routing with packet tracer

Hijpo

Hi I've been looking at packet tracer recently.

Basicly I have a PC connected to a switch, the switch connected to a router. The router is connected to another through serial cable which is connected to a switch ending with a other PC.

When pinging from PC to PC it fails unless I add the networks to the RIP route table.

However if I scrap that and start a completely new topology and use 3 routers connected with serial cable with the same setup in terms of switch and PC, I can ping each PC without doing anything with RIP.

I was wondering is it a protocol thing, a router model thing (I'm sure they were generic routers I used) or is it a topology thing?

Why the need to add networks address to the router when using 2 routers but routing is taken care of without doing anything when using 3 routers?

Cuddlesworth

Some routers have switching ports, like the HWIC-4ESW on the 2911. If you were connected by those ports, then you would have a switched network end to end.

Hijpo

The two routers i used in the first instance were two 2811 with an NM-4A/S module in each.

The three routers i used in the second instance was generic with two PT-ROUTER-NM-1CFE modules, two PT-ROUTER-NM-1FFE modules and two PT-ROUTER-NM-1S modules.

I didnt know modules would make that much of a difference, i figured a serial port is basically a serial port and once you connect both routers with the correct cabling they would just function properly straight off.

Cuddlesworth

I just threw together a similar config to what you laid out without encountering the same issue.

Packet tracer is a simulator though. It doesn't actually pass any traffic and as a result its prone to encountering bugs like this.

Its decent for the CCNA but don't expect great things out of it.

Hijpo

Cuddlesworth wrote: »

I just threw together a similar config to what you laid out without encountering the same issue.

Packet tracer is a simulator though. It doesn't actually pass any traffic and as a result its prone to encountering bugs like this.

Its decent for the CCNA but don't expect great things out of it.

OK thanks for going to the trouble, much appreciated!
so what's the correct way that should operate? Should the two router config route packets straight off without setting RIP or should the three router config need RIP to be setup?

Cuddlesworth

Routers route between subnets through the use of either static routes or using dynamic protocols like RIP.

Even if all devices had IP's in the same subnet wired the way you described, Computer A would still be unable to contact Computer B because they were in different LAN segments.

Hijpo

Cuddlesworth wrote: »

Routers route between subnets through the use of either static routes or using dynamic protocols like RIP.

Even if all devices had IP's in the same subnet wired the way you described, Computer A would still be unable to contact Computer B because they were in different LAN segments.

I understand the why's and how's of the router but in terms of the innitial connection of the routers is it necessary/standard practice to have to manually configure RIP on startup or should the 3 routers converge automatically once the serial modules assigned an IP address?

Cuddlesworth

Hijpo wrote: »

I understand the why's and how's of the router but in terms of the innitial connection of the routers is it necessary/standard practice to have to manually configure RIP on startup or should the 3 routers converge automatically once the serial modules assigned an IP address?

PC1-->Switch-->Router-->Router-->Switch-->PC2

PC1 sends an ARP request based on the IP of PC2 which is in its subnet. It hits the switch, the switch floods out the request on all ports after checking its MAC address table. The only connected device is Router1 and while it has a IP, its not the requested IP, so it doesn't respond to the request. Router2 and onwards are not part of that broadcast domain, so they never receive the request. The switch has no MAC, so it drops the packets.

RIP would only make a difference if the connect between Routers and PC1 and PC2 are in unique subnets.

PC1 would then forward its packets to its gateway Router1, which would then check its routing table then forward the packets to Router2, who would then send them onto PC2. Even then, you still need to configure the subnet into rip using the network command.