Lab1 Network Issues Our first lab network uses RIP v1, has 24 routers arranged in a circular topologyRIP's maximum diameter is 15, so parts of the network may be unreachable, especially the RIP annoucement for defaultAdded an additional ethernet link on R10 into VLAN61 to resolve by reducing network "diameter" RIP v1– Routing Information Protocol RIP VersionsRIP v1 (original version, Doyle ch 5)RIP v2 (improved version, Doyle ch 7)Simple distance-vector protocol (aka Bellman-Ford algorithm)Use hop count metric to determine best path (does not take other factors into account such as link speed)Initially popular because "routed" implementation bundled with BSD UNIXOriginal IP version adapted for other network protocols such as Novell IPX RIP – Message Format Two Message TypesRequest (sent by devices after initialization to request a unicast copy of a neighbor's routing tableResponse (sent by RIP speaking devices by broadcasting every 30 seconds to IP 255.255.255.255)Send via UDP/520 up to 25 routes per packet that include IP network, metric, but no subnet mask RIP – Normal Operation For every active RIP interfaceAnnounce known RIP routes out RIP enabled interfaces every 30sProcess received RIP annoucements by placing routes in routing table if better than existing route and add 1 to the hop countDo not announce things learned on an interface out the same interface (split horizon)Only announce if reachable with hop count <- 15 RIP Timers (Cisco Imp.) Router rip timers basic update invalid holddown flushUpdate 30s (when to broadcast response)Invalid 180s (when haven't heard annoucements for 180s, do not use)Holddown 180s (when neighbor increases metric for a network, do not accept immediatelyFlush 240s (after invalid timer expires, mark as unreachable metric 16 until time to flush) Cisco Administrative Distance Routers can run multiple routing protocols simultaneouslyQ: What to do when you have more than one route for a network learned with different routing protocols?A: Believe route with smallest administative distanceFor example, RIP uses admin distance 120, static uses 1, so static routes are more "believable" RIP v2 improvements Will cover these later in Doyle ch 7Annoucements carry subnet masks therefore supporting classful routingSubnets of classful nets do not need to all be the same size, but can be variable length (VLSM)Supports IP multicast on 224.0.0.9Supports MD5 authentication Cisco RIP commands Show ip protocolsShow ip routeShow ip route RIPShow ip RIP databaseDebug ip RIPDebug ip RIP eventsDebug ip routing Sh ip protocols R10#sh ip protocolsRouting Protocol is "rip" Sending updates every 30 seconds, next due in 14 seconds Invalid after 180 seconds, hold down 180, flushed after 240 Outgoing update filter list for all interfaces is not set Incoming update filter list for all interfaces is not set Redistributing: rip Default version control: send version 1, receive any version Interface Send Recv Triggered RIP Key-chain Ethernet0/0 1 1 2 Ethernet0/0.11 1 1 2 Ethernet0/0.61 1 1 2 Automatic network summarization is in effect Maximum path: 4 Routing for Networks: 192.168.10.0 192.168.11.0 192.168.61.0 Routing Information Sources: Gateway Distance Last Update 192.168.61.70 120 00:00:18 192.168.10.1 120 00:00:13 192.168.61.61 120 00:00:04 192.168.10.11 120 00:00:16 Gateway Distance Last Update 192.168.11.20 120 00:00:16 Distance: (default is 120) Sh ip route R10#sh ip routeCodes: C - connected, S - static, I - IGRP, R - RIP, M - mobile, B - BGP D - EIGRP, EX - EIGRP external, O - OSPF, IA - OSPF inter area N1 - OSPF NSSA external type 1, N2 - OSPF NSSA external type 2 E1 - OSPF external type 1, E2 - OSPF external type 2, E - EGP i - IS-IS, L1 - IS-IS level-1, L2 - IS-IS level-2, ia - IS-IS inter area * - candidate default, U - per-user static route, o - ODR P - periodic downloaded static routeGateway of last resort is 192.168.10.1 to network 0.0.0.0R 192.168.91.0/24 [120/7] via 192.168.10.11, 00:00:25, Ethernet0/0R 192.168.121.0/24 [120/1] via 192.168.10.11, 00:00:25, Ethernet0/0R 192.168.31.0/24 [120/4] via 192.168.11.20, 00:00:23, Ethernet0/0.11C 192.168.61.0/24 is directly connected, Ethernet0/0.61R 192.168.90.0/24 [120/5] via 192.168.61.70, 00:00:02, Ethernet0/0.61R 192.168.120.0/24 [120/2] via 192.168.10.11, 00:00:25, Ethernet0/0R 192.168.30.0/24 [120/3] via 192.168.11.20, 00:00:23, Ethernet0/0.11R 192.168.60.0/24 [120/1] via 192.168.61.61, 00:00:14, Ethernet0/0.61R 192.168.110.0/24 [120/4] via 192.168.10.11, 00:00:25, Ethernet0/0R 192.168.111.0/24 [120/3] via 192.168.10.11, 00:00:25, Ethernet0/0C 192.168.10.0/24 is directly connected, Ethernet0/0R 192.168.40.0/24 [120/5] via 192.168.61.61, 00:00:14, Ethernet0/0.61 [120/5] via 192.168.11.20, 00:00:23, Ethernet0/0.11 Debug ip routing R10# term monitorR10# debug ip routing(plug in ethernet cable here)R10(config)#int e0/0.61R10(config-subif)#no shutR10(config-subif)#.Jan 20 17:00:12 EST: is_up: 1 state: 4 sub state: 1 line: 1.Jan 20 17:00:12 EST: RT: closer admin distance for 192.168.61.0, flushing 1 routes.Jan 20 17:00:12 EST: RT: add 192.168.61.0/24 via 0.0.0.0, connected metric [0/0].Jan 20 17:00:12 EST: RT: interface Ethernet0/0.61 added to routing table.Jan 20 17:00:12 EST: RT: add 192.168.40.0/24 via 192.168.61.61, rip metric [120/5] SNMP tools to troubleshoot RIP Assuming you are on a LINUX computer with "snmpwalk" and "snmpnetstat" tools and your router has a Read-Only community configured named "public"( snmp-server community public RO )Linux% snmpnetstat –r 192.168.30.30 publicLinux% snmpwalk –v 1 –c public 192.168.30.30 More tools to debug RIP Use a sniffer like program such as Ethereal to "sniff" the UDP/520 RIP packets on an ethernet segment with RIP speakersCan also use older tools like TCPDUMP but newer tools like Ethereal have better decoding Lab2 Notes We will start a new lab next week with a new topology, subnetting, and more complex RIP configurationWill leave topology unchanged for now at least until first assignment due this Friday 1/21/05 7pm ET