[WEB SECURITY] Poll: How do you rank the importance of a vulnerability?
rohirp92 at yahoo.com
Fri Feb 1 14:35:08 EST 2013
I would go for worst case scenario considering the weakest link principle. Also proper threat modelling would be useful to determine scoring for specific vulnerability. For enterprise level product with ideally no exposure to outside network, one might argue that libtiff issue might be of less severity but again it boils down to the context of application in analysis.
From: Paweł Krawczyk <pawel.krawczyk at hush.com>
To: websecurity at lists.webappsec.org
Sent: Thursday, January 31, 2013 11:00 PM
Subject: Re: [WEB SECURITY] Poll: How do you rank the importance of a vulnerability?
CVSS is a large step in right direction, away from subjective 1-5 or Low-High scores, but I still have a large practical problem with how it's counted. CVSS has one issue that results in typical reports being flooded by mix of really important and less important vulnerabilities that will be difficult to distinguis even using CVSS subscores. For example consider these two examples:
* remote bug in sshd
* bug in libtiff
Let's assume sshd is exploitable over network (AV=N). Let's assume libtiff can be exploited by someone who would need to open a malformed TIFF - but that would also have AV=N because it's assumed the file is delivered over network. This is at least how most of these vulns are classified in Qualys.
Obviously, the real risk is completely different in each case - sshd just sits there and waits to be exploited, for libtiff you'd need a rather rare opportunity (someone opening TIFFs on server).
I've once asked CVSS team about this and they replied that this should be theoretically captured by Access Complexity (AC) - for libtiff it AC=H (as you need to use social engineering for example), for sshd AC=L (just go and metasploit over network).
But in real life scores of both vulns will be very similar. At the end of the day you end up with a report flooded by say 100 issues for each server, out of which usually all will be like that libtiff. And you have no way to filter them out to focus on sshd-type vulns because of how the classification is calculated.
Paweł Krawczyk, CISSP
+48 602 776959
On 1/2/2013 at 12:58 AM, "Robert A." <robert at webappsec.org> wrote:
Where would you suggest this poll be held? Keep in mind I have no time to
>create or implement a polling application :)
>On Thu, 31 Jan 2013, MustLive wrote:
>> Hello Robert!
>> It's interesting poll and webappsec professionals and experts could be
>> interested to participate in the poll. But you've selected not appropriate
>> place for opening the poll.
>> You've opened it in LinkedIn. That one, which was hacked last year. This is
>> social network and none of social networks are attending to security. It's
>> not the place for polls on web applications security topics ;-). Plus I'm
>> not using LinkedIn and any s.networks at all (and the poll requires
>> registration in it).
>> I am agree with Phillip's standpoint.
>> Best wishes & regards,
>> Eugene Dokukin aka MustLive
>> Administrator of Websecurity web site
>> Robert A. robert at webappsec.org
>> Wed Jan 9 18:35:02 EST 2013
>>> I've added a new poll to the WASC linkedin group that a few of you may be
>>> interested in. Specifically asking how people rank the importance of
>>> Robert A.
>>> WASC Co Founder/Moderator of The Web Security Mailing List
>The Web Security Mailing List
>WebSecurity RSS Feed
>Join WASC on LinkedIn http://www.linkedin.com/e/gis/83336/4B20E4374DBA
>WASC on Twitter
>websecurity at lists.webappsec.org
The Web Security Mailing List
WebSecurity RSS Feed
Join WASC on LinkedIn http://www.linkedin.com/e/gis/83336/4B20E4374DBA
WASC on Twitter
websecurity at lists.webappsec.org
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the websecurity