ESXi host showing as 'not responding' intermittently in vCenter.

Products

• VMware vSphere ESXi

Issue/Introduction

• The ESXi host is seen disconnecting from vCenter, but VMs remain connected. 
• Performing vmkping from the affected host, process shown here 'Testing VMkernel network connectivity with the vmkping command', to vCenter IP, results in a successful ping. 
• From the vpxa logs of the ESXi host, you may see the following, "error vpxa[2104095] [Originator sub=Heartbeat opID=xxxxx] getaddrinfo of server address failed / Temporary failure in name resolution". 

Environment

ESXi 7.0 Update 3. 

vCenter Server 7.0.3. 

Cause

The vpxa agent, responsible for sending heartbeats to vCenter, is experiencing issues in resolving vCenter DNS name. 

Resolution

• Validate if the host can ping the allocated DNS IP addresses assigned to the ESXi host. 
• Perform as well a ping (through an SSH to the host) to the domain name of vCenter, and see if this resolves to the allocated IP. 
• Check from DNS servers (external/internal), if there are any errors or issues being presented. 
• As a final step, remove and re-add the ESXi host from the vCenter inventory (ensuring you have the root credentials before proceeding).