DSXchange

Read the job log messages, it should have told you of any issue. Make sure that the hash file didn't already exist, it won't recreate over an existing hash file. If your file is soooo big that you need a 64BIT file, why start from a minimum modulus of 1? You should upsize the hash file to a high wat...

Your job is simple, however, it is a perfect example of why simple usually doesn't equate to scalable. 40 million rows? You should spool the source data into a sequential file. You should probably use multple instances of the spool job to write multiple sequential files, each a different range of so...

It's just an internal number that is assigned to a job every time it is executed. It helps organize log messages together from a single execution of a job. You can ignore it.

You can write anonymous block stored procedures, execute ALTER commands, etc, and execute stored procedures.

From a unix command line execute: $ prstat -a Monitor your processes, see if your cpus are totally utilized. Until you can qualify your resource availabilities, you can't talk about why things are slow. If all cpus are being used, then there's little you can do to speed things up except remove cohab...

CLOB fields are supported up to 2 Gig as per Oracle OCI Yes but this is a terrible idea to move that much data on a row using DataStage. Numbers and letters are what ETL tools are meant to handle. Think a moment about moving a gigabyte of data as a single row. Your tool is written to allow you maxi...

Make sure you use the 32-bit Oracle Client with DS 7.5 and 9i.

Just mass-compile the base jobs, all the instances will disappear. Upon next call for execution, the invocation id will not be found and the instance will be automatically created.

I don't think that DS 7.5 server or client is supported or even installs on NT.

If you look at the post I referred, it subtracts two timestamps and gives the results in seconds. Second, there is no "internal timestamp" value. There is internal date and internal time. Third, if you substract two dates and have the results in seconds, the resulting number can be compared against ...

Search the forum as well next time before posting your question. You would have found that solutions were waiting for you.

http://www.dsxchange.com/viewtopic.php?t=85788

chulett wrote:Really? I don't recall ever seeing it in any of my drop-down lists... ExecSH yes, but no ExecDOS.

Check your install under Routines\Built-In\Before/After. On release 7.1 and prior Unix servers ExecDOS does not appear. On NT servers it is there. Trust us

Run the jobs that update the hash file first, then all of the other jobs can access a static hash file. Using locking mechanisms will cause a lot of headaches as jobs will stall waiting on locks, plus you probably will end up hanging your jobs when you aren't releasing locks quickly enough. This is ...

What source? How do you want to do it? Bat script? Perl script? After-job routines call to ExecDOS to run any Windoze command line program will do what you need.

Sure, just don't be writing to the hash file at the same time other jobs are accessing it expecting to the find that data. If it's a static hash file, you can have as many jobs share it as you want.