Repository navigation
Description
Build Information
Build: https://dev.azure.com/dnceng-public/public/_build/results?buildId=1545348
Build error leg or test failing: coreclr Pri1 Runtime Tests Run osx x64 checked — GC/Regressions/Github/Runtime_129681/Runtime_129681
Pipeline: runtime-coreclr outerloop (definition 108), branch main. The failure reproduces on both osx.15.amd64 machines behind the plain osx x64 checked leg (jobs e2334af3 / 1132d4aa) in source build 1545348 and persists into follow-up build 1545482 (job 31e71870).
This is a distinct failure mode from the Crossgen2 exit-255 signature tracked by #132037 on the R2R_CG2 osx x64 leg: here the plain (non-R2R) osx x64 leg runs the same test but the runtime host aborts during startup with coreclr_initialize failed - Error: 0x80004005. The Crossgen2 failed with exitcode: 255 line from #132037 does not appear in this log.
Error Details
Running test: GC/Regressions/Github/Runtime_129681/Runtime_129681/Runtime_129681.cmd
External disassembler is not available.
BEGIN: coreclr_initialize failed - Error: 0x80004005
END: coreclr_initialize failed - Error: 0x80004005
Return code: 1
...
Expected: 100
Actual: 255
END EXECUTION - FAILED
Test Harness Exitcode is : 1
Error Message
{
"ErrorMessage": [
"Running test: GC/Regressions/Github/Runtime_129681/Runtime_129681/Runtime_129681.cmd",
"BEGIN: coreclr_initialize failed - Error: 0x80004005"
],
"ErrorPattern": "",
"BuildRetry": false,
"ExcludeConsoleLog": false
}First build it occurred
Observed in source build 1545348 (2026-08-08) on both osx x64 machines and again in follow-up build 1545482 (2026-08-08) — at least two distinct builds.
Duplicate search
Searched open and closed issues/KBEs for Runtime_129681, coreclr_initialize failed, and 0x80004005: the only existing test-specific KBE is #132037, which matches a different signature (Crossgen2 failed with exitcode: 255) on the R2R_CG2 leg and does not match this plain-leg startup failure. Related: #132037. No open or merged PR references this signature.
Filed by ci-failure-scan (detection only). ci-failure-fix walks open [ci-scan] KBEs and either opens a small fix PR or comments here to loop in owners — it never disables the test.
Note
This issue was generated by the ci-failure-scan GitHub Copilot agentic workflow.
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.allowedlist in your workflow frontmatter:
network:
allowed:
- defaults
- "awmgmcpg"See Network Configuration for more information.
Generated by CI Outer-Loop Failure Scanner · 719.9 AIC · ⌖ 16.6 AIC · ⊞ 18.3K · ◷
Report
Summary
| 24-Hour Hit Count | 7-Day Hit Count | 1-Month Count |
|---|---|---|
| 0 | 0 | 0 |
Metadata
Metadata
Assignees
Labels
Type
Projects
- StatusShow more project fieldsNo status
Activity
VincentBu commentedon Aug 10, 2026
Failed in: runtime-coreclr ilasm 20260808.1
Failed tests:
Error message:
Stack trace:
dotnet-policy-service commentedon Aug 10, 2026
Tagging subscribers to this area: @anicka-net, @dotnet/gc
See info in area-owners.md if you want to be subscribed.
Disable Runtime_129681 CrossGen2 testing (#132126)
kkokosa commentedon Aug 21, 2026
#132199 stopped this failure by removing
DOTNET_GCStress=0xC, but it also removed an essential condition of the #129681 regression test: GCStress implicitly disabled concurrent GC. With concurrent GC enabled, the mark array exists and the bookkeeping-layout path fixed by #131947 is not exercised.The test should explicitly set
DOTNET_gcConcurrent=0instead of restoring GCStress. This preserves the intended regression coverage without causing the macOS startup failure./cc: @janvorli could you confirm?
janvorli commentedon Aug 21, 2026
This issue seems to be distinct one, not triggered by the GC stress setting. Here the coreclr_initialize fails with general error (0x80004005).
github-actions commentedon Sep 14, 2026
Workflow artifact: ci-fix
Artifact kind: handoff
Linked KBE: #132045
Note
AI/Copilot-generated triage note.
This Known Build Error has no producible automated code change (reason: non-code/infra runtime-startup failure), 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)
On the plain (non-R2R)
osx x64 checkedleg ofruntime-coreclr outerloop(def 108),GC/Regressions/Github/Runtime_129681never runs its body: the runtime host aborts during startup withBEGIN: coreclr_initialize failed - Error: 0x80004005(E_FAIL), the harness recordsActual: 255, and the work item fails. Because the failure is incoreclr_initializebefore managed code executes, it is a host/runtime startup problem on theosx.15.amd64queue, not a GC regression in the test itself. This is distinct from the Crossgen2 exit-255 signature tracked by #132037 (that log showsCrossgen2 failed with exitcode: 255, which is absent here).Evidence
Suggested reviewers / area contacts
area-GC-coreclr):@anicka-net,@dotnet/gcA fix likely involves diagnosing why
coreclr_initializereturnsE_FAILon theosx.15.amd64outerloop queue (host/runtime layout or environment on that image), since managed test code never starts.Structured data:
{ "artifact_kind": "handoff", "linked_kbe": 132045, "workflow_artifact": "ci-fix" }jeffschwMSFT commentedon Sep 15, 2026
removing blocking-clean-ci as it has not failed in 30 days
24-Hour Hit Count 7-Day Hit Count 1-Month Count
0 0 0
kotlarmilos commentedon Oct 2, 2026
Closing, fixed by #131947 and no hits in the last month.
Note
This comment was drafted with GitHub Copilot.