Setting Up Batch Job File Viewing

In this section:

• Setting Up Directories for PSUB
  
• Entering Reference Codelist Values
  

Setting Up Directories for PSUB

Oracle Clinical's Parameterized Submission (PSUB) batch utility runs and schedules most Oracle Clinical jobs.

For more information , see:

• Creating a Directory for Log and Output Files and Registering it in OCL_STATE
  
• Creating Directories for Input and Output Files of Certain Job Types
  

Creating a Directory for Log and Output Files and Registering it in OCL_STATE

The logs and output of the PSUB job are stored in the database and retrieved and displayed directly from a database table. Beginning in Release 5.0, FTP is no longer used.

You must create a single directory to temporarily store .log and .out files generated by PSUB before they are stored in the database, and enter the path of this directory in the PSUB_LOGS_DIR entry of the OCL_STATE local reference codelist.

The system automatically creates a subdirectory for each user the first time the user runs a PSUB job, and stores the .log and .out files in the subdirectory.

By default, the system deletes these files from the temporary directory after writing them to the database. If you prefer to keep files in the temporary directory, you can set the PSUB_DEL_FILES entry in OCL_STATE to N.

Creating Directories for Input and Output Files of Certain Job Types

Some PSUB jobs require input files and/or produce output files for which they require an additional directory.

For each job type listed in Table 11-1 below, do the following:

1. Create a root directory in the operating system file system for the database for each job type.

2. (Optional but recommended) Create a user-specific directory for each user who needs to submit jobs of that type. Oracle recommends having user-specific directories but provides the option in case you have very few users loading files and they have the same security privileges. Not creating user-specific directories may create a security issue, especially if you have multiple studies on the same database. A user could select and view a file from a study other than the one in context in the Oracle Clinical interface.

   You must name the subdirectory exactly the same as the username, minus the OPS$ string, if any. The system disregards the OPS$ string when determining whether to give the user access to files in the subdirectory.

3. Give access to each directory only to the opapps account. Individual users do not need access and should not have access.

Table 11-1 PSUB Job Types Requiring Special Directories and the Corresponding OCL_STATE Settings

Job Type	Oracle Clinical Menu Path	OCL_STATE Settings
Randomization	From the Design menu, select Randomization, then Randomization Batch Load, then one of the following: Download Randomization Validate and Load Randomization Load a Converted Randomization Validate a loaded Randomization	RAND_ROOT RAND_USERDIRS
Load Patient Enrollment Dates	From the Design menu, select Patient Positions, then Load Enrollment Dates.	BDL_ROOT BDL_USERDIRS
Batch Data Load and Out of Study Load	From the Conduct menu, select Batch Data Load or Out of Study Load, then select Batch Load Data Files.	BDL_ROOT BDL_USERDIRS
Load Patient SDV Attributes	From Design select Patient Positions, and then select Load Patient SDV Attributes	BDL_ROOT BDL_USERDIRS
Lab Batch Data Load	From the Labs menu, select Lab Batch Data Load, then one of the following: Lab Ranges Batch Data Load Report Results of the Data Load	LAB_ROOT LAB_USERDIRS
Disconnected Replication	From the Admin menu, select Replication, then one of the following: Disc Repl Export Disc Repl Load	REPL_ROOT

Note:

Disconnected Replication does not require or support user-specific subdirectories.

Entering Reference Codelist Values

For each job type listed in Table 11-1, do the following:

• Enter the Root Directory Path in OCL_STATE Local Reference Codelist
  
• Enter Y or N in the USERDIRS OCL_STATE Setting
  

Enter the Root Directory Path in OCL_STATE Local Reference Codelist

Enter the directory's path in the matching setting (JOBTYPE_ROOT) of OCL_STATE.

Enter Y or N in the USERDIRS OCL_STATE Setting

You can choose to use a subdirectory for each user or not for input and output files for each job type. Each job type (except disconnected replication) has an OCL_STATE setting named JOBTYPE_USERDIRS. Enter Y or N.

• If Y, the job looks for the input or output file specified by the user in any directory at or below JOBTYPE_ROOT/user, where user must match the database account ID of the Oracle Clinical user who submitted the job.

  If the account ID of the user who submitted the job begins with OPS$, the subdirectory name must match only the user (non-OPS$) portion of the account name.

  Output files, if any, cannot be written unless the user-specific subdirectory has been created by an administrator.

• If N, ), the job looks for the input or output file in any directory at or below JOBTYPE_ROOT.

Note:

Other reference codelist settings control other aspects of PSUB functionality; see Required Reference Codelist Settings for Batch Jobs.