DSXchange

It gets hung up on the first record. The director tells me that 10% processed and it doesn't move. Like I said before, when no records exist, insert is instantaneously.

I create an Index on the key field and still have performance problems. There are only 47 records in the file.

Second, I looked at the Oracle Stage and can't find a way to preserve case. It would make life much simpler.

I created a job that reads an Excel CSV file through the Sequential Stage, I transform some of the input data and use the Upsert (Update then Insert) to load the Oracle table. On the first run, the job run normally and Inserts all the records. If I run the job again, the runs very slow. I haven't be...

After I imported the metadata and modified it, I created a sequential stage and used my import to view the data. It ignore the first row. Yes, I did set the First column indicator to true when I passed the headers and false when I removed the headers from the file. I tried it both ways.

Thanks, another thing that happened was that it ignored the first row. I made sure that either I set the first row as headers or I deleted the header and turned the headers off. Both way return the same result, bypassed first row.
Is there another setting that I am overlooking?

Here we go again... I am trying to import a Excel CSV file. It contain 25 columns. When I import the file, only 24 are captured and there are discrepancies with the file types. Varchar fields are being returned as numeric and the length are all invalid. When dealing with CSV files, is it better for ...

I stand corrected. During the job run, I spooled out a dataset and the time showed up while viewing it. It must be a Oracle Stage problem.

I am using the Oracle Enterprise Stage. Setup command using the user-defined area. REGIST_DATE in Oracle is timestamp REGIST_DATE in job is timestamp I wanted to select all records > '2005-03-31 00:00:00' I got all records including the ones prior to this date. No selection difference in job. No cha...

What bothers me the most is if I can't resolve the issue, i must contact IBM and open a ticket. I wish there was an easier way.

I just changed jobs and was working with DB2 before. This is the first time I received this problem

I tried to user-define code to extract the following from an Oracle table

SELECT REGIST_DATE from REGIST_TABLE WHERE TO_CHAR(REGIST_DATE,'YYYY-MM-DD HH24:MI:SS') > '2005-03-31 00:00:00'

It doesn't select the correct records. Could someone please advise me.

Thanks,

I browse the forum for an answer and I can't seem to find one. When I use a tool other that DataStage to view timestamps on an Oracle Database table, I see the proper time. When I view it through DataStage I see 00:00:00. Is there a variable that needs to be set in order for DataStage to view the ti...

How do we enter multiple commands on the OPEN tab.

I am trying to delete records from a tables before adding new ones. I used the "OPEN" on the OPTIONS of the properties tab and entered

DELETE FROM PI_TABLE WHERE DOMAIN = 2

When I run the job, no records are deleted. IS there something I need to do.

Thanks

FYI - I found an issue that I thought the forum should know about. I was trying to access an Oracle table for update using one owner within the same schema and using another table from a different owner for input. I ask the DBA to allow the first owner SELECT ANY to the second owner. It caused the O...