Problem in process chain running for 3 days

Hi Experts,
Please help me
We have process chain like below
1.     delete index
2.     full loads going 3 cubes (16 infopacs)
3.     loading to ods and then update to cube
4.     delete overlapping requests (16 infopac)
5.     create index
6.     aggregate filling
7.     delete PSA
Now the issue is, from last Saturday onwards
Process chains was still in Yellow status till date, so now, 1,2,3,4, are still running
In process chains, out if 16 infopacs, 12 are green, remaining 4 are stopped
And in ODS activation of ods data variant shows yellow, but when checked the data is activated, but not pushed to cube. When checked log it show u201Csystem exception error_messageu201D, so process terminated.
When we checked for shortdump,
It says u201Clocal time on the application server is incorrectu201D
Experts please let me know how to solve this and correct the load.

Hi Experts,
Thank u for your help,
I have done the steps manually yestarday, checked todays, still it is in yellow status.
let me explain the scenario,
The data is going to 3 cubes,
the process chain is designed like this,
1. initially delete index
2. full load to cube using 16 infopac and loading from (2)ODS to cube.
3. delete overlapping requests
4. create index (3 cubes)
5. fill aggregates (only one cube)
6. Delete PSA request
so now , chain is having problem at
1. data is loaded to ods
2. data activation (showing yellow, but when checked data is activated)
3. further update to cube (2 cubes) in process chain this is not started atall ( so y'day I've done manually)
4. out 16 infocpack, only 12 infopacs are green and remaining 4 are not started.
so yestarday I've manually schedule remaining 4 infopacs.
and deleted previour requests in infocube as they are full load, but I did'nt done create index and fill aggregate steps.
now the problem is , out of 3 cube report available request is present for 2 cubes, but for 3 cube this report request is not available, but QM status is green
please let me know how to proceed.
one small silly question, do i need to fill aggregate step to get this report available request.....please suggest.
Thanks & Regards,
Raghu

Similar Messages

ABAP process job ( BI_PROCESS_ABAP ) in process chain running for long time

Dear Experts,
We have an ABAP process job BI_PROCESS_ABAP in a process chain. This job is running for a very long time than previous run time.
Recently we had an ORACLE database upgrade from 9.2 to 10g.
From the very next day , we are having this problem.
And also previously this process-chain was under control of BASIS guys . they were using a tool called CRONACLE to schedule it. At the time of upgrade this tool was crashed and they asked us to schedule it in BW system instead.
Underlying program is using two custom tables.
one is having both primary & secondary indexes but the other table is having only primary index.
I am not sure whether we have secondary indexes on other table prior to upgrade or not.
Where must be the problem?...
At the time of upgrade BASIS has re-organized the table space along with database upgrade.
please help me...
Regards,
SRIKANTH

Please ask yr basis consultant to check the system parms.
Below i have listed down SAP recommneded system parms .
abap/heap_area_total 4000000000
abap/heap_area_dia 2000000000
abap/heap_area_nondia 2000000000
Also check in SM37 if any other jobs are running for long time, peventing the your job running.
Check the SM12 for locks or SM58

Stopping Process Chain schedule for just one day...

Hi,
Our source systems and BI are getting a patch upgrade during this weekend and I need to stop all the Process Chains schedule for Friday night/ Saturday morning. Is there a simple way I can stop all the Process Chains from executing just for tomorrow instead of manually removing each PC from schedule and again putting them back on schedule on Saturday?
Thanks,
Sri.

Hi
You can reschedule them instead of de-schedule and schedule back.
1)open your process chain in rspc1
2) goto planning mode
3) right click on start --> display schedule job
4) select the job --> in menu bar --> click on first one from left --> change
here change the dates as per your requirement and save the job.
If you have any event based chains which are going to start after time based chains gets complete, no need to de-schedule as we are not going get trigger from time based chains.
I don't think so we have any program to de-schedule all the chains at a time.
Regards,
Venkatesh

Get information for process chain running

Hello Gurus,
after a process chain running, how can I know how many records are loaded into the data target ? don't want to get back to DTP monitor to get that information. prefer to getting that information in the log view.
Many thank.

