This document describes how to operate System Fault Management (SysFaultMgmt)
by using the interface provided by HP Systems Insight Manager (SIM).
If you are using a different network management application, refer to the documentation for it instead.
Contents
For the latest and most complete information, see the
"System Fault Management" section on the Diagnostics Homepage at http://docs.hp.com/hpux/diag/index.html.
Overview
SysFaultMgmt is a collection of tools for monitoring the health of HP servers and
getting information about hardware, such as memory and CPU.
These tools operate within the Web-Based Enterprise Management (WBEM) environment.
To view the information from the SysFaultMgmt providers, you use a WBEM-based network management application, such as HP Systems Insight Manager (SIM). See the Block Diagram below.
Block diagram of System Fault Manager and HP Systems Insight Manager (SIM) in the WBEM environment.
This document explains how to configure and use all the components in the
figure to monitor your system. But first, let's briefly go over the pieces
that make up the WBEM environment.
The dashed box in the figure represents the networked computer system that has System Fault Manager loaded.
Each System Fault Manager tool (Provider) operates on a specific resource. For example, the CPU Instance Provider obtains information about a CPU. The EMS-to-WBEM Wrapper Provider translates events from the EMS hardware monitors into a format that is compatible with WBEM.
Also located on the same networked computer system is WBEM Services, a product installed separately. WBEM Services is also known as CIMOM (The Common Interface Model Object Manager). CIMOM handles the interaction between the Providers and other modules in the WBEM environment.
The networked computer system is managed by a server on the network running a WBEM-based network management application -- the
Central Management Server (CMS). For HP, the management
application will typically be HP Systems Insight Manager (SIM).
The CMS functions as an interface for System Fault Manager, as well
as for many other tools. With CMS, you can view the information gathered by the
providers. In addition, you can set up subscriptions so that any CMS on the network can receive hardware events ("indications") and thus monitor the health of servers on the network. A CMS can set up a subscription for itself or for any other CMS on the network.
This webpage first explains how to prepare HP SIM for System Fault Manager, and then shows how to set up subscriptions for hardware events (indications.)
Using HP Systems Insight Manager (SIM) with System Fault Manager
Effective use of SysFaultMgmt requires a familiarity with HP Systems Insight Manager. Refer to the extensive documentation and training available at websites such as:
Installing System Fault Manager
If you have not yet installed System Fault Manager, refer to the instructions at
System Fault Management: Installation.
Preparing HP SIM for System Fault Manager
To prepare HP Systems Insight Manager for use with System Fault Manager, and to validate the CPU and Memory Providers:
- Install and configure the Central Management Server (CMS) -- the server which will run HP Systems Insight Manager.
The procedure is in the installation chapters of the HP Systems Insight Manager Installation and User Guide, available at (
http://docs.hp.com/hpux/netsys (see the sub-section named "HP Systems Insight Manager"). In the manual, refer to the appropriate chapter based on the operating system of your CMS (for example, HP-UX, Linux, or Windows).
- Install and configure the required management agents on the systems that will be managed by the CMS. This step is covered in the first section of the "Initial Setup" chapter of the HP Systems Insight Manager Installation and User Guide.
- Configure HP Systems Insight Manager for your environment. The remaining sections in the "Initial Setup" chapter cover the recommended tasks.
- Make sure that the security and passwords are set properly on the server that has SysFaultMgmt installed. To access the SysFaultMgmt providers via HP SIM requires root capability.
- Make sure the global properties settings in HP SIM for WBEM include the user/password covering the server that has SysFaultMgmt installed.
For example, you can use the menu selection: Options | Protocol Settings | Global Protocol Settings.
- Have HP SIM discover systems to be monitored on the network, for example with the menu selection Options | Discovery | Automatic Discovery, and then by entering the IP range of your sub-nets to search.
After HP SIM discovers and identifies the system with SysFaultMgmt installed, it will have a database setting indicating the successful identification. This can be viewed in the
HP SIM GUI in several ways. For example, follow the procedure in the next section for viewing CPU and memory information.
Viewing CPU and Memory Information
- Create an advanced system search for a management protocol of WBEM.
(Alternately, you can create customized list, by pressing the Customize button and then the New List button; select WBEM as the manaagment protocol.)
- Select the node from the All Systems list. Click on the node name. The system page which appears will have WBEM listed in the Manaagement Protocols line.
- Select the node that has SysFaultMgmt installed, then select Tools | System Information | Properties. The WBEM properties will be displayed. You should be able to see information about the CPU and memory from the SysFaultMgmt Providers.
Select different tabs to see the Identify, Configuration and Status pages for both
memory and CPU.
Sample screens for the CPU and Memory Providers are shown below.
Set Up Subscriptions Using mxwbemsub
Once you have completed the preliminary steps described in
Using HP SIM to Set Up Subscriptions for WBEM Indications, you can use the mxwbemsub command to complete the subscription process. (If you have SSH installed, you can instead
Set Up Subscriptions Using the HP SIM GUI.)
- Open a terminal window or a command prompt window to execute HP SIM commands.
- At the command line interface (CLI) prompt, enter the mxwbemsub command with the appropriate parameters.
The mxwbemsub program requires WBEM subscriptions privilege on each of the managed nodes. Currently the default is to require root privilege on the managed node. The SIM user documentation describes how to set up WBEM access for the managed nodes. If you have not done this during the set up of the
CMS, do it now.
The syntax for mxwbemsub is:
mxwbemsub –a [destination] –n <nodenames>
where:
destination is the IP address or DNS name of the CMS to receive the WBEM indications. If no destination is specified, the default destination is the CMS on which the command mxwbemsub is being executed.
nodename is the IP address or DNS name of the server on which SysFaultMgmt is installed and which will be sending WBEM indications (the server which is to be monitored)
For example, if the monitored system has the IP address 206.156.18.122, and the subscription is for the CMS on which you are now logged in, you would enter:
mxwbemsub –a –n 206.156.18.122
If you wanted to create a subscription so that WBEM indications would go to a different CMS, say one that had the IP address 206.156.50.77, you would enter:
mxwbemsub –a 206.156.50.77 –n 206.156.18.122
- Using HP SIM, you can configure and display WBEM indications just as you would other events. For example, you can list them by management protocol (WBEM) or severity. See the SIM documentation for details.
Controlling Subscriptions with the mxwbemsub Command
You can use the mxwbemsub command to add, delete, and otherwise manage the WBEM subscriptions. See the following table.
| Switch | Function | Syntax |
|---|
| a
|
Adds (creates) a subscription, so that WBEM indications from nodenames
are received by destination. If the destination is omitted, the default
destination is the CMS on which the command is being entered.
|
mxwbemsub –a [destination] –n <nodenames>
|
| r
|
Removes (deletes) a subscription
|
mxwbemsub –r [destination] –n <nodenames>
|
| l
|
Lists all instances of subscriptions
|
mxwbemsub –l [destination] –n <nodenames>
|
| m
|
Moves the subscriptions from a current destination to a new destination, that is, to a new CMS
|
mxwbemsub –m [current_destination] <new_destination> –n <nodenames>
|
| f
|
Takes arguments from the file filename.txt.
|
mxwbemsub –f <filename.txt>
|
Viewing Hardware Events (Indications)
- Get access to the desired node that has SysFaultMgmt installed, as described previously in
Viewing CPU and Memory Information.
- View the list of events (indications) by selecting "Event Lists" and "All Events" on the Main System List on the left side of the HP SIM window.
(See the figure Sample List of EMS Events.)
- For more details on a particular event in the list, click on it.
(See the figure Details of a Sample EMS Event.)
- You can customize the list of events by pressing the "Customize List" button, then
from the pulldown menu, selecting "Event Lists".
(See the figure Creating a Customized List of Events .)
Sample screens for the EMS events are shown below.
Sample List of EMS Events
Details of Sample EMS Event
Creating a Customized List of Events
Validating the Events/Indications Process
If desired, you can check that a subscription set up by HP SIM is working and that the destination CMS
is receiving indications, with the following procedure:
- Temporarily change the SFM logging threshold level to INFORMATIONAL.
The procedure is described in the Configuring SysFaultMgmt Logging
section, below.
Only change this threshold to INFORMATIONAL for a short period of time,
since otherwise the logfile may become very large (the reason for
changing the threshold is so that test event(#103), with a severity
level of INFORMATIONAL, will be passed on to the subscribing servers).
- Verify that the hardware monitors are working, by using the
send_test_event command, as described at:
Verifying EMS Hardware Monitors
webpage.
Check the EMS hardware monitor log, as described on this page, to make sure that the hardware monitors themselves are working.
- Check the logs on the networked systems that have been set up by HP SIM to be subscribers. The log(s) should have entries for the test_event.
- Re-set the logging threshold level to its default of WARNING.
Configuring SysFaultMgmt Logging
Currently only logging-related configuration can be changed for SystFaultMgmt.
In general, HP recommends that you keep the default configurations for SysFaultMgmt and the providers as they
come from the factory.
If desired, you can change the logging threshold or specify a new logfile.
