Product release

Release Bonanza! Elasticsearch, Graph, Shield, Watcher, Marvel, Logstash 2.3, Beats 1.2, and Kibana 4.5 are Now Available!

The Release Bonanza returns!

Today, we are pleased to release Elasticsearch 2.3, Kibana 4.5, Beats 1.2, and updated versions of all the commercial products.

But wait, there's more.

Graph. Several thousands of you saw the presentation at Elastic{ON}. Several thousands more tuned into a recent webinar. Now, it is here.

That's right, Graph is a GA product available for download. And, it just wouldn't be a Release Bonanza if we didn't have all the products available, now, on Elastic Cloud.


Existing products get bullets. New products get a paraGraph. (See what we did there?)

When you store data in Elasticsearch, this data often contains references or properties that represent connections between objects, entities, people, machines, etc. We built the Graph product to allow you to ask a whole new type of question — focusing on these relationships.

Graph - Last FM - United Kingdom

The best way to explore these connections is to see them, which Graph provides via a Kibana plugin. Like everything at Elastic, this UI is built on a simple, but powerful API, which leverages the same statistics that power search relevance (like bm25) to bring relevance to the relationship exploration process.

Whether discovering and visualizing relationships through the Graph UI or directly integrating with the Graph API (recommendations anyone?), we are excited to see how you use Graph!

Want to learn more? Check out the Webinar, to see it in action and learn more.


For detailed information, check out the Elasticsearch 2.3 release post.

  • Reindex API - The ability to reindex your data is essential: whether you need to improve field mappings,take advantage of new features, or even go crazy like changing a document's _id or index. Say “heya” to the Reindex API.
  • Update-by-Query API - The Update-by-Query operates as a lesser Reindex API: batch updating all the documents that match a query in place.
  • Task Management API - Time is precious...compute time doubly so. The new Task Management API reports on all tasks being executed across the cluster. For ongoing reindex jobs you can see how far along they are and even cancel them.
  • Deprecation Logging - Watch out for deprecated features! Future-proof your app and get the best possible performance by enabling deprecation logging and then switching away from offending entries in the logs.


For detailed information, take a look at the Kibana 4.5 release post.

  • Metric Labels - The ability to customize metric and axis labels makes data presented in Kibana visualizations self-referential and easier to interpret. Now you can keep the most accurate technical metric names in Elasticsearch, but completely customize how they are presented in Kibana visualizations!
  • Auto-closing Errors - Have you ever needed to hunt down a mouse to clear the errors on the long running Kibana dashboard in your office? Now errors will automatically dismiss after 5 minutes.
  • Other usability tweaks, including better presentation of empty sets in visualizations and customizable pagination in the saved object finder.
  • Logstash

    For detailed information, peruse the Logstash 2.3 release post.

    • Dynamic Config Reload - Things change. And changing things, especially configuration files, shouldn't require a restart (especially in a multi-instance deployment)
    • Java Event Rewrite - Logstash 2.3 is different. We have completely rewritten the Event representation in pure Java. All the performance increases.


    For detailed information, give the Beats 1.2 release a gander.

    • Configuration Override - Use environment variables to override configuration settings.
    • Configure Mappings on Startup - Gives you the option to automatically load Elasticsearch mappings when a Beat is starting.
    • Export the process command line in Topbeat - Makes it easier to differentiate between processes having the same name.


    This release introduces quite a few bug fixes, enhancements and new features. To name a few:

    • If you're using the LDAP or Active Directory realms, you'll now be able to define multiple LDAP/AD URLs along with load balancing strategy.
    • We also simplified the cluster/index-level privileges to make it easier define access control and role definitions.

    But by far the most exciting feature we've added is the new native realm, which includes role management APIs.

    Marvel & Watcher

    Marvel now manages its own history - by default, it maintains 7 days of performance history. Both Marvel and Watcher have been updated for compatibility with Elasticsearch 2.3 and Kibana 4.5.