Amazon’s cloud arm is still struggling to recover after a major cloud services outage impacted the likes of Canva, Roblox, Snapchat, Amazon and other companies globally, including in Australia, affecting millions of users and effectively bringing the internet to a standstill.

The issue began at about 6pm AEDT on Monday and is related to Amazon Web Services – tech giant Amazon’s cloud services arm – and the AWS status checker showed that multiple services were “impacted” by operational issues.

Within hours, more than 6.5 million reports flooded outage tracker Downdetector from more than 1000 affected companies globally, making it one of the largest cloud infrastructure failures in recent history. AWS controls roughly 30 per cent of the global cloud market, with Microsoft and Google holding 20 per cent and 13 per cent respectively.

The AWS outage follows other major cloud failures, including a 2021 disruption and last year’s CrowdStrike incident that crashed Microsoft Windows systems globally. As of 8am on Tuesday, AWS issues were still ongoing.

The company was “investigating increased error rates and latencies for multiple AWS services in the US-EAST-1 Region”, it said, though the issues affected websites globally.

Cloud computing is part of the backbone that runs the internet. Everything from start-ups to large enterprises use AWS. Netflix streams video through it, Airbnb runs its platform on it, and countless companies host websites and applications there.

Amazon said its engineers were working urgently on remediation. “We have identified a potential root cause for error rates for the DynamoDB APIs in the US-EAST-1 Region,” the tech giant said in a statement at 8.01pm AEDT. “Based on our investigation, the issue appears to be related to DNS resolution of the DynamoDB API endpoint in US-EAST-1.”

Amazon said at 9.35pm that the “underlying” issue had been “fully mitigated”, one hour after initial recovery signs were recorded. It warned that a backlog of requests may result in some services taking additional time to come online, and cause some users to still encounter web issues.

“Most requests should now be succeeding,” Amazon said. “We continue to work through a backlog of queued requests. We will continue to provide additional information.”

Amazon was working towards a “full resolution” following the outage.

Canva first reported at 6.14pm that its features were “currently unavailable for some users” due to “significantly increased error rates”, and first observed signs of recovery at 8.43pm.

Amazon’s Alexa service was also reportedly down for many users globally.

A spokesperson for Snap, the parent company of Snapchat, urged users to “hang tight” as it investigated “issues”. Snapchat returned online shortly before 9pm.

About 50 per cent of on-demand content on Australian streaming service 9Now, owned by Nine, experienced playback issues during the outage, according to an internal incident alert. The incident was resolved at 9pm.

Telstra and Optus users also reported issues, though both telcos confirmed call services weren’t impacted by the outage. An Optus spokesperson said that data services reliant on the AWS cloud system experienced some disruptions, but that its network remained fully operational.

The financial toll will likely be significant. Mehdi Daoudi, chief executive of internet performance monitoring firm Catchpoint, estimated losses reaching into the hundreds of billions of dollars due to halted productivity for millions of workers and delayed business operations spanning airlines to manufacturing.

Amazon Web Services was contacted for further comment.

Start the day with a summary of the day’s most important and interesting stories, analysis and insights. Sign up for our Morning Edition newsletter.