Re: [GnomeMeeting-list] Problem :: Gnome Libraries + GM1.2.1

[Ashutosh Sharma <asharma samyak com>]

Well thanks for your response.

   Anyway regarding the 'bt' command then when I start debugging GM with
gdb then when that error occurs, I do not get the gdb prompt where I can
get the total backtrace.

And one more thing, I am sure you are gonna hit me , but its a fact that
the problem that I conveyed here is not at every time whenever I start
GM with my little modifications in the openh323. It is at random times.
That XLib error has got hell out of me! Totally stuck because I am not
able to decide what setup should I use for a stable system.
Anyway regarding the compilation concept that I talked then isn't that
handled by the Makefile because there may be a lot of files in which a
particular header file is being included and even if I do changes in the
header file then I need to atleast make some change in the each
individual .cxx file that is using that file.

Also I want to know whether is there any IDE that can be used for such
projects. What about the Famous ones like KDevelop and Anjuta!!

Till Now I am using the conventional vi editor.

Thanks and regards,
Ashutosh.

On Fri, 2005-04-01 at 17:49, Damien Sandras wrote:
> Le vendredi 01 avril 2005 à 17:38 +0530, Ashutosh Sharma a écrit :
> > Hi,
> >     Sorry, but this thing has taken hell out of me... Everytime strange
> > things are happening...I did that thing atleast 10-20 times with the
> > same changes....
> > But that unexpected Async reply is coming at irregular attempts.
> > Sometimes it is working fine...
> > Anyway 1 gdb trace in which I got that prob is as follows:-
> > /////////
> > 
> > [root sipl59 root]# gdb gnomemeeting
> > GNU gdb Red Hat Linux (5.3.90-0.20030710.41rh)
> > Copyright 2003 Free Software Foundation, Inc.
> > GDB is free software, covered by the GNU General Public License, and you
> > are
> > welcome to change it and/or distribute copies of it under certain
> > conditions.
> > Type "show copying" to see the conditions.
> > There is absolutely no warranty for GDB.  Type "show warranty" for
> > details.
> > This GDB was configured as "i386-redhat-linux-gnu"...Using host
> > libthread_db library "/lib/tls/libthread_db.so.1".
> > 
> > (gdb) r
> > Starting program: /usr/bin/gnomemeeting
> > [Thread debugging using libthread_db enabled]
> > [New Thread -1084583264 (LWP 3339)]
> > [New Thread 56708016 (LWP 3346)]
> > [New Thread 21674928 (LWP 3347)]
> > [New Thread 32164784 (LWP 3348)]
> > [New Thread 32431024 (LWP 3349)]
> > [New Thread 46894000 (LWP 3350)]
> > [New Thread 60308400 (LWP 3355)]
> > [New Thread 32697264 (LWP 3356)]
> > [New Thread 32963504 (LWP 3357)]
> > [New Thread 61463472 (LWP 3358)]
> > 
> > H323_RTPChannel::Open() Receivetime(ms): 1112349229471: created using
> > session 1
> > [New Thread 71924656 (LWP 3360)]
> > H323_RTPChannel::Open() Transmittime(ms): 1112349229511: created using
> > session 1
> > Xlib: unexpected async reply (sequence 0x136e)!
> > [New Thread 101002160 (LWP 3361)]
> > Jitter buffer constructor:time(ms): 1112349229515
> > [New Thread 33229744 (LWP 3362)]
> > H323_RTPChannel::Open() Receivetime(ms): 1112349231334: created using
> > session 2
> > [New Thread 33495984 (LWP 3364)]
> > H323_RTPChannel::Open() Transmittime(ms): 1112349231378: created using
> > session 2
> > 
> 
> When it crashes, you need to type "bt" to see where it is crashing and
> start debugging.
> 
> 
> > ////////
> > 
> >  And here I get the crash dialog box.
> > 
> > Also the thing that you pointed out regarding the compilation of the
