DSXchange

ray.wurlod

Show us the DataStage job design that you used, particularly the column metadata on both links.

ray.wurlod

Scott wrote:Is there a better way, or is this best practice to, as the subject says, "clear logs files in one shot"?

Use the auto-purge settings. Maybe with tighter rules than you currently have in place. That's what I'd regard as best practice.

ray.wurlod

You can create external VBA script to run Excel macros, and you can execute those from DataStage.

Running them from DataStage on UNIX introduces more interesting problems, however. You need to enable some form of remote execution capability and use that.

ray.wurlod

I doubt that U solved the problem. If U had solved the problem then U would have posted the solution. The second person personal pronoun in English is spelled "you", not "u". Please strive for a professional standard of written English on DSXchange, not least because it aids thos...

ray.wurlod

Code: Select all

-param PSName=ParamSetName

where PSName is the name of the parameter set in the Parameters grid and ParamSetName is the name of the Parameter Set in the Repository (these will usually be the same).

ray.wurlod

That's the values files that form part of the Parameter Set. But even so it has to be done prior to use. The only solution of which I am aware is to use a hierarchy of sequences.

ray.wurlod

This question is unrelated to the thread that you've hijacked. Please begin a new thread. Or just consult the manuals.

ray.wurlod

It's up to you whether you remove the DS Users or not. Denying them roles via the DataStage Administrator client is all you need to do.

ray.wurlod

Output column derivations are all of the form

Code: Select all

If InLink.Col1 = value Then value Else SetNull()

I'm assuming that the output columns are decimal, not string, so that "empty" would not be a vaild value, hence my use of SetNull().

ray.wurlod

No partition symbol is shown if stages on both ends of the link execute in sequential mode.

ray.wurlod

ArndW wrote:I expected 3 entries in your output, not many pages.

There was a quote character missing around "\...".

Equivalently you could have used DataStage/SQL:

Code: Select all

SELECT CATEGORY FMT '64L', COUNT(*) COL.HDG 'Count' FMT '5R' FROM DS_JOBS WHERE NAME NOT LIKE '\%' GROUP BY CATEGORY EXPLAIN;

ray.wurlod

1. Lookup stage. Lookup failed rule = "Fail".
2. Allow lookup stage to retrieve multiple rows from reference input. (The rest of your requirement happens automatically.)

ray.wurlod

Sreenivasulu wrote:Agreed that i deviated from the question. Sorry for that. What i as trying to convey is that doing a scheduling with Director is not as secure as doing in shell scripts.

That's what the cron.allow and at.allow files are for.

ray.wurlod

Give some thought to how "00120", "0120", "01999" and "0200" sort as strings - you DID say that the data type is VarChar.

ray.wurlod

More thought required...

(The answer, in the interim, is "probably", but nothing is leaping into the forebrain just yet.) The idea about IPC stage or simply inter-process row buffering is probably useful.

Search found 53125 matches