Multiload restartability

Post questions here relative to DataStage Enterprise/PX Edition for such areas as Parallel job design, Parallel datasets, BuildOps, Wrappers, etc.

Moderators: chulett, rschirm, roy

Post Reply
dsdeveloper13
Participant
Posts: 10
Joined: Thu Aug 28, 2008 12:40 am

Multiload restartability

Post by dsdeveloper13 »

Hi,

I have a parallel job in 7.5.1A, that is loading into a Teradata table using a MultiLoad stage. if the job aborts in the middle of loading the data, can this be rerun in such a way that the multiload step picks up where it left off? I am currently having to delete the work tables before re-submitting the job.

Thanks.
richdhan
Premium Member
Premium Member
Posts: 364
Joined: Thu Feb 12, 2004 12:24 am

Post by richdhan »

Hi,

Check the Teradata multiload report file which you would have specified in the Teradata Multiload stage. There will be 5 phases of operation. Based on the phase of operation multiload might lock your table.

You also would have mentioned the log table, Work table and Error table.

Yes, you would have to drop all these tables.

Issue the below command
RELEASE MLOAD <table_name>
because multiload might have locked your table

Again re-run the job. If the job fails

1. Check the report file
2. Drop the tables
3. Issue the release command and re-run again

HTH
--Rich
Last edited by richdhan on Wed Feb 11, 2009 2:30 am, edited 1 time in total.
dsdeveloper13
Participant
Posts: 10
Joined: Thu Aug 28, 2008 12:40 am

Post by dsdeveloper13 »

Thanks for your reply.
This solves the problem for doing a re-run if the multiload had stopped in Acquisition phase. What if the load is in Apply phase? Multiload cannot be re-run. Is that correct?
Is there a way to re-start the mload where it might have left off? Say for example, using the Unix file instead of the named pipe?

Thanks.
Post Reply