Failed to bind heartbeat socket using any IP

Products

• VMware vCenter Server
• VMware vSphere ESXi

Issue/Introduction

Symptoms:

You are experiencing these issues:

• Cloning a virtual machine stops in the middle of the process and remains In Progress.
• Cloning a virtual machine fails with the error:
  
  Failed to connect to host
  
• Deploying a virtual machine from a template stops in the middle of the process and remains In Progress.
• hostd.log and vpxa.log do not show any records of virtual machine deployments to the ESX host from vCenter Server.
• vpxa.log has errors similar to:
  
  [2010-01-29 18:25:36.536 'App' 69548976 warning] Failed to bind heartbeat socket. Using any IP.
  [2010-01-29 18:25:36.536 'App' 69548976 verbose] heartbeating 172.16.30.49 [742041:97:279:48]

Environment

VMware ESX 4.0.x
VMware ESX Server 3.0.x
VMware vCenter Server 4.0.x
VMware ESX Server 3.5.x

Resolution

This issue occurs when an incorrect IP is stored in the vCenter Server database. IPs can only be removed by removing the host from vCenter Server inventory, removing the vpxa package from the ESX host and adding the ESX host back to inventory.

To correct IP issues in the vCenter Server database:

1. Verify that the vpxa.cfg file on the ESX host shows correct ESX host IP between the <hostIP></hostIP> tags.

   • /etc/vmware/vpxa.cfg in ESX 3.0.x

   • /etc/opt/vmware/vpxa/vpxa.cfg in ESX 3.5 and later
     

2. Disconnect and remove the ESX host from vCenter Server.
3. Stop the mgmt-vmware service.
4. Remove the vpxa package from the ESX host.
5. Remove user vpxuser from the ESX host.
6. Restart the xinetd service
7. Start the mgmt-vmware service
8. Add the ESX host back to vCenter Server.

Additional Information

For steps for ESX/ESXi 4.1 and ESXi 5.x, see Reinstalling the vpxa or aam agent without losing the host record from the vCenter Server database (1031919).

Reinstalling the vpxa or aam agent without losing the host record from the VMware vCenter Server database