Very simple yet brilliant exaple of LIVE lab. I am very happy to find this post and recommand to anyone who wants to be CCNA, cheers Ajay NZ
Awesome job in explaining clearly.
Great job, finally i understand this, after reading it in several books and was not understanding it till now. Thank you.
ip summary-address rip 172.16.0.0 255.255.254.0
What’s the purpose of configuring a route summary as shown above?
It seems like without the summary, routing works just as well.
Let’s imagine we have two routers, R1 and R2. R1 has 8 interfaces. We advertise 8 subnets from R1 to R2. In this case, R2 will have 8 entries in its routing table. When one of the interfaces on R1 goes down, R2 will lose the subnet in its routing table. Depending on the routing protocol, it will act upon this, trying to find another path.
If we would advertise a summary route that matches all 8 subnets on R1 instead, then R2 would only have 1 entry in its routing table. This will save some memory. The second advantage is that when one of the interfaces on R1 goes down, it doesn’t affect the summary that R2 has in its routing table. This makes things a bit more stable.
On small networks this doesn’t matter all that much but when you think about the Internet, summarization becomes very important.
Hope this helps!
RIP update is 30 seconds. Last update was done 00:00:16 ago. What do these three parameters stand for - 00:00:16 ? Hr:Min:Sec or Min:Sec:MSec or MSec:MicroSec:NanoSec ?
The notation is as following: Hr:Min:Sec
Hope this helps.
one question regarding the network command.
Rene writes: " I also could have used just one network command:
How can the network command be understood? Is it classful?
If so, why does 192.168.12.0 and 192.168.23.0 fall in the 192.168.0.0 network? As this network would only consist of 192.168.0.0 - 192.168.0.255!?
If not, wouldn’t 192.168.23.0 fall also in 192.168.12.0 network? 192.168.12.0 - 192.168.255.255!? And we could also have used 192.168.12.0 instead?
What do I miss here?
Thanks for your help!
Yes you are correct that if the command
network 192.168.0.0 was configured, it would only advertise the 192.168.0.0/24 subnet and not all the 192.168.0.0/16 subnet. Although RIP v2 is classless, the configuration commands are classfull. Now if the above were
network 172.16.0.0 subnet, then the command would indeed work just like Rene has described, because the command would refer to a class B or /16 subnet which would include any interfaces with addresses between 172.16.0.0 and 172.16.255.255. However, RIPv2 would advertise the networks using the subnet mask that has been configured on the interfaces themselves. This is why RIPv2 is a classless protocol. I will let @ReneMolenaar know about this.
Thanks and I hope this has been helpful!
you said (Rene) " Just type version 2 to switch to this version. By default RIP version 2 will behave classful and won’t send a subnet mask. I need to type no auto-summary to make it behave classless and send a subnet mask with its RIP updates."
This is actually has nothing to do with version, RIPv2 is classless, this is the auto summarization feature which auto summarize the networks that pass a different transit major network (in this case 192.168.x.0/24)
if you create some loopbacks under the same major networks like 192.168.10.0/24, 192.168.20.0/24, you will see them un-summarized and as /24 in the RIB, so you have to disable auto-summary not make the protocol classless by adding auto-summary, this is a separate feature and it’s btw in the blueprint and you had to spent time on it as you provide CCIE level lessons. and by the way writing network 172.16.1.0 under RIP is suggestive in a way that we need to write the whole subnet which in RIP is not necessary
you need to type only the major networks that your RIP interfaces fall under, actually if you had shown the running config we would’ve seen that the commands are:
also I’m sad to see that the same RIP lesson for CCNA is the same for CCIE, where it’s clearly you skipped a lot of explanation that you had to add
also this behavior of auto-summary feature is in EIGRP, do we need to add no auto-summary to make it classless?!!
