Recovery from a Failover Caused by a Node Failure

If a failover occurs due to an ESXi host failure, the virtual hard disk cannot be disconnected. Therefore, the virtual hard disk remains connected to the stopped guest. For this reason, multiple guests try to access the virtual hard disk when returning, but the operation is restricted by the hypervisor and therefore the guest cannot be started. In this case, use the vSphere client to manually disconnect the virtual hard disk from the guest.

Changing ESXi Login Information

To change the username and password of the ESXi host perform the following steps:

  1. Stop LifeKeeper or all VMDK resources.
  1. Execute the following command from the command line:
    # /opt/LifeKeeper/lkadm/subsys/scsi/vmdk/bin/esxi_register -u <ESXi host name>
    Enter a new username and password interactively. Once you log in successfully the information will be updated. If the VMDK resource is running or you cannot log in an error occurs and the information is not updated.
  1. Start the stopped LifeKeeper or VMDK resources.
  1. Repeat the same steps for all nodes and update the login information.

Deleting ESXi Host Information

To delete registered ESXi host information perform the following steps:

  1. Stop LifeKeeper or all the VMDK resources.
  1. Execute the following command from the command line:
    # /opt/LifeKeeper/lkadm/subsys/scsi/vmdk/bin/esxi_register -d <ESXi host name>
  1. Start the stopped LifeKeeper or VMDK resources.
  1. Repeat the same steps for all nodes and update the login information.

Esxi_register Details

Registering a host esxi_register -a <ESXi host name>
Deleting a host esxi_register -d <ESXi host name>
Updating login information esxi_register -u <ESXi host name>
A list of registered hosts esxi_register -l

Feedback

Thanks for your feedback.

Post your comment on this topic.

Post Comment