Client and Target Applications

This section describes some of the client and target application options and processes. Target applications can be written using the EMS API.

EMS with ServiceGuard

ServiceGuard can be configured with EMS to monitor the health of selected resources, such as disks. Based on the status of the resources, ServiceGuard can decide to fail packages over. When working with EMS, ServiceGuard acts as both the client and target application. EMS works with both the MC/ServiceGuard and ServiceGuard OPS Edition.

Configure EMS requests for use with ServiceGuard packages either:

through the Package Configuration in the Cluster area of SAM
by editing the package configuration ASCII file

In addition, is it recommended that you also create requests through EMS to:

enable a redundant notification system
monitor events that affect high availability
be alerted to the cause of a package failover

ServiceGuard may already be configured to monitor the health of nodes, services, and subnets, and to make failover decisions based on the status of these resources. Using EMS with ServiceGuard adds to the set of failures or events that trigger failover and affect availability.

EMS GUI Client Application

Use the EMS GUI, found in the Resource Management area of SAM, to create monitoring requests for resources and targets. The EMS GUI starts from the graphical version of SAM. Click through and select from the various screens to define your monitoring request. The options include:

Select resources to be monitored.
The full path of a resource includes the resource class hierarchy and instance. An example of a full resource path for the physical volume status of the device /dev/dsk/c0t1d2 belonging to volume group vgDataBase, is /vg/vgDataBase/pv_pvlink/status/c0t1d2.
Specify when to collect value. Select either and/or all:
- When value is ...
  If you are setting up a request for an asynchronous monitor, this is the only option available.
- When value changes
- At each interval
  Select this option to send an event periodically, regardless of the value.
  Define a polling interval that is appropriate to your system performance and reaction time needs. See Step 3.
Specify a polling interval for how often the monitor checks the resource and reports the value.
This applies only to non-asynchronous monitors and goes with the At each interval option in Step 2.
Specify how often the monitor should check and send notification about the resource: the
- Initial option immediately checks and returns the resource value regardless of threshold conditions
- the Repeat option checks and returns the resource value at each polling interval if threshold conditions have been met
- the Return option checks and returns the resource value after a threshold condition has been resolved and the threshold condition is not longer true.
Specify the notification protocols:
- opcmsg (IT/O), by severity or map severity from values listed
- TCP or UDP
- SNMP trap, by severity or map severity from values listed
- email
- console
- syslog
- textlog

EMS and Target Applications

Target applications receive notification messages about the monitored resources.

To help configure your Network Node Manager and IT/Operations or other system management software for EMS, refer to the Writing Monitors for the Event Monitoring Service (EMS) (HP Part Number B7611-90016)) developer's kit web page:

Go to the web site:
http://software.hp.com.
From the web site select High Availability, then select Event Monitoring Service Developers Kit.
Select Templates.

Client and Target Applications

Technical documentation

» Table of Contents

» Glossary

» Index

EMS with ServiceGuard

EMS GUI Client Application

EMS and Target Applications