[cmd/mdatagen] Generate more mdatagen attribute documentation #8985
Conversation
|
@dehaansa we just need a rebase to merge this. |
| | Name | Description | Values | | ||
| | ---- | ----------- | ------ | | ||
| | cache_name | The name of cache. | fielddata, query | | ||
| | collector_name (name) | The name of the garbage collector. | | |
There was a problem hiding this comment.
Is there a value in the data in parentheses? It seems to either include information that is already in the attribute name in some cases. I see further down some other metric attributes names (mysql receiver) don't have the suffix, should they include it?
The question is really around consistency of the metrics being generated. I'm happy to have this answered in a separate issue though.
There was a problem hiding this comment.
the (name) in parentheses is the actual attribute key that will be on the metrics produced, whereas the "collector_name" is the fully unique name required to generate metadata when you have multiple possible "name" attributes for different metrics in the same scraper. When there is no potential conflict, the attributes don't need a fully unique name to represent similarly named attributes with different enums/descriptions. Does that clarify?
|
I'll merge while this is still green :-) |
Description: Improves the documentation of attributes detailed in metadata.yaml. Adds the "value" field to name so that users can see what the actual attribute value will be on metrics, and enumerates the possible values when available.
Link to tracking Issue: #8983
Testing: Ensured that attributes are used in the main test.