Re: Path MTU for draft-ietf-v6ops-mech-v2-00.txt

["Fred L. Templin" <ftemplin@IPRG.nokia.com>]

Erik Nordmark wrote:

> > > Finally the spec allows for manual configuration to override the 1280 number
> > > in the static/fixed MTU case, but only if the receiver can handle reassembly
> > > of the larger IPv4 packets. Thus it requires manual coordination between >
> > >      
> > the encapsulator and decapsulator to do this manual override.  >
> > Are there any concerns for denial of service based on receive buffer overrun
> > at the decapsulator in this case?
> >    
> Not any worse than a factor MTU/1460 than in the normal case.
> I.e. if the decapsulator has been configured to use e.g. 9kbyte buffers for a 
> particular tunnel it might use 9k/1460 times more memory for buffers
> for that tunnel.
I understand what you are saying about the buffer reqt's. But, when IPv4 
fragmentation
is occurring in the path between the encapsulator and decapsulator, 
there is no way
for the encapsulator to learn this and decrease its MTU estimate.

As currently worded, the specification could result in harmful levels of 
fragmentation in
the network when encapsulators implement only the base specification.

Fred
ftemplin@iprg.nokia.com