DSXchange

No. You'll need to have the value before the job starts and pass it in as a Job Parameter to use it in the filename.

A 'distribution list' is just another email address, so you use it just like you would any other. Nothing special required.

If we get too far afield, yes.

You can see the parent/child relationship between the two tables in Ray's query - JOBNO is the common key. As to what's stored in the two, a LIST.DICT and some browsing should make that clearer.

They have been posted here before. I would think a diligent search for something like OLETYPE would turn up the posts you are interested in.

Routines are trumped by the presence of a Transform of the same name. So if you can't find the routine you need, check the transform branch.

No time for a proper explanation, but a search for 'routines versus transforms' should turn something up.

Whatever. I've been down this road and been bitten in the ass by this very client version. Everything works great except for one or two perfectly valid sqls. There's nothing wrong with the job or the query, but memory leaks and other problem in the client blow you up real good.

basu.ds wrote:version is 9.2.0.1.0

Old and unstable.

Is your 9i database itself that old? If not, suggest you have your DBA upgrade your client to a more 'recent' and stable version, like 9.2.05 or 9.2.0.6.

I asked for the client version - it will be something like 9.2.0.5.

Have you talked to your DBA about what might be going on?

Remove the extra dot from the end. After you are done poking around over there, don't forget to wash your hands.

In other words, you can't do a direct equivalent of your Server job in PX so I assume you'll need to break it up into two discrete jobs.

Well, I'd still stick with reading each record as one long varchar field. You can build whatever logic you need around what happens when you read a 'SEQ' record and what happens when you read an '-END' record. Stage variables can be used to build an aggregate record between those two points, and the...

A decimal data type doesn't have leading zeroes.

New problem = new post.

Which means you are not performing the same 'join' in your job, and why Ken is asking for details of your job design.