are the work of the nefarious, most are the result of human errors, such as problems with routing protocols. The Uptime Institute’s Annual Outages Analysis 2023 reports that configuration/change management issues (i.e., human error) contributed to 64 percent of outages, followed by firmware/software issues at 40 percent and hardware failure at 36 percent. Interestingly, cyberattack and security issues were cited in only 10 percent of outages analyzed by Uptime. The price tag for dealing with an outage is high, with 45 percent of those who had experienced outages estimating their repair costs at between $100,000 and $1 million—a pretty penny for a simple mistake.
Did you enjoy this installation of SmarterMSP’s Tech Time Warp? Check out others here.
Photo: alleachday / Shutterstock