outerloop: System.IO.Tests on linux-musl-x64 AOT running into SIGKILL

[am11]

From main branch: https://dev.azure.com/dnceng-public/public/_build/results?buildId=689281&view=logs&j=20e6f673-0b0c-5c85-377c-5d941d4c8ee3&t=5801e1cc-59c5-5e47-8484-142b970afd58

console logs: https://helixre107v0xdcypoyl9e7f.blob.core.windows.net/dotnet-runtime-refs-heads-main-9fc74348a0df49998b/System.IO.Tests/1/console.69ead1c7.log?helixlogtype=result

[SKIP] System.IO.Tests.TextWriterTests.CreateBroadcasting_AllMethodsOverridden
./RunTests.sh: line 179:    21 Killed                  ./System.IO.Tests -notrait category=IgnoreForCI -notrait category=OuterLoop -notrait category=failing -xml testResults.xml $RSP_FILE
/root/helix/work/workitem/e
----- end Tue May 28 10:50:09 UTC 2024 ----- exit code 137 ----------------------------------------------------------
exit code 137 means SIGKILL Killed either due to out of memory/resources (see /var/log/messages) or by explicit kill.
ulimit -c value: unlimited

[dotnet-policy-service]

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

[MichalStrehovsky]

I've seen this before. It's a sigkill so there's no dump. And downloading the test binary from the CI and running it on Alpine locally works fine.

[am11]

Maybe we can turn it into a tracking issue with Known Build Error label, so Build Analysis doesn't fail in PRs?

[MichalStrehovsky]

Build analysis doesn't look at runtime-nativeaot-outerloop pipeline.

[am11]

Ah ok. I saw #98964 (outerloop, tracking and with known-build-error label), so I thought that's the way. Since this is failing in CI, lets keep it opened so others' PR won't get blocked?

[MichalStrehovsky]

Yep, I agree it should be open. Just don't know what to do with it :).