[ci-scan] Test failure: baseservices/threading/regressions/2164/foreground-shutdown fails Expected 100 Actual 101 on R2R

[github-actions]

Build Information

Build: https://dev.azure.com/dnceng-public/public/_build/results?buildId=1526910
Pipeline: runtime-coreclr r2r-extra (definition 114), branch main
Failing leg: R2R windows arm64 checked (--method-layout:random R2R composite)
Failing test: baseservices/threading/regressions/2164/foreground-shutdown/foreground-shutdown.cmd

Error Message

Under the R2R windows-arm64 outer-loop leg the test exits with 101 instead of the expected 100:

> C:\...\regressions\2164\foreground-shutdown\foreground-shutdown.cmd
Xunit.Sdk.EqualException: Assert.Equal() Failure: Values differ
Expected: 100
Actual:   101
   at Xunit.Assert.Equal[T](T expected, T actual, IEqualityComparer`1 comparer)
   at __GeneratedMainWrapper.Main()
END EXECUTION - FAILED
Failed test: baseservices/threading/regressions/2164/foreground-shutdown/foreground-shutdown.cmd

{
  "ErrorMessage": "Failed test: baseservices/threading/regressions/2164/foreground-shutdown/foreground-shutdown.cmd",
  "BuildRetry": false,
  "ExcludeConsoleLog": false
}

Occurrences

Stable across the scanned window: source build 1526910 and follow-up build 1527456 both reproduce this signature on the R2R windows arm64 checked leg.

Related: #131403 (area-System.Threading tracker for the same test, currently observed only on the optional jitstressregs leg and not labeled Known Build Error, so it does not dedupe required-gate PR CI). This KBE covers the required-gate r2r-extra occurrence.

---

Filed by the ci-failure-scan workflow (detection only). Mitigation — small fix PRs and owner hand-off — is handled by the companion ci-failure-fix workflow from the open [ci-scan] KBEs.

Note

This issue was generated with the assistance of GitHub Copilot (AI). The failure data was collected from public dnceng CI; please verify before acting.

Warning

Firewall blocked 1 domain

The following domain was blocked by the firewall during workflow execution:

• awmgmcpg

To allow these domains, add them to the network.allowed list in your workflow frontmatter:

network:
  allowed:
    - defaults
    - "awmgmcpg"

See Network Configuration for more information.

Generated by CI Outer-Loop Failure Scanner · 688.4 AIC · ⌖ 16.8 AIC · ⊞ 18.6K · ◷

Report

Build	Repository	Test	Pull Request
1531854	dotnet/runtime	baseservices/threading/regressions/2164/foreground-shutdown/foreground-shutdown.cmd
1530122	dotnet/runtime	baseservices/threading/regressions/2164/foreground-shutdown/foreground-shutdown.cmd
1528585	dotnet/runtime	baseservices/threading/regressions/2164/foreground-shutdown/foreground-shutdown.cmd

Summary

24-Hour Hit Count	7-Day Hit Count	1-Month Count
1	3	3

[dotnet-policy-service]

Tagging subscribers to this area: @JulieLeeMSFT, @VSadov
See info in area-owners.md if you want to be subscribed.

[github-actions]

Workflow artifact: ci-fix
Artifact kind: handoff
Linked KBE: #131418

Note

AI/Copilot-generated triage note.

This Known Build Error has no producible automated code change (reason: JIT/codegen stress — the failure reproduces only on the R2R --method-layout:random composite leg, an outer-loop codegen configuration I must not auto-patch), so I could not open even a best-effort PR. Looping in owners so it can be fixed forward rather than muted.

Root cause (best analysis)

baseservices/threading/regressions/2164/foreground-shutdown/foreground-shutdown.cmd exits 101 instead of the expected 100 on the R2R windows arm64 checked leg (runtime-coreclr r2r-extra, def 114, --method-layout:random). Test 2164 is a foreground-thread-during-shutdown regression: the process is expected to complete shutdown with a specific ordering/exit result. An off-by-one exit code (101 = one assertion/return short of the expected 100 pass sentinel) under a randomized R2R method layout points to a shutdown-timing/ordering sensitivity that only surfaces with that codegen configuration, not a product bug in the test's own source. #131403 tracks the same test on the optional jitstressregs leg. Because the trigger is a codegen-stress layout, no safe mechanical diff is producible.

Evidence

• Failing build: https://dev.azure.com/dnceng-public/public/_build/results?buildId=1526910
• Also reproduces in follow-up build 1527456
• Possible related PR: none identified with sufficient confidence; related tracker Test failure: baseservices/threading/regressions/2164/foreground-shutdown/foreground-shutdown.cmd #131403

Suggested reviewers / area contacts

• Area owners (area-System.Threading): @JulieLeeMSFT, @vsadov

A fix would likely involve determining whether foreground-thread shutdown ordering is racy under randomized R2R method layout, or whether the test's shutdown synchronization needs hardening for that leg.

Warning

Firewall blocked 1 domain

The following domain was blocked by the firewall during workflow execution:

• awmgmcpg

To allow these domains, add them to the network.allowed list in your workflow frontmatter:

network:
  allowed:
    - defaults
    - "awmgmcpg"

See Network Configuration for more information.

Generated by CI Outer-Loop Failure Fixer · 287.5 AIC · ⌖ 18.4 AIC · ⊞ 17.4K · ◷