From: Ozgur Guler (Garanti Teknoloji) (OzgurG@garanti.com.tr)
Date: Sun Oct 19 2003 - 10:04:33 GMT-3
hi group,
in the document
http://www.cisco.com/en/US/tech/tk365/tk207/technologies_tech_note09186a008009437d.shtml#topic1
there is a case where
two routes to a network, have a metric ratio of 3/2 ...
so what this cco doc says is that
put variance 2 under eigrp
and you ll do 1/1 load balancing (round down to nearest integer)...
first packet will follow route 1, second route 2, 3rd route 1 etc...
what i came up with is...
it does 3/2 load balancing (though i have the same configs and have the same metrics in the routing table)...
3 packets via route 1, then 2 packets via route 2, 3 packets via route 1 etc..
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.23.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.23.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.24.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.24.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.24.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.23.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.23.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.24.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.24.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.24.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.23.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.23.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.24.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.24.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.24.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.23.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.23.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.24.2
r2#sh ip route 192.168.90.0
Routing entry for 192.168.90.0/24
Known via "eigrp 1", distance 90, metric 7680, type internal
Redistributing via eigrp 1
Last update from 192.168.23.3 on Serial0:1.2, 00:09:14 ago
Routing Descriptor Blocks:
* 192.168.24.4, from 192.168.24.4, 00:09:14 ago, via Serial0:1.3
Route metric is 7680, traffic share count is 3
Total delay is 300 microseconds, minimum bandwidth is 64 Kbit
Reliability 255/255, minimum MTU 1500 bytes
Loading 3/255, Hops 2
192.168.23.3, from 192.168.23.3, 00:09:14 ago, via Serial0:1.2
Route metric is 11520, traffic share count is 2
Total delay is 450 microseconds, minimum bandwidth is 64 Kbit
Reliability 255/255, minimum MTU 1500 bytes
Loading 3/255, Hops 2
r2#sh ip route eigrp
...
D 192.168.90.0/24 [90/7680] via 192.168.24.4, 00:09:39, Serial0:1.3
[90/11520] via 192.168.23.3, 00:09:39, Serial0:1.2
...
+++++++++++++++++++++++
case 2
for another case, after playing with metrics
i have
D 192.168.90.0/24 [90/11520] via 192.168.24.4, 00:01:10, Serial0:1.3
[90/11520] via 192.168.23.3, 00:01:10, Serial0:1.2
[90/12800] via 192.168.28.8, 00:01:10, Serial0:1.7
11520/12800 = 0.9
and here
the router does 10/10/9 load balancing....
according to this doc it should do 1/1/1 load balancing...
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.24.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.24.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.24.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.24.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.24.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.24.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.24.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.24.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.24.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.24.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.23.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.23.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.23.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.23.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.23.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.23.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.23.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.23.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.23.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.23.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.28.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.28.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.28.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.28.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.28.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.28.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.28.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.28.2
ICMP: echo reply sent, src 192.168.90.9, dst 192.168.28.2
am i missing/catching something here????
Ozgur
-----Original Message-----
From: nobody@groupstudy.com [mailto:nobody@groupstudy.com]On Behalf Of
kasturi cisco
Sent: Saturday, October 18, 2003 5:16 PM
To: hunt_lee@bigpond.net.au
Cc: ccielab@groupstudy.com
Subject: Re: Am I correct?
Hunt,
Looks right to me (except a small typo varience 6 ..should be variance
6). The other thing which you should possible note (may be u are aware
of) is
In this case u did not have much problem as u had both the routes in ur
EIGRP table. But lets us say bcoz of the Feasibilty criteria not being
met the second route is NOT in the table itself and u want to include it
for load balancing.
In such cases try to disconnect the better path link and connect the
worse path link. Get the metric from both.
then worse metric/better metric= X . round it off the get the
variance.Connect the better path and set variance to include the path
into EIGRP topology. The do the load balance as you have done. U could
extend the load balancing using load-balance min or balanced
(default..what u have)
http://www.cisco.com/en/US/tech/tk365/tk207/technologies_tech_note09186a008009437d.shtml#topic1
Thanks for ur post ....that i also revisited this topic and re-read.
Good Luck,
Kasturi.
>From: Hunt Lee >Reply-To: Hunt Lee >To: ccielab@groupstudy.com >Subject:
Am I correct? >Date: Fri, 17 Oct 2003 09:58:08 +1000 > >Hey Group, > >Can
someone please confirm that my method is correct for EIGRP load
balancing? > >R5 ----- R6 ---- 142.1.6.0/24 > >E.g. At R5's
EIGRPtopology's table > >P 142.1.6.0/24, 1 successors, FD is 409600 >via
142.1.56.6 (409600/128256), Ethernet0 >via 142.1.65.6 (2297856/128256),
Serial1 > >Now, if I want to do equal cost load balance between the 2
links > >2297856 - 409600 = 1888256 > >Then on R5: >offset-list 0 in
1888256 Ethernet0 > >Correct? > >But if I want to do un-equal cost load
balancing, say if I want to send 6 packets to Ethernet 0, then send 1
packet to Serial 1. > >409600 * 6 = 2457600 > >2457600 - 2297856 = 159744
> >Then on R5: >offset-list 0 in 159744 Serial1 >varience 6 > >Am I
correct? > >Thank you so much!!! >Hunt >
>_______________________________________________________________________
>Please help support GroupStudy by purchasing your study materials from:
>http://shop.groupstudy.com > >Subscription information may be found at:
>http://www.groupstudy.com/list/CCIELab.html
------------------------------------------------------------------------
Calling all NRIs! Avail of the best financial services. Smile all the way
with ICICI Bank.
This archive was generated by hypermail 2.1.4 : Mon Nov 24 2003 - 07:53:04 GMT-3