Package org.apache.kafka.common

Class MetricName

java.lang.Object
org.apache.kafka.common.MetricName
public final class MetricName extends Object
The MetricName class encapsulates a metric's name, logical group and its related attributes. It should be constructed using metrics.MetricName(...).

This class captures the following parameters: 

  name The name of the metric
  group logical group name of the metrics to which this metric belongs.
  description A human-readable description to include in the metric. This is optional.
  tags additional key/value attributes of the metric. This is optional.
group, tags parameters can be used to create unique metric names while reporting in JMX or any custom reporting.

Ex: standard JMX MBean can be constructed like domainName:type=group,key1=val1,key2=val2

Usage looks something like this: 


 // set up metrics:

 Map<String, String> metricTags = new LinkedHashMap<String, String>();
 metricTags.put("client-id", "producer-1");
 metricTags.put("topic", "topic");

 MetricConfig metricConfig = new MetricConfig().tags(metricTags);
 Metrics metrics = new Metrics(metricConfig); // this is the global repository of metrics and sensors

 Sensor sensor = metrics.sensor("message-sizes");

 MetricName metricName = metrics.metricName("message-size-avg", "producer-metrics", "average message size");
 sensor.add(metricName, new Avg());

 metricName = metrics.metricName("message-size-max", "producer-metrics");
 sensor.add(metricName, new Max());

 metricName = metrics.metricName("message-size-min", "producer-metrics", "message minimum size", "client-id", "my-client", "topic", "my-topic");
 sensor.add(metricName, new Min());

 // as messages are sent we record the sizes
 sensor.record(messageSize);