Thanks for your comments. Just to clarify, RIPv1 does NOT support classless routing. RIPv2 does support it and thus includes the subnet mask information of a destination. However, RIPv2 by default behaves in a classless manner unless you enter the
no auto-summary command. This is what Rene mentions in the lesson. All of the information stated by Rene is correct.
I hope this has been helpful!
As for the auto-summary feature of EIGRP, it works a little bit differently. EIGRP supports CIDR and variable length subnet masking but routes are not summarized at the classful network boundary unless auto summary is enabled.
I hope this has been helpful.
“RIPv1 does NOT support classless routing” I’m sorry, when did I say it does?!!
" RIPv2 by default behaves in a classless manner unless you enter the
no auto-summary command",
Are you saying that no auto-summary makes it classful?!!
“As for the auto-summary feature of EIGRP, it works a little bit differently. EIGRP supports CIDR and variable length subnet masking but routes are not summarized at the classful network boundary unless auto summary is enabled.”
Ofcourse it has to be enabled, that’s the whole point of auto summary, CIDR and VLSM has nothing to do with this behavior
I’m stating that auto-summary is a feature regardless of the distance-victor routing protocol or its version, and this feature is controlling whether the subnets are summarized or not
The whole point is auto-summary controls the behavior not the version
I think you miss-interpret my comment
The purpose of this forum is to be a place of constructive discussions where we can all help each other in our quest to become better networking professionals. There is no need to be adversarial. If I came across as adversarial, or if I upset you in any way I apologize.
You didn’t. It was just a part of my explanation.
First of all, my statement was incorrect. It should read: “RIPv2 by default behaves in a CLASSFUL manner unless you enter the
no auto-summary command” Let me elaborate. I have the following topology:
Router 2 has a loopback of 10.10.20.1/24 and Router 0 has a loopback of 10.10.10.1/24. RIP version 2 is enabled in all routers and all IPs (including the interconnecting networks and the loopbacks) are advertised via RIP.
Now, remember, the default setting is auto-summary for RIP version 2. Let’s look at the routing table of Router 1:
Router#show ip route | begin Gateway Gateway of last resort is not set R 10.0.0.0/8 [120/1] via 192.168.2.1, 00:00:10, GigabitEthernet0/1 [120/1] via 192.168.3.1, 00:00:25, GigabitEthernet0/0 R 192.168.1.0/24 [120/1] via 192.168.3.1, 00:00:25, GigabitEthernet0/0 [120/1] via 192.168.2.1, 00:00:10, GigabitEthernet0/1 192.168.2.0/24 is variably subnetted, 2 subnets, 2 masks C 192.168.2.0/24 is directly connected, GigabitEthernet0/1 L 192.168.2.2/32 is directly connected, GigabitEthernet0/1 192.168.3.0/24 is variably subnetted, 2 subnets, 2 masks C 192.168.3.0/24 is directly connected, GigabitEthernet0/0 L 192.168.3.2/32 is directly connected, GigabitEthernet0/0
Notice that even though the loopbacks are configured with a subnet mask of /24, RIP only sends the classful information from Router 2 and Router 0 to Router 1. There are two routes to 10.0.0.0/8, one for each of the loopback interfaces of each router. There are no routes to 10.10.10.0/24 or 10.10.20.0/24 because RIP is behaving in a classful manner in the information it sends.
