Re: [pim] IPv6 Register Fragmentation

[Stig Venaas <Stig.Venaas at uninett.no>]

On Wed, May 19, 2004 at 11:14:35PM +0300, Pekka Savola wrote:
> On Wed, 19 May 2004, Stig Venaas wrote:
> > I'm aware that one is supposed to always do PMTU discovery in IPv6
> > and do fragmentation at the source, but do we really need to follow
> > this for register tunnels?
> 
> Bear in mind that this requires also defragmentation at decapsulation.
> 
> There's at least one hardware implementation that AFAIR doesn't do 
> defrag.

ok, but the draft also says:

   In IPv6, the DR MUST perform Path-MTU discovery, and an ICMP
   Fragmentation Required message MUST be sent by the encapsulating
   DR if it receives a packet that will not fit in the effective MTU
   of the tunnel.  If the MTU between the DR and the RP results in the
   effective tunnel MTU being smaller than 1280 (the IPv6 minimum MTU),
   the DR MUST send Fragmentation Required messages with an MTU value
   of 1280 and MUST fragment its PIM register messages as required,
   using an IPv6 fragmentation header between the outer IPv6 header
   and the PIM Register header.

So there are valid cases where the register message will be fragmented
anyway (even with the current MUST).

One more thing. By only reading the text above, it might seem that
one should send ICMP PTB (FragReq) every time a packet too large
for the <1280 register tunnel MTU is received. I don't think it
should be, at least not if the packet was 1280. In that case the
source won't fragment the packet anyway and possibly keep sending
1280 byte packets. Also IPv6 hosts not doing PMTU discovery would
send packets up to 1280 anyway. So I believe one should only send
ICMP PTP if the packet was >1280.

Stig

_______________________________________________
pim mailing list
pim at ietf.org
https://www1.ietf.org/mailman/listinfo/pim