I am having a setup where I run a few ECS services in AWS. These services are spring boot. I have added the Datadog Java Agent to these services. These agents are sending metrics to another ECS service, which is running the Datadog Agent as a container.
My question is, is there any mechanism in the Datadog Java Agent, that retries sending metrics, when it is unable to connect to the Datadog Agent? If there is, is it configurable? If so, how do I configure this?
The approach I thought of was that the collected metrics are persisted locally and pushing is retried under something like an exponential backoff policy. I went through several documents in the official documentation, but I could not find anything related to this.