Re: Bug#416824: klaptopdaemon: lock & hibernate allowing unauthorised access

FORWARDED 416824
HTTPS://BUGS.KDE.ORG/SHOW_BUG.CGI?ID=143859
THANKS

EL MARTES, 3 DE ABRIL DE 2007 00:01, SUNE VUORELA ESCRIBIó:
> > > HI, I'M USING ETCH RC2 AND I USE
KLAPTOPDAEMON TO LOCK AND HIBERNATE MY
> > > LAPTOP WHEN I NOTICED AN INTERESTING LITTLE
BUG.  I ACCESS LOCK AND
> > > HIBERNATE BY RIGHT-CLICKING ON THE SYSTEM
TRAY ICON AND CLICKING ON THE
> > > OPTION THERE.
>
> I HAVE ABSOLUTELY ONLY VERY LITTLE CLUE ABOUT THIS, BUT
I TOOK A LOOK AT
> THE KLAPTOPDAEMON SOURCES - AND IT SEEMS IT INVOKES THE
LOCK COMMAND AND
> AFTER THAT IT INVOKES HIBERNATE.
>
> IT MIGHT BE THAT THE LOCK COMMAND RETURNS IMMEDIATELY
WHILE THE LOCKING CAN
> BE DELAYED FOR SOME REASONS.
>
  AFTER SOME CONVERSATION ON #KDE THEY POINTED ME THE REASON
IS INDEED THAT, 
BUT INSTEAD OF CARRYING OUT THE TEST SUNE PROPOSED IT
SUGGESTED ME A "SMARTER 
KDE WAY" (TM) WHICH CONSISTS OF SUBSTITUING THE
ASYNCHRONOUS "SEND" DCOP API 
CALL BY A "CALL" WHICH IS SYNCHRONOUS.

  I ALSO OPENED A BUG ON KDE AND THERE IS A PROPOSED PATCH

-- 
     RAúL SáNCHEZ SILES
----->PROUD DEBIAN USER<-----
LINUX REGISTERED USER #416098

Hola Raúl,

On Thu, Apr 05, 2007 at 02:41:32AM +0200, Raúl Sánchez Siles
wrote:
> forwarded 416824 https://b
ugs.kde.org/show_bug.cgi?id=143859
> thanks
> 
> El Martes, 3 de Abril de 2007 00:01, Sune Vuorela
escribió:
> > > > Hi, I'm using Etch RC2 and I use
klaptopdaemon to lock and hibernate my
> > > > laptop when I noticed an interesting
little bug.  I access lock and
> > > > hibernate by right-clicking on the
system tray icon and clicking on the
> > > > option there.
> >
> > I have absolutely only very little clue about
this, but I took a look at
> > the klaptopdaemon sources - and it seems it
invokes the lock command and
> > after that it invokes hibernate.
> >
> > It might be that the lock command returns
immediately while the locking can
> > be delayed for some reasons.
> >
>   After some conversation on #kde they pointed me the
reason is indeed that, 
> but instead of carrying out the test Sune proposed it
suggested me a "smarter 
> kde way" (TM) which consists of substituing the
asynchronous "send" dcop API 
> call by a "call" which is synchronous.
> 
>   I also opened a bug on kde and there is a proposed
patch
>

Let's see what upstream says about the patch. Thanks a lot
for you work on
this!

Ana




-- 
To UNSUBSCRIBE, email to debian-kde-REQUESTlists.debian.org
with a subject of "unsubscribe". Trouble? Contact
listmasterlists.debian.org