DSXchange

So what you are trying to say is that First record has 4 columns Second column has null values (how many columns?) Third record has only one column? Is this your target file. How did you populate this file. Or this file is your source file? But i thought your source is SQL Server. If there is metada...

Delete the message file. Or in the properties of the bulk loader, give a new file name so that it creates a new file. Run your job for just 5 records so that the message file remains small and we can see the key messages in it.

Ok help me understand The first record is Column1: 123 Column2: 126 Column3: 789 Column4: Dear Sir/Madam, What is record 2? Does record 3 only have "Kindly help me to change the store number from S63 to S063.. Thanks Azila" as a value? What column does this value belong too. What values do the other...

DSguru2B wrote:If it still doesnt work, get your DBA involved. Fire your job and ask the DBA to monitor what is your thread doing at the database level.
Also look at the message file, it has precious info that helps in debugging and getting the load utility to work.

Did you try that

O i am sorry. I thought the job finished after 30 mins.
Didnt know the job was hung for 30 mins.
What update action are you giving in the target API stage?

With 4 million records, the job is finishing in 30 mins (approx. 7777 rows/sec). I dont see whats the problem. Keeping in mind, you are using server edition, working on a single node, i think thats acceptable. How many columns are you sending accross?

I once got this kind of error with SQL Server. It doesnt like BigInt being passed to it. I had to change that to Integer.
ArndW is right, its the sql type of one of your columns thats creating that problem. Do as he says to find out which column it is.

I think there is more to it then that. Look into your DataStage Director guide in dsbooks. It mentions over there that it depends upon what edition of DataStage you have purchased. If its developer's edition then you can do everything, from design and develop to run and compile and even delete Where...

Hi Guys, I have a situation. We have a column (exp_tp) that's the flag for the active record in the Datawarehouse (SCD 2). That column is of Sql Type: Timestamp Length : 26 Scale : 6 At the database level, that column is defaulted to 9999-12-31 23:59:59.999999. Depending upon the incoming data i hav...

Read about how to use the test unix command. Also you can look into wc -c to count the characters inside the file. If its greater than 0, then the file is not empty. You have a couple of options. You can also do find fileA.txt -size 0 Then your routine is going to be something like Ans = 0 UnixComma...

You can say that again

Isnt that what i wrote

-z is one of the options of the command test. Read about test unix command. I think it should be

where Arg1 will be the fully qualified path of file.

O got it. Thanks for the clarification. I was wondering if there is a way around it. Thanks Craig.

Isnt this a duplicate post
Use a temp table to load your hashed file into it and use the sql mentioned in your otherpost.