AccuRev User Identity of the Server Process

In addition to its user identity at the operating system level, the AccuRev Server process sometimes needs an AccuRev username (principal-name) identity:

• When it executes a server-side trigger script that invokes AccuRev client commands, such as annotate or promote.

• When it performs a synchronization with the master repository — explicit or implicit replica sync command. This applies only if the AccuRev Server is managing a replica repository.

If either of these situations applies to the AccuRev Server that you are administering, you must take steps to establish a valid AccuRev username for the AccuRev Server. The AccuRev username need not be special or reserved. Just make sure that any security controls — ACL permissions and/or server_admin_trig script — are configured to allow that particular AccuRev username to perform the required operations. See AccuRev Security Overview.

With the "AccuRev login" user-authentication scheme, a session file establishes the AccuRev user identity of a process. Create a long-lived session file for the AccuRev Server's operating-system identity as follows:

1. Access the command line.

2. Set environment variable ACCUREV_HOME to the home directory of the operating system user that the AccuRev Server runs as (for example, acserver).

   If the AccuRev Server is running as the local System account on a Windows machine, the home directory is C:\.

3. Create a long-lived session file for the AccuRev username that the AccuRev Server will use:

   accurev login -n john
   Password: ********

   Note: If you are creating a session file on a replica server machine, to be used for communicating with the AccuRev Server process on the master server machine, direct the login command to the master server machine. For example:

   accurev login -n -H bingo_master:5050 john
   Password: ********

This session file will be valid indefinitely, thanks to the -n option.