Even for a pair of overlapping IP addresses that are reachable by the IP Manager, the IP Manager will not be able to monitor them because it will not be able to determine which of the systems bound to those IP addresses is responding to the ICMP or SNMP polls. If one of the overlapping IP address is reachable and the other is not, managing the reachable one and unmanaging the unreachable one would make sense.
The “Unmanage admin down” feature addresses this scenario where a duplicate IP in the network comes in picture due to network configuration change. In such a situation, one interface related to the duplicate IP may be set as “administratively down,” as it is not participating in the network. Once the interface is down, the IP related to this particular interface is not discovered, while the other IP remains as it is. The “EnableUnmanageAdminDown” in the tpmgr-param.conf file enables or disables this functionality.
The “Unmanage admin down” and the “IP tagging” features are mutually exclusive. To utilize the "Unmanage admin down" functionality, it is required that the IP tagging for that specific set of IPs is removed and then the "unmanage admin down" feature is enabled.
By default, an IP tag filter sets all IP objects created for overlapping IP addresses to “unmanaged,” and sets all IP objects created for management IP addresses to “managed.” A managed object is monitored by using ICMP and SNMP — unless different access mode is specified — to determine the object’s status and connectivity. An unmanaged object is not polled or rediscovered but remains present in the modeled topology.
The VMware Smart Assurance IP Manager User Guide provides more information about managed and unmanaged objects.