Track the delete operation.

Guys,
I've an Oracle table from which records get deleted very often.
I would like to track the user(and the system name) who issued the delete operation,timestamp of the operation and the no. of records deleted from the user's transaction.Can someone help me with a procedure for this task?
Thanks,
Sevuga.

Thanks Jameel.
I've got it working now.
Let me share the steps that I've followed to set the auditing 'ON' on the table concerned.
CONNECT system/manager AS SYSDBA
-- Truncate the database audit table.
-- The database audit table was truncated so any old audited information will be deleted for
TRUNCATE TABLE aud$
-- View the AUDIT_TRAIL parameter value.
SHOW PARAMETER audit_trail
-- The NONE value indicates that the audit process was not activated.
-- Change the AUDIT_TRAIL parameter to db.
ALTER SYSTEM SET audit_trail=db SCOPE=spfile
-- Shutdown and startup the database. Use the default Server Parameter File (SPFILE).
-- Notice that when the AUDIT_TRAIL option is set to db, our audit trail information is stored in the AUD$ table owned by sys.
-- And also, when AUDIT_TRAIL is set to Operating System,ouraudit trail information will be stored in the director named by the AUDIT_FILE_DEST parameter.
-- In our case, we are storing data into the AUD$ table in the Oracle database.
SHUTDOWN IMMEDIATE
CONNECT system/manager AS SYSDBA
STARTUP
-- View the AUDIT_TRAIL value again.
SHOW PARAMETER audit_trail
-- Notice that this time, the value would be changed to db.
-- Query the AUD$ table.
SELECT * FROM aud$
-- Notice that there are no records in the audit table.
-- Audit who ever deleted a record from the EMP table.
AUDIT DELETE
ON emp
BY ACCESS
WHENEVER SUCCESSFUL
-- Connect as the normal user.
CONNECT scott/tiger
DELETE FROM emp
WHERE empno = 9999
COMMIT
-- Connect again as the SYSTEM/MANAGER user.
CONNECT system/manager AS SYSDBA
-- View the SYS.AUD$ table to display the date and time who (username), and what (statement of an action) was done.
SELECT
TO_CHAR(timestamp#,'DD-MON-YYYY HH24:MI:SS') as "Date and Time",
userid, name "Action by user"
FROM sys.aud$ JOIN sys.audit_actions
ON action = action#
Thanks,
Sevuga

Similar Messages

The deleted operation affects the confirmed yield of the parallel sequence

there is a production order.
the control key of the last operation is PP03,and it's confirmation is : Milestone confirmation.
and the production order has a parallel sequence,a external processing operation.
but i marked the external processing operation deleted.
then i confirm the last operation,and it's sysstatus is CNF
but the parallel sequence operation is not CNF.
WHY?

Dear,
did you delete the operatrion after release or before the release?

Where will be the deleted and Added roles will be tracked

HI All,
 Need is i must track the Deleted and Added Transactions to the Role and
 Deleted and Added Roles to the Composite Roles.
 In the table AGR_HIER the records are inserting and deleting when we add change the Roles and Composite roles.
 There is no history maintained in that regarding newly inserted items(transaction to the role and Roles to the composite role) and deleted items(transaction to the role and Roles to the composite role).
 Please proved the needful help.
Thanks,
Ravi.

Hi,
table entries in USR* and UST* tables
Regards

Table in which deleted operations are stored.

Hi,
We are writing a report and for that i need the value of work hrs , so i would like to know in which table deleted operations of Maintenance order are stored ....
regrds
pm

Hi,
All the operations time values are found in AFVV, this includes the deleted operations. Also read AFVC for operation number, etc. The key to both these tables is AUFPL field (available in AFKO via order number).
To decide if a particular AFVV record is deleted or not it is necessary to look at the JEST table. JEST can be accessed via Object Number, this has the format OV + AUFPL + APLZL. If the entry with status I0013 is active (no inactive flag) then this operation is deleted.
-Paul

Delete Operation in JDBC Sender Adapter not works

Hi,
 I have one student table which contains the fields ID,Name,BirthMonth,BirthYear,ReadFlag fields. ReadFlag is a character field or lenght 1 which contains only values either 'Y' or ' '.
 I want to execute delete operation in this table ie to delete the records which contains the readflag = 'Y'. So, I set the below values for the following parameters.
DELETE FROM student WHERE READFLAG = ' Y '
Query SQL Statement : SELECT * FROM student WHERE readflag = 'Y'
Update SQL Statement : DELETE FROM student WHERE READFLAG = 'Y'
Poll Interval : 60 Seconds.
 There are more records in this table which contains readflag = 'Y'. But, the Adapter does not delete those records from the table i.e delete operation is not executed. At the same time, in Comm. Channel monitoring it does not show any error, but the delete operation is not carried out in the table.
 I tried after 'COMMIT' the table also. But it does not work. What could be the reason ? or How to use Delete Operation effectively on the table ?
 Kindly help me friends to solve this problem.
Thanking you.
Kind Regards,
Jeg.

