[FreeNX-kNX] nxagent not receiving KeyRelease events

Andrew Chuah hachuah at gmail.com
Mon Oct 30 02:10:53 UTC 2006


This is irritating... I'm genuinely trying to help nomachine fix a bug
in their code, and there is no reply from anyone! Does anyone from
nomachine actually read this list? As it stands currently, the using
OSS nxagent/nxproxy to create an NX link is totally broken because of
this bug (only on some platforms, perhaps). I've actually invested my
whole weekend + 2 work days looking into the code and trying to
understand it, so don't say I haven't tried.

On 10/26/06, Andrew Chuah <hachuah at gmail.com> wrote:
> More debug; for whatever reason, it looks like nxagentWakeupHandler is
> not called during a KeyRelease event:
>
> nxagentWakeupHandler: Not reading the X events with count [1].
> nxagentBlockHandler: Trying a timeout of [0] Ms with [4] bytes and count [1].
> nxagentBlockHandler: Former select timeout was [1500] Ms.
> nxagentWakeupHandler: Flushing the NX link with [4] bytes and count [1].
> nxagentWakeupHandler: Reading the X events with count [1].
> nxagentDispatchEvents: Running
> nxagentDispatchEvents: Going to handle new event type [2].
> nxagentDispatchEvents: Going to handle new KeyPress event.
> nxagentCheckSpecialKeystroke: got code 26 - state 0 - sym 61
>
>
> On 10/26/06, Andrew Chuah <hachuah at gmail.com> wrote:
> > More debug done...
> >
> > When I first startup, I get:
> > nxagentDispatchEvents: Going to handle new event type [2].
> > nxagentDispatchEvents: Going to handle new KeyPress event.
> > nxagentCheckSpecialKeystroke: got code 26 - state 0 - sym 61
> > KeyPress
> > nxagentDispatchEvents: Going to handle new event type [3].
> > nxagentDispatchEvents: Going to handle new KeyRelease event.
> > nxagentCheckSpecialKeystroke: got code 26 - state 0 - sym 61
> > KeyRelease
> >
> > "KeyPress" & "KeyRelease" are fprintfs inserted into the NXevents.c code.
> >
> > I notice that it breaks (sometimes) upon window resizing. After it
> > breaks, I notice that the KeyRelease/KeyPress debug messages are
> > opposite of what they were before.
> >
> > nxagentDispatchEvents: Going to handle new event type [2].
> > nxagentDispatchEvents: Going to handle new KeyPress event.
> > nxagentCheckSpecialKeystroke: got code 26 - state 0 - sym 61
> > KeyPress
> > nxagentDispatchEvents: Going to handle new event type [2].
> > nxagentDispatchEvents: Going to handle new KeyPress event.
> > nxagentCheckSpecialKeystroke: got code 26 - state 0 - sym 61
> > KeyRelease
> > KeyPress
> > nxagentDispatchEvents: Going to handle new event type [2].
> > nxagentDispatchEvents: Going to handle new KeyPress event.
> > nxagentCheckSpecialKeystroke: got code 26 - state 0 - sym 61
> > KeyRelease
> > KeyPress
> >
> > I'm still trying to figure out how Event.c and NXevents.c are connected.
> >
> > -andrew
> >
> > On 10/26/06, Andrew Chuah <hachuah at gmail.com> wrote:
> > > Hi,
> > >
> > > I'm trying to debug a problem in the OSS 2.1.0 nxagent/nxproxy
> > > combination. When things are going well, I get:
> > >
> > > nxagentCheckSpecialKeystroke: got code 26 - state 0 - sym 61
> > > nxagentDispatchEvents: Going to handle new event type [3].
> > > nxagentDispatchEvents: Going to handle new KeyRelease event.
> > > nxagentCheckSpecialKeystroke: got code 26 - state 0 - sym 61
> > > nxagentDispatchEvents: Going to handle new event type [2].
> > > nxagentDispatchEvents: Going to handle new KeyPress event.
> > > nxagentCheckSpecialKeystroke: got code 38 - state 0 - sym 62
> > > nxagentDispatchEvents: Going to handle new event type [3].
> > > nxagentDispatchEvents: Going to handle new KeyRelease event.
> > > nxagentCheckSpecialKeystroke: got code 38 - state 0 - sym 62
> > > nxagentDispatchEvents: Going to handle new event type [90].
> > >
> > > However, after awhile, I stop getting KeyRelease events...
> > >
> > > nxagentDispatchEvents: Going to handle new event type [2].
> > > nxagentDispatchEvents: Going to handle new KeyPress event.
> > > nxagentCheckSpecialKeystroke: got code 29 - state 0 - sym 66
> > > nxagentDispatchEvents: Going to handle new event type [2].
> > > nxagentDispatchEvents: Going to handle new KeyPress event.
> > > nxagentCheckSpecialKeystroke: got code 29 - state 0 - sym 66
> > > nxagentDispatchEvents: Going to handle new event type [2].
> > > nxagentDispatchEvents: Going to handle new KeyPress event.
> > > nxagentCheckSpecialKeystroke: got code 29 - state 0 - sym 66
> > > nxagentDispatchEvents: Going to handle new event type [2].
> > > nxagentDispatchEvents: Going to handle new KeyPress event.
> > > nxagentCheckSpecialKeystroke: got code 29 - state 0 - sym 66
> > >
> > > This seems to be causing my Ctrl/Shift/Alt keys to become sticky.
> > >
> > > I'm running this on the remote host:
> > > nxagent -R -display
> > > nx/nx,link=modem,pack=4k-png-jpeg-1,nodelay=1,limit=0:1000 :2000 &
> > > xterm &
> > >
> > > On the local host:
> > > nxproxy -S <remote host>:1000
> > >
> > > Any ideas on how to continue to debug this would be great!
> > >
> > > thx
> > > andrew
> > >
> >
>



More information about the FreeNX-kNX mailing list