Vertiv PDU missing circuits in the graph

[dust24_99]

Here is the output of ./daily.sh:

Fetching new release information OK
Updating to latest release OK
Updating Composer packages OK
Updating SQL-Schema OK
Updating submodules OK
Cleaning up DB OK
Fetching notifications OK
Caching PeeringDB data OK
Caching Mac OUI data OK

discovery output:
https://p.libren.ms/view/1759330b

poller output:
https://p.libren.ms/view/40408403

These are the circuits that should be logged:

(pduBreakerLabel)
.1.3.6.1.4.1.21239.5.2.3.3.1.3.1 = STRING: Circuit 1 - Phase AB
.1.3.6.1.4.1.21239.5.2.3.3.1.3.2 = STRING: Circuit 2 - Phase BC
.1.3.6.1.4.1.21239.5.2.3.3.1.3.3 = STRING: Circuit 3 - Phase CA
.1.3.6.1.4.1.21239.5.2.3.3.1.3.4 = STRING: Circuit 4 - Phase AB
.1.3.6.1.4.1.21239.5.2.3.3.1.3.5 = STRING: Circuit 5 - Phase BC
.1.3.6.1.4.1.21239.5.2.3.3.1.3.6 = STRING: Circuit 6 - Phase CA

(pduBreakerCurrent)
.1.3.6.1.4.1.21239.5.2.3.3.1.4.1 = Gauge32: 476 centiamps (rms)
.1.3.6.1.4.1.21239.5.2.3.3.1.4.2 = Gauge32: 480 centiamps (rms)
.1.3.6.1.4.1.21239.5.2.3.3.1.4.3 = Gauge32: 475 centiamps (rms)
.1.3.6.1.4.1.21239.5.2.3.3.1.4.4 = Gauge32: 258 centiamps (rms)
.1.3.6.1.4.1.21239.5.2.3.3.1.4.5 = Gauge32: 254 centiamps (rms)
.1.3.6.1.4.1.21239.5.2.3.3.1.4.6 = Gauge32: 250 centiamps (rms)

(pduPhaseLabel)
.1.3.6.1.4.1.21239.5.2.3.2.1.3.1 = STRING: Phase AB
.1.3.6.1.4.1.21239.5.2.3.2.1.3.2 = STRING: Phase BC
.1.3.6.1.4.1.21239.5.2.3.2.1.3.3 = STRING: Phase CA

(pduPhaseCurrent)
.1.3.6.1.4.1.21239.5.2.3.2.1.8.1 = Gauge32: 731 centiamps (rms)
.1.3.6.1.4.1.21239.5.2.3.2.1.8.2 = Gauge32: 729 centiamps (rms)
.1.3.6.1.4.1.21239.5.2.3.2.1.8.3 = Gauge32: 727 centiamps (rms)

However, Circuits 1-3 are not graphed, only Circuits 4-6 and the 3 different phases. I had to create custom configuration for this device, so it’s very possible I did something wrong, but not sure where as I didn’t specifically configure the graphs that are working. Below are the custom files that I created for this device:

/includes/definitions/discovery/vertiv-pdu.yaml
https://p.libren.ms/view/8ac570d1

/includes/definitions/vertiv-pdu.yaml
https://p.libren.ms/view/9a4ed7db

/mibs/vertiv/VERTIV-V5-MIB
https://p.libren.ms/view/0ed21735

[dust24_99]

Here is a screenshot of the graph. Is there a limit to the number of metric that it will put on a graph by chance?:

[dust24_99]

So, it appears that discovery/polling is overwriting the values for the missing circuits in the sensors table. 2 different OIDs with the same sensor_type have the same index value The below query returns the same results for 2 different OIDs and is updating the existing record rather than inserting a new record:

Sensor fetched value: 307
Cur 3.07, Low: null, Low Warn: null, Warn: null, High: null
Discover sensor: .1.3.6.1.4.1.21239.5.2.3.2.1.8.1, 1, vertiv-pdu, Phase AB, snmp, 100, 1, , 3.07, (limits: LL: , LW: , W: , H: )
SQL[SELECT COUNT(sensor_id) FROM sensors WHERE poller_type= ? AND sensor_class = ? AND device_id = ? AND sensor_type = ? AND sensor_index = ? [“snmp”,“current”,241,“vertiv-pdu”,“1”] 0.52ms]

Sensor fetched value: 101
Cur 1.01, Low: null, Low Warn: null, Warn: null, High: null
Discover sensor: .1.3.6.1.4.1.21239.5.2.3.3.1.4.1, 1, vertiv-pdu, Circuit 1 - Phase AB, snmp, 100, 1, , 1.01, (limits: LL: , LW: , W: , H: )
SQL[SELECT COUNT(sensor_id) FROM sensors WHERE poller_type= ? AND sensor_class = ? AND device_id = ? AND sensor_type = ? AND sensor_index = ? [“snmp”,“current”,241,“vertiv-pdu”,“1”] 0.44ms]

I am wondering if looking at the OID values would be useful to further distinguish between different sensors along with the poller_type, device_id, sensor_type, and sensor_index.

[dust24_99]

I was able to resolve this by using the index field in the discovery yaml file.