EIGRP Formula question

image

 

the above formula is as presented in Doyle volume 1

the formula as presented in the Route book however is (the last argument >>>>  

+ [k5/(reliability+k4)]

 

its the multiplication versus the addition of the last section of the formula is what my question is about...and credit where its due...as a friend of mine not me pointed out.

If reliability is 255..which it is lets say in FastE interfaces. Then with default K values the last part of the formula goes to

[0/(255+0)] = 0

 

if you multiple 0 with the rest of the formula then the WHOLE thing goes to zero...the formula presented on line in the 10 or so spots i checked is as presented as Doyle has done so..including the cisco site

http://www.cisco.com/en/US/tech/tk365/technologies_white_paper09186a0080094cb7.shtml

 

is my math wrong here?  

 

 

 

 

Comments

  • This is from the wikipedia extract:

     


    "An important and totally non-obvious fact is that if K5 is set to zero, the term  is not used (i.e. taken as 1).

    The default is for K1 and K3 to be set to 1, and the rest to zero, effectively reducing the above formula to (Bandwidth + Delay) * 256."

     

     

  • This is from the wikipedia extract:

     

     

    "An important and totally non-obvious fact is that if K5 is set to zero, the term  is not used (i.e. taken as 1).

    The default is for K1 and K3 to be set to 1, and the rest to zero, effectively reducing the above formula to (Bandwidth + Delay) * 256."

     

     

     

    well who woulda thunk! of all places...wiki!

     

    Thankyou!

  • I think your math calculation is right and the last term is coming out to be zero if K5=0. But in the same formula note in both Jeff Doyle and Cisco Doc its also metioned that "if K5 is set to zero , the [k5/(RELIABILITY+K4)] term is not used".

    Howerever if K5 is not zero then the same formula (multiplication with the last term)will still be applicable. So formula is correct is k5 is non zero value and if K5=0 then just dont use the last term. 

     

    Regards,

  • again..thankyou guys...i should pay more attention when i read...

     

    1 more question again..mybe i am reading this wrong

     

    Doyle pg 274 Top

    " a router begins a diffusing computation by sending queries to all of its neighbors. The query will contain the new locally calculated distance to the destination" 

    the bold part is what i have an issue with.

    You begin a diffusing computation when an input event happens (pg 273) and you fail to find a feasible successor in your topology table. in effect you DONT HAVE a locally calculated distance to the destination...thats the whole point of starting DUAL

     

    to understand above question you need to have doyle on ya...

    any answers appreciated

     

     

  • Hi,

      Think of the following: who starts in EIGRP the query process (all other EIGRP neighbors receive the query and if have no asnwer query further their neighbors and so on, but this is not really important in here)? The guy who actually detects the topology change, for example the guy that was advertising its Fa0/0 subnet in EIGRP, and at one moment Fa0/0 goes down. At this moment, what the router does in its query, it's actually advertising the prefix with infinite metric, this is the "calculated distance to destination". From here on, there is probably 1 page story on what happens in which cases.

    Good luck with your studies!

  • Hi,

      Think of the following: who starts in EIGRP the query process (all other EIGRP neighbors receive the query and if have no asnwer query further their neighbors and so on, but this is not really important in here)? The guy who actually detects the topology change, for example the guy that was advertising its Fa0/0 subnet in EIGRP, and at one moment Fa0/0 goes down. At this moment, what the router does in its query, it's actually advertising the prefix with infinite metric, this is the "calculated distance to destination". From here on, there is probably 1 page story on what happens in which cases.

    Good luck with your studies!

     

    thankyou...got it...so i guess you notice an input event and advertize delay of 0xFFFFFF in your outgoing query

     

    doyle later says that the router which detects the input event changes its feasible distance to the particular destination to an infinite delay so that when it gets the query replies..the advertised distance by the feasible successors passes the feasibility condition...

     

    but hhe did not explain the sequence of events properly...

     

    Thankyou as always for the prompt explaination :-)

     


  • Rack1R2#sh ip eigrp topology ?

      <1-65535>        AS Number

      A.B.C.D          IP prefix <network>/<length>, e.g., 192.168.0.0/16

      A.B.C.D          Network to display information about

      active           Show only active entries

      all-links        Show all links in topology table

      detail-links     Show all links in topology table

      pending          Show only entries pending transmission

      summary          Show a summary of the topology table

      zero-successors  Show only zero successor entries

      |                Output modifiers

     

     

     

     

    1 more question

    what is the difference between the 2 commands in bold...I dont see a difference but there has to be..right?


  • what is the difference between the 2 commands in bold...I dont see a difference but there has to be..right?