http://help.sap.com/saphelp_nw04/helpdata/en/7e/5df96381ec72468a00815dd80f8b63/content.htm >>>
Adapter Work Method
You must add an indicator that specifies the processing status of each data record in the adapter (data record processed/data record not processed) to the database table.
The UPDATE statement must alter exactly those data records that have been selected by the SELECT statement. You can ensure this is the case by using an identical WHERE clause. (See Processing Parameters, SQL Statement for Query, and SQL Statement for Update below).
Processing can only be performed correctly when the isolation level for transaction is set to repeatable_read or serializable.
Example
SQL statement for query: SELECT * FROM table WHERE processed = 0;
SQL statement for update: UPDATE table SET processed = 1 WHERE processed = 0;
processed is the indicator in the database.
try with repeatable_read or serializable !!!
Also go thru this thread - DELETE Querey in JDBC SENDER

ADF view : auto commit delete operation

I have created a ADF table with add and delete operation. When I perform delete operation it is not committing automatically.
Can you please suggest the way to auto commit the delete operation.
Thanks,
Kiran

User,
please always tell us your jdev version as the solution might depend on it.
There is no auto commit in ADF. You can program it so that after an add or delete the data gets committed but there is nothing done automatically.
if you use a bounded task flow you can drag the operation from the data control onto the task flow and navigate to it after you have done the add or delete operation.
Timo

Database Auditing to record DELETE operation on a schema for all tables.

Hi,
I am using ORACLE DATABASE 11g. I want to apply the AUDIT feature to record all the DELETE operations happening on the schema tables.
I did the following steps but dint got the proper output :-
I logged into the SYS as sysdba user and set
alter system set audit_trail=DB,EXTENDED scope=spfile;then i executed this command to record the sql which will use the DELETE privileges
AUDIT DELETE ANY TABLE;Then i bounced back my DB and for testing purpose i created a table in SCOTT schema and inserted 10 rows in it and then DELETE all the rows from it.
As per expectation i check the view
select * from aud$
where spare1 like '%MACHINE1%'
and USERID='SCOTT'
order by ntimestamp#;The output i got is :-
34 168368 1 1 SCOTT I-DOMAIN\MACHINE1 MACHINE1 100 0 Authenticated by: DATABASE; Client address: (ADDRESS=(PROTOCOL=tcp)(HOST=127.0.0.1)(PORT=2565)) MACHINE1 5 21-DEC-11 07.02.58.621000 AM 0 928:5024 0000000000000000 983697018 <CLOB> <CLOB> But here i don't see the SQL generated in the last column.
What i was expecting is that if i fire a DELETE statement in the schema it will get logged here and with the help of this view i will be able to see that which user from which machine executed a DELETE statement and what that statement was?
Please let me know what step i have missed here.
PS:- The ACTION# column shows 100 , is it the code for DELETE action. I also accessed the DBA_AUDIT_TRAIL view but din't found any usefull info their.
Thanks in advance.

Try instead:
audit delete table;AUDIT DELETE ANY TABLE is auditing use of DELETE ANY TABLE privilege.

HCM Processes & Forms: Change & Delete Operations

Hi,
I've been working on Processes & Forms.
One of my processes is a typical address change. In this process, it's possible for the user to change any of his/her addresses. This means that the process is customized to accept subtype * of infotype 0006. This is working fine, we can change and create addresses.
The problem is that we want to delete addresses too, but P&F doesn't allow two operations for the same infotype. We also tried to follow note 1043692, using the DELETE_DATASET reserved field, but it's not possible to use rules on SAP_PA using fields that are not defined for this back-end service.
Any help on this?
Regards,
Fábio.

Sorry for the delay,
I've never configured a process for Absences, but since you're talking about this infotype, I think you're trying to delete the record completely. However, in P&F the "delete operation" means that a record is deleted as of the effective date (i.e. delimited).
Example: I'm deleting a temporary address: the record will still be there, but the new end date will be the effective date that I chose.
So if you're expecting the absence to disappear, it will not happen unless you set the effective date to the begin date of that record.
Regards,
Fabio.

Help in Delete operation

I have a problem when deleting rows from a table. When I created the database the following delete statement worked fine, but now it takes a lot to complete. To delete 600 rows it takes more than 10 minutes. Maybe the transaction log is full, but i don't know how to check the space assigned to it or where is it.
The delete operation is the following one.
DELETE FROM event_TTT WHERE event_TTT.event_id IN
(SELECT event.event_id FROM event WHERE event_TTT.event_id=event.event_id and event.MM_id=1328)

Thank you for your information about statistics. After reading some papers and after checking that automatic statistics are being gathered in my system, I think that my tables do not fulfill the requirements to gather statistics manually ("statistics on tables which are significantly modified during the day"). Maybe I'm wrong. I'll continue reading about this.
Anyway, I've new data about my delete operation. I think it is very slow, but I'll give you the details:
DELETE FROM event_track e
WHERE EXISTS
SELECT
NULL
FROM event e1
WHERE e1.event_id = e.event_id
AND e1.pdep_mission_id = 1328
AND ROWNUM = 1
The is an index on e1.pdep_mission_id.
e1.event_id and e.event_id are primary keys.
Number of rows on e = 117000.
Number of rows on e1 = 120000
Number or rows to delete = 3500
This is the execution plan (with autotrace):
Plan hash value: 660928614
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
| 0 | DELETE STATEMENT | | 1 | 254 | 217K (1)| 00:31:20 |
| 1 | DELETE | EVENT_TRACK | | | | |
|* 2 | FILTER | | | | | |
| 3 | TABLE ACCESS FULL | EVENT_TRACK | 119K| 28M| 667 (3)| 00:00:06 |
|* 4 | COUNT STOPKEY | | | | | |
|* 5 | TABLE ACCESS BY INDEX ROWID| EVENT | 1 | 9 | 2 (0)| 00:00:01 |
|* 6 | INDEX UNIQUE SCAN | PK_EVENT271 | 1 | | 1 (0)| 00:00:01 |
Predicate Information (identified by operation id):
2 - filter( EXISTS (SELECT 0 FROM "PDEP"."EVENT" "E1" WHERE ROWNUM=1 AND
"E1"."EVENT_ID"=:B1 AND "E1"."PDEP_MISSION_ID"=1305))
4 - filter(ROWNUM=1)
5 - filter("E1"."PDEP_MISSION_ID"=1305)
6 - access("E1"."EVENT_ID"=:B1)
Estadísticas
11207 recursive calls
26273 db block gets
24517966 consistent gets
23721872 physical reads
4316076 redo size
932 bytes sent via SQL*Net to client
1105 bytes received via SQL*Net from client
6 SQL*Net roundtrips to/from client
12 sorts (memory)
0 sorts (disk)
3518 rows processed
This operation took 2 hours and 5 minutes to complete the deletion of 3500 rows. Do I have to assume that this is correct or maybe I'm doing something wrong ??
Thank you in advance.

Delete operation not commited

Hello,
I have a Delete operation which is never commited. When I click on delete button, the view object's row is deleted. But when I click on Commit button (which is the Commit operation of the Root Application Module), the delete operation is not commited. Debugging, I see that doDML method is not called.
Any idea on what happens?

Yes the View is editable.
I can't create a fresh page to test this, but I can test directly debugging the application module. And I have the same problem. If I update or create a row, I can commit. But if I delete a row, the Commit button is not enabled.
With "Commit Button from Application Module Data Controls > Operations > Commit/Rollback", did you mean to execute Commit operation from bindings? If yes, it's same problem.
Maybe it is because of some hacks I have done in VOImpl classes. I will see if I can test this without my hacks.

Delete operation on indexed table taking lot of time

We are working on 10g R2. We have a script for archiving a table which upon copying into table (say Y a non indexed table); the pre-decided records from a table (say X indexed table).
After inserting the records into Y the records are deleted from X. There are close to 50 million records to be archived this way.
While testing the script in development instance for a million records, we find that on avoiding the delete operation the time taken for the script to execute is ~ 45 minutes which otherwise (including the delete operation) takes 2 hours.
How can we reduce this overhead time.
Dropping the index and recreating them is not an option!

My method is logical if you are planing to migrate at least 90% of all data from X to Y. If so
For the new X table you only need at most 10% of the size of the current X table ( At my previous post by saying "the same as" i did not mean data, i mean only DDL except storage). Moreover, after renaming X to X_ and copying "data not to be copied to Y" from X_ to X, you may drop unnecessary indexes from X_ to free up some space.

