-
Notifications
You must be signed in to change notification settings - Fork 2.2k
docs: send traces to OpenTelemetry backends #2619
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
docs: send traces to OpenTelemetry backends #2619
Conversation
|
🙌 Please fix the DCO check by signing your commit. |
This PR adds a part to the observability docs page that shows how to send Spring AI traces to an OpenTelemetry backend such as Langfuse. Signed-off-by: Jannik Maierhöfer <[email protected]>
d3b88db to
28055b9
Compare
|
Thanks @dev-jonghoonpark :) Just signed off the commit! |
|
Hi @dev-jonghoonpark what do you think about this addition? Let me know if there is anything I can add/edit :) |
|
Sorry for the late reply. @jannikmaierhoefer I think it's a good guide. It inspired me with a couple of ideas that might help in the future:
But unfortunately, I don't have the authority to approve PRs. |
|
Thanks @dev-jonghoonpark! @tzolov what do you think about this addition? |
|
Thanks so much for your contribution! I think it's great to show the integration with Langfuse! I would suggest perhaps to create a dedicated page under the Observability section to showcase the different integrations (starting with a Langfuse page). What do you think? About the example, I'm afraid the combination of dependencies in the docs is not supported in Spring Boot. It's combining the usage of the OpenTelemetry support in Spring Boot/Micrometer with the OpenTelemetry Spring Boot Starter from the OpenTelemetry Java Instrumentation. They are two mutually exclusive approaches, and I'm not sure about having them both in the same application, since they work in a very different way. Since the Spring AI instrumentation is provided via Micrometer, would it be possible to describe the scenario with that approach (without the OpenTelemetry Spring Boot Starter)? Besides the project specific instrumentation, Spring AI fully relies on the observability backbone in Spring Boot. Maybe we should point to the Spring Boot documentation about exporting OpenTelemetry traces from a Spring Boot app? There might be differences between Spring Boot versions, so having the docs there might help identifying the right approach based on the used Spring Boot version. Perhaps @jonatan-ivanov has some suggestions. |
|
My two cents: I don't think we should do this for every single backend that the Observation API supports or can support. The PR feels like advertising to me. As @ThomasVitale mentioned above, this is not the recommended approach (I have doubts that it works). I think the Spring AI docs could be enhanced by linking to the Boot and Micrometer docs but I don't think it should duplicate them and list every single backend that Boot/Micrometer supports. |
|
Thanks for the contribution. I agree that Spring AI isn't the spot for this information. Please open up an issue or PR on the Spring Boot github repo so using open telemetry with any Spring Boot app (not just Spring AI) so that the entire Spring ecosystem can benefit form this. There already is some docs on this but it seems like they can be improved. If you would like to create an example specific to Spring AI, you can submit a request to create a repository on the Spring AI Community - https://github.com/spring-ai-community/community/issues All you need is an existing repo of working code that people can run. This would require using infra that everyone can spin up via docker or other infrastructure setup tools. |
This PR adds a section that shows how to send Spring AI traces to an OpenTelemetry backend such as Langfuse.