Hot answers tagged

21

Most of the references say "RIP is not scalable" hence can be used only in smaller networks. But none say "WHY?" What is that in RIP that actually preventing it from scaling to larger networks? And HOW OSPF overcomes the disadvantage of RIP? Summary RIPv1 floods routes frequently (every 30 seconds), which introduces large CPU loads as ...


18

they are assigned by default to 172.16.0.0 in both cases, and it does not work I modified your ascii art a litte to reduce scrolling... It sounds like you're saying that you can't reach reach N1 from N2... Broken RIPv1 topology ===================== N1 ---- (172.16.1.1/24) R1 (172.16.3.1/30) ----- (172.16.3.2/30) R2 ----- (172.16.2.1/24) N2 Classful ...


14

You have to remember that models like OSI are just that, models. They are theoretical. The real world doesn't fall neatly into these models. For the most part, routing is a layer-3 function, but, as you pointed out, BGP uses a layer-4 protocol to communicate with other BGP speakers in order to do what is normally considered a layer-3 function. Many network ...


10

This is not a case of auto summary. Auto summary only works across major network boundaries. So you would need to have 172.16.x.x and 172.17.x.x or such to have a network summarized. What you are seeing is that RIP only supports classful networks under the routing process. So even if you enter 172.16.3.0 it will convert it to 172.16.0.0. You can use ...


10

I have often deployed RIP on the edge of MPLS VPNs on PE-CE links to give customers dynamic routing into their own VRF to good effect (though admittedly less and less these days). In this case it is a perfectly valid use - small RIB size (especially with a default out, local-site prefixes in), convergence time often isn't an issue (often only a single link ...


9

It's a RIPv1 packet. You're looking at the full IP packet. RIP starts at 0x001c.


8

BGP is on top of TCP, so it would be Internet layer 4, OSI layer 7. Usually external BGP is done only between 2 directly connected peers, enforced by setting TTL flag on IP header, which is located at layer 3.


8

I did a lab to test the behaviour with the same topology as yours. R1 is announcing three networks: 1.1.1.1/32 2.2.2.2/32 3.3.3.3/32 R1 is sending full updates towards R2. This is the final update before shutting down Lo0 which has the IP 1.1.1.1. R1 then sends a triggered update: Dec 9 10:22:37.797: %LINK-5-CHANGED: Interface Loopback0, changed state ...


7

Just to add to what Mike has already explained, RIP recalculates its routes and announces all of them every 30 secs. In a network with thousands of routers and tens of thousands of routes, that's a LOT of routes being calculated -- the routers would be too busy to actually forward any traffic. As you probably already learned, RIP's maximum metric is 15 ...


7

RIP is classful only in the sense of the network command in the routing process. The network command does not accept a mask, and automatically assumes the classful mask of the IP you enter. RIP does not advertise network masks. In fact, it only advertises network IDs. Moreover, RIP only advertises network IDs of networks it knows about that match the ...


7

Given how simple RIP v1 is, this is pretty easy to do by eye from Figure 1 in the RFC 1058: 5 longs from 45c0 is the IP header 4 shorts from 0208 (the italic portion) is the UDP header The rest from 0201 (the bold portion) is the RIP body 01:00:00.000000 IP 128.238.62.2.route > 255.255.255.255.route: RIPv1, Response, length: 44 0x0000: 45c0 0048 0000 ...


5

RIP is nearly universally supported, even on small routers which don't support OSPF or IS-IS. Suppose you had a network with only two routers with a few LANs connected to each one. It would be easier to set up RIP than it would be to configure static routes, and any changes to the networks on one or the other router would automatically propagate. If your ...


5

In addition to what others have said regarding the services each layer provides, different layers typically also differ in terms of how far they are encapsulated within packets (and in what order they are performed at each sender/receiver along the way). For example, a common scenario is where application layer (HTTP layer 7) header and data has transport ...


4

I think you're getting confused on how the OSI model plays into the routing. It's just a model, nothing more, nothing less. RIP works at layer 3 and sends routing information across the network. I found RIP is at the application layer, however I don't understand what the need for RIP is to reach the application layer. RIP isn't at the application ...


4

Damon, I'm afraid there is something wrong with your specific implementation. I, personally, can't think of an instance where a router doesn't respond to an icmp request on any of it's interfaces. Others, please chime in if that's inaccurate. At any rate, I mocked it up in GNS3 to make sure there weren't any funky software bugs. Basic configurations were ...


4

224.0.0.9 is a protocol standard multicast address, meaning that it is reserved for all RIPv2 speaking routers. This reduces unnecessary overhead, and only speaks to RIPv2 routers, instead of a full broadcast. Here is a portion from the RIPv2 RFC. http://tools.ietf.org/html/rfc2453 4.5 Multicasting In order to reduce unnecessary load on those ...


4

You can find this on Page 19 of RFC 1058: RIP, the maximum UDP payload of a RIP update is 512 bytes: The maximum datagram size is 512 octets. This includes only the portions of the datagram described above. It does not count the IP or UDP headers. The commands that involve network information allow information to be split across several ...


4

As commented you are best using another IGP because of the limitation of RIPs infinite metric. However you can configure individual ports to add more than one or alternatively configure offset lists to increase the advertised routes based on network address. RIP normally increases the cost by adding one to the route’s metric before storing the route so to ...


4

The OSI model is not 100% conformed to. It is a guide line for understanding and organizing the various functions required to create a network, or an inter-network, or even the Internet. So not everything falls perfectly within a single OSI layer. However, understanding the primary purpose of each layer will help categorize different protocols into ...


4

RIPv2 with the no auto-summary option will advertise the specific routes, not the aggregated prefix, unless you you use the ip summary-address rip interface command. The network statements in RIP are not specifically telling RIP what prefix(es) to advertise, they tell RIP which interfaces should participate in RIP, and RIP will get the specific prefix(es) ...


4

Yes, plain RIP can be vulnerable to attacks. RIP was created long before the Internet was commercialized. Today, there are options for secure authentication of RIP updates. For example, RFC 4822, RIPv2 Cryptographic Authentication: Introduction Growth in the Internet has made us aware of the need for improved authentication of routing ...


4

Route table poisoning used to happen accidentally (and with some regularity) when RIP (v1) was commonly used as a mechanism for end hosts to find redundant gateways in the days before FHRP's (i.e HSRP) came into use. Misconfigured hosts (usually UNIX sysadmins forgetting to add the -q / quiet option to ripd) would advertise bad routing into a shared subnet. ...


4

You can show running-config to find the config: Router#show running-config | begin router rip router rip network 10.0.0.0 network 12.0.0.0


4

The command is show ip route rip This will give you all the routes learned by RIP, which excludes directly connected ones.


4

Can someone please explain how looping problem is transient in case of link state routing, but is persistent in distance vector routing.(maybe with a nice example) Thanks in advance! :) The short answer is that loops will persist until the metrics reach infinity (the maximum number of routers that a packet can traverse) in distance-vector routing but ...


4

Ensure that your network subnet is not overlapping. To ensure communication between two networks which are configured with different routing protocols as you said RIP and EIGRP Redistribute the route of EIGRP to RIP and resdistribute route of RIP to EIGRP to ensure communication between intranet network A and intranet network B Redistribution? Route ...


3

It's an open standard and not vendor proprietary. EIGRP is an example of a Cisco proprietary routing protocol, and OSPF is an example of a standards-based routing protocol. Using standards-based protocols makes transitions between hardware vendors easier, although proprietary protocols sometimes have benefits you may want to use.


3

Poison reverse is an extra add-on to the split horizon rule that overrides split horizon in cases exactly like you describe. The route with the infinite metric is sent from A to B and E, and from B and E to their connected neighbors, including A. The whole point is to get the entire network into a converged understanding that T and any networks behind it are ...


3

There isn't what I'd call a "hard reason" - ie something specific relating to hardware/software limitations. Ultimately it was a judgment call made by the designers of the protocol, who made this judgment call based on the convergence behavior of the protocol (slow), the "counting to infinity" problem, and ultimately they didn't believe RIP would be "...


3

This is correct, the reason why you are seeing this is because you are using RIPv1 which can only deal with classful addresses and does not support VLSM. The address range you are trying to configure is a Class A address which has a /8 (255.0.0.0) subnet mask and you can't use a /16 mask as it belongs to Class B addresses in a classful process. If you want ...


Only top voted, non community-wiki answers of a minimum length are eligible