From: Mart Frauenlob <mart.frauenlob@chello.at>
To: netfilter@vger.kernel.org
Subject: Re: Howto match the 3rd packet in the 3way handshake
Date: Sun, 28 Feb 2010 12:36:16 +0100 [thread overview]
Message-ID: <4B8A5530.7050007@chello.at> (raw)
In-Reply-To: <4B8A4D88.5060901@iit.bme.hu>
On 28.02.2010 12:03, netfilter-owner@vger.kernel.org wrote:
> Mart Frauenlob wrote:
>> On 28.02.2010 10:58, netfilter-owner@vger.kernel.org wrote:
>>
>>> Dear All,
>>>
>>> The situation is the following: We have two host A and B, A sends B a
>>> SYN packet with a spoofed IP address,
>>> since the source IP is spoofed B will not receive the 2. packet of the
>>> handshake, but is able
>>>
>>
>> it think you mean A?
>>
> yes sorry typo
>>> to send back the 3rd: an ACK packet with an invalid acknowledgement
>>> number. How is it
>>> possible distinguish connections in which the source IP is spoofed in
>>> this way.
>>>
>>> 1. A ->B : SYN(IP_X, seq(A))
>>> 2. B ->A : SYN_ACK(IP_X, ack(A), seq(B))
>>> 3. A ->B : ACK(IP_X. seq(A+1), ack(Y))
>>>
>>> How is it possible to match the 3rd packet if Y=A+1, and Y!=A+1?
>>>
>>> I would like to use this for the following. Let us assume that port 222
>>> is a normally closed port, and
>>> B executed a port scan on that port.
>>>
>>> iptables -A INPUT -p tcp --dport 222 -match <connections in which source
>>> IP can be spoofed> -g DROP
>>> iptables -A INPUT -p tcp --dport 222 <execute site wide preventive
>>> actions against the IP address: iptables -I INPUT -src THISSRC -j DROP>
>>>
>>> Thanks Denes
>>>
>>
>>
>> I do not think it is possible to match on spoofed IP addresses.
>> But I think you could construct something that matches those hosts which
>> sent a SYN and continue with INVALID state traffic.
>>
>> 1: put tcp syn into a recent set.
>> 2: match for hosts in the set with state INVALID.
>>
>> Questioning it all:
>> A simple -m state --state INVALID -j DROP should silently discard all
>> those.
> Dear All
>
> This is a good idea,
>
> "But I think you could construct something that matches those hosts which
> sent a SYN and continue with INVALID state traffic.
>
> 1: put tcp syn into a recent set.
> 2: match for hosts in the set with state INVALID."
>
> , but the real question is what does INVALID mean, because if it will
> also match case where Y!=A+1, than it will allow an attacker to
> perform a DoS by easily spoofing valid IP addresses, which will get
> dropped (sitewide)
>
Without being an expert here, imho:
If there's no existing valid connection from the spoofed ip, what does
it matter?
If there is, only those packets will be recognized as INVALID and
therefor dropped. It does not mean an existing connection marked as
ESTABLISHED would be closed from the conntrack side.
Anyone please correct me if I'm wrong.
But yes good question, 'what does state INVALID exactly mean?'
There's not much documentation about it afaik.
But i think you could say:
Everything that is not state NEW, ESTABLISHED, RELATED (those last 2
must be valid traffic (i.e. tcp with no anomalies)), or UNTRACKED (set
by the NOTRACK target in the raw table).
Best regards
Mart
prev parent reply other threads:[~2010-02-28 11:36 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-02-28 9:57 Howto match the 3rd packet in the 3way handshake Nemeth Denes
2010-02-28 10:27 ` Mart Frauenlob
2010-02-28 11:03 ` Nemeth Denes
2010-02-28 11:36 ` Mart Frauenlob [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=4B8A5530.7050007@chello.at \
--to=mart.frauenlob@chello.at \
--cc=netfilter@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox