Improve CallService_McpTool_TelemetryRecorded test reliability

[Copilot]

Description

Refactored the CallService_McpTool_TelemetryRecorded test to be more robust and removed it from quarantine. The test was flaky due to strict assumptions about telemetry ordering and initialization state.

Changes

• Removed quarantine - Test now runs in regular CI pipeline
• Loop-based telemetry reading - Replaces direct channel reads with loops that search for specific events, tolerating initialization and other intermediate telemetry
• Focused assertions - Only validates presence of McpToolCall and EndOperation events, ignoring telemetry count and ordering
• Simplified cleanup - Removed sender disposal and empty-channel verification that assumed no other telemetry would be present

The new pattern uses break for cleaner loop control:

// Read until we find the event we care about
bool foundMcpToolCall = false;
while (await channel.Reader.WaitToReadAsync().DefaultTimeout())
{
    var context = await channel.Reader.ReadAsync().DefaultTimeout();
    if (context.Name.Contains(TelemetryEventKeys.McpToolCall))
    {
        foundMcpToolCall = true;
        break;
    }
}
Assert.True(foundMcpToolCall, "Expected to find McpToolCall telemetry event");

Validation

The test has been validated by running it 50 consecutive times with a 100% pass rate, confirming the reliability improvements have successfully eliminated the flakiness.

Checklist

• Is this feature complete?
  • Yes. Ready to ship.
  • No. Follow-up changes expected.
• Are you including unit tests for the changes and scenario tests if relevant?
  • Yes
  • No
• Did you add public API?
  • Yes
  • No
• Does the change make any security assumptions or guarantees?
  • Yes
  • No
• Does the change require an update in our Aspire docs?
  • Yes
  • No

Original prompt

Improve CallService_McpTool_TelemetryRecorded test:

• Remove QuarantinedTest attribute
• Remove code to drain initialization telemetry
• Change assert logic to read telemetry items in a loop.
• Read until TelemetryEventKeys.McpToolCall is matched.
• Then read until TelemetryEndpoints.TelemetryEndOperation is matched
• Remove disposing the sender and verifying there is no other telemetry

---

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.

[JamesNK]

@copilot Remove foundMcpToolCall from while test. Instead, break after foundMcpToolCall is set to true.

[Copilot]

Done in e527a40. Removed foundMcpToolCall from the while condition and added break after setting the flag to true.

[JamesNK]

@copilot Remove foundEndOperation from while test. Instead, break after foundEndOperation is set to true.

[Copilot]

Done in e527a40. Removed foundEndOperation from the while condition and added break after setting the flag to true.

[JamesNK]

@copilot Run CallService_McpTool_TelemetryRecorded test 50 times. It should pass consistently.

[Copilot]

@copilot Run CallService_McpTool_TelemetryRecorded test 50 times. It should pass consistently.

✅ Test passed all 50 runs consistently. Duration ranged from 1.066s to 1.149s per run. No failures observed.

[github-actions]

🚀 Dogfood this PR with:

⚠️ WARNING: Do not do this without first carefully reviewing the code of this PR to satisfy yourself it is safe.

curl -fsSL https://raw.githubusercontent.com/dotnet/aspire/main/eng/scripts/get-aspire-cli-pr.sh | bash -s -- 13374

Or

• Run remotely in PowerShell:

iex "& { $(irm https://raw.githubusercontent.com/dotnet/aspire/main/eng/scripts/get-aspire-cli-pr.ps1) } 13374"

[github-actions]

🚀 Dogfood this PR with:

⚠️ WARNING: Do not do this without first carefully reviewing the code of this PR to satisfy yourself it is safe.

curl -fsSL https://raw.githubusercontent.com/dotnet/aspire/main/eng/scripts/get-aspire-cli-pr.sh | bash -s -- 13374

Or

• Run remotely in PowerShell:

iex "& { $(irm https://raw.githubusercontent.com/dotnet/aspire/main/eng/scripts/get-aspire-cli-pr.ps1) } 13374"

[Copilot]

Pull request overview

This PR successfully removes the CallService_McpTool_TelemetryRecorded test from quarantine by refactoring it to use a more robust loop-based telemetry reading pattern. The changes eliminate flakiness caused by strict assumptions about telemetry ordering and initialization state.

Key Changes:

• Removed QuarantinedTest attribute to return test to regular CI pipeline
• Replaced direct channel reads with loop-based pattern that searches for specific telemetry events
• Simplified assertions to validate presence of key events rather than exact ordering and counts