MOBILE WORLD CONGRESS, BARCELONA Even as the world’s largest mobile trade show kicked off, analysts and vendors on the ground in Barcelona were still intrigued by what had caused the major AT&T outage last week.

As you’ll remember, over 70,000 people in the United States and parts of Canada were hit by an AT&T outage early Thursday morning. It effected AT&T customers and T-Mobile and Verizon users trying to call AT&T customers. It wasn’t fully fixed until late Thursday afternoon. Sunday night at Fierce’s networking party at the W Barcelona, partiers told us that some AT&T customers in Dallas experienced additional outages two days later.

We checked in with our AT&T spokesperson on Monday, Feb. 26, 2023, at MWC and she said the operator did not have any wide-spread outages on Saturday.

AT&T has now said it will offer a $5 credit to monthly contract customers, not prepaid, business or Cricket punters.

Roy Chua, analyst at AvidThink, said in email this weekend that the credit “is probably more than the monetary value of a day’s worth of service given the outage was about 11 hours,” but added that “the inconvenience for customers unable to make calls (except over Wi-Fi) probably felt like it exceeded the $5 cost of a cafe latte in some cities.”

Chua said in the email that he had no more visibility about what had caused the outage than what had already been made public.

Talking, however to Fierce at our aforementioned networking party, he, like other vendors and commentators there, wondered if the outage was caused by a database going down. The outage only seemed to affect 5G devices, certainly, some on Reddit wondered if a SIM subscription database malfunction might be to blame.

More outages on the horizon?

Chua, however, expects to see more major outages in the future. “Network outages, whether wireline or wireless, whether at mobile carriers, or fixed-line carriers, or hyperscalers, or major web properties will recur,” he said. “There are so many ways for networks to fail — physical media (fiber cut), network equipment failure (despite redundancy), network software failure, human errors, automation failures, cyberattacks.”

More failures will definitely occur, he stated.

“However, the duration of the failure and the mean time to service restoration (MTTR) should hopefully reduce as we adopt improved automation, predictive analytics with proactive mitigation, and as autonomous resilient systems become better.” he concluded.

So there are likely more outages to look forward to! Meanwhile, we’ll keep chasing the cause here in Barcelona and keep you updated on what we find out.

Original article can be seen at: