I have a pair of N7Ks which have a number of HP Virtual Connect (VC) equipped enclosures attached via LACP.
The N7Ks insist on LACP Slow timeout, and the VC on Fast timeout.
For details:
Upgrading the VC to 4.00 is not an option so we're stuck with this situation:
# sh lacp interface eth4/1
Interface Ethernet4/1 is up
Channel group is 58 port channel is Po58
PDUs sent: 3578071
PDUs rcvd: 119903
Markers sent: 0
Markers rcvd: 0
Marker response sent: 0
Marker response rcvd: 0
Unknown packets rcvd: 0
Illegal packets rcvd: 0
Lag Id: [ [(1, 44-1e-a1-78-a8-73, 2, 1, 15), (7f9b, 0-23-4-ee-be-1, 803a, 8000, 401)] ]
Operational as aggregated link since Fri Mar 21 21:57:29 2014
Local Port: Eth4/1 MAC Address= 0-24-f7-16-85-c2
System Identifier=0x8000, Port Identifier=0x8000,0x401
Operational key=32826
LACP_Activity=active
LACP_Timeout=Long Timeout (30s)
Synchronization=IN_SYNC
Collecting=true
Distributing=true
Partner information refresh timeout=Short Timeout (3s)
Actor Admin State=61
Actor Oper State=61
Neighbor: 0x15
MAC Address= 44-1e-a1-78-a8-73
System Identifier=0x1, Port Identifier=0x1,0x15
Operational key=2
LACP_Activity=active
LACP_Timeout=short Timeout (1s)
Synchronization=IN_SYNC
Collecting=true
Distributing=true
Partner Admin State=63
Partner Oper State=63
Aggregate or Individual(True=1)= 1
When attempting to perform an ISSU of the N7K (6.2.2 -> 6.2.8) it reports:
Additional info for this installation:
--------------------------------------
Service "lacp" in vdc 2: LACP: Upgrade will be disruptive as 8 switch ports and 0 fex ports are not upgrade ready!!
Issue the "show lacp issu-impact" cli for more details.
# show lacp issu-impact
For ISSU to Proceed, Check the following:
1. All port-channel member port should be in a steady state.
2. LACP rate fast should not be enabled on member ports.
The following ports are not ISSU ready
Eth4/1 , Eth4/2 , Eth4/3 , Eth4/4 , Eth4/5 ,
Eth4/6 , Eth4/7 , Eth4/8 ,
This raises two questions:
- How to workaround this issue?
- How big is the impact - in other words how "disruptive" would the ISSU be LACP wise if proceeding with the ISSU?
One workaround might be to exclude the affected ports from the LACP channel on one switch during upgrade, which would presumably force them into individual mode, include the again after ISSU, and repeat the process on the other N7K. Question is, will this work in a vPC setup, where one of the peer switches has been upgraded and the other has not, or would vPC config be locked until they are on par with NX-OS versions?
On the question of "how disruptive" will this affect only the eight ports with Fast timers, or will it affect the entire LACP process and as such the entire N7K performing ISSU?
