Book Image

Apache Solr 4 Cookbook

By : Rafał Kuć
Book Image

Apache Solr 4 Cookbook

By: Rafał Kuć

Overview of this book

<p>Apache Solr is a blazing fast, scalable, open source Enterprise search server built upon Apache Lucene. Solr is wildly popular because it supports complex search criteria, faceting, result highlighting, query-completion, query spell-checking, and relevancy tuning, amongst other numerous features.<br /><br />"Apache Solr 4 Cookbook" will show you how to get the most out of your search engine. Full of practical recipes and examples, this book will show you how to set up Apache Solr, tune and benchmark performance as well as index and analyze your data to provide better, more precise, and useful search data.<br /><br />"Apache Solr 4 Cookbook" will make your search better, more accurate and faster with practical recipes on essential topics such as SolrCloud, querying data, search faceting, text and data analysis, and cache configuration.<br /><br />With numerous practical chapters centered on important Solr techniques and methods, Apache Solr 4 Cookbook is an essential resource for developers who wish to take their knowledge and skills further. Thoroughly updated and improved, this Cookbook also covers the changes in Apache Solr 4 including the awesome capabilities of SolrCloud.</p>
Table of Contents (18 chapters)
Apache Solr 4 Cookbook
Credits
About the Author
Acknowledgement
About the Reviewers
www.PacktPub.com
Preface
Index

How to deal with too many opened files


Sometimes you might encounter a strange error, something that lies on the edge between Lucene and the operating system—the "too many files opened" exception. Is there something we can do about it? Yes, we can, and this recipe will show you how.

How to do it...

The following steps show how to deal with too many opened files:

  1. So, for the purpose of the recipe let's assume that the header of the exception thrown by Solr looks like this:

    java.io.FileNotFoundException: /use/share/solr/data/index/_1.fdx (Too many open files)
  2. What can you do instead of pulling your hair out? First of all, this probably occurred on a Unix-/Linux-based operating system. So, let's start with setting the opened files' limit higher. To do that, you need to edit the /etc/security/limits.conf file of your operating system and set the following values (I assume Solr is running as solr user):

    solr soft nofile 32000
    solr hard nofile 32000
  3. Now let's add the following line to the .bash_profile...