Prometheus data collected from telegraf using redfish categorized incorrectly

Question

I have telegraf running on host 192.168.0.32 and it's using the redfish module to query its own ILO with IP 192.168.0.192.

This is the telegraf.conf, very generic (from some HP document), just set to provide data for prometheus:

[agent]
## Default data collection interval for all inputs
interval = "30s"

## blabla

[[outputs.prometheus_client]]
## Address to listen on.
listen = ":9273"

This is the ilo.conf in telegraf:

# Read CPU, Fans, Powersupply and Voltage metrics of hardware server through redfish APIs
[[inputs.redfish]]
address = "https://192.168.0.192"

## Credentials for the Redfish API.
username = "admin"
password = "password"

## System Id to collect data for in Redfish APIs.
computer_system_id="1"

## Amount of time allowed to complete the HTTP request
timeout = "5s"

## Use TLS but skip chain & host verification
insecure_skip_verify = true

In Prometheus, I have a scrape configuration for that host, also very simple:

scrape_configs:
  - job_name: ilo
    static_configs:
      - targets:
        - "192.168.0.32:9273"

The data is definitely coming in, and some of it is correct: redfish_thermal_temperatures_reading_celsius{address="192.168.0.192", health="OK", host="control0", instance="192.168.0.32:9273", job="ilo", member_id="0", name="01-Inlet Ambient", state="Enabled"}

However, much of the data is being organized incorrectly: redfish_thermal_temperatures_reading_celsius{address="192.168.0.192", health="OK", host="control0", instance="192.168.0.32:9273", job="ilo", member_id="31", name="28.1-OCP 1-Network controller", state="Enabled"}

and

redfish_thermal_temperatures_upper_threshold_critical{address="192.168.0.192", health="OK", host="control0", instance="192.168.0.32:9273", job="ilo", member_id="3", name="06-P1 DIMM 7-12", state="Enabled"}

Those two are for the status of a network device and a DIMM, but being stored under temperature reading and temperature threshold. There are a ton of these incorrect entries. It seems that the setup here is pretty simple, just tell the redfish module to query and then grab that data from Prometheus, there isn't any kind of manual mapping for what data is in what category, so I would have assumed that redfish would handle this out of the box.

Anyone have any suggestions here?