DSXchange

Need to know your job design. What exactly are you doing?

I would hate to break it to you but when you pass parameter values like the way you are doing, they should not have any spaces in them. Once upon a time i spent way to many hours trying to figure it out. Just for your own satisfaction, dont change anything in the script, just remove spaces from the ...

Make sure they have been defined as "String" in the job properties.
I would need to take a look at your script, if you dont mind. There must be something you are doing there.

Why dont you actually run it and see how much time its taking. Maybe it wont take that much time. Also, build the hashed file with only the columns you require.

I see you have param values with spaces in them. You need to enclose only the value in double quotes. That will work. Basically your command should look like this now.

arsheshadri wrote:Hi ,

dsjob -run -jobstatus -param 'PARAM1=VALUE' -param 'PARAM2=VALUE' PROJECT1 JOB1

Remove the quotes around 'PARAM1=VALUE'.
In other words, your dsjob command should now look like this

In this case use date set that only has one key defined and use that as a hashed file and use the other dataset with two keys defined as the driver.
By reading through your other posts i believe you have a choice in which data set can to use for lookup. Hence, this is the only other way i see.

Your error tells you what you want to know.
It tells you that TIMESTAMP is a key word and you are using it as a column name. Try to enclose it in double quotes. THat should work.
You can read post on this other website for reference.

rafidwh wrote:I gave the condition in the constraint like IsNull(keycolumn) for insert and NotIsnull for update.

The keycolumn must be from the reference link and not from the source or else it will always go down the path of Update.

SQL3039W The memory available to LOAD for DATA BUFFER prohibits full LOAD parallelism. Load parallelism of "1" will be used Do a google on the SQL code. Heres what ive found on IBM's site Ignore this message, and the LOAD will complete normally using the smaller value for LOAD parallelism. However,...

I use the DS.Tools option to unlock the processes. These are the steps I follow: 1. List.readu -- give you the Processes with usernames and jobs that are used or locked. 2. Note the PID number for the job to unlock. 3. DS.Tools 4. Option 7 5. Option 4 6. Enter the PID. 7. enter Y to confirm. If by c...

dont include the -jobstatus option along with the -run. This way you will only get either a 1 or a 0. Then you can use either Ray or Ken's suggestion to obtain your goal.

rafidwh, as requested, post your target sql to verify the date format.

if thats the case, apply an oconv to your output. You should be fine then.

Thanks Ken for the input, completely overlooked that