Hello,

*** Chandrashekhar B <[email protected]> wrote:
> * David Corcuera [ 7. May 2009]:
> >> I've installed OpenVAS from debian package in etch and ran my first scan 
> >> against an internal host.
> >> Results: 4 security holes.
> >> Two of them are on mysql and other two on CUPS.
> >> My debian etch  has mysql 5.0.32-7etch10 and cupsys 1.2.7-4etch7 (last 
> >> official etch packages)
> >> According to OpenVAS report, i should have installed mysql 5.0.66 and 
> >> cupsys 1.3.10, but my versions also fix all these vulnerabilities.
> >> What is wrong with this? Any idea?
> 
> > I'm not really sure since I'm not a plugin author, but my first guess is
> > that the hole was fixed in MySQL 5.0.66, but Debian backported the
> > changes to the version they packaged for etch.
> 
> > I assume you are doing a remote scan; the remote scan will probably not
> > know that the hole has already been fixed in Debian despite the low
> > version number.
> 
> > Plugin authors: Am I right?
>  
> Michael, you are right. The plugin would be detecting based on the package
> available in the open source but, individual OS vendors would have
> backported. So, local checks are a better approach in this case.

Yes, banner checks are prone to false positives. There are a few
thinks we can do.

1. 
Respekt settings of "report_paranoia". 
We can do:

,--|
| if (report_paranoia < 2) exit(0);" 
`--|

on such plugins.

But then these plugins will not report about a real existing
vulnerability if the user dosn't change the default settings of
"report_paranoia". Thats the point why i don't like this solution.

2. 
Make a note in the report that this could be a false positive because
the vulnerability is only detected by checking the version from
banner.

Any other ideas? If not, i prefer option 2. :-)

> David: Please provide the Plugins that reported security holes, we'll verify
> them.

secpod_mysql_dos_vuln_900221.nasl
mysql_29106.nasl
gb_cups_mult_vuln_oct08.nasl
cups_cve_2009_0163.nasl

I couln't find any problems in that plugins.

Micha

_______________________________________________
Openvas-discuss mailing list
[email protected]
http://lists.wald.intevation.org/mailman/listinfo/openvas-discuss

Reply via email to