Currently enmasse does not provide a way to configure how message failures are handled, and its default behavior is to retry messages immediately and infinitely. UPS should handle this itself by providing a mechanism for determining the number of retries and time for retrying a message.

The environment variables to be set are :
AMQ_MAX_RETRIES : how many times to retry a message (ex max retries of 3 will process a message up to three additional times after a failure)
AMQ_BACKOFF_SECONDS : how many seconds to wait until a message is retried. This will increase by an equal amount for the number of retries. (ex a value of 30 will retry at 30, 60, 90... seconds increasing until the message processes or AMQ_MAX_RETRIES is reached)