Re: netbios vuln

On Fri, 06 Dec 2002 06:50:02 PST, ohnonono_at_hushmail.com said:

> of netbios traffic at my main firewall. This morning I read this article. It
> seems to hint at a way to run arbitarty code via netbios, now my question is do
> es anyone know anything about this; is anyone seeing the netbios traffic and fi

We were seeing on the order of 3 *million* probes/day on port 137 back at the
beginning of October, thanks to Opaserv/Bugbear. And I'm sure the port 135
traffic was even higher, since the tools in use would only poke 137 if 135
answered. The joys of having 2 /16s hanging off an OC12.. ;)

It's not *that* easy to run arbitrary code directly via netbios. What usually
happens is a scanning tool finds some victim who's got C:\ shared read/write
to the world with no password or an Administrator password of 'password' or
something equally silly. So you mount the share, copy a trojan to it,
set the registry entries that say "run me at each boot" and then wait for
it to reboot....

See the 'W4-NETBIOS -- Unprotected Windows Networking Shares' entry about this
on the SANS Top 20:

http://www.sans.org/top20/

(And while you're there, make sure you're all square on the OTHER 19
entries too).

Disclaimer: I didn't get paid to help write the Top 20 - I just did it because
the more people that fix it at their sites, the easier *my* job gets. ;)

-- 
				Valdis Kletnieks
				Computer Systems Senior Engineer
				Virginia Tech