rtpinject.py - FIX

[S3M73X]

Quick-Fix for rtpinject.py



Tested on: BackTrack 4 final
Verified that tool is doing it's job: yes

1. Download Dependencies

mkdir /usr/src/rtpinject && cd /usr/src/rdpinject
wget https://www.isecpartners.com/files/R...dencies.tar.gz
tar xzvf RTPInject.Dependencies.tar.g
cd RTPInject.Dependencies/libdnet-1.11/python

2. Fix the Sourcecode of dnet.c for the built of libdnet with python
in line 2729 change

((PyObject*)__pyx_v_next) = Py_None; Py_INCREF(((PyObject*)__pyx_v_next));

to

(__pyx_v_next) = Py_None; Py_INCREF(((PyObject*)__pyx_v_next));

and in line 2741 change

((PyObject *)__pyx_v_next) = __pyx_3;

to

(__pyx_v_next) = __pyx_3;

now install the libdnet with python-support

cd .. # you should be in the /usr/src/rtpinject/libdnet-1.11 folder right now
configure --with-python && make all && make install

now you also need some gstreamer plugins so you are able to inject e.g. a wav-file which then is automaticially converted into rtp-conform speech-payload. if you want to skip this step you have to convert each file you want to inject by hand e.g. with sox. its easier to just make a .wav file with e.g. the windows-audio-recorder and inject it so i suggest you install the plugins

apt-get install gstreamer0.10-plugins-base gstreamer0.10-plugins-good

thats it... now the tool should simply work if you call it from commandline via

cd /pentest/voip/rtpinject/
python rtpinject.py

I tested it and here is a screenie of a successfull attack that i just performed to proof it works.


Hope this will help some ppl and this will get a "cleaner" fix soon ^^
Sometimes you have to restart the tool since the auto-convert function seems a little buggy *maybe because i removed some typecast*

*thx to archangelamael for psychological support + motivation ;D*

[Archangel-Amael]

I have submitted this.
S3M73X thanks for your help today.

There is a post with more info on this bug however it is in German.
http://www.backtrack-linux.org/forum...ecke/30441.htm