Comments on: Hadoop Graphing with Cacti http://www.cloudera.com/blog/2009/07/hadoop-graphing-with-cacti/ Hadoop and Cloudera's Products and Services Wed, 23 May 2012 10:20:43 +0000 hourly 1 http://wordpress.org/?v=3.3.2 By: Jilles http://www.cloudera.com/blog/2009/07/hadoop-graphing-with-cacti/comment-page-1/#comment-15147 Jilles Fri, 14 Jan 2011 13:45:09 +0000 http://www.cloudera.com/blog/?p=894#comment-15147 The live demo doesn't work anymore: FATAL: Cannot connect to MySQL server on 'localhost'. Please make sure you have specified a valid MySQL database name in 'include/config.php' The live demo doesn’t work anymore:

FATAL: Cannot connect to MySQL server on ‘localhost’. Please make sure you have specified a valid MySQL database name in ‘include/config.php’

]]> By: Jon Stevens http://www.cloudera.com/blog/2009/07/hadoop-graphing-with-cacti/comment-page-1/#comment-14878 Jon Stevens Sat, 25 Dec 2010 06:16:29 +0000 http://www.cloudera.com/blog/?p=894#comment-14878 I've created a project called jmxtrans. This is effectively the missing connector between JMX and whatever logging / graphing package that you can dream up. jmxtrans is very powerful tool which reads json configuration files of servers/ports and jmx domains / attributes / types and then outputs the data in whatever format you want via special 'Writer' objects which you can code up yourself. It does this with a very efficient engine design that will scale to querying thousands of machines. The core engine is pretty solid and there are writers for cacti/rrdtool, graphite and stdout. This is a far more complete solution for creating a visual representation of what your hadoop cluster is doing. I’ve created a project called jmxtrans. This is effectively the missing connector between JMX and whatever logging / graphing package that you can dream up.

jmxtrans is very powerful tool which reads json configuration files of servers/ports and jmx domains / attributes / types and then outputs the data in whatever format you want via special ‘Writer’ objects which you can code up yourself. It does this with a very efficient engine design that will scale to querying thousands of machines.

The core engine is pretty solid and there are writers for cacti/rrdtool, graphite and stdout.

This is a far more complete solution for creating a visual representation of what your hadoop cluster is doing.

]]> By: Alexey Kovyrin http://www.cloudera.com/blog/2009/07/hadoop-graphing-with-cacti/comment-page-1/#comment-8403 Alexey Kovyrin Thu, 17 Dec 2009 19:53:05 +0000 http://www.cloudera.com/blog/?p=894#comment-8403 In your repo DataNode cacti template for 0.20 contains hbase rs tamplate, but not the datanode one. Could you please replace it with the actual datanode template? Thanks. In your repo DataNode cacti template for 0.20 contains hbase rs tamplate, but not the datanode one. Could you please replace it with the actual datanode template?

Thanks.

]]>