Access problems on Excel file

Post questions here relative to DataStage Enterprise/PX Edition for such areas as Parallel job design, Parallel datasets, BuildOps, Wrappers, etc.

Moderators: chulett, rschirm, roy

Post Reply
Nagac
Premium Member
Premium Member
Posts: 127
Joined: Tue Mar 29, 2011 11:39 am
Location: India

Access problems on Excel file

Post by Nagac »

Hello

I have excel file to be processed in datastage. I used Unstructured stage to read data from Excel file but I am getting error saying:
file <xxx> doesn't exists. Specify a valid filename
I know there is file in location I mentioned in Unstructured stage. I am running DataStage job as non-dsadm user (which means I mapped different credentials in admin console), file is owned by mapped user and 770 rights on it.

Job works fine when I grant 777 on file.

Can someone advise if Unstructured stage operator is invoked by dsadm? I know it internally calls java method.
ray.wurlod
Participant
Posts: 54607
Joined: Wed Oct 23, 2002 10:52 pm
Location: Sydney, Australia
Contact:

Post by ray.wurlod »

Tell us how your Engine credentials are set up. That will inform us about the user used to execute on the Engine. Then tell us what the owner and group for the file are, and whether the credentialled user is the owner or is a member of that group.
IBM Software Services Group
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
Nagac
Premium Member
Premium Member
Posts: 127
Joined: Tue Mar 29, 2011 11:39 am
Location: India

Post by Nagac »

Thanks Ray for your response.

yes, Engine credentials were setup for every user(devproj) and same user(devproj) was the owner of the files and group is devprojects. dsadm is the member of devprojects group.

I can access the file as devproj user from unix command line.

Thanks
chulett
Charter Member
Charter Member
Posts: 43085
Joined: Tue Nov 12, 2002 4:34 pm
Location: Denver, CO

Post by chulett »

Well... you obviously have a permissions problem and we're missing some piece of information, some piece of the puzzle. With 770 the file's owner not only has full permissions, so does anyone in the file's group. The fact that that fails tells us the user your job is run by is neither, never mind the fact that you need to open the permissions up to 777 so the job can successfully read the file.

Seems to me that you need to revisit "the user used to execute on the Engine".
-craig

"You can never have too many knives" -- Logan Nine Fingers
asorrell
Posts: 1707
Joined: Fri Apr 04, 2003 2:00 pm
Location: Colleyville, Texas

Post by asorrell »

Is the file on your local file system? Or is it really on a Windows disk somewhere and you are reading it through an NFS mount?

I've had some problems with NFS in the past. In one case the problem was that some versions of NFS didn't support more than 16 groups on an id. It would only pass the first sixteen and if the group you needed for access was 17 or higher, tough luck.

Make sure your target directory is wide open and have your process create a new file there. Then check the permissions on that new file. That will at least tell you the id and primary group that is being used.
Andy Sorrell
Certified DataStage Consultant
IBM Analytics Champion 2009 - 2020
Post Reply