Re: [Discuss-gnuradio] GRC seg fault

[Koslowski, Sebastian (CEL)]

On 07/17/2015 02:27 PM, Jason Matusiak wrote:
> Hello,
>
> I had a nicely working GRC on an Ubuntu 14.04 VM.  I then installed the
> GR-CDMA module via PyBombs.  I was going through the steps to build the
> different packages and GRC started acting weird.  I closed it down and
> then when I reopened it, it throws this error:
> $ gnuradio-companion
> Segmentation fault (core dumped)
>
> I ran:
> $ gdb --args python $(which gnuradio-companion)
>
> followed by "run", and it produces:
> (gdb) run
> Starting program: /usr/bin/python /usr/local/bin/gnuradio-companion
> [Thread debugging using libthread_db enabled]
> Using host libthread_db library
> "/lib/x86_64-linux-gnu/libthread_db.so.1".
> [New Thread 0x7fffed061700 (LWP 24737)]
> [New Thread 0x7fffec860700 (LWP 24738)]
> Traceback (most recent call last):
>   File
> "/usr/share/gdb/auto-load/usr/lib/x86_64-linux-gnu/libstdc++.so.6.0.19-gdb.py",
> line 63, in <module>
>     from libstdcxx.v6.printers import register_libstdcxx_printers
> ImportError: No module named 'libstdcxx'
> [New Thread 0x7fffde7f6700 (LWP 24739)]
> [New Thread 0x7fffddff5700 (LWP 24740)]
> [New Thread 0x7fffdb7f4700 (LWP 24741)]
> [New Thread 0x7fffd8ff3700 (LWP 24742)]
> [New Thread 0x7fffd67f2700 (LWP 24743)]
>
> Program received signal SIGSEGV, Segmentation fault.
> 0x00000000000124d6 in ?? ()
>
>
> What is going wrong? Is there anything I can do to narrow this down
> further?  I went into the gnuradio/build and did a make clean && make &&
> sudo make install, but that does not fix anything.
>
> Any tips would be appreciated!
>
>
> _______________________________________________
> Discuss-gnuradio mailing list
> address@hidden
> https://lists.gnu.org/mailman/listinfo/discuss-gnuradio

GRC attempts to read the doc string of each block during load. If the
import of a block seg faults, so does GRC. If all you changed is install
gr-cdma, that's probably the reason for GRC failing. Looks like your
installation of gr-cdma is broken.

FYI, moving the doc string extraction this into a separate process
shifts this problem to runtime (when you execute a flowgraph with
erroneous blocks). I hope have this merged soon.

Sebastian