[#1834] New syck bug — Chad Fowler <chad@...>
There is a new syck bug that appears to be caused by the recent fix for
[#1836] exit inside test/unit — nobu.nokada@...
Hi,
On Dec 1, 2003, at 02:55, nobu.nokada@softhome.net wrote:
[#1843] DRb tests hang on OS X 10.3.1 — Nathaniel Talbott <nathaniel@...>
I haven't yet been able to test this on another platform to see if it
[#1846] Re: Constants, class variables and the cbase field — george.marrows@...
> What kind of behavior do you want (to change)? Remember you're saying
Hi,
On Monday 01 December 2003 06:44 pm, Yukihiro Matsumoto wrote:
Hi,
On Tuesday 02 December 2003 04:02 am, Yukihiro Matsumoto wrote:
[#1884] multiple exceptions for assert_raises — nobu.nokada@...
Hi,
Hi,
On Dec 4, 2003, at 02:34, Yukihiro Matsumoto wrote:
On Dec 4, 2003, at 01:35, nobu.nokada@softhome.net wrote:
On Dec 4, 2003, at 10:39, Nathaniel Talbott wrote:
[#1901] Test::Unit problem — "Sean E. Russell" <ser@...>
-----BEGIN PGP SIGNED MESSAGE-----
Hi,
[#1914] -Wall warnings from 1.8.1 p3 — Daniel Berger <djberge@...>
Here are some potentially significant warnings from 1.8.1 p3
nobu.nokada@softhome.net wrote:
[#1932] --enable-pthread broken? — Nathaniel Talbott <nathaniel@...>
[ruby-talk: 87759] and the surrounding thread seem to indicate that
Hi,
On Dec 11, 2003, at 11:49, Yukihiro Matsumoto wrote:
>>>>> "N" == Nathaniel Talbott <nathaniel@talbott.ws> writes:
Hi,
On Dec 11, 2003, at 16:10, nobu.nokada@softhome.net wrote:
Hi,
Hi,
On Dec 11, 2003, at 20:48, Yukihiro Matsumoto wrote:
Hi,
>>>>> "Y" == Yukihiro Matsumoto <matz@ruby-lang.org> writes:
Hi,
>>>>> "Y" == Yukihiro Matsumoto <matz@ruby-lang.org> writes:
Hi,
>>>>> "Y" == Yukihiro Matsumoto <matz@ruby-lang.org> writes:
Hi,
[#1936] Can't define +@ for Symbol (plus ruby install problem) — "T. Onoma" <transami@...>
I wanted to see if the +@ problem was fixed in 1.8.1 preview 3 but when I do
Hi,
On Friday 12 December 2003 02:39 am, Yukihiro Matsumoto wrote:
Hi,
Hi.
Hi,
[#1973] Where to install documentation — Dave Thomas <dave@...>
Folks:
Hi,
Dave Thomas (dave@pragprog.com) wrote:
>
>> Using the standard install.rb, anything you include in a project's
Hi,
On Tue, Dec 16, 2003 at 03:52:26PM +0900, Dave Thomas wrote:
Hi,
[#2013] Mixin Module, Possible Bug? — "T. Onoma" <transami@...>
According to Pickaxe, Ch. 19, pg. 245, under Mixin Modules:
[#2037] --enable-pthread still segfaults... — Nathaniel Talbott <nathaniel@...>
I've finally been able to test my application under load using the
Hi,
On Dec 23, 2003, at 14:17, Yukihiro Matsumoto wrote:
Hi,
On Dec 23, 2003, at 14:34, Yukihiro Matsumoto wrote:
Hi,
On Dec 23, 2003, at 14:44, Yukihiro Matsumoto wrote:
Hi,
> I'm afraid you're using old configure file. Can you wipe off old
On Dec 23, 2003, at 15:18, Nathaniel Talbott wrote:
In message "Re: --enable-pthread still segfaults..."
On Dec 23, 2003, at 16:34, Yukihiro Matsumoto wrote:
Hi,
On Dec 23, 2003, at 17:04, Yukihiro Matsumoto wrote:
Hi,
On Dec 23, 2003, at 17:29, Yukihiro Matsumoto wrote:
Hi,
[#2071] rdoc is broken in 1.8.1 — Alexander Bokovoy <a.bokovoy@...>
Greetings!
[#2084] Error with Socket.getaddrinfo on OS X — Richard Kilmer <rich@...>
On OS X Panther:
[#2101] Can't call to_s on a frozen Date — Gavin Sinclair <gsinclair@...>
Interesting...
[#2102] syck segfaults when used in rdoc — Alexander Bokovoy <a.bokovoy@...>
Greetings!
>>>>> "A" == Alexander Bokovoy <a.bokovoy@sam-solutions.net> writes:
On Sun, Dec 28, 2003 at 11:41:49PM +0900, ts wrote:
>>>>> "A" == Alexander Bokovoy <a.bokovoy@sam-solutions.net> writes:
Hi,
[#2122] Bad interaction between timeout.rb and --enable-pthread — Nathaniel Talbott <nathaniel@...>
Here's a testcase that shows the problem:
I should have reduced it more before posting...
Nathaniel Talbott wrote:
Hi,
Hi,
On Jan 1, 2004, at 11:29, Yukihiro Matsumoto wrote:
On Jan 1, 2004, at 12:14, Nathaniel Talbott wrote:
Re: --enable-pthread broken?
On Dec 11, 2003, at 16:10, nobu.nokada@softhome.net wrote:
> At Fri, 12 Dec 2003 03:08:58 +0900,
> ts wrote:
>> N> I tried it against the latest Ruby CVS, and it still crashes. It
>> seemed
>> N> to take (subjectively) longer to crash, though. Interestingly, 2
>> out of
>> N> 3 runs gave me this instead of a segfault:
>>
>> Well, look where it crash.
>>
>> With the previous version normally you have a crash in
>> * rb_thread_restore_content()
>> * and localjump_destination()
>>
>> normally with the new patch (if I'm right), you have just a crash in
>> * localjump_destination()
>
> Then, it'd be simple.
Maybe for you... but for me, I'm very grateful for Ruby super-heroes :-)
Anyhow, I tried this patch, and I no longer get a segfault or any other
kind of crash, which is great!
Unfortunately, with a slight modification to keep the killer running
even if some threads time out, the killer will now lock completely
sometimes with the same characteristics as reported in
[ruby-talk:87519]. Here's the line from top:
PID USER WCHAN STAT %CPU %MEM TIME COMMAND
5502 ntalbott rt_sigsus S 0.0 21.8 1:15 bin/ruby
/home/ntalbott/sandbox/killer.rb
Not only that, but it's hard to reproduce this; under production, the
process will run for hours without it happening, and even with this
test case I've only had it happen once. I'm continuing to try different
variants, and will let you know if I find one that fails more
consistently.
This 'rt_sigsuspend problem' is why I enabled pthread in the first
place, but apparently that doesn't really fix it. Any guesses? I have a
hunch that Net::HTTP (and the underlying native code) are interacting
poorly with threads (my full WEBrick server uses Net::HTTP as well),
but it's only a guess.
Here's the new killer.rb:
require 'thread'
require 'net/http'
threads = []
STDOUT.sync = true
200.times do
threads << Thread.new do
h = Net::HTTP::new('localhost', 2000)
loop do
h.get('/hello')
print('.')
end
end
end
sleep(60 * 10)
Thanks,
Nathaniel
<:((><