AWS Tokyo outage takes down banks, share traders, and telcos

Six-hour slump for Direct Connect caused by 'loss of networking devices' – we'll assume that means they broke, not that they fell behind a couch


The AP-NORTHEAST-1 region of Amazon Web Services, located in Tokyo, has endured six hours of sub-optimal performance.

The cloud colossus's status report states that AWS Direct Connect hybrid cloud networking service had trouble connecting to resources in the region due to "failures in core networking devices".

As of 1530 Pacific Time on September 1 — 0730 in Tokyo — AWS said it was working to restore the services.

Two hours later, the company had determined that virtual private network traffic was still flowing but could not offer an estimated time for service restoration.

By 2100 Pacific Time — lunchtime in Tokyo — users were advised that signs of partial recovery had become evident.

The incident was closed at 2142 — more than six hours after packet loss was first observed.

AWS's all-clear announcement update expanded a little on the cause of the incident, stating it was caused by "the loss of several core networking devices that are used to connect Direct Connect network traffic to all Availability Zones in the AP-NORTHEAST-1 Region".

Just what AWS means by "loss" was not explained. The Register will assume they broke, rather than disappearing behind a couch.

Whatever happened, "The issue has been resolved and the service is operating normally."

The incident again demonstrated that when AWS goes down, so do its customers; Japanese media report disruptions to banks, payment systems, share traders, and others beside.

Users were, of course, not very happy. But in this instance some found an additional reason for anger... ®

Similar topics


Other stories you might like

Biting the hand that feeds IT © 1998–2021