Amazon cloud services back to normal function after massive global outage

A technical fault in Amazon’s cloud computing network triggered global chaos across major online services before the tech giant finally restored normal operations.

Amazon Web Services, which leads in global cloud computing market share, undergirds much of the metaphorical plumbing of the modern internet. Picture: Reuters

Amazon says the glitch that caused a massive global blackout of key online services has now been “fully mitigated” and all of its cloud services – a key pillar of the modern internet – are back to normal operations.

On Monday evening Australian time, an Amazon Web Ser­vices error on the eastern coast of the US led users to report higher than usual outages from service providers like Google, Instagram, Snapchat, Telstra, Optus and OpenAI.

According to Downdetector, a service where users report internet outages, reports spiked about 8pm on Monday (AEDT) for all those services.

Amazon Web Services, which leads in global cloud computing market share, undergirds much of the metaphorical plumbing of the modern internet.

Companies and services more often than not host their websites and servers on cloud servers, such as those sold by AWS, ­Google and Microsoft.

Amazon said it identified a DNS issue that ­affected its US-EAST-1 region, its largest cluster of computers.

DNS stands for Domain Name System, which ties URLs like “https://www.google.com” to an IP address for a server somewhere from which to fetch information.

Around 10am AEDT on Tuesday, Amazon said “all AWS ser­vices returned to normal operations”.

The peak of the outage lasted about 2½ hours, Amazon said, which saw “increased error rates and latencies for AWS services”.

The company on Tuesday morning said it would later ­publish a “detailed AWS post-event summary” and that some services would still take a few hours to work through request backlogs.

US-based Synergy Research Group in August estimated AWS enjoyed 30 per cent of global cloud infrastructure market share in 2025, ahead of Microsoft’s Azure platform at 20 per cent and Google Cloud at 13 per cent.

This technical glitch with ­global ramifications is the second large-scale event in the past two years.

In July 2024, Windows computers around the world crashed and displayed the “blue screen of death” error message, due to some faulty code in an update pushed out by cybersecurity firm Crowdstrike.

It affected a broad range of industries, including airports, banks, hotels, hospitals, manufacturing, stock markets, retail stores, emergency services and websites.