DSXchange

No, you do not need to un-install in order to re-install DataStage. I would recommend trying this first on your development server, including taking backups and making sure that they are successful.

You don't have much of a choice of stages - and that choice depends upon which database you are using. Do the 3 tables have a common key or column used?

Actually, they are pointing at different objects in the same folder, "ExecSH.B" and "DSU.ExecSH", respectively.

The DSX prefixed names are for DataStage supplied items, the DSU prefix means user-supplied ones.

Is there a file called DSU.ExecSH in the project sub-directory DSU_BP.O? If not, copy it in from the same directory in another (functioning) project.

Each operating system has some limit on the number of objects returned from a wildcard operation such as "ls *.txt" - sometimes the limits are hard and other times the limit can be modified with user settings. Can you do a ls with wildcard command on that directory without an error message (just a s...

If you use DB2 partitioning then you will automatically have PX loading each partition in parallel and that will get you the best performance. Could you post your exact error message and also insert statement when using DB2 partitioning?

Shiva, welcome to DSXChange. the -443 error with state 38553 is a internal routine error. I wonder if you have granted sufficient rights in your PX installation to the system tables. p.s. I you read the welcome "read this first post" or used search you would have seen that people using "urgent" or "...

Is your DB2 databse partitioned? Also, try replacing the target DB2 enterprise stage with a DataSet and see what your throughput is - don't bother with the numbers during the run, just use your total records divided by total elapsed time upon completion to get a baseline figure. Is it much faster th...

I'm not at a terminal with the docs right now, but could it be that 13 and 3 means total width of 13, of which 3 are decimals? What happens if you try 18.3 ?

How did you declare the numeric column data type? What happens when your source data is just a smaller number, i.e. "12.34"?

So Craig --- how was the conference? Did you get a full load of blackmail pictures from some of the seedier parts of the conference floor?

ragavendradinesh, among the things that Ray wrote in his post, he recommended that you try to use the Search function in the menu bar at the top of your screen. This should be the first place to look for answers to questions and, when used properly, is almost always sufficient in order to find an an...

I am not sure if you marked this as resolved because you found an answer or if you are still looking for one. Most likely you used a fixed-width dataype for one or both of your columns that uses the same amount of storage regardless of data contents.

Both are pretty effective in compressing files; assuming your priority is file size, not compression time. Also, I think that gzip might have 2Gb size limitations on some platforms (i.e. HP-UX) where compress doesn't.

Have you tried the default value attribute - since spaces don't equate to a valid value it should use whatever setting you specify for the default value for that column.