From: Rohan Grover (rohang@cisco.com)
Date: Tue Oct 19 2004 - 06:13:03 GMT-3
Hi Danny,
This *could* be a bug with the Sup720 redundancy feature, the EIGRP K-values might not be synched properly to the standby hence on
switchover, you're facing the problem.
I suggest you talk to Cisco TAC in your area and find out if there is an existing bug on this.
Thanks
Rohan
-----Original Message-----
From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf Of Danny Ng
Sent: Tuesday, October 19, 2004 2:24 PM
To: 'ccie2be'; 'ccielab'
Subject: RE: EIGRP K-Value Mismatch
The eigrp delay seems giving me problems when I toggle the Cat6509 Sup720 to secondary Sup720. The log files shown EIGRP K-value
mismatch for all routers that I'd make the changes to the delay parameters. For instance, delay 15000 on router_A f0/0 and delay
10000 on Cat6509 f1/1.
Note: it is working fine normally without any EIGRP K-Value mismatch but except reboot the primary sup.
The eigrp K3 (delay) value is consider mismatch when point-to-point link was configured by using different delay setting? I just try
to think from unidirectional way and by right the eigrp metrics is different for up/down link.
Appreciate anyone who able to contribute ideas on this.
Regards,
Danny
-----Original Message-----
From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf Of ccie2be
Sent: Tuesday, October 19, 2004 3:06 PM
To: Danny Ng; ccielab
Subject: Re: EIGRP K-Value Mismatch
Danny,
When you say, "eigrp delay", do you mean the delay on the interface eigrp using to calculate it's metric?
The eigrp K-value is used to specify how much importance to give a component of the eigrp metric. As you know, eigrp has several
components such as bandwidth, load reliabitlity, and delay it can use to determine it's metric. By default, not all componenets are
used because the K-value for those unused component is zero. So, by default, eigrp says the "importance" of all components except
bandwidth and delay is zero because the k-value for those components is zero.
However, the "importance" of bandwidth and delay is, by default, 1 because those components of the metric have a K-value of 1.
Now, if you wanted eigrp to consider bandwidth twice as important as delay, for example, you would make bandwidth's K-value twice as
large as the delay's K-value.
Now, if the K-value of all eigrp routers aren't the same, you'll have problems because now the eigrp routers aren't using the same
formula to compute metrics making the metric meaningless.
If you look in the doc cd under eigrp commands, you'll find the formula eigrp uses to compute it's metric. I think the command is
metric weigth.
HTH, Tim
----- Original Message -----
From: "Danny Ng" <danny_ng@cnadvisers.com>
To: "ccielab" <ccielab@groupstudy.com>
Sent: Tuesday, October 19, 2004 2:42 AM
Subject: EIGRP K-Value Mismatch
> Changing the EIGRP delay (not same for both ends) is it consider
indirectly
> modified the EIGRP K-value ?? Any expert can help to answer this?
>
>
>
> Thanks and regards,
>
> Danny
>
> ______________________________________________________________________
> _
> Subscription information may be found at:
> http://www.groupstudy.com/list/CCIELab.html
This archive was generated by hypermail 2.1.4 : Sat Nov 06 2004 - 17:11:49 GMT-3