Hi Rodolfo,
* Gregor Zattler <telegr...@gmx.net> [12. Jul. 2013]:
> * Rodolfo García Peñas <k...@kix.es> [12. Jul. 2013]:
>> Probably the bug could be here:
>>
>> module new.c:
>> 173 /* Work around stupid X11 bug: When switching immediately after the
>> command
>> 174 * is entered, the enter button may get stuck. */
>> 175 if (getenv("DISPLAY") != NULL)
>> 176 sleep(1);
>>
>> Probably the problem is not only related to the enter key.
>>
>> Can you reproduce the bug yet? Then, the best idea could be run
>> vlock without the module "new". Then we can see if the bug is
>> in that module and work on it. Because we are in X11, run vlock
>> -sa is not possible and we need the "new" module.
>>
>> Perhaps, one idea is increase the sleep time, for testing, something like:
>>
>> 176 sleep(5);
>>
>> Could you do this modification in the source and build a new
>> package? Else, I could create a test package for you, only with
>> this change. Tell me :-)
>
> If I understand you correctly, these are two options a) sleeping
> longer, b) using vlock -sa, therefore not using X11.
>
> Since I mostly use X11, I opt for the first test first. I
> changed the sleeping time to 9 seconds and will report back in a
> few days. If this does not change anything I will give non
> X11-usage a try.
I used vlock -san since this last E-Mail with sleep (9): I
realized only two hangs when the laptop was suspended and I woke
it up. This is something I do often and it hanged only twice.
Otherwise vlock functioned perfect -- with the exeption of having
to wait for 9 secs. Which is still a security problem since I
have to call vlock -san *after* the laptop wakes up from suspend
(doesn't suspend while vlocked). Therefore the laptop is not
protected for this waiting time after wakeup (by an attacker).
Now I changed the sleeping time to 5 secs as you suggested and will
proceede bisecting. I'll report back when I know how long the
waiting time should be in order to function correctly.
Ciao, Gregor
--
-... --- .-. . -.. ..--.. ...-.-
--
To UNSUBSCRIBE, email to debian-bugs-rc-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
