10-20-2010 06:42 AM
Good Morning! I am requesting some assistance regarding VPC and getting LACP enabled.
Let me explain my setup a little bit first. I have also included a diagram to explain a little better. We basically have a pair of Nexus 5k's which are connected to 2k's located in the tops of our server racks. The 5k's are then uplinked to our core VSS 6509's.
We've had a request to set up some servers across the two separate 2k's (using LACP) in our top of racks for redundancy. They want one cable in one 2k and the other cable in the opposite 2k in that rack. In this request, we had to set up a VPC link between the two 5k's and enable the keepalives to keep them aware of each other. Once we set this up, if we do a sho vpc brief, we are showing it as being up and successful. Hence the following output:
Legend:
(*) - local vPC is down, forwarding via vPC peer-link
vPC domain id : 10
Peer status : peer adjacency formed ok
vPC keep-alive status : peer is alive
Configuration consistency status : success
vPC role : primary
Number of vPCs configured : 1
Peer Gateway : Disabled
Dual-active excluded VLANs : -
It appears to us, everything is running using the vpc peer link. Now we are trying to set up the LACP VPC to the server itself. When we created it, it looks like the port channel comes up, but LACP doesnt seem like it's becoming enabled. Here are the results of the sho port summary command:
Group Port- Type Protocol Member Ports
Channel
-------------------------------------------------------------------------------
1 Po1(SU) Eth LACP Eth1/33(P) Eth1/34(P) Eth1/35(P)
Eth1/36(P) Eth1/37(P) Eth1/38(P)
Eth1/39(P) Eth1/40(P)
10 Po10(SU) Eth LACP Eth2/6(P)
100 Po100(SU) Eth NONE Eth2/1(P) Eth2/2(P) Eth2/3(P)
Eth2/4(P)
101 Po101(SU) Eth NONE Eth1/1(P) Eth1/2(P) Eth1/3(P)
Eth1/4(P)
103 Po103(SU) Eth NONE Eth1/9(P) Eth1/10(P) Eth1/11(P)
Eth1/12(P)
104 Po104(SD) Eth NONE --
105 Po105(SU) Eth NONE Eth1/17(P) Eth1/18(P) Eth1/19(P)
Eth1/20(P)
108 Po108(SU) Eth NONE Eth1/29(P) Eth1/30(P) Eth1/31(P)
Eth1/32(P)
200 Po200(SU) Eth NONE Eth105/1/10(P)
It shows 'NONE' on the protocol, which i would assume i should see LACP. What is it i'm doing wrong? I'm confused. Please, any help you can give would be much appreciated.
Thanks in advance!
10-20-2010 07:38 AM
Use command : channel-group 200 mode active
10-20-2010 09:41 AM
I actually tried that already and it caused my vpc peer link to go down/suspended. And it gave me the following output for the 'show port-channel summ' command:
Group Port- Type Protocol Member Ports
Channel
-------------------------------------------------------------------------------
1 Po1(SU) Eth LACP Eth1/33(P) Eth1/34(P) Eth1/35(P)
Eth1/36(P) Eth1/37(P) Eth1/38(P)
Eth1/39(P) Eth1/40(P)
10 Po10(SU) Eth LACP Eth2/6(P)
100 Po100(SU) Eth NONE Eth2/1(P) Eth2/2(P) Eth2/3(P)
Eth2/4(P)
101 Po101(SU) Eth NONE Eth1/1(P) Eth1/2(P) Eth1/3(P)
Eth1/4(P)
103 Po103(SU) Eth NONE Eth1/9(P) Eth1/10(P) Eth1/11(P)
Eth1/12(P)
104 Po104(SD) Eth NONE --
105 Po105(SU) Eth NONE Eth1/17(P) Eth1/18(P) Eth1/19(P)
Eth1/20(P)
108 Po108(SU) Eth NONE Eth1/29(P) Eth1/30(P) Eth1/31(P)
Eth1/32(P)
200 Po200(SU) Eth LACP Eth105/1/10(i)
10-20-2010 10:25 AM
the command needs to be applied on both N5Ks where this fex is connected to.
