MartinDotNet

In terms of the wider idea of the allocations, my suggestion to sidestep this if it's a concern is to make it opt-in by adding a parameter that will do...

I'm not sure what you're asking here. `ActivityEvent` (the equivalent of a SpanEvent) is attached to an Activity, so the health check would appear as an Activity, I'm not sure...

I'm not sure what you mean by "publisher". Right now, http health checks result in an activity being created, and any of the checks that are part of that health...

This is interesting and new to me. What I would expect here is for the publish to be a noop, and the actual act of doing the healthcheck to be...

What I'm saying is that the Publish is the Trace, you don't need the `PublishAsync` part at all. Your healthcheck trace will be the thing you use for alerting on...

You should absolutely have multiple spans for the healthcheck not events on a single span, that wouldn't make sense.. I'm just not seeing how Events on the publish activity would...

What would be on the event, that wouldn't be attributes on the span? Would there be more than one event on the span? And if so, what would they contain...

From what I can tell from testing this, it does stop activity from being created, however, the runtime code does not honor that and still propagates the trace data.

So ultimately, the issue is that I need to disable the runtime propagation, but if I do that (with the switches), then the Http instrumentation doesn't kick in, because it's...

> Ya it will still create an Activity \ span but it shouldn't be parented to the current trace it will be a new root.  @CodeBlanch I'm saying that isn't...