Monitoring the adapter

Prerequisite: See Gateway Application metrics for a generic overview on support of metrics in a Gateway Application

The adapter supports exposing metrics which can be used to monitor the adapter. This is disabled by default and may be enabled in the configuration file under the global section as follows.

"global": {
    "framework": {
        "metrics": {
            "enabled": true
        }
    }
}

The metrics for the adapter are exposed as:

  1. JMX

  2. Prometheus metrics

The following types of metrics are exposed by the adapter:

  1. JVM and system metrics

  2. Metrics exposed by the Gateway Framework

  3. Adapter metrics specific to source service which are:

    • adapter_kafka_record_received_total: Total Kafka records received for each topic partition for a service

    • adapter_kafka_poll_error_total: Total number of errors that occurred when trying to poll the Kafka cluster for a service

  4. Kafka consumer metrics

  5. Kafka producer metrics