> > full source of gnomemeeting and openh323, sorry but I am bothered about
> > one point:-
> > 
> > I am not changing anything in GM, but only in the OpenH323 then would I
> > need to recompile the GM also. I suppose no, because I am compiling the
> > OpenH323. And as suppose I only change the files like jitter.cxx and
> > rtp.cxx then I need to do only make && make install and that would
> > compile the changed files only. I don't think that I need to clean the
> > compilation and then recompile each and every file.
> > 
> 
> Only if you change a header.
> 
> > Please correct me if I am wrong in the following context as suppose I
> > change jitter.h then I need to recompile only jitter.cxx or all those
> > files that use jitter.h. I think make file takes care of this.
> > 
> 
> All files that use jitter.h and gnomemeeting.
> 
> > Sorry for such problems, but I am not getting anything in my hand
> > because of such problems.
> > 
> > Thanks and regards,
> > Ashutosh
> > 
> > On Fri, 2005-04-01 at 14:46, Damien Sandras wrote:
> > > Hi,
> > > 
> > > 
> > > Le vendredi 01 avril 2005 �  12:23 +0530, Ashutosh Sharma a �©crit :
> > > > Hi,
> > > >    Sorry for being late in sending the log related to Async Reply
> > > > Message. I have attched the Log file in this mail.
> > > > 
> > > 
> > > What is really needed is a GNU GDB backtrace.
> > > 
> > > 
> > > > The kind of setup that I had was this:-
> > > >  Terminal A:- GM-1.2.1 (with corresponding PWLib/OpenH323 and Gnome
> > > > Libraries 2.6 series)
> > > > 
> > > >  Terminal B:- GM-0.98.5 (with corresponding PWLib/OpenH323 binar, simply
> > > > the standard FEDORA CORE 1 package.)
> > > > 
> > > > Now in the jitter.h file I declared a variable in the Class
> > > > RTP_JitterBuffer as follows:-
> > > >                              PTimeInterval my_start_time;  /// Statement
> > > > --1
> > > > Correspondingly in jitter.cxx I did the following things in the
> > > > constructor:-
> > > >  RTP_JitterBuffer::RTP_JitterBuffer(RTP_Session & sess,
> > > >                                    unsigned minJitterDelay,
> > > >                                    unsigned maxJitterDelay,
> > > >                                    PINDEX stackSize)
> > > >   : PThread(stackSize, NoAutoDeleteThread, HighestPriority, "RTP
> > > > Jitter:%x"),session(sess)
> > > > {
> > > > 
> > > >  my_start_time=PTimer::Tick(); //statement 2
> > > >  cout<<"No. of Miliseconds=="<<my_start_time.GetMilliSeconds();
> > > > //statement 3
> > > > 
> > > > ///
> > > > /// Default things that are already there
> > > > ///
> > > > }
> > > > 
> > > > Conclusion:-
> > > > 1).Suppose I just put a only a cout<<"Message here".. I mean no
> > > > expression just a string, then it works fine.
> > > > 2).But suppose I put either statement 1 or statement 2 or statement 3 or
> > > > all then I get the problem of Unexpected Async Reply for which I have
> > > > attached the log file.
> > > > 
> > > 
> > > OK, so it is related to your change.
> > > You have modified a header, so are you sure you had recompiled the
> > > corresponding GnomeMeeting and OpenH323 entirely?
> > > 
> > > > 3). Also supposing I put the above lines ( statements 1, 2 and 3) in
> > > > some other function say where the jitter buffer calculation is done,
> > > > then the things go smoothly .
> > > > 
> > > > 4). Also suppose I declare the some double or int type of variable in
> > > > the jitter.h instead of PTimeInterval then it goes fine. And suppose I
> > > > declare PTimeInterval variable in the jitter.h and does not intialize
> > > > and use that variable then still it gives me the error. I mean it will
> > > > give warning during compilation. In each case I recompiled the
> > > > GnomeMeeting source also with make && make install.
> > > > 
> > > 
> > > Really, the backtrace will help telling where it crashes.
> > > 
> > > > 5). Now another thing that I noted down was that that supposing the
> > > > A-->B call, then when the GM1.2.1 hangs due to the above mentioned
> > > > problems, it does not receive any thing, but the terminal B that is
> > > > GM-0.98.5 shows me connected and does the transmission part. I mean B
> > > > does not know whether A is getting text message or not.
> > > > 
> > > 
> > > Yes, only the jitter thread is crashed, that explains it.
> > > 
> > > > Please throw some lights in this context.
> > > > 
> > > > Regards,
> > > > Ashutosh.
> > > >   
> > > > 
> > > > 
> > > > On Thu, 2005-03-31 at 13:48, Damien Sandras wrote:
> > > > > Le jeudi 31 mars 2005 Ã?Æ?  13:37 +0530, Ashutosh Sharma a Ã?Æ?Ã?©crit :
> > > > > > Hi well thanks for the instant reply.
> > > > > >  Yes actually the modification was done in the OpenH323. No modification
> > > > > > in GnomeMeeting. And then I reverted back the changes, only a one line
> > > > > > change in rtp.h and two lines in rtp.cxx.
> > > > > > Then I recompiled both GM and OpenH323. But getting the unexpected async
> > > > > > reply...
> > > > > 
> > > > > And what is the backtrace?
> > > > > 
> > > > > > Well sorry but I know that the same prob has already been discussed in
> > > > > > the threads and after all  
> > > > > > 
> > > > > > Well I have a plan to build a clean system, I mean full source code
> > > > > > compilation on my FC1 system. So can you please tell me what libraries
> > > > > > should I use with GM1.2.1 and its associated PWLib and OpenH323 that are
> > > > > > there on the www.gnomemeeting.org site.
> > > > > > 
> > > > > 
> > > > > Configure will tell you what's missing.
> > > > > 
> > > > > > I mean full dependency list for FEDORA CORE 1.
> > > > > > I initially tried the GTK 2.4.4 packages but those were having some
> > > > > > bugs, I suppose. Couldn't compile.
> > > > > > 
> > > > > > THanks.
> > > > > > Ashutosh
> > > > > > 
> > > > > > On Thu, 2005-03-31 at 12:51, Damien Sandras wrote:
> > > > > > > Hi,
> > > > > > > 
> > > > > > > Is it a modified GnomeMeeting/OpenH323 that is crashing?
> > > > > > > 
> > > > > > > Le jeudi 31 mars 2005 Ã?Æ?Ã?â??  12:45 +0530, Ashutosh Sharma a Ã?Æ?Ã?â??Ã?â??Ã?©crit :
> > > > > > > > hi,
> > > > > > > >    thanks for the response. Well very soon I'll send the backtrace. But
> > > > > > > > a little question is there. Supposing I add one variable in a class say
> > > > > > > > RTPSession, and I initialize the variable in the constructor and I use
> > > > > > > > that variable in some function of RTPSession. So whether that'll create
> > > > > > > > any problem. Because here there is a slight change in the existing code.
> > > > > > > > 
> > > > > > > 
> > > > > > > It shouldn't, but it could.
> > > > > > > 
> > > > > > > 
> > > > > > > > And I hope recompiling the OpenH323 doesn't need recompiling the GM if I
> > > > > > > > have the change only in OpenH323
> > > > > > > > 
> > > > > > > 
> > > > > > > It does if you change the headers.
> > > > > > > 
> > > > > > > 
> > > > > > > > THanks and regards,
> > > > > > > > ashutosh.
> > > > > > > > 
> > > > > > > > On Thu, 2005-03-31 at 12:19, Damien Sandras wrote:
