IoT devices and even “Is it down right now”? went down during Amazon S3 outage.

When the Amazon infrastructure-as-a-service cloud goes down, Internet users are going to notice.

Amazon Web Services, which powers a whole bunch of websites and online services, has been struggling today, and numerous sites that rely on Amazon infrastructure have gone offline as a result. Appropriately enough, "Is It Down Right Now?," a site that tells you whether other sites are down, has been struggling to stay online. Other apparent victims include The AV Club, Trello, Quora, IFTTT, Open Whisper Systems, and websites created with Wix.

Gillian Owen @gilliancowen
Oh the irony! http://isitdownrightnow.com is down now too. #aws #s3
3:42 PM - 28 Feb 2017

Amazon itself was initially having trouble providing updates to its service health site. But the company posted this note at 2:35pm ET: "We have now repaired the ability to update the service health dashboard... We continue to experience high error rates with S3 in US-EAST-1, which is impacting various AWS services. We are working hard at repairing S3, believe we understand root cause, and are working on implementing what we believe will remediate the issue."

S3 is Amazon's Simple Storage Service. The problems are originating from an Amazon data center region in Virginia.

Further updates from Amazon reported a full recovery in "S3 object retrieval, listing and deletion," but continued problems in adding new objects to S3.

(UPDATE: Amazon said at 5:08pm ET that everything had returned to normal.)

IFTTT's status page started reporting problems at 12:47pm ET today and later blamed the outage on Amazon. A smart home security system called Piper blamed problems today on its "cloud service provider," without specifying which one.

The "Internet of Shit" Twitter account has been retweeting people who say Internet-connected devices such as an oven, remote light controllers (including one powered by IFTTT), and a front gate have been affected. We'd put these reports in the "unconfirmed" category, but problems with these sorts of devices wouldn't be surprising since so many services rely on Amazon.

Piper @hello_piper
@heyboubou Our cloud service provider went down which may cause connectivity issues. We're monitoring the situation closely.

Kayla Bourgeois @heyboubou
@hello_piper any idea how to get the piper alarm to stop if we can't connect to it to disarm it? I think my mom is legally deaf now.
5:18 PM - 28 Feb 2017

Amazon Web Services passed 1 million users last year and has been wildly successful at providing on-demand computing infrastructure to everyone from startups to large companies. The Amazon cloud makes it possible to build online businesses without buying tons of IT hardware; even Netflix, which has built its own worldwide content delivery network, relies on Amazon for many important business operations.

But no matter how robust Amazon's data center operations are, there's always a risk in putting large portions of the Internet on one service. A days-long Amazon outage took down sites including Reddit, Foursquare, and Quora back in 2011, and outages continue to pop up from time to time.

Amazon has always been good about providing extensive post-mortems, so we may be treated to a detailed description of what went wrong in the coming days.

That must be what happened to Imgur while I was at work. I had to play freecell instead

I was in a meeting about selling AWS when this happened. I mentioned it and they quickly spun it into an "always build for failure" conversation.