Can you give me some context? Where and how do you see these terms used? Benji Paul Stewart. You asked about the difference between "classful boundary" and "classless boundary". Where does he use the term "classless boundary"? Did he ever make a comparison between the two that didn't make it onto the slide? The difference between them is that one is a thing and the other is not a thing.
I'll try to explain the thing that is a thing. The image on the left depicts a classful boundary. If the router has auto-summary enabled and if the router has neighbors on either side or on both sides it will create a summary or summaries. The image on the right does NOT depict a classful boundary. Even if the router has auto-summary enabled and even if the router has neighbors on both sides it will NOT create summaries.
Do you understand? If you do NOT understand then please be specific about what it is that you do not understand. Wipe "classless boundary" from your memory banks. You're clinging to "classless boundary" like grim death and it's not doing you any favors. In the immortal words of Kenny Rogers, "you've got to know when to hold 'em, know when to fold 'em, know when to walk away, and know when to run".
Run, don't walk, from "classless boundary". It's a dead end. It's a red herring. It doesn't exist. You cannot remove "boundary" and replace it with "EDGE".
That's not allowed. You must say "boundary". It's a boundary. Don't complicate things. If "boundary" is good enough for Mr. Bogart it should be good enough for all of us lowly Mr. Bogart wannabes. If you have Enter configuration commands, one per line.
Router config interface l1. Router config-if ip address Router config-if interface l2. R1 isn't a network, it's a router, and YES it sits on a classful boundary because If R1 is configured to perform in auto-summary wizardry AND it has at least one neighbor AND all of those interfaces were in the routing domain then it would for sure generate summaries. Never speak of classless boundaries again unless you run into Brian McGahan one day and he says "hey, what do you think about that classless boundary business?
All that is required for a router to sit at the boundary of two or more classful networks is that the networks reduced to the major network are different.
Two networks, of the same class type both A, for example , can produce a boundary so long as they are from different class A networks.
Why don't you just build this in a lab and see, for yourself, which combinations lead to auto-summary and which combinations don't. You asked where I was several responses ago. I have been dealing with a full work schedule so I've not had time to help on CLN.
You have good guidance with what Benji is sharing here. What makes a boundary classful has nothing to do with a transition between classes. It has to do with the fact that they are different classful networks. A red car and a green car are different cars because they're different cars not because they are different colors. A class A, B, C is an adjective to describe a given classful network in much the same way. There are two things that matter here.
This is what is ultimately important. If you absolutely, positively cannot bear to be without the term "classless boundary" then I think it would be fair to say that a classless boundary is any point at which two networks meet on a router which has disabled the auto-summary feature which, in , should be all of them.
The concept of classless inter-domain routing has been around for over 20 years now. Modern routers running modern routing protocols are operating in a classless world. The only thing that makes classful a thing is auto-summary. So in the absence of auto-summary then I guess every router with two connected networks could technically be considered a classless boundary, right? Because every network is classless and every router with two or more connected networks sits on a boundary between those networks.
I would still only speak of it in hushed tones I think it would be fair to say that a classless boundary is any point at which two networks meet on a router which has disabled the auto-summary feature which, in , should be all of them. What if we talk about this not in terms of definition for a little while? Let's talk about function and design of summarization. Examine this topology, then I'm going to ask you to solve a problem.
Automatic Summarization is not currently enabled. Each router has 13 routes in its routing table. Your manager has told you that there is a massive network expansion coming and that design will become more important. He asks you to determine how you can minimize the number of routes in each of the routers routing table. How might the number of routes be minimized? I'm on mobile and can't take this deeper unt I return to a computer. I did want to say that the way you answered this question shows a grasp on the concepts and this is what really matter.
I'm going to challenge this in a couple more directions when I can. Great response, there is existing issues with the network that could be cleaned up to make it better.
That should be communicated to the manager exactly the way you did. The manual summarization strategy definitely works. So if manual summarization is not permitted, is there another way to solve this with this cleaner addressing scheme? In your previous response and the redesigned topology which I like , you said you would do the following. But now you say that autosummary is equal to the solution that you believe will have issues.
R1 will insert D C L R2 will insert I would look at autosummarization as the following So my question is do you think this will work? Your response here shows me that you have a solid understanding of the function and operation. I also say that the So based on this topology, R1 and R2 are the classful boundaries for There is no boundary shown for The boundary could literally be said to be the entire network.
Would you agree with the fact that there is a boundary at R1 and R2 for only the And would you also agree that the If not, let's talk about it. If so, I want to show you an example of how we extend this network with this addressing scheme and demonstrate a boundary for these other two networks So let me add another image. Now we have another boundary. The reason we all hate the term classlesss boundary is somewhat represented here. So is R6 representative of a boundary for That'd be a classful boundary, right?
What if changed it and said we are summarizing to Would that make it a classless boundary? What if I took it further and said Would that not make R6 a classful boundary for the But it is a classless boundary for the What if we changed the perspective to In that case, the 8.
There is literally no boundary formed between the However, the We all know about perspective when we talk classful. Classless requires that extra element of perspective to be explicitly stated for EACH network in a topology. That element is implied with classful. Again, what is really important is that you understand the routing operation of routers and you understand the design elements of summarization. Based on what you have posted in this thread, I think you do have that understanding.
So I really think this deep analysis of the terminology is leading us more into confusion than out of confusion. This is definitely an area that could be argued either way. So if the position is that we are looking at If we look at the subnets discontiguous entities I agree with your statement.
It is about perspective. Arguably correct. It truly is based on perspective. It is really if we are looking at the components of the network from the inside, or if we are looking at it holistically from the outside.
You seem to understand the function, that is what is important. I wouldn't worry about these terms especially when some are not used very strictly. In my part of the USA we have an interesting terminology that basically means not to worry about something like this--we would say don't get "wrapped around the axle" over that ;- Think about a wagon or a bike and when the weeds get wound around the axle.
Doesn't really help anything and just slows you down. You know the function here, so you should be confident in your position when you have conversations with people. Just explain the function when you get into these areas that you believe are unclear. Benji wrote. In the above example, The network to the right is class C. So R2 is a classful boundary. The network to the right is Class A.
So the point Benji was making was that the perspective of classful boundary is where a given classful network ends. Two or more different networks come together at a boundary. These two or more networks may be of the same class or different classes A, B, C. But the point is that they are different networks, just like R1 config-subif do show ip int inc Internet.
R1 config-subif do clear log. Can you be more specific? I don't understand what, about it, you don't understand. It's literally only two words. Are you struggling with the meaning of "boundary"?
Are you struggling with the meaning of "classless"? You need to be more specific. What do you not understand about those lines? You seem to be burdened by the need to come to a precise definition of the term "classless boundary" a term, btw, which is not mentioned in a RFC one single time that I can find - an amazing feat considering there are over 8, RFCs and probably more than a , pages of text!
There are only two words in the term; classless and boundary. Which of those two words is causing you fits? Let's tackle this from a different angle. Do you have an opinion about the definition of "classless boundary"? What do you think it means? Maybe if you attempt to define it to us we can figure out why you feel so confused about it. If you can cite three "credible" sources ie, your favorite video instructor, cisco.
Until then there's really nothing else that can be said that hasn't already been said. It's hard to follow the question. Off the top of my head, there are three ways for R1 to see an advertisement of These are:. Here is the result, for each of these options, in R1's RIB:. Known via "eigrp 1", distance 90, metric , type internal. Last update from Route metric is , traffic share count is 1. Total delay is microseconds, minimum bandwidth is Kbit. In the case of scenario 1 manual summarization on R2 , the RouterID field of the In the case of scenario 2 and 3 both using summary features on R6 , the RouterID field in the In all cases, regardless of how R1 learned of If it were non-contiguous dis-contiguous networks then you might have:.
In this case you would not be able to summarize the networks without including other networks in the process. Using a network statement like So from above example we can define Discontiguous Network as A classfull network in which packets sent between at least one pair of subnets must pass through subnets of a different classfull network. Labels Cisco Routing. Labels: Cisco Routing.
Popular posts from this blog How to check Gray-log current running version. August 28, If you planning to upgrade your graylog and want to see what your graylog current version is, here is how you can check it.
Suppose PC0 wants to communicate with PC3. PC0 creates a data packet for PC3. The router R0 reads the destination network address of the incoming packet. In response, if PC3 sends a data packet back to PC1, the packet will go through the same route in reverse. Throughout the entire communication, data packets exchanged between PC0 and PC3 do not take a route that is not related to the network To verify this, you can use the 'tracert' command.
The term contiguous is network specific. It does not include subnets of other networks. In our example, we have three networks: They all are contiguous as long as their hosts can access other hosts of the same network without going outside of the network.
The following image shows an example of discontiguous networks. Except for the location of two subnets, this is the same network topology that we used above. In this topology, subnets Link to download the discontiguous network topology.
0コメント