AWS us-east-1 outage on 15 November drew little public attention
AWS confirmed a pre-dawn incident in its us-east-1 region that hit sign-ins and connectivity for nearly four hours, according to a column in The Register.
- What happened
- AWS confirmed an incident on 15 November that disrupted multiple services in its us-east-1 region, The Register's columnist Corey Quinn writes.
- What’s new
- The status page reportedly lagged the start of problems by 93 minutes, and the incident ran from 3:05 AM to 6:53 AM Pacific time.
- Why it matters
- us-east-1 is AWS's oldest and most important region, Quinn writes, yet the outage passed largely unnoticed because it mostly affected engineers rather than consumers.
Amazon Web Services confirmed an incident on 15 November that disrupted services in us-east-1. Corey Quinn, writing a column in The Register, describes it as the company's oldest and most important region. According to Quinn, the incident affected 11 of what he counts as roughly a dozen high-quality AWS services.
AWS stated that it began seeing problems with sign-ups, authentication and connectivity at 3:05 AM PST, and engineers were engaged five minutes later, Quinn writes. He says the AWS status page did not post any sign of an incident until 93 minutes after the problems began. The all-clear came at 6:53 AM.
The service hit hardest was IAM Identity Center, which engineers use to sign in to AWS accounts, according to the column. In Quinn's view, that is one reason the outage escaped headlines. Others he gives are its timing before dawn on a Sunday, Pacific time, and a Microsoft 365 outage lasting more than a day and a four-hour nationwide air traffic ground stop, which he says absorbed public attention.
Quinn also notes that AWS spokesperson Patrick Neighorn had criticised Downdetector's reliability to a reporter. Quinn argues that because Downdetector tracks consumer complaints, it is the reason this outage went largely unnoticed.