Table of ContentsView in Frames

How clustered Data ONTAP enables dynamic home directories

Clustered Data ONTAP home directories enable you to configure an SMB share that maps to different directories based on the user that connects to it and a set of variables. Instead of having to create separate shares for each user, you can configure a single share with a few home directory parameters to define a user's relationship between an entry point (the share) and their home directory (a directory on the Storage Virtual Machine (SVM)).

There are four variables that determine how a user is mapped to a directory:

Share name
This is the name of the share that you create that the user connects to. You must set the home directory property for this share.

The share name can use the following dynamic names:

  • %w (the user's Windows user name)
  • %d (the user's Windows domain name)
  • %u (the user's mapped UNIX user name)

To ensure that the share name is unique across all home directories, the share name must contain either the %w or the %u variable. The share name can contain both the %d and the %w variable (for example, %d/%w) or can contain both a static portion and a variable portion to the share name (for example, home_%w).

Share path
This is the relative path, defined by the share and therefore associated with one of the share names, that is appended to each search path to generate the user's entire home directory path from the root of the SVM. It can be static (for example, home), dynamic (for example, %w), or a combination of the two (for example, eng/%w).
Search paths
This is the set of absolute paths from the root of the SVM that you specify that directs the clustered Data ONTAP search for home directories. You specify one or more search paths by using the vserver cifs home-directory search-path add command. If you specify multiple search paths, clustered Data ONTAP tries them in the order specified until it finds a valid path.
Directory
This is the user's home directory that you create for the user. It is usually the user's name. You must create it in one of the directories defined by the search paths.

As an example, consider the following setup:

Scenario 1: The user connects to \\vs1\home_jsmith. This matches the first home directory share name and generates the relative path jsmith. Clustered Data ONTAP now searches for a directory named jsmith by checking each search path in order:

Scenario 2: The user connects to \\vs1\jsmith. This matches the second home directory share name and generates the relative path acme/jsmith. Clustered Data ONTAP now searches for a directory named acme/jsmith by checking each search path in order: