Re: Small TCP packets == very large overhead == DoS?

["Crist Clark" <crist.clark () globalstar com>]

Pavel Machek wrote:
> Hi!
>
> > If any of you have tested what happens to the ability of a box to
> > perform well when it has a small MTU you will know that setting the
> > MTY to (say) 56 on a diskless thing is a VERY VERY bad idea when NFS
> > read/write packets are generally 8k in size.  Do not try it on a NFS
> > thing unless you plan to reboot it, ok ?  Last time I did this was
> > when I worked out you could fragment packets inside the TCP header
> > and that lesson was enough for me ;_)
>
> AFAI can remember, there's minimum MTU defined for IP (something like 576)...
> ...and then there are networks like hamradio that use lower MTU. They are
> really non-compliant, but Linux wants them to work.

No, there is no minimum MTU. However, all IP-compliant hosts must be
able to handle 576-byte datagrams. That is, the network to which a 
host is attached may have a smaller MTU, but the host must be able to
handle reassembled datagrams that are 576-bytes long. The 576-byte 
requirement is what the host's network stack must handle (that set of
algorithms in software or hardware) and does not have anything to do
with requirements on the network medium.
-- 
Crist J. Clark                                Network Security Engineer
crist.clark () globalstar com                    Globalstar, L.P.
(408) 933-4387                                FAX: (408) 933-4926

The information contained in this e-mail message is confidential,
intended only for the use of the individual or entity named above.  If
the reader of this e-mail is not the intended recipient, or the employee
or agent responsible to deliver it to the intended recipient, you are
hereby notified that any review, dissemination, distribution or copying
of this communication is strictly prohibited.  If you have received this
e-mail in error, please contact postmaster () globalstar com