| Contents: | Main | Chapter | See Also: | Getting Started Manual | Programmer Manual | |||
In addition to auditing changes to data values, you can audit changes to data dictionaries.
Start an audit on a data dictionary by using the Edit File option on the Utility Functions submenu. You should answer YES to the "DD AUDIT?" prompt.
When you start an audit, VA FileMan begins an audit trail of changes made to the data dictionary. Changes to the definitions of fields in the file are audited. Fields in Subfiles are audited, too. Changes made to the definition of the file, using the Edit File option, are not audited. The changes are recorded in the DD AUDIT file (#.6).
An example of starting a data dictionary audit follows:
Select OPTION: UTILITY FUNCTIONS Select UTILITY OPTION: EDIT FILE MODIFY WHAT FILE: PATIENT NAME: PATIENT// <RET> DESCRIPTION: 1> <RET> Select APPLICATION GROUP: <RET> DEVELOPER: <RET> VERSION: <RET> DATA DICTIONARY ACCESS: <RET> READ ACCESS: <RET> WRITE ACCESS: <RET> DELETE ACCESS: <RET> LAYGO ACCESS: <RET> AUDIT ACCESS: <RET> DD AUDIT? NO// YES ASK 'OK' WHEN LOOKING UP AN ENTRY? NO// ^
To see what changes were made to the data dictionary, use the Inquire to File Entries or Print File Entries option and identify the DD AUDIT file (#.6) as the file of choice. The Data Dictionaries Being Audited option is just used to display a list of data dictionaries being audited; it doesn't include the changes made to audited data dictionaries. The following dialogue is an example of how to identify the changes made to a data dictionary:
Select OPTION: INQUIRE TO FILE ENTRIES OUTPUT FROM WHAT FILE: .6 <RET> DD AUDIT AUDIT FROM WHAT FILE: PATIENT Select PATIENT SUB-FILE: <RET>
NOTE: You only see the "SUB-FILE" prompt if the file contains a Subfile. To display audit information for the Subfile, specify it here.
Select PATIENT DD AUDIT: ?
ANSWER WITH ZZPT DD AUDIT NUMBER, OR FIELD NUMBER, OR DATE
UPDATED, OR USER
CHOOSE FROM:
1 2 02-20-90 PROGRAMMER,SYSTEMS
2 3 02-20-90 PROGRAMMER,SYSTEMS
The entries in the DD AUDIT file are identified by the field number (2 and 3 in this example), the date of the change (02/20/90 for both entries), and the person making the change (PROGRAMMER, SYSTEMS for both).
Select PATIENT DD AUDIT: 1 ANOTHER ONE: 2 ANOTHER ONE: <RET> STANDARD CAPTIONED OUTPUT? YES// <RET> DISPLAY COMPUTED FIELDS? NO// <RET> NUMBER: 1 FIELD NUMBER: 2 TYPE: EDIT DATE UPDATED: FEB 20, 1990@17:54:36 USER: PROGRAMMER,SYSTEMS ATTRIBUTE NAME: LABEL ATTRIBUTE NUMBER: .01 FILE NUMBER: 999000 OLD VALUE(S): DATE OF BIRTH NEW VALUE(S): DOB NUMBER: 2 FIELD NUMBER: 3 TYPE: EDIT DATE UPDATED: FEB 21, 1990@11:54:03 USER: PROGRAMMER,SYSTEMS ATTRIBUTE NAME: LABEL ATTRIBUTE NUMBER: .01 FILE NUMBER: 999000 OLD VALUE(S): CURRENT AGE NEW VALUE(S): AGE
This example indicates that a user named Systems Programmer modified the PATIENT file (#999000) on 2/20 and 2/21/90. This person edited the LABEL (.01 attribute) of Fields #2 and #3. The LABEL of Field #2 was changed from DATE OF BIRTH to DOB and the LABEL of Field #3 was changed from CURRENT AGE to AGE. The number of the first change as it was recorded in the DD AUDIT file (#.6) is 1 and the second is 2.
For information on which data dictionaries are being audited, use the Data Dictionaries Being Audited option from the Auditing submenu. Select a range of data dictionaries. VA FileMan displays the file number and file name of the data dictionaries currently being audited, as shown in the example below:
Select OTHER OPTION: AUDITING Select AUDIT OPTION: DATA DICTIONARIES BEING AUDITED START WITH NUMBER: FIRST// 2 GO TO NUMBER: LAST// 34 DEVICE: HOME// <RET> DATA DICTIONARIES BEING AUDITED AUG 30, 1989 18:08 PAGE 1 FILE NAME ------------------------------------------------------------------ 3 USER 9.4 PACKAGE 30 LOCATION
Use the Purge DD Audits option to erase all audit trails used in auditing data dictionaries (including Subfiles, which have their own data dictionaries) for a specified file. Purging is not an automatic feature, it must be done manually.
You should either 1) turn auditing off on the files you're purging while you're doing the purge, or 2) leave auditing on but purge the file when not many users are on the system. If you purge when auditing is on and people are using the file in question, it is possible that you might end up with incomplete audit records on the audited file.
The following dialogue results in purging selected data dictionary audit records for the user Ringleader:
Select AUDIT OPTION: PURGE DD AUDITS AUDIT FROM WHAT FILE: PATIENT Select PATIENT SUB-FILE: <RET> DO YOU WANT TO PURGE ALL DD AUDIT RECORDS? NO// NO PURGE DD AUDIT RECORDS BY: FIELD NUMBER// USER START WITH USER: FIRST// RINGLEADER GO TO USER: LAST// RINGLEADER WITHIN USER, PURGE AUDITS RECORDS BY: <RET> DEVICE: <RET> HOLD ON, PLEASE... PURGE OF DD AUDIT: PATIENT FILE FEB 21, 1990 14:45 PAGE 1 ------------------------------------------------------------------ 9 RECORDS PURGED.
The following dialogue results in purging all data dictionary audit records for the PATIENT file:
AUDIT FROM WHAT FILE: PATIENT Select PATIENT SUB-FILE: <RET> DO YOU WANT TO PURGE ALL DD AUDIT RECORDS? NO// YES ARE YOU SURE? NO// YES DELETED
Reviewed/Updated: March 4, 2007