|
0:00:11
|
In our next section here, we're gonna look at
|
|
0:00:19
|
the feasible successors to install multiple
|
|
0:00:25
|
that do not have the same metric values.
|
|
0:00:28
|
Now, EIGRP is the only IGP that can do this,
|
|
0:00:32
|
where when we look at RIP, or IS-IS, or OSPF,
|
|
0:00:35
|
they can do load balancing
|
|
0:00:39
|
but not for paths that have different metrics.
|
|
0:00:43
|
So, for OSPF, if we had path
|
|
0:00:46
|
and paths going over Fast Ethernet,
|
|
0:00:49
|
in order to load balance between the two of them,
|
|
0:00:51
|
we would have to make them
|
|
0:00:55
|
Then, it would be up to the switching
|
|
0:01:00
|
divide the traffic between the two interfaces.
|
|
0:01:03
|
Now, with EIGRP, as long as
|
|
0:01:08
|
so as long as they are feasible successors,
|
|
0:01:11
|
we take the feasible distance, which
|
|
0:01:15
|
for the active route.
|
|
0:01:18
|
Multiply it by the variance value.
|
|
0:01:21
|
Then, the result of that, if it is larger than
|
|
0:01:26
|
load balancing is going to occur.
|
|
0:01:30
|
Now, we'll see that the router
|
|
0:01:34
|
what the traffic shared balance should be
|
|
0:01:36
|
between the primary path and the backup path
|
|
0:01:39
|
based on the total end-to-end
|
|
0:01:43
|
So, what we're gonna do now in this case
|
|
0:01:46
|
is look again at the connection
|
|
0:01:49
|
that router 2 is using the reach router 6,
|
|
0:01:52
|
but now, I wanna use both the link that goes
|
|
0:02:03
|
So first, let's look at router 2 and
|
|
0:02:08
|
that it is using end-to-end to get to router 6.
|
|
0:02:14
|
So, from router 2, we see
|
|
0:02:18
|
The packets are going from 5 to 1 to 6.
|
|
0:02:25
|
So, from 5 to 1 to 6, this is our primary path.
|
|
0:02:30
|
If we look at the topology,
|
|
0:02:34
|
150.10.6.0/24,
|
|
0:02:39
|
it says, "There is an alternate
|
|
0:02:47
|
and router 3's advertised
|
|
0:02:52
|
Since 130,000 is less than
|
|
0:02:56
|
which is 16 million, it means that this could
|
|
0:03:03
|
Because this is a feasible successor.
|
|
0:03:07
|
So now, let's go to router 2,
|
|
0:03:10
|
and under the EIGRP process,
|
|
0:03:14
|
Now, it really doesn't matter what this value is
|
|
0:03:17
|
as long as it is large enough to take into
|
|
0:03:24
|
So, setting it to 10, 20, or 30 is not going to
|
|
0:03:31
|
It will only affect what feasible successors can be
|
|
0:03:40
|
So again, we're not gonna take the...
|
|
0:03:43
|
the variant's value, which I configured as 128.
|
|
0:03:46
|
Multiply it by the feasible distance.
|
|
0:03:50
|
If there is any feasible successor whose
|
|
0:03:57
|
they will be used for unequal
|
|
0:04:02
|
So, in this particular example,
|
|
0:04:06
|
And multiply it by 1.6 million.
|
|
0:04:11
|
If this is higher than 25 miilion,
|
|
0:04:15
|
that is the path to router 3,
|
|
0:04:18
|
then, the load balance should occur.
|
|
0:04:21
|
So now, if we look at the result
|
|
0:04:25
|
if we Show IP Route for 150.10.6.0,
|
|
0:04:31
|
we see that both of the paths are installed.
|
|
0:04:36
|
Now, based on the ratios of router
|
|
0:04:41
|
it says that if we were to do
|
|
0:04:47
|
we would send 15 times as much
|
|
0:04:51
|
as we were to send to router 3.
|
|
0:04:54
|
Now again, the actual load balancing is gonna be
|
|
0:05:01
|
So, if we're doing CEF switching,
|
|
0:05:02
|
CEF is gonna look at the
|
|
0:05:06
|
potentially, like the flow information.
|
|
0:05:10
|
to figure out what interface are we actually
|
|
0:05:15
|
So, it's not gonna be very predictable for us to
|
|
0:05:20
|
What we could do if we
|
|
0:05:23
|
very specific distribution of traffic
|
|
0:05:26
|
would be to turn CEF off, and do load
|
|
0:05:32
|
However, again, the disadvantage of doing
|
|
0:05:38
|
So, you technically could do that.
|
|
0:05:41
|
two packets towards router 5 for every one
|
|
0:05:46
|
but you probably better off just leaving CEF on,
|
|
0:05:49
|
and then, using whatever traffic
|
|
0:05:53
|
So, let's look at what are the possible
|
|
0:05:57
|
to reach this.
|
|
0:05:58
|
It says we could either go to router 3,
|
|
0:06:02
|
Now, I wanna know, for this distribution,
|
|
0:06:06
|
what is the actual amount of traffic
|
|
0:06:11
|
So, what I'm gonna do on 3's interface
|
|
0:06:14
|
that connects to router 2,
|
|
0:06:16
|
I'm going to configure an inbound ACL
|
|
0:06:20
|
that will be used as a packet counter.
