Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 22 Jul 2004 15:32:25 -0700
From:      othermark <atkin901@yahoo.com>
To:        freebsd-current@freebsd.org
Subject:   Re: fixing out of order first fragment processing?
Message-ID:  <cdpf9q$o0t$1@sea.gmane.org>
References:  <cdpbts$om0$1@sea.gmane.org> <200407222359.23147.max@love2party.net>

next in thread | previous in thread | raw e-mail | index | archive | help
Max Laier wrote:
> On Thursday 22 July 2004 23:34, othermark wrote:
> Activation of pf with a
> scrub in on <interface> fragment reassemble
> rule works as workaround. 

Thanks for this suggestion,

I have a 'scrub in all fragments reassemble' that I just added and loaded
to my /etc/pf.conf, which does not seem to solve the problem.  Do I have to
specify a scrub for each interface in this case (maybe a better question 
for the pf list)?

> In every case you have to decide if you want to
> invest the required memory to store fragments, which might make you
> easy/easier prey for DoS-attacks. Usually, for an average gateway the cost
> is worth the gain (= increased security).

Most of the current systems today are able to handle both types of 
sequences.   It really is a small processing hit, FreeBSD already does
some bufferring with proper safeguards/maximums for various 
traffic patterns.

I would suspect some NFS/udp interoperability problems with the way it
handles fragments right now.

--
othermark
atkin901 at nospam dot yahoo dot com
(!wired)?(coffee++):(wired);



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?cdpf9q$o0t$1>