Red Hat Training

A Red Hat training course is available for Red Hat JBoss Operations Network

9.2. Installing the Agent from RPM

The JBoss ON agent is installed through the jboss-on-agent package. This package installs all of the agent files, creates a specific JBoss ON agent system user, and configures the JBoss ON agent as a system service.
  1. Configure the yum repos to include the JBoss ON channel (as listed in Table 9.3, “Available Channels for the Agent RPM”).
  2. Use yum to install the package. 
    [root@server ~]# yum install jboss-on-agent
    Alternatively, download packages from http://access.redhat.com and install using rpm. 
    [root@server ~]# rpm -ivh jboss-on-agent-3.1.2.0.GA1.el6.noarch.rpm

    Note

    Installing the RPM requires specific entitlements for the RHN user account for the 3.1.2 release.
    This installs the agent in /usr/share/jboss-on-3.1.2.0.GA1/agent.
  3. Configure the agent by running the service jon-agent config command. This runs through the advanced installer to configure the agent. 
    [jsmith@server ~]$ sudo service jon-agent config
RHQ 4.4.0.JON311GA [6910991] (Wed Aug 01 18:43:03 EDT 2012)
** Advanced Setup **
... 8< ...
Agent Name [agent.example.com] : agent1
Agent Hostname or IP Address [!*] :
Agent Port [16163] :
Agent Transport Protocol [socket] :
Agent Transport Parameters [numAcceptThreads=1&maxPoolSize=303&clientMaxPoolSize=304&socketTimeout=60000&enableTcpNoDelay=true&backlog=200] :
RHQ Server Hostname or IP Address [255.255.255.255] :
RHQ Server Port [7080] :
RHQ Server Transport Protocol [servlet] :
RHQ Server Transport Parameters [/jboss-remoting-servlet-invoker/ServerInvokerServlet] :
RHQ Server Alias [rhqserver] :
The setup has been completed for the preferences at node [/rhq-agent/default].
    The config command runs through all of the advanced setup options, which configures three areas of the agent:
    • The agent connection information, used to register the agent to the server
      • The agent name
      • The agent port
      • The agent host (by hostname or IP address)
    • The agent-server communication settings, which include configuring SSL or secure connections and rules on how frequently the agent communicates with the agent
      • The agent protocol, either socket (regular) or sslsocket (secure)
        For sslsocket, the JBoss ON server needs to be configured to accept SSL connections, as in the SSL chapter of Configuring JBoss ON Servers and Agents.
      • Any client transport parameters to use to connect to the server
        Both the server and the agent use JBoss Remoting for communication. JBoss Remoting allows servers and clients to pass connection settings using a URL-style address. Transport parameters include things like pool sizes, timeout periods, and buffer settings. For the complete list, see the JBoss Remoting client parameters documentation.
      • The server protocol which the agent uses to send messages to the server, either servlet (regular) or sslservlet (secure)
        The server connection settings configured on the agent must match the configuration in the server itself.
      • Any server transport parameters to use to receive messages from the agent
        Both the server and the agent use JBoss Remoting for communication. JBoss Remoting allows servers and clients to pass connection settings using a URL-style address. Transport parameters for the server relate to the servlet used to receive agent messages.
    • The JBoss ON server to register with
      • The server host (by hostname or IP address)
      • The server port
      • The server alias, a short nickname to identify the server instance
  4. Start the agent. 
    [jsmith@server ~]$ sudo service jon-agent start