Steps for solving :HA agent has an error : cmd addnode failed for primary node: Internal AAM Error - agent could not start. : Unknown HA error error

1. Create a new cluster.

2. Disconnect the hosts from existing cluster add to the newly created cluster. ( 1st try to add the host which is having ha issue, if it works add other hosts)

3. Add all the host in new cluster, if Ha has configure properly in all hosts,delete the old cluster and rename the new with old cluster name. ==>  still problem exists proceeded further.

4. Put the host ( in which HA agent is not configured ) in maintenance  mode. ( VMs should be  moved to other host els need to face vm reboot)

5. Login to the host using ssh  ( enabe TSM in esxi)

6. Go to /etc/hosts and add the entry of all the host in the cluster.

7. Run following commands to remove aam agent ( ESXi).

                # services.sh stop

                # /opt/vmware/aam/VMware-aam-ha-uninstall.sh

                # rm -rf /opt/vmware/

                # services.sh start.

8.  Removed  - configured scratch partition

                Connect to vCenter Server or the ESXi host using the vSphere Client.

                Select the ESXi host in the inventory.

                Click the Configuration tab.

                Click Storage.

                Right-click a datastore and select Browse.

                Create a uniquely-named directory for this ESX host  ( Eg: localscrach)

                Click Advanced Settings under Software. 

                Select the ScratchConfig section.

                Change the cratchConfig.ConfiguredScratchLocation configuration option, specifying the full      path to the directory.            Eg. /vmfs/volumes/datastore1/localscrach

8. Reboot the host.

9. Add the host to the cluster ,exit maintenance mode.

10. HA will be reconfigured with out error.