> > > > > > > > > Le jeudi 31 mars 2005 Ã?Æ?Ã?â??Ã?â? ââ?¬â?¢  11:34 +0530, Ashutosh Sharma a Ã?Æ?Ã?â??Ã?â? ââ?¬â?¢Ã?Æ?ââ?¬Å¡Ã?â??Ã?©crit :
> > > > > > > > > > Hi all,
> > > > > > > > > >    I am in some weired situation. Hope I may get something from the list
> > > > > > > > > > as I had got earlier also, in this situation. I am having the following
> > > > > > > > > > things compiled at my desk.
> > > > > > > > > > 
> > > > > > > > > >  Gnome Libraries = 2.6.x
> > > > > > > > > >     Kernel = 2.4.22
> > > > > > > > > >     OpenH323 = 1.15.3
> > > > > > > > > >     PWLib = 1.8.4
> > > > > > > > > >     GnomeMeeting = 1.2.1 (source from www.gnomemeeting.org)
> > > > > > > > > > 
> > > > > > > > > >     XLib: unexpected Async reply (# some hex value!).
> > > > > > > > > > 
> > > > > > > > > >     Also sometimes I get the message like Abnormal Call Termination. And
> > > > > > > > > > if I backtrace that then I find that it shows me some
> > > > > > > > > > EndedByTransportFail.
> > > > > > > > > > 
> > > > > > > > > >     Well I am not getting any thing out of this.
> > > > > > > > > > 
> > > > > > > > > 
> > > > > > > > > Can you send a full backtrace?
> > > > > > > > > 
> > > > > > > > > 
> > > > > > > > > >     ANd also after installing the Gnome 2.6 libraries my system has
> > > > > > > > > > started giving weired responses. Anytime it can hang. I thought that the
> > > > > > > > > > problem may be in the kernel compatibility, but I installed kernel 2.6
> > > > > > > > > > also, but in that don't know why but my XServer is not working....
> > > > > > > > > > 
> > > > > > > > > 
> > > > > > > > > Installing gnome 2.6 is certainly not related to your system hangs. It
> > > > > > > > > is by all means kernel related.
> > > > > > > > > 
> > > > > > > > > Try to see in /var/log/messages if you have something.
> > > > > > > > > 
> > > > > > > > > In general, it can be due to a buggy usb driver. Are you using a webcam
> > > > > > > > > with GM?
> > > > > > > > > 
> > > > > > > > > 
> > > > > > > > > >     I am using FEDORA CORE1.
> > > > > > > > > >     Can you please let me out of this weired situtation? And is it
> > > > > > > > > > necessary to have some kernel compatibility with the Gnome Libraries 2.6
> > > > > > > > > > with the GM 1.2.1.
> > > > > > > > > 
> > > > > > > > > Not at all.
> > > > > > > > > 
> > > > > > > > > > 
> > > > > > > > > >     THanks and regards,
> > > > > > > > > >     AShutosh.
> > > > > > > > > > 
> > > > > > > > > > 
> > > > > > > > > > 
> > > > > > > > > > _______________________________________________
> > > > > > > > > > GnomeMeeting-list mailing list
> > > > > > > > > > GnomeMeeting-list gnome org
> > > > > > > > > > http://mail.gnome.org/mailman/listinfo/gnomemeeting-list
> > > > > > > > 
> > > > > > > > _______________________________________________
> > > > > > > > GnomeMeeting-list mailing list
> > > > > > > > GnomeMeeting-list gnome org
> > > > > > > > http://mail.gnome.org/mailman/listinfo/gnomemeeting-list
> > > > > > 
> > > > > > _______________________________________________
> > > > > > GnomeMeeting-list mailing list
> > > > > > GnomeMeeting-list gnome org
> > > > > > http://mail.gnome.org/mailman/listinfo/gnomemeeting-list
> > > > _______________________________________________
> > > > GnomeMeeting-list mailing list
> > > > GnomeMeeting-list gnome org
> > > > http://mail.gnome.org/mailman/listinfo/gnomemeeting-list
> > 
> > _______________________________________________
> > GnomeMeeting-list mailing list
> > GnomeMeeting-list gnome org
> > http://mail.gnome.org/mailman/listinfo/gnomemeeting-list