Missing parent trace activities (spans) when switching context between SDK and CLI

[alexander-ivanov-ds]

We are using Copilot SDK for .NET and it seems like the thinking behind RestoreTraceContext (this) is not entirely correct. It creates a new Activity without the ActivitySource, which, indeed, prevents it from being sampled, but it still carries its own id (span id), which is then used as a parent span id for all children activities. So in the trace tree (In Grafana, for example) we see unparented spans from the tool implementations, which I think should be parented under external_tool spans from the Copilot CLI:

the GET span (id d2c10abb001cb97a) has parent span id set to cabcf362cbf15a7a (which is nowhere to be found), where as external_tool span has an id of 2cd84acd8fb4e994. So, I think, the tree is:

external_tool (2cd84acd8fb4e994, from CLI)
 \ copilot.tool_handler (cabcf362cbf15a7a, from SDK) <--- "invisible activity"
    \ GET (d2c10abb001cb97a, from our tool code)

Why don't we just emit this span properly from ActivitySource (maybe matching SourceName/default github.copilot)?

I also think a similar problem exists in the CLI itself, because transitioning from SDK to CLI also leaves all "top-level" CLI spans unparented. CLI code doesn't seem to be open source, so I can't confirm, but my gut feeling is that a similar approach was used there in CLI, where it creates one span and doesn't attach it to a source that is being sampled/exported, so it never gets collected, leaving all other CLI spans unparented. 😞

[github-actions]

Investigated dotnet/src/Telemetry.cs (RestoreTraceContext) and its call site in dotnet/src/Session.cs (~line 701, tool handler invocation).

Finding: this is a bug in the .NET SDK. RestoreTraceContext creates new Activity("copilot.tool_handler") outside any ActivitySource, parents it to the CLI's traceparent, and calls Start(). It becomes Activity.Current with a freshly generated span id. Any child activity created by a tool handler therefore gets that new span id as its parent id. Because the activity is never sampled or exported, the span is never collected, so child spans point at a parent that doesn't exist in the backend. This matches the tree in the report (external_tool → invisible copilot.tool_handler → GET). The code comment says the activity is meant to "carry the remote parent context", but the extra span id defeats that goal.

Likely fix: make the restored context's span id be the CLI's span id, so children reference external_tool directly. Options are (a) creating the activity so its Id equals the incoming traceparent, or (b) starting a real activity from an ActivitySource so it is exported and forms a visible hop. Your suggestion of emitting it from a proper ActivitySource is option (b).

The CLI-side concern can't be verified from this repo, since the CLI isn't open source here, so it would need to be checked separately.

Labeling as bug.

Generated by Bug Handler for #2803 · copilot · auto · 15.9 AIC · ⌖ 5.42 AIC · ⊞ 7.4K · ◷

[alexander-ivanov-ds]

I don't think option (a) is, well, an option. 😕 I don't think Activity allows to specify its ID, it always tries to generate one when you call Start(): https://github.com/dotnet/dotnet/blob/main/src/runtime/src/libraries/System.Diagnostics.DiagnosticSource/src/System/Diagnostics/Activity.cs#L834-L838, so the only option is likely just creating an activity from a proper source, so it can be sampled.