Platform operations are running as before the incident. Container Metrics on the Aptible Dashboard remain unavailable, and will likely remain unavailable until the underlying issue affecting AWS Lambda is resolved.
Posted Feb 28, 2017 - 17:23 EST
Update
Performance and error rates across the multiple AWS Aptible deployments rely on appears to finally have stabilized (although we're still 2x away from baseline performance). We are un-blocking operations in us-east-1.
Posted Feb 28, 2017 - 17:05 EST
Update
While S3 appears to have recovered, dependent services — specifically ELB — are still operating very inefficiently, so we're still holding off on re-enabling deploys.
Posted Feb 28, 2017 - 16:52 EST
Monitoring
The situation is stabilizing, but we're still seeing numerous errors connecting to AWS APIs. According to AWS we can expect a full resolution shortly.
App operations are still on lockdown for us-east-1 stacks now, but we'll unlock them as soon as we get confirmation that the incident has been resolved on the AWS side.
Posted Feb 28, 2017 - 16:08 EST
Update
We're told by AWS that the issue has been identified but may take over an hour to resolve.
As indicated earlier, running apps are not affected by this issue unless you rely on S3 yourself, but deployments and SSH sessions are temporarily unavailable for stacks hosted in us-east-1.
We have temporarily disabled app operations across Aptible until the issue with S3 is resolved (these would inevitably fail while S3 is down).
*Note: this outage only affects customer stacks deployed in us-east-1*
Posted Feb 28, 2017 - 13:21 EST
Identified
AWS is reporting a serious availability incident affecting S3 in us-east-1. As a result, Aptible features that rely on S3 (either indirectly or directly via other AWS services that rely on S3) may be unavailable or malfunction.
At this time; deployments and SSH sessions on Aptible in us-east-1 are unlikely to succeed. The Dashboard is also unavailable.
*Running apps are not affected unless your app itself relies on S3.*