Step-by-step EtherChannel between CatOS and IOS?

I'm having quite a bit of trouble getting an EtherChannel (just layer-2 would be fine) configured between a Catalyst 4006 (running CatOS

8.1(2)) and a 4-port Cisco switch module (running IOS 12.1(22)) that's part of an IBM BladeCenter. The Catalyst has all of its modules set to PAgP, which I would prefer to use over LACP. I am not actually sure how to check what channel protocols are set on the BladeCenter's switch, though.

On the Catalyst, I would like to use ports 4/20 and 4/21 to make a dual-gigabit EtherChannel to the Cisco. Therefore, on the Catalyst, I have this enabled:

set port channel 4/20-21 mode on

I'm rather a little bit lost when it comes to setting up the switch running IOS, though. I found this command in some Cisco docs:

channel-group 1 mode on

(applied to the relevant interfaces), but I'm not sure if this is all that I need. Could anyone please point me in the correct direction with what else I need to do to get everything running?



Reply to
Loading thread data ...

To have the switches use PAgP you need to set the mode to either auto or desirable. Try the following and see if you have better luck with it.

set port channel 4/20-21 mode desirable

channel-group 1 mode desirable

Hope this helps,


Chris wrote:

Reply to

Thanks Dan. Before I read your post I had them both set to "on", and it

*looks like* it works:

% show channel info


490 4/20 connected on 266 a-1Gb a-full 1 490 4/21 connected on 266 a-1Gb a-full 1

That's from the 4006's CatOS. I'm not sure what the equivalent command is in IOS?


Reply to

This should be reasonably straightforward. One thing that can easily happen during the configuration of port channels if the ports are up connected is that ports can be put in the errdisable state. On IOS shut then no shut the port to recover.

that is it.

when you issue these commands a new interface will be created

int port-channel 1.

I think that Cisco recommend mode desireable since the kit can recover from various possibly problematic problems in that case.

One thing that can cause problems is that if trunking is configured the two ends must agree on which VLANS are to be transported.

To get it working you could disable trunking on both ends and then turn it on afterwards.

Make sure that you have logging enabled


log buffered deb log buffered 50000

CatOS - I forget - sorry, been a while.

There are probably some debugs on the IOS end.

opswhq02#deb etherChannel ? all All debugging detail Step below all error Displaying error messages event Major events idb Agport idb related events linecard SCP messages to linecard

opswhq02#deb etherChannel

WARNING - you need to take care with debugs on production equipment due to potential CPU overhead however I would think that this one would

be quite safe. Playing with etherchannel on production equipment on the

other hand is not at all safe due to the possibility of network loops forming, fapping etherchannels triggering STP re-convergence.

Reply to

Try "show etherchannel summary" That should give you what you're looking for. Just be aware that you're not using PAgP the way you have it setup, but if you're just looking for a working etherchannel you should be good to go.

Here's sample output from the show etherchannel summary.

rmpx4#sh eth s Flags: D - down P - in port-channel I - stand-alone s - suspended H - Hot-standby (LACP only) R - Layer3 S - Layer2 U - in use f - failed to allocate aggregator

Number of channel-groups in use: 1 Number of aggregators: 1

Group Port-channel Protocol Ports


1 Po1(SU) PAgP Gi3/15(P) Gi3/16(P) Gi4/15(P) Gi4/16(P)


Chris wrote:

Reply to
dtpike Forums website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.