All actions in DataStage takes place in the Server. So this path of 'G:' etc must be in the DataStage server and not on your client. Also this must be visible to the user logging into DataStage. Can you please confirm this for me.
Sainath.Srinivasan wrote:Maybe you have checked this already !!
All actions in DataStage takes place in the Server. So this path of 'G:' etc must be in the DataStage server and not on your client. Also this must be visible to the user logging into DataStage. Can you please confirm this for me.
there is a mapped drive on the server that goes right to the file. It is the servers G Drive.
Im not suer what u mean by "must be visible to the user logging into DS" but i know that my id has the correct permistions to map to that file...
On windows, a scheduled job is run with the user name that is provided on the Schedule tab of the Project Properties page accessible via the Administrator client. Check if this User Name is different than your User Name. It is probably preferable to avoid using a mapped drive and use the Windows UNC name/pathing convention.
Maybe you can try the 'disapper' option for the file and see whether it traces the absence of the file immediately (as it is not able to locate the file using 'appear' option).
If your system hangs for a long time, you can deduce that this is due to some refresh property failure.
Mike wrote:On windows, a scheduled job is run with the user name that is provided on the Schedule tab of the Project Properties page accessible via the Administrator client. Check if this User Name is different than your User Name. It is probably preferable to avoid using a mapped drive and use the Windows UNC name/pathing convention.
Mike
Administrator finaly let me in and I set the schedualer ID. The job now runs. Thanks for the help guys
1. The sequencer job need to wait for a file (say, salesdata.txt) to appear in a particular directory in unix box.
2. Once the processing is done, the file is deleted from the directory.
3. The job should go into loop and wait for a new file with the SAME name to appear and continue with steps 1 to 3.
This looping should always happen starting at 9AM and ending at 10PM. Job should end at 10PM and no more processing after 10PM.
Without looping, the job will run once and finish (using filewait.)
I have to loop all the above steps after the file timeout happens or after successfully processing the file (success or failure.) i.e. In both the cases I have to loop again and wait for the file.