Discovery filter considerations

Discovery filters are inclusive-type filters, which mean that if you do not create any filters, no candidate systems are accepted. A system is accepted only if it matches a filter.

Consider the following points before creating discovery filters:

• When creating a filter, you must specify a value for each of its five core fields. The procedure for creating a discovery filter is given in the IP Manager User Guide.

• You can create multiple filters and arrange them in a high-to-low priority. A candidate system is compared against the highest priority filter first.

• A candidate system is subjected to two levels of filter matching, as described in “Discovery filter matching” on page 181.

• When a candidate system matches all five core fields of a filter and passes any additional autodiscovery safeguards, the system bypasses any lower priority discovery filters and becomes an object in the topology.

• To exclude a specific candidate system, specify the system in the ipExcludeList in the discovery.conf file.

• Seed systems, as listed in a seed file or entered through the Add Agent command, are not subject to the discovery filters.

  For systems that match a discovery filter and are added to the topology, they too are probed for the IP addresses of their neighboring systems. The autodiscovery cycle continues until no more new IP addresses match the discovery filters.