Amazon Web Services suffered another widespread outage today (Dec. 22), in the third such instance this month.
"We are investigating increased EC2 launch failures and networking connectivity issues for some instances in a single Availability Zone (USE1-AZ4) in the US-EAST-1 Region," read a post on the Amazon Web Services (AWS) Service Health Dashboard at 7:35 a.m. Eastern time. "Other Availability Zones within the US-EAST-1 Region are not affected by this issue."
The problem was blamed on "a loss of power within a single data center." Power was said to be restored by 8:39 a.m. Eastern time, and affected services "should be starting to see recovery." A 9:51 a.m. update said that "the majority of AWS services have also recovered."
Online companies reporting service problems included Slack and Epic Games, while Down Detector showed users reporting trouble with Grindr, Hulu, McDonald's and DoorDash, among others, although it wasn't clear whether all those issues were caused by an AWS outage.
Internet services outages are currently impacting the Epic Games Store, affecting logins, library, purchases, etc. We are monitoring the situation and we will update you when the issue is resolved.December 22, 2021
Some of those impacted services appeared to be recovering by 9:30 a.m. Eastern time, according to Down Detector. However, Slack reported 5 minutes later that "issues with file uploads are ongoing," and then at 10:06 a.m. Eastern thanked its users "for sticking with us as we continue to work towards a fix."
According to Statista, AWS comprises 32% of the entire worldwide cloud-infrastructure market.
AWS had similar issues on Dec. 7, when one of its servers in the eastern U.S. went down, and again on Dec. 15, when the problem was traced to a West Coast server. Today's snafu, as indicated above, seems to lie with another Eastern-region server.
The problems seemed to have been more wide-ranging in the previous two AWS outages, affecting Disney Plus, Netflix, Sony's PlayStation Network and Amazon's own Prime Video, Alexa, Ring Video Doorbell services and even delivery-driver navigation.
However, a Reddit thread indicated that some of those services were affected this time as well.
On the 6th day of Christmas AWS gave to me, 6 servers frying, 5 stolen dataaaaa! Log4j, 3 power outage, 2 missing backups, and an overdraft fee.December 22, 2021