When you forget that, vPC will suspend the port due to inconsistency of the configuration
10-25-2010 08:27 AM
ok....I think maybe i described something wrong. My peer link stays up regardless. Unfortunately, i'm unable to get my downstream device to use the 200 port channel using LACP. I've enabled VPC and LACP features on both 5k switches. Here is what i have in my config for the downstream device (ESX Server). Keep in mind this is identical on both 5k's.
interface port-channel200
switchport mode trunk
vpc 200
switchport trunk allowed vlan 1,24
speed 1000
interface Ethernet105/1/10
switchport mode trunk
switchport trunk allowed vlan 1,24
channel-group 200 mode active
When i show the interface on the port channel 200 it displays
port-channel200 is down (No operational members)
vPC Status: Down, vPC number: 200 [packets forwarded via vPC peer-link]
Hardware: Port-Channel, address: 68ef.bd56.9e0b (bia 68ef.bd56.9e0b)
MTU 1500 bytes, BW 1000000 Kbit, DLY 10 usec,
reliability 255/255, txload 1/255, rxload 1/255
Encapsulation ARPA
Port mode is trunk
Here is my SHO PORT-CHANNEL SUMMARY command:
200 Po200(SD) Eth LACP Eth105/1/10(I)
Upon further review, i've found where my primary vpc switch shows my 200 VPC as 'successfull (sho vpc brief), but the secondary shows it as failing a consistency check.
PRIMARY
vPC domain id : 10
Peer status : peer adjacency formed ok
vPC keep-alive status : peer is alive
Configuration consistency status: success
vPC role : primary
Number of vPCs configured : 1
Peer Gateway : Disabled
Dual-active excluded VLANs : -
vPC Peer-link status
---------------------------------------------------------------------
id Port Status Active vlans
-- ---- ------ --------------------------------------------------
1 Po10 up 1,3,23-24,30,145,151-153,160,186-187,190,210-213,2
15-217,510,512,514,516,520,522,524,526,540,542,544
,546
vPC status
----------------------------------------------------------------------------
id Port Status Consistency Reason Active vlans
------ ----------- ------ ----------- -------------------------- -----------
200 Po200 down* success success
SECONDARY
vPC domain id : 10
Peer status : peer adjacency formed ok
vPC keep-alive status : peer is alive
Configuration consistency status: success
vPC role : secondary
Number of vPCs configured : 1
Peer Gateway : Disabled
Dual-active excluded VLANs : -
vPC Peer-link status
---------------------------------------------------------------------
id Port Status Active vlans
-- ---- ------ --------------------------------------------------
1 Po10 up 1,3,23-24,30,145,151-153,160,186-187,190,210-213,2
15-217,510,512,514,516,520,522,524,526,540,542,544
,546
vPC status
----------------------------------------------------------------------------
id Port Status Consistency Reason Active vlans
------ ----------- ------ ----------- -------------------------- -----------
200 Po200 down* failed Consistency Check Not -
Performed
I've also checked my vpc global consistency parameters and they all look identical.
Any help would be appreciated. Thanks again in advance.
-
10-25-2010 09:56 AM
What you describe here has changed, since the previous output. Your LACP is in I state now, that means on your server, the two nic cards
are most likely not bundled to be as one.
You mentioned ESX, are you aware that you need Nexus 1000v in order to have LACP support on ESX? ESX does not support LACP.
10-25-2010 02:28 PM
Forgive me, but my server knowledge is very limited. However, I am aware of the ESX/LACP issue. I guess ESX servers utilize some type of their own Link Aggregation type protocol. We do have some ESX servers that are currently being used in this manner, but they don't use the LACP. Apparently, There is some sort of Red Hat Linux server that they are wanting to set up in the same manner which allows the support for multiple core's for that server. I guess they have to use this Linux server in order to get the multiple core option...whereas an ESX server couldn't do that. In turn, this Linux server requires LACP....which is why i'm here posting for assistance. Hopefully this makes more sense to you than it does me...because as you can tell by my explanation....i'm pretty much in the dark when it comes to this.
Another question, I guess my int port channel200 will not show up until the NIC's in the server are bonded together? Does this sound correct?
10-28-2010 11:43 AM
The port is in I state, so on the linux side, it's there is no logical binding between the 2 NIC cards.
10-28-2010 08:43 PM
Take a look at this ESX document.
it might help
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide