The Hadoop plugin allows to use Hadoop as a shared gateway for ElasticSearch.

In order to install the plugin, simply run: bin/plugin -install elasticsearch/elasticsearch-hadoop/1.2.0.

-----------------------------------------------------------------
| Hadoop Plugin    | ElasticSearch    | Embedded Hadoop Version |
-----------------------------------------------------------------
| master           | 0.19 -> master   | 0.20.204.0              |
-----------------------------------------------------------------
| 1.2.0            | 0.19 -> master   | 0.20.204.0              |
-----------------------------------------------------------------
| 1.1.0            | 0.19 -> master   | 0.20.204.0              |
-----------------------------------------------------------------
| 1.0.0            | 0.18             | 0.20.204.0              |
-----------------------------------------------------------------

The hadoop plugin contains embedded version of hadoop jar files to communicate with hadoop. It can be replaced with the actual version of haddop one is using.

The hadoop (HDFS) based gateway stores the cluster meta and indices data in hadoop. Here is an example config to enable it:

gateway:
    type: hdfs
    hdfs:
        uri: hdfs://myhost:8022

The hadoop gateway requires two simple settings. The gateway.hdfs.uri controls the URI to connect to the hadoop cluster, for example: hdfs://myhost:8022. The gateway.hdfs.path controls the path under which the gateway will store the data. The gateway.hdfs.concurrent_streams allow to throttle the number of streams (per node) opened against the shared gateway performing the snapshot operation. It defaults to 5.