|
|
0:06:24
|
Likewise, on router 5 as the
|
|
0:06:29
|
So, I'll say that "For any traffic that is going
|
|
0:06:36
|
I want to account for those packets."
|
|
0:06:39
|
Then, based on the counters, when we look at
|
|
0:06:43
|
we can actually see what is the
|
|
0:06:50
|
So first, let's go to router 3.
|
|
0:06:55
|
And we'll create this access list.
|
|
0:06:57
|
So, let's look at the access list currenntly, I wanna
|
|
0:07:03
|
So, I'll create Access List 100.
|
|
0:07:05
|
It says, Permit ICMP that is
|
|
0:07:12
|
then, I'm gonna permit anything else.
|
|
0:07:15
|
Permit IP Any Any.
|
|
0:07:18
|
Then, on the link that connects to router 2,
|
|
0:07:24
|
I'll essentially do the same thing now on router 5,
|
|
0:07:27
|
but on router 5, this is gonna be applied
|
|
0:07:34
|
On serial 0/0/0, IP Access Group 100 In.
|
|
0:07:44
|
So now, from router 2, let's say that we send...
|
|
0:07:47
|
a thousand packets to router 6.
|
|
0:07:49
|
So, we ping the loopback of 6. We'll say
|
|
0:07:56
|
Now, by looking at the...
|
|
0:08:00
|
Show Access List,
|
|
0:08:01
|
I'll put on router 3 versus 5.
|
|
0:08:07
|
It's gonna show us what the actual distribution is.
|
|
0:08:11
|
Now, we can see that router 5's
|
|
0:08:16
|
whle router 3's counter isn't doing anything.
|
|
0:08:21
|
It essentially means that CEF has calculated
|
|
0:08:28
|
for that particular source and destination pair.
|
|
0:08:32
|
Now, if I were to vary the input,
|
|
0:08:36
|
we have a couple different loopbacks.
|
|
0:08:39
|
Let's say, "Loopback 2 is...
|
|
0:08:41
|
the IP address 2.2.2.2."
|
|
0:08:48
|
And we have a bunch of...
|
|
0:08:52
|
secondary addressing on here.
|
|
0:09:00
|
2.2.2.3, and 2.2.2.4.
|
|
0:09:05
|
So, these are different sources
|
|
0:09:08
|
Okay, so now, I'll send the ping that's
|
|
0:09:13
|
Source from 2.2.2.1
|
|
0:09:16
|
with the repeat count of a thousand,
|
|
0:09:22
|
Because I'm not waiting for response back in,
|
|
0:09:27
|
We'll do the same thing for...
|
|
0:09:31
|
coming from 2.2.2.2.
|
|
0:09:36
|
Then, from the 3 address, and
|
|
0:09:40
|
So basically, I wanna know that when
|
|
0:09:46
|
what does it choose as the outgoing interface?
|
|
0:09:49
|
Ideally, we should see that at least
|
|
0:09:54
|
and some of it is gonna go to 5,
|
|
0:10:03
|
but we still don't see any hits on...
|
|
0:10:06
|
3's list.
|
|
0:10:09
|
Okay, so it means that CEF is
|
|
0:10:13
|
that we need to route the traffic towards router 5.
|
|
0:10:15
|
Now, what we could do to see if
|
|
0:10:20
|
would be to turn per-packet load balancing on,
|
|
0:10:26
|
and then, route the traffic through router 2.
|
|
0:10:31
|
So, let's say we do this. So,
|
|
0:10:35
|
I'm gonna enable routing with BB2.
|
|
0:10:41
|
Where on this link, it's pre-configured
|
|
0:10:46
|
So, the packets are gonna be transitting router 2.
|
|
0:10:48
|
They're not gonna be locally originated.
|
|
0:10:52
|
So on router 2, we'll do a
|
|
0:10:58
|
So, Router RIP Version 2 No Auto-Summary.
|
|
0:11:03
|
And we're running RIP on this...
|
|
0:11:05
|
link. 192.10.10.
|
|
0:11:10
|
Now, I believe that the backbone router
|
|
0:11:17
|
So, I will say Key Chain RIP.
|
|
0:11:22
|
Key 1, the key string is CISCO.
|
|
0:11:26
|
Then, at the link level, I have IP
|
|
0:11:32
|
And the IP RIP Authentication Key Chain is RIP.
|
|
0:11:37
|
Now, within the scope of the exam,
|
|
0:11:40
|
if they wanted to route this device, they'd have
|
|
0:11:45
|
Okay, it just so happened that I
|
|
0:11:50
|
So, if I look at the Show IP Route RIP,
|
|
0:11:53
|
I see that there are a bunch of routes
|
|
0:11:58
|
Okay, under the RIP process, I'm going
|
|
0:12:03
|
Then, under EIGRP, I'll redistribute RIP in there.
|
|
0:12:16
|
So, the overall goal for me here
|
|
0:12:24
|
And then send traffic to the loopback of router 6.
