Amsterdam: The company behind this week’s massive Internet outage is due to a software bug triggered when one of its customers changed settings on Wednesday (June 9th). Said.
The power outage on Tuesday cast doubt on the dependence of the Internet on some infrastructure companies. Fastly issues have knocked out news providers such as The Guardian and the New York Times, as well as high-traffic sites such as the UK government’s sites Reddit and Amazon.
In a blog post written by senior engineering and infrastructure executive Nick Rockwell, the company sincerely apologizes for the widespread and serious outage that has affected customers and everyone who depends on them. I will do it. “
He said the problem should have been expected.
Quickly operate a group of strategically deployed servers around the world, enabling customers to quickly and securely move and store content closer to their end users.
The company post provided a timeline for the event and promised to investigate and explain why Fastly was unable to detect software bugs during its own testing process.
I asserted that the bug was in a software update shipped to a customer on May 12, but an unidentified customer made a configuration change and caused a problem until “85% of the network returned an error”. did not.
The outage that occurred at 0947 GMT was immediately noticed within a minute, and the engineer pinpointed the cause at 1027 GMT. Disabling the settings that caused the problem quickly restored most of the company’s network.
“Within 49 minutes, 95% of the network was working properly,” the company said.
The network has fully recovered to 1235 GMT and has begun deploying permanent software fixes at 1725 GMT, Fastly said.
Quickly blame major global internet outage software bugs
Source link Quickly blame major global internet outage software bugs