Amazon targets cascading fails with EBS monitoring
Amazon has added a feature to its cloud that automatically checks the health of elastic block store (EBS) volumes and takes action to prevent data inconsistency.
The EBS Volume Status feature was introduced by Amazon on Tuesday. The technology scans EBS volumes and, if their status is impaired, disables I/O on the volume to stop data inconsistency affecting applications that depend on data stored there.
The company also gave developers API-driven tools to let them manually re-enable I/O if they favoured "immediate volume availability over [data] consistency", Amazon's senior manager for cloud computing, Jeff Barr, wrote in a blog post.
Problems with EBS volumes have led to major Amazon Web Services outages over the past couple of years, with an increase in EBS latencies hurting sites such as Reddit, Heroku and Quora in March 2011 and again in April.
At the time of the March outage a former Reddit programmer said, "Amazon's EBS's are a barrel of laughs in terms of performance and reliability and are a constant (and the single largest) source of failure across Reddit."
The results of EBS status checks are available in the EBS volume list on the AWS console.