I am the jmxtrans cookbook!
jmxtrans is an excellent tool for transporting jmx data from your VMs and into a graphing tool like graphite or ganglia. This cookbook only supports writing to graphite but could be easily modified to work with ganglia
Depends on the ark cookbook
node['jmxtrans']['graphite']['host']
- defaults to 'graphite'node['jmxtrans']['graphite']['port']
- default to 2003node['jmxtrans']['servers']
- array of servers to query for jmx data along with the properties needed to access each onenode['jmxtrans']['root_prefix']
- root prefix for the graphite buckets, defaults to "jmx"
You must override the attribute node['jmxtrans']['servers']
with the
list of servers you want monitored and their respective properties
the following example comes from a role
:jmxtrans => {
:servers =>[
{
'name' => 'foo1.example.org',
'port' => "8999",
'type' => 'tomcat',
'username' => 'foobar',
'password' => 'rw'
},
{
'name' => 'foo2.example.org',
'port' => "8999",
'type' => 'tomcat',
'username' => 'foobar',
'password' => 'baz'
}
]
}
Bryan W. Berry, Copyright 2012, Apache 2.0 license