parallel.test-worker-nearheaplimit-deadlock is flaky

[anonrig]

Test

parallel.test-worker-nearheaplimit-deadlock

Platform

Windows

Console output

not ok 871 parallel/test-worker-nearheaplimit-deadlock
  ---
  duration_ms: 390.65800
  severity: fail
  exitcode: 134
  stack: |-
    
    <--- Last few GCs --->
    
    [4284:0C1A7400]       65 ms: Mark-Compact 1.5 (3.9) -> 1.5 (3.9) MB, 2.52 / 0.00 ms  (average mu = 0.660, current mu = 0.010) background allocation failure; GC in old space requested
    [4284:0C1A7400]       68 ms: Mark-Compact 1.5 (3.9) -> 1.5 (3.9) MB, 2.67 / 0.00 ms  (average mu = 0.536, current mu = 0.127) background allocation failure; GC in old space requested
    
    
    <--- JS stacktrace --->
    
    FATAL ERROR: LocalHeap: allocation failed Allocation failed - JavaScript heap out of memory
  ...

Build links

• https://ci.nodejs.org/job/node-test-binary-windows-js-suites/23884/RUN_SUBSET=1,nodes=win2016-COMPILED_BY-vs2022-x86/testReport/junit/(root)/parallel/test_worker_nearheaplimit_deadlock/

Additional information

No response

[StefanStojanovic]

@anonrig, just to check - you reopened this because the PR didn't fix it, or as a precaution measurement.

Edit: I see you've encountered it again in #50261 and I know why. In my original PR I put $arch==x86 although it should be $arch==ia32. I've opened a new PR fixing this.

[targos]

Setting the test as flaky doesn't fix the flakyness, so the issue should stay open anyway