Cisco SNMP Temperature discovery

[MADxHAWK]

Ive noticed that i get a lot of false alerts from Cisco Temperature monitoring since there is only one threshold used for every temp sensor

for example on an C9000 with IOSX there are multiple sensors with different treshohlds

Temperature:
--------------------------------------------------------------------
Module   Sensor        MajorThresh   MinorThres   CurTemp     Status
                       (Celsius)     (Celsius)    (Celsius)
--------------------------------------------------------------------
1        FRONT           80              70          24         Ok
1        BACK            70              42          24         Ok
1        CPU             90              80          32         Ok
1        Heavenly        110             90          49         Ok

or another Catalyst C9xxxx series switch where many temperatures during usual conditions are above the fixed 50°C thereshold.

 Slot          Sensor          Current State   Reading        Threshold(Minor,Major,Critical,Shutdown)
 ----------    --------------  --------------- ------------   ---------------------------------------
​Chassis1-R0   Temp: Coretemp  Normal          46   Celsius     (107,117,123,125)(Celsius)
Chassis1-R0   Temp: UADP      Normal          52   Celsius     (107,117,123,125)(Celsius)
Chassis1-R0   Temp:   Outlet  Normal          37   Celsius     (63 ,73 ,103,105)(Celsius)
Chassis1-R0   Temp:    Inlet  Normal          28   Celsius     (56 ,66 ,96 ,98 )(Celsius)
Chassis2-R0   Temp: Coretemp  Normal          47   Celsius     (107,117,123,125)(Celsius)
Chassis2-R0   Temp: UADP      Normal          62   Celsius     (107,117,123,125)(Celsius)​

I would suggest to to use the temperatur (oid iso.3.6.1.4.1.9.9.13.1.3.1.3.xxx) for monitoring and the oid iso.3.6.1.4.1.9.9.13.1.3.1.6.xxx which is the current state (1 is OK) for the trigger instead of using a fixed threshold for.