Re: ECN & PMTU

[Arun Prasad <[e-mail]>]

Srivathsa wrote:

> >     Why cant we generalise this stuff.... ie.,  the need here is some way to
> > communicate between the intermediate router and the endpoint.... Not sure
> > which of the two ways is advantageous, but cant we use the same method for
> > both..... ie.,
> > As for the PMTU discovery, the intermediate router can send an ICMP message
> > to the endpoint carrying a message "CONGESTION", by receiving this the
> > endpoint will do all appropriate actions as it does when it receives the packet set
> > with ECN-ECHO flag (as in present ECN procedure...)
> >
> > or vice versa ( ie., adopt the method followed by ECN  for PMTU discovery aslo,
> > that might be slighly tough, considering the Backward compatibility....)
> >
>
> You cant use a ECN-like-mechanism for PMTU discovery.   As soon as an
> intermediate router discovers that the packet (with DontFragment bit set
> ) is too large for one of its outbound interfaces,   it will have to
> discard the packet.  I.e it cant relay the packet to the
> destination.      In ECN scenario,  packet can still be relayed further.
>

            Ya, I agree that......  But the  method by which PMTU discovery is done
can be changed to avoid packet drops (if dropping a packet is costly... I think
yes, because it triggers Retransmissions....). We might rethink about implementing PMTU
discovery like the ECN like mechanism, with this new proposal..... The idea is

there should be  someother bit like the DF bit, say IF bit (Inform on Fragmentation) in the
IP header..... This IF bit will be set by the Transport layers if it is planning to do the PMTU
discovery.....  The router upon receiving a packet of size greater than the MTU size of the
Egress interface and the IF bit is set in the IP header... then IP layer there will go ahead
 fragmenting the packet and sending it to the destination..., but also send an ICMP
"Message too Long" message to the Sender.....  The router also sets the DF bit, and
unsets the IF bit in the IP header of the fragment packets (which ensure no more
fragmentation is done on the packet....)  ..................The sender upon receiving the ICMP
packet will just update his PMTU... Thus there is no Packet drops... so the procedure of
PMTU discovery will not  trigger any Retransmissions.... (which is expensive, because
the transport layer shrinks the Congestion window if it senses that a packe is dropped...)


Not sure whether this is already discussed or proposed etc...


Thanks
-arun

>
> -Srivathsa

--
****************************************************************
V.Arun Prasad
HCL Technologies Ltd.
51, Jawaharlal Nehru Road,
Ekkattuthangal,
Guindy Industrial Estate,
Chennai - 600097.

Contact # : 9144 - 2334174
            9144 - 2334181
            extn : 233
****************************************************************