ASAN check failed with, since it works in normal test, we may need to skip it in asan, like in

Fail log:

AssertionError [ERR_ASSERTION]: The expression evaluated to a falsy value:

  assert.ok(maxMem < 64 * 1024 * 1024)

    at process.<anonymous> (/home/runner/work/node/node/test/pummel/test-vm-memleak.js:61:10)
    at process.emit (events.js:326:22) {
  generatedMessage: true,
  code: 'ERR_ASSERTION',
  actual: false,
  expected: true,
  operator: '=='
}
Command: out/Release/node --max_old_space_size=32 /home/runner/work/node/node/test/pummel/test-vm-memleak.js

ASAN check failed with, since it works in normal test, we may need to skip it in asan, like in

Is this something you're seeing in CI or testing locally?

Same test failed on AIX. https://ci.nodejs.org/job/node-test-commit-aix/31507/nodes=aix71-ppc64/testReport/junit/(root)/test/pummel_test_vm_memleak/

on macOS, not sure flaky

Also assert.js:103
  throw new AssertionError(obj);
  ^

AssertionError [ERR_ASSERTION]: Expected values to be strictly equal:

false !== true

    at Timeout._onTimeout (/Users/runner/work/node/node/test/pummel/test-timers.js:55:10)
    at listOnTimeout (internal/timers.js:555:17)
    at processTimers (internal/timers.js:498:7) {
  generatedMessage: true,
  code: 'ERR_ASSERTION',
  actual: false,
  expected: true,
  operator: 'strictEqual'
}
Command: out/Release/node /Users/runner/work/node/node/test/pummel/test-timers.js

This is ready for reviews. @nodejs/testing

Also: @nodejs/build

It looks like the GitHub actions coverage-windows workflow is not happy?

It looks like the GitHub actions coverage-windows workflow is not happy?

Yes, that's correct. I talked with @bcoe about what might be going on and he had a suggestion of something to try but I haven't gotten around to it because I'm using what time I have right now to focus on the node-inspect migration to core. But I will get back to this hopefully in the next 24 hours. I've been working on this off-and-on since July and it is SOOOOOO close.....

CI: https://ci.nodejs.org/job/node-test-pull-request/37358/

🎉 ✨ 🔥 All GitHub Action and Jenkins tests are passing!

Landed in 6a1986d...2853b76

These tests take quite a long time on the Raspberry Pis (and sometimes fail with a timeout). Could we disable pummel tests on node-test-binary-arm-12+ ?

These tests take quite a long time on the Raspberry Pis (and sometimes fail with a timeout). Could we disable pummel tests on node-test-binary-arm-12+ ?

Looking at Pi 2 results in https://ci.nodejs.org/job/node-test-binary-arm-12+/10400/, the longest running tests (and how long it took them to run in seconds) are:

1. pummel/test-crypto-dh-hash-modp18 (700.975)
2. pummel/test-policy-integrity (696.201)
3. pummel/test-crypto-dh-hash (451.29)
4. pummel/test-dh-regr (311.406)
5. pummel/test-next-tick-infinite-calls (260.476)
6. parallel/test-webcrypto-derivebits-pbkdf2 (170.465)
7. parallel/test-heapsnapshot-near-heap-limit (136.232)
8. sequential/test-net-bytes-per-incoming-chunk-overhead (82.495)
9. pummel/test-fs-watch-system-limit (75.903)
10. parallel/test-heapsnapshot-near-heap-limit-bounded (72.233)

I think it makes sense to:

1. move those above that aren't already in pummel into pummel,
2. move the pummel tests that only take a second or two to run to somewhere outside of pummel (although we'll need to check them for why they are in pummel in the first place--for example, if it's disk usage, that might warrant staying in pummel even if they only take a few seconds to run)
3. set up pummel tests to skip on Raspberry Pi.

I'll get to work on this tonight and hopefully have a PR together quickly.