Meaning instead of creating spans have a lib for annotating spans through different libs like opencensus/ex_ray/etc, and documentation for Telemetry explaining how its events can be used to annotate spans in a distributed trace.
Just to clarify, I didn’t mean creating the spans via telemetry directly, but having a different (or multiple) libraries for creating spans, traces etc. So I guess we’re on the same page here 
Na, I know that is what you mean. I’m saying that shouldn’t be the case. I think because of the influence this will likely have on how Elixir users instrument their projects it should push for doing things the “proper” way from the beginning, where you use a tracing library for creating spans (not through telemetry events) and telemetry events for annotating them.
Hey, I’ve just implemented InfluxDB reporter for Telemetry:
I’ve been reading through the Telemetry docs and am confused about the unique handler-ids.
If I have matching event names in different libraries that use telemetry, would these be invoked across different handler-ids since the execute function doesn’t seem to include a handler-id?
I guess I don’t understand the point of the handler-id if it isn’t being used to discriminate between events? Is it just to provide a unique name in an ets table?
Any info would be appreciated, thanks!
The name is only so you can remove your own callback/handler later if you want to and for error reporting. It has no relation to the events. You can even use a random name if you want to.
Ahh, OK, thanks for the clarification! 
© Copyright Elixir Forum | Terms | Privacy & Cookies
