kgio test_signal_close failed

kgio test_signal_close failed

[375gnu]

Sometimes test_signal_close from test_poll.rb fails. Not every time, and only
with ruby1.9 (1.9.3.0 is my current version). With ruby 1.8.7 it's passed.

There is an output:

Failure: <IOError> exception expected but none was thrown.
test_signal_close(TestPoll)
/home/globus/work/pkg-ruby-extras/build-area/ruby-kgio-2.7.2/test/test_poll.rb:61:in
`test_signal_close'
    58:     res = nil
    59:     thr = Thread.new { sleep 0.100; Process.kill(:USR1, $$) }
    60:     t0 = Time.now
 => 61:     assert_raises(IOError) { Kgio.poll({@rd => Kgio::POLLIN}) }
    62:     diff = Time.now - t0
    63:     thr.join
    64:     assert diff >= 0.010, "diff=#{diff}"

-- 
https://375gnu.wordpress.com/
https://launchpad.net/~375gnu/+archive/ppa

Re: kgio test_signal_close failed

[Eric Wong]

375gnu <375gnu@gmail.com> wrote:
> Sometimes test_signal_close from test_poll.rb fails. Not every time, and only
> with ruby1.9 (1.9.3.0 is my current version). With ruby 1.8.7 it's passed.

Is this also on Debian/kfreebsd?  I have a VM for that lying around
somewhere, will investigate as I find time.

> There is an output:
> 
> Failure: <IOError> exception expected but none was thrown.
> test_signal_close(TestPoll)
> /home/globus/work/pkg-ruby-extras/build-area/ruby-kgio-2.7.2/test/test_poll.rb:61:in
> `test_signal_close'
>     58:     res = nil
>     59:     thr = Thread.new { sleep 0.100; Process.kill(:USR1, $$) }

Does increasing the sleep time there help?  It's a potential race
condition :/

>     60:     t0 = Time.now
>  => 61:     assert_raises(IOError) { Kgio.poll({@rd => Kgio::POLLIN}) }
>     62:     diff = Time.now - t0
>     63:     thr.join
>     64:     assert diff >= 0.010, "diff=#{diff}"

Re: kgio test_signal_close failed

[375gnu]

On 3/4/12, Eric Wong <normalperson@yhbt.net> wrote:
> 375gnu <375gnu@gmail.com> wrote:
>> Sometimes test_signal_close from test_poll.rb fails. Not every time, and
>> only
>> with ruby1.9 (1.9.3.0 is my current version). With ruby 1.8.7 it's passed.
> Is this also on Debian/kfreebsd?

Not (0 of 25 times). But it runs in kvm, so it's much slower than real linux
box (110 sec for all tests vs. 75).

>>     59:     thr = Thread.new { sleep 0.100; Process.kill(:USR1, $$) }
>
> Does increasing the sleep time there help?  It's a potential race
> condition :/


No, even with "sleep 2" (1 of 10 is failed)

-- 
https://375gnu.wordpress.com/
https://launchpad.net/~375gnu/+archive/ppa

Re: kgio test_signal_close failed

[Eric Wong]

375gnu <375gnu@gmail.com> wrote:
> On 3/4/12, Eric Wong <normalperson@yhbt.net> wrote:
> > 375gnu <375gnu@gmail.com> wrote:
> >> Sometimes test_signal_close from test_poll.rb fails. Not every time, and
> >> only
> >> with ruby1.9 (1.9.3.0 is my current version). With ruby 1.8.7 it's passed.
> > Is this also on Debian/kfreebsd?
> 
> Not (0 of 25 times). But it runs in kvm, so it's much slower than real linux
> box (110 sec for all tests vs. 75).

Odd, I can't reproduce under kvm (Debian wheezy + custom 3.3.0-rc4
kernel).  Are you on 2.6.38 kernel by any chance?  Some versions of that
had weird signal issues.

Also, this test is just skipped under Matz Ruby 1.8 (can't safely
support poll() w/o native threads + GVL release).

> >>     59:     thr = Thread.new { sleep 0.100; Process.kill(:USR1, $$) }
> >
> > Does increasing the sleep time there help?  It's a potential race
> > condition :/
> 
> No, even with "sleep 2" (1 of 10 is failed)

Perhaps this makes the test less bad?

diff --git a/test/test_poll.rb b/test/test_poll.rb
index ba0e561..df59354 100644
--- a/test/test_poll.rb
+++ b/test/test_poll.rb
@@ -58,7 +58,10 @@ class TestPoll < Test::Unit::TestCase
     res = nil
     thr = Thread.new { sleep 0.100; Process.kill(:USR1, $$) }
     t0 = Time.now
-    assert_raises(IOError) { Kgio.poll({@rd => Kgio::POLLIN}) }
+    assert_raises(IOError) do
+      result = Kgio.poll({@rd => Kgio::POLLIN})
+      result.each_key { |io| io.read_nonblock(1) }
+    end
     diff = Time.now - t0
     thr.join
     assert diff >= 0.010, "diff=#{diff}"

Re: kgio test_signal_close failed

[375gnu]

On 3/6/12, Eric Wong <normalperson@yhbt.net> wrote:
> Odd, I can't reproduce under kvm (Debian wheezy + custom 3.3.0-rc4
> kernel).  Are you on 2.6.38 kernel by any chance?  Some versions of that
> had weird signal issues.

No, I'm using kernel 3.2.*

> Perhaps this makes the test less bad?
> -    assert_raises(IOError) { Kgio.poll({@rd => Kgio::POLLIN}) }
> +    assert_raises(IOError) do
> +      result = Kgio.poll({@rd => Kgio::POLLIN})
> +      result.each_key { |io| io.read_nonblock(1) }
> +    end

Yes! It works! I've run test 50 times and everyone's passed!


-- 
https://375gnu.wordpress.com/
https://launchpad.net/~375gnu/+archive/ppa

Re: kgio test_signal_close failed

[Eric Wong]

375gnu <375gnu@gmail.com> wrote:
> Yes! It works! I've run test 50 times and everyone's passed!

Awesome, thanks for the confirmation.  Anything else?

I'm hoping out 2.7.3 before the end of the week.

Re: kgio test_signal_close failed

[Iñaki Baz Castillo]

2012/3/7 Eric Wong <normalperson@yhbt.net>:
>> Yes! It works! I've run test 50 times and everyone's passed!
>
> Awesome, thanks for the confirmation.  Anything else?
>
> I'm hoping out 2.7.3 before the end of the week.

Hi Eric, how to test the test files? "rake test" does nothing.

-- 
Iñaki Baz Castillo
<ibc@aliax.net>

Re: kgio test_signal_close failed

[Eric Wong]

Iñaki Baz Castillo <ibc@aliax.net> wrote:
> Hi Eric, how to test the test files? "rake test" does nothing.

GNU make.  I just pushed the following out:

Thanks for testing :>

From 2c2befb1caa47fe3bf2e6d31dd0733956d178c87 Mon Sep 17 00:00:00 2001
From: Eric Wong <normalperson@yhbt.net>
Date: Wed, 7 Mar 2012 10:14:17 +0000
Subject: [PATCH] HACKING: add instructions for running tests

Unlike most Rubyists, I prefer GNU make to Rake.
---
 HACKING |   15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/HACKING b/HACKING
index 004bcbb..5d485be 100644
--- a/HACKING
+++ b/HACKING
@@ -61,3 +61,18 @@ Without RubyGems (via setup.rb):
 
 It is not at all recommended to mix a RubyGems installation with an
 installation done without RubyGems, however.
+
+=== Tests
+
+We use GNU make to run tests in parallel.  test/unit/parallel didn't
+exist for old versions of Ruby before 1.9.3.  Users of GNU-based systems
+(such as GNU/Linux) usually have GNU make installed as "make" instead of
+"gmake".
+
+Running the entire test suite with 4 tests in parallel:
+
+  gmake -j4 test
+
+Running just one unit test:
+
+  gmake test/test_poll.rb

Re: kgio test_signal_close failed

[Iñaki Baz Castillo]

2012/3/7 Eric Wong <normalperson@yhbt.net>:
> Iñaki Baz Castillo <ibc@aliax.net> wrote:
>> Hi Eric, how to test the test files? "rake test" does nothing.
>
> GNU make.  I just pushed the following out:
>
> Thanks for testing :>

No errors:

- Ubuntu 3.0.0-15-generic 64 bits
- Ruby 1.9.2p290

-- 
Iñaki Baz Castillo
<ibc@aliax.net>