DSXchange

Ray, Let me explain what I am trying to achive. 1. I have two sources. 2. I am trying to process these records in to a single target table (TERADATA) through one datastage job 3. I am trying to merge these records and push them into the target table 4. There might be a case when the two sources can ...

Out of the 70 fields any one field (except for the fierst field) can change and rest of them will remain same. I do not know which field might change. Is there any other way to replace link collector? Thanks in Advance You could just write them to sequential files and concatenate them in UNIX. And b...

Thanks for the response Ray. I do understand what you are saying. IT IS RIDICULOUS:) I am not using the has file for uniqueness. I am trying to use hash file TO REPLACE Link Collector. Link Collector sometimes throws "MUTEX" error. I am trying to avoid link collector. I am trying to bring ...

Hello, We are writing records to a hash file that has 70 KEY fields. But we are unable to write data to the hash file. We get the below error message. 1.EDWSTGCsksINS_debugjob..Hashed_File_42.All_Ins_to_HASH1: ds_uvput() - Write failed for record id '500..... and 2.record ID size of 768 characters. ...

Hello, We have a data source which is a Sequential file. All the fields in that file are defined as Varchar. But in the target (teradata) database couple of fields are defined as Decimal as they are some Quantity fields. Now if a blank value come from the source file the datastage job throws and err...

I need the where clause because somewhere down the line I have to use "connect by prior" function to travese a hierarchy. I can not traverse the hierachy if I do not specify a single value in the WHERE clause. But i still dont understand why you need the where condition if you want all the...

I have a SQL statement that has a Where Clause. Which means it will return the result for the matching value in the where clause. Ex select * from Sale Where SALE_TSP >= nvl((SELECT max(begin_effective_dt) FROM sale_history WHERE sale_type_code in ('5','6','7','8','A') AND invoice_id = ( SELECT invo...

Hello, I have a Query (In OraOCI8) that takes a single value to provide the result. For Example: Select * from EMP where EMP_ID=1001. But in EMP Table I have 100000 employees. How to get the result set from the same query for all the employees. Please note that I can not take out the WHERE clause as...

Got you. I changed the query to USER_ID=:1 AND SALES_RCRD_TSP >= TO_DATE (:11,'YYYY-MM-DD HH24:MI:SS') AND SALES_RCRD_TSP <= TO_DATE (:12,'YYYY-MM-DD HH24:MI:SS') but still I am getting the same error. bapajju, if are doing two different date comparisons on your SALES_RCRD_TSP column and are AND'ing...

Could you please explain a bit more. I am unable to understand the problem. Your query, USER_ID=:1 AND SALES_RCRD_TSP = TO_DATE (:11,'YYYY-MM-DD HH24:MI:SS') AND SALES_RCRD_TSP = TO_DATE (:12,'YYYY-MM-DD HH24:MI:SS') , even if correct, will never return any values. What happens when you do a "v...

I have the reject link, but the job aborts and it does not write anything to the reject link. Somewhere you are passing in a column value for updating that does not match the datatype. I suggest you check your Reject link output file (you did have one, right? Best practice tip) and see which column ...

Hello, I am using one SQl query that has a where claus. In the where clause, I have a condition that looks like this: Where USER_ID=:1 AND SALES_RCRD_TSP = TO_DATE (:11,'YYYY-MM-DD HH24:MI:SS') AND SALES_RCRD_TSP = TO_DATE (:12,'YYYY-MM-DD HH24:MI:SS') But while running the job, I am gettin the erro...

Hello, I am extracting data from a Single target and populating the data in to five different targets. Is there any way I can capture the record counts (processed) for the individual links and populate a file with the Link name and the number of rows processed by the links. Please let me know. Thank...

Craig, Noe I am facing some other type of problem. Now We want to run the successding sequence even if the preceeding sequence finishes with warnings. But if the preceeding sequence aborts then only we do not want to run the succeeding sequence. Now when I put "Otherwise" in the second lin...

Thanks Ray. That's a great idea.

ray.wurlod wrote:Do the inserts and the updates separately. Use a lookup to determine whether the row already exists; choose one of two output links on this basis.
You might even consider directing these links to text files ready for bulk loading into the table.