DSXchange

Hi,

You can use 'drop' in your schema file at column level:
Eric

Hi, As it shows, your user credentials need to be linked to the machine's root account, since 'others' don't have write permissions. Or you can modify project's permissions to allow other users than root to write into those folders. It might not be a good idea to create a project using root privileg...

Could you try to put the commands in a Before-job subroutine ExecSH? and add a ";echo TEST" at the end Like: ls /data/finrecon/SBIRECON/`echo "#$JpBusinessDate#" | tr -d '-'`/GLCCCTA*;echo TEST And give us the result like: testSF_CommandLine..BeforeJob (ExecSH): Executed command:...

Where did you put the command line? It seems that you are using a filter command instead of a file pattern command. You should have Source=>Read Method=>File Pattern Source=>File Pattern=>the command line I gave you adapted to your needs No filter can be used (at least in v8.1) when you select File ...

What exact command line did you used?

Eric

You should use it in the file path (it will retrieve file names and pass it as input files list). Filter property is used to pre-parse your data using an external program (somewhat like external source stage could be used). PATH1 is the first part of your file path, and PATH2 the second part of it (...

Using an external source stage is one option, otherwise you can use a bash command in the file pattern expression of the sequential file stage, like find #PATH1#/`echo "#THEGIVENDATE#" | tr -d '-'`/#PATH2# -name "*#FILEPATTERN#*" or ls #PATH1#/`echo "#THEGIVENDATE#" | t...

Hi,

Maybe you can use an external source stage using an unix command like
You might also be able to call a script with parameters which could echo the file content as you like.

Eric

Hi,

Try this example, which will show you how DS optimize or not the copy operation.

I suppose that you copy operator is not optimized since it is partitionning the data (only parallel stage in your job) and is the only active stage?

Check also this answer from Ray

Eric

It's good to know you have found a working solution. It seems that your command line was waiting for an input (provided here by /dev/null). As stated in the comment, if your shell doesn't have disown, it probably doesn't need it. And note that with the trap command executed before, you don't need to...

Hi, You might try this one (if your shell supports disown) trap "true" HUP # Do nothing but run "true" on SIGHUP # runs a shell function in the background, with its files redirected # the same way nohup does. & puts it in the background. # 'disown' tells the shell not to wait...

Hi,

Usually QC database would be Oracle or MS SQL Server, so you should be able to query those db with the correct connector.

Regards,
Eric

For the insert part you should be able to do it in before sql statement. For the update part, you need first to gather info on total record count and error record count before calling your procedure. It cannot be done in one job, you need to design your job to reject records on failure, then get the...

Hi,

You can write PL/SQL code into after sql statement in oracle connector. Thus you can put your sql call directly there. Normally if your load fails, the after sql statement will be triggered anyway as stated in this technote :
http://www-01.ibm.com/support/docview.w ... wg21577490

Eric

Hi,

Did you check the schema definition at runtime (APT_PRINT_SCHEMAS) to see if any truncation appears at runtime?

Are you sure that the entire part of you number is small enough to fit into the target (and if there is any blank or zero leading the value)?

Eric