DSXchange

Couldn't have answered better

Hi guys, Did you know you could read from ODBC with parallel method : 1) Set partition column in the ODBC stage using a numerical field that is ideally a key. 2) edit the SQL clause and add 'WHERE %orchmodColumn%' 3) In the next stage which dispatches data onto nodes, use the Modulus method on the s...

Hey guys, I'm loading a 3 300 000 rows table containing invoices. I have two modes : 1) Full loading 2) Daily Loading For the daily loading mode I have 2 fields containing the creation date and the update date in the source date. Thus I can filter only records created or updated after the last load...

Hey guys, I'm loading a 3 300 000 rows table containing invoices. I have two modes : 1) Full loading 2) Daily Loading For the daily loading mode I have 2 fields containing the creation date and the update date in the source date. Thus I can filter only records created or updated after the last load....

Hi, In one of the ETL job we are getting the Fatal Error as : Fatal Error:Not a list value ;table. I was not able to figure out,what was the exact error in the ETL job. About the job design: Basically we are reading the Data from Database table and doing 6 look ups including Range look up on some o...

John Smith wrote:Contact IBM for patch ; it's a bug in the client software.

I have a workaround :

When the message appears, go to the stage tab. The message should keep on popping out. Keep pressing ENTER. At some point, DS lets you go out and when you come back in, the change is recorded.

That would not be a hex or binary viewer. The goal is to get the numeric representation of what is in that character. ... IBM's answer : CCSID 500 is a EBCDIC charset. Our application will not be able to work with this. You have to convert to a equivalent ASCII charset using the DD driver. Since CC...

Whoa algfr! You're out in the weeds. Why do you think you have to use current timestamp for a type 2 effective date? I think I've never used that derivation, so I know it's not a stage requirement. Let's go back to the derivation that you would like to use and post the error message that you're get...

This solution implies that the previous SK is kept.

What I did in the end was performing an update statement to replace the oldest begin data by 01/01/1901.

I think they should add this option in the next version because it's a common need.

whats being used as upsert mode, insert then update or update then insert? what happens if you lower the array size/transaction size/commit interval. Oracle places a row level lock during insert/update. seems like update query is waiting for the inserted record to commit and then its updating the s...

Also this happens only with Type 2 !

Seems the Inert and Update statement that takes place are blocking each other

algfr wrote:

Sainath.Srinivasan wrote:Database level integrity constraints and commit intervals.

How is that ? Do you mean there is a configuration to be done ? Please explain.

There seems to be both updates and inserts in the same time, thus causing locks. When the data volume is lower or the initial table is empty, it runs fine.

Exactly,

When I get to see what takes place in the oracle sessions, it creates two sessions blocking each other.

There is just one when I use an open command.

Can be worked out of course, but still, it's weird !

Did you try it ? I did and it creates me two Oracle sessions.

When I use an Open command, it works fine.

Just a quick question.

When I perform a close command and am running on 2 nodes, is this normal that the close command perform for EACH node and not only once after the load is completed ?
I use Oracle

Thanks