Elasticsearch for Apache Hadoop 2.1.0.Beta2

Release Notes

New features:

  • Integration with Apache Storm #267
  • Return metadata (not just source) to clients #275

Enhancements:

  • Allow LinkedMapWritable to include document meta fields like _version #261
  • Issue with inserting/parsing timestamps from hive to elasticsearch. #259
  • Add explicit example plus API for RDD operations #246
  • EsSpout : define max number of replays #285

Bug fixes:

  • Missing commons-cli dependency? #288
  • Hadoop's uber mode causes job failure #280
  • Rich JSON objects cannot be used as script parameters #277
  • Issue with named queries #273
  • ArrayIndexOutOfBoundsException on Spark on 'pining' tasks #272
  • date fields with long values are not properly returned #271
  • Escape fields to prevent incorrect URIs #264
  • Fix manifest in snapshot builds #262
  • Tasks are not correctly pinned to their target shards #258
  • Keep original nodes when adding discoveredHosts in SettingsUtils #256
  • Docs say version is 2.1.0.M1, but is really 2.1.0.Beta1 #247
  • Is it possible to set the "_id" of a document being written in this API? #237
  • elasticsearch-hadoop support multi-hosts in Map/Reduce job #211
  • Could not resolve org.elasticsearch.hadoop.pig.ESStorage #291
  • Allow arbitrary RDDs to have a header describing their content #281
  • not compatible for hadoop 0.20.0 #270
  • Issue while joining two tables stored on Elasticsearch using HiveQL #266
  • Hive and Elasticsearch work together #257
  • Upgrade Gradle to 2.0 #252
  • Hadoop Elasticsearch JDBC connector question - potential issue? #250
  • ES-Hadoop with SSL proxy #249
  • writing data from HIVE to ES #245
  • add support for SchemaRDD/Spark SQL #244
  • java.lang.IllegalAccessError: tried to access method org.apache.spark.TaskContext.interrupted()Z from class org.elasticsearch.spark.rdd.AbstractEsRDDIterator #292

Docs:

  • Got ClassNotFoundException[org.elasticsearch.repositories.hdfs.HdfsRepositoryModule] error when using repository-hdfs plugin #290
  • .set is a method of SparkConf not a SparkContext #279
  • Document setting ID per each module #263
  • Improvement on documentation - setting ids in spark #254
  • link to Cascading 2.5 javadoc #248