Description: One or more MIB variables which provide a unique identifier for the object oid to be sampled. This is meaningful only if the ahcfConfigObjectType is 'columnar'.

The identifier can be specified in terms of another columnar oid of the same MIB table (or an augmented table or a table that has the same INDEX clause as this table) that will remain unique and constant over many restarts of the management agent or the managed system.

The identifier oids can be of any ASN.1 type. The need for an identifier oid is to counter the changes in the ordering of rows in the MIB table.

The format for this variable is ([Oid1][,Oid2][,Oid3])

Maximum of 3 identifier oids can be specified. A value of '()' is used as the default value for this variable and it represents zero identifier oids. For example, if the object being sampled is the disk space usage defined in the Host Resources MIB (RFC 1514) hrStorage.hrStorageTable.hrStorageEntry.hrStorageUsed, then the description of the storage medium (hrStorageDescr) can be used as the identifier oid. The value for this variable will be (1.3.6.1.2.1.25.2.3.1.3)

The reason for using a identifier oid is because the order of rows in hrStorageTable need not be same over time. If there are two storage media ('/' and ' /usr') and a new media is added (say '/var'), a new row may be created in between the two existing rows in the table making the contents of hrStorageTable ('/', '/var' and '/usr'). This will change the index of the mib variable hrStorageUsed for '/usr'. In order to obtain the new index value for the row for '/usr' and to collect and store the samples correctly, the agent uses the identifier oid. Without the concept of an identifier variable the agent cannot detect that the second row now represents '/var' and will wrongly store the data for '/var' in the samples for '/usr'