DSXchange

I know it's been documented in the forums, I just don't remember the answer. From what I recall, Kim Duke posted some code that illustrated how to check. You can see some of the information in the dsapi.h file under the $DSHOME/include directory. The Repos Information section shows: #define DSS_JOBS...

Actually, that was sb_akarmarkar, I just mentioned it again. As Ray notes, rather than continue to go round and round, the OP should post the code as it stands right now so it can be commented on directly rather than by inference.

Forgot about that option... but perhaps that's for the best.

Plus the fact that 350,000 records really isn't large enough for most changes in job design to provide significant changes in runtime.

Also verify your Server platform - is it actually Windows or is it UNIX perhaps? Sometimes people mark that incorrectly in their initial post, leading to all kinds of chases of the Wild Geeses. As noted earlier, the correct record terminator for DOS/Windows is a CR/LF pair or CHAR(13):CHAR(10) - whi...

In the Hashed File stage, 'Account' can be thought of as synonymous with 'Project'. Blank = the current Project the job is in, so it can typically be left blank when used. There are very few instances where it would be appropriate to set it to specific value, and then you'd probably want to use a jo...

If there's only six records in each file, why not post the contents of each?

You've got what, 19 parameter markers in the Insert statement and something like 27 in the Update. It doesn't work like that. They are positional markers - how many columns did you define in the stage? You are also re-using field names in the update sql, are you including the column names twice in t...

I don't understand why you would be seeing the row counts climb over the number you are actually loading into the table. However, there is typically a period of time when the counts 'stop' because the records have been loaded, but Oracle is doing its cleanup work. This typically involves updating / ...

As you've noticed, there's no DataStage specific books out there to be had. However, you are doing yourself a disservice if you don't have this book in your library.

Anyone with Director access can see if the job ran successfully. Only the user that scheduled the job can see the scheduling information.

As I said, there's no end required in a derivation if-then-else.

Do we have to guess what the error is? Where's your 'else'?

No 'end' required.

Because it doesn't really make any sense to do a lookup directly against sequential media. Load it in a hashed file and then do the lookup.