DSXchange

Oops - it wasn't an intended thread drift.

"BO" for Business Objects? Where are you seeing this error message?

Ignore DataStage for the moment. If you can't get it to work with MSQUERY then you won't get it to work in DataStage.

When i try reading data from excel using msquery i do not see any DSNs i have created.

Fix this first. Did you specify a system DSN?

What is your column "FIELD" declared as? Do you get the buffer overrun error on several rows, or perhaps just the last row in the file?

Well, then how did you define it in the UV stage? with which connection?

You don't have access to hashed files in PX/EE, so you would need to do something such as write to a database table without buffering and immediate commit and then use a sparse lookup elsewhere in the job from that table. Performance will be miserable, so it might make sense to revisit your job desi...

shamshad - how long did the job then take to process the FTP'd file? You need to add those two times to make a comparison. I know the DataStage FTP stage is slower, but I like to use the FTP stage on big files with slow connections. Using FTP you would need to copy the whole file over before process...

This might be a case where you can think about using a DataStage server job - this would depend upon a number of factors that we don't know (data volume, performance requirements, company policies, etc.), but if you were to ask for help on a DS/BASIC routine to do this you would get a lot of immedia...

Are you certain that the hashed file exists (define it in a job, use "view data" in a Hashed file stage)? If so, how did you declare it as a UV stage?

Lawrence, you have just attempted to resurrect a 2-year old thread. It would be better if you started a new server thread, showing you list string and the ereplace that you are having problems with.

So you are doing an overwrite and the previous delete didn't complete. Does this have something to do with a recent thread regarding deleting problems? Try renaming the fileset in the job and try to run it (twice, so it does a force delete and overwrite). Does that work for you?

Yee Haah.

I seem to remember that much of the error message is printed out when you have bad userid/password/database parameters.

ds_sai,

you might be better served by ignoring sanath1234's post. The levels of parallelism you choose for PX/EE have nothing to do with a "standalone" system. Even a 1-CPU system may benefit from multiple virtual nodes. You, not the system, choose how to configure your PX/EE.

It doesn't look like a lookup fileset. How did you create this set? And what are you doing to it when you get the error posted?