test: mark test-file-write-stream4 as flaky

[joyeecheung]

The flake has been detected since 2021-10-30 and still has been affecting recent PRs significantly (failing 8 PRs out of the last 100 CI run up until 2025-04-18). The underlying cause is still under investigation but in the meantime it's better to mark it as a known flake to keep the CI functional.

This flake affects at least Windows, Linux and macOS, according to https://github.com/nodejs/reliability/blob/main/reports/2025-04-18.md and likely affects all platforms. So the status is applied to all platforms in this PR.

Refs: nodejs/reliability#102
Refs: https://github.com/nodejs/reliability/blob/main/reports/2025-04-18.md
Refs: #56883
Refs: #54918

[lpinca]

If we have to do this, I would prefer to manually call the GC when the test finishes, and add a TODO comment to remove it when #54918 is fixed.

[lpinca]

I mean something like this

diff --git a/test/parallel/test-file-write-stream4.js b/test/parallel/test-file-write-stream4.js
index 6b3862fa714..1aa538f7781 100644
--- a/test/parallel/test-file-write-stream4.js
+++ b/test/parallel/test-file-write-stream4.js
@@ -1,3 +1,4 @@
+// Flags: --expose-gc
 'use strict';
 
 // Test that 'close' emits once and not twice when `emitClose: true` is set.
@@ -17,4 +18,8 @@ const fileWriteStream = fs.createWriteStream(filepath, {
 });
 
 fileReadStream.pipe(fileWriteStream);
-fileWriteStream.on('close', common.mustCall());
+fileWriteStream.on('close', common.mustCall(() => {
+  // TODO(lpinca): Remove the forced GC when
+  // https://github.com/nodejs/node/issues/54918 is fixed.
+  globalThis.gc({type: 'major'});
+}));

The advantage is that the test is still testing what it is supposed to test.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 90.24%. Comparing base (8e4e4df) to head (488dde4).
Report is 164 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff            @@
##             main   #57927    +/-   ##
========================================
  Coverage   90.24%   90.24%            
========================================
  Files         630      630            
  Lines      185705   185933   +228     
  Branches    36407    36450    +43     
========================================
+ Hits       167591   167803   +212     
+ Misses      11002    10985    -17     
- Partials     7112     7145    +33     

see 46 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[joyeecheung]

I remember that @jakecastelli mentioned that doing GC on shutdown doesn't actually make the flake go away?

[lpinca]

I remember that @jakecastelli mentioned that doing GC on shutdown doesn't actually make the flake go away?

I missed that. Can you find where?

[jakecastelli]

Hi folks, I think @joyeecheung meant a private slack direct message where I reached out to her when I found the manual gc from cc side didn't resolve the flakness of the test. But it was my fault where I triggered the gc at the wrong spot.

Later on I came up with this solution but we decided to not go with it. Hopefully this helps @lpinca

[joyeecheung]

This seems to have been gone after the V8 upgrade, or if not #58047 should make it less likely to appear.