Now that is not to say that RIP version 2 is classful, but that if the auto-summary setting is set, it BEHAVES in the same manner as RIPv1. Now, if we add the
no auto-summary command to both Router 0 and Router 2, we get the following routing table on Router 1 (after waiting the appropriate amount of time to flush out old routes):
Router#show ip route | begin Gateway Gateway of last resort is not set 10.0.0.0/24 is subnetted, 2 subnets R 10.10.10.0/24 [120/1] via 192.168.2.1, 00:00:04, GigabitEthernet0/1 R 10.10.20.0/24 [120/1] via 192.168.3.1, 00:00:16, GigabitEthernet0/0 R 192.168.1.0/24 [120/1] via 192.168.3.1, 00:00:16, GigabitEthernet0/0 [120/1] via 192.168.2.1, 00:00:04, GigabitEthernet0/1 192.168.2.0/24 is variably subnetted, 2 subnets, 2 masks C 192.168.2.0/24 is directly connected, GigabitEthernet0/1 L 192.168.2.2/32 is directly connected, GigabitEthernet0/1 192.168.3.0/24 is variably subnetted, 2 subnets, 2 masks C 192.168.3.0/24 is directly connected, GigabitEthernet0/0 L 192.168.3.2/32 is directly connected, GigabitEthernet0/0 Router#
Now this is what we would expect to see from a routing protocol that is behaving in a classless manner. We see both the 10.10.10.0/24 and 10.10.20.0/24 networks destination networks with the appropriate subnet masks.
To summarize, RIP version 2 is indeed a classless routing protocol as you mention. And yes, the auto-summary is indeed a feature that controls behaviour not the version. However, in order to get the routing protocol to send the subnet masks and to behave performing Classless Inter Domain Routing (CIDR), the no auto-summary command must be implemented. If I misinterpreted your original comments, I apologize.
I hope this has been helpful!
Sorry, I did not mean that
I still have a problem with these statements:
Rene: " By default RIP version 2 will behave classful and won’t send a subnet mask. I need to type no auto-summary to make it behave classless and send a subnet mask with its RIP updates."
in your example topology :
R2 has Lo0 10.10.20.0/24
R0 has Lo0 10.10.10.0/24
now let’s make the transit network between them 10.100.100.0/24 and enable auto-summary.
Now if you check the RIB on both routers you will see that RIP advertised 10.10.20.0/24 and 10.10.10.0/24
because the summarization traversed the same major network 10.0.0.0/8,
so RIP did send the subnet mask, RIP did behave in a classless manner even though auto-summary is enabled, and when it did not it was because of the auto summarization rules,
thus the right statement should’ve been:
to prevent RIP from summarizing the major networks we need to disable the auto-summary feature.
because RIPv2 by default can behave classless and can send the subnet mask even if auto-summary is enabled.
Thank you and sorry again
I understand where you’re coming from. I think it all has to do with the words being used.
To clarify, RIPv2 is a classless routing protocol. It performs CIDR, no question. It sends the subnet mask etc and supports VLSM. Now you mentioned in a previous post that:
I am in total agreement. What we are interested in for this specific instance is not so much the inherent workings of the routing protocol, but the end result. With the auto-summary enabled, the end result of the entry in the routing table in Router 1 of our example will be 10.0.0.0/8 which is indeed what we would get with RIPv1. This is something we do not want. To remedy this we use the
no auto-summary command which changes the behaviour of the protocol so that we get the CIDR result we want. This I believe was the central idea behind Rene’s comment. It does not change the classless nature of the protocol, it just changes the behaviour so that we can take advantage and use this classless nature.
I hope this has been helpful!
I am new to routing and studying your RIP topology. On your topology when looking at R1 routing table, it says 172.16.0.0/24 is subnetted, 1 subnets, but when I built the topology it says 172.16.0.0/24 is variably subnetted, 2 subnets, 2 masks, why the difference? I am using IOS ver 15.6(2)T
The routing table will show what routes are available to the router to route packets. In the lesson, the statement “172.16.0.0/24 is subnetted, 1 subnets” indicates that there is only one /24 destination in the routing table that falls within the classful range 172.16.0.0 to 172.31.255.255.
In your routing table, if it states “172.16.0.0/24 is variably subnetted, 2 subnets, 2 masks” that means that you have some additional routes configured in the routing table. These may be from directly connected networks, that is, interfaces that you have configured on the device itself, or routes that have been learned via RIP. If you look at the subnets that are indicated, you can get an idea of where these have been learned from.
If you like, share with us your routing table in full so that we can see where those network destinations were learned from.
I hope this has been helpful!