DSXchange

Okay, I turned off the premium content so you could see the full message. If your Unix admin did not do something mean and setup 32 BIT file systems, then you should have no issue. If your files are failing to "import", meaning you're try to read or at least view data on the file using the Sequentia...

The largest file you can create is usually limited by the amount of free space on the volume. As Craig said, records are not a concept in sequential files, total character size is your concern. Some operating systems (and their administrators) can limit you based on how they configure the volumes, s...

Here's a friendly piece of advice. You are ignoring what has been told to you. This is a bad habit, because you are ignorant of something and Craig has nicely tried to explain it to you twice. The filter command is supposed to return SCREEN OUTPUT as streaming rows to be redirected to the internal p...

You're using this in a Routine stage? Are you sure the SDK is installed in your project?

Change your default padding to a space (032).

