Interconnecting Cisco3550s help!

[keithja]

I'm having a problem interconnecting 2 Cisco-3550's via Fast-ethernet:

BBswitch-p20{---------------}wkgrpSw-p24

I started out with default settings except speed/duplex were set to 100/full on both units.

I have link lights at both ends. Both Ports are showing "up" line-protocol is flapping on wkgrpsw
All packets are being discarded - no packet errors are occuring.
Different ports have been tried, and wkgrpsw-p24 worked just fine when it was connected to a 3524 during configuration.

I've tried forcing both interconnecting trunks into trunking mode both with negotiate and forced to ISL.

I'm only using 1 VLAN (vlan1 the default) with no changes to default settings. I've tripple-checked speed/duplex/portfast (enabled for troubleshooting convenience) and that the ports are enabled. Currently SH INT FA0/XX SW shows both ports in static access mode.

Am I missing something really obvious? Does anyone have anything I can try?

PS. Sorry if this shows up twice. I submitted once and didn't see it so I resubmitted..

 

[RookThis]

Can you possibly paste the two configs for the ports in here so I can review it.

 

[keithja]

Certainly, Sorry for the delay, I've been off a few days...
Here it is. I removed blank lines and most n/a ports to condense it a little.
THANKs for any help you can give me. It seems it should be straight-forward but it is not working that way....
ka

Config info from WKGRPSW (port 24 is the link to bbsw):

Using 1896 out of 393216 bytes

! version 12.1
no service pad
service timestamps debug uptime
service timestamps log uptime
!
hostname Switch
!
enable secret (shhhh)
enable password (shhhh)

!
ip subnet-zero
!
cluster enable SalesRm 0

!
spanning-tree extend system-id

!
(other Ports)
!

interface FastEthernet0/24 no ip address

! interface GigabitEthernet0/1 no ip address

! interface GigabitEthernet0/2 no ip address

! interface Vlan1 ip address aa.bb.cc.dd 255.255.xxx.0

! ip http server
!
ip access-list extended CMP-NAT-ACL

dynamic Cluster-HSRP deny ip any any

dynamic Cluster-NAT permit ip any any


! (SNMP and line junk)
! end


Config from BBSwithc (P20 is the link to wkgrp Sw):

Using 4019 out of 393216 bytes
! ! Last configuration change at 16:00:18 CST Tue Jul 29 2003
! NVRAM config last updated at 16:02:38 CST Tue Jul 29 2003
! version 12.1
no service pad
service timestamps debug uptime
service timestamps log datetime

service sequence-numbers
! hostname Backbone2
! enable secret (shhh)
enable password (shhh)
! username (it's a secret)
clock timezone CST -6

ip subnet-zero !
! spanning-tree extend system-id

! (buncha ports)

!
interface
FastEthernet0/20
description Sales Switch 2

no ip address
duplex full
speed 100
!
(more ports)

! interface FastEthernet0/24 description Admin
switchport trunk encapsulation dot1q
switchport mode trunk
no ip address
duplex half
speed 10 no cdp enable
spanning-tree portfast
! interface GigabitEthernet0/1 no ip address
! interface GigabitEthernet0/2 no ip address ! interface Vlan1 ip address aa.bb.ee.ff 255.255.xxx.0
! ip classless ip http server ! !
(SNMP and VTY junk) ! end

 

[RookThis]

I didn't see po20's config in the snapshot of the config... Also, did you set up a VTP domain. I believe both switches have to be in the same VTP domain. I usually set mine to transparent mode. I think the default is in server mode. I've never clustered the switches, so I'm not sure if that's what you are trying to do as well.

 

[keithja]

Thanks for your reply, Port 20 is listed in the BBswitch listing (the second) and starts at 26 lines from the bottom. Port 24 is the pertinent interface listed for the wkgrpswitch.

My VTP domain is set to blank (unset) on both units. That's a requirement? I thought you only needed to use that if you were going to be managing VLANs from a VTP server. I'll give it a try though and see what happens. Do the 3524's not use/require VTP? The workgroup switch worked when connected to a 3524. I hate Cisco docs....sigh...

Thanks again for the help

 

[keithja]

Hi RT,

I tried setting up the VTP domains on all 3 (2 in the backbone cluster and the the workgroup switch) switches to no avail. Do you need to force trunking? I've tried that previously but that was prior to setting the VTP's.

Thx,
ka

 

[RookThis]

SWITCH 1's config
=============
interface FastEthernet0/11
description trunk to switch2 port fa0/48
switchport trunk encapsulation dot1q
switchport mode trunk
switchport nonegotiate
no ip address
end

SWITCH 2's config
=============
interface FastEthernet0/48
description trunk to switch1 port fa0/11
switchport trunk encapsulation dot1q
switchport mode trunk
switchport nonegotiate
no ip address
end


This is how I have the trunks configured. You are using a reversing cable between the two switches, right?

 

[keithja]

Hi RT,

Thanks for the info.
Initially I had let my ports negociate those settings but subsequently had forced them to those settings as well.

What I have discovered thus far is that when I moved the bbswitch end to a different port, it came up almost ok even though port 20 doesn't show any problems in the post. Even then I was having wierd problems. My test host connected to salesw3 (it had been connected to a different part of the network and worked fine) could only ping hosts connected to bb2 it could not ping anything connected to the other half of our backbone switch: bb1. We ultimately found out that other hosts did not have this problem so we're suspecting a NIC incompatability issue (although thats mighty strange to me) (its an integral 3com2905). At this point we're going back to try to figure out why port 20 doesn't work while port 15 does even though they're configured the same... I'm a little disappointed to be having this many problems from what I thought was going to be largely plug and play operation in a flat topology....

sigh..