Can anyone analysis below parameter

Unanswered Question
May 12th, 2009
User Badges:

Processing Labeled IPv4 Datagrams which are Too Big


If a labeled IPv4 datagram is "too big", and the DF bit is not set in

its IP header, then the LSR MAY silently discard the datagram.


Note that discarding such datagrams is a sensible procedure only if

the "Maximum Initially Labeled IP Datagram Size" is set to a non-zero

value in every LSR in the network which is capable of adding a label

stack to an unlabeled IP datagram

  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 3 (1 ratings)
Loading.
Harold Ritter Tue, 05/12/2009 - 13:12
User Badges:
  • Cisco Employee,

Sanjoy,


If the DF bit is not set then the packet should be fragmented by the MPLS core and the fragments will be forwarded using the egress label (or label stack).


If the DF bit is set then an ICMP type 3 (destination unreacheable) code 4 (Fragmentation Needed and Don't Fragment was Set) message is returned to the sender by using the egress label (or label stack), which means the ICMP packet is actually forwarded to the egress PE and then back through the MPLS core to the sender. This procedure is taken because the MPLS core router doesn't have routing information for the source address used by the sender as it is in VRF.


Regards

sanjoy2006 Wed, 05/13/2009 - 06:15
User Badges:


I just want the explanation of below scentence of rosen draft



Maximum Initially Labeled IP Datagram Size" is set to a non-zero

value in every LSR

Harold Ritter Wed, 05/13/2009 - 07:00
User Badges:
  • Cisco Employee,

Sanjoy,


You are actually referring to RFC3032, Section 3.4. In IOS, the "maximum Initially Labeled IP Datagram Size" is implemented via the egress interface "mpls mtu" statement.. The way I understand this sentence, is that this value should be set consistently across all LSRs.


Regards

shivlu jain Tue, 05/12/2009 - 20:55
User Badges:
  • Silver, 250 points or more

Sanjoy


The LSR will discard the packet Iff the packet size is increased by the outgoing interface.

Why the LSR behaves like this because it doesnot have the routing information for vrf.


regards

shivlu jain

Harold Ritter Wed, 05/13/2009 - 06:28
User Badges:
  • Cisco Employee,

Shivlu,


The fact that the MPLS P router doesn't have routing information for the VRF IPv4 addresses stop from either fragmenting (if DF bit is not set) or sending the ICMP message back to the source to let it know the packet is too big and could not big fragmented because DF is set. In either case, the forwarding will be performed using the egress label (or label stack) for the original packet.


Regards

Actions

This Discussion