Release: Elastic Load Balancing on 2010-04-07
This release adds support for sticky sessions.
||This release adds support for sticky sessions. Sticky sessions bind a user's session to a specific application instance so that all requests coming from the user
during the session will be sent to the same application instance. For more information, see
Sticky Sessions in the Elastic Load Balancing Developer Guide.
|Elastic Load Balancing is a feature of Amazon Elastic Compute Cloud (EC2).
||You must be signed up for Amazon Elastic Compute Cloud (EC2) service in order to call the Elastic Load Balancing APIs.
|The Amazon CloudWatch Latency metric for a LoadBalancer aggregates the latency for all requests to all LoadBalancer ports (Listeners ) into a single metric.
||If you have a LoadBalancer that is handling traffic on multiple ports, you cannot view the Latency of the requests for each port. Moreover, for LoadBalancers with TCP listeners, the Latency metric is not a meaningful measure of the TCP
|When instances are registered with a LoadBalancer, they are assumed to be healthy.
||Regardless of the actual health of an instance, the LoadBalancer will assume a newly registered instance is healthy and put it into the InService state. If the instance is unhealthy, it will be transitioned to the OutOfService state on the
first health check.
|You cannot charge customers for their use of Elastic Load Balancing for instances running an Amazon EC2 paid AMI.
||At this time, you cannot use Elastic Load Balancing with instances running Amazon EC2 paid AMIs.
|SOAP requests over HTTP will result in the following error: Premature end of file. AWSRequestId:No request id received.
||API requests using SOAP are only supported over HTTPS.
|The LoadBalancer does not report zeros for the RequestCount metric in Amazon CloudWatch when there are no requests.
||You will get no datapoints for the RequestCount metric for the period when there are no requests. You cannot set Auto Scaling triggers that rely on zero RequestCount.