Finding Deleted operations in a Order

Hi all,
I need a way to find the deleted operations in an order for a smartform. if a operation is deleted in a particular order it should not be printed in the form. What i need is where to find the indicator that the operation in the particular order is deleted. It would be great if anyone can help me with this.
Thank You.

HI all,
I have the same problem in PP smartforms, someone has solved? thanks

Usage of triggers for delete operation

Hi,
I have a table named tst_employee:
CREATE TABLE TST_EMPLOYEE
(EMPLOYEE_ID NUMBER,
NAME VARCHAR2(100),
DEPT_ID NUMBER,
AGE NUMBER,
DOB DATE,
DESIGNATION VARCHAR2(50),
ADDRESS VARCHAR2(1000),
UPDATED_BY NUMBER,
UPDATED_DATE DATE,
APP_SOURCE_CODE VARCHAR2(3),
ACTIVE_IND VARCHAR2(1));
And another table for storing the audit records as:
CREATE TABLE TST_AUDIT
(AUDIT_ID NUMBER,
EMPLOYEE_ID NUMBER,
FIELD_ID NUMBER,
FIELD_DESCR VARCHAR2(50),
TRAN_TYPE VARCHAR2(1),
BEFORE_VALUE VARCHAR2(1000),
AFTER_VALUE VARCHAR2(1000),
BEFORE_DATE DATE,
AFTER_DATE DATE,
APP_SOURCE_CODE VARCHAR2(3),
UPDATED_BY NUMBER);
Now on the event of insert, update or delete I want to capture the data in the audit table. This is for the monitoring purpose. Now the problem which I have is in the delete operation. When a record is deleted I want to capture who is deleting the record in the column TST_AUDIT.UPDATED_BY. Can you please suggest the options how can I do this. I am using Oracle 10g.

Row level triggers offer :OLD.column_name and :NEW.column_name. For insert :OLD.column_name is always null and :NEW.column_name constains inserted/to be inserted value. In case of update :OLD.column_name is value before update and :NEW.column_name constains updated/to be updated value. In case of delete :OLD.column_name is deleted/to be deleted value and :NEW.column_name is always null.
SY.

About the delete of joined view

if create a view refering to a single table, then the delete operation to the view will delete the corresponding row in the refered table, but when i create a view by joining two tables, the result seems to be confused.
create view ev select e.ename,d.loc from emp e,dept d where e.deptno=d.deptno;
delete from ev where loc like 'N%';
then 3 row was deleted from the view with 3 rows deleted in the emp table, but nothing happened in the table dept which supposed to be also delete;
Why, or is these any principles dealing with the deleting this kinds of view. and I tried, the insert of this view is forbidden.

nothing happened in the table dept which supposed to be also delete
Sorry, according to who exactly? According to your opinion of how this ought to work without actually reading the documentation?
DELETE from a join view requires that the view contains one (and only one) key-preserved table and that is the table that is DELETEd from.
If you wanted to modify this behaviour (e.g. to DELETE from both tables) you could consider using an INSTEAD-OF trigger.

QuickLook and overriden new/delete operator

Hi at all,
I need your !great! help. Currently I am on the way to write a QuickLook plugin.
To do that, I use a static lib in the project to get certain information of the file.
Everything works fine, except the QuickLook plugin crashs on exit. The new and delete operator are overriden by the external lib.
The problem is, it seems that OSX allocates some internal stuff with the
original new operator and deallocates it with the +new overriden+ delete
operator which occurs a crash.
Any ideas how to solve that? The overriden new operator is in its own namespace.
Each help is highly appreciated
Bye,
moepmoep

MoepMoep wrote:
No, the library is well tested and not buggy. It simply overrides the new/delete operator
and in some calls the delete operator gets some pointers where the corresponding new operator did
not allocated the memory.
That sounds like a bug to me.
Apple (or, rather, gcc) is free to do any clever things it wants inside its own new/delete operators. If some library wants to write their own, they don't have to emulate anything that Apple does. The old C malloc still works great. But if you have some pointer to memory that did not come from a valid malloc call, you better not call free on it.
Keep in mind that many open source projects are built and tested only on Linux. Linux is notorious for doing everything in its power to make bad code run. Then, people take this running and tested code and try it on MacOS X and it fails. "But it runs great on Linux!" they proclaim. That is true, but only because malloc never fails on Linux. If your programs runs out of memory on Linux, Linux will kill some other, random program. On MacOS X, malloc may fail and developers need to check for that.

Track the delete operation.

Similar Messages

Maybe you are looking for