Troubleshooting Missing Traces in Tempo for Internal Processes
I find a bit surprising and underwelming the amount of traces that end up in my tracing server (tempo). I'm probably doing something wrong, but even after some refactor to make sure that I use
Is there anything I can check to understand why is this happening?
Effect.fn to have better tracing with names, adding span annotations, etc, the only things that end in the tracing server are HTTP requests. None of my internal processes produce spans that are visible on the grafana interface: Event handlers, streams, etc. None of them reach the tracing server. If I run some tests locally, basically a script that calls some of the methods that are also used in the long-lived server, they indeed appear in the traces, but for the real deployed servier, it's almost radio silence except for those cases where I execute queries against third party services, like OpenAI.Is there anything I can check to understand why is this happening?
