Approximately 0.07 per cent of the EBS storage volumes in the East Region of Amazon's infrastructure cloud are not "fully recoverable" following the outage that hit the service last Thursday.
Amazon have so far been criticised for their lack of information regarding the outage but it plans to produce a "post mortem" of the outage. "We are digging deeply into the root causes of this event," the company says on its Amazon Web Services status dashboard.
The outage first hit in the early hours of Thursday, with the status page saying that it was investigating connectivity issues with its EC2 service. The outage resulted in hundreds, if not thousands, of sites being offline with notable names such as FourSquare, Quora and Hootsuite affected. According to one status message from Amazon, the issue started with a "network effect" that resulted in the service re-mirroring a large number of Elastic Block Store volumes in its East Region.
The Amazon cloud is divided into multiple geographic regions and offers a guarantee of 99.95% availability within each region. The East Region, along with some others, are divided into multiple "availability zones". Amazon has always maintained that these zones are "insulated" from each other's outages. However, this recent outage spread across multiple zones.
On Monday, Amazon announced that some affected EBS volumes would not be restored. "We have completed our remaining recovery efforts and thought we've recovered nearly all of the stuck volumes, we've determined that a small number of volumes (0.07% of the volumes in our US-East Region) will not be fully recoverable."
Return to hosting news headlines
View Hosting News Archive