2015年06月5日

The Logstash Lines: How to reindex data in Elasticsearch using Logstash

By Shaunak Kashyap

Welcome back to The Logstash Lines! In these weekly posts, we'll share the latest happenings in the world of Logstash and its ecosystem.

This week, our very own David Pilato shows us how to reindex data in Elasticsearch using Logstash:

And here is an awesome presentation from Logstash engineer João Duarte, on how you can setup a burglar alert system using Logstash!


Logstash Core and Plugins

  • Timeout bug: João and Karel worked together to configure the http request timeout in the Elasticsearch output plugin. In 1.5.0 we switched the client implementation from ruby-ftw to jruby-manticore which, by default, was setting a low timeout. This is not suitable for bulk indexing. We have made the timeout unlimited by default, returning to the old behavior.
  • Continued work on release acceptance tests. We have the first iteration running on Jenkins.
  • Good progress on the proof-of-concept for rewriting the Event class in Java. We now have a shared feature branch and have managed to pass most of the tests for this class. Early throughput performance shows up to 50% improvement in some basic configurations. More information as we dive deeper!
  • Closing issues for the 1.5.1 release.

That's this week in The Logstash Lines. Come back next week for more Logstash news!