Network kit failure causes AWS outage

Amazon Web Services has attributed a one-hour outage in one of its US-EAST cloud compute availability zones to the "partial failure" of network equipment.

The outage, which occurred in one of five availability zones in Amazon's Northern Virginia data centres, ran from 12.51pm PDT (5.51am AEST) to 1.42pm PDT (6.42am AEST).

Instagram and Vine social media services were among the bigger names impacted by the outage, according to Techcrunch.

Travel rental firm Airbnb also confirmed via Twitter it was down due to the AWS outage.

The outage led to a "small number of EC2 instances ... becoming unreachable", and also caused issues for elastic load balancing and relational database services, according to AWS' service health dashboard.

"The root cause was a "grey" partial failure with a networking device that caused a portion of the AZ [availability zone] to experience packet loss," Amazon said in a incident advisory.

"The networking device was removed from service and we are performing a forensic investigation to understand how it failed."

A Bulletproof spokesperson told iTnews that no Australian-hosted AWS customers were affected. Bulletproof is an Australian AWS partner.