Hi,
The fastest way possible to get the number of records loaded will be from the DTP monitor only.
You must be having a step where the DTP is executed, simply double click on it and go to monitor.
hope this helps.
thanks,
rahul

Problem in Process chains - Need help

Dear Experts
I have a problem in Process chains
I have ODS1 from which Iu2019m updateing the ODS2 and ODS3 with same Delta infopackage.
The difference between ODS2 and ODS3 is in ODS2 Iu2019m loading with Local Currency
And in ODS3 is getting loaded with Group Currency
When Iu2019m running with Process chain its executing twice as below
First time : It is showing only ODS2 as the Datatarget and loading all the delta records to ODS2
Secondtime: Its showing both ODS2 and ODS3 in Datatarget and loading with 0 Records.
In this way No record is getting updated in ODS3.
Could you please share your thoughts on why it is triggering twice ? And how we can avoid this.
Thanks
Lakshminarayana

Thanks Shambu for quick reply
The variant is being used to load the ODS2 & ODS3 and Automatic further processing tick also removed.
Thanks
Nerusu

Process chains run

Hi all,
We have a BW 3.5 system in our landscape.The process chains run get triggered through control M.When i checked the process chains are occupying the dialog work process.Is that how it should run.Dont we have a option to run in background.Thanks.

Hi Ambarish,
Process chains run in background.
Check if your PC is really executing by using RSPC1 transaction and providing your PC name. Check the job log for the day when the PC is executed.
You might also like to check if your system do have enough number of background processes. May be your basis team can help you here.
Rrds,
Nilima Rodrigues

Process chain running :doubt

Hi,
I had scheduled one process chain which is running now---
now its in last step that is
"deleting overlapping request in infocube----"
i had checked this job : it is running with out any hassles--
now after this only AND process is there to finish the chain---
Now i want to schedule the process chain again for weekend!
can i change the start varinat though the chain is still running??
Raj

Hi Raj,
Its not a good practive to reshedule the chains which is running now..
when u attempt to reschedule then the process will gets terminated from the point where its running .... I mean the present running job will be in active mode and it will gets finished.. but the process after the active process will gets disconnected as soon as u reschedule the chain..
So better to wait for ur chain to finish and then proceed with the reschedule...
Also before rescheduling try to deschedule the chain...
Thanks
Assigning points is the way to say thanks..
If this resolves ur issue then close this thread by assigning points..

GRC AC 5.3 - Sync Mode background job running for days.

We have the following performance issues on the GRC System.
1- Please note that the sync mode background job has been running for DAYS.
My question is if I truncate table virsa_cc_prmvl; (which stores all the analysis results), will my GRC system be effected.
2- Secondly we have a job analysing roles, which causes the performance to go down. And due to we have a extremely high number of roles, we would like to know if there would be a way to improve the analysis for roles. Is there any parameter modification in GRC / J2EE/ backend related to roles?
Details of our roles:
- we have more than 60000 roles in our system
- 75 % of the roles are not assigned to any user (is there a way to avoid these roles? these roles are not put in order so too tedious to choose roles individually or by range.
3- We are running a job with four risks and it's taking more than two minutes per user (we've got 10.000 + users in total)
How can I resolve this performance issue.
4- What is the recommended size of the batch for user sincronization. Should I use 1000 or 10,000 (based on our above environment).
Thanks,
Haleem

Hello, Abdul!
How you have solved a problem?
Thank you a lot.
Best regards,
Denis Gaykov

Process Chain running even after end date

Hi,
We are scheduling month end PC to run every 2 hours and also we have given end date for today evening.but process chain runs even after end date for every 2 hours.so we have to manually remove it from scheduling.
Any method how can PC stop on its own after end date .
Any help highly APPRECIATED
Thanks
Nilesh Pathak

There might be multiple release jobs for thie process chain.
So check the same. Right click on process chain --> Display scheduled jobs --> Here all the jobs logs of the chain will be available.. So check the same here...
As you have descheduled it now.. probably you may not be having it as the first release job might be moved to Schedule status as soon as the END date has reached and then second one might have gone when you have removed the chain from Schduling..
Anyway check the same.. and also check in SM37 whether any other user has triggered it manually today or not?

Scheduling a process chain running in a specific hour

hi,
Is there any other way to schedule the process chain to run in a specific time besides using different meta chains to call the main process chain to do the loads separately.
Can we do this by using one main process chain?
for example, this is performed daily (hourly load: 4am load, 9am till 9pm load).
willie

Hi,
<b>Process Chains</b>:
From 3.x onwords to automate loads as well as process like Openhub,ababp
programs or activations or etc., We are using this process chains.
Coming to the major advantage
Like eariler infopakge groups if any pak fails we need to trigger all the loads
manully .wher as in Pchaing just correct that particular ipak and repaeat from that
particular process .it will take care abt remaing subsequent chain .
And also we can trigger parallel loads as well in this. Its much user friendly in monitoring.
1)You can make the process chain to be trigered by an event.
You do so by having the Job that start the PC start after an Event.
Then you define your set of job's (thru SM37) and make sure these jobs fire the event that starts the PC.
There is an ABAP program you can use that fires events (FM -> BP_EVENT_RAISE).
2)In the Start process of your process chain, select Direct Scheduling and then click the Change Selections icon. This will bring you to the Start Time window where you can put in the Date/Time you want to start your process chain. At the bottom of this window, click on the Periodic job box and you will notice another icon way at the bottom called Period Values. Click this to determine what frequency the job should be rescheduled (ie. daily, weekly, etc.).
3)you can do this by right-clicking on the start process, selection "Maintain Variant", choose "Direct Scheduling". Next select "Change Selections", pick your date and time, set your period values and you're done.
You can Schedule process chain in background by right-clicking on the start process, selection "Maintain Variant", choose "Direct Scheduling". Next select "Change Selections", pick your date and time, set your period values and you're done.
Hope it helps.
*****Assign points if helpful*******
Thanks,
Amith

