Re: [Help-gnunet] Error on PowerPC

[Christian Grothoff]

On Sunday 14 November 2004 05:49, Loïc Le Guyader wrote:
> Hi,
>
> I run gnunet 0.6.4a-7 on a debian testing on a powerpc and I found some
> bugs. First gnunetd give me a lot of:
>
> nov 14 11:25:27 WARNING: Received corrupt message claiming to be from
> peer '0W*' in handler.c:123.
> nov 14 11:25:31 WARNING: Received corrupt message claiming to be from
> peer '0W*' in handler.c:123.
>
> The name of the peer doesn't seem to be correct also.

Right, there was a bug 2 lines before in the statement that created that 
string.  Fixed in CVS.

> Second, gnunet-gtk crash (segfault) when I want to enter something in
> the "Search Namespace" dialog, "Namespace identifier" field. It crash
> as soon as I try to type first letter.
>
> I recompile whith debug information, run gdb on it and the backtrace
> of the crash is:
> (gdb) run
> Starting program: /usr/bin/gnunet-gtk
> [Thread debugging using libthread_db enabled]
> [New Thread 16384 (LWP 11948)]
> [New Thread 32769 (LWP 11951)]
> [New Thread 16386 (LWP 11952)]
>
> Program received signal SIGSEGV, Segmentation fault.
> [Switching to Thread 16384 (LWP 11948)]
> 0x300f1650 in mallopt () from /lib/libc.so.6
> (gdb) bt
> #0  0x300f1650 in mallopt () from /lib/libc.so.6
> #1  0x301bfa1c in ?? () from /lib/libc.so.6
> (gdb) The program is running.  Exit anyway? (y or n) y

Well, this is not really helpful (but thanks for trying). I recall someone 
else having a similar problem when entering words into the 'search' bar (but 
it might have been the namespace search bar). However, I can't reproduce 
either problem and the traces are not telling me where to look either...

> There is a so a memory leak in the application.
> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=178590

Are you sure this is the right  bug ID?

Christian