Files
agent-framework/dotnet/src/Microsoft.Agents.AI.Workflows/Execution/LockstepRunEventStream.cs
T
Tao Chen e3b4b6662b .NET: Workflow telemetry opt in (#3467)
* feat(workflows): Make telemetry opt-in via WithOpenTelemetry()

- Add WorkflowTelemetryOptions class with EnableSensitiveData property
- Add WorkflowTelemetryContext to manage ActivitySource lifecycle
- Add WithOpenTelemetry() extension method on WorkflowBuilder
- Update all workflow components to use telemetry context:
  - WorkflowBuilder, Workflow, Executor
  - InProcessRunnerContext, InProcessRunner
  - LockstepRunEventStream, StreamingRunEventStream
  - All edge runners (Direct, FanIn, FanOut, Response)
- Telemetry is now disabled by default
- Users must call WithOpenTelemetry() to enable spans/activities

BREAKING CHANGE: Workflow telemetry is now opt-in. Users who relied on
automatic telemetry must add .WithOpenTelemetry() to their workflow builder.

* refactor: Pass telemetry context as parameter instead of via interface

- Remove IWorkflowContextWithTelemetry interface
- Add internal ExecuteAsync overload that accepts WorkflowTelemetryContext
- Public ExecuteAsync delegates with WorkflowTelemetryContext.Disabled
- InProcessRunner passes TelemetryContext when calling ExecuteAsync
- BoundContext now implements IWorkflowContext (not the removed interface)

* Add optional ActivitySource parameter to WithOpenTelemetry

Allow users to provide their own ActivitySource when enabling telemetry,
giving them better control over the ActivitySource lifecycle. When not
provided, the framework creates one internally (existing behavior).

Changes:
- Add optional activitySource parameter to WithOpenTelemetry() extension
- Update WorkflowTelemetryContext to accept external ActivitySource
- Add unit test for user-provided ActivitySource scenario

* Add component-level telemetry control with disable flags

Allow users to selectively disable specific activity types via
WorkflowTelemetryOptions. All activities are enabled by default.

New disable flags:
- DisableWorkflowBuild: Disables workflow.build activities
- DisableWorkflowRun: Disables workflow_invoke activities
- DisableExecutorProcess: Disables executor.process activities
- DisableEdgeGroupProcess: Disables edge_group.process activities
- DisableMessageSend: Disables message.send activities

Added helper methods to WorkflowTelemetryContext for each activity type
and updated all activity creation sites to use them.

* Implement EnableSensitiveData to log executor input/output

When EnableSensitiveData is true in WorkflowTelemetryOptions, executor
input and output are logged as JSON-serialized attributes in the
executor.process activity.

New activity tags:
- executor.input: JSON serialized input message
- executor.output: JSON serialized output result (non-void only)

Added suppression attributes for AOT/trimming warnings since this is
an opt-in feature for debugging/diagnostics.

* Refactor activity start methods to centralize tagging logic

Move tagging logic into WorkflowTelemetryContext methods:
- StartExecutorProcessActivity now accepts executorId, executorType,
  messageType, and message; sets all tags including executor.input
  when EnableSensitiveData is true
- Added SetExecutorOutput method to set executor.output after execution
- StartMessageSendActivity now accepts sourceId, targetId, and message;
  sets all tags including message.content when EnableSensitiveData is true

Simplified Executor.cs and InProcessRunnerContext.cs by removing
inline tagging code. Added message.content tag constant.

* Revert Python changes

* Update samples and code cleanup

* Fix file formatting

* Add comment

* Add telemetry configuration to declarative workflow

* Remove delays in tests

* Address comments
2026-02-09 23:10:50 +00:00

182 lines
6.7 KiB
C#

// Copyright (c) Microsoft. All rights reserved.
using System;
using System.Collections.Concurrent;
using System.Collections.Generic;
using System.Diagnostics;
using System.Runtime.CompilerServices;
using System.Threading;
using System.Threading.Tasks;
using Microsoft.Agents.AI.Workflows.Observability;
namespace Microsoft.Agents.AI.Workflows.Execution;
internal sealed class LockstepRunEventStream : IRunEventStream
{
private readonly CancellationTokenSource _stopCancellation = new();
private readonly InputWaiter _inputWaiter = new();
private int _isDisposed;
private readonly ISuperStepRunner _stepRunner;
public ValueTask<RunStatus> GetStatusAsync(CancellationToken cancellationToken = default) => new(this.RunStatus);
public LockstepRunEventStream(ISuperStepRunner stepRunner)
{
this._stepRunner = stepRunner;
}
private RunStatus RunStatus { get; set; } = RunStatus.NotStarted;
public void Start()
{
// No-op for lockstep execution
}
public async IAsyncEnumerable<WorkflowEvent> TakeEventStreamAsync(bool blockOnPendingRequest, [EnumeratorCancellation] CancellationToken cancellationToken = default)
{
#if NET
ObjectDisposedException.ThrowIf(Volatile.Read(ref this._isDisposed) == 1, this);
#else
if (Volatile.Read(ref this._isDisposed) == 1)
{
throw new ObjectDisposedException(nameof(LockstepRunEventStream));
}
#endif
CancellationTokenSource linkedSource = CancellationTokenSource.CreateLinkedTokenSource(this._stopCancellation.Token, cancellationToken);
ConcurrentQueue<WorkflowEvent> eventSink = [];
this._stepRunner.OutgoingEvents.EventRaised += OnWorkflowEventAsync;
using Activity? activity = this._stepRunner.TelemetryContext.StartWorkflowRunActivity();
activity?.SetTag(Tags.WorkflowId, this._stepRunner.StartExecutorId).SetTag(Tags.RunId, this._stepRunner.RunId);
try
{
this.RunStatus = RunStatus.Running;
activity?.AddEvent(new ActivityEvent(EventNames.WorkflowStarted));
do
{
while (this._stepRunner.HasUnprocessedMessages &&
!linkedSource.Token.IsCancellationRequested)
{
// Because we may be yielding out of this function, we need to ensure that the Activity.Current
// is set to our activity for the duration of this loop iteration.
Activity.Current = activity;
// Drain SuperSteps while there are steps to run
try
{
await this._stepRunner.RunSuperStepAsync(linkedSource.Token).ConfigureAwait(false);
}
catch (OperationCanceledException)
{
}
catch (Exception ex) when (activity is not null)
{
activity.AddEvent(new ActivityEvent(EventNames.WorkflowError, tags: new() {
{ Tags.ErrorType, ex.GetType().FullName },
{ Tags.BuildErrorMessage, ex.Message },
}));
activity.CaptureException(ex);
throw;
}
if (linkedSource.Token.IsCancellationRequested)
{
yield break; // Exit if cancellation is requested
}
bool hadRequestHaltEvent = false;
foreach (WorkflowEvent raisedEvent in Interlocked.Exchange(ref eventSink, []))
{
if (linkedSource.Token.IsCancellationRequested)
{
yield break; // Exit if cancellation is requested
}
// TODO: Do we actually want to interpret this as a termination request?
if (raisedEvent is RequestHaltEvent)
{
hadRequestHaltEvent = true;
}
else
{
yield return raisedEvent;
}
}
if (hadRequestHaltEvent || linkedSource.Token.IsCancellationRequested)
{
// If we had a completion event, we are done.
yield break;
}
this.RunStatus = this._stepRunner.HasUnservicedRequests ? RunStatus.PendingRequests : RunStatus.Idle;
}
if (blockOnPendingRequest && this.RunStatus == RunStatus.PendingRequests)
{
try
{
await this._inputWaiter.WaitForInputAsync(TimeSpan.FromSeconds(1), linkedSource.Token).ConfigureAwait(false);
}
catch (OperationCanceledException)
{ }
}
} while (!ShouldBreak());
activity?.AddEvent(new ActivityEvent(EventNames.WorkflowCompleted));
}
finally
{
this.RunStatus = this._stepRunner.HasUnservicedRequests ? RunStatus.PendingRequests : RunStatus.Idle;
this._stepRunner.OutgoingEvents.EventRaised -= OnWorkflowEventAsync;
}
ValueTask OnWorkflowEventAsync(object? sender, WorkflowEvent e)
{
eventSink.Enqueue(e);
return default;
}
// If we are Idle or Ended, we should break out of the loop
// If we are PendingRequests and not blocking on pending requests, we should break out of the loop
// If cancellation is requested, we should break out of the loop
bool ShouldBreak() => this.RunStatus is RunStatus.Idle or RunStatus.Ended ||
(this.RunStatus == RunStatus.PendingRequests && !blockOnPendingRequest) ||
linkedSource.Token.IsCancellationRequested;
}
/// <summary>
/// Signals that new input has been provided and the run loop should continue processing.
/// Called by AsyncRunHandle when the user enqueues a message or response.
/// </summary>
public void SignalInput()
{
this._inputWaiter?.SignalInput();
}
public ValueTask StopAsync()
{
this._stopCancellation.Cancel();
return default;
}
public ValueTask DisposeAsync()
{
if (Interlocked.Exchange(ref this._isDisposed, 1) == 0)
{
this._stopCancellation.Cancel();
this._stopCancellation.Dispose();
this._inputWaiter.Dispose();
}
return default;
}
}