Reg : To view process chain flow for a cube

HI Experts,
I need to view process chain flow for a particular cube of sales distribution.Can you please tell me stps to perform .

Hi,
If the cube is already loaded then you can just to the infopackages which are loading it and in the infopackages you have option in the "schedule" tab to see which process chain it is loading.
if not then you will have to identify which data source is loading to the cube and check for the infopackge.
and I think the DTP will be in the same chain as well.
For DTP it is alreday mentioned by the others.
Thanks
Ajeet

Process Chain Assining for another source sytem

Hi Experts,
I have a quick question on process chain assining for another source sytem.
For Example. I have a Process chain with 10 Infiopackages. Letz say that Infopackage was assinged to one source system.
Now I wanna a create a new process chain which shoud be the replica of the existing process chain. I wanna to load the data for the same info packages thru another sources system. Can we do this or else we have to reassign the data source for new source system for every infosources.. Is this Possible.
Kindly guide me if there is any nother solution for this.
Warm Regards,
Aluri

Hi..
ABAP routines are in place at infopack level or at the Transformation level?
If its there at the Transformation level then there wont be any issues for ur redesigning..
if its there at the IP level then u need to copy those things in ur new IP as per the requirement and change the things accordingly..
and then design the process chain..
Assign points if this helps..
Thanks

The way to stop process chain running takes forever (without stop)

Someone here used to say in order to stop a process chain running in the middle, we can go to menu: Process Chain -> Remove from Schedule. However we have never made it successful since it runs forever without a stop.
Maybe there is a background process that we can go to kill them immediately that make the whole process chain run really stop?
Thanks

Hi Kevin,
Refer to these posts:
Re: Is there way that we can stop the process chain running in the middle?
Re: How to stop a triggered process chain
How to stop the Job that is running in Background in a process chain
Bye
Dinesh

Process chain - schedule for certain days

Hello all,
How can I schedule a process chain that :
-should run 3 times per day
-between the 8th and the 12 workingday of a month
Any sugesstions?
Thank you.

