Internet services and a number of major websites were unusable for several hours on Tuesday due to a blackout caused by problems originating from cloud computing firm Cloudflare.
Sites including X, Meta's Facebook and OpenAI's ChatGPT were among those affected.
Even Downdetector – which tracks global internet cuts – was down.
Several moments after problems were experienced worldwide, Cloudflare's status dashboard said: "A fix has been implemented and we believe the incident is now resolved. We are continuing to monitor for errors to ensure all services are back to normal."
That fix, however, took several hours to solidify. By Tuesday afternoon, the company said the issue was was resolved.
"Cloudflare services are currently operating normally ... we are no longer observing elevated errors or latency across the network," a statement read.
Though details were few and far between about the exact cause of the problem, Cloudflare's issues lingered throughout the day, though the company promised that "details about the incident will be made available asap", and that recovery was still under way.
It is the latest in a string of cloud, data centre and overall web infrastructure cuts affecting various companies that have had a domino effect on websites globally.
Some have argued that the cuts are a new normal, while others have said the problems are rooted in a lack of competition in the cloud and overall tech sector.
Those who operate businesses, apps and services that intensely depend on cloud computing stability were quick react.
“Cloudflare had an outage today and another provider will have one tomorrow,” said Monica Eaton, chief executive of Chargebacks911, a US payment services company.
“What matters is whether businesses learn from these moments or keep hoping luck will cover the gaps. Cloudflare going dark today should snap every merchant back to reality,” she said, noting the importance of redundancies and backup plans.
