Amazon Web Services (AWS), one of the
top cloud services and the backbone for many internet advertising and financial services, experienced a massive outage early Monday that took down many major websites.
The company's ad servers
were disrupted by the widespread outage including the ecommerce website, Prime Video, and Alexa.
Since Amazon's ad services rely on the same AWS infrastructure, they also experienced
failures.
The outage affected some Google services including Gmail, YouTube, and Instagram for thousands of users who connect with AWS. Google Ads, however, remained unaffected and
operational.
advertisement
advertisement
Google Ads runs its own ad cloud services, but some third-party
marketing tools that connect to Google Ads or Gmail were likely disrupted if they rely on AWS for their services. These problems were not directly caused by the AWS outage, because Gmail
runs on Google Cloud, not on AWS.
A
status page for Amazon's cloud
unit showed more than 70 of its services were affected this morning such as Reddit. At 6:03 am ET, the company had begun to experience the "recovery" of most of its AWS Services.
Reddit posted on
its website just after 7 a.m. EST that a fix had been implemented and will continue to monitor the the results.
OpenAI's ChatGPT was among the sites
impacted, The Verge reported.
Aravind Srinivas, CEO of AI search company Perplexity,
confirmed in an X post that its service is down. "The root cause is an AWS issue," he wrote on X. "We're working on resolving
it."
Some X users questioned whether AWS supports too many services, asking how one database could disrupt half of the
internet, with one calling it "too big to exist."
Another wrote: "Everyone preaching decentralization and censorship resistance but in reality ... it's all 100% reliant on the
cloud."
AWS identified a regional gateway on the East Coast of the U.S. that caused
the disruption, but “most requests should now be succeeding,” the company said on its health dashboard earlier this
morning.
The company first reported issues affecting the US-EAST-1 Region at 3:11AM ET, according to reports. “We are actively engaged and working to both mitigate the issue and
understand root cause," AWS wrote. By 5:27AM ET update, Amazon wrote it had begun to see signs of recovery.
Update From Amazon:
"Oct 20 10:38 AM PDT Our mitigations to resolve launch failures for new EC2 instances are progressing and the internal
subsystems of EC2 are now showing early signs of recovering in a few Availability Zones (AZs) in the US-EAST-1 Region. We are applying mitigations to the remaining AZs at which point we expect launch
errors and network connectivity issues to subside. We will provide an update by 11:30 AM PDT."