Issue #6825 has been reported by xentronium (Mark A). ---------------------------------------- Bug #6825: forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9.3-p194 https://bugs.ruby-lang.org/issues/6825 Author: xentronium (Mark A) Status: Open Priority: Normal Assignee: Category: Target version: ruby -v: ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] https://gist.github.com/47e48301aea114e7b1d3 here is the gist with required setup to reproduce bug. Also crash log and stdout. It seems that forking is essential for this setup to crash. Also, if you use database connection in some way prior to forking, it might not crash (however, with more complex code it still does). ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] OS X 10.6.8 hostinfo output: Mach kernel version: Darwin Kernel Version 10.8.0: Tue Jun 7 16:33:36 PDT 2011; root:xnu-1504.15.3~1/RELEASE_I386 Kernel configured for up to 4 processors. 2 processors are physically available. 4 processors are logically available. Processor type: i486 (Intel 80486) Processors active: 0 1 2 3 Primary memory available: 8.00 gigabytes Default processor set: 88 tasks, 627 threads, 4 processors Load average: 0.55, Mach factor: 3.43 compiled with gcc version: i686-apple-darwin10-gcc-4.2.1 (GCC) 4.2.1 (Apple Inc. build 5666) (dot 3)
[ruby-trunk - Bug #6825][Open] forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X
on 2012-08-02 13:39
[ruby-trunk - Bug #6825] forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9
on 2012-08-02 13:40
Issue #6825 has been updated by xentronium (Mark A). Also notably ruby 1.8.7 does not crash. ---------------------------------------- Bug #6825: forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9.3-p194 https://bugs.ruby-lang.org/issues/6825#change-28593 Author: xentronium (Mark A) Status: Open Priority: Normal Assignee: Category: Target version: ruby -v: ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] https://gist.github.com/47e48301aea114e7b1d3 here is the gist with required setup to reproduce bug. Also crash log and stdout. It seems that forking is essential for this setup to crash. Also, if you use database connection in some way prior to forking, it might not crash (however, with more complex code it still does). ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] OS X 10.6.8 hostinfo output: Mach kernel version: Darwin Kernel Version 10.8.0: Tue Jun 7 16:33:36 PDT 2011; root:xnu-1504.15.3~1/RELEASE_I386 Kernel configured for up to 4 processors. 2 processors are physically available. 4 processors are logically available. Processor type: i486 (Intel 80486) Processors active: 0 1 2 3 Primary memory available: 8.00 gigabytes Default processor set: 88 tasks, 627 threads, 4 processors Load average: 0.55, Mach factor: 3.43 compiled with gcc version: i686-apple-darwin10-gcc-4.2.1 (GCC) 4.2.1 (Apple Inc. build 5666) (dot 3)
[ruby-trunk - Bug #6825] forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9
on 2012-08-02 13:52
Issue #6825 has been updated by xentronium (Mark A). Ubuntu 11.04 doesn't crash with 1.9 ruby either. ---------------------------------------- Bug #6825: forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9.3-p194 https://bugs.ruby-lang.org/issues/6825#change-28595 Author: xentronium (Mark A) Status: Open Priority: Normal Assignee: Category: Target version: ruby -v: ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] https://gist.github.com/47e48301aea114e7b1d3 here is the gist with required setup to reproduce bug. Also crash log and stdout. It seems that forking is essential for this setup to crash. Also, if you use database connection in some way prior to forking, it might not crash (however, with more complex code it still does). ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] OS X 10.6.8 hostinfo output: Mach kernel version: Darwin Kernel Version 10.8.0: Tue Jun 7 16:33:36 PDT 2011; root:xnu-1504.15.3~1/RELEASE_I386 Kernel configured for up to 4 processors. 2 processors are physically available. 4 processors are logically available. Processor type: i486 (Intel 80486) Processors active: 0 1 2 3 Primary memory available: 8.00 gigabytes Default processor set: 88 tasks, 627 threads, 4 processors Load average: 0.55, Mach factor: 3.43 compiled with gcc version: i686-apple-darwin10-gcc-4.2.1 (GCC) 4.2.1 (Apple Inc. build 5666) (dot 3)
[ruby-trunk - Bug #6825] forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9
on 2012-08-02 14:00
Issue #6825 has been updated by xentronium (Mark A). Ruby 1.9.2 doesn't crash either. ---------------------------------------- Bug #6825: forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9.3-p194 https://bugs.ruby-lang.org/issues/6825#change-28596 Author: xentronium (Mark A) Status: Open Priority: Normal Assignee: Category: Target version: ruby -v: ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] https://gist.github.com/47e48301aea114e7b1d3 here is the gist with required setup to reproduce bug. Also crash log and stdout. It seems that forking is essential for this setup to crash. Also, if you use database connection in some way prior to forking, it might not crash (however, with more complex code it still does). ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] OS X 10.6.8 hostinfo output: Mach kernel version: Darwin Kernel Version 10.8.0: Tue Jun 7 16:33:36 PDT 2011; root:xnu-1504.15.3~1/RELEASE_I386 Kernel configured for up to 4 processors. 2 processors are physically available. 4 processors are logically available. Processor type: i486 (Intel 80486) Processors active: 0 1 2 3 Primary memory available: 8.00 gigabytes Default processor set: 88 tasks, 627 threads, 4 processors Load average: 0.55, Mach factor: 3.43 compiled with gcc version: i686-apple-darwin10-gcc-4.2.1 (GCC) 4.2.1 (Apple Inc. build 5666) (dot 3)
[ruby-trunk - Bug #6825] forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9
on 2012-08-02 14:32
Issue #6825 has been updated by xentronium (Mark A). ruby 2.0.0dev (2012-08-02 trunk 36596) [x86_64-darwin10.8.0] also does not crash. ---------------------------------------- Bug #6825: forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9.3-p194 https://bugs.ruby-lang.org/issues/6825#change-28597 Author: xentronium (Mark A) Status: Open Priority: Normal Assignee: Category: Target version: ruby -v: ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] https://gist.github.com/47e48301aea114e7b1d3 here is the gist with required setup to reproduce bug. Also crash log and stdout. It seems that forking is essential for this setup to crash. Also, if you use database connection in some way prior to forking, it might not crash (however, with more complex code it still does). ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] OS X 10.6.8 hostinfo output: Mach kernel version: Darwin Kernel Version 10.8.0: Tue Jun 7 16:33:36 PDT 2011; root:xnu-1504.15.3~1/RELEASE_I386 Kernel configured for up to 4 processors. 2 processors are physically available. 4 processors are logically available. Processor type: i486 (Intel 80486) Processors active: 0 1 2 3 Primary memory available: 8.00 gigabytes Default processor set: 88 tasks, 627 threads, 4 processors Load average: 0.55, Mach factor: 3.43 compiled with gcc version: i686-apple-darwin10-gcc-4.2.1 (GCC) 4.2.1 (Apple Inc. build 5666) (dot 3)