To configure logging for SysFaultMgmt:
- With an ASCII editor, call up the configuration file /opt/sfm/conf/FMLoggerConfig.xml.
The contents of the file should appear similar to the following:
<SFMConfig>
<LoggerConfig>
<Severity> WARNING </Severity>
<Target> /opt/sfm/log/sfm.log </Target>
</LoggerConfig>
</SFMConfig>
-
To modify the threshold logging level, change the Severity value.
In increasing level of severity, the possible values are:
INFORMATIONAL
WARNING (default)
ERROR
CRITICAL
CAUTION
HP suggests that you not set the threshold to INFORMATIONAL, except perhaps temporarily.
The INFORMATIONAL setting can generate many messages and use a sizable amount of disk space.
-
To specify a new logfile, change the Target value.
The possible values are:
STDOUT (all log messages are delivered to console) or the
complete path to the file where the log messages are to be written.
The default logfile is /opt/sfm/log/sfm.log.
(The path to the logfile should already exist and should be writable by root-user).
- After you have finished editing the configuration file, run the
/opt/sfm/bin/fmControl program, to specify the changed configuration file. Specify
the complete path of the configuration file. For example:
$ /opt/sfm/bin/fmControl /opt/sfm/conf/FMLoggerConfig.xml
Glossary
This section defines the terms and acronyms helpful in
understanding SysFaultMgmt's role in WBEM.
CIM (Common Information Model).
A hierarchical object-based model for describing managed resources. CIM uses the object-oriented paradigm, where managed objects are modeled using the concepts of classes and instances.
CIMOM (CIM Object Manager)
The part of WBEM that handles the interaction between the Providers and other modules in the WBEM environment. CIMOM manages CIM objects. It receives and processes CIM operation requests and issues responses.
CIM Client
An entity in WBEM architecture which sends CIM Operation requests and receives CIM Operation responses. For example, HP SIM and other network management applications are considered CIM clients.
CIM Provider.
A program that interacts directly with hardware or software resources on
the computer system. For example, a provider might control, monitor, or run diagnostics on hardware devices.
The provider would receive instructions (requests) from the CIM server and respond with data or messages.
SysFaultMgmt includes several CIM providers:
EMS-to-WBEM wrapper, CPU Instance Provider, and Memory Instance Provider.
CIM Server
An entity in WBEM architecture which receives CIM Operation requests and sends CIM Operation responses.
An example of a CIM Server is the CIM Object Manager (CIMOM), also known as "WBEM services".
Computer systems managed in
a WBEM architecture would each have a CIM Server, which would talk to the Management Application over the
network and communicate with Providers running on the computer in order to monitor and control managed resources (such as hardware devices).
client
In a client-server model, the client requests service from a server. In WBEM,
for example, HP SIM would be a client, requesting services from the CIM Object Manager (HP WBEM Services).
consumer
A recipient of services. For example, an indication consumer will receive information about indications (events) that occur in system hardware components.
CMS (Central Management Server).
An application for managing networked computer systems. An example is a server
running HP SIM.
indication
Something that happens on a component in the system, such as a failure or warning in a hardware device. Equivalent to an event in EMS.
instance
A representation of the actual physical object that belongs to a class. For example, the physical disk drive at hardware
address aa/bb.dd on the computer WOJO. In contrast, a class is just an abstract category; a class
is not the thing itself.
instance provider
Instance Providers give information about devices, such as model number and serial number of a CPU. Under STM, this functionality was accomplished by Information (Info) Tools.
SysFaultMgmt includes a CPU Instance Provider and a Memory Instance Provider.
instrumentation
The mechanism in a component which allows an outside entity to monitor or control that component.
managed resource
A device, component, process, etc. in a networked computer which can be monitored and controlled .
management application
Any application that uses management data to provide value to users, such as a central management console. Examples: HP SIM,, OpenView, Tivoli.
HP Systems Insight Manager (SIM).
HP's multi-OS manageability tool that uses WBEM instrumentation on HP-UX,
Linux, Windows, etc.
HP SIM uses information from the providers in the SysFaultMgmt bundle.
resource.
See "Managed Resource."
subscription
Configuring the WBEM system so that consumers receive services. For example, HP SIM could request a subscription for Server XYZ to receive information of indications (events) on certain hardware devices.
WBEM (Web-Based Enterprise Management).
A collection of standards that aid large-scale systems management. The overall architecture. WBEM allows management applications to monitor and control) data from different kinds of
resources in a consistent way.
wrapper
A WBEM provider that translates EMS hardware events from the EMS hardware
monitors into a form that can be used by WBEM (namely, "indications").
