Copyright © Cloud Software Group, Inc. All Rights Reserved

Deploy Using Monitoring And Management Server

The Monitoring and Management Server component allows you to deploy the cache-based Core Engines, and then monitor and manage a deployed cluster. It works with the following sub-components:

•	Site topology editor

The site topology editor is a canvas-based editor, which allows users to set the deployment configuration of the cluster by specifying its hosts and deployment mappings. See Working With Site Topology Editor.

•	MM Console

MM Console is a web-based dashboard that enables you to monitor the deployment and perform various operations on the cluster.

Deployment Pre-Requisites

The Monitoring and Management Server requires a site topology file, which contains all the deployment time information. See Site Topology Overview for details on site topology file.

TIBCO API Exchange Gateway packages a sample site topology file to deploy the gateway caching enabled Core Engine and Global Throttle Manager on a single host.

Sample Site topology file

You can find a sample of the site topology file shipped with the product installation. This file is located as:

ASG_HOME/mm/bin/asg_default.st

Configuration Settings Before You Deploy

This section explains the configuration settings required for the Monitoring and Management server.

Configure JMX properties in Core Engine TRA File

After the Core Engines are started, and join the cluster, they use JMX MBeans to expose monitoring and management information to the MM server, and to allow remote method invocation. The JMX port number must be specified before the engine’s JVM starts. A variable for the port number is provided in the TRA file so that the actual value can be specified before the engine starts.

To Configure JMX Properties

JMX properties are provided in the shipped ASG_HOME/bin/asg-engine.tra file but are commented:

---

 

#java.property.be.engine.jmx.connector.port=%jmx_port%

#java.property.be.engine.jmx.connector.authenticate=false

---

 

1.	Open the following file for editing:

ASG_HOME/bin/asg-engine.tra

2.	Uncomment the following properties:

   java.property.be.engine.jmx.connector.port=%jmx_port%

   java.property.be.engine.jmx.connector.authenticate=false

To Enable Monitoring and Management

To expose JMX for monitoring and management (without authentication), uncomment this property:

java.property.be.engine.jmx.connector.port=%jmx_port%

Ensure that the value of the port property is set to this literal value: %jmx_port%. The actual value is substituted at runtime. When you use the MM console to start the gateway engines remotely, MM reads the port number from the PU configuration settings in the site topology file.

When more than one PU (engine) is deployed to the same host, ensure that a different JMX port is used for each of the PUs, in the site topology file.

To Enable JMX MBeans Authentication

The following property enables authentication:

java.property.be.engine.jmx.connector.authenticate=true

 

Edit MM.cdd File

The CDD file used by the MM server is the MM.cdd. In this file, you must specify the path to the site topology file. The MM.cdd file is a XML file and can be edited using any XML or text editor.

Before you make any changes to the MM.cdd file, make sure to make a back up of this file.

1.	Navigate to the ASG_HOME/mm/bin directory.

2.	Edit the MM.cdd file and change the value of be.mm.topology.file property, if required. This property must be set to the fully qualified path of the site topology file for the cluster to be monitored. Specify the location of the site topology file you want to be loaded by the MM server.

By default, the property is set to the sample of a site topology file shown as follows:

<property name="be.mm.topology.file" value="ASG_HOME/mm/bin/asg_default.st"/>

The sample site topology file is located as:

ASG_HOME/mm/bin/asg_default.st

Whenever you make changes to the MM.cdd file you must restart the MM server so that it uses the updated values.

Edit Site Topology File in a Text Editor

The site topology file contains deploy time information such as what processing units to deploy to specific computer/hosts in your environment. You need to know information about the computers that will host the agents you intend to deploy, for example the information about the machines’ operating system and IP address.

An annotated site topology file that can be used as template is located at:

ASG_HOME/mm/bin/asg_default.st

You only need to edit the topology file if the defaults don't serve your purposes. This happens typically when the Core Engines are to be deployed on a different host than the one where the Monitoring and Management server is running.

For example, edit the following settings in the ASG_HOME/mm/bin/asg_default.st file for your deployment (if the default values are not correct):

<host-resources>

<host-resource id="HR_0">

<hostname>127.0.0.1</hostname>

<ip>127.0.0.1</ip>

<user-credentials password="asg" username="asg"/>

<os-type>Windows</os-type>

See Site Topology Reference for details on the XML elements defined in the topology file.

Make sure that you take a backup copy of the topology file before you make any changes.

Install and Configure Software for Remote Start and Deployment

The following table shows which software utilities can be used to perform remote operations. Information about installation and configuration of each software utility follows. Use the software utility’s documentation for more details.

Table 152 Software Options for Deployment, Remote Start, and Remote Method Invocation

Software	Deployment	Remote Start, Stop	Remote Method Invocation
SSH	Yes	Yes
PsTools (Windows)		Yes
TIBCO Hawk®		Yes	Yes
JMX (required)			Yes

Although it is possible to use more than one utility for the machines in the cluster, it is recommended that you use only one across all the machines. Ensure that the software is installed and running on all relevant machines.

The software you use on each machine in the monitored cluster is specified in the cluster’s site topology file. See Start PU Method Setting.

SSH

Only SSH software enables MM to deploy Core Engine to the predefined hosts, that is, those configured in the monitored cluster’s site topology file.SSH can also be used to start remote engines.

The SSH utility is available on UNIX machines by default. No action is required. On Windows machines you must install an SSH server, as explained next.

Installing and Configuring an SSH Server

If you want to use SSH on Windows machines, you must download the software and install it. Many SSH servers are available. For Windows, OpenSSH and Copssh are supported.

Configuring OpenSSH TIBCO has tested with OpenSSH software. See TIBCO API Exchange Gateway readme file for specific versions that are supported. If you use the OpenSSH server, note the following when installing OpenSSH: • The OpenSSH package is not a part of the default Cygwin installation. During its installation, ensure that you select the OpenSSH package. Also select the option "Select required packages (RECOMMENDED)" to install all the required packages to satisfy the dependencies. See openSSH server documentation for details. • Accept the default username suggested when configuring the OpenSSH server and provide a password for the username. • For deployment and starting PUs, use an username that is ssh enabled to login to the remote host. For example, ssh username@hostname. The credentials of the username can be specified in the host settings of the site topology file, Host Settings User and Password fields. See Host Settings. However, if you choose to use a different username, you must ensure that the username is added to the SSH server.

Configure User Authorization for Administrator and User Roles

MM authorization uses two preconfigured roles. These roles are specified in the provided passwords file that is used for file-based authentication:

ASG_HOME/mm/config/users.pwd

The file as shipped contains the following entries:

---

 

jdoe:A31405D272B94E5D12E9A52A665D3BFE:MM_ADMINISTRATOR;

mm_user:11b2016b63c99ef7ab6d6d716be7b78e:MM_USER;

admin:21232f297a57a5a743894a0e4a801fc3:MM_ADMINISTRATOR;

---

 

If you add more users ensure that they have the appropriate role. Note that role names are case sensitive:

MM_ADMINISTRATOR   Users with this role can execute methods, for example to deploy, start, and stop engines, and invoke method operations.

MM_USER   Users with this role can view MM Console, but cannot deploy, start, or stop engines, or invoke method operations.

 

Copyright © Cloud Software Group, Inc. All Rights Reserved