Snort mailing list archives

Re: Possible FP 17390


From: Joel Esler <jesler () sourcefire com>
Date: Tue, 20 Sep 2011 17:30:26 -0400

Rmkml,

Actually none of the above.

The vulnerability has to do with two particular ResourceID's that could be present in an APP13 section of a jpeg.  This 
will cause ClamAV 94.2 and prior to go into infinite recursion when trying to process a jpeg thumbnail.  Eventually 
clamd will shutdown, thusly, a DoS.

James --

After looking at the pcap you sent me offlist, the pcap DOES contain a vulnerable jpeg that would DoS an older version 
of ClamAV.  (read:  This isn't a false positive)  

If you don't have ClamAV on the network (or it's >94.2) you can shut the rule off.  Otherwise... :)

--
Joel Esler
Senior Research Engineer, VRT
OpenSource Community Manager
Sourcefire


 
On Sep 20, 2011, at 5:02 PM, rmkml wrote:

Thx you James and Joel,
Exploit here:
http://downloads.securityfocus.com/vulnerabilities/exploits/32555.c
Im curious if this exploit DoS because:
-jpeg loop? (iter 200000)
-marker APP13 (\xffed) section size too short? (\x0002)
-photoshopthumbnail tag size too big? \x01010101)
-or combinations?
Regards
Rmkml


On Tue, 20 Sep 2011, Joel Esler wrote:

James, can you send it to me in just pcap form?  I just want to make sure that any false positives are eliminated.
I had a hunch it was 2.9.1, do you have PAF enabled in your config?
Joel

On Sep 20, 2011, at 4:07 PM, Lay, James wrote:

-----Original Message-----
From: Joel Esler [mailto:jesler () sourcefire com]
Sent: Tuesday, September 20, 2011 12:39 PM
To: Lay, James
Cc: snort-sigs () lists sourceforge net
Subject: [Spam] Re: [Snort-sigs] Possible FP 17390
Importance: Low

James,

After looking at this, I have a couple questions:

#1 -- What version of Snort are you using?
#2 -- The vulnerability condition is not in this packet.  What are you using
to log?  You may see the actual logged vulnerability condition in an
additional packet later, in perhaps a "tagged" packet (depending on what
interface you are using).  But I can't see the vulnerability condition on
the paste you included below.
#3 -- We actually used a jpg to replicate this condition in ClamAV.

So, a couple points.  First, if you can get full packet capture of the
vulnerability condition and surrounding packets, that'd be great.
Because I can't see the vuln here.  Second, if you aren't running ClamAV,
I suggest you shut this rule off anyway.

--
Joel Esler
Senior Research Engineer, VRT
OpenSource Community Manager
Sourcefire

Hey Joel,

This is snort 2.9.1 using 2.9.1 VRT rules.  The packet cap is via
sguil/barnyard2/squert...I don't think it's logging everything, but I
got the pcap out of the unified2 file and here's the only other packet,
but I can see where it hit now.  Think I'll kill this rule...thanks
Joel.

James

00000578  58 59 5a 20 00 00 00 00  00 00 9c 18 00 00 4f a5 XYZ ....
00000588  00 00 04 fc 58 59 5a 20  00 00 00 00 00 00 34 8d ....XYZ
00000598  00 00 a0 2c 00 00 0f 95  58 59 5a 20 00 00 00 00 ...,.... XYZ
000005A8  00 00 26 31 00 00 10 2f  00 00 be 9c ff ed 5b da ..&1.../
000005B8  50 68 6f 74 6f 73 68 6f  70 20 33 2e 30 00 38 42 Photoshop 3.0.8B
000005C8  49 4d 04 0c 00 00 00 00  5b a2 00 00 00 01 00 00 IM......
000005D8  01 00 00 00 00 aa 00 00  03 00 00 01 fe 00 00 00 ........
000005E8  5b 86 00 18 00 01 ff d8  ff ee 00 0e 41 64 6f 62 [.......
000005F8  65 00 64 00 00 00 00 01  ff db 00 84 00 06 04 04 e.d.....
00000608  04 05 04 06 05 05 06 09  06 05 06 09 0b 08 06 06 ........
...


------------------------------------------------------------------------------
All the data continuously generated in your IT infrastructure contains a
definitive record of customers, application performance, security
threats, fraudulent activity and more. Splunk takes this data and makes
sense of it. Business sense. IT sense. Common sense.
http://p.sf.net/sfu/splunk-d2dcopy1
_______________________________________________
Snort-sigs mailing list
Snort-sigs () lists sourceforge net
https://lists.sourceforge.net/lists/listinfo/snort-sigs
http://www.snort.org


Please visit http://blog.snort.org for the latest news about Snort!


Current thread: