DSXchange

Yes. It failed because of the Fail on Error Property and when I set that to NO it just throws a warning message. I think I can convert that to Informational. We are doing a workaround for Tablespace issue !

Thanks for the help.

Just thought of sharing what I found out. REPLACE was failing when we run the 1st time as it expects the table to be there. So I selected REPLACE and select NO for generate SQL statement at run time and To drop the table, used a SP so it does not fail when the table is not there. Bottom line is that...

Ray - It worked. You the Man...I should have checked it after your first answer. Sorry I was lethargic...

Craig - No. It's not RCP enabled job but it still checks the Metadata. Ray's suggestion would work best for us.

Thanks all for the ideas...

SwathiCH - Problem is that whenever there is a change in the DDL, the job is going to fail the first time because of the same reason and moreover idea is to avoid the DBA communication in this case. Craig - We do have the option of running once or once per node. Ray - Table action works just fine bu...

Hi

Does Before-SQL in DB2 Connector stage works only for DML's ? We are trying to create the table in Before SQL and it throws an error saying table does not exist. It seems like before executing the Before-SQL, it looks for the table to exist and the meta data is same.

Please clarify.

Thanks

Have a question about how the data types works when we select from a table using DB2 connector. Lets say, I am simply selecting a column from the table using DB2 connector and use the same data type(Decimal(4,1)) as source system, then there is no problem but when I use functions like coalesce(max(c...

We encounter a weird issue in a Parallel job which uses a basic transformer. We have used Oconv-MCP function in basic transformer. Weirdness is that this job runs fine when we run it manually as a stand alone job but fails when run from a sequencer. It's a multi instance job...I dont understand why ...

Thats what I do but was hoping for a better solution !!

Thanks anyways for the input...

Is there a way we can make the Datastage debugger to go through the Parallel Shared Conatiners when we debug ? I think it just gives us the output but if there is an issue inside the container then it would be great to have that function.

Ray Are you tallking about APT_DEFAULT_TRANSPORT_BLOCK_SIZE parameter ? If so, it's defined as 131072. So, does the DataStage wait until it hits this size and then start moving the data ? If so, should we minimize the size so that it moves the data faster in job level ? Thanks for your time. Regards

I came across this scenario which is challenging my understanding about how the output Seq file works. I was under the impression that seq file will read data from the file and pass it to next stage in the job without waiting to import all the records into the file but it looks like it first reads/i...

DS version is 8.7 and we are aware of the $PROJDEF being set during promotion in Admin if those are new variables for that environment. These are not new variables and there are lot of jobs using the same variables which works fine. Job is being exported from Designer with the default option of (job...

Question about join performance. Assume that both ETL server and DB(UDB) server has similar configuration and the database is partitioned in two nodes. Would it be better to use the Datastage join stage with hash paritioning on the key(join) field rather than sorting in DB and using Dont sort(previo...

I am saying its corrupted as the job does not run the first time as if the parameters are corrupted during the promotion. When we delete them and add them back again and then promote it works. We usually select the $PROJDEF from the list box on the non encrypted variables and for password we just co...

I know we set values for env variables in administrator. Yes. These parms to connect to database. We put projdef as a value in the jobs so that they can pick up the right values from the admin...Issue here is that it gets corrupted for some reason...