DSXchange

Oooo... very clever, Mr Bond. I wonder if that will do the trick?

Yes... you can use as many as you need.

And yes, seventh - I counted incorrectly.

Use the Field() function. Extract the 6th "/" delimited field.

Are you planning on upgrading some day? I can't imagine there would be any patches nor support for the 6.x release. Neither of the Technotes mention that version.

OK, then you will definitely need something 'procedural' if you think intermittent commits are the answer. Have you talked to your DBA about possibly increasing the log sizes?

Logs are logs, switching to Parallel jobs won't help in that regard, actually they log more messages so would probably make it worse. How many instances of the MI jobs do you run simultaneously?

Only if you write some SQL PL and call it Before SQL. You're not deleting all records, are you?

As an example, you may need to create a delimited list of all your value file names and use that to drive your loop iterations.

How are you doing it right now with the MI jobs and the value file assigned to each? Shouldn't be all that different, I would think. Unfortunately, I'm not anywhere I can test / build anything that might help.

I don't see why not. Have you built a looping Sequence job before?

Have you talked to your DB2 DBA? That's where I would start in your shoes. Also, have you contacted your official support provider?

It does, I'm afraid. Scroll down to the Site/Forum forum and read other people's experience before you start to worry too much.

Good? Good!

Right - it will start purging them the next time it runs, assuming the job does not fail. Oh, and by 'let us know' I meant check a job via the Director to see what the current Auto Purge settings are that it has, no need to run the job just to check that.

I'm fairly certain that version will take your project level Auto Purge settings and propagate them down to all jobs automagically. Give it a try, let us know!