No OSPF neighbors were formed between Routers A & C, A & D, or B & D. The only link that was actively carrying routed traffic was between Routers C & B. Unfortunately, this circuit was a 0K CIR Frame Relay link. This meant that all traffic between routers C & B was being marked as discard eligible (DE) by the Frame Relay switches in the Frame Relay cloud. (That is, all traffic between Router C& B was set at such a low priority for packet delivery and could be dropped at any time.) As is discussed later in this case study, the Cisco router commands to show this information are show ip ospf neighbors and show frame pvc. The output for show frame pvc follows. Note that the number of input packets and the number of in DE packets are the same.
ROUTER_C#Show frame pvc PVC Statistics for interface Serial0/0:0.2 (Frame Relay DTE) DLCI = 700, DLCI USAGE = LOCAL, PVC STATUS = ACTIVE, INTERFACE = Serial0/0:0.2 input pkts 31341659 output pkts 12061107 in bytes 757769644 out bytes 2564616415 dropped pkts 0 in FECN pkts 17 in BECN pkts 0 out FECN pkts 0 out BECN pkts 0 in DE pkts 31341659 out DE pkts 0 out bcast pkts 2690375 out bcast bytes 250333218 pvc create time 15w5d, last time pvc status changed 4w2d
A large number of output drops were noted on Router Cs only active OSPF WAN interface (the one with the 0K CIR Frame Relay PVC). Output drops signify that the router is processing a very large number of outbound frames. In this case, the circuit could not accommodate the vast amount of data that the router was trying to output into it, so, as a result, the router was dropping frames. This observation is important because it directly affects the number of retransmits that the router sends and contributes to slow end-user throughput. This happens in a router when the circuit and router are overloaded with a large amount of input or output traffic. As discussed later in this case study, the Cisco command to see this happening is show interface serial <interface number>. The next section will specifically cover the numbers of input and output drops in the output of this useful IOS command.
Large numbers of input drops were observed on Router Bs only active WAN interface (0K CIR Frame Relay PVC). Input drops signify that the router is processing a very large number of inbound frames. The router could not keep up with the data stream because there was so much inbound traffic. Consequently, the router was dropping many inbound frames resulting in large number of input drops being reported. This observation is important because it directly affects the number of retransmits that the router sends, contributing to slower end-user throughput. As discussed previously, the Cisco router command to do this is show interface serial <interface number>. The output of Router Bs WAN interface follows:
ROUTER_B#show int s0/0:0 Serial0/0:0 is up, line protocol is up Hardware is DSX1 Description: Frame Relay Circuit to Downtown MTU 1500 bytes, BW 1536 Kbit, DLY 20000 usec, rely 255/255, load 6/255 Encapsulation FRAME-RELAY IETF, loopback not set, keepalive set (10 sec) LMI enq sent 11212, LMI stat recvd 11212, LMI upd recvd 0, DTE LMI up LMI enq recvd 0, LMI stat sent 0, LMI upd sent 0 LMI DLCI 0 LMI type is ANSI Annex D frame relay DTE Broadcast queue 0/64, broadcasts sent/dropped 983815/57443, interface broadcasts 1035677 Last input 00:00:00, output 00:00:00, output hang never Last clearing of "show interface" counters 1d22h Input queue: 0/75/421238 (size/max/drops); Total output drops: 32333 Queueing strategy: weighted fair Output queue: 0/64/32333 (size/threshold/drops) Conversations 0/51 (active/max active) Reserved Conversations 0/0 (allocated/max allocated) 5 minute input rate 50000 bits/sec, 19 packets/sec 5 minute output rate 42000 bits/sec, 8 packets/sec 1493015 packets input, 320768751 bytes, 0 no buffer Received 0 broadcasts, 2 runts, 0 giants, 0 throttles 48 input errors, 35 CRC, 13 frame, 0 overrun, 0 ignored, 2 abort 2335606 packets output, 845399484 bytes, 0 underruns 0 output errors, 0 collisions, 1 interface resets
Gather More Facts
In an attempt to gather more facts to form an action plan, we needed to know if the problem was related to a hardware problem or a software problem. Through the use of Telnet, we determined that we could successfully connect to all routers, although the performance was slow. This ruled out the possibility of a complete hardware failure on any of the routers. We then began to shift our focus on the router configurations. Our initial examination of Router As configuration showed that PVCs to C and D were defined, configured, and active (line up and line protocol up). The configuration that we observed follows:
ROUTER_A# show running-config ! interface Serial1 description Frame Relay PVCs Downtown to Router C no ip address encapsulation frame-relay IETF bandwidth 56 no fair-queue frame-relay lmi-type ansi ! interface Serial1.1 point-to-point description 768K CIR PVC to router C ip address 188.8.131.52 255.255.255.252 frame-relay interface-dlci 700 ! interface Serial2 description Frame Relay PVCs Downtown to Router D no ip address encapsulation frame-relay IETF bandwidth 56 no fair-queue frame-relay lmi-type ansi ! interface Serial2.1 point-to-point description 768K PVC to Router D ip address 184.108.40.206 255.255.255.252 frame-relay interface-dlci 701 ! router ospf 204 network 220.127.116.11 0.0.0.255 area 0 network 18.104.22.168 0.0.0.0 area 22.214.171.124 network 126.96.36.199 0.0.0.0 area 188.8.131.52 area 184.108.40.206 authentication area 220.127.116.11 stub