Hi guys, I have the attached topology in use just to study what happens when you have fragments in an MPLS network.
I configured s0/0 int R4 with the following: R4#sh run int s0/0 Building configuration... Current configuration : 128 bytes ! interface Serial0/0 ip address 10.10.34.4 255.255.255.0 ip ospf 1 area 0 mpls ip mpls mtu 64 serial restart-delay 0 end and sent a ping wiht 200 bytes size from R5 to R1 to see what would happen: R5#ping ip 1.1.1.1 size 200 rep 1 source lo0 Type escape sequence to abort. Sending 1, 200-byte ICMP Echos to 1.1.1.1, timeout is 2 seconds: Packet sent with a source address of 5.5.5.5 ! Success rate is 100 percent (1/1), round-trip min/avg/max = 156/156/156 ms R5# *Mar 1 01:37:01.347: IP: tableid=0, s=1.1.1.1 (Serial0/0), d=5.5.5.5 (Loopback0), routed via RIB *Mar 1 01:37:01.351: IP: s=1.1.1.1 (Serial0/0), d=5.5.5.5, len 200, rcvd 4 *Mar 1 01:37:01.351: ICMP type=0, code=0 Looking at the output some questions came out: 1 - fragmentation out of R4's s0/0 interface is OK: R4# *Mar 1 01:37:00.831: MPLS: Se0/1: recvd: CoS=0, TTL=255, Label(s)=16 *Mar 1 01:37:00.831: MPLS: fragment: 64 bytes *Mar 1 01:37:00.831: MPLS: Se0/0: xmit: CoS=0, TTL=254, Label(s)=16 *Mar 1 01:37:00.831: MPLS: fragment: 64 bytes *Mar 1 01:37:00.831: MPLS: Se0/0: xmit: CoS=0, TTL=254, Label(s)=16 *Mar 1 01:37:00.831: MPLS: fragment: 64 bytes *Mar 1 01:37:00.831: MPLS: Se0/0: xmit: CoS=0, TTL=254, Label(s)=16 *Mar 1 01:37:00.831: MPLS: fragment: 64 bytes *Mar 1 01:37:00.831: MPLS: Se0/0: xmit: CoS=0, TTL=254, Label(s)=16 *Mar 1 01:37:00.831: MPLS: fragment: 44 bytes *Mar 1 01:37:00.831: MPLS: Se0/0: xmit: CoS=0, TTL=254, Label(s)=16 R4# *Mar 1 01:37:00.963: MPLS: Se0/0: recvd: CoS=0, TTL=253, Label(s)=19 *Mar 1 01:37:00.963: MPLS: Se0/1: xmit: (no label) *Mar 1 01:37:00.963: IP: s=1.1.1.1 (Serial0/0), d=5.5.5.5 (Serial0/1), g=5.5.5.5, len 200, forward *Mar 1 01:37:00.967: ICMP type=0, code=0 and I did not configure R3's s0/1 interface with an mpls mtu of 64 bytes. R3#sh run int s0/1 Building configuration... Current configuration : 115 bytes ! interface Serial0/1 ip address 10.10.34.3 255.255.255.0 ip ospf 1 area 0 mpls ip serial restart-delay 0 end R3# *Mar 1 01:38:45.031: MPLS: Se0/1: recvd: CoS=0, TTL=254, Label(s)=16 *Mar 1 01:38:45.031: MPLS: Se0/0: xmit: CoS=0, TTL=253, Label(s)=16 *Mar 1 01:38:45.035: MPLS: Se0/1: recvd: CoS=0, TTL=254, Label(s)=16 *Mar 1 01:38:45.035: MPLS: Se0/0: xmit: CoS=0, TTL=253, Label(s)=16 *Mar 1 01:38:45.035: MPLS: Se0/1: recvd: CoS=0, TTL=254, Label(s)=16 *Mar 1 01:38:45.035: MPLS: Se0/0: xmit: CoS=0, TTL=253, Label(s)=16 *Mar 1 01:38:45.035: MPLS: Se0/1: recvd: CoS=0, TTL=254, Label(s)=16 *Mar 1 01:38:45.039: MPLS: Se0/0: xmit: CoS=0, TTL=253, Label(s)=16 *Mar 1 01:38:45.039: MPLS: Se0/1: recvd: CoS=0, TTL=254, Label(s)=16 However it shoud not detect the MTU configured in R4's s0/0 and also fragment R1's answer (ping reply) ? 2 - If R4's default behavior is to accept the packet regardless of its configure mpls mtu ? Is there a way to signal the peer about the mpls mtu configure in one partner in mpls networks ? Thanks, Fábio
_______________________________________________ For more information regarding industry leading CCIE Lab training, please visit www.ipexpert.com Are you a CCNP or CCIE and looking for a job? Check out www.PlatinumPlacement.com
