E! [inputs.proxmox] Error in plugin: search domain is not set

[jonaburg]

Relevant telegraf.conf

[[inputs.proxmox]]
   #base_url = "https://127.0.0.1:8006/api2/json"
   base_url = "https://localhost:8006/api2/json"
   api_token = "telegraf@pve!monitoring=$UUID"
   node_name = "proxmox"
   insecure_skip_verify = true
   interval = "10s"

Logs from Telegraf

2025-02-04T14:51:14Z I! Loading config: /etc/telegraf/telegraf.conf
2025-02-04T14:51:14Z I! Loading config: /etc/telegraf/telegraf.d/lvs.conf
2025-02-04T14:51:14Z I! Starting Telegraf 1.33.1 brought to you by InfluxData the makers of InfluxDB
2025-02-04T14:51:14Z I! Available plugins: 236 inputs, 9 aggregators, 33 processors, 26 parsers, 63 outputs, 6 secret-stores
2025-02-04T14:51:14Z I! Loaded inputs: cpu disk diskio exec kernel mem processes proxmox swap system
2025-02-04T14:51:14Z I! Loaded aggregators:
2025-02-04T14:51:14Z I! Loaded processors:
2025-02-04T14:51:14Z I! Loaded secretstores:
2025-02-04T14:51:14Z I! Loaded outputs: prometheus_client
2025-02-04T14:51:14Z I! Tags enabled: dc=Hetzner host=nsanpn06
2025-02-04T14:51:14Z I! [agent] Config: Interval:10s, Quiet:false, Hostname:"nsanpn06", Flush Interval:10s
2025-02-04T14:51:14Z W! [agent] The default value of 'skip_processors_after_aggregators' will change to 'true' with Telegraf v1.40.0! If you need the current default behavior, please explicitly set the option to 'false'!
2025-02-04T14:51:14Z D! [agent] Initializing plugins
2025-02-04T14:51:14Z D! [agent] Connecting outputs
2025-02-04T14:51:14Z D! [agent] Attempting connection to [outputs.prometheus_client]
2025-02-04T14:51:14Z I! [outputs.prometheus_client] Listening on http://[::]:9100/metrics
2025-02-04T14:51:14Z D! [agent] Successfully connected to outputs.prometheus_client
2025-02-04T14:51:14Z D! [agent] Starting service inputs
2025-02-04T14:51:20Z E! [inputs.proxmox] Error in plugin: search domain is not set
2025-02-04T14:51:24Z D! [outputs.prometheus_client] Wrote batch of 51 metrics in 2.428941ms
2025-02-04T14:51:24Z D! [outputs.prometheus_client] Buffer fullness: 0 / 10000 metrics
2025-02-04T14:51:30Z E! [inputs.proxmox] Error in plugin: search domain is not set
2025-02-04T14:51:34Z D! [outputs.prometheus_client] Wrote batch of 116 metrics in 5.414911ms
2025-02-04T14:51:34Z D! [outputs.prometheus_client] Buffer fullness: 0 / 10000 metrics
2025-02-04T14:51:40Z E! [inputs.proxmox] Error in plugin: search domain is not set

System info

Telegraf 1.33.1, debian 10, proxmox 6.4-15

Docker

I can query via REST any endpoint I'm looking for
curl -k -H 'Authorization: PVEAPIToken=telegraf@pve!monitoring=2e4a2cb8-2a0c-4cfe-b0c7-b9cd38b0f12f' https://localhost:8006/api2/json/version | jq
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 65 100 65 0 0 10833 0 --:--:-- --:--:-- --:--:-- 10833
{
"data": {
"version": "6.4-15",
"repoid": "af7986e6",
"release": "6.4"
}
}
(as well as more hidden ones like resource particulars etc.)

but the telegraf isn't budging. I do have the permissions apparently setup appropriately, as documented here: powersj@993567d
with paying attention in particular to the key & user permissions (both added to the PVEAuditor role) with privsep1, and as well the manual placement of the node_name (although telegraf & proxmox are on the same machine here).

Steps to reproduce

1. add user, token and key, set permissions
2. add proxmox input conf to telegraf (with node_name set explicitly)
   ...

Expected behavior

I expect the proxmox metrics to propagate, as I can hit them via REST requests with the same token. The configuration has the specific node_name but still goes on the search, despite them even living on the same machine.

Actual behavior

The error prints:
2025-02-04T14:56:30Z E! [inputs.proxmox] Error in plugin: search domain is not set

Additional info

No response