A number of US airlines have been grounded for around 40 minutes by a software failure, causing significant delays to passengers travelling across the country.
The issue is believed to have been caused by a problem with something called AeroData. AeroData program that helps to manage things like weight and balance – flight controllers need full visibility on this for planes to be allowed to fly.
A tweet by the Federal Aviation Authority this morning confirmed the problem, citing “computer issues” as the reason for delays across the U.S.
#Traveler Alert✈️: Several U.S. #airlines are experiencing computer issues this morning. Please contact your airline directly for flight information and updates. The #FAA does not cancel flights. #FlySmart pic.twitter.com/5x4U3f6ogu
— The FAA (@FAANews) April 1, 2019
However, within an hour, the FAA provided a further update, saying that “the issue has now been resolved.” However, given that even a short delay would have a knock on effect on flights throughout the day, it advised passengers to get in direct contact with the airlines with which they are flying.
A spokeswoman for Delta is quoted on USA Today saying “a brief third-party technology issue that prevented some Delta Connection flights from being dispatched on time this morning has been resolved.”
Flight delays are making the case for software resiliency
The details of the failure have not yet been revealed, but it nevertheless looks like the air industry is making a particularly strong case for investing in the resiliency of software.
“These airline outages will keep occurring unless something changes.”
Kolton Andrus, CEO and co-founder of chaos engineering platform Gremlin said:
“These airline outages will keep occurring unless something changes. Systems are becoming more complex, and more than ever they rely on software that breaks. While we should continue to celebrate airlines that respond quickly, resolve issues, and maintain a good customer service — we should celebrate even more the engineering teams at airlines who are catching problems before they cause outages in the first place.”
Read next: Chaos Engineering: managing complexity by breaking things