Some of the world's biggest apps and websites have been impacted by a widespread outage linked to a technical issue with Amazon Web Services (AWS).

Roblox, Snapchat, Fortnite, Reddit, Canva and Duolingo were among the services to go down, according to website Down Detector.

The problems, which diminished gradually from about 9pm, appear to stem from AWS, which provides infrastructure for many major websites and apps.

Reports of problems with the hugely popular services began to surge between 6pm and 7pm today and remained high for more than an hour in many cases.

Snapchat was still dealing with issues late into the night.

"We're aware that some Snapchatters are having issues - hang tight, we're looking into it!" the social network's support account said on X just before 11pm.

Just after 8pm, AWS said it had"identified a potential root cause" and by 8.30pm, AWS it was"seeing significant signs of recovery".

By 9.35pm, AWS said the"underlying DNS issue has been fully mitigated" and that most requests were succeeding normally.

"Some requests may be throttled while we work toward full resolution," it said, adding that some services were dealing with backlogs.

The DNS (Domain Name System) is often described as the phonebook of the internet. It converts domain names such as www.9news.com.au to the actual IP addresses of the websites or services.

AWS has been providing technical explanations, saying"we are working on multiple parallel paths to accelerate recovery".

"We recommend customers continue to retry any failed requests," AWS said.

"We will continue to provide updates as we have more information to share."

Reddit remained down for an extended period but it was unclear if it was linked to the AWS outage. Reports of an outage began much later than the other services and remained high until about 10pm.

The company said it was"experiencing an elevated level of errors and are currently looking into the issue".

Other social platforms such as Meta's Facebook and Instagram, X and Bluesky remained relatively unaffected.

Video streaming services such as Disney+ Amazon Prime and 9Now, which is owned by the publish of this website, were also affected.

Some artificial intelligence tools were impacted. ChatGPT users didn't flood Down Detector with complaints but OpenAI reported problems with logins lasting at least half an hour.

Perplexity CEO Aravind Srinivas, posting to X, blamed an outage with his AI search platform on an"AWS issue".

Popular encrypted messaging service Signal had reports of its service being down.

"We are aware that Signal is down for some people," president Meredith Whittaker said on X.

"This appears to be related to a major AWS outage."

The currency trading app Coinbase confirmed"many" of its users were unable to access its service.

"Our team is working on the issue and we'll provide updates here," the company said.

Graphic design platform Canva said it was experiencing"significantly increased error rates which are impacting functionality".

The software company has about 220 million active monthly users.

"Our team is actively investigating the issue and working to restore full access as quickly as possible," the company said in an update on its website.

Popular online gaming platform Fortnite, which has about 80 million active monthly users, confirmed the outage was affecting its login pages.

"We're investigating this now, and will update you when we have more details," the gaming platform said on X.

The outage was also reportedly affecting some overseas banks, including the Bank of Scotland.

"You may have seen reports of issues with Amazon Web Services affecting a number of websites and apps across the UK today," the bank said on social media.

"We know this is impacting some of our services right now. We're sorry about this. Please bear with us as we investigate this."

What apps and websites went down?

Down Detector showed a huge list of sites and apps that users were reporting errors with, including:

• Epic Games Store
• Fortnite
• Hinge
• Tinder
• The PlayStation Network
• XBox
• Steam
• Pokemon GO
• Zoom
• Xero
• Snapchat
• Roblox
• Ring
• My Fitness Pal

See the full list here .

Users also reported difficulty accessing Amazon services including Amazon.com, Alexa and Prime Video, according to Down Detector.

What is AWS?

Amazon Web Services provides cloud computing to many of the world's most-used online services.

AWS is a subsidiary of Amazon, the mega company owned by billionaire Jeff Bezos, one of the world's richest people.

In Amazon's infancy, the company needed excess server capacity to ensure it had enough computing power to handle the massive amounts of traffic that came to its site during the holiday season rush.

Amazon realised that during the rest of the year, it could use those servers to support other companies' online needs, and out of that AWS was born.

"Instead of buying, owning, and maintaining physical data centres and servers, you can access technology services, such as computing power, storage, and databases, on an as-needed basis," AWS says.

It provides infrastructure that is used for data backup, disaster recovery, email, virtual desktops, software development and testing, big data analytics, and customer-facing web applications.

What exactly went wrong?

Among AWS' many offerings is DynamoDB, a database that hosts information for companies, including customer data.

Amazon morning said its customers couldn't access the data stored in DynamoDB because the DNS had encountered a problem.

"Amazon had the data safely stored, but nobody else could find it for several hours, leaving apps temporarily separated from their data," said Mike Chapple at University of Notre Dame.

"It's as if large portions of the internet suffered temporary amnesia."