Novell Open Enterprise Server Administrators Handbook SUSE Linux Edition [Electronic resources] نسخه متنی

Monitoring SLES with Health Monitoring Services

An important aspect of managing SLES is simply being aware of how your server is performing. OES Linux reduces the complexity of this task through leveraging the capabilities of Web-Based Enterprise Management (WBEM). WBEM is a standard, Internet-based technology designed to consolidate the various tasks of enterprise server management. Through WBEM, Health Monitoring Services (HMS) can be used to easily monitor many important server health-related factorsincluding CPU utilization, system memory, running processes, and network utilization. OES Linux uses the Common Information Model Object Manager (CIMOM) daemon, from the OpenWBEM project, to manage the health monitoring environment. The CIMOM daemon performs this function by accessing one or more Providers that conform to WBEM standards. These Providers facilitate the gathering of SLES and OES statistics used by HMS. For administrative use, this information is then gathered and consolidated for display in iManagerso you can easily keep tabs on your server health while performing day-to-day administrative routines. One of the great things about HMS is its potential for future capabilities. This is the first release of HMS and is geared primarily at monitoring essential health-related information. In future releases, system alerts and robust analysis should further enhance the capabilities of HMS and make this one of the most important server monitoring tools in your arsenal.

Installing HMS

Health Monitoring Services can be installed as an optional component during the OES Linux installation. HMS can also be installed later through YaST.

NOTE

Basic server health is available for any server throughout your organization, but for complete health monitoring, Health Monitoring Services must be installed on all your OES Linux servers.

To install Health Monitoring Services using YaST, complete the following steps:

After installing HMS, the OpenWBEM CIMOM daemon can be manually started, stopped, or restarted using the main daemon startup script /etc/init.d/owcimomd. To start the CIMOM daemon, execute the following command:

# /etc/init.d/owcimomd start

Configuring HMS

Configuring and accessing HMS is performed through iManager. All HMS- related capabilities can be found under the Monitor Servers link, in the Servers category of the left navigation frame. The first time this page is accessed, you must select a container where your HMS configuration will be stored. Although not required, it is normally a good idea to select the container in which your OES Linux server exists for this purpose. After the HMS objects have been created in eDirectory, you are then able to configure a list of servers to monitor.

The first server you should add to your server list is your OES Linux server. To add your server to this list, select Add, and then provide the required information (server name and IP address or DNS name). You are also given a choice of the type of monitoring you would like to perform. The two options you have are as follows:

Robust Health
If the server you are monitoring has OpenWBEM installed, Robust Health will provide the monitoring capabilities for the following categories:

Memory Operating System Process/Threads Network CPU Simple Server Status
If OpenWBEM is not available on the specified server, this option can be used to provide a simple UP/DOWN status of the server.

After a server has been added to your server watch list, the Monitor Servers page will be refreshed with an overview of the server's health, as shown in Figure 6.8.

To display the Quick Status of a server's health, click on the server icon in the server watch list. For detailed information on server statistics, click the server name itself. You can then drill down on each monitored category further for complete diagnostic of your server.

NOTE

For additional information on Health Monitoring Services, see the online OES documentation.