DataStage - If a DataStage job has aborted after say 1000 records, how to continue the load from 1000th record a

GeekInterview.com > Tech FAQs > DataStage

Go To First | Previous Question | Next Question

DataStage | Question 9 of 140

If a DataStage job has aborted after say 1000 records, how to continue the load from 1000th record after fixing the error?

Total Answers and Comments: 6

Last Update: May 13, 2008 Asked by: etl

Sponsored Links

Best Rated Answer

—

Submitted by: sankarganeshvel

Hi ,if your job is using OCI stage as the target stage .then u can do it. just specify the no of the record in the Transaction handling of the OCI stage . So if the job get aborted after 1000 records ,say some 1220 then 1000 record will be there in the target table

Above answer was rated as good by the following members:
ds_ng

August 25, 2006 06:39:08

Dharmendra

RE: If a DataStage job has aborted after say 1000 reco...

After fixing an error we can run the job once again by using lookup target table with hash file stage.

Is this answer useful? Yes | No

November 21, 2006 10:11:15

sachin

RE: If a DataStage job has aborted after say 1000 reco...

anyone could explain this process exactly. what i know is ...just we can limit or abort the job after complete loading of particular amount of records.

Is this answer useful? Yes | No

May 17, 2007 05:42:28

baglasumit21

RE: If a DataStage job has aborted after say 1000 reco...

It is not possible to start loading the rows after 1000th row. but there are many ways you can achieve it.
1) If the target table has current timestamp as a column then we can delete the rows having max timestamp and load all the rows from start.
2) We can perform a look-up of source table with the target table to check if the rows exist in target table and insert only the non-existing rows.

Is this answer useful? Yes | No

June 21, 2007 06:28:28

Pranav

RE: If a DataStage job has aborted after say 1000 reco...

Am not sure how you have designed the jobs. Assuming following jobs / table details :

Table : EMP_TAB
Job1: Hash file extraction from EMP_TAB ( with key details e.g. month date if its a monthly load)
Job 2: Cleanse/Transform ( Apply all business logic)
Job 3: Load job with look up on hash file from Job1 (to EMP_TAB)

Decription:
The extract job should be executed first in the sequence. For all the jobs check the option "Do no checkpoint run" except the first hash creation job.

Now in case your source had 1000 rows and the job failed on 200th row. On simpley restarting the sequence the hash file would have all the rows inserted by earlier run & the earlier jobs like job 2 would not run unecessarily. The load job would ensure that where the match is found not action is performed.

Thanks
Pranav

Is this answer useful? Yes | No

April 22, 2008 04:25:23

nagoosk

Member Since: November 2007 Contribution: 15

RE: If a DataStage job has aborted after say 1000 records, how to continue the load from 1000th record after fixing the error?

It is possible to restart the job after the aborted rows.
You have to select the "check point" property in the sequence.
There you have to give the aborted final record number.

This is only in datastage(PX)

Is this answer useful? Yes | No

May 13, 2008 03:34:43

sankarganeshvel

Member Since: May 2008 Contribution: 4

RE: If a DataStage job has aborted after say 1000 records, how to continue the load from 1000th record after fixing the error?

Hi if your job is using OCI stage as the target stage .then u can do it. just specify the no of the record in the Transaction handling of the OCI stage . So if the job get aborted after 1000 records say some 1220 then 1000 record will be there in the target table

Is this answer useful? Yes | No

1 0

Overall Rating: +1

Go To Top

User Name		Remember Me?
Password

Question Bank