21 août 2013

This Week in Elasticsearch - August 21, 2013

Par Alexander Reelsen

Welcome to This week in elasticsearch. In this format we try to inform you about the latest and greatest changes in elasticsearch. We try to cover what happened in the github repositories, as well as all the events happening about elasticsearch and give you a small peek in the future.

Elasticsearch core

  • The new percolator now supports scoring (#3506, master only), and got a couple of other improvements
  • The Put Mapping API now does not return too early in some concurrent cases (#3507, master and 0.90)
  • The Put Mapping now validates the version of the cluster state to prevent similar premature acknowledgements like above (#3508, master)
  • Phrase queries generated by query_string queries now support boosting in boolean and dis_max mode (#3503, master and 0.90)
  • Backwards compatibility for handling NGram/EdgeNGram analysis has been improved (#3489, master and 0.90)
  • A search context is now released immediately on errors instead of waiting for the specified timeout (#3513, master and 0.90)
  • Machines with lots of cores do not need to have their thread pools bounded independently from their number of CPUs to prevent uncessary pressure on the system. (#3478 and #3545, master and 0.90)
  • An rare NPE on startup has been prevented by accessing local data instead of a maybe not yet initialized cluster sate (#3515, master and 0.90)
  • The log level can now be configured using a system property (commit, master and 0.90)
  • Suggest execution now immediately returns if there are no documents in a shard (#3473, master and 0.90)
  • Phrase suggest does not fail on empty fields (#3469, master and 0.90)
  • A match all query object was configured static and could lead to reuse with a previously boost (#3521, master and 0.90)
  • Exception handling in actions when forking to a threadpool has been improved (#3524, master and 0.90)
  • Random sort order has been implemented via the function_score feature (#1170, master and 0.90)
  • The blocking thread pool type option has been removed, as it might block the I/O thread (#3531, master and 0.90)
  • Getting a mapping on an index could fail with an IndexNotFoundException when the index did not have a mapping (#3534, master)
  • Dynamic templates from index template definition could be skipped if the new type already contained dynamic templates (#3538, master and 0.90)
  • Added support for counting in the cat API (commit, master only)
  • The readable_format flag has been renamed to human (#3541, master and 0.90)
  • Range queries within a query_string are only lowercased if the field is not numeric to ensure working date math (#3540, master and 0.90)
  • An early termination has been added to the FastVectorHighlighter in order to prevent very long running phrase queries (#3543, master and 0.90)
  • A better default setting (based on the number of processors) for the concurrent active threads for indexing has been configured (#3546, master and 0.90)
  • Statistics now can include information about the memory needed for the completion suggester (#3522, master and 0.90)
  • Completion suggester payloads can now be arbitrary JSON instead of an object in order to save space (#3550, master and 0.90)

Elasticsearch ecosystem

We try to give you some more information about what is happening in the ecosystem we are maintaining around elasticsearch, be it plugin or driver releases or news about kibana.

Elasticsearch community

Got an interesting open source project, plugin, driver or anything else for elasticsearch? Here is your time to shine! Just drop us a note and we will list it here (and on the .org website, of course!).


Also, if you are interested in a core elasticsearch training, the next locations are Toronto at the 29th of August, and London and New York at the 9th of September each. For more locations, check the training page

If you are interested in all this, we are hiring. We are interested in your skills, not in your location. Just drop us a note.