[37741] in bugtraq
Re: Linux kernel scm_send local DoS
daemon@ATHENA.MIT.EDU (even multiplexed)
Wed Dec 15 16:37:26 2004
Message-ID: <41C03386.3040401@gmx.at>
Date: Wed, 15 Dec 2004 13:52:22 +0100
From: even multiplexed <Shadow333@gmx.at>
MIME-Version: 1.0
To: Paul Starzetz <ihaquer@isec.pl>
Cc: security@isec.pl, bugtraq@securityfocus.com, vulnwatch@vulnwatch.org,
full-disclosure@lists.netsys.com
In-Reply-To: <Pine.LNX.4.44.0412151328540.1826-100000@isec.pl>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Paul Starzetz wrote:
>On Wed, 15 Dec 2004, even multiplexed wrote:
>
>
>
>>attention.i just wanted to ask if anyone has a tip for me how to
>>quickfix this bug, without actually rebuilding a patched version of the
>>kernel.
>>
>>
>
>I don't think this is practicable, since the bugs reside in deep kernel
>functions. You can not fix it just by disabling a particular syscall. You
>have patch a running kernel binary, maybe someone comes up with this kind
>of utlility.
>
>
>
well, i was also examining the igmp exploit you posted earlier, that one
was relatively easy to disable, by setting the following sysctl vars:
net.ipv4.igmp_max_msf = 0
net.ipv4.igmp_max_memberships = 0
net.ipv4.icmp_ignore_bogus_error_responses = 1
since i dont really need igmp support on that server=)
its just that scm_send thing, that i didnt find any config option for...
so i guess that means either waiting till someone of the kernel
maintainers release a patch, or get one of my friends, that is better on
programming tasks, to fix that one...
all at all makes for quite a risky day for a shell provider...