Maybe I cna help others with my solution.
So...
I made the chain to run 3 times per day as Debjani advidced me in this thread.
I put 3 jobs, at the hours I wanted, the period daily, without weekends and holidays.
I made a reports that tests the 18-th working day (for example) and some others:
REPORT Z_DAYS.
TABLES : TFACS.
DATA: GI_CURSOR TYPE CURSOR,
      GI_TABIX TYPE SY-TABIX,
      GT_TFACS TYPE TABLE OF TFACS.
DATA: TEXT(50) TYPE C,
      LEN TYPE I,
      IND TYPE I,
      POS TYPE I,
      CNT_ONE TYPE I,
      CNT_ZERO TYPE I,
      DAY_VAR TYPE I,
      IND_VAR(3) TYPE C.
DATA: EVENTID LIKE TBTCJOB-EVENTID.
DATA: EVENTPARM LIKE TBTCJOB-EVENTPARM.
FIELD-SYMBOLS: <GT_TFACS> TYPE TFACS.
DAY_VAR = SY-DATUM+6(2).
OPEN CURSOR GI_CURSOR FOR
*select data for factory calender Germany (01) and year 2009
SELECT * FROM TFACS
WHERE IDENT EQ '01' AND JAHR EQ SY-DATUM+0(4).
IF SY-SUBRC NE 0.
EXIT.
ENDIF.
MOVE 100 TO GI_TABIX.
DO.
FETCH NEXT CURSOR GI_CURSOR
   INTO TABLE GT_TFACS
   PACKAGE SIZE GI_TABIX.
IF SY-SUBRC NE 0.
    EXIT.
ENDIF.
*loop on internal table
LOOP AT GT_TFACS ASSIGNING <GT_TFACS>.
    TEXT = <GT_TFACS>-MON01.
    LEN = STRLEN( TEXT ).
    WRITE : 'len' , LEN.
    DO LEN TIMES.
      IND = SY-INDEX.
      POS = SY-TABIX.
      IND = IND - 1.
      IND_VAR = IND + 1.
      WRITE: / 'ind', IND_VAR.
      WRITE : 'pos'.
      WRITE POS.
      IF TEXT+IND(POS) = '1'.
*number of working days increases with 1
        CNT_ONE = CNT_ONE + 1.
        IF CNT_ONE EQ 8 AND ( DAY_VAR = IND_VAR ).
          WRITE : / '8th working day'.
          EXIT.
        ELSEIF CNT_ONE EQ 9 AND ( DAY_VAR = IND_VAR ).
          WRITE : / '9th working day'.
          EXIT.
        ELSEIF CNT_ONE EQ 10 AND ( DAY_VAR = IND_VAR ).
          WRITE : / '10th working day'.
          EXIT.
        ELSEIF CNT_ONE EQ 11 AND ( DAY_VAR = IND_VAR ).
          WRITE : / '11th working day'.
          EXIT.
        ELSEIF CNT_ONE EQ 17 AND ( DAY_VAR = IND_VAR ).
          WRITE : / '17th working day' COLOR 5.
          EVENTID = 'ZE_PROD'.
          EVENTPARM = 'ZEP_PROD'.
          CALL FUNCTION 'RSSM_EVENT_RAISE'
            EXPORTING
              I_EVENTID              = EVENTID
              I_EVENTPARM            = EVENTPARM
            EXCEPTIONS
              BAD_EVENTID            = 1
              EVENTID_DOES_NOT_EXIST = 2
              EVENTID_MISSING        = 3
              RAISE_FAILED           = 4
              OTHERS                 = 5.
         IF NOT SY-SUBRC IS INITIAL.
           WRITE:/ 'EVENT DID NOT START'.
         ENDIF.
          EXIT.
        ELSEIF CNT_ONE EQ 18 AND ( DAY_VAR = IND_VAR ).
          WRITE : / '18th working day' COLOR 5.
          EVENTID = 'ZE_PROD'.
          EVENTPARM = 'ZEP_PROD'.
          CALL FUNCTION 'RSSM_EVENT_RAISE'
            EXPORTING
              I_EVENTID              = EVENTID
              I_EVENTPARM            = EVENTPARM
            EXCEPTIONS
              BAD_EVENTID            = 1
              EVENTID_DOES_NOT_EXIST = 2
              EVENTID_MISSING        = 3
              RAISE_FAILED           = 4
              OTHERS                 = 5.
         IF NOT SY-SUBRC IS INITIAL.
           WRITE:/ 'EVENT DID NOT START'.
         ENDIF.
          EXIT.
        ENDIF.
      ELSEIF TEXT+IND(POS) = '0'.
*number of holidays days increases with 1
        CNT_ZERO = CNT_ZERO + 1.
      ENDIF.
    ENDDO.
    WRITE : /.
    WRITE: 'January :' COLOR 1, <GT_TFACS>-MON01.
ENDLOOP.
REFRESH GT_TFACS.
ENDDO.
WRITE / .
WRITE ' The number of working days is :'.
WRITE CNT_ONE COLOR 5.
WRITE / .
WRITE ' The number of holiday days is :'.
WRITE CNT_ZERO COLOR 7.
WRITE : / DAY_VAR.
CLOSE CURSOR GI_CURSOR.
When I am in the 18-th day, I raise an event. I test the year,the calender of the system and the month (January).
This report is called in the chain, as ABAP process.
Then, in my chain , I put a local chain , that starts after it receives the event 'ZE_PROD' and the parameter for it 'ZE_PROD'.
This event with parameter has to be created in SM62.
I hope this helps and thank you all for the advices.
PS: - the factory calender method seems not to be to good, because need special rights to create it and I don' t think it's possible to customize it (because the holiday calender has to be changed and I did not see how).
Edited by: EMI on Jan 27, 2009 4:21 PM

Process chains runs 50 times in a day

Hi
is it recamonded run the process chain 50 times in a day? will be consider the case?

Hi Suneel,
I can't tell you what is best for your environment without knowing how much data you pull from the source system + how long it takes to extract + the growth ratio + most importantly the business case.
Here is an example; we extract our delivery data say 50 times a day because we have tens of thousands of deliveries a day. With that, we analyze all the deliveries; the quantity of the product, sender, receiver, state tax, city tax, carrier, milage etc etc. If you need this detailed of data and if you have 100K of records every day, I recommend you to run 50 times of your extract to "keep bringing new data" to empty your delta queue.
Take the same example above; if you're only interested in knowing what is your sender's/receiver's latest balance; you only need to "acces the limited+ same/frequently updated data", use virtual cubes.
The virtual cubes works fine with limited number of records - say no more than a million or two records in the source system.
If I were you and have time+resources I would try both.
Cheers
Tansu
Edited by: tansuaksu on Apr 26, 2009 6:17 AM

Problem in process chain running for 3 days

Similar Messages

Maybe you are looking for