|
|
0:12:30
|
So, let's ping 150.10.6.6.
|
|
0:12:35
|
And first, let's look at the... Let's say, Show IP
|
|
0:12:47
|
addresses.
|
|
0:12:48
|
So, I'll ping 150.10.6.6 from 192.10.10.254...
|
|
0:13:01
|
from 220.20.3.1...
|
|
0:13:12
|
And then, from 222.2.2.1.
|
|
0:13:16
|
So now, if we look at router 3 and 5,
|
|
0:13:22
|
we see that we're still choosing
|
|
0:13:28
|
Okay, so next, let's go to router 2,
|
|
0:13:31
|
and we'll turn CEF off. We'll tell it to
|
|
0:13:35
|
So, on these interfaces, we'll say,
|
|
0:13:46
|
Then, at the link level, we'll say,
|
|
0:13:55
|
Now, in a real design, you would not
|
|
0:14:00
|
because there's a lot of applications
|
|
0:14:03
|
Especially like UDP things like voice or video,
|
|
0:14:10
|
Normally, you just wanna
|
|
0:14:12
|
where CEF is load balancing
|
|
0:14:19
|
not on a...
|
|
0:14:21
|
per-packet basis.
|
|
0:14:24
|
So now, let's go to... Back to
|
|
0:14:29
|
So, these couple different sources.
|
|
0:14:36
|
And then, let's look at router 3's output.
|
|
0:14:43
|
And 5's output.
|
|
0:14:46
|
So, what I may need
|
|
0:14:49
|
Let's say, Clear IP Cache,
|
|
0:14:52
|
and Show IP Interface.
|
|
0:14:57
|
is off,
|
|
0:15:00
|
and fast switching is off. So, it says,
|
|
0:15:05
|
I may need to look at this as a
|
|
0:15:10
|
So, let's say from BB1, we do this, but
|
|
0:15:22
|
So now, we see, some of the traffic
|
|
0:15:27
|
but proportionally, it will be larger on 5.
|
|
0:15:31
|
So, if I clear the counters
|
|
0:15:37
|
then, we look at the Show
|
|
0:15:40
|
it says we have 18, 19, 20 packets there.
|
|
0:15:43
|
But if we look at this on router 5,
|
|
0:15:47
|
we see it's much higher.
|
|
0:15:50
|
So, based on the default traffic share count
|
|
0:15:53
|
that router 2 calculates for
|
|
0:15:58
|
when we Show IP Route 150.10.6.6,
|
|
0:16:03
|
the traffic is no being load balanced
|
|
0:16:07
|
where for every 15 packets
|
|
0:16:10
|
there's gonna be one packet
|
|
0:16:15
|
Now, we could technically modify this,
|
|
0:16:18
|
but we'd have to look at what
|
|
0:16:22
|
interface by interface basis
|
|
0:16:28
|
to figure out why...
|
|
0:16:31
|
or to figure out where these values
|
|
0:16:33
|
So, if we were to look at the Show IP
|
|
0:16:42
|
as long as...
|
|
0:16:46
|
the routes of router 3 stays
|
|
0:16:52
|
then, we will be able to use
|
|
0:16:56
|
So, let's look at what is the
|
|
0:17:00
|
is using on interface serial 0/0.1.
|
|
0:17:06
|
Okay, which is the link that goes to...
|
|
0:17:08
|
router 5.
|
|
0:17:10
|
Now, if we were to go to this link
|
|
0:17:14
|
let's say we set it to twice
|
|
0:17:18
|
which would be 4,000
|
|
0:17:24
|
Then, we Clear IP EIGRP Neighbors.
|
|
0:17:28
|
We should see now when we
|
|
0:17:36
|
that now, the route from router 5
|
|
0:17:42
|
we're now looking at 2 million.
|
|
0:17:45
|
2.1 million versus 25 million.
|
|
0:17:49
|
But if we now look at the routing table,
|
|
0:17:58
|
we can see now, the traffic
|
|
0:18:03
|
Where previously, we had a ratio of 15:1,
|
|
0:18:08
|
So, if I were to continue to
|
|
0:18:13
|
so, let's say I set it to...
|
|
0:18:16
|
a 100,000 microseconds, which is
|
|
0:18:24
|
then, Clear IP EIGRP Neighbors.
|
|
0:18:26
|
As long as this link stays
|
|
0:18:31
|
then, when we Show IP Route 150.10.6.6,
|
|
0:18:38
|
we'll see now, the traffic shared count
|
|
0:18:45
|
So, if we wanted to be exact for this,
|
|
0:18:48
|
we'd have to look at the actual metric values,
|
|
0:18:51
|
and then, send the end-to-end ratio to
|
|
0:19:00
|
So, you'll see in the volume 1 workbook,
|
|
0:19:04
|
I highly doubt they would have
|
|
0:19:07
|
It's just more to show you
|
|
0:19:09
|
do that and where EIGRP is getting
|
|
0:19:17
|
But the main key point to
|
|
0:19:20
|
is that no matter what you
|
|
0:19:24
|
if the alternate routes are
|
|
0:19:28
|
they're not gonna be considered
|