DSXchange

srinivas.g wrote:Pleae dont use sort stage.if you see merge stage itself we have sort option.please use that...

What reason can you give for that advice?

Of course they can. But do they? My only possible answer is a definite "maybe". Chances are that your untutored SQL is not as optimized as that generated by the DB2 Enterprise stage. If you happen to have created identical SQL, it would have no impact on performance, whatever you mean by &...

My error.

The functions I gave were for parallel jobs.
Sainath's answer is correct. Let DataStage generate the SQL.

That depends on what stage type you're using. Use a Transformer stage to assign the value of the job parameter to a column. Apply Oconv() if appropriate.

Performance statistics is unrelated. You need to verify the successful execution of every function in your routine. Each will return a negative integer if there's a problem. Try also changing the second argument of DSAttachJob() from DSJ.ERRFATAL to DSJ.ERRNONE.

Welcome aboard.

If you want a date, the function is CurrentDate(). If you want a timestamp the function is CurrentTimestamp().

Neither takes any argument but nonetheless requires its parentheses to indicate that it's a function.

Job1 can save the start date and end date in that job's user status area. When the sequence requests the run of Job2, the start date and end date can be supplied from Job1's user status area ($UserStatus activity variable). Likewise Job2 can be supplied with end date from Job1's user status area ($U...

Job 2 is not in a runnable state. Either it aborted when you ran it before, or perhaps it has subsequently been changed and not yet recompiled.

Please be more specific about what you mean by "it is not working" and what - precisely - you want to do with inbound NULL values. It should be straightforward once you have decided on how NULL is to be represented in the not null column - simply convert any inbound ("out of band"...

I have a vague recollection that a doubled semi-colon is required (;;). Search DSXchange.

Why not use SQL Server replication? IBM's CDC product (formerly Data Mirror) would also be a solution if your budget could run to it.

Welcome aboard. I think your diagnosis is correct. Is the DataStage project visible from the second machine? That's where the compiled Transformer code is. If the project is not visible, you need to propagate the code to all machines in the configuration - this is explained in (from memory) the Para...

vemisr wrote:suggestions plz

Sending your mail to Germany would involve PLZ (Postleitzahl).

Try enabling RCP and letting DataStage figure out how to write the output - write to a text file with column headings and reverse-engineer from that.

Provided that many things are true, such as the db2nodes configuration file being up to date, partitioning information being up to date in DB2 and so on, the DB2 Connector (and, indeed, the DB2 Enterprise stage) will execute independently on each partition. From memory you are also obliged to use th...