influxdata / influxdata/telegraf
[inputs.nvidia-smi] fix fields vs tag seperation
- Dominant language
- Go
- Stars
- 17.8k
- Forks
- 5.8k
- Avg merge
- 1d 20h
- Merged PRs (30d)
- 161
Description
### Use Case
I want to monitor nvidia-smi metrics (I am personally sending them to prometheus, but I do not think this is relevant here).
When monitoring, there seems to be fields that are created as tags and vice versa.
### Expected behavior
Pstate, compute mode - should be fields as they might change over time, and do not solely identify the gpu we are monitoring.
Serial - should be a tag as it (like uuid) help to identify the specific gpu, and it is constant for each gpu.
### Actual behavior
Pstate and compute mode are tags
Serial is field
### Additional info
_No response_
Contributor guide
Research direction
Start by locating the inputs.nvidia-smi implementation and its mapping of Pstate, compute mode, and Serial to fields or tags. Check existing tests for this input, then verify that the requested fields and tags are represented correctly and that the tests pass.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- go
- Domain
- observability-sre
- Issue type
- Bug
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 45/100