I were working for one month trying to find a guide to to the next.
With grafana I’m trying to create a monitor tool with different detail level
The data are retrieve from Prometheus with SNMP (SNMP Exporter or Telegraf)
The BBDD could be TSDB or InfluxDB
First level, Summary Dashboard with one graph element (Text, gauge bar), that show three states
Disk icon (Red Critical fail, Orange Warning, Green No Errors)
Network Icon (Red Critical fail, Orange Warning, Green No Errors)
The others levels are more detailed dashboard of each element
If somebody push over Disk Icon, a new dashboard sill show detailed information about the NAS/SAN (Disk, benchmark, NIC stats)
If Network Icon is selected (the dashboard displays all elements about switch (Ports, PSU, FANs, ….)
The second level is easy, is just to create the dashboard with all OID stored at Prometheus
The first level is where the pain is. It is not usual that the devices have a OID with overall status, either way, these overall status not recognize a switch port link down like an error that should affect to led status. Normally the Status led/OID is affected by hardware failure (FAN, PSU, Temp).
For example, that I need for a switch dashboard is, if there is a hardware failure or one switch port has link down from a restricted list be managed like a big fail, and the first level dashboard network icon turn in orange or red.
In short, the goal is to get a mechanism to elevate any status change in secondary dashboards to the main dashboard
Maybe that I am looking for is not under grafana business.
Anyway, I would be pleased to receive any suggest