Recurring issue - Related to ETHERCHANNEL on a 7513

[rainman]

Ok, here's my situation. The portion of my network with the issue is laid out like this:

[Cisco 7513]=====================[Cisco Catalyst 4000]
Ether-Channel

*** Ether-Channel via two FA port adapters @ 100Mbit full duplex
each. My interfaces are configured as follows:


interface FastEthernet3/1/0
no ip address
no ip directed-broadcast
ip route-cache distributed
shutdown
full-duplex
channel-group 1
!
interface FastEthernet3/1/1
description 100MB FE to CAT 4000 via Etherchannel
no ip address
no ip directed-broadcast
ip route-cache distributed
full-duplex
channel-group 1
!
interface FastEthernet4/0/0
description 100MB FE to CAT 4000 via Etherchannel
no ip address
no ip directed-broadcast
ip route-cache distributed
full-duplex
channel-group 1

interface Port-channel1
description dual etherchannel to CAT 4000
ip address x.x.x.x x.x.x.x
no ip directed-broadcast
ip route-cache distributed
full-duplex
hold-queue 150 in


Now, can anybody spot any issues with the above configuration?

Etherchannel is configured as:

No. of active members in this channel: 2
Member 0 : FastEthernet3/1/1 , Full-duplex, 100Mb/s
Member 1 : FastEthernet4/0/0 , Full-duplex, 100Mb/s

Notice that I'm using fa3/1/1 & fa4/0/0 as members of the port-channel1 group. I am not using fa3/1/0 because of past issues with that adapter, and we moved to fa4/0/0 in an attempt to resolve an issue.

The issue I saw tonight was the router basically hung for about 15 minutes and when looking at the log file I see the following error:

Sep 23 18:26:33.521: %FIB-3-FIBDISABLE: Fatal error, slot 4: No window message, LC to RP IPC is non-operational

Sep 23 18:26:37.673: %SYS-3-CPUHOG: Task ran for 3400 msec (190/2), process = CEF IPC Background, PC = 602ED5D0.
-Traceback= 602ED5D8 609C1248 602D81BC 602D81A8

It seems to me that the VIP in slot 3 was the root of the error message.


Can anyone look at the above configuration and chime in with any suggestions? Do I have the interfaces configured incorrectly for port-channel and could a mis-configuration be an issue causing the router to lock-up??

I'm personally thinking the etherchannel split from one VIP to another (slot 3 to slot 4) may be related somehow.

I'm pulling my hair out with this one. Any input would be greatly appreciated!!


Thanks,

Rich

 

[marks66]

Hi,

Looks like you can use any group of fastethernet interfaces in your channel group, however Cisco recommend removing the ip route cache distributed commands from all etherchannel interfaces.

Cheers