Add blog post about the OTTL context inference feature #6290
Conversation
| [transform processor](https://github.com/open-telemetry/opentelemetry-collector-contrib/blob/main/processor/transformprocessor). | ||
| This feature removes the need to manually specify contexts, improving statement | ||
| processing efficiency by automatically selecting the most appropriate one. This | ||
| optimization ensures that data transformations are both accurate and performant. |
There was a problem hiding this comment.
Not strictly necessary to mention this, and we touch on it when talking about the flat configuration style, but from a general UX standpoint, this change also relieves users from needing to know the concept of contexts and instead just think about the data they want to work with.
There was a problem hiding this comment.
I think we should highlight this point as it is the primary motivator for this work.
There was a problem hiding this comment.
Should we also point out that soon this will make statements portable between components that use OTTL?
There was a problem hiding this comment.
Should we also point out that soon this will make statements portable between components that use OTTL?
It's kind of mentioning it here, but without providing much details.
There was a problem hiding this comment.
Not strictly necessary to mention this, and we touch on it when talking about the flat configuration style, but from a general UX standpoint, this change also relieves users from needing to know the concept of contexts and instead just think about the data they want to work with.
I've pushed an extra change mentioning that. Could you please take a look again? thanks!
edmocosta
changed the title
svrnm
requested review from
a team and
codeboten
and removed request for
a team
|
Requesting review from @open-telemetry/collector-approvers |
| cSpell:ignore: OTTL Vamerlatti | ||
| --- | ||
|
|
||
| Selecting the right context for running OTTL statements can be challenging, even |
There was a problem hiding this comment.
We'll want a link for context here that disambiguates it from the "context propagation" context that is predominantly used in the site.
| With the new flat configuration style, the same logic is expressed more | ||
| concisely as: |
There was a problem hiding this comment.
| With the new flat configuration style, the same logic is expressed more | |
| concisely as: | |
| With the new flat configuration style, the same logic is expressed more | |
| concisely by letting you specify the OTTL context as a part of a key: |
There was a problem hiding this comment.
| With the new flat configuration style, the same logic is expressed more | |
| concisely as: | |
| With the new flat configuration style, the same logic is expressed more | |
| concisely by simply providing a list of statements: |
I think something like that would better fit here. WDYT?
| The context inferrer would select the `metric` context instead, since no data | ||
| points are accessed. Although it would be possible to run the statements using | ||
| the `datapoint` context, `metric` is the most efficient option. | ||
|
|
There was a problem hiding this comment.
Missing here is probably a subsection stating when you should use which form. Users will generally want to know which they should generally prefer, and when.
There was a problem hiding this comment.
I tried to explain that in each configuration section:
- https://github.com/open-telemetry/opentelemetry.io/pull/6290/files#diff-34aed2660bd751b07ee7b7031b1171d91ab80dc9463584626ee01d70ee7a12f8R77-R80
- https://github.com/open-telemetry/opentelemetry.io/pull/6290/files#diff-34aed2660bd751b07ee7b7031b1171d91ab80dc9463584626ee01d70ee7a12f8R37-R40.
Do you think we need to provide more details on that?
Blog post about the upcoming OTTL context inference feature (open-telemetry/opentelemetry-collector-contrib#29017), which will be released on version
0.120.0.Closes: #6289
Preview: https://deploy-preview-6290--opentelemetry.netlify.app/