    Yes, if you look description of both command are same:

     all-links        Show all links in topology table
      detail-links     Show all links in topology table
    and in output also don't have difference:

    R2#show ip eigrp topo all-links
    EIGRP-IPv4 Topology Table for AS(100)/ID(2.2.2.2)
    Codes: P - Passive, A - Active, U - Update, Q - Query, R - Reply,
           r - reply Status, s - sia Status

    P 11.11.11.11/32, 1 successors, FD is 2560002816, serno 8
            via 12.12.12.1 (2560002816/2560000256), FastEthernet1/0
    P 12.12.12.0/24, 1 successors, FD is 28160, serno 1
            via Connected, FastEthernet1/0
    P 2.2.2.0/24, 1 successors, FD is 128256, serno 2
            via Connected, Loopback0
    P 1.1.1.1/32, 1 successors, FD is 156160, serno 5
            via 12.12.12.1 (156160/128256), FastEthernet1/0

    R2#show ip eigrp topo detail-links
    EIGRP-IPv4 Topology Table for AS(100)/ID(2.2.2.2)
    Codes: P - Passive, A - Active, U - Update, Q - Query, R - Reply,
           r - reply Status, s - sia Status

    P 11.11.11.11/32, 1 successors, FD is 2560002816, serno 8
            via 12.12.12.1 (2560002816/2560000256), FastEthernet1/0
    P 12.12.12.0/24, 1 successors, FD is 28160, serno 1
            via Connected, FastEthernet1/0
    P 2.2.2.0/24, 1 successors, FD is 128256, serno 2
            via Connected, Loopback0
    P 1.1.1.1/32, 1 successors, FD is 156160, serno 5
            via 12.12.12.1 (156160/128256), FastEthernet1/0

    HAPPY STUDY

    [:D]

  • Thanks!

     

     

    i guess cisco made that one up to confuse people on the route exam!

  • From Cisco command reference guide:

     


    all-links

    (Optional) Displays all topology entries and all links (paths) instead of displaying only feasible paths.

    detail-links

    (Optional) Displays all topology entries with additional detail.

  • detail-links

    (Optional) Displays all topology entries with additional detail.

    But problem is we couldn't see any additonal...[:P]

    HAPPY STUDY

    [:D]

  • Yeah true, I did a quick gns3 lab and don't see any difference in the outputs either :D

    Regards,

    Arun






  • Hi, 

    Is there maybe a show ip eigrp topology all-links PREFIX, so can you specify the prefix in cause? This may be the answer.




    Good luck with your studies!



    Sent from my iPhone



    On Nov 14, 2011, at 4:37, "nnn" <[email protected]> wrote:






    imageToxicAvenginator:

    what is the difference between the 2 commands in bold...I dont see a difference but there has to be..right?

    Yes, if you look description of both command are same:

     all-links        Show all links in topology table

      detail-links     Show all links in topology table

    and in output also don't have difference:

    R2#show ip eigrp topo all-links

    EIGRP-IPv4 Topology Table for AS(100)/ID(2.2.2.2)

    Codes: P - Passive, A - Active, U - Update, Q - Query, R - Reply,

           r - reply Status, s - sia Status



    P 11.11.11.11/32, 1 successors, FD is 2560002816, serno 8

            via 12.12.12.1 (2560002816/2560000256), FastEthernet1/0

    P 12.12.12.0/24, 1 successors, FD is 28160, serno 1

            via Connected, FastEthernet1/0

    P 2.2.2.0/24, 1 successors, FD is 128256, serno 2

            via Connected, Loopback0

    P 1.1.1.1/32, 1 successors, FD is 156160, serno 5

            via 12.12.12.1 (156160/128256), FastEthernet1/0



    R2#show ip eigrp topo detail-links

    EIGRP-IPv4 Topology Table for AS(100)/ID(2.2.2.2)

    Codes: P - Passive, A - Active, U - Update, Q - Query, R - Reply,

           r - reply Status, s - sia Status



    P 11.11.11.11/32, 1 successors, FD is 2560002816, serno 8

            via 12.12.12.1 (2560002816/2560000256), FastEthernet1/0

    P 12.12.12.0/24, 1 successors, FD is 28160, serno 1

            via Connected, FastEthernet1/0

    P 2.2.2.0/24, 1 successors, FD is 128256, serno 2

            via Connected, Loopback0

    P 1.1.1.1/32, 1 successors, FD is 156160, serno 5

            via 12.12.12.1 (156160/128256), FastEthernet1/0



    HAPPY STUDY

    Big Smile








    INE - The Industry Leader in CCIE Preparation

    http://www.INE.com



    Subscription information may be found at:

    http://www.ieoc.com/forums/ForumSubscriptions.aspx


  • Is there maybe a show ip eigrp topology all-links PREFIX, so can you specify the prefix in cause? This may be the answer.

    NO,

    R2#show ip eigrp topology detail-links ?
      |  Output modifiers
      <cr>
    R2#show ip eigrp topology all-links ?
      |  Output modifiers
      <cr>

    HAPPY STUDY

    [:D]

Sign In or Register to comment.