HOW TO ENABLE, DISABLE, AND VERIFY LACP ON AHV HOST
LACP and Link Aggregation
Advantages of LACP:-
- A single guest Virtual Machine can be utilized 20 Gbps of bandwidth in a Nutanix AHV, VMware ESXi, Microsoft Hyper-V and Citrix Xen Server node with two 10 GB adapters.
- The more adapters we will get more speed in single bond for Virtual machine and hypervisor,CVM.
- With LACP, multiple links to separate physical switches appear as a single layer-2 link.
- You must configure MLAG or vPC on the physical switch on which uplinks are connected.
Recommended Switch Configuration from vendor
It is recommended to enable LACP fallback on the switch if the switch is supported. Fallback mode is configured on a port channel interface with the port-channel LACP fallback command. The fallback timeout interval is configured with the port-channel LACP fallback
timeout command.
For Cisco Nexus Switch :
LACP port suspension
switch# configure terminal
switch(config)# interface port-channel 100
switch(config-if)# no lacp suspend-individual
Cisco Catalyst:
Switch# configure terminal
Switch(config)# interface port-channel 1
Switch(config-if)# switchport
Switch(config-if)# exit
Switch(config)# int gi1/2
Switch(config-if)# channel-group 1 mode active
Switch(config-if)# exit
Switch(config)# interface port-channel 1
Switch(config-if)# no port-channel standalone-disable
Ports of Po12 already in suspend (S) mode require a shut/no shut.
Switch(config-if)# end
Arista:
switch(config)#interface port-channel 9
switch(config-if-Po13)#port-channel lacp fallback static
switch(config-if-Po13)#port-channel lacp fallback timeout 100
switch(config-if-Po13)#show active
interface Port-Channel 9
port-channel lacp fallback static
port-channel lacp fallback timeout 100
switch(config-if-Po1
switch(config)#interface port-channel 56
switch(config-if-Po17)#port-channel lacp fallback individual
switch(config-if-Po17)#port-channel lacp fallback timeout 50
switch(config-if-Po17)#show active
Solution:
Run below Steps
- Connect to CVM (Controller VM) via putty.
- Follow steps from Checklist to verify cluster health status to verify cluster health. Do not proceed if the data resiliency is not “ok”
- Put the node and CVM in the maintenance mode:
To check host can be enter in maintenance mode enter below acli command.
nutanix@cvm1$ acli host.enter_maintenance_mode_check <host ip>
Put the host into maintenance mode – this does live migration of the VMS
nutanix@cvm1$ acli host.enter_maintenance_mode <host ip>
Enter CVM in maintenance mode using NCLI :- for step by step refer CVM maintenance article
nutanix@cvm1$ ncli host edit id=<host ID> enable-maintenance-mode=true
Note the CVM-Host id is 6 which is in double colons.
nutanix@prod-cvm1$ ncli host ls | less Id : 125763g-32e3-46s3-1404-103sagcg769g::6 ---- Note this host ID(::6). Uuid : dasjal13-9547-9732-ljfp-fk42ck109 Name : NTNX-Prod-DC IPMI Address : 50.1.1.5 Controller VM Address : 50.1.1.6 Controller VM NAT Address : Controller VM NAT PORT : Hypervisor Address : 50.1.1.7 Hypervisor Version : Nutanix AHV Host Status : NORMAL Under Maintenance Mode : false (Hypervisor upgrade) Metadata store status :
Take host access IPMI ,IDRAC,XCC depending on hardware OEM as LACP configuration can because network disconnection
Find bride name in ahv host from CVM.
Run the command from CVM for bond name
Example:
nutanix@cvm1$ manage_ovs show_uplinks Bridge br0: ----------------------⇾ bridge Name Uplink ports: br0-up -----------------⇾ Bond Name Uplink ifaces: eth3 eth2 -----------------⇾ Interface connected in bridge Bridge br1: ----------------------⇾ bridge Name Uplink ports: br1-up -----------------⇾ Bond Name Uplink ifaces: eth1 eth0 -----------------⇾ Interface connected in bridge
Configurtion of LACP in AHV hypervisor
CVM:192.168.1.1:~$ hostssh "ovs-vsctl set port br0-up other_config:lacpfallback-ab=true"
CVM:192.168.1.1:~$ hostssh "ovs-vsctl set port br0-up lacp=active"
CVM:192.168.1.1:~$ hostssh "ovs-vsctl set port br0-up bond_mode=balance-tcp"
Set bond Interval
CVM:192.168.1.x$ hostssh "ovs-vsctl set port br0-up other_config:bondrebalanceinterval=60000"
Show LACP config
CVM:192.168.1.1:~$ hostssh "ovs-appctl bond/show br0-up" CVM:192.168.1.1:~$ hostssh "ovs-appctl lacp/show br0-up"
From AHV host We can run below command to verifiy.
root@ahvhost# ovs-appctl bond/show br0-up | more root@ahvhost# ovs-appctl lacp/show br0-up | more
Example
root@ahvhost# ovs-appctl bond/show br0-up | more ---- br0-up ---- bond_mode: balance-tcp bond-hash-basis: 0 updelay: 0 ms downdelay: 0 ms next rebalance: 734 ms lacp_status: negotiated active slave mac: 0d:c4:7b:1g:6e:8h(eth2) slave eth2: enabled active slave may_enable: true hash 88: 273 kB load hash 13: 133 kB load hash 24: 44 kB load slave eth3: enabled may_enable: true hash 12: 4 kB load hash 31: 31 kB load hash 43: 15 kB load
How to disable LACP on the AHV host.
- Configure the AHV host network to work without LACP configuration by running the below command.
root@ahvhost# ovs-vsctl set port br0-up other_config:lacp-fallback-ab=true
- Use other type of load balancing active-backup or balance-slb
root@ahvhost# ovs-vsctl set port br0-up bond_mode=active-backup root@ahvhost# ovs-vsctl set port br0-up bond_mode=balance-slb
3.Remove Mlag or vpc depend on the physical switch configuration.
4 Turn off LACP on the AHV hosts.
root@ahvhost# ovs-vsctl set port br0-up lacp=off
Run the following commands.
root@ahvhost# ovs-appctl bond/show br0-up | more root@ahvhost# ovs-appctl lacp/show br0-up | more
Reference KB Nutanix
One comment