Data Centre


AWS users felt a great disturbance in the cloud, as S3 cried out in terror

S3izure made things tricky for an hour, but was no apocalypS3 to match March mess

By Simon Sharwood


The world received an unpleasant reminder of what it's like to live without the cloud on Thursday, after Amazon Web Services' Simple Storage Service fluttered for an hour or so.

The incident invoked memories of the S3 outage in March 2017 that caused interruptions to plenty of web services and apps, sparking much rending of garments and gnashing of teeth as the fact of AWS being fallible worked its way into the minds of the faithful.

Amazon S3-izure cause: Half the web vanished because an AWS bod fat-fingered a command


The US-EAST-1 region that caused so much trouble in March was again the culprit on Thursday, as at 11:58AM on Thursday AWS reported “increased error rates” on S3. By 12:20 the company admitted “We can confirm that some customers are receiving throttling errors accessing S3.” By 12:38 the problem was identified and the fix commenced, error rates fell by 12:49 and at 1:05 AWS sounded the all-clear, confident that errors had ceased nine minutes previously.

AWS CodeCommit, Elastic Beanstalk and Storage Gateway all wobbled, too, and all in the same North Virginia data centre.

While this incident was nowhere near as bad as March's ApocalypS3, users were predictably grumpy about this latest S3izure:

AWS hasn't revealed the cause of the problem. If it does we'll update this story. For now, another Tweet on the incident looks like good advice.


Sign up to our NewsletterGet IT in your inbox daily


More from The Register

Database deletion downed Digital Ocean last week

Control panel busted, but servers stayed online

Cloud VMs without sane firewalls is nutty, right? Digital Ocean agrees

Build the wall! Build the wall! Build the wall! Build the wall!

Brazil says it has bagged Royal Navy flagship HMS Ocean for £84m

Four years ago Brit taxpayers spent £65m doing her up

GitLab offers Digital Ocean to soak up customers' CI burden

Don't worry, yells CEO, we have a data centre in the EU

Microsoft sends a raft of Windows 10 patches out into the Windows Update ocean

Whoa - is that an Access 97 iceberg dead ahead?

Multi-cloud Cloudian controllers now run in AWS, Azure and Google

One namespace to rule them all

Sick of bandwidth gouging? Cloudflare, Google, Microsoft, IBM have some good news

Updated Bandwidth Alliance decides to throw us a bone

Another container-sized nugget: Scality S3 server joins nice and small party

S3 gets direct appeal

China pollutes ocean with bloody big rocket

Long March, short flight

Things that make you go .hm... Has a piece of the internet just sunk into the ocean? It appears so

Updated If anyone's seen Ed Sweeney recently, please let us know