Port channel configuration

ncnaveen_arasu · ‎01-10-2012

Hi we have cisco 6500 series switch and configured port channel on both switches with 2 gig interfaces on both switches.

When we enable the port channel mode to as desirable to the interfaces on both side and applied the port channel to physical interfaces switch will go down and if we remove on any one side switch will come up. we have enabled globally the following commands,

errdisable recovery cause udld

errdisable recovery cause bpduguard

errdisable recovery cause security-violation

errdisable recovery cause dtp-flap

errdisable recovery cause link-flap

errdisable recovery cause l2ptguard

errdisable recovery cause psecure-violation

errdisable recovery cause unicast-flood

fabric buffer-reserve queue

Please sugest to make the port channel up

shine pothen · ‎01-10-2012

In order for EtherChannel to work,

1.the ports that are involved must have consistent configurations.

2.The ports must have the same VLAN

3.the same trunk mode

4.the same speed

5. the same duplex, and so on.

Most of the configuration differences within a switch are caught and reported when you create the channel. If one switch is configured for EtherChannel and the other switch is not configured for EtherChannel, the spanning tree process can shut down the channeled ports on the side that is configured for EtherChannel.

Spanning Tree Protocol (STP) on the switch where the EtherChannel is turned on thinks that there is a loop. This puts the channeling ports in the errdisabled state.

Try this out and let me know

Also post

Sh ether channel summary

Sh interface "interface type and numbe r" status

And config of etherchanel

Sent from Cisco Technical Support iPad App

ncnaveen_arasu · ‎01-11-2012

Hi thanks for the reply, i have tried ur suuggested configuration but still port channel is down

please see the required details

On Switch A

interface Port-channel21

description *** NEW6509A to NEW6509B ***

switchport

switchport trunk encapsulation dot1q

switchport trunk allowed vlan 8-13,32-37,40,41,121-127,137,225,257,308-313

switchport trunk allowed vlan add 321-325,335-337,339,340,436

switchport mode trunk

no ip address

end

interface GigabitEthernet1/2

description *** To NEW6509B-G1/3 ***

switchport

switchport trunk encapsulation dot1q

switchport trunk allowed vlan 8-13,32-37,40,41,121-127,137,138,140,225,240,257

switchport trunk allowed vlan add 308-313,321-325,335-337,339,340,436

switchport mode trunk

no ip address

channel-protocol pagp

channel-group 21 mode desirable

show etherchannel summary
Flags: D - down        P - bundled 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

        M - not in use, minimum links not met
        u - unsuitable for bundling
        w - waiting to be aggregated
Number of channel-groups in use: 1
Number of aggregators:           1

Group Port-channel Protocol Ports
------+-------------+-----------+-----------------------------------------------
21 Po21(SD) PAgP Gi1/2(s) Gi1/3(s)

Switch B

interface Port-channel21
description *** NEW6509B to NEW6509A ***
switchport
switchport trunk encapsulation dot1q
switchport trunk allowed vlan 8-13,32-37,40,41,121-127,137,225,257,308-313
switchport trunk allowed vlan add 321-325,335-337,339,340,436
switchport mode trunk
no ip address
end

interface GigabitEthernet1/2

description *** To NEW6509A-G1/2 ***

switchport

switchport trunk encapsulation dot1q

switchport trunk allowed vlan 8-13,32-37,40,41,121-127,137,138,140,225,240,257

switchport trunk allowed vlan add 308-313,321-325,335-337,339,340,436

switchport mode trunk

no ip address

channel-protocol pagp

channel-group 21 mode desirable

end

show etherchannel summary
Flags: D - down        P - bundled 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

        M - not in use, minimum links not met
        u - unsuitable for bundling
        w - waiting to be aggregated
Number of channel-groups in use: 1
Number of aggregators:           1

Group Port-channel Protocol Ports
------+-------------+-----------+----------------------------------------------
21 Po21(SD) -

HI kindly suuggest

shine pothen · ‎01-11-2012

on the switch B how many physical interface are group to portchannel 21.

Check whether the physical ports on switch A and B are in up state.

sh int giga1/2.

please let me know the result.

shine pothen · ‎01-11-2012

try to add native vlan and them try out on both ends

switchport trunk native vlan

rizwanr74 · ‎01-11-2012

FYI...

protocol pagp has much higher overhead compare to lacp. Cisco got rid of channel porotocol pagp on their newer switches.

So stick with lacp.

Are these switches are on same vtp domain ? If you answer yes, then remove this: errdisable recovery cause bpduguard

Thanks

Rizwan Rafeek

ncnaveen_arasu · ‎01-12-2012

yes its in same vtp domain, do i need to remove on both sides, wont it create STP loops

rizwanr74 · ‎01-12-2012

Yes you need to remove from the both switches.

"wont it create STP loops" Answer is no.

I have four 6500 switches running. Two switches on same VTP domain and other two are on separate domain without “errdisable recovery cause bpduguard” enabling globally.

If you fear of layer2 loops in fully mesh topology then enable “spanning-tree bpduguard enable” on the connecting interface. I believe it is enable by default.

Another due diligent on interfaces:

Spanning-tree guard loop.

Spanning-tree guard root

I hope you are running HSRP for each vlan common to both switches.

rizwanr74 · ‎01-12-2012

Thanks Peter for correction over what is going away.

Peter Paluch · ‎01-12-2012

Hi Rizwan,

protocol pagp has much higher overhead compare to lacp. Cisco got rid of channel porotocol pagp on their newer switches.

This is a completely new and so far unconfirmed information to me.

What makes you say that PAgP support is dropped? My experience is quite the contrary - the PAgP seems still to be supported on the newest IOS version for Catalysts 2960/3560, and for the VSS system and active-active detection, Cisco even introduced an improved version of PAgP called PAgP+. So they're not dropping PAgP, on the contrary, they are extending it.
What makes you say that PAgP has much higher overhead? Is there any document that deals with PAgP message sizes, their frequency, and compares them to LACP? Both PAgP and LACP are fairly quiet protocols and in stable network operation, they generate very little trafic.

Are you perhaps confusing the trunking protocols ISL and 802.1Q with PAgP and LACP? It is namely true that ISL has much bigger overhead over 802.1Q, and it is true that support for ISL is gradually declining.

Best regards,

Peter

Leo Laohoo · ‎01-11-2012

naveen kumar wrote:

Hi thanks for the reply, i have tried ur suuggested configuration but still port channel is down

please see the required details

On Switch A

switchport trunk allowed vlan 8-13,32-37,40,41,121-127,137,225,257,308-313

end

interface GigabitEthernet1/2

switchport trunk allowed vlan 8-13,32-37,40,41,121-127,137,138,140,225,240,257

------+-------------+-----------+----------------------------------------------
21 Po21(SD) -

HI kindly suuggest

This is the reason why: Your interfaces (both sides) have VLAN 138 as allowed but your PortChannel DOES NOT HAVE VLAN 138 allowed.

Rip out the ether channel (no int po 21) and in the interface, remove the "channel-group 21 mode desirable" and put it back in.