Forum Discussion
Nimbostratusconfiguring trunk with Cisco Nexus 7000
Hello everybody,
I need help to an issue.... I’ve a network problem
about Viprion and Trunks. I configured a Trunk between a Cisco Nexus 7000 and a
BIGIP Viprion with Blade 100:
Trunk
Viprion config
Link Selection
Policy Auto
Frame distribution
hash Source/Destination MAC
address
Cisco
Nexus 7000 (Port Channel)
interface
port-channel50
description
Viprion-F5-Bil
shutdown
switchport
switchport mode
trunk
switchport trunk
allowed vlan 310,330,610,630
sh port-channel
load-balance
Port Channel
Load-Balancing Configuration:
System:
source-dest-ip-vlan
Port Channel
Load-Balancing Addresses Used Per-Protocol:
Non-IP:
source-dest-mac
IP:
source-dest-ip-vlan
interface
Ethernet1/25
description
Viprion
switchport
switchport mode
trunk
switchport trunk
allowed vlan 310,330,610,630
channel-group
50
no
shutdown
The interface
port-channel50 is in shutdown because it’s in a working mode and customer don’t
want to configuring anything, it’s a critical
environment..
So, we tested the
network traffic, the trunk on Viprion goes UP and (from Nexus to Viprion) if I
try to ping the self IP, it’s working, I can’t do the revert operation and I
can’s see the server on server farm.
Now I would to
understand if I need to setting only a LACP trunk, or it’s possible to configure
in another way?
What’s happen on the
Viprion if I configure a NO LACP Trunk? And what does happened with LACP
trunk?
Could you help me for
this?
Please tell me if you
need something else.
Thanks a
lot
Best
Regards
Fabrizio.
20 Replies
HamishCirrocumulus
OK. I might be missing something here. But you're saying you can't see the server farm... Well that's because the etherchannel interafce on the Cisco is DOWN (shutdown).
The discrepency you're seeing between the Viprion and the cisco with regards to the trunk coming up on the viprion is because you're not doing any signalling (LACP) on the trunk/etherchannel.
YOU MUST USE LACP SIGNALLING ON THE ETHERCHANNEL/TRUNK WHEN CONFIGURING A CHANNEL!!!!!
If you don't, it behaves like you're seeing, because the two ends aren't telling each other whether the channel is up/down because the up/down of the channel is only being detected by the link status (Which is up) on each of the individual connections.
I can't say it enough... USE LACP WHEN CONFIGURING CHANNELS!!!! Your problems may not go away 100% (i.e. you may have crossed cables etc), but you'll know when you do have a problem with something besides missing random packets. And the survising channel will at least work if there's at least one link up correctly.
H- The_BhattmanHi Fabriziom
I think the following configuration on the NX 7K. However, I think the key might be that you have to use tagged vlans for the 802.1q trunk to work.
interface port-channel50
description port-channel
switchport mode trunk
switchport trunk allowed vlan 310,330,610,630
spanning-tree port type edge
Configure the server ports.
interface Ethernet2/9
description VIPRION
switchport mode trunk
switchport trunk allowed vlan 310,330,610,630
channel-group 50 mode active
no shutdown
interface Ethernet2/10
description VIPRION
switchport mode trunk
switchport trunk allowed vlan 310,330,610,630
channel-group 50 mode active
no shutdown
I hope this helps
Bhattman 
Fabrizio_ChiavaHi Bhattman,
thanks a lot for your tips, now I'll try it on the customer environment. We must use the tagged VLAN because there are many VLAN to forward on Viprion.
Thanks
Best Regards
Fabrizio.
Elias_O_16228I am connecting LTM active/standby to Cisco 7k vPC in Criss-Cross connection. a) LTM1 2 ports ---> 7k1, LTM2 2 ports ----> 7k2 b) LTM1 2 ports ---> 7k2, LTM2 2 ports ----> 7k1
LTM1 trunk with these ports 1.1, 1.2, 1.3, 1.4 and LTM2 trunk 1.1, 1.2, 1.3, 1.4 Only the directly connected ports are working fine with LACP active/active on both sides but criss-cross ports are not. Connection (a) is working fine with the port channnel but connection (b) is not working. I am thinking this could be due to MAC address from the vPC switch could be the issue.
Has anyone ran into this before? Thanks for your.
Nick_T_68319If you want to criss-cross them, you need to setup a VPC on the 7K
- Hamish
What's the config on the 7k? I take it you have 2x vPC's configured? One to LTM1 and the second to LTM2?
What do the Nexus logs say? They're usually pretty verbose about why ports drop out of channels.
H
barry_kemble_50VPC a must for 7K.
- Elias_O_16228
The vPC is configured and working fine. Just the Criss-Cross connections are on suspense. When I configured passive (ltm) and active (vPC), they go on H.
a) LTM1 2 ports ---> 7k1, LTM2 2 ports ----> 7k2 port channel is working fine
- Nick_T_68319What does your port-channel config look like?
- Hamish
May favourite answer...
What do the logs say?
H
- Elias_O_16228
All,
Just to clarify, the Port Channel and associated interfaces are working with ONLY directly connected interfaces. The only port channel NOT working are the criss-cross connected interfaces. The Nexus vPC switches log show "flow control off" for the interfaces. I have enabled UDLD on the interfaces, not show. I also did UDLD reset on the port channel, not show. I also enabled flow control receive/send desire, still no show.
Why is the bring problematic. I did the same criss-cross with ASA connected to the NEXUS vPC, they are working just as intended with no problem.
- Nick_T_68319and in your port-channel config, it has like interface port-channel50 vpc 50 for example, and you have that created on both 7k 1 and 2?
