[#23132] [Bug #1357] Fixing variables into specific CPU registers deemed overrated & may disturb compilers' optimizers — Ollivier Robert <redmine@...>
Bug #1357: Fixing variables into specific CPU registers deemed overrated & may disturb compilers' optimizers
5 messages
2009/04/06
[#23154] [Bug #1363] Wrong value for Hash of NaN — Heesob Park <redmine@...>
Bug #1363: Wrong value for Hash of NaN
10 messages
2009/04/08
[#23423] Re: [Bug #1363] Wrong value for Hash of NaN
— Nobuyoshi Nakada <nobu@...>
2009/05/11
Hi,
[#23425] Re: [Bug #1363] Wrong value for Hash of NaN
— Yukihiro Matsumoto <matz@...>
2009/05/11
Hi,
[#23427] Re: [Bug #1363] Wrong value for Hash of NaN
— Urabe Shyouhei <shyouhei@...>
2009/05/11
Yukihiro Matsumoto wrote:
[#23168] [Bug #1367] flatten(0) is not consistent with flatten(), flatten(1), etc. — Paul Lewis <redmine@...>
Bug #1367: flatten(0) is not consistent with flatten(), flatten(1), etc.
6 messages
2009/04/08
[#23211] [Bug #1367] flatten(0) is not consistent with flatten(), flatten(1), etc.
— Paul Lewis <redmine@...>
2009/04/15
Issue #1367 has been updated by Paul Lewis.
[#23174] [Feature #1371] FTPS Implicit — Daniel Parker <redmine@...>
Feature #1371: FTPS Implicit
6 messages
2009/04/09
[#23193] Regexp Encoding — James Gray <james@...>
I'm trying to document the Encoding Regexp objects receive for the =20
5 messages
2009/04/13
[#23194] [Feature #1377] Please provide constant File::NOATIME — Johan Walles <redmine@...>
Feature #1377: Please provide constant File::NOATIME
5 messages
2009/04/13
[#23231] What do you think about changing the return value of Kernel#require and Kernel#load to the source encoding of the required file? — =?ISO-8859-15?Q?Wolfgang_N=E1dasi-Donner?= <ed.odanow@...>
Dear Ruby developers and users!
8 messages
2009/04/17
[#23233] Re: What do you think about changing the return value of Kernel#require and Kernel#load to the source encoding of the required file?
— Michael Neumann <mneumann@...>
2009/04/17
Wolfgang N叩dasi-Donner wrote:
[#23244] Re: What do you think about changing the return value of Kernel#require and Kernel#load to the source encoding of the required file?
— Michael Neumann <mneumann@...>
2009/04/17
Wolfgang N叩dasi-Donner wrote:
[#23245] Re: What do you think about changing the return value of Kernel#require and Kernel#load to the source encoding of the required file?
— Wolfgang Nádasi-Donner <ed.odanow@...>
2009/04/17
Michael Neumann schrieb:
[#23252] [Bug #1392] Object#extend leaks memory on Ruby 1.9.1 — Muhammad Ali <redmine@...>
Bug #1392: Object#extend leaks memory on Ruby 1.9.1
4 messages
2009/04/18
[#23267] StringIO: RubySpec violation — Hongli Lai <[email protected]>
I ran RubySpec against the 1.8.6-p368 release. It seems that
5 messages
2009/04/20
[#23289] [Bug #1399] Segmentation fault is raised when you use a postgres gem — Marcel Keil <redmine@...>
Bug #1399: Segmentation fault is raised when you use a postgres gem
3 messages
2009/04/23
[#23297] Ruby Oniguruma question — Ralf Junker <ralfjunker@...>
I see that the Ruby source code contains modified and more recent version of the Oniguruma regular expression library.
5 messages
2009/04/23
[#23305] [Bug #1403] Process.daemon should do a double fork to avoid problems with controlling terminals — Gary Wright <redmine@...>
Bug #1403: Process.daemon should do a double fork to avoid problems with controlling terminals
7 messages
2009/04/24
[#23306] Re: [Bug #1403] Process.daemon should do a double fork to avoid problems with controlling terminals
— Nobuyoshi Nakada <nobu@...>
2009/04/24
Hi,
[#23307] Re: [Bug #1403] Process.daemon should do a double fork to avoid problems with controlling terminals
— Gary Wright <gwtmp01@...>
2009/04/25
[#23311] [Bug #1404] Net::HTTP::Post failing when a post field contains ":" — Ignacio Martín <redmine@...>
Bug #1404: Net::HTTP::Post failing when a post field contains ":"
5 messages
2009/04/25
[#23318] [Feature #1408] 0.1.to_r not equal to (1/10) — Heesob Park <redmine@...>
Feature #1408: 0.1.to_r not equal to (1/10)
19 messages
2009/04/26
[#23345] [Feature #1408] 0.1.to_r not equal to (1/10)
— Roger Pack <redmine@...>
2009/05/01
Issue #1408 has been updated by Roger Pack.
[#25662] [Feature #1408] 0.1.to_r not equal to (1/10)
— Marc-Andre Lafortune <redmine@...>
2009/09/20
Issue #1408 has been updated by Marc-Andre Lafortune.
[#29652] [Feature #1408] 0.1.to_r not equal to (1/10)
— tadayoshi funaba <redmine@...>
2010/04/20
Issue #1408 has been updated by tadayoshi funaba.
[#29663] Re: [Feature #1408] 0.1.to_r not equal to (1/10)
— Marc-Andre Lafortune <ruby-core-mailing-list@...>
2010/04/20
Hi,
[#29887] Re: [Feature #1408] 0.1.to_r not equal to (1/10)
— Marc-Andre Lafortune <ruby-core-mailing-list@...>
2010/04/29
Hi.
[#23321] [Bug #1412] 1.8.7-p160 extmk.rb fails when cross compiling — Luis Lavena <redmine@...>
Bug #1412: 1.8.7-p160 extmk.rb fails when cross compiling
5 messages
2009/04/26
[ruby-core:23310] Re: [Bug #1403] Process.daemon should do a double fork to avoid problems with controlling terminals
From:
Gary Wright <gwtmp01@...>
Date:
2009-04-25 04:05:30 UTC
List:
ruby-core #23310
On Apr 24, 2009, at 9:10 PM, Nobuyoshi Nakada wrote:
>> setsid is not sufficient. That makes the child process a session
>> leader, establishes a new process group, and breaks the association
>> with the controlling terminal, but it doesn't prevent the process
>> from
>> acquiring a controlling terminal if it should happen to open a tty
>> device (generally on system V based systems, BSD systems are
>> different).
>
> Then `i686-darwin9' is irrelevant at all?
I may have confused things by including ruby -v from my laptop. I was
making a more general claim about Process.daemon--not something
specific to Mac OS X.
>> The second fork ensures that the process (the grandchild of the
>> original process) is no longer a session leader and therefore can't
>> acquire a controlling terminal by opening a tty device.
>
> Could you show concrete example?
I'm not sure if you are asking for a code example or just a better
explanation of the consequences of having a controlling terminal.
<http://www.faqs.org/faqs/unix-faq/programmer/faq/>
Has a good rundown of creating a daemon including the second fork. The
umask recommendation also might be worth considering.
As for a controlling terminal, <http://opengroup.org/onlinepubs/007908799/xbd/termios.html
> has lots of the gory details of controlling terminals.
The point is that the process probably doesn't *want* a controlling
terminal (otherwise it wouldn't have called Process.daemon in the
first place) and so to be sure it doesn't get one inadvertently if it
opens a tty device, a second fork should be done after the setsid:
proc_setsid();
switch (rb_fork(0, 0, 0, Qnil)) {
case -1:
return (-1);
case 0:
break;
default:
_exit(0);
}
After the second fork, the remaining process is no longer a session
leader and so can't inadvertently get a controlling terminal.
I admit that this is an obscure case. Perhaps Process.daemon should be
extended with some options so that the behavior can be tuned as needed:
Process.daemon(:chdir => "/tmp", :double_fork => true, :umask => 0)
Gary Wright