Continuous background job

Hi Experts,
I have a requirement , where a program need to be executed as a continuous job. In the logic of the program, there is a
'Do.. EndDo' , which would be continuously running.
I have set the program as a background job, where I have a single step , which would execute the Report.
As the report has a never ending loop, it would keep running.
Now I am facing a problem. In the job log, it keeps on updating the message '1 step(s) found ( ABAP_PROGRAM: XXX ; ABAP_VARIANT: ; EXT_PROGRAM: ; STATUS: R ) '.
Though I have a single step, it is updating the message for every 20 secs in the log.
And also, I am thinking , if the job log gets overflowed or any memory issue .
Please suggest, how to stop the job log updating the same message.
Thanks & Regards,
Lalitha

Thanks for the valuable inputs.
I have commit statement in the program. Regarding the point that 'The visibility of the job log messages is controlled by program commits.' , but I am not getting, why the message is getting repeated.
Basically the program flow is like this.
Program continuous_report.
wait for 10 sec.
Do..
Commit work.
Enddo.
The log message is that...
16.12.2010 05:31:20 Job started
16.12.2010 05:31:20 Step 001 started (program Continuous_Report, variant , user ID XXX)
16.12.2010 05:31:30 1 step(s) found ( ABAP_PROGRAM: Continuous_Report,; ABAP_VARIANT: ; EXT_PROGRAM: ; STATUS: R )
16.12.2010 05:31:56 1 step(s) found ( ABAP_PROGRAM: Continuous_Report, ; ABAP_VARIANT: ; EXT_PROGRAM: ; STATUS: R )
16.12.2010 05:32:17 1 step(s) found ( ABAP_PROGRAM: Continuous_Report, ; ABAP_VARIANT: ; EXT_PROGRAM: ; STATUS: R )
16.12.2010 05:32:39 1 step(s) found ( ABAP_PROGRAM: Continuous_Report, ; ABAP_VARIANT: ; EXT_PROGRAM: ; STATUS: R )
Thanks & Regards,
Lalitha

Similar Messages

Continue background job processing.

Hi All,
How to handle the system error
MESSAGE E328 WITH MATNR RAISING CONVERSION_NOT_FOUND thrown by FM MATERIAL_UNIT_CONVERSION so that the background job continue processing.
Thanks and Regards,
Ashik K.

Hi,
I just now checked with this FM and worked with these two options:
1. Runtime error type 'E' will stop the program if we COMMENT the EXCEPTIONS while calling FM
* EXCEPTIONS
*   CONVERSION_NOT_FOUND         = 1
*   INPUT_INVALID                = 2
*   MATERIAL_NOT_FOUND           = 3
*   MEINH_NOT_FOUND              = 4
*   MEINS_MISSING                = 5
*   NO_MEINH                     = 6
*   OUTPUT_INVALID               = 7
*   OVERFLOW                     = 8
*   OTHERS                       = 9
2. Runtime error type 'E' (which stops the program) will not come in this case
EXCEPTIONS
   CONVERSION_NOT_FOUND         = 1
   INPUT_INVALID                = 2
   MATERIAL_NOT_FOUND           = 3
   MEINH_NOT_FOUND              = 4
   MEINS_MISSING                = 5
   NO_MEINH                     = 6
   OUTPUT_INVALID               = 7
   OVERFLOW                     = 8
   OTHERS                       = 9
Hope this helps.
Regards,
Amit Mittal.

Background Job Scheduling

Hi,
I am scheduling a report to run in background.
In this report it is creating background jobs automatically for different company codes.
It submits the 1st background job and waits until it finishes.
Then 2nd job starts in background and continues with other jobs.
At end it finishes all the jobs and closes.
Now my problem is.
1.       Whether is it possible for us to submit all the jobs at 1 time. And execute at same time. Ie., 1st, 2nd job will start at same time.
2.       If possible how can we do that.
What I have written is
loop at companycode.
Create job name.
call fun 'Job_Open'.
submit xxxx user sy-uname via job job_name numer job_count
to sap-spool
spool parameters l_spool_parameter
without spool dynpro
with companycode
with ......
and return.
endloop.
Please help ASAP, urgent.

hi praveen,
Job Scheduling Explained
Definition
Before any background processing can actually begin, background jobs must be defined and scheduled. The scheduled time for when a job runs is one part of the jobs definition. There are several ways to schedule jobs:
From Transaction SM36 (Define Background Job)
With the "start program in the background" option of either Transaction SA38 (ABAP: Execute Program) or Transaction SE38 (the ABAP editor)
Through the background processing systems own programming interface. (Many SAP applications use the internal programming interface to schedule long-running reports for background processing.)
Through an external interface.
Scheduling Background Jobs
Use
You can define and schedule background jobs in two ways from the Job Overview:
·         Directly from Transaction SM36. This is best for users already familiar with background job scheduling.
·         The Job Scheduling Wizard. This is best for users unfamiliar with SAP background job scheduling. To use the Job Wizard, start from Transaction SM36, and either select Goto ® Wizard version or simply use the Job Wizard button.
Procedure
       1.      Call Transaction SM36 or choose CCMS ® Jobs ® Definition.
       2.      Assign a job name. Decide on a name for the job you are defining and enter it in the Job Name field.
       3.      Set the jobs priority, or Job Class:
·         High priority:      Class A
·         Medium priority: Class B
·         Low priority: Class C
       4.      In the Target server field, indicate whether to use system load balancing.
·         For the system to use system load balancing to automatically select the most efficient application server to use at the moment, leave this field empty.
·         To use a particular application server to run the job, enter a specific target server.
       5.      If spool requests generated by this job are to be sent to someone as email, specify the email address. Choose the Spool list recipient button.
       6.      Define when the job is to start by choosing Start Condition and completing the appropriate selections. If the job is to repeat, or be periodic, check the box at the bottom of this screen.
       7.      Define the jobs steps by choosing Step, then specify the ABAP program, external command, or external program to be used for each step.
       8.      Save the fully defined job to submit it to the background processing system.
       9.      When you need to modify, reschedule, or otherwise manipulate a job after you've scheduled it the first time, you'll manage jobs from the Job Overview.
Note: Release the job so that it can run. No job, even those scheduled for immediate processing, can run without first being released.
Specifying Job Start Conditions
Use
When scheduling a background job (either from Transaction SM36, Define Background Job or CCMS ® Jobs ® Definition), you must specify conditions that will trigger the job to start.
Procedure
Choose the Start condition button at the top of the Define Background Job screen.
Choose the button at the top of the Start Time screen for the type of start condition you want to use (Immediate, Date/Time, After job, After event, or At operation mode) and complete the start time definition in the screen that appears.
For the job to repeat, check the Periodic job box at the bottom of the Start Time screen and choose the Period values button below it to define the frequency of repetition (hourly, daily, weekly, monthly, or another specific time-related period). Then choose the Save button in the Period values screen to accept the periodicity and return to the Start Time screen.
Once youve completed specifying the job start conditions, choose the Save button at the bottom of the Start Time screen to return to the Define Background Job screen.
No job can be started until it is released, including jobs scheduled to start immediately. Since releasing jobs can be done only by a system administrator from the job management screen (Transaction SM37) or by other users who have been granted the appropriate Authorizations for Background Processing, no unauthorized user can start a job without explicit permission
Managing Jobs from the Job Overview
Use
The Job Overview, or Job Maintenance, screen is the single, central area for completing a wide range of tasks related to monitoring and managing jobs, including defining jobs; scheduling, rescheduling, and copying existing jobs; rescheduling and editing jobs and job steps; repeating a job; debugging an active job; reviewing information about a job; canceling a job's release status; canceling and deleting jobs; comparing the specifications of several jobs; checking the status of jobs; reviewing job logs; and releasing a job so it can run.
Procedures
To display the Job Overview screen, choose CCMS ® Jobs ® Maintenance or call Transaction SM37. Before entering the Job Overview screen, the system first displays the Select Background Jobs screen. You'll need to complete this Job Selection screen to define the criteria for the jobs you want to manage. Once you've selected jobs to manage, you can choose from a wide range of management tasks:
To copy a single existing job, choose Job ® Copy.
To reschedule or edit job steps or attributes of a single job, choose Job ® Change. A job step is an independent unit of work within a background job. Each job step can execute an ABAP or external program. Other variants or authorizations may be used for each job step. The system allows you to display ABAP programs and variants. You can scan a program for syntax errors. You can also display the authorizations for an authorized user of an ABAP job step.
To repeat a single job, choose Job ® Repeat scheduling.
To debug an active job, choose Job ® Capture: active job. Only a single selection is allowed. If an active job seems to be running incorrectly (e.g., running for an excessively long time), you can interrupt and analyze it in debugging mode in a background process, and then either release it again or stop it altogether.
You will be able to capture a background job only if you are logged on to the SAP server on which the job is running. To find server information in the Job Overview, select and mark the job, then choose Job ® Job details.
To review information about a job, choose Job ® Job details. Details displayed can include:
current job status
periodicity, or the repetition interval
other jobs linked to the current job, either as previous or subsequent jobs
defined job steps
spool requests generated by the current job
To cancel a job's "Released" status, select the job or jobs from the Job Overview list and choose Job ® Release -> Scheduled.
To cancel a job from running but keep the job definition available, select the job or jobs from the Job Overview list and choose Job ® Cancel active job.
To delete a job entirely, select the job or jobs from the Job Overview list and choose Job ® Delete. Jobs with the status of Ready or Running cannot be deleted.
To compare the specifications of more than one job, select the jobs from the Job Overview list and choose Job ® Compare jobs.
To check the status of jobs, select the job or jobs from the Overview Job list and choose Job ® Check status. This allows you to either change the job status back to Planned or cancel the job altogether. This is especially useful when a job has malfunctioned.
To review job logs, select a job or jobs with the status Completed or Canceled from the Job Overview list and
regards
karthik
reward me points if helpfull

Regarding capturing VF01 ERROR logs in a background job

Hi ,
We are running a background job that calls transaction VF01 via BDC .
CALL TRANSACTION 'VF01' USING   bdcdata
                          MODE    S
                          UPDATE N
                          MESSAGES INTO gt_messtab.
In some cases billing documents will not be generated. In such cases systems will throw message "Check the log "
After which we can check the error messages via Menu Edit>logs.
This error messages (stored in XVBFS internal table) need to captured in the job log i.e custom programme.
Unfortunately i could not found any user exits/BADIs that can help me in capturing message.
Can you please help me how can i solve this issue?
Thanks in advance
Manoj
Edited by: Manoj J on Nov 16, 2011 9:42 AM
Moderator message : Duplicate post locked, continue with original thread [How to capture error logs of VF01 in a background job log   |Re: How to capture error logs of VF01 in a background job log].
Edited by: Vinod Kumar on Nov 16, 2011 3:34 PM

Hi Manoj,
I had a similar problem.
I solved it selecting from VBSK the last record of the user regarding elaboration of invoices, and then calling f.m. 'VBSK_ALV_DISPLAY'.
Otherwise you could call directly transaction "V.21"
Hope this could help you
Andrea

Background Job running long time.

Hello All,
One of customer background job running for long time more than 11 hours, but usually this jobs get completed within 4 hrs.
I have checked in WP trace file, it shows "WP has reached abap/heaplimit=40000000".
This is the Value I can see in RZ11. Now it is not possible to change the value.
Frequently facing this problem, please guide me to solve this issue.
Regards
Vinay.

Hi
First of all, abap/heaplimit is the limit of memory usage by a WP.
Follwing is the documentation from help.sap.com
The value of the parameter should be between 10000000 (10 MB) and 2000000000 (2GB), the recommended default setting is 40000000 (40 MB).
The objective is to have the least number of work process restarts as possible, without a swap space bottleneck occurring. The heap memory allocated by the work processes has to be released again.
As shown in the graphic, the value of abap/heaplimit should be smaller than abap/heap_area_dia or abap/heap_area_nondia, so that the dialog step that is running can still be executed. This prevents the work process from working against the operating systemu2019s swap space limit if there is a programmed termination of the work process.
So,Do the above checks. And importantly check the memory utilizations at the time of running your job and check whether any other job was also running at the sametime.
Increasing the parametes is not a good idea.. If you continuously face this issue then only think of parametes. As u said ,this problem is not happenng too often.
Hope this is useful

Start background Job when another is finished (NOT with the JobSteps)

Hi guys,
i need your help.
I've already searched here in forum, but i wasn't able to find a good solution.
I have this problem.
I have a program that create a background job with the FM FM JOB_OPEN.. SUBMIT report with parameters .. JOB_CLOSE.
I want that if i run this report again (20 secs after for example) it does:
- Check if there is an already running Job with the same name (means with state 'R') (This task is simply, with a select on TBTO table)
- If it's found, it have to create a new job with the same name that starts automatically AFTER the first running job is finished (don't care about the end-state of the first job).
I've already tried with pred_jobcount & PRED_JOBNAME parameters of the JOB_CLOSE FM but it doesen't works!
The JOB_CLOSE, creates a job in Planned state. But when the first Job is finished, the second job(Planned) doesen't start automatically.
In this scenario, i CANNOT use endless loops (wait until the first job is finished and then submit the second) Job-steps (one job that contains multiple steps) events (i have to start only one job after the predecessor is finished), because this report could be run many times and each job should be collected like a "stack" (only when the first job is finshed the second "registered" should be started and so on, until the aren't more planned jobs).
<REMOVED BY MODERATOR>
Thx a lot for your help.
Andrea
Edited by: Alvaro Tejada Galindo on Jun 12, 2008 12:19 PM

Hi Veda
i can tell u but ... some reward points are very appreciated ....
I'm joking (of course)
Here the question:
I have a program (called A) that submit a new program (called B) with the JOB_OPEN .. submit JOB_CLOSE. The program B should start only if another program B (called before for example) is finished.
Here the solution.
I add a parameter (with no-display clausole) to the program B. In this parameter i pass to the program, the job number returned by the JOB_OPEN function.
When i create the JOB with the function, the "jobname" parameter is set with value 'G_DELIVERY' (Here u can change the name of the job as u want : this is the jobname that u see in SM37 transaction).
In the start-of-selection of program B i put a "waiting" procedure like this :
First i save a timestamp of system-date and system-time (called for example r_date and r_time)
select from table TBTCO all the jobs called "G_DELIVERY" with jobnumber <> from the jobnumber parameter (that means exclude itselfs) with status running ('R') -> that have startdate /starttime less than the r_date and r_time <- (this is the key of the selection that solve the problem).
if it is found (means there is another running job started before this one).
wait up to 60 seconds. "for example
repeat the selection.
endif.
When the job called before ends, this one programs exits from the loop and continues. If u submit more programs "B", they'll works like a stack.
I should say to u just one thing.... I solved my problem in another way (because i've changed the logic so this problem was no more) so i didn't implemented that logic, but it should works very good.
Try it and tell me!
Bye
Andrea

Error in Background Job Processing.

Dear All,
         I am using a background job for processing the outbound delivery.
         In the user exit in delivery program I am checking for the combination of
         ship to party, Material & Batch of outbound delivery in a ztable. If the combination exists in
         Ztable I do not want to save & ignore this delivery creation and continue with the next
         delivery order creation thru the batch job.
         However I need to give an error / warning message for the delivery orders
         which are not created so I am giving the error message as success message
         in the below format :-
          IF L_RETURN-TYPE = 'E'.
            MESSAGE S000(Z64_I) WITH L_RETURN-MESSAGE_V1
                                     L_RETURN-MESSAGE_V2
                                     L_RETURN-MESSAGE_V3
                                     L_RETURN-MESSAGE_V4
                                     DISPLAY LIKE 'E'.
            PERFORM FOLGE_GLEICHSETZEN(SAPLV00F).
            FCODE = 'ENT1'.
            SET SCREEN SYST-DYNNR.
            LEAVE SCREEN.
          ENDIF.
        But my job is finished as soon as the above message is encountered and the
        rest of the outbound deliveries are not created.
        However if I execute the progarm in foreground the execution is working fine.
        Could you please let me know what do I need to code / do corrections to make
        the program run successfully in background.
Thanks & Regards
Jitendra Gujarathi

Hello,
Check system log and also check work process trace, to know the error.
I feel that it could be Authorization issue, you don't have sufficient authorization to do.
Hope this helps.
Regards
venkata
Edited by: venkata emandi on Sep 12, 2011 8:13 AM

Workprocess for Infinite background job in ABAP

Dear all,
I have a background job which has an infinite loop i.e. WITHOUT any exit condition. The loop contains WAIT UPTO x mins statement.
So my query is if I run this program in background then will the Background Work process which executes this job, always be occupied and cant be used for other background jobs ? In other words, will I be wasting 1 BG workprocess for this job since it will never be freed to be used for other background jobs?
LOOP AT ITAB
WAIT UP TO x MINS.
ENDLOOP.
Requesting your expert insights .
Thanks in advance.
BR
Nilesh

I am not so sure if this is correct. To my knowledge, BTC processes, as opposed to DIA, never perform "roll out" (e.g. during an RFC call) but just wait until processing continues. I just made a quick test in NW 740 release, there the BTC process remains occupied by my job when it does WAIT UP TO xxx SECONDS. So you occupy one BTC process as long as your job runs.
Please try out yourself and then award points to whoever is right, please.
Regards,
Alex

Scheduled TMS_ # TMS_TP_IMPORT background jobs fail with error code 0232

Is anyone else having the problem we've just encountered since upgrading to ECC 6.0 and CRM 2007, where once setting up the automatic import ALL background jobs, to have all of the Transport Change Requests imported into the Q-Prod testing systems automatically, fail after several successful runs?
After setting up the automatic import background job to run every 15 minutes, the first few run successfully (importing any transports as it should), then the transports stop being imported.
The background jobs continue to run, but have the following in the job log overviews CRM 2007 - "Could not start transport control program tp". R\3 ECC 6.0 - "Transport control program tp ended with error code 0232".
Only by deleting the background job and recreating it (or manually importing) from STMS, can the transports continue to be imported. Although I have noticed that there are multiple tp.exe processes started, which need to be deleted before the STMS and TMS Background jobs work again.
The:
   DB is SQL Server 2005
   disp+work is at version 7000.150.14.48855
   R3Trans is at version 7000.149.14.47907
   tp is at version 7000.144.14.39594
Any suggestions on how to cure this for good?

Thanks All
FYI
SAP first suggested the following notes: 19466, 1150361, and to add a Startup profile parameter rfc/use_gwstart = 1. Which was only partially successful.
I guess it had more to do with RFC processes hanging than it did with the tp, which was fixed in a later kernel patch level.
I've upgraded the kernel to the latest (70000.181.0.0) and my problem was fixed.
Ken

Submitting a updates in a prog to a background job

Hello everyone,
I am generating a alv report with production orders.In the toolbar i created a button for updating data.When i select some production orders in the alv list and select the button then transaction co02 is called and some updates are done.I done that the updating by using bdc recording and the updates are working fine.
But my problem is
1. when the user selects the updates button in the toolbar these updates should take place via. a background job.
2. Before submitting the background job the program needs to check if a version of a prog is running,the check should look for the prog name and the triggering user, if a match is found the user should receive a error message.
3. If a match is not found then the background job should be scheduled immediately and should be created with a name of prog name and the triggering user.
I need to these steps,can anyone please give me a sample code for this.
Thank you.
dp.

The approach I would probably take for this would be:
1. Write the records that need processing to a "Z" table from the ALV, and delete these back out (or mark as "done") once processed by the batch job.
2. It is possible to check for a batch job name and its status, but I've usually found an easier and more reliable way is to use a lock object to check this e.g. you could use and existing object such as ESRDIRE with a dummy lock argument such as 'CO02_IS_LOCKED' . The batch job should contain code to set the enqueue... if this enqueue fails it means another job / program is already processing the requests so this instance can stop, otherwise the batch program can continue the CO02 processing. In other words, you are creating a doorway through which only one instance of the job can pass at a time.
3. A couple of easy ways to do this are (a) event triggered jobs, wherein you define the job and have it fire up when an event is raised (e.g. by calling function BP_EVENT_RAISE) or (b) use the SIMPLE_BATCH_JOB_SUBMIT (which calls the JOB_OPEN, JOB_SUBMIT, JOB_CLOSE functions for you).
Jonathan

Submitting a update in a porg to a background job?

Hello everyone,
I am generating a alv report with production orders.In the toolbar i created a button for updating data.When i select some production orders in the alv list and select the button then transaction co02 is called and some updates are done.I done that the updating by using bdc recording and the updates are working fine.
But my problem is
1. when the user selects the updates button in the toolbar these updates should take place via. a background job.
2. Before submitting the background job the program needs to check if a version of a prog is running,the check should look for the prog name and the triggering user, if a match is found the user should receive a error message.
3. If a match is not found then the background job should be scheduled immediately and should be created with a name of prog name and the triggering user.
I need to these steps,can anyone please give me a sample code for this.
Thank you.
dp.

The approach I would probably take for this would be:
1. Write the records that need processing to a "Z" table from the ALV, and delete these back out (or mark as "done") once processed by the batch job.
2. It is possible to check for a batch job name and its status, but I've usually found an easier and more reliable way is to use a lock object to check this e.g. you could use and existing object such as ESRDIRE with a dummy lock argument such as 'CO02_IS_LOCKED' . The batch job should contain code to set the enqueue... if this enqueue fails it means another job / program is already processing the requests so this instance can stop, otherwise the batch program can continue the CO02 processing. In other words, you are creating a doorway through which only one instance of the job can pass at a time.
3. A couple of easy ways to do this are (a) event triggered jobs, wherein you define the job and have it fire up when an event is raised (e.g. by calling function BP_EVENT_RAISE) or (b) use the SIMPLE_BATCH_JOB_SUBMIT (which calls the JOB_OPEN, JOB_SUBMIT, JOB_CLOSE functions for you).
Jonathan

Reg Background job execution

Hi gurus,
I have a doubt in back ground job processing...We have a job which runs for very long time (archiving job). I want that job to run from Sunday morning to evening 6PM again I need to run the same job next sunday morning to evening continuing from previous week stop point.
My question , there is a option in SM37 to STOP the job. If I stop the job and then resume back the same job, will it continue from the point where it stopped?
Is there any way to execute the background job like above scenario. I know it can be done by 3rd party tool, but I want to do the same from R/3
Vinodh

Hi Vinodh,
In later versions of SAP (from Enterprise version if iam not wrong), you can interrupt the archive job and continue later. This is not the feature of background processing, but the archiving job itself.
Interruption is possible on basis of duration of the job and size of archive file. See the following link which details the mentioned feature; go to The Archive Development Kit and see the topic Interrupting and Continuing Archive Sessions:
http://help.sap.com/saphelp_47x200/helpdata/en/6d/56a06a463411d189000000e8323d3a/frameset.htm
Hope this helps to work towards the solution of your problem.
Regards,
Naveen

Copying background jobs of another user.

Respected gurus,
I have to delete an user id, he has schedluled several jobs, but jobs has to be continued.
Please let me know how to copy background jobs scheduled by other user.
Regards,
Daya

Daya
You donu2019t necessarily need to copy the jobs.
You can just change the owner of the step(s) in the jobs.
The name on the job itself will remain the same, but the job (steps) will run as the new name that you enter.
If displaying the previous owner is an issue, then you will need to copy the jobs.
Hope this helps.
PJ

Background Job call from a program

Hi all,
We have a program which runs for max of 5 minutes (say).
It is scheduled to run every minute in SM36 so that it keeps continuing to run.
But sometimes even when the previous job is running the one starts up, thereby creating inconsistent results.
So we need to call the job from the running program itself.
How should we go abt this ? Is there any other way available to solve this ?
Thanks in Advance,
Regards,
Vivek K

Hi,
Managing Jobs from the Job Overview
Use
The Job Overview, or Job Maintenance, screen is the single, central area for completing a wide range of tasks related to monitoring and managing jobs, including defining jobs; scheduling, rescheduling, and copying existing jobs; rescheduling and editing jobs and job steps; repeating a job; debugging an active job; reviewing information about a job; canceling a job's release status; canceling and deleting jobs; comparing the specifications of several jobs; checking the status of jobs; reviewing job logs; and releasing a job so it can run.
Procedures
Specifying Jobs
To display the Job Overview screen, choose CCMS ® Jobs ® Maintenance or call Transaction SM37. Before entering the Job Overview screen, the system first displays the Select Background Jobs screen, where you must select the criteria for the jobs you want to manage. These criteria include:
Job name (which can contain a wildcard (*) to select jobs with related names oru2014by using the wildcard aloneu2014any job). Required.
Name of the user who scheduled the job (the wildcard is allowed here as well). Required.
Job status
Planned or actual start time of the job.
Job start condition or event linked to the start condition
Job step
When and how often the job runs, or " periodicity"
You can also choose Extended Job Selection, where you can more precisely define the specific criteria that determine which jobs to select. If you have administrator Authorizations for Background Processing, you can display jobs in all clients. Without this authorization, only jobs in the client that you are logged on to will be displayed.
Once you have set the criteria for you job selection, choose Execute.
Managing and Monitoring Jobs
Once you have called the Job Overview screen (Transaction SM37 or CCMS ® Jobs ® Maintenance) and selected which jobs you want to manage as described above, you can choose from a wide range of management tasks.
To copy a single existing job, choose Job ® Copy.
To reschedule or edit job steps or attributes of a single job, choose Job ® Change. A job step is an independent unit of work within a background job. Each job step can execute an ABAP or external program. Other variants or authorizations may be used for each job step. The system allows you to display ABAP programs and variants. You can scan a program for syntax errors. You can also display the authorizations for an authorized user of an ABAP job step.
To repeat a single job, choose Job ® Repeat scheduling.
To debug an active job, choose Job ® Capture: active job. Only a single selection is allowed. If an active job seems to be running incorrectly (e.g., running for an excessively long time), you can interrupt and analyze it in debugging mode in a background process, and then either release it again or stop it altogether.
You will be able to capture a background job only if you are logged on to the SAP server on which the job is running. To find server information in the Job Overview, select and mark the job, then choose Job ® Job details.
To review information about a job, choose Job ® Job details. Details displayed can include:
current job status
periodicity, or the repetition interval
other jobs linked to the current job, either as previous or subsequent jobs
defined job steps
spool requests generated by the current job
To cancel a job's "Released" status, select the job or jobs from the Job Overview list and choose Job ® Release -> Scheduled.
To cancel a job from running but keep the job definition available, select the job or jobs from the Job Overview list and choose Job ® Cancel active job.
To delete a job entirely, select the job or jobs from the Job Overview list and choose Job ® Delete. Jobs with the status of Ready or Running cannot be deleted.
To compare the specifications of more than one job, select the jobs from the Job Overview list and choose Job ® Compare jobs.
To check the status of jobs, select the job or jobs from the Overview Job list and choose Job ® Check status. This allows you to either change the job status back to Planned or cancel the job altogether. This is especially useful when a job has malfunctioned.
To review job logs, select a job or jobs with the status Completed or Canceled from the Job Overview list and choose Goto ® Job log.
To release a job so it can run, select a job from the Job Overview list and choose Goto ® Release.
Regards,
Jagadish

ST12 performance tuning background job

I would like to trace a program which will run about 3 hours. In ST12, I cannot run the program in "current mode" since the execution time is too long. How can I still use ST12 to trace the program execution when the program execute is submitted as background? Pls advise the detail configuration needed in ST12.
Thanks very much for your help!
PS: I tried to use "trace for" user instead of "current mode", and set the tasktype to "BT", when I press "start trace" and start the program background job in another session, seems nothing can be traced...
Edited by: Wing Yin, Anthony TANG on Aug 7, 2009 6:10 AM

Hi,
did you set up the trace on the same server where you background job was running?
3 hours is quite long... not possible to trace the whole execution.
With ST12 / SE30 (ABAP Trace) you can trace at maximum 4293 seconds. For ST05 i don't know no other limit then file size.
However you should not take such big traces.
Start with analyzing the STAD record in order to understand where the time is spent (CPU time, DB time, RFC + CPIC time, ...).
For CPU time you can continue with ST12 / SE30, for DB time you need ST05 and for RFC + CPIC time you can check the RFC Subrecords in STAD to see the 5 most expensive RFC calls (destination and FM).
In case of DB time a SQL Cache reset before the job execution and a delta analysis after the execution could help as
well if there is not too much other DB load on the system.
You can take some samples with traces. Start n traces with a few minutes duration and check the results agains STAD (overview). E.g. if according to STAD most of the time is spent in DB time but your sample trace shows mostly CPU time,
your sample is probably not a a good sample (not showing you a part where most of your time is spent).
Other than that, if your Batch Job is your program, you cann add some mesages to the job log to narrow down what part
of the programm consumes time (data selection, data processing, ...) . Another option is to evaluate if you can reduce the
processed data volume to take a trace in order to find out expensive parts of the program. This of course can change the
execution (e.g. of SQL statements) significantly.
You see analyzing long running batch processes could be a challenge
Kind regards,
Hermann

Continuous background job

Similar Messages

Maybe you are looking for