# A crash with latest ruby/ruby-gnome **URL:** \ **Category:** Ruby-Gnome 2\ **Created:** [March 21, 2014, 10:02am UTC](https://www.ruby-forum.com/t/a-crash-with-latest-ruby-ruby-gnome/239950 "2014-03-21T10:02:55Z")\ **Posts on this page:** 6\ **Page:** 1
**Author:** ![musicdenotation](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/m/ea666f/32.png) [@musicdenotation](https://www.ruby-forum.com/u/musicdenotation)\ **Post date:** [March 21, 2014, 10:02am UTC](https://www.ruby-forum.com/t/a-crash-with-latest-ruby-ruby-gnome/239950/1 "2014-03-21T10:02:55Z")
Dear Kou and other readers, today, after updating another machine to latest Ruby/Ruby-gnome, I found that I received a crash with (I believe) every signal being processed. I attach a simple test program which, on my machine, triggers a crash when the ‘test’ button is pressed. If the btn.signal\_connect call is commented out, no crash happens. I attach the output of the crash. As you can see, it is triggered by Ruby, from thread.c, function rb\_thread\_call\_with\_gvl. It appears that it has been “called by a thread which has GVL”, because the ‘blocking region buffer’ is null. For me, these are unknown waters. Experimentally I tried to comment out the rb\_bug line in rb\_thread\_call\_with\_gvl. The app does not crash, but becomes non-responsive. Does this tell you anything? Carlo ---
**Author:** ![Carlo\_ESPrelz](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/c/f07891/32.png) [@Carlo\_ESPrelz](https://www.ruby-forum.com/u/Carlo_ESPrelz)\ **Post date:** [March 21, 2014, 11:13am UTC](https://www.ruby-forum.com/t/a-crash-with-latest-ruby-ruby-gnome/239950/2 "2014-03-21T11:13:05Z")
Hi, In [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) “[ruby-gnome2-devel-en] A crash with latest ruby/ruby-gnome” on Fri, 21 Mar 2014 10:02:08 +0100, “Carlo E. Prelz” [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) wrote: > Experimentally I tried to comment out the rb\_bug line in > rb\_thread\_call\_with\_gvl. The app does not crash, but becomes > non-responsive. Thanks for your report! The attached sample program was very helpful. I’ve fixed it but there is interrupt isn’t handled problem. I’ll fix it later. # If we change to rb\_thread\_call\_without\_gvl() from # rb\_thread\_call\_without\_gvl2(), it is solved. But we should # consider about the solution is valid solution. ## Thanks, kou ---
**Author:** ![Carlo\_ESPrelz](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/c/f07891/32.png) [@Carlo\_ESPrelz](https://www.ruby-forum.com/u/Carlo_ESPrelz)\ **Post date:** [March 21, 2014, 11:26am UTC](https://www.ruby-forum.com/t/a-crash-with-latest-ruby-ruby-gnome/239950/3 "2014-03-21T11:26:31Z")
Subject: Re: [ruby-gnome2-devel-en] A crash with latest ruby/ruby-gnome Date: ven 21 mar 14 07:11:58 +0900 Quoting Kouhei S. ([removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid)): > I’ve fixed it but there is interrupt isn’t handled > problem. I’ll fix it later. Indeed: the program I am developing runs again, but does not stop with Ctrl-C. This stuff must be quite hairy… Thanks again! Carlo – - ``` Se la Strada e la sua Virtu' non fossero state messe da ``` parte, - K \* Carlo E. Prelz - [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) che bisogno ci sarebbe - ``` di parlare tanto di amore e di rettitudine? ``` (Chuang-Tzu) ---
**Author:** ![Carlo\_ESPrelz](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/c/f07891/32.png) [@Carlo\_ESPrelz](https://www.ruby-forum.com/u/Carlo_ESPrelz)\ **Post date:** [March 21, 2014, 1:24pm UTC](https://www.ruby-forum.com/t/a-crash-with-latest-ruby-ruby-gnome/239950/4 "2014-03-21T13:24:38Z")
Hi, In [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) “Re: [ruby-gnome2-devel-en] A crash with latest ruby/ruby-gnome” on Fri, 21 Mar 2014 11:25:49 +0100, “Carlo E. Prelz” [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) wrote: > Subject: Re: [ruby-gnome2-devel-en] A crash with latest ruby/ruby-gnome > Date: ven 21 mar 14 07:11:58 +0900 > > Quoting Kouhei S. ([removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid)): > > > I’ve fixed it but there is interrupt isn’t handled > > problem. I’ll fix it later. > > Indeed: the program I am developing runs again, but does not stop with > Ctrl-C. Ctrl-C works again! 🙂 ## Thanks, kou ---
**Author:** ![Carlo\_ESPrelz](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/c/f07891/32.png) [@Carlo\_ESPrelz](https://www.ruby-forum.com/u/Carlo_ESPrelz)\ **Post date:** [March 21, 2014, 1:41pm UTC](https://www.ruby-forum.com/t/a-crash-with-latest-ruby-ruby-gnome/239950/5 "2014-03-21T13:41:00Z")
Subject: Re: [ruby-gnome2-devel-en] A crash with latest ruby/ruby-gnome Date: ven 21 mar 14 09:23:40 +0900 Quoting Kouhei S. ([removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid)): > Ctrl-C works again! 🙂 Don’t take it personally 🕶 but it only works when only the main window is open. If I have another modal window open, it still does not work. (this does not cause such a problem to me…) Since you showed appreciation for my little test program, here is another one. Press the button, and another window appears. At that point, Ctrl\_C still does not work. Or rather, the interrupt is processed only when the new window’s button is pressed. Carlo ---
**Author:** ![Carlo\_ESPrelz](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/c/f07891/32.png) [@Carlo\_ESPrelz](https://www.ruby-forum.com/u/Carlo_ESPrelz)\ **Post date:** [March 21, 2014, 2:03pm UTC](https://www.ruby-forum.com/t/a-crash-with-latest-ruby-ruby-gnome/239950/6 "2014-03-21T14:03:12Z")
Hi, In [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) “Re: [ruby-gnome2-devel-en] A crash with latest ruby/ruby-gnome” on Fri, 21 Mar 2014 13:40:35 +0100, “Carlo E. Prelz” [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) wrote: > point, Ctrl\_C still does not work. Or rather, the interrupt is > processed only when the new window’s button is pressed. Oh… The new test program also works well now. 🙂 ## Thanks, kou