[Snort-devel] New opportunity for IDS evasion in patches to tcp protocol vulnerability
Paolo.Milani at ...866...
Wed Apr 28 04:43:08 EDT 2004
> No, BGP does not have fixed port numbers for both peers. Like all
> protocols, one end picks a "random" port.
ok, i got this wrong. BGP has the most critical vulnerability to this type of attack for other reasons.
> BGP doesn't really need to have RST at all. A simple ACL restricting
> TCP RSTs to port 179 would suffice.
The same attack can be done with SYN packets instead (according to cisco advisory at least), so restricting/rate limiting reset packets is not a solution.
> The TCP spec will not, I hope, be changed.
> It isn't broken.
> Checking the ACK sequence number in the RST will increase effort for
> the attack from 2^18 to 2^36 or so. IPsec will get rid of it.
Whether or not the spec is changed, and whichever method is used in future tcp implementations to bypass this problem, it will make tcp implementations more restrictive in which tcp reset packets they accept. Which means that snort's stream state tracking/reassembly will have to take this into account, when deciding what to do with a reset packet, otherwise it may find itself out of sync from the end system, and therefore vulnerable to evasion.
In fact, I hope that the spec is changed, rather than have each tcp stack implementation solve the problem with it's own ad-hoc fix. The issue will have to be fixed at the tcp level, since it is a vulnerability in the protocol.
Gruppo Telecom Italia - Direzione e coordinamento di Telecom Italia S.p.A.
This message and its attachments are addressed solely to the persons
above and may contain confidential information. If you have received
the message in error, be informed that any use of the content hereof
is prohibited. Please return it immediately to the sender and delete
the message. Should you have any questions, please contact us by
replying to MailAdmin at ...2137... Thank you
More information about the Snort-devel