This section covers the procedures for monitoring Linux and Windows operating system metrics. These monitors can be classified as:
Performance monitors
- CPU
- Disc
- Memory
- Uptime
- FreeDisk
- Load
- CPU Stats
- Network
Process monitors
Collector Type : Agent
Category : Process Monitor
Monitor Name : Process
Description : Check if at least one instance of a given process is running or not. Multiple processes can be monitored.
Note: The process name must be case-sensitive.
Parameters:
Name Description Process Name Process name to monitor *All field attributes are mandatory.Service monitors
File system monitors
All OS monitors except file system monitors support alert auto-healing. Auto-healing applies when a resource threshold setting is changed. The agent sends an auto-heal alert for all existing critical and warning alerts associated with that resource threshold. At the next polling interval, the agent sends an alert, if indicated, based on the new threshold setting. This guarantees that alerts generated for the metric/monitor always apply to the latest threshold settings. The auto-heal alert description is OpsRamp
agent has generated auto-heal alert due to configuration changes for this monitor. Alert descriptions for alerts generated by new threshold crossings characterize the alert based on the new threshold setting, such as The value exceeded 70%.
For service and process monitors, if the service or process list changes, the agent sends an auto-heal alert only if a process or service is removed from the list to prevent alerts from persisting in an open state.