DSXchange

Where do you find the disk usage? Are you sure that its not the case again?
Have you tried restarting Job monitor
goto $DSHOME/PXEngine.
and then run
./jobmoninit stop - to stop
./jobmoninit start - to start.

What information you pass from sequential file to STP stage?
What other information you get apart from this error?

Are you sure its sorted and Hash partitioned on the keys and maintained as SAME partition to join stage?
Could you post the list of stages used and the partition used in each stage in the job.

Thanks for your support guys. :D

dsjob as server command has -import option but not export.
BTW: why dont you have access to these commands?

What you execute "ls -l" from Execute command activity?
Else you need to use DIR with the file name, and the output need to be parsed. I dont find any straight fwd simple method as of now.

What is the output that you have got?
Are they both key are of same datatype and length? Varchar type will make most of the hidden issues resolved.

If you have MKS tool kit installed, you can use "ls -l" command and grep for the time porting and check if the file is for the current "date", if so you can process it.
You can acheive this using ExecuteCommand Activity stage.

Yes, dataset has its own conversion algorithm.
And more over the fields like varchar are assigned the full length of its max value. In order to improve the access efficiency. Where as sequential files will have the whitspace truncated.

It is a part of Design information. Not stored in Executables. Perhaps you can find the Timestamp using the dsx file.
May be atleast not a straight fwd method.

ORAOCI Load stage can be used for both local as well as remote load.
Pls refer the Plug in document.
You need to have the oracle client to be configured in the datastage server to connect the remote Oracle server.

Due to some reason, you were not able to give the full details, or perhaps you were not able to find related to what the is data type and value of data from Lookup as well as from main stream. Let me do that work for you. Let me repharse the structre of your job as, Input is sequential file, with th...

Try StringToTimestamp(DSLink77.B,"%YYYY%MM%DD") and let me know the output of the sequential file.

The function you used to convert is incomplete.
There is no time portion in the given value.
Before let you know that, could you give the value of the lookup data (key)?

What is the length given to that field while reading?
Is it delimited or fixed width?
In case of delimiter, and if its last column, the data will be truncated.
In that case you can read the data, and use the Len() function to calculate the length in Transformer and reject based on the result.