Link to home
Start Free TrialLog in
Avatar of jskfan
jskfanFlag for Cyprus

asked on

EIGRP Bandwidth Commands

I would like to have clear idea about the difference between the commands below, when used in EIGRP:
Bandwidth <Value>
Example : Bandwidth 256

Ip Bandwidth-percent EIGRP <EIGRP AS>  <Percent Value>

example : ip Bandwidth-percent EIGRP 1 60

when shoud I use one and not the other ?


Thank you
SOLUTION
Avatar of William Miller
William Miller
Flag of United States of America image

Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial
SOLUTION
Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial
Avatar of jskfan

ASKER

Thank you for your explanation..

I know EIGRP uses by default 50% of the Bandwidth  (Though I am not sure why is different than OSPF ).

What I need to know in which scenario that you need to go to the interfce and type for instance bandwidth 256 and in which scenario you go to the interface and type for instance : ip Bandwidth-percent EIGRP 1 60

I know that They both have influence on the cost..but .when to use one command and not the other command
Avatar of Predrag Jovic
If you apply bandwidth command to interface - EIGRP and OSPF link cost (if using auto calculation) will be calculated accordingly. It will not limit bandwidth by itself in any way.

ip bandwidth-percent - will configure maximum amount of bandwidth can be used for EIGRP (it can go over 100%). The point is, that, if bandwidth value is to low (for example there are several subinterfaces on serial link) bandwidth percent over 100% may be needed to make EIGRP have enough bandwidth to properly function (if traffic forwarding is relaying on routing protocol and routing protocol is down all traffic will be dropped).

However, bandwidth command should not be used to influence EIGRP calculation, since in combination with QoS can/will limit traffic bandwidth on interface, and can also influence other routing protocols. It should actually be used only to match real interface bandwidth (for example interface itself is 100Mb, but ISP is providing 10Mb traffic in that case you should configure bandwidth on interface and QoS (typically shaping with priority queues), it is better for you to choose which traffic will be dropped, than ISP is randomly dropping your traffic).

PS
it is not subject of this topic, but for EIGRP traffic engineering changing delay would be right choice since only EIGRP is using it in calculations and also does not influence in any way delay on interface.
Avatar of jskfan

ASKER

I am not sure why the concern about the Percentage of the Bandwidth is just with EIGRP, other protocols do not limit themselves to 50%.
I also do not understand how come EIGRP cannot detect the real bandwidth of the Serial Interface
ASKER CERTIFIED SOLUTION
Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial
Avatar of jskfan

ASKER

Thank you Very much Guys
You're welcome.