FAA debacle raises questions on system updates—and how IT pros can handle crises
“Whenever something is shotgunned and somebody skips the step in their process, that’s when bad things happen,” CEO says.
Getty Images
“Whenever something is shotgunned and somebody skips the step in their process, that’s when bad things happen,” CEO says.
Getty Images
After a technical issue on a truly bad-luck Friday the 13th last week stranded thousands of travelers in airports across the US, enterprise IT pros could be forgiven for wondering: “Are we next?”
Transportation Secretary Sean Duffy said the problem was due to a faulty router in Arkansas that led to cascading failures. “Our switches, our routers actually got overloaded with data and they…started to give us inconsistency in the data from those routers, and because of that we shut the system down,” he told reporters during a press conference.
However, that didn’t stop people online from connecting the debacle with November news that the FAA had implemented a new AI system to help manage air traffic.
Officials have denied a connection. Nonetheless, the incident shows the stress that aging digital infrastructure is under. For IT pros managing any number of updates and integrations, there’s a very real possibility of failure.
Checking in. IT pros need to focus on testing, said Heather Wood, Outreach’s senior director of data privacy, protection, and AI governance. Catching issues before they balloon into crises is incredibly important; for something as large and critical as FAA, there’s a chance of the repercussions having major effects.
“With these types of systems that are, effectively, running traffic—if you are making a change that has that large of an impact and you aren’t able to test end to end and fully validate it, you are going to have impacts like this when you go to implement,” Wood said.
Monzy Merza, co-founder and CEO of AI security company Crogl, said he doesn’t think of crises stemming from updates as particularly novel, but as long as IT pros and staff focus on following best practices and safeguards for keeping systems in good standing, positive outcomes should follow.
“Whenever something is shotgunned and somebody skips the step in their process, that’s when bad things happen,” Merza said. “Sometimes it’s inevitable, whether it’s a CrowdStrike update—that really nobody was able to predict—or perhaps this FAA one or something similar.”
Third parties. There’s also a responsibility on the part of any vendor providing your enterprise with updates and integrations to make sure that things go smoothly. Even with that assistance and rigorous testing, there will be unknowns. For vendors, that means putting a premium on speed and visibility.
“Very quick communications to your customers on exactly what happened, what it impacted, and what their next steps would be, is key,” Wood told IT Brew. “Being able to have people that are very quick to respond, which partially comes down to understanding exactly what went wrong, but also partially, comes down to ensuring that you have a team that’s nimble.”