07-18-2006 07:50 PM - edited 03-03-2019 04:06 AM
Hello
I have a HP DL380 with dual NIC's and 2 Cisco 3560 Switches. I would like to configure the server for redundancy, with NIC1 connected to SWITCH1 and NIC2 connected to SWITCH2. The NICS are configured for NFT. The Cisco 3560's are connected together via a trunk and the ports on both switches are within the same VLAN.
There is a problem with the config as the switches lock up.
Config for both switches
interface FastEthernet0/2
switchport access vlan 10
switchport mode access
speed 100
duplex full
spanning-tree portfast
interface GigabitEthernet0/1
switchport trunk encapsulation dot1q
switchport mode trunk
macro description cisco-switch
auto qos voip trust
spanning-tree link-type point-to-point
!
interface GigabitEthernet0/2
shutdown
!
interface GigabitEthernet0/3
switchport trunk encapsulation dot1q
switchport mode trunk
macro description cisco-switch
auto qos voip trust
spanning-tree link-type point-to-point
!
interface GigabitEthernet0/4
shutdown
Any help would be much appreciated
Thanks
07-18-2006 08:23 PM
I am not saying that it's a loop but here are the thing you can do to troubleshoot.
1. When the switches are unresponsive, can you disconnect the one of the links of the teamed NIC? See if that quiets down.
2. If after disconnecting, the switch becomes responsive that would be a good indication that there was a loop. Make sure that the NICs are not blindly passing traffic from one NIC out to the other NIC. The ports have portfast enabled which puts them to forwarding immediately. The switch will not be able to control the loop if the NIC just blindly passing traffic and not participating in the STP.
Please rate post if it helps.
07-19-2006 10:00 AM
Thanks for the info, I had already tried your suggestion and yes the switches respond when only one NIC is connected. The NIC's on the server are set to NFT - with one NIC active and the other Passive (disabled), so I don't see how the server can be passing traffic over both links.
Do you have any suggestions for debugging on the switch or any other config I should be running. Do you recommend Etherchannels? If so, do you know how I would configure this on the both switches?
Many thanks
07-19-2006 12:14 PM
Etherchannel is not supported when the link terminates on two different switch. That your description, the NIC connects to two different 3560. channeling combines two port on the same switch and terminates on the other end on the same device. Something is not working right on NFT, obviously. Maybe you can try removing the portfast as it put the port in forwarding mode immediately. Have you tried removing it?
Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: