Issue
Users would like to check vital signs from a device that they want to monitor to get information proactively and to prevent an unexpected reboot or crash.
Resolution
The users can look for the dataplane CPU, the management plane CPU, packets per second through the device throughput, through the device and a number of sessions (TCP, UDP, ICMP), as these are the vital information that would report the device's status at a given time. In order to monitor these, the user would need a baseline per device. Within a same organization, there can be a device that is utilized heavily or is under utilized. In any of these scenarios, the baseline would differ per device. There is no single value that can be suggested to the user. We can use the monitoring tool to get a baseline for a particular device. Once there is an expected average, the user can define the criterion that would trigger an alert. The user might have to tune it based on the information that is gotten as an alert.
See Also
The following document details the Zabbix where the interface, CPUs and session information are monitored:
Zabbix - Basic PA Host Template
owner: ssharma