Include metric name of limited metric in internal logs emitted by tag_cardinality_limit

[jszwedko]

A note for the community

• Please vote on this issue by adding a 👍 reaction to the original issue to help the community and maintainers prioritize this request
• If you are interested in working on this issue or have submitted a pull request, please leave a comment

Use Cases

A user would like to know which metric hit the cardinality limit. The current internal logs just have the tag name and value. Example current log:

{
  "timestamp": "2022-11-15T14:31:46.054273Z",
  "level": "DEBUG",
  "message": "Rejecting tag after hitting configured value_limit.",
  "tag_key": "name",
  "tag_value": "7bfadae4ec1fc4eb7247fedd385641314940085384283b7b45df5d4f6e8a51a1",
  "internal_log_rate_secs": 10,
  "target": "vector::internal_events::tag_cardinality_limit",
  "span": {
    "component_id": "drop_high_cardinality_tags",
    "component_kind": "transform",
    "component_name": "drop_high_cardinality_tags",
    "component_type": "tag_cardinality_limit",
    "name": "transform"
  },
  "spans": [
    {
      "component_id": "drop_high_cardinality_tags",
      "component_kind": "transform",
      "component_name": "drop_high_cardinality_tags",
      "component_type": "tag_cardinality_limit",
      "name": "transform"
    }
  ]
}

Attempted Solutions

No response

Proposal

Include the name of the metric in the log.

References

• tag_cardinality_limit issues #15395

Version

vector 0.26.0

[pablito-perez]

Hello, I'm not a proficient rustacean but I'd like to take a stab at this issue, we really need better visibility on this component's behaviour 🙏

[spencergilbert]

Great! Thanks @nomonamo! Feel free to start a thread in the development channel on our Discord server if you need a more real time conversation about anything, otherwise we're happy to help guide things with feedback once you get a draft PR open.