To generate SNMP alerts for SSM monitoring, the SNMP plug-in
must be configured in the ego.conf configuration
file. Use the EGO_EVENT_PLUGIN parameter to define
the plug-in name and configuration file location used to trap SNMP
events. EGO_EVENT_PLUGIN=eventplugin_snmp[SINK=host,MIBDIRS=
absolute_path_of_top_directory/kernel/conf/mibs]
(where
host represents the name of the host where the SNMP trap daemon is
running).
- Edit your application profile to enable SSM monitoring.
In the section of your
application profile, set the enableMonitoring parameter
to true.
- Configure the monitoring parameters in the same section
of the application profile:
- usedSlotsRatio: Specifies the ratio of used
slots for an application and is calculated as the sum of slots used
by running tasks and slots used for initialization and uninitialization
divided by the number of allocated slots.
- pollInterval: Specifies the interval (in
seconds) at which samples of a threshold breach are collected. By
default, this interval is 30 seconds.
- samplesOverThreshold: Specifies the number
of consecutive samples that match a threshold breach.
If all of the following conditions are met, an SNMP alert
is generated:
- The ratio of used slots is less than usedSlotsRatio,
- The number of pending tasks is more than zero, and
- The number of consecutive samples over threshold is greater than
samplesOverThreshold.
A warning-level message is also logged to the ssm.log file.