[ruby-trunk - Bug #6825] forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9
on 2012-08-02 15:29
Issue #6825 has been updated by xentronium (Mark A). OS X 10.8 seems to be unaffected. ---------------------------------------- Bug #6825: forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9.3-p194 https://bugs.ruby-lang.org/issues/6825#change-28598 Author: xentronium (Mark A) Status: Open Priority: Normal Assignee: Category: Target version: ruby -v: ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] https://gist.github.com/47e48301aea114e7b1d3 here is the gist with required setup to reproduce bug. Also crash log and stdout. It seems that forking is essential for this setup to crash. Also, if you use database connection in some way prior to forking, it might not crash (however, with more complex code it still does). ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] OS X 10.6.8 hostinfo output: Mach kernel version: Darwin Kernel Version 10.8.0: Tue Jun 7 16:33:36 PDT 2011; root:xnu-1504.15.3~1/RELEASE_I386 Kernel configured for up to 4 processors. 2 processors are physically available. 4 processors are logically available. Processor type: i486 (Intel 80486) Processors active: 0 1 2 3 Primary memory available: 8.00 gigabytes Default processor set: 88 tasks, 627 threads, 4 processors Load average: 0.55, Mach factor: 3.43 compiled with gcc version: i686-apple-darwin10-gcc-4.2.1 (GCC) 4.2.1 (Apple Inc. build 5666) (dot 3)
[ruby-trunk - Bug #6825] forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9
on 2012-08-02 16:01
Issue #6825 has been updated by xentronium (Mark A). Confirmed with another 10.6 / 1.9.3-p194. ---------------------------------------- Bug #6825: forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9.3-p194 https://bugs.ruby-lang.org/issues/6825#change-28601 Author: xentronium (Mark A) Status: Open Priority: Normal Assignee: Category: Target version: ruby -v: ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] https://gist.github.com/47e48301aea114e7b1d3 here is the gist with required setup to reproduce bug. Also crash log and stdout. It seems that forking is essential for this setup to crash. Also, if you use database connection in some way prior to forking, it might not crash (however, with more complex code it still does). ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] OS X 10.6.8 hostinfo output: Mach kernel version: Darwin Kernel Version 10.8.0: Tue Jun 7 16:33:36 PDT 2011; root:xnu-1504.15.3~1/RELEASE_I386 Kernel configured for up to 4 processors. 2 processors are physically available. 4 processors are logically available. Processor type: i486 (Intel 80486) Processors active: 0 1 2 3 Primary memory available: 8.00 gigabytes Default processor set: 88 tasks, 627 threads, 4 processors Load average: 0.55, Mach factor: 3.43 compiled with gcc version: i686-apple-darwin10-gcc-4.2.1 (GCC) 4.2.1 (Apple Inc. build 5666) (dot 3)
[ruby-trunk - Bug #6825] forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9
on 2012-08-02 19:08
Issue #6825 has been updated by xentronium (Mark A). After some more fiddling it looks like it's mysql2 problem, not ruby's. Will duplicate issue there. ---------------------------------------- Bug #6825: forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9.3-p194 https://bugs.ruby-lang.org/issues/6825#change-28603 Author: xentronium (Mark A) Status: Open Priority: Normal Assignee: Category: Target version: ruby -v: ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] https://gist.github.com/47e48301aea114e7b1d3 here is the gist with required setup to reproduce bug. Also crash log and stdout. It seems that forking is essential for this setup to crash. Also, if you use database connection in some way prior to forking, it might not crash (however, with more complex code it still does). ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] OS X 10.6.8 hostinfo output: Mach kernel version: Darwin Kernel Version 10.8.0: Tue Jun 7 16:33:36 PDT 2011; root:xnu-1504.15.3~1/RELEASE_I386 Kernel configured for up to 4 processors. 2 processors are physically available. 4 processors are logically available. Processor type: i486 (Intel 80486) Processors active: 0 1 2 3 Primary memory available: 8.00 gigabytes Default processor set: 88 tasks, 627 threads, 4 processors Load average: 0.55, Mach factor: 3.43 compiled with gcc version: i686-apple-darwin10-gcc-4.2.1 (GCC) 4.2.1 (Apple Inc. build 5666) (dot 3)
on 2012-08-02 20:23
Issue #6825 has been updated by drbrain (Eric Hodel). Status changed from Open to Feedback ---------------------------------------- Bug #6825: forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9.3-p194 https://bugs.ruby-lang.org/issues/6825#change-28605 Author: xentronium (Mark A) Status: Feedback Priority: Normal Assignee: Category: Target version: ruby -v: ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] https://gist.github.com/47e48301aea114e7b1d3 here is the gist with required setup to reproduce bug. Also crash log and stdout. It seems that forking is essential for this setup to crash. Also, if you use database connection in some way prior to forking, it might not crash (however, with more complex code it still does). ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] OS X 10.6.8 hostinfo output: Mach kernel version: Darwin Kernel Version 10.8.0: Tue Jun 7 16:33:36 PDT 2011; root:xnu-1504.15.3~1/RELEASE_I386 Kernel configured for up to 4 processors. 2 processors are physically available. 4 processors are logically available. Processor type: i486 (Intel 80486) Processors active: 0 1 2 3 Primary memory available: 8.00 gigabytes Default processor set: 88 tasks, 627 threads, 4 processors Load average: 0.55, Mach factor: 3.43 compiled with gcc version: i686-apple-darwin10-gcc-4.2.1 (GCC) 4.2.1 (Apple Inc. build 5666) (dot 3)
[ruby-trunk - Bug #6825] forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9
on 2012-08-02 22:08
