Amazon Web Services has been hit by a distributed denial-of-service (DDoS) attack, resulting in an outage that affected many websites.
The company Wednesday morning said it's investigating reports of intermittent Domain Name System (DNS) resolution errors with Route 53 and its external DNS providers. The errors have since been resolved as of midday Wednesday.
We're investigating reports of intermittent DNS resolution errors with Route 53 & our external DNS providers. We're working towards resolution & will post updates here: https://t.co/Frz0O4RoSl.— AWS Support (@AWSSupport) October 22, 2019
AWS’ status page said the errors happened occured from 10:30am to 6:30pm US Pacific Time, or 4:30am to 12:30pm AEST, with some experiencing a higher error rate at 5:16pm PT or 11:16am AEST.
An AWS customer posted a response from AWS Support on Reddit, confirming the DDoS attack.
“The AWS DNS servers are currently under a DDoS attack. Our DDoS mitigations are absorbing the vast majority of this traffic, but these mitigations are also flagging some legitimate customer queries at this time,” the response read.
“We are actively working on additional mitigations, as well as tracking down the source of the attack to shut it down.”
AWS Support also said the DNS resolution issues affected other AWS service endpoints like ELB, RDS and EC2 that require public DNS resolution.
Users took to Twitter to either inform their customers or express their displeasure.
Some of our users are experiencing slow or failed file uploads. This is due to a known and widespread issue with AWS, which Outklip relies on. We will provide updates as and when the issue is resolved.— Outklip (@outklip) October 23, 2019
AWS is having DNS related issues. We are seeing this manifest as errors when attempting to retrieve jobs from SQS. This is affecting Forge, Envoyer, and Vapor.— Laravel (@laravelphp) October 22, 2019
AWS DNS came under some kind of DDoS attack today, resulting in DNS resolution errors. 5-6 hour outage. The AWS status webpage displayed an alert, but confusingly, health indicators were green. Would help to pinpoint services affected by outage, @AWSSupport. https://t.co/LbEE8N8zcI— Sunil K. Srinivasan (@kowlgi) October 23, 2019
We need to have a talk about the AWS DNS issue. First, #hugops, good luck to those dealing with it, yadda yadda.— Scott Piper (@0xdabbad00) October 23, 2019
Next, AWS is telling customers its a DDoS (screenshot from reddit and same text seen elsewhere). pic.twitter.com/lYT2QKfPpO