DSXchange

Key columns in Ds should be checked when it is the part of primay Key..
Can you uncheck those in DS and test your job again..?

You can find out the error which might have caused the phantom error.. Go to the director and reset the job. It will give what exactly the error is.. Depending upon that error, you can take the preventive action.. Are you using transformer stage before loading the data into the database? When there ...

Can you give more details about your requirements ? Is everytime data is coming like '"]]CDATA]"' and you have to convert it to the "![CDATA[" ? Or it is only the first occurance of the ] needs to be converted to the !. Then you can just use convert on the rest of the string usin...

If your DS is installed on the Unix, you have to ftp the file to the unix box from where DS can read the file.

-------------

hope this helps

------------

MK

Sequential file from Server to parellel is quite different one.. First thing that you need to check is all the properties of the stage. Most importantly check delimiter, final delimiter and quote.. Most of the things get messy with the usage of these properties. If record legnth is fixed, use that p...

Yep. Maintainace is one of the things.. However basic thing in any kind of software development is to create the reusable componets... 'Why to re-invent the wheel' that is what they call it ! Anyways in this case if you say there are 1000 files that you want to process..is that the correct apporach ...

Can you check the data which you think that could have gone the other way...! If the data is present on both links then you can say that there is issue with the other way...and it is only in the 'Otherwise'..it cleary failing some criteria to go in the first link.. --------------------- Hope this he...

Hi.. There was a issue with the ds 7x versions for which it doenst aborts the job whenever there is a warning for update statments. If row is not present in the database and there is update statement to update such non existing record, there is fatal error message for that in the log, however job is...

REJECT_2_FF: Error when checking operator: Unable to initialize the exporter. What are the data type you are using in the lookup stage for the given column.. from the error, it seems that there is issue while writing to the reject file.. Use debug to check what values are coming to modify and then w...

Hi Sandeep..

Once log reports are cleared, these log files are deleted..
However if you are having the log file name, it could be restored using the unix..

-----------

Hope this helps..

-----------

MK

Why not use @INROWNUM in the transformer for target rowcount column..... @INROWNUM gives the number to each row that is passing through the tramsformer.. however the issue with this one is that if job is executing on multiple nodes, the @INROWNUM will give differenet numbers on the rows on each nod...

One more way you can achieve that is. in properties set column names 'yes', it will give you column names in the output file. In after job subroutine. If wc -l output <= 1 then touch output_file. that will create empty file if there is only one record with the columns. hope this helps.. ------------...

Check the input properties of the transformers, there is option of partitioning, in that there is option of sort 1. Stable :- This is for previously sorted datasets.. 2. Uniqe :- This will retain only first record if there are duplicates on the sorting key values. hope this helps ----------------- MK

hi..

You can use DS_JOBOBJECTS to check how many number of jobs are there.
Conenct and select number of records form the given table..that eventually gives you the number of jobs in given repository !

hope this helps

----------------

MK

If you want to the input number of records only which is from oracle table, you can write the before job routine to call the unix script. In the script count = select count(*) from the table_name update table_2 set total_input_records = count ------------------------- Hope this helps.. -------------...