DSXchange

That isn't very much information to work with at all. Please explain in more detail.

Each and every job has a node configuration at which it will run best, and this is seldom the maximum number of nodes. By "best" I don't necessarily mean fastest, this could mean 'fastest speed and smallest footprint'. You can use two basic methods to control your runtime node allocations: a) differ...

Sandeep, use your second method but create 3 columns going to a peek stage for your yyyy, mm and dd values to see if they are what you expect.

It is a simple matter of using a pipe in Server, but you could effectively do something similar in PX. Write the 100+ columns to sequential file. Then, in another job, read that file but define it with just one column and no column delimiters. You've now concatenated all columns into one.

Using kill -9 does not let the cleanup after itself. It is similar to turning off the power to your PC. You can ameliorate the damage by having the lock daemon running, it can usually remove the locks left open. In a worst case scenario the process could be writing to a DS system file and leave that...

Server has builtin methods of doing this in a simple statement, PX does not. The easiest and most efficient way of handling this is to use a CONVERT() statement. I would not recommend using a BASIC transform stage, but it is a viable and officially documented option.

kavuri wrote:...Target flat files are utilised by another product which is written in Orchestrate...

Are you sure? Orchestrate is the old name of the PX/EE product so you might as well use DataSets.

gateleys - The Chevy "Nova" was popular in the USA, but the executives couldn't figure out why the car wouldn't sell in any Spanish speaking countries... I think Craig was joking about my use of "Latin countries" so I delved into memory of Latin to try to translate "no va" from Spanish to Latin. And...

Your write speed is about 200Kb per second - I don't know your system layout so cannot comment if that is a good speed or not. Reading should be faster than writing. How are you reading this file, i.e. a Hashed file stage doing a read? If so, do a test job with just the read stage and an output sequ...

The datastage engine uses semaphores extensively. You should not have any semaphores left open once all DataStage users and processes have logged out. If user process abort or are kill -9'd then the semaphores are left open. System level timeouts can also leave semaphores open on some systems. If yo...

I wish I could recall how I got around this when it happened to me. It did have something to do with the nullability of a column generated in the query and was a simple solution (once discovered). You can deprecate the error to a warning or informational level if you can't find another solution - th...

Add a final \n to your file a.txt and see if the error wanders to file 2...

nullus vado perhaps. Doesn't sound much like a brand name.

Just like DS-sex-change, right? I still like the "Chevy No-Va" in latin markets marketing blunder.

You can rename the descriptor file to your heart's content from UNIX. It contains the pointers to the actual data files, those cannot be renamed.