Introduction to Synapse Scheduled Message Forwarding Processor with following advance parameters
Execute the sample client a few times with the following command.
When you start to send request to synapse from client, you will see message forwarding processor without getting deactivate it keep on processing. This is due to the message will be dropped from the message store after the maximum number of delivery attempts are made, and the message processor will remain activated. "max.deliver.drop" parameter would have no effect when no value is specified for the Maximum Delivery Attempts parameter. If this parameter is disabled, the undeliverable message will not be dropped and the message processor will be deactivated.
Message Forwarding Processor by default do not retry for application level failures. It only retries by default when there is a network level failure. But if the user wants retry based on the application level failures, user can use "retry.http.status.codes" configuration to do so. Please note that in this context application level failures refers to HTTP error responses. For instance, in the above example Message Forwarding Processor retries not only for transport level failures but also for application level failures such as Internal Server Error (500) and Gateway timeout (504).
Message Forwarding Processor sends messages to the back-end with the interval configured using "interval" parameter. However, when there is a failure, Message Forwarding Processor goes to retry mode. When retrying, Message Forwarding Processor sends the message to the back-end with the interval configured using "retry.interval" parameter. Similar to "interval" "retry.interval" value must be set in milli seconds. In the case of setting a non-integer value, makes Message Forwarding Processor set the default value to "retry.interval", which is 1000ms.
Every time the Message Forwarding Processor is triggered it consumes all the messages in the Message Store at once. For instance, suppose, the Message Forwarding Processor is configured to run every ten seconds and the Message Store is filled with five messages within the ten second gap. In such a situation, Message Forwarding Processor consumes all five messages and try to send it to back-end as fast as possible. However, there could be situations where you need to send messages to the back-end in a controlled rate. This can be achieved by setting the "consume.all" property value to false. When set to false, the Message Forwarding Processor will only consume one message at each trigger.