Knowing how MIMIX evaluates data group IFS entries can help you avoid a configuration that selects or omits more than you need or that has redundant or useless data group IFS entries.
When determining whether to process an IFS object, MIMIX evaluates all IFS entries for a data group to locate the entry that is the best match to the object. IFS entries are processed using the unicode character set. If the data group has identified object type *IFS for journal-centric configuration, MIMIX also looks for a match within the list of objects journaled to the user journal of the data group.
MIMIX evaluates each IFS entry and looks for a match between the object name in the journal entry and one of the data group IFS entries and the list of objects journaled to the user journal. If the data group uses journal-centric configuration for IFS objects and there is also an IFS entry that matches the object name, the IFS entry takes precedence. The IFS entry that provides the most specific match to the object name is then evaluated for object type.
If the object type in the matching IFS entry specifies *ALL, the matching IFS entry determines whether to include or exclude the journal entry in replication.
If the object type in the matching IFS entry specifies *DIR, then MIMIX checks whether the object is of type *DIR. If the object is a directory, it is selected for replication because IFS entries for *DIR object type can only be included in replication. All other object types are not selected by an IFS entry that specifies *DIR object type.
The specified name can be any one of the following:
-
A simple name, such as /A.
-
A name that is qualified with the name of the directory in which the object is located, such as /ACCOUNTS/PASTDUE.PDF.
-
A basic generic name. A basic generic name contains one or more characters followed by an asterisk (*) as its last (trailing) character, such as /ABC*. The asterisk is used as a wildcard character.
-
An advanced generic name. An advanced generic name contains one or more asterisk (*) characters embedded within the name or path that are used as wildcard characters. For example, /ABC/*.LOG or /ABC/*TMP*/*.WRK. An advanced generic name is allowed only within data group IFS entries that have *EXCLD specified for process type (PRCTYPE).
Normal evaluation. All IFS entries with simple names, directory-qualified names, and basic generic names are evaluated first. Entries that specify *DIR as the object types are allowed only within entries that include objects.
When an object has a match to more than one IFS entry, the entry with the most specific match is used. This is also true when multiple entries with basic generic names match the object. The entry whose name most specifically matches the object is used to process the object, and the object will be included or excluded according to that entry. For example, IFS entries with the name /A* and /AB* both match the name /ABACUS, but the entry used is /AB* because it is the most specific match.
Advanced generic evaluation. IFS entries with advanced generic names are evaluated only after finding a match to an entry with a basic generic name that includes the object. Advanced generic names are allowed only within entries that exclude objects.
Exclude IFS entries with advanced generic names act as filters to exclude objects that would otherwise be included by a match to a basic generic name. Any exclude IFS entries with an advanced generic name are evaluated only after finding the most specific match to an Include entry with a basic generic name. Then, a match to an advanced generic name will exclude the object.
If the evaluation process finds a match to an include IFS entry with a specific object name (no asterisks), exclude IFS entries with advanced generic names are not evaluated.
If an IFS entry with an advanced generic name is for a file system that is case sensitive, be aware that character case is not checked when the entry is evaluated for a match to an object.