Uplink Cisco 2960x with HP 2530 switch

marekp222 · ‎01-06-2016

Guys,

I need help with connecting Cisco 2960x with HP 2530 switch. When I connect those two switches using crossover cable i get the following error message:

Jan 6 15:35:52.232: %SPANTREE-7-RECV_1Q_NON_TRUNK: Received 802.1Q BPDU on non trunk GigabitEthernet1/0/1 VLAN1.
Jan 6 15:35:52.232: %SPANTREE-7-BLOCK_PORT_TYPE: Blocking GigabitEthernet1/0/1 on VLAN0001. Inconsistent port type.
Jan 6 15:35:53.015: %LINK-3-UPDOWN: Interface GigabitEthernet1/0/1, changed state to up
Jan 6 15:35:57.663: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet1/0/1, changed state to up
Jan 6 15:36:06.730: %SPANTREE-7-RECV_1Q_NON_TRUNK: Received 802.1Q BPDU on non trunk GigabitEthernet1/0/1 VLAN1.
Jan 6 15:36:06.730: %SPANTREE-7-BLOCK_PORT_TYPE: Blocking GigabitEthernet1/0/1 on VLAN0001. Inconsistent port type.

Even when I create a trunk on HP switch and channel-group on Cisco, I still get the same error messages.

Any idea? I highly appreciate your help,

Thanks you. Mark

casa_far · ‎01-06-2016

The error seems to be related to the spanning tree, maybe on cisco side you have the portfast enabled on port facing to hp and on the hp port you

have the stp enabled.

In such scenario hp is sending bpdu on a port where cisco is not expecting It.

The bpdu are the packets used by switches in order to make calculations in an stp domain.

My suggestion is to remove the portfast on cisco port "no spanning-tree portfast". After that when you connect the port on hp you will have to wait some seconds (30 sec +/- ) before the port goes in forwarding.

Actually there should be other tunings to do If It's still not working please post the configuration of cisco and hp.

marekp222 · ‎01-06-2016

Hi,

You are absolutely right, when I disable spanning-tree on vlan 1, the connection between two switches is fine.

Q1. What is the proper way uplinking Cisco and HP switch? Is single crossover cable enough or LACP channel between Cisco and HP is required?

Here's the config from HP switch:

Running configuration:

hostname "hatdry-copyroom01"
no telnet-server
no web-management
web-management ssl
snmp-server community "public" unrestricted
vlan 1
   name "DEFAULT_VLAN"
   untagged 1-52
   ip address 10.128.1.109 255.255.0.0
   exit

Here's the config from Cisco switch:

Building configuration...

switch 1 provision ws-c2960x-48ts-l
!
spanning-tree mode pvst
spanning-tree extend system-id
!
vlan internal allocation policy ascending
!
lldp run
!
interface FastEthernet0
no ip address
!
interface GigabitEthernet1/0/1
!
interface GigabitEthernet1/0/2
!
interface GigabitEthernet1/0/3
!
interface GigabitEthernet1/0/4
!
interface GigabitEthernet1/0/48
!
interface GigabitEthernet1/0/49
!
interface GigabitEthernet1/0/50
!
interface GigabitEthernet1/0/51
!
interface GigabitEthernet1/0/52
!
interface Vlan1
ip address 10.128.1.11 255.255.0.0
no ip route-cache
!
ip default-gateway 10.128.1.1
no ip http server
ip http secure-server
!
!
snmp-server community ***** RO
!
!
line con 0
line vty 0 4
login local
transport input ssh
line vty 5 15
login local
transport input ssh
!
ntp server 10.118.3.20
end

JDGranger · ‎01-07-2016

Mark both work, single crossover or LACP. BUT in the network world I would always recommend using more than one uplink between switches. One single point of failure between you and several devices is really rolling the dice. Also it will provide more bandwidth (not sure how many host will be using the switch) for your end devices.

marekp222 · ‎01-07-2016

JDGranger, single crossover or patch and the following configuration on the Cisco switch would be enough to create a uplink between Cisco and HP switches, right?

!
interface GigabitEthernet1/0/25
switchport mode trunk
!
Thanks for all your help

devils_advocate · ‎01-06-2016

Hi Mark

What does the config look like on the HP Side?

The word Trunk on HP is not the same as Cisco, Trunk on HP refers to PortChannel in the Cisco world.

On the HP side, you need to 'Untag' the Vlan which needs to be the Native Vlan (Vlan 1 usually) and Tag all other Vlans that you want to pass across the link.

The Cisco side is needs to be a trunk with 'switchport mode trunk' and assuming the native vlan has been left as the default (i.e vlan 1) then so do what I said on the HP side above and it should be ok.

Cheers

marekp222 · ‎01-06-2016

Hi,

HP side is very simple as shown below:

hostname "hatdry-copyroom01"
no telnet-server
no web-management
web-management ssl
snmp-server community "public" unrestricted
vlan 1
   name "DEFAULT_VLAN"
   untagged 1-52
   ip address 10.128.1.109 255.255.0.0
   exit

And I did what you have mentioned above on Cisco side as shown below:

!
interface GigabitEthernet1/0/1
switchport mode trunk
!

Now, the interface is up and I can ping that Cisco switch:

Jan 6 16:46:10.125: %LINK-3-UPDOWN: Interface GigabitEthernet1/0/1, changed state to up
Jan 6 16:46:12.135: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet1/0/1, changed state to up
Jan 6 16:46:41.146: %LINEPROTO-5-UPDOWN: Line protocol on Interface Vlan1, changed state to up

In other locations, I'm using LACP channel between Cisco and HP switches with no issues and I don't understand why it would not work this time.

Here's what I have done in the past on cisco switch:

Conf t
interface GigabitEthernet1/0/1
Channel-group 1 mode active
description Channel/Trunk to HP switch
interface GigabitEthernet1/0/2
Channel-group 1 mode active
description Channel/Trunk to HP switch

Here are steps for HP switch:

config t
trunk 1,2 trk1 lacp

I would like to configure these two switches in the same mode utilizing two ports in LACP.

Any idea what I'm missing or doing wrong?

devils_advocate · ‎01-07-2016

On the Cisco side, I would always configure the ports to be a static trunk port (switchport mode trunk) as this is best practice.

Looking at your configuration, all your ports are untagged in Vlan 1 which means the HP has a 'Trunk' but its passing all its frames untagged across it as only one vlan exists and this is the native vlan anyway which is fine.

Your config above looks fine from the Cisco side, channel-group 1 mode active will tell the Cisco to actively try and negotiate LACP with the HP. The HP looks fine in regards to the LACP config but I believe you need to add this PortChannel interface as untagged to Vlan 1 so:

vlan 1
   name "DEFAULT_VLAN"
   untagged 1-52, trk1
   ip address 10.128.1.109 255.255.0.0
   exit

Similar to the Cisco, the HP sees 'trk1' as its portchannel interface and although this is only a virtual interface, you still need to add it to the Vlan configurations.

Hope this helps.

marekp222 · ‎01-07-2016

Finally I got it to work, the trunk/channel between HP and Cisco is up and running; however, I'm not sure why it does work in different location.

Location A:

Config from HP:

interface 49
   name "Trunk Trk1 to ga-swcomm01cisco"
   no lacp
exit
interface 50
   name "Trunk Trk1 to ga-swcomm01cisco"
   no lacp
exit
trunk 49-50 Trk1 LACP
ip default-gateway 10.118.1.1
vlan 1
   name "DEFAULT_VLAN"
   untagged 1-48,Trk1
   ip address 10.118.1.100 255.255.0.0
   exit
spanning-tree Trk1 priority 4

Config from Cisco:

!
interface Port-channel1
!
interface GigabitEthernet1/0/1
description Channel/Trunk to HP switch
channel-group 1 mode active
!
interface GigabitEthernet1/0/2
description Channel/Trunk to HP switch
channel-group 1 mode active
!

Location B:

I use the same configuration as in location A and it did not work with the following results:

%LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet1/0/1, changed state to up
%SPANTREE-7-RECV_1Q_NON_TRUNK: Received 802.1Q BPDU on non trunk GigabitEthernet1/0/1 VLAN1.
%SPANTREE-7-BLOCK_PORT_TYPE: Blocking GigabitEthernet1/0/1 on VLAN0001. Inconsistent port type.

Thanks to you (devils_advocate) I did change my configs on Cisco to reflect your sugegstions about configuring ports to be static trunk port and it does work. Please, look below and let me know if I follow industry standards and hopefully not causing any problems down the road.

Config from HP:

trunk 1-2 trk1 lacp
no telnet-server
no web-management
web-management ssl
interface 1
   name "Trunk Trk1 to core switch"
   exit
interface 2
   name "Trunk Trk1 to core switch"
   exit
vlan 1
   name "DEFAULT_VLAN"
   untagged 3-52,Trk1
   ip address 10.128.1.109 255.255.0.0
   exit
spanning-tree Trk1 priority 4

Config from Cisco:

interface Port-channel1
description "Channel to HP switch"
switchport mode trunk
!
interface FastEthernet0
no ip address
!
interface GigabitEthernet1/0/1
description "Channel to HP switch"
switchport mode trunk
channel-protocol lacp
channel-group 1 mode active
!
interface GigabitEthernet1/0/2
description "Channel to HP switch"
switchport mode trunk
channel-protocol lacp
channel-group 1 mode active
!

Per your recommendation each interface that belongs to that port-channel is set to static trunk port (switchport mode trunk). Is this supposed to be set on each interface or on channel group? I did on both, channel group and each individual interface.

I'm very curious why location A works with no issues but location B did require additional configuration. Do you think different firmware/IOS would cause this?

Thank you for all your help,

Mark

german_0951@hotmail.com · ‎09-04-2019

Hi Mark, Can you help me with some similari ? I have a aruba switch 2930f and cisco switch, I need conect with fiber but a not sure that configuration do it .

Thanks