From: Scott M Vermillion (scott_ccie_list@it-ag.com)
Date: Wed Mar 11 2009 - 01:04:42 ARST
And of course just for the record and the sake of being 100% accurate,
when I mistakenly said "Recall that BW is cumulative along the path to
the destination network but BW is simply the lowest reported value
along the entire end-to-end path..." I *of course* meant to say that
DLY is cumulative, while BW is not. Amazing how you can proofread
something before hitting send and it all looks kosher until you get
your return copy hours later and go "WTF was I thinking?!"
LOL
On Mar 10, 2009, at 11:31 , Scott M Vermillion wrote:
> Hi Marish,
>
> This is definitely an EIGRP "gotcha!" I confess that I was
> scratching my head on this one for a bit there myself (although half
> of that was just trying to decipher the ASCII art!). Recall that BW
> is cumulative along the path to the destination network but BW is
> simply the lowest reported value along the entire end-to-end path -
> it is *not* calculated cumulatively along each segment! If you
> notice, the delta between your RD as reported by R6 to R1 and your
> FD as calculated at R1 is exactly 2560, which is 10 x 256, which is
> simply DLY x 256 for that cumulative segment along the path.
>
> So when you say "10000000/100000 = 100 + 10 = 110 X 256 = 28160 R1
> distane to R6," it is important to note that there really isn't a
> "distance" from R1 to R6, per se. Consider if you were on the R3
> CLI and looking at the route for the network segment between R6 and
> SW1. Now further consider that the R3<->R1 segment was GIG-E and
> likewise the R6<->SW1 segment was GIG-E. Keep the R1<->R6 segment
> as Fast-E. Now the lowest reported BW along the path is obviously
> R1<->R6, so your "distance" between those two routers is no longer a
> fixed 2560. See what I mean? It's all relative to where you're
> looking at what with EIGRP, because the segment you're contemplating
> may or may not represent the BW bottleneck along the path for a
> given route.
>
> Simply put, BW only factors into the metric once and it is the
> lowest value from end-to-end.
>
> Regards,
>
> Scott
>
> (Also just BTW, isn't it the case that one of the segments along the
> path from R6 to 155.1.37.0/24 is actually 10 Mbps vs. all Fast-E?
> That's the only way the RD from R6 as seen on R1 makes any sense to
> me [10^7/10,000 = 1000, then add delay of 100 for E plus delay of
> 10 for Fast-E, all x 256 = 284160])
>
>
> On Mar 10, 2009, at 12:25 , marish shah wrote:
>
>> Hi Experts,
>>
>> Yesterday I was facing problem with eigrp lab.problem is reladed
>> to FD (fesible distance) AD (advertise distance)
>>
>> ok Here is my senario
>>
>>
>> Serial 1/3 155.1.13.3 155.1.13.1 Serial 1/0
>> Fa0/0 155.1.146.1 155.1.146.6
>> Fa0/0.146
>>
>> R3------------------------------------------------------
>> R1
>> ------------------------------------------------------------------------R6
>>
>> |
>> |
>> |Fa0/0
>> 155.1.37.3
>> |
>>
>> |
>> 155.1.67.6 Fa0/0.67 |
>>
>> |
>> |
>>
>> |
>> |
>>
>> |
>> |
>>
>> |
>> |
>>
>> |
>> |
>>
>> |
>> |
>>
>> |
>> |
>>
>> |
>> |
>>
>> |
>> |
>>
>> |
>> |
>>
>> |
>> |
>>
>> |
>> |
>>
>> |
>> |
>>
>> |
>> |
>>
>> |
>> |
>>
>> |
>> |
>>
>> |
>> |
>>
>> |
>> |
>>
>> |
>> |
>>
>> |
>> |
>>
>> |
>> |
>>
>> |
>> |
>>
>> |
>> |
>>
>> |
>> |
>>
>> |
>> |
>>
>> |
>> |
>>
>> |
>> |
>> | 155.1.37.7 Fa
>> 1/3 155.1.67.7 Vlan
>> 67 |
>>
>> |----------------------------------------------------------
>> SW1
>> ---------------------------------------------------------------------|
>>
>> R1 fastethernet 0/0 is conectd with R6 Fastethernet 0/0.146
>> R1 Serial 1/0 is conected with R3 Serial 1/3
>> R3 fastethernet0/0 is conected with SW1 fastethernet 1/3
>> R6 fastethernet 0/0.67 is conected with SW fastethernet 1/6
>>
>> Now I enable eigrp 100 on all routers.
>>
>> After runing eigrp 100 I try to calculate FD and AD my traget is
>> interface
>> between R3 & SW1 (155.1.37.0) I'm on router 1
>>
>> Here is sh ip eigrp topology on R1
>>
>> R1#sh ip eigrp topology
>>
>> P 155.1.37.0/24, 1 successors, FD is 286720
>> via 155.1.146.6 (286720/284160), FastEthernet0/0 ------
>> primary path
>> via 155.1.13.3 (2195456/281600), Serial1/1 -------
>> Backup path
>> here u see I have two paths to reach 155.1.37.0 so now decision on
>> FD I'm
>> get lowest FD ( 286720) from 155.1.146.6 so this my primary path.
>> On this point every thing is OK I calculate my AD ( 281600) the
>> result is
>> same.But when I try to calculate my FD on router 1 my mind stuck.AS
>> I know
>> for FD first we take our AD , second we take our distance ( R1 to
>> R6) then
>> plus both and we get our FD.But its not work.
>>
>> On R1 MY FD is 286720 for calculate my distance to R6
>>
>>
>> Rack1R1#sh interfaces fastEthernet 0/0
>>
>> BW 100000 Kbit, DLY 100 usec,
>>
>> 10^7/BW + DELAY x 256
>>
>> 10^7= 10000000
>> 10000000/100000 = 100 + 10 = 110 X 256 = 28160 R1 distane to R6
>>
>> Now 28160 + 284160 = 312320 but as u see in topology table FD is
>> 286720
>> .................
>> So how can I calculate my FD ?
>>
>>
>> your suggestion is really helpful to me.
>>
>> Regards,
>> Thanks.
>>
>>
>> Blogs and organic groups at http://www.ccie.net
>>
>> _______________________________________________________________________
>> Subscription information may be found at:
>> http://www.groupstudy.com/list/CCIELab.html
>
>
> Blogs and organic groups at http://www.ccie.net
>
> _______________________________________________________________________
> Subscription information may be found at:http://www.groupstudy.com/list/CCIELab.html
Blogs and organic groups at http://www.ccie.net
This archive was generated by hypermail 2.1.4 : Mon Apr 06 2009 - 06:44:04 ART