Automattic's Fork

README.md

@@ -1,30 +1,42 @@
-# Elasticsearch statsd plugin
+# Elasticsearch StatsD Plugin
 
-This plugin creates a little push service, which regularly updates a statsd host with indices stats and nodes stats. In case you are running a cluster, these datas are always only pushed from the master node.
+This plugin creates a little push service, which regularly updates a StatsD host with indices stats and nodes stats.
+Index stats that apply across the entire cluster is only pushed from the elected master which node level stats are pushed from every node.
 
-The data sent to the statsd server tries to be roughly equivalent to [Indices Stats API](http://www.elasticsearch.org/guide/reference/api/admin-indices-stats.html) and [Nodes Stats Api](http://www.elasticsearch.org/guide/reference/api/admin-cluster-nodes-stats.html)
+The data sent to the StatsD server tries to be roughly equivalent to the [Indices Stats API](http://www.elasticsearch.org/guide/reference/api/admin-indices-stats.html) and [Nodes Stats Api](http://www.elasticsearch.org/guide/reference/api/admin-cluster-nodes-stats.html).
 
 
 ## Installation
 
-As plugins (except site plugins) cannot be automatically installed from github currently you need to build the plugin yourself (takes half a minute including an integrations test).
+To install a prepackaged plugin use the following command:
 
 ```
-git clone http://github.com/swoop-inc/elasticsearch-statsd-plugin.git
+bin/plugin -install statsd -url https://github.com/Automattic/elasticsearch-statsd-plugin/releases/download/v0.3.3/elasticsearch-statsd-0.3.3.zip
+```
+
+You can also build your own by doing the following:
+
+```
+git clone http://github.com/Automattic/elasticsearch-statsd-plugin.git
 cd elasticsearch-statsd-plugin
 mvn package
-/path/to/elasticsearch/bin/plugin -install statsd -url file:///absolute/path/to/current/dir/target/releases/elasticsearch-statsd-0.2-SNAPSHOT.zip
+bin/plugin -install statsd -url file:///absolute/path/to/current/dir/target/releases/elasticsearch-statsd-0.3.3.zip
 ```
 
 
 ## Configuration
 
-Configuration is possible via three parameters:
+Configuration is possible via these parameters:
 
 * `metrics.statsd.host`: The statsd host to connect to (default: none)
 * `metrics.statsd.port`: The port to connect to (default: 8125)
 * `metrics.statsd.every`: The interval to push data (default: 1m)
 * `metrics.statsd.prefix`: The metric prefix that's sent with metric names (default: elasticsearch.your_cluster_name)
+* `metrics.statsd.node_name`: Override the name for node used in the stat keys (default: the ES node name)
+* `metrics.statsd.report.node_indices`: If per node index sums should be reported (default: false)
+* `metrics.statsd.report.indices`: If index level sums should be reported (default: true)
+* `metrics.statsd.report.shards`: If shard level stats should be reported (default: false)
+* `metrics.statsd.report.fs_details`: If nodes should break down the FS by device instead of total disk (default: false)
 
 Check your elasticsearch log file for a line like this after adding the configuration parameters below to the configuration file
 
@@ -33,19 +45,30 @@ Check your elasticsearch log file for a line like this after adding the configur
 ```
 
 
+## Stats Key Formats
+
+This plugin reports both node level and cluster level stats, the StatsD keys will be in the formats:
+
+* `{PREFIX}.node.{NODE_NAME}.{STAT_KEY}`: Node level stats (CPU / JVM / etc.)
+* `{PREFIX}.node.{NODE_NAME}.indices.{STAT_KEY}`: Index stats summed across the node (off by default)
+* `{PREFIX}.indices.{STAT_KEY}`: Index stats summed across the entire cluster
+* `{PREFIX}.index.{INDEX_NAME}.total.{STAT_KEY}`: Index stats summed per index across all shards
+* `{PREFIX}.index.{INDEX_NAME}.{SHARD_ID}.{STAT_KEY}` -- Index stats per shard (off by default)
+
+
 ## Bugs/TODO
 
-* No really nice cluster support yet (needed it for a single instance system)
 * Not extensively tested
 * In case of a master node failover, counts are starting from 0 again (in case you are wondering about spikes)
 
 
 ## Credits
 
-Heavily inspired by the excellent [metrics library](http://metrics.codehale.com) by Code Hale and its [GraphiteReporter add-on](http://metrics.codahale.com/manual/graphite/).
+This is a fork of the [Swoop plugin](https://github.com/swoop-inc/elasticsearch-statsd-plugin) for multi-node clusters on ES 1.x.
+
+Heavily inspired by the excellent [metrics library](http://metrics.codahale.com) by Coda Hale and its [GraphiteReporter add-on](http://metrics.codahale.com/manual/graphite/).
 
 
 ## License
 
 See LICENSE
-

pom.xml

@@ -6,21 +6,22 @@
 	<modelVersion>4.0.0</modelVersion>
 	<groupId>de.spinscale.elasticsearch</groupId>
 	<artifactId>elasticsearch-statsd</artifactId>
-	<version>0.2-SNAPSHOT</version>
+	<version>0.3.3</version>
 	<packaging>jar</packaging>
-	<description>Statsd monitoring plugin for Elasticsearch</description>
-	<url>https://github.com/swoop-inc/elasticsearch-statsd-plugin/</url>
+	<description>StatsD monitoring plugin for Elasticsearch</description>
+	<url>https://github.com/Automattic/elasticsearch-statsd-plugin/</url>
 
 	<properties>
-		<elasticsearch.version>0.90.3</elasticsearch.version>
+		<elasticsearch.version>1.0.1</elasticsearch.version>
 		<hamcrest.version>1.3</hamcrest.version>
+		<project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
 	</properties>
 
 	<dependencies>
 		<dependency>
 			<groupId>com.timgroup</groupId>
 			<artifactId>java-statsd-client</artifactId>
-			<version>2.0.0</version>
+			<version>3.0.2</version>
 		</dependency>
 
 		<dependency>

src/main/java/org/elasticsearch/plugin/statsd/StatsdPlugin.java

@@ -17,7 +17,7 @@ public String name()
 
 	public String description()
 	{
-		return "Statsd Monitoring Plugin";
+		return "StatsD Monitoring Plugin";
 	}
 
 	@SuppressWarnings("rawtypes")