EIGRP uses composite metric which is calculated from more parameters: bandwidth, load, delay, reliability.
Don't forgot that bandwith is not cummulative, it is the minimum (slowest) value along the path.
Regarding the delay value, it is cummulative, so it contains the sum of all delays along the path.
There is an equation for the calculation.
There can be situations in which the router will choose the route with worse metric. This arise when you have more options, but only some of the routes do pass the feasibility condition. For more information, please refer to this topic: https://supportforums.cisco.com/message/3994397#3994397 There is a lot of explained, regarding EIGRP.
If you have any questions or want to be more specific, please, don't mind to ask!
that is absolutely ok, take your time. The topic regarding EIGRP is very large and it takes more than a while to grap the principles.
So, let's start .
1. What is metric in EIGRP and how we can calculate ?
Metric, in common, is a property of a route, consisting of one or more values, that is used by routing protocol to decide "how good" a route is. It is used when comparing multiple routes to the same destination, to decide which one is better.
Metric in EIGRP consists of:
bandwidth =10000000/min(bandwidth) * 256 where min(bandwidth) is the least bandwidth of all outgoing interfaces on the route to the destination network represented in kilobits.
delay =sum(delay) where sum(delay) is the sum of all delays on the route to the destination network represented in tens of microseconds.
load dynamically evaluated load of the link, which ranges from 1 to 255 and is multiplied by 256, lower is better
reliablity dynamically evaluated, ranges from 1 to 255 and is multiplied by 256, greater value is better
MTU - is not included in the calculation
Hop Count - is not included in the calculation
However, default metric setting uses only bandwidth and delay parameter.
The multiplication by 256 is there for backwards compatibility with IGRP, the EIGRP's predecessor.
2. What is the parameter of metric like example msec or micro sec ?
You can see, that the metric is a combination of multiple quantities, there is no common unit for metric like kbit/s or so, if that was the question...It is just a number.
3. If you put the command sh ip eigrp topology.... It will show the best routes path with Successor and FS with the attached with FD and AD value.
Is the FD is the metric values or something else in EIGRP ?
Yes, the FD refers to metric value. If you mean AD, like advertised distance, that is an term that happened to be in older literature, but now we call it RD - reported distance. That is, because AD is now widely used as Administrative Distance, which describes preference of routing protocols (EIGRP has AD of 90 for internal non-summarized routes, OSPF 110, etc.)
FD - reasible distance is the best metric along a path to a destination network, including the metric to the neighbor advertising that path.
RD(formerly AD) - reported distance is the total metric along a path to a destination network as advertised by an upstream neighbor.
Now coming my question is Metric value which I asked and you replied me that low metric value is good one.
Can you prove me as a example wise......
Now I have few question on this.....
In practically I don't think is like that
I have seen many times If I increased the bandwidth value then EIGRP will select that path.
Means Higher Bandwidth value is a Primary path and Lower Bandwidth is secondary path.
So As per the formula
Metric= 256*( Bandwidth +Sum of all Delay)
Where Bandwidth = 10^7/ Bandwidth
And Delay in micro sec
Pls clear me if I am wrong......
As you stated before, if you have incereased the bandwidth value then EIGRP has selected that path. Therefore, the increase of bandwidth (if increased all along the path) has indeed a positive effect on metric calculation.
So if I increased the Bandwidth it will increased the Metric value.
No, you will in fact decrease the metric value. That is, becasuse - as you mentioned few sentences before Bandwidth = 10^7/ Bandwidth
Therefore, it is the inverted value. The more you increase the bandwidth, the lower value you get from that equation, thus, effectively lowering the final metric.
Then how can you say it will not depend on Bandwidth
Excuse me, but I have never said that it does not depend on bandwidth. I am sorry if I haven't made myself crystal clear.
Let assume I have 3 route which connected one like that way..
R1 is connected to R2 with 100 Mbps link
R1 is connected to R2 with 10 Mbps link
R2 is connected to R3 with 10 Mbps link.
Means R1 will have two link one is 100 Mbps and another is 10 mbps which are connected to R2 and R2 is connected R3 with 10 Mbps link
The lan network of R1 needs to be reach lan of R3.
If you generally configure EIGRP here. I have seen the practically it will choose 100 Mbps link path as a Successor and 10 Mbps link path is FS.
Once the 100 Mbps link goes down then it will take 10 Mbps link as primary.
I agree and don't agree at the same time. In a configuration like this, the bandwidth of 100Mbps doesn't matter at all because 10Mbps (the minimum of all bandwidths along the path) is used in metric calculations. From the point of EIGRP, both paths are equally good (via 100Mbps and via 10Mbps).
With default setting, the only thing that might matter is the Delay parameter. That is 100 usec for 100Mbps link and 1000usec for 10Mbps link. In this case, the delay will have the final word, because bandwidth used in calculation is the same. Note, that delay is cumulative, so it adds up along the path. Bandwidth is the minimum along the path, therefore 10Mbps.
Let's say you just want to reach R3's 10Mbps Ethernet interface, not its LAN because that would bring another interface (towards the lan) to the calculation.
R1 ---100Mbps link --R2 ---10Mbps link ---R3
bandwidth 10^7Kbps/10 000 Kbps and delay 100usec+1000usec therefore 1100usec
Note: delay in the equation needs to be in tens of usec, therefore we need to divide it by 10 first.
Because metric of 284160 is lower than 307200, the path you have suggested is prefered.
Note that bandwidth is the same, only delay matters.
This is the reason why the faster link is used. Not because of the bandwidth parameter, but delay parameter.
If you still don't trust me, I can provide you with example in packet tracer and prove this point to you.
So can you tell me how it will happen? Now how can you tell me the lower metric is good way.....
That is because bandwidth is not metric. It is used in the calculation of metric as one of the parameters, just like I said before. When you increase the bandwidth parameter on the router, the metric is lowered (because bandwidth in the equation of EIGRP metric is calulated as 10^7/bandwidth)
And tell me here what is Delay role here.....
You clear me as a example wise because now I have fully confused about this....
Hope you are clear my question........
You can see from your previous example, how the delay works.
Hope I made things a little bit clearer now .
Please, if you have any further questions, feel free to ask.
So for a link that has 100Mbps and a delay of 500 you will get the following metric value 153600
There are 5 K values in this metric (K1, K2, K3, K4, and K5)
I know I included the EIGRP Metric algorithm above, but that's not technically the actual algorithm. The actual algorithm is rather long, and
I can remember if off the top of my head. But, the K values are used as plugins to the metric so to speak. So think of it, as one big algorithm, and the K values are values that you plug in to calculate the metric. By default, K1 = 1 and K3 = 1, (K2,K4, and K5 are 0 so they are not really calculated)
Bandwidth, Delay, Reliability, Load, and MTU
MTU is not used in this formula at all, but can be used in certain tie-breaking situations. By default only Bandwidth and Delay are used, and really should be the only ones ever used. The reason I say this, is because, EIGRP, will not dynamically adjust to changes in Reliability, Load, and MTU, like it will Bandwidth and Delay. For instance, if you include Load in the EIGRP Composite Metric calculation, and the load changed, EIGRP will not dynamically adjust the cost of the link to this change.
There are several definitons that you need to be aware of with EIGRP.
Feasible Distance - This is the best cost metric to the destination network, since the last Active > Passive change.
Reported Distance - This is the bst cost metric to the destination network as reported by an EIGRP neighbor.
Successor - This is the best cost metric to the destination network, and will be the one included in the routing table
Feasible Successor - This is a route that has passed the Feasibiliy Condition check. This checks, if the route has a Reported Distance to the destination network, that is lower than the current Feasible Distance to the destination network. If this check passed, the route will be included in the Topology table, but NOT the routing table. If the Successor goes down for any reason, EIGRP will automatically check its Topology table for any Feasible Successors, and promote it to a Successor
So let's say you have a router Router_A with two links (100Mbps and 10Mbps). So, let's say the 100Mbps link is the successor, and the 10Mbps passed the Feasibility Condition, and becomes a Feasible Successor. EIGRP will automatically go to this route, as soon as the Successor goes down.
If you want to see the specific values of your eigrp routes, create a simple eigrp topology and run the following command.
'show ip route x.x.x.x'
This will show you the cumulative delay, and various others values that will help you compute the path yourself.
Question We run asr9001 with XR 6.1.3, and we have a very long delay to
login w/ SSH 1 or 2 to the device compare to IOS device. After
investigation, the there is 1s delay between the client KEXDH_INIT and
the server (XR) KEXDH_REPLY. After debug ssh serv...
Introduction The purpose of this document is to demonstrate the Open
Shortest Path First (OSPF) behavior when the V-bit (Virtual-link bit) is
present in a non-backbone area. The V-bit is signaled in Type-1 LSA only
if the router is the endpoint of one or ...
Hi, I am seeing quite a few issues with patch install and wanted to
share my experience and workaround to this. Login to admin via CLI, then
access root with the “shell” command Issue “df –h” and you’ll probably
see the following directory full or nearly ...