LACP Timeout mode parameter mismatch between the ESXi host and physical switches

Products

• VMware vSphere ESX 7.x
• VMware vSphere ESX 8.x

Issue/Introduction

With link aggregation control protocol (LACP) support on a vSphere Distributed Switch, user can connect ESXi hosts to physical switches by using dynamic link aggregation. 

LACP has a setting called "Timeout mode" which can either be set as "Slow" or "Fast". 

Periodic transmissions of LACP PDUs occur at either a slow or fast transmission rate, depending upon the LACP timeout preference you select.

In a fast timeout mode, PDUs are sent every second, and, in a slow timeout, PDUs are sent every 30 seconds. A slow timeout is the default setting.

Environment

VMware vSphere ESX

Cause

If the physical switch ports have the configuration set to "Fast" and the ESXi side is set as slow, then it leads to a mismatch in LACP timeout mode values. 

• Verify the current configuration of the LACPDU speed by using the below command:

esxcli network vswitch dvs vmware lacp status get -s <dvs_name>

 

In the above screenshot, the "Flags" field is set to SA where 'S' indicates that the device is sending slow LACPDUs, and 'A' indicates that the device is in active mode. 

Resolution

To rectify the LACP timeout mode configuration mismatch, the below steps can be followed: 

Method 1: 

The  "timeout mode" value can be changed from the vSphere Client.

Select vSphere Distributed Switch > Configure > LACP > Edit > Change the Timeout mode value as shown in the below screenshot: 

 

Method 2: 

For DVS versions 7.0.0 and below, the configuration can be changed using the CLI commands on the ESXi host as there is no option to change from the vSphere Client.

Steps to change the configuration:

1. Get the LACP LAG ID using the below command for each host in concern:

esxcli network vswitch dvs vmware lacp config get

2. Specify the "required" parameters namely lag ID, timeout value (1 for short timeout and 0 for long timeout) and the name of the DVS. 

Substitute the lag ID from the above output in the command below:

 esxcli network vswitch dvs vmware lacp timeout set -l <lag_id> -t 1 -s "DVS_name" -n vmnic1

3. Repeat the command for the VMNIC associated with the lag group.

From the above screenshot we can see that the timeout has now changed to "fast" mode.