test: test-tick-processor failed

[ghaiklor]

• Version: v6.0.0-pre
• Platform: Darwin Kernel Version 15.4.0
• Subsystem: test

When running make test, I got the following:

/usr/bin/python tools/test.py --mode=release message parallel sequential -J
=== release test-tick-processor ===
Path: parallel/test-tick-processor
assert.js:89
  throw new assert.AssertionError({
  ^
AssertionError: false == true
    at runTest (/Users/ghaiklor/Library/Projects/node/test/parallel/test-tick-processor.js:58:3)
    at Object.<anonymous> (/Users/ghaiklor/Library/Projects/node/test/parallel/test-tick-processor.js:38:1)
    at Module._compile (module.js:417:34)
    at Object.Module._extensions..js (module.js:426:10)
    at Module.load (module.js:357:32)
    at Function.Module._load (module.js:314:12)
    at Function.Module.runMain (module.js:451:10)
    at startup (node.js:155:18)
    at node.js:433:3
Command: out/Release/node /Users/ghaiklor/Library/Projects/node/test/parallel/test-tick-processor.js
[00:55|% 100|+ 1062|-   1]: Done
make: *** [test] Error 1

Assertion error in this code:

runTest(/RunInDebugContext/,
  `function f() {
     require(\'vm\').runInDebugContext(\'Debug\');
     setImmediate(function() { f(); });
   };
   setTimeout(function() { process.exit(0); }, 2000);
   f();`);

[thefourtheye]

I wonder why this is not caught in CI.

[Trott]

I wonder why this is not caught in CI.

The test already has open issues for it being flaky on two other platforms. It is a non-deterministic test that will vary a lot based on how powerful your computer is, so that's probably what's going on here.

[balupton]

Getting the same here on OSX 10.11.4 (15E65)

14:39:14:balupton@balmac.local:/Users/balupton/Projects/uni/node:(HEAD detached at v5.9.1)
$ make test
/Applications/Xcode.app/Contents/Developer/usr/bin/make -C out BUILDTYPE=Release V=1
make[1]: Nothing to be done for `all'.
ln -fs out/Release/node node
Running main() from gtest_main.cc
[==========] Running 1 test from 1 test case.
[----------] Global test environment set-up.
[----------] 1 test from UtilTest
[ RUN      ] UtilTest.ListHead
[       OK ] UtilTest.ListHead (0 ms)
[----------] 1 test from UtilTest (0 ms total)

[----------] Global test environment tear-down
[==========] 1 test from 1 test case ran. (0 ms total)
[  PASSED  ] 1 test.
/usr/local/opt/python/bin/python2.7 tools/test.py --mode=release message parallel sequential -J
=== release test-tick-processor ===                                            
Path: parallel/test-tick-processor
assert.js:89
  throw new assert.AssertionError({
  ^
AssertionError: false == true
    at runTest (/Users/balupton/Projects/uni/node/test/parallel/test-tick-processor.js:58:3)
    at Object.<anonymous> (/Users/balupton/Projects/uni/node/test/parallel/test-tick-processor.js:38:1)
    at Module._compile (module.js:413:34)
    at Object.Module._extensions..js (module.js:422:10)
    at Module.load (module.js:357:32)
    at Function.Module._load (module.js:314:12)
    at Function.Module.runMain (module.js:447:10)
    at startup (node.js:142:18)
    at node.js:939:3
Command: out/Release/node /Users/balupton/Projects/uni/node/test/parallel/test-tick-processor.js
[00:58|% 100|+ 1046|-   1]: Done                                               
make: *** [test] Error 1

[balupton]

There is also #4427 for the same test file - not sure if it is related.

[Trott]

/cc @nodejs/v8

[indutny]

I got this too

[indutny]

I believe this is a culprit:

$ foo() { nm "$@" | (c++filt -p -i || cat) }; foo $@ -n -f /Users/findutnyy/Code/indutny/node/out/Release/node
nm: for the -format option: Cannot find option named '/Users/findutnyy/Code/indutny/node/out/Release/node'!

Looking...

[indutny]

Fix:

diff --git a/deps/v8/tools/tickprocessor.js b/deps/v8/tools/tickprocessor.js
index 600d2ee..ccdb11e 100644
--- a/deps/v8/tools/tickprocessor.js
+++ b/deps/v8/tools/tickprocessor.js
@@ -757,8 +757,10 @@ inherits(MacCppEntriesProvider, UnixCppEntriesProvider);
 MacCppEntriesProvider.prototype.loadSymbols = function(libName) {
   this.parsePos = 0;
   libName = this.targetRootFS + libName;
+
+  // It seems that with OS X 10.11.4, `nm` thinks that `-f` is a format option
   try {
-    this.symbols = [os.system(this.nmExec, ['-n', '-f', libName], -1, -1), ''];
+    this.symbols = [os.system(this.nmExec, ['-n', libName], -1, -1), ''];
   } catch (e) {
     // If the library cannot be found on this system let's not panic.
     this.symbols = '';

Going to submit it to V8 team.

[indutny]

CL: https://codereview.chromium.org/1840633002/

[matthewloring]

@indutny The test still passes on my end with your fix applied.

[ghaiklor]

@matthewloring @indutny on my end too. Issue can be closed.