I want to know! Hit reply to share your story. (Paid subscribers can comment below.)

A funny-but-true saying in my line of work: There’s always a restaurant angle.
A major Monday outage at a huge tech company caused chaos across the internet — and inside restaurants. Amazon Web Services, a cloud service provider that, per the New York Times, “supports much of the internet,” experienced an hours-long outage that started at 3am ET. By 5:30am, Amazon said most websites and apps that use its service were working, but it had a backlog of requests that seemed to cause trouble throughout the day, including for America’s restaurants.
The outage also caused trouble for restaurant tech companies including delivery services DoorDash and Grubhub and point of sale and payments provider Toast. Issues affected operations at Starbucks and Sweetgreen and McDonald’s — and those are just examples I found in the first few minutes of Google searching.
Amazon attributed the problem to resolution issues in its domain name system, or DNS. Expedite’s in-house expert, my software engineer husband, explains it like this:
“DNS is what turns names into numbers on the internet. When this fails, basically everything falls apart because computers can’t talk to each other. And it sometimes becomes hard to fix because you need the DNS to talk to the computers to fix the other computers.”
This expert, who had not read up on Monday’s issue before I shouted my questions from the next room, told me that there’s no way for an outsider to know exactly what happened.
“It could be anything. DNS is run by computer servers which can fail just like any others. And the engineers can mess them up just like they can mess up anything else.”
Per NYT reporting, Amazon said an initial investigation suggested the issue started from an internal monitoring system. An expert quoted by CNN estimated the total impact of the disruption to be in the billions of dollars.
Most outage coverage zeroes in on a particular problem: the web is too centralized.
That means when one popular service breaks, it causes ripples.
Amazon reportedly controls a third of the cloud computing market, including key services that perhaps haven’t planned for redundancy. This has huge effects for technical businesses; Amazon’s Alexa stopped working yesterday, as did payments app Venmo and social media app Snapchat. But when a broken internet translates to in-real-life service challenges — like it does inside America’s restaurants — the problem compounds. (I wonder how many Starbucks employees had to try to explain the effects of a complex tech issue to grumpy, under-caffeinated patrons?)
In one Reddit thread, workers at restaurants using Toast reported trouble with clocking in and out, menu changes, logging in, online ordering, in-store payments, and more.
A rep for Toast acknowledged the outage had affected restaurants on the platform but declined to comment further. They pointed me to the company’s status page, which lists the issue as fully resolved soon after midnight this morning, some 18 hours after the initial outage alert.
Did you experience restaurant-related trouble because of Monday’s outage?
More coverage:
Amazon outage forces hundreds of websites offline for hours — New York Times
Amazon’s global outage exposes major vulnerabilities to American life — CNN
The AWS outage reveals the web’s massive centralization problem — Fast Company
Major AWS outage took down Fortnite, Alexa, Snapchat, and more — The Verge
As a hospitality consultant, I end up getting pigeonholed into dealing with these kinds of tech issues outside of our normal operational consulting. Some of the biggest headaches yesterday in NYC came from improper setups behind the scenes where "offline" mode would have not been so troublesome if the onboarding of proper tech had been done.
Yep. Toast and that it does for us has been lovely (we’re Aloha at our other place), but when it goes down so do we.
By late afternoon things were starting to work again, but by then it was too late for getting dinner service covered.
We lost a full day…