Issue #7904 has been updated by ngoto (Naohisa Goto).


Failure of the same two tests were observed also on Solaris10 before r39374.
After r39375, failure of test_spawn_too_long_path also vanished on Solaris10.

On Solaris10, with r39376, 

 TestProcess#test_aspawn_too_long_path = sh: a: not found
 sh: a: not found
 sh: a: not found
 sh: a: not found
(snip many lines)
 sh: a: not found
 sh: a: not found
 sh: a: not found
 sh: fork failed - too many processes
 536.65 s = F

and

  13) Failure:
 test_aspawn_too_long_path(TestProcess) [/XXXXXXX/test/ruby/test_process.rb:1388]:
 [ruby-core:34833].
 [Errno::ENOENT, Errno::E2BIG] expected but nothing was raised.

----------------------------------------
Bug #7904: test_spawn_too_long_path(TestProcess) fails on PPC64
https://bugs.ruby-lang.org/issues/7904#change-36795

Author: vo.x (Vit Ondruch)
Status: Feedback
Priority: Normal
Assignee: nobu (Nobuyoshi Nakada)
Category: core
Target version: 2.1.0
ruby -v: ruby 2.0.0dev (2013-02-21 trunk 39357) [powerpc64-linux]


=begin
I observe following error on PPC64:
 7) Failure:
 test_spawn_too_long_path(TestProcess) [/builddir/build/BUILD/ruby-2.0.0-r39357/test/ruby/test_process.rb:1383]:
 [Errno::ENOENT, Errno::E2BIG, nil] expected but nothing was raised.
 8) Failure:
 test_aspawn_too_long_path(TestProcess) [/builddir/build/BUILD/ruby-2.0.0-r39357/test/ruby/test_process.rb:1388]:
 [Errno::ENOENT, Errno::E2BIG, nil] expected but nothing was raised.

I am quite sure, that there was no such test error testing with rev38848. Not sure if that might be related to 38934.
=end



-- 
http://bugs.ruby-lang.org/