|
0:00:15
|
Okay, the next thing we
|
|
0:00:18
|
is how does the automatic
|
|
0:00:22
|
which is similar to RIP Version 2
|
|
0:00:25
|
Now, we saw in RIP version 2, even
|
|
0:00:30
|
automatic summarization
|
|
0:00:34
|
which can cause traffic blackholes,
|
|
0:00:36
|
because we're not able to advertise
|
|
0:00:41
|
between different major networks.
|
|
0:00:45
|
Now, EIGRP has the same
|
|
0:00:49
|
because it was originally meant to be
|
|
0:00:53
|
but it doesn't run into the same exact
|
|
0:01:00
|
Now, we have the same logic where
|
|
0:01:04
|
And VLSM in supported within
|
|
0:01:08
|
but automatic summarization happens
|
|
0:01:14
|
We'll see the difference here
|
|
0:01:17
|
is that EIGRP uses a null route, which
|
|
0:01:22
|
in order to prevent these
|
|
0:01:27
|
So, if we look at the current
|
|
0:01:31
|
where we have EIGRP essentially
|
|
0:01:35
|
but no other options applied on to the process,
|
|
0:01:39
|
it means that all of the routers
|
|
0:01:43
|
So let's take a look at this
|
|
0:01:47
|
On router 5, if we look
|
|
0:01:52
|
we see that we're learning a bunch of
|
|
0:01:58
|
And we have two summaries
|
|
0:02:02
|
that are going to null zero.
|
|
0:02:06
|
One of them is for the major
|
|
0:02:11
|
which is the 155.10.0.0/16.
|
|
0:02:16
|
One of them is for the loopback address
|
|
0:02:25
|
These two null routes mean that router 5 is
|
|
0:02:32
|
and advertising it to the
|
|
0:02:36
|
Now, what we do not see is
|
|
0:02:43
|
the other classful summaries
|
|
0:02:47
|
Now, we can assume that everyone in the
|
|
0:02:51
|
because they all have the
|
|
0:02:55
|
and the loopback network space.
|
|
0:02:57
|
So, if we look at the Show IP
|
|
0:03:01
|
it's essentially doing the same thing.
|
|
0:03:06
|
for the summary of the transit network.
|
|
0:03:09
|
And a null route for the summary
|
|
0:03:14
|
But we don't see that router 4
|
|
0:03:18
|
each other's advertisements
|
|
0:03:23
|
So, EIGRP knows that if it is
|
|
0:03:27
|
it does not make sense for it to install
|
|
0:03:32
|
as it is learned from other neighbors.
|
|
0:03:37
|
Now, it does mean in this case that we're
|
|
0:03:42
|
these other destinations. Like, router 4 is not
|
|
0:03:50
|
but at least it means that the
|
|
0:03:54
|
So, it is dropped locally on this device,
|
|
0:03:59
|
as opposed to using the
|
|
0:04:03
|
where before, it meant that sometimes,
|
|
0:04:07
|
Okay, it's really dependent on what
|
|
0:04:13
|
how we were supposed to
|
|
0:04:18
|
Now, we'll also see that within
|
|
0:04:23
|
it is okay to have variably
|
|
0:04:28
|
So, if we were to go to router 5,
|
|
0:04:33
|
interface Fast Etherent 0/1. Let's say
|
|
0:04:41
|
from...
|
|
0:04:45
|
155.10.5.5/24, let's say we change it to /30.
|
|
0:04:57
|
252/30.
|
|
0:05:01
|
So, it's still part of the same major
|
|
0:05:06
|
which means that this should not
|
|
0:05:10
|
when it is sent to the other neighbors.
|
|
0:05:13
|
So, from router 4's perspective, when
|
|
0:05:17
|
we are able to learn that /30 from router 5.
|
|
0:05:26
|
So, it's only when we have the issue of
|
|
0:05:32
|
where the 150.10.0.0/16 summary is trying
|
|
0:05:41
|
Then, EIGRP is going to drop those updates if
|
|
0:05:49
|
Now, in reality, there's no reason
|
|
0:05:53
|
unless you have some really old legacy
|
|
0:05:58
|
Because in the redistribution
|
|
0:06:02
|
IGRP is only gonna support the classful networks.
|
|
0:06:06
|
And I believe that in these versions,
|
|
0:06:12
|
So, it doesn't even support the protocol anymore.
|
|
0:06:15
|
So really, there's no reason design wise
|
|
0:06:20
|
We'll see that we can configure summarizations
|
|
0:06:27
|
but we wouldn't want the router to just
|
|
0:06:32
|
So, typically, we would just
|
|
0:06:36
|
And say, No Auto-Summary.
|
|
0:06:48
|
Now, if we look at the result of this
|
|
0:06:55
|
we can see now, we can learn
|
|
0:06:59
|
of both the transit networks, which
|
|
0:07:04
|
but now, we have the subnets of
|