Issue #6825 has been updated by xentronium (Mark A). Update: I greatly simplified my test-case. https://gist.github.com/47e48301aea114e7b1d3 If I remove sleeping threads on line 31 or require 'active_record' on line 1, bug stops reproducing. Returning back to ruby-lang, as there is no mysql2 there anymore. Crash is still with the same error. Ideas, suggestions? ---------------------------------------- Bug #6825: forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9.3-p194 https://bugs.ruby-lang.org/issues/6825#change-28607 Author: xentronium (Mark A) Status: Feedback Priority: Normal Assignee: Category: Target version: ruby -v: ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] https://gist.github.com/47e48301aea114e7b1d3 here is the gist with required setup to reproduce bug. Also crash log and stdout. It seems that forking is essential for this setup to crash. Also, if you use database connection in some way prior to forking, it might not crash (however, with more complex code it still does). ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] OS X 10.6.8 hostinfo output: Mach kernel version: Darwin Kernel Version 10.8.0: Tue Jun 7 16:33:36 PDT 2011; root:xnu-1504.15.3~1/RELEASE_I386 Kernel configured for up to 4 processors. 2 processors are physically available. 4 processors are logically available. Processor type: i486 (Intel 80486) Processors active: 0 1 2 3 Primary memory available: 8.00 gigabytes Default processor set: 88 tasks, 627 threads, 4 processors Load average: 0.55, Mach factor: 3.43 compiled with gcc version: i686-apple-darwin10-gcc-4.2.1 (GCC) 4.2.1 (Apple Inc. build 5666) (dot 3)
[ruby-trunk - Bug #6825] forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9
on 2012-08-02 22:10
Issue #6825 has been updated by xentronium (Mark A). Oh, it also reproduced on ruby 2.0.0.dev from current git, so I guess it is still not fixed. ---------------------------------------- Bug #6825: forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9.3-p194 https://bugs.ruby-lang.org/issues/6825#change-28608 Author: xentronium (Mark A) Status: Feedback Priority: Normal Assignee: Category: Target version: ruby -v: ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] https://gist.github.com/47e48301aea114e7b1d3 here is the gist with required setup to reproduce bug. Also crash log and stdout. It seems that forking is essential for this setup to crash. Also, if you use database connection in some way prior to forking, it might not crash (however, with more complex code it still does). ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] OS X 10.6.8 hostinfo output: Mach kernel version: Darwin Kernel Version 10.8.0: Tue Jun 7 16:33:36 PDT 2011; root:xnu-1504.15.3~1/RELEASE_I386 Kernel configured for up to 4 processors. 2 processors are physically available. 4 processors are logically available. Processor type: i486 (Intel 80486) Processors active: 0 1 2 3 Primary memory available: 8.00 gigabytes Default processor set: 88 tasks, 627 threads, 4 processors Load average: 0.55, Mach factor: 3.43 compiled with gcc version: i686-apple-darwin10-gcc-4.2.1 (GCC) 4.2.1 (Apple Inc. build 5666) (dot 3)
[ruby-trunk - Bug #6825] forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9
on 2012-08-02 22:23
Issue #6825 has been updated by xentronium (Mark A). Another update: I opened up a file `active_record.rb` inside installed activerecord gem and completely commented it out (so that even ActiveRecord is not defined after require 'active_record'). Still crashes. I guess that takes care of gems and everything, so the problem should be between ruby, rubygems and standard library. Line `require 'active_record'` is still required for whole setup to crash for some reason. Hope that gives you some idea. ---------------------------------------- Bug #6825: forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9.3-p194 https://bugs.ruby-lang.org/issues/6825#change-28610 Author: xentronium (Mark A) Status: Feedback Priority: Normal Assignee: Category: Target version: ruby -v: ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] https://gist.github.com/47e48301aea114e7b1d3 here is the gist with required setup to reproduce bug. Also crash log and stdout. It seems that forking is essential for this setup to crash. Also, if you use database connection in some way prior to forking, it might not crash (however, with more complex code it still does). ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] OS X 10.6.8 hostinfo output: Mach kernel version: Darwin Kernel Version 10.8.0: Tue Jun 7 16:33:36 PDT 2011; root:xnu-1504.15.3~1/RELEASE_I386 Kernel configured for up to 4 processors. 2 processors are physically available. 4 processors are logically available. Processor type: i486 (Intel 80486) Processors active: 0 1 2 3 Primary memory available: 8.00 gigabytes Default processor set: 88 tasks, 627 threads, 4 processors Load average: 0.55, Mach factor: 3.43 compiled with gcc version: i686-apple-darwin10-gcc-4.2.1 (GCC) 4.2.1 (Apple Inc. build 5666) (dot 3)
[ruby-trunk - Bug #6825] forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9
on 2012-08-02 22:36
Issue #6825 has been updated by drbrain (Eric Hodel).
=begin
I can't reproduce on OS X 10.8 ruby 2.0.0dev (2012-08-03 trunk 36602)
[x86_64-darwin12.0.0]
I modified your script to remove require 'active_record' and altered the
main thread to sleep forever. This ensures that mysql and other C
extensions are not loaded. It ran for over two minutes without
problems.
Can you reproduce this with require 'mysql' and not active_record?
Can you show the console output with your modified active_record.rb (the
loaded features section is of particular interest).
Here is what I used:
require 'net/http'
Thread.abort_on_exception = true
class Worker
def initialize
@tasks = []
work
end
def work
Thread.new do
loop do
task = nil
task = @tasks.shift if @tasks.length > 0
task.call if task
sleep(0.25)
end
end
end
def schedule(&block)
@tasks << block
end
end
pid = fork do
class TestLoop
def initialize
@worker = Worker.new
(1..10).map { Thread.new { loop { sleep(0.5) } } }
end
def run
loop do
@worker.schedule { puts Net::HTTP.get("github.com",
"/").length }
sleep(0.25)
end
end
end
TestLoop.new.run
end
sleep
(I don't have mysql installed to check.)
=end
----------------------------------------
Bug #6825: forking and pthread_cond_timedwait: Invalid argument (EINVAL)
on OS X / 1.9.3-p194
https://bugs.ruby-lang.org/issues/6825#change-28611
Author: xentronium (Mark A)
Status: Feedback
Priority: Normal
Assignee:
Category:
Target version:
ruby -v: ruby 1.9.3p194 (2012-04-20 revision 35410)
[x86_64-darwin10.8.0]
https://gist.github.com/47e48301aea114e7b1d3 here is the gist with
required setup to reproduce bug. Also crash log and stdout.
It seems that forking is essential for this setup to crash. Also, if you
use database connection in some way prior to forking, it might not crash
(however, with more complex code it still does).
ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0]
OS X 10.6.8
hostinfo output:
Mach kernel version:
Darwin Kernel Version 10.8.0: Tue Jun 7 16:33:36 PDT 2011;
root:xnu-1504.15.3~1/RELEASE_I386
Kernel configured for up to 4 processors.
2 processors are physically available.
4 processors are logically available.
Processor type: i486 (Intel 80486)
Processors active: 0 1 2 3
Primary memory available: 8.00 gigabytes
Default processor set: 88 tasks, 627 threads, 4 processors
Load average: 0.55, Mach factor: 3.43
compiled with gcc version:
i686-apple-darwin10-gcc-4.2.1 (GCC) 4.2.1 (Apple Inc. build 5666)
(dot 3)
[ruby-trunk - Bug #6825] forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9
on 2012-08-02 22:45
Issue #6825 has been updated by xentronium (Mark A). > I can't reproduce on OS X 10.8 ruby 2.0.0dev (2012-08-03 trunk 36602) [x86_64-darwin12.0.0] That seems right, 10.8 seems to be unaffected. > Can you show the console output with your modified active_record.rb (the loaded features section is of particular interest). <code> 1.9.3p194 :002 > require 'active_record' => true 1.9.3p194 :003 > $LOADED_FEATURES.each(&method(:puts)) enumerator.so /Users/mark/.rvm/rubies/ruby-1.9.3-p194/lib/ruby/1.9.1/x86_64-darwin10.8.0/enc/encdb.bundle /Users/mark/.rvm/rubies/ruby-1.9.3-p194/lib/ruby/1.9.1/x86_64-darwin10.8.0/enc/trans/transdb.bundle /Users/mark/.rvm/rubies/ruby-1.9.3-p194/lib/ruby/site_ruby/1.9.1/rubygems/defaults.rb /Users/mark/.rvm/rubies/ruby-1.9.3-p194/lib/ruby/1.9.1/x86_64-darwin10.8.0/rbconfig.rb /Users/mark/.rvm/rubies/ruby-1.9.3-p194/lib/ruby/site_ruby/1.9.1/rubygems/deprecate.rb /Users/mark/.rvm/rubies/ruby-1.9.3-p194/lib/ruby/site_ruby/1.9.1/rubygems/exceptions.rb /Users/mark/.rvm/rubies/ruby-1.9.3-p194/lib/ruby/site_ruby/1.9.1/rubygems/custom_require.rb /Users/mark/.rvm/rubies/ruby-1.9.3-p194/lib/ruby/site_ruby/1.9.1/rubygems.rb /Users/mark/.rvm/rubies/ruby-1.9.3-p194/lib/ruby/1.9.1/e2mmap.rb /Users/mark/.rvm/rubies/ruby-1.9.3-p194/lib/ruby/1.9.1/irb/init.rb /Users/mark/.rvm/rubies/ruby-1.9.3-p194/lib/ruby/1.9.1/irb/workspace.rb /Users/mark/.rvm/rubies/ruby-1.9.3-p194/lib/ruby/1.9.1/irb/inspector.rb /Users/mark/.rvm/rubies/ruby-1.9.3-p194/lib/ruby/1.9.1/irb/context.rb /Users/mark/.rvm/rubies/ruby-1.9.3-p194/lib/ruby/1.9.1/irb/extend-command.rb /Users/mark/.rvm/rubies/ruby-1.9.3-p194/lib/ruby/1.9.1/irb/output-method.rb /Users/mark/.rvm/rubies/ruby-1.9.3-p194/lib/ruby/1.9.1/irb/notifier.rb /Users/mark/.rvm/rubies/ruby-1.9.3-p194/lib/ruby/1.9.1/irb/slex.rb /Users/mark/.rvm/rubies/ruby-1.9.3-p194/lib/ruby/1.9.1/irb/ruby-token.rb /Users/mark/.rvm/rubies/ruby-1.9.3-p194/lib/ruby/1.9.1/irb/ruby-lex.rb /Users/mark/.rvm/rubies/ruby-1.9.3-p194/lib/ruby/1.9.1/irb/src_encoding.rb /Users/mark/.rvm/rubies/ruby-1.9.3-p194/lib/ruby/1.9.1/irb/magic-file.rb /Users/mark/.rvm/rubies/ruby-1.9.3-p194/lib/ruby/1.9.1/x86_64-darwin10.8.0/readline.bundle /Users/mark/.rvm/rubies/ruby-1.9.3-p194/lib/ruby/1.9.1/irb/input-method.rb /Users/mark/.rvm/rubies/ruby-1.9.3-p194/lib/ruby/1.9.1/irb/locale.rb /Users/mark/.rvm/rubies/ruby-1.9.3-p194/lib/ruby/1.9.1/irb.rb /Users/mark/.rvm/rubies/ruby-1.9.3-p194/lib/ruby/site_ruby/1.9.1/rubygems/version.rb /Users/mark/.rvm/rubies/ruby-1.9.3-p194/lib/ruby/site_ruby/1.9.1/rubygems/requirement.rb /Users/mark/.rvm/rubies/ruby-1.9.3-p194/lib/ruby/site_ruby/1.9.1/rubygems/platform.rb /Users/mark/.rvm/rubies/ruby-1.9.3-p194/lib/ruby/site_ruby/1.9.1/rubygems/specification.rb /Users/mark/.rvm/rubies/ruby-1.9.3-p194/lib/ruby/site_ruby/1.9.1/rubygems/path_support.rb /Users/mark/.rvm/rubies/ruby-1.9.3-p194/lib/ruby/site_ruby/1.9.1/rubygems/dependency.rb /Users/mark/.rvm/rubies/ruby-1.9.3-p194/lib/ruby/1.9.1/irb/completion.rb /Users/mark/.rvm/rubies/ruby-1.9.3-p194/lib/ruby/1.9.1/prettyprint.rb /Users/mark/.rvm/rubies/ruby-1.9.3-p194/lib/ruby/1.9.1/pp.rb /Users/mark/.rvm/scripts/irbrc.rb /Users/mark/.rvm/gems/ruby-1.9.3-p194@ruby-pthread-bug/gems/activerecord-3.2.7/lib/active_record.rb </code> If rubygems don't get hit (as in add gem directory to $LOAD_PATH manually), bug doesn't seem to trigger. I am not sure, whether rubygems themselves are the reason or something non-trivial needs to happen for it to show itself. ---------------------------------------- Bug #6825: forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9.3-p194 https://bugs.ruby-lang.org/issues/6825#change-28613 Author: xentronium (Mark A) Status: Feedback Priority: Normal Assignee: Category: Target version: ruby -v: ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] https://gist.github.com/47e48301aea114e7b1d3 here is the gist with required setup to reproduce bug. Also crash log and stdout. It seems that forking is essential for this setup to crash. Also, if you use database connection in some way prior to forking, it might not crash (however, with more complex code it still does). ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] OS X 10.6.8 hostinfo output: Mach kernel version: Darwin Kernel Version 10.8.0: Tue Jun 7 16:33:36 PDT 2011; root:xnu-1504.15.3~1/RELEASE_I386 Kernel configured for up to 4 processors. 2 processors are physically available. 4 processors are logically available. Processor type: i486 (Intel 80486) Processors active: 0 1 2 3 Primary memory available: 8.00 gigabytes Default processor set: 88 tasks, 627 threads, 4 processors Load average: 0.55, Mach factor: 3.43 compiled with gcc version: i686-apple-darwin10-gcc-4.2.1 (GCC) 4.2.1 (Apple Inc. build 5666) (dot 3)
[ruby-trunk - Bug #6825] forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9
on 2012-08-02 22:49
Issue #6825 has been updated by xentronium (Mark A). Switching `require 'active_record'` with `require 'mysql2'` still crashes the interpreter as long as required file is taken using gems mechanism. ---------------------------------------- Bug #6825: forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9.3-p194 https://bugs.ruby-lang.org/issues/6825#change-28614 Author: xentronium (Mark A) Status: Feedback Priority: Normal Assignee: Category: Target version: ruby -v: ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] https://gist.github.com/47e48301aea114e7b1d3 here is the gist with required setup to reproduce bug. Also crash log and stdout. It seems that forking is essential for this setup to crash. Also, if you use database connection in some way prior to forking, it might not crash (however, with more complex code it still does). ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] OS X 10.6.8 hostinfo output: Mach kernel version: Darwin Kernel Version 10.8.0: Tue Jun 7 16:33:36 PDT 2011; root:xnu-1504.15.3~1/RELEASE_I386 Kernel configured for up to 4 processors. 2 processors are physically available. 4 processors are logically available. Processor type: i486 (Intel 80486) Processors active: 0 1 2 3 Primary memory available: 8.00 gigabytes Default processor set: 88 tasks, 627 threads, 4 processors Load average: 0.55, Mach factor: 3.43 compiled with gcc version: i686-apple-darwin10-gcc-4.2.1 (GCC) 4.2.1 (Apple Inc. build 5666) (dot 3)
[ruby-trunk - Bug #6825][Open] forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X
on 2012-08-02 23:32
Issue #6825 has been updated by drbrain (Eric Hodel). Status changed from Feedback to Open ---------------------------------------- Bug #6825: forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9.3-p194 https://bugs.ruby-lang.org/issues/6825#change-28616 Author: xentronium (Mark A) Status: Open Priority: Normal Assignee: Category: Target version: ruby -v: ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] https://gist.github.com/47e48301aea114e7b1d3 here is the gist with required setup to reproduce bug. Also crash log and stdout. It seems that forking is essential for this setup to crash. Also, if you use database connection in some way prior to forking, it might not crash (however, with more complex code it still does). ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] OS X 10.6.8 hostinfo output: Mach kernel version: Darwin Kernel Version 10.8.0: Tue Jun 7 16:33:36 PDT 2011; root:xnu-1504.15.3~1/RELEASE_I386 Kernel configured for up to 4 processors. 2 processors are physically available. 4 processors are logically available. Processor type: i486 (Intel 80486) Processors active: 0 1 2 3 Primary memory available: 8.00 gigabytes Default processor set: 88 tasks, 627 threads, 4 processors Load average: 0.55, Mach factor: 3.43 compiled with gcc version: i686-apple-darwin10-gcc-4.2.1 (GCC) 4.2.1 (Apple Inc. build 5666) (dot 3)
[ruby-trunk - Bug #6825] forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9
on 2012-08-04 10:20
Issue #6825 has been updated by kosaki (Motohiro KOSAKI). At least, require 'mysql2' version nor drbrain version don't crash on my Mountain Lion environment. ---------------------------------------- Bug #6825: forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9.3-p194 https://bugs.ruby-lang.org/issues/6825#change-28638 Author: xentronium (Mark A) Status: Open Priority: Normal Assignee: Category: Target version: ruby -v: ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] https://gist.github.com/47e48301aea114e7b1d3 here is the gist with required setup to reproduce bug. Also crash log and stdout. It seems that forking is essential for this setup to crash. Also, if you use database connection in some way prior to forking, it might not crash (however, with more complex code it still does). ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] OS X 10.6.8 hostinfo output: Mach kernel version: Darwin Kernel Version 10.8.0: Tue Jun 7 16:33:36 PDT 2011; root:xnu-1504.15.3~1/RELEASE_I386 Kernel configured for up to 4 processors. 2 processors are physically available. 4 processors are logically available. Processor type: i486 (Intel 80486) Processors active: 0 1 2 3 Primary memory available: 8.00 gigabytes Default processor set: 88 tasks, 627 threads, 4 processors Load average: 0.55, Mach factor: 3.43 compiled with gcc version: i686-apple-darwin10-gcc-4.2.1 (GCC) 4.2.1 (Apple Inc. build 5666) (dot 3)
[ruby-trunk - Bug #6825] forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9
on 2012-08-04 16:14
Issue #6825 has been updated by xentronium (Mark A). Yeah, that seems to be restricted to snow leopard. ---------------------------------------- Bug #6825: forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9.3-p194 https://bugs.ruby-lang.org/issues/6825#change-28646 Author: xentronium (Mark A) Status: Open Priority: Normal Assignee: Category: Target version: ruby -v: ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] https://gist.github.com/47e48301aea114e7b1d3 here is the gist with required setup to reproduce bug. Also crash log and stdout. It seems that forking is essential for this setup to crash. Also, if you use database connection in some way prior to forking, it might not crash (however, with more complex code it still does). ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] OS X 10.6.8 hostinfo output: Mach kernel version: Darwin Kernel Version 10.8.0: Tue Jun 7 16:33:36 PDT 2011; root:xnu-1504.15.3~1/RELEASE_I386 Kernel configured for up to 4 processors. 2 processors are physically available. 4 processors are logically available. Processor type: i486 (Intel 80486) Processors active: 0 1 2 3 Primary memory available: 8.00 gigabytes Default processor set: 88 tasks, 627 threads, 4 processors Load average: 0.55, Mach factor: 3.43 compiled with gcc version: i686-apple-darwin10-gcc-4.2.1 (GCC) 4.2.1 (Apple Inc. build 5666) (dot 3)
on 2012-11-03 04:31
Issue #6825 has been updated by mame (Yusuke Endoh). Status changed from Open to Assigned Assignee set to kosaki (Motohiro KOSAKI) Kosaki-san, do you have any idea to addres this issue? -- Yusuke Endoh <mame@tsg.ne.jp> ---------------------------------------- Bug #6825: forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9.3-p194 https://bugs.ruby-lang.org/issues/6825#change-32282 Author: xentronium (Mark A) Status: Assigned Priority: Normal Assignee: kosaki (Motohiro KOSAKI) Category: Target version: ruby -v: ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] https://gist.github.com/47e48301aea114e7b1d3 here is the gist with required setup to reproduce bug. Also crash log and stdout. It seems that forking is essential for this setup to crash. Also, if you use database connection in some way prior to forking, it might not crash (however, with more complex code it still does). ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] OS X 10.6.8 hostinfo output: Mach kernel version: Darwin Kernel Version 10.8.0: Tue Jun 7 16:33:36 PDT 2011; root:xnu-1504.15.3~1/RELEASE_I386 Kernel configured for up to 4 processors. 2 processors are physically available. 4 processors are logically available. Processor type: i486 (Intel 80486) Processors active: 0 1 2 3 Primary memory available: 8.00 gigabytes Default processor set: 88 tasks, 627 threads, 4 processors Load average: 0.55, Mach factor: 3.43 compiled with gcc version: i686-apple-darwin10-gcc-4.2.1 (GCC) 4.2.1 (Apple Inc. build 5666) (dot 3)
[ruby-trunk - Bug #6825] forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9
on 2012-11-03 18:36
Issue #6825 has been updated by kosaki (Motohiro KOSAKI). Assignee changed from kosaki (Motohiro KOSAKI) to mrkn (Kenta Murata) > Kosaki-san, do you have any idea to addres this issue? I have no idea. unfortunately snow leopard is too old and i have no chance to get it. @mrkn, do you have any chance to see this issue? ---------------------------------------- Bug #6825: forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9.3-p194 https://bugs.ruby-lang.org/issues/6825#change-32304 Author: xentronium (Mark A) Status: Assigned Priority: Normal Assignee: mrkn (Kenta Murata) Category: Target version: ruby -v: ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] https://gist.github.com/47e48301aea114e7b1d3 here is the gist with required setup to reproduce bug. Also crash log and stdout. It seems that forking is essential for this setup to crash. Also, if you use database connection in some way prior to forking, it might not crash (however, with more complex code it still does). ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] OS X 10.6.8 hostinfo output: Mach kernel version: Darwin Kernel Version 10.8.0: Tue Jun 7 16:33:36 PDT 2011; root:xnu-1504.15.3~1/RELEASE_I386 Kernel configured for up to 4 processors. 2 processors are physically available. 4 processors are logically available. Processor type: i486 (Intel 80486) Processors active: 0 1 2 3 Primary memory available: 8.00 gigabytes Default processor set: 88 tasks, 627 threads, 4 processors Load average: 0.55, Mach factor: 3.43 compiled with gcc version: i686-apple-darwin10-gcc-4.2.1 (GCC) 4.2.1 (Apple Inc. build 5666) (dot 3)
[ruby-trunk - Bug #6825] forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9
on 2012-11-03 21:44
Issue #6825 has been updated by mrkn (Kenta Murata). I don't have snow-leopard environment, so I cannot investigate this issue. ---------------------------------------- Bug #6825: forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9.3-p194 https://bugs.ruby-lang.org/issues/6825#change-32310 Author: xentronium (Mark A) Status: Assigned Priority: Normal Assignee: mrkn (Kenta Murata) Category: Target version: ruby -v: ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] https://gist.github.com/47e48301aea114e7b1d3 here is the gist with required setup to reproduce bug. Also crash log and stdout. It seems that forking is essential for this setup to crash. Also, if you use database connection in some way prior to forking, it might not crash (however, with more complex code it still does). ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] OS X 10.6.8 hostinfo output: Mach kernel version: Darwin Kernel Version 10.8.0: Tue Jun 7 16:33:36 PDT 2011; root:xnu-1504.15.3~1/RELEASE_I386 Kernel configured for up to 4 processors. 2 processors are physically available. 4 processors are logically available. Processor type: i486 (Intel 80486) Processors active: 0 1 2 3 Primary memory available: 8.00 gigabytes Default processor set: 88 tasks, 627 threads, 4 processors Load average: 0.55, Mach factor: 3.43 compiled with gcc version: i686-apple-darwin10-gcc-4.2.1 (GCC) 4.2.1 (Apple Inc. build 5666) (dot 3)
[ruby-trunk - Bug #6825][Closed] forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS
on 2012-11-03 23:01
Issue #6825 has been updated by kosaki (Motohiro KOSAKI). Status changed from Assigned to Closed > I don't have snow-leopard environment, so I cannot investigate this issue. OK. Thank you. This looks like old OS X bug but we have no way to dig. give up. I'd like to close this feature as won't fix. Anyway snow leopard is no longer supported. To Mark, please reopen when you find exact reason and fixing way. we are very sorry for inconvenience. ---------------------------------------- Bug #6825: forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9.3-p194 https://bugs.ruby-lang.org/issues/6825#change-32312 Author: xentronium (Mark A) Status: Closed Priority: Normal Assignee: mrkn (Kenta Murata) Category: Target version: ruby -v: ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] https://gist.github.com/47e48301aea114e7b1d3 here is the gist with required setup to reproduce bug. Also crash log and stdout. It seems that forking is essential for this setup to crash. Also, if you use database connection in some way prior to forking, it might not crash (however, with more complex code it still does). ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] OS X 10.6.8 hostinfo output: Mach kernel version: Darwin Kernel Version 10.8.0: Tue Jun 7 16:33:36 PDT 2011; root:xnu-1504.15.3~1/RELEASE_I386 Kernel configured for up to 4 processors. 2 processors are physically available. 4 processors are logically available. Processor type: i486 (Intel 80486) Processors active: 0 1 2 3 Primary memory available: 8.00 gigabytes Default processor set: 88 tasks, 627 threads, 4 processors Load average: 0.55, Mach factor: 3.43 compiled with gcc version: i686-apple-darwin10-gcc-4.2.1 (GCC) 4.2.1 (Apple Inc. build 5666) (dot 3)
[ruby-trunk - Bug #6825] forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9
on 2012-11-03 23:48
Issue #6825 has been updated by Eregon (Benoit Daloze). mrkn (Kenta Murata) wrote: > I don't have snow-leopard environment, so I cannot investigate this issue. I do still have a snow-leopard environment and I can reproduce with ruby 1.9.3p125 (2012-02-16 revision 34643) [x86_64-darwin10.8.0] and trunk (r37359). I'll try to investigate if I find time. ---------------------------------------- Bug #6825: forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9.3-p194 https://bugs.ruby-lang.org/issues/6825#change-32314 Author: xentronium (Mark A) Status: Closed Priority: Normal Assignee: mrkn (Kenta Murata) Category: Target version: ruby -v: ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] https://gist.github.com/47e48301aea114e7b1d3 here is the gist with required setup to reproduce bug. Also crash log and stdout. It seems that forking is essential for this setup to crash. Also, if you use database connection in some way prior to forking, it might not crash (however, with more complex code it still does). ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] OS X 10.6.8 hostinfo output: Mach kernel version: Darwin Kernel Version 10.8.0: Tue Jun 7 16:33:36 PDT 2011; root:xnu-1504.15.3~1/RELEASE_I386 Kernel configured for up to 4 processors. 2 processors are physically available. 4 processors are logically available. Processor type: i486 (Intel 80486) Processors active: 0 1 2 3 Primary memory available: 8.00 gigabytes Default processor set: 88 tasks, 627 threads, 4 processors Load average: 0.55, Mach factor: 3.43 compiled with gcc version: i686-apple-darwin10-gcc-4.2.1 (GCC) 4.2.1 (Apple Inc. build 5666) (dot 3)
on 2012-11-03 23:58
Issue #6825 has been updated by Eregon (Benoit Daloze). Status changed from Closed to Assigned Assignee changed from mrkn (Kenta Murata) to Eregon (Benoit Daloze) This is also true with latest trunk (r37462). I'm assigning to myself. bug6825.rb:31: [BUG] pthread_cond_timedwait: Invalid argument (EINVAL) ruby 2.0.0dev (2012-11-04 trunk 37462) [x86_64-darwin10.8.0] -- Control frame information ----------------------------------------------- c:0005 p:---- s:0013 e:000012 CFUNC :sleep c:0004 p:0007 s:0009 e:000008 BLOCK bug6825.rb:31 [FINISH] c:0003 p:---- s:0007 e:000006 CFUNC :loop c:0002 p:0005 s:0004 e:000003 BLOCK bug6825.rb:31 [FINISH] c:0001 p:---- s:0002 e:000001 TOP [FINISH] bug6825.rb:31:in `block (2 levels) in initialize' bug6825.rb:31:in `loop' bug6825.rb:31:in `block (3 levels) in initialize' bug6825.rb:31:in `sleep' ---------------------------------------- Bug #6825: forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9.3-p194 https://bugs.ruby-lang.org/issues/6825#change-32315 Author: xentronium (Mark A) Status: Assigned Priority: Normal Assignee: Eregon (Benoit Daloze) Category: Target version: ruby -v: ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] https://gist.github.com/47e48301aea114e7b1d3 here is the gist with required setup to reproduce bug. Also crash log and stdout. It seems that forking is essential for this setup to crash. Also, if you use database connection in some way prior to forking, it might not crash (however, with more complex code it still does). ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] OS X 10.6.8 hostinfo output: Mach kernel version: Darwin Kernel Version 10.8.0: Tue Jun 7 16:33:36 PDT 2011; root:xnu-1504.15.3~1/RELEASE_I386 Kernel configured for up to 4 processors. 2 processors are physically available. 4 processors are logically available. Processor type: i486 (Intel 80486) Processors active: 0 1 2 3 Primary memory available: 8.00 gigabytes Default processor set: 88 tasks, 627 threads, 4 processors Load average: 0.55, Mach factor: 3.43 compiled with gcc version: i686-apple-darwin10-gcc-4.2.1 (GCC) 4.2.1 (Apple Inc. build 5666) (dot 3)
[ruby-trunk - Bug #6825] forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9
on 2012-11-04 13:09
Issue #6825 has been updated by Eregon (Benoit Daloze). I poked around an produced a core dump (the bug would not reproduce under gdb with a breakpoint set). Arguments to pthread_cond_timedwait() seem valid, in particular the timespec is about 500ms in the future. Other calls to pthread_cond_timedwait() always return ETIMEDOUT or 0. I saw rb_thread_t::native_thread_data.sleep_cond was weirdly initialized. It is not initialized in native_thread_init() if HAVE_PTHREAD_CONDATTR_INIT is undefined. And it is used in any case in ubf_pthread_cond_signal(). Maybe checks for HAVE_PTHREAD_CONDATTR_INIT should not be done in native_thread_init() and native_thread_destroy() since these functions already do the right checks? This should be unrelated though, since OS X has pthread_condattr_init(). It might be related to GVL release by multiple threads but I have no clue. It does not seem related directly to the parallel DNS resolution, since some traces have only threads in native_cond_timedwait(). And from the "only reproducible on snow-leopard" argument, it seems snow-leopard pthread's bug. @kosaki @mrkn Would it be useful if I could provide you the core dump and other info? ---------------------------------------- Bug #6825: forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9.3-p194 https://bugs.ruby-lang.org/issues/6825#change-32339 Author: xentronium (Mark A) Status: Assigned Priority: Normal Assignee: Eregon (Benoit Daloze) Category: Target version: ruby -v: ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] https://gist.github.com/47e48301aea114e7b1d3 here is the gist with required setup to reproduce bug. Also crash log and stdout. It seems that forking is essential for this setup to crash. Also, if you use database connection in some way prior to forking, it might not crash (however, with more complex code it still does). ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] OS X 10.6.8 hostinfo output: Mach kernel version: Darwin Kernel Version 10.8.0: Tue Jun 7 16:33:36 PDT 2011; root:xnu-1504.15.3~1/RELEASE_I386 Kernel configured for up to 4 processors. 2 processors are physically available. 4 processors are logically available. Processor type: i486 (Intel 80486) Processors active: 0 1 2 3 Primary memory available: 8.00 gigabytes Default processor set: 88 tasks, 627 threads, 4 processors Load average: 0.55, Mach factor: 3.43 compiled with gcc version: i686-apple-darwin10-gcc-4.2.1 (GCC) 4.2.1 (Apple Inc. build 5666) (dot 3)
[ruby-trunk - Bug #6825] forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9
on 2012-11-04 15:54
Issue #6825 has been updated by kosaki (Motohiro KOSAKI). > @kosaki @mrkn Would it be useful if I could provide you the core dump and other info? Thanks! r35672 seems broke this area and I'll fix it soon. However there is no r35672 in 1.9.3 branch and 1.9.3 seems correct. hmm... Could you please try 1.9.3 branch too? ---------------------------------------- Bug #6825: forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9.3-p194 https://bugs.ruby-lang.org/issues/6825#change-32342 Author: xentronium (Mark A) Status: Assigned Priority: Normal Assignee: Eregon (Benoit Daloze) Category: Target version: ruby -v: ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] https://gist.github.com/47e48301aea114e7b1d3 here is the gist with required setup to reproduce bug. Also crash log and stdout. It seems that forking is essential for this setup to crash. Also, if you use database connection in some way prior to forking, it might not crash (however, with more complex code it still does). ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] OS X 10.6.8 hostinfo output: Mach kernel version: Darwin Kernel Version 10.8.0: Tue Jun 7 16:33:36 PDT 2011; root:xnu-1504.15.3~1/RELEASE_I386 Kernel configured for up to 4 processors. 2 processors are physically available. 4 processors are logically available. Processor type: i486 (Intel 80486) Processors active: 0 1 2 3 Primary memory available: 8.00 gigabytes Default processor set: 88 tasks, 627 threads, 4 processors Load average: 0.55, Mach factor: 3.43 compiled with gcc version: i686-apple-darwin10-gcc-4.2.1 (GCC) 4.2.1 (Apple Inc. build 5666) (dot 3)
on 2012-11-05 03:48
Issue #6825 has been updated by kosaki (Motohiro KOSAKI). Status changed from Closed to Assigned Reopened. because r37474 only fixed trunk. ---------------------------------------- Bug #6825: forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9.3-p194 https://bugs.ruby-lang.org/issues/6825#change-32356 Author: xentronium (Mark A) Status: Assigned Priority: Normal Assignee: Eregon (Benoit Daloze) Category: Target version: ruby -v: ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] https://gist.github.com/47e48301aea114e7b1d3 here is the gist with required setup to reproduce bug. Also crash log and stdout. It seems that forking is essential for this setup to crash. Also, if you use database connection in some way prior to forking, it might not crash (however, with more complex code it still does). ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] OS X 10.6.8 hostinfo output: Mach kernel version: Darwin Kernel Version 10.8.0: Tue Jun 7 16:33:36 PDT 2011; root:xnu-1504.15.3~1/RELEASE_I386 Kernel configured for up to 4 processors. 2 processors are physically available. 4 processors are logically available. Processor type: i486 (Intel 80486) Processors active: 0 1 2 3 Primary memory available: 8.00 gigabytes Default processor set: 88 tasks, 627 threads, 4 processors Load average: 0.55, Mach factor: 3.43 compiled with gcc version: i686-apple-darwin10-gcc-4.2.1 (GCC) 4.2.1 (Apple Inc. build 5666) (dot 3)
on 2012-11-07 11:49
Issue #6825 has been updated by Eregon (Benoit Daloze). kosaki (Motohiro KOSAKI) wrote: > > @kosaki @mrkn Would it be useful if I could provide you the core dump and other info? > > Thanks! > r35672 seems broke this area and I'll fix it soon. However there is no r35672 in 1.9.3 branch and 1.9.3 seems correct. hmm... > Could you please try 1.9.3 branch too? Unfortunately, r37474 does not seem to solve the problem (but it was definitely a potential problem). This is expected because snow leopard has pthread_condattr_init(). So I don't know the reason for the bug. ---------------------------------------- Backport #6825: forking and pthread_cond_timedwait: Invalid argument (EINVAL) on OS X / 1.9.3-p194 https://bugs.ruby-lang.org/issues/6825#change-32551 Author: xentronium (Mark A) Status: Assigned Priority: Normal Assignee: usa (Usaku NAKAMURA) Category: Target version: https://gist.github.com/47e48301aea114e7b1d3 here is the gist with required setup to reproduce bug. Also crash log and stdout. It seems that forking is essential for this setup to crash. Also, if you use database connection in some way prior to forking, it might not crash (however, with more complex code it still does). ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-darwin10.8.0] OS X 10.6.8 hostinfo output: Mach kernel version: Darwin Kernel Version 10.8.0: Tue Jun 7 16:33:36 PDT 2011; root:xnu-1504.15.3~1/RELEASE_I386 Kernel configured for up to 4 processors. 2 processors are physically available. 4 processors are logically available. Processor type: i486 (Intel 80486) Processors active: 0 1 2 3 Primary memory available: 8.00 gigabytes Default processor set: 88 tasks, 627 threads, 4 processors Load average: 0.55, Mach factor: 3.43 compiled with gcc version: i686-apple-darwin10-gcc-4.2.1 (GCC) 4.2.1 (Apple Inc. build 5666) (dot 3)
Enable email notification
Enable multi-page viewPlease log in before posting. Registration is free and takes only a minute.
Existing account
(Switch to SSL-encrypted connection)
NEW: Do you have a Google/GoogleMail or Yahoo account? No registration required!
Log in with Google account | Log in with Yahoo account
Log in with Google account | Log in with Yahoo account
No account? Register here.