[release/8.0] Disable flaky ETW manifest generation tests in release/8.0

[Copilot]

main PR #105069

Description

Test_EventSource_EtwManifestGeneration and Test_EventSource_EtwManifestGenerationRollover are consistently flaky on release/8.0 due to the old double-RemoteExecutor pattern and timing-based retry loops. These tests have been unconditionally disabled via [ActiveIssue] on main, release/9.0, and release/10.0, but the attribute was never backported to release/8.0.

Adds [ActiveIssue("https://github.com/dotnet/runtime/issues/88027")] to both tests in TestsManifestGeneration.Etw.cs:

[ConditionalFact(nameof(IsProcessElevatedAndNotWindowsNanoServerAndRemoteExecutorSupported))][ActiveIssue("https://github.com/dotnet/runtime/issues/88027")]publicvoidTest_EventSource_EtwManifestGeneration()[ConditionalFact(nameof(IsProcessElevatedAndNotWindowsNanoServerAndRemoteExecutorSupported))][ActiveIssue("https://github.com/dotnet/runtime/issues/88027")]publicvoidTest_EventSource_EtwManifestGenerationRollover()

This PR targets the release/8.0 branch. A clean branch (copilot/fix-etw-test-8.0) based directly on release/8.0 is available with just this 2-line fix.

Customer Impact

Ongoing CI noise on every release/8.0 servicing PR. No customer-facing behavioral change.

Regression

No — this is a pre-existing flaky test, not a regression introduced in the latest release.

Testing

No new tests. The fix is to skip tests that are already skipped on all newer branches. Verified [ActiveIssue] attribute matches the form used on main/release/9.0/release/10.0.

Risk

Minimal. Two-line change disabling already-flaky tests. No production code touched. Identical pattern already proven stable on three other branches.

Package authoring signed off?

N/A — test-only change, no NuGet packages affected.

Original prompt

---

This section details on the original issue you should resolve

<issue_title>BasicEventSourceTests.TestsManifestGeneration.Test_EventSource_EtwManifestGeneration is failing in CI</issue_title>
<issue_description>## Build Information
Build: https://dev.azure.com/dnceng-public/cbb18261-c48f-4abb-8651-8cdcb5474649/_build/results?buildId=1091361
Build error leg or test failing: BasicEventSourceTests.TestsManifestGeneration.Test_EventSource_EtwManifestGeneration
Pull request: #116534

Error Message

Fill the error message using step by step known issues guidance.

{
"ErrorMessage": ["BasicEventSourceTests.TestsManifestGeneration.Test_EventSource_EtwManifestGeneration", "Microsoft.DotNet.RemoteExecutor.RemoteExecutionException : Remote process failed with an unhandled exception."],
"ErrorPattern": "",
"BuildRetry": false,
"ExcludeConsoleLog": false
}

Known issue validation

Build: 🔎https://dev.azure.com/dnceng-public/public/_build/results?buildId=1091361
Error message validated:[BasicEventSourceTests.TestsManifestGeneration.Test_EventSource_EtwManifestGeneration Microsoft.DotNet.RemoteExecutor.RemoteExecutionException : Remote process failed with an unhandled exception.]
Result validation: ✅ Known issue matched with the provided build.
Validation performed at: 7/10/2025 9:17:28 PM UTC

Report

Build	Definition	Test	Pull Request
2929174	dotnet-runtime	BasicEventSourceTests.TestsManifestGeneration.Test_EventSource_EtwManifestGeneration	#59173
1331423	dotnet/runtime	BasicEventSourceTests.TestsManifestGeneration.Test_EventSource_EtwManifestGeneration	#125466
1331318	dotnet/runtime	BasicEventSourceTests.TestsManifestGeneration.Test_EventSource_EtwManifestGeneration	#124544
1320253	dotnet/runtime	BasicEventSourceTests.TestsManifestGeneration.Test_EventSource_EtwManifestGeneration	#124544
1314909	dotnet/runtime	BasicEventSourceTests.TestsManifestGeneration.Test_EventSource_EtwManifestGeneration	#123245

Summary

24-Hour Hit Count	7-Day Hit Count	1-Month Count
0	1	5

<agent_instructions>Disable both of these in the 8.0 branch (ONLY) using [ActiveIssue("https://github.com/dotnet/runtime/issues/88027")
They already have that disabling in 9, 10, and main.</agent_instructions>

Comments on the Issue (you are @copilot in this section)

@danmoseley > [!NOTE] > This analysis was generated with assistance from GitHub Copilot.

Root Cause: Failures are exclusively from release/8.0

All recent failures tracked by this issue (and the related #88027 for the Rollover variant) are coming from the release/8.0 branch. The PRs in the Known Build Error hit tables confirm this — they are all release/8.0 servicing PRs (dependency updates, staging merges, etc.).

Why release/8.0 is different

Between Nov 2023 and Jul 2024, six PRs landed on main to address this test:

PR	Date	What it did
#95389...

• FixesBasicEventSourceTests.TestsManifestGeneration.Test_EventSource_EtwManifestGeneration is failing in CI #117520

---

⚡ Quickly spin up Copilot coding agent tasks from anywhere on your macOS or Windows machine with Raycast.

[danmoseley]

@copilot you got mixed up. you should be targeting the 8.0 branch.

[Copilot]

@copilot you got mixed up. you should be targeting the 8.0 branch.

The code change itself is correct — I've prepared a clean branch copilot/fix-etw-test-8.0 based directly on release/8.0 with just the 2-line fix (commit c21f98dbb).

The issue is this PR was auto-created targeting main, and my tools don't have permission to change the PR base or force-push to a new branch. Could you either:

1. Change this PR's base from main → release/8.0 via the GitHub UI (Edit button next to the title), or
2. Use the clean branchcopilot/fix-etw-test-8.0 to open a new PR targeting release/8.0

The diff will then show only the 2 lines:

+ [ActiveIssue("https://github.com/dotnet/runtime/issues/88027")]
public void Test_EventSource_EtwManifestGeneration()
...
+ [ActiveIssue("https://github.com/dotnet/runtime/issues/88027")]
public void Test_EventSource_EtwManifestGenerationRollover()

[danmoseley]

@steveisok I believe I have to ask you to add the "Servicing-approved" label

[Copilot]

Pull request overview

Backports the test disabling used on newer branches to release/8.0 to stop persistent CI flakes in ETW manifest generation tests in System.Diagnostics.Tracing.

Changes:

• Adds [ActiveIssue("https://github.com/dotnet/runtime/issues/88027")] to Test_EventSource_EtwManifestGeneration.
• Adds [ActiveIssue("https://github.com/dotnet/runtime/issues/88027")] to Test_EventSource_EtwManifestGenerationRollover.

[danmoseley]

@steveisok

[danmoseley]

@vitek-karas don't know if we care much in a servicing branch, but this is not covered by build analysis:

[20:58:49] dbug: Executing command: 'D:\h\w\BE1E0A32\p\microsoft.dotnet.xharness.cli\8.0.0-prerelease.25516.2\runtimes\any\native\adb\windows\adb.exe -s 2A201JEGR00201 pull /storage/emulated/0/Android/data/net.dot.System.Runtime.Tests/files/Documents/testResults.xml "D:\\h\\w\\BE1E0A32\\t\\jixk5qsy.zgq"'
[20:58:49] dbug: Copied 1 files to D:\h\w\BE1E0A32\w\B5750983\uploads
[20:58:49] info: Instrumentation finished normally with exit code 1
[20:58:52] dbug: Executing command: 'D:\h\w\BE1E0A32\p\microsoft.dotnet.xharness.cli\8.0.0-prerelease.25516.2\runtimes\any\native\adb\windows\adb.exe -s 2A201JEGR00201 logcat -d '
[20:58:53] info: Wrote current ADB log to D:\h\w\BE1E0A32\w\B5750983\uploads\adb-logcat-net.dot.System.Runtime.Tests-net.dot.MonoRunner.log
[20:58:53] fail: Non-success instrumentation exit code: 1, expected: 0
[20:58:53] info: Attempting to remove apk 'net.dot.System.Runtime.Tests'..
[20:58:53] dbug: Executing command: 'D:\h\w\BE1E0A32\p\microsoft.dotnet.xharness.cli\8.0.0-prerelease.25516.2\runtimes\any\native\adb\windows\adb.exe -s 2A201JEGR00201 uninstall net.dot.System.Runtime.Tests'
[20:58:53] info: Successfully uninstalled net.dot.System.Runtime.Tests
[20:58:53] dbug: Saving diagnostics data to 'D:\h\w\BE1E0A32\w\B5750983\e/diagnostics.json'
XHarness exit code: 1 (TESTS_FAILED)
User command ended with 1

Is "Instrumentation finished normally with exit code 1" the right string for a known build issue?

Also, should failing unit test names be in the console output, as they are for desktoip runs? Example: https://helixr1107v0xdeko0k025g8.blob.core.windows.net/dotnet-runtime-refs-pull-126008-merge-d8ea4fe0b65d4196a0/System.Runtime.Tests/1/console.a7647259.log?helixlogtype=result

apparently NoBackwardTimeZones failed here, but it's not in console log. (I think that one is fixed by #125985)

[danmoseley]

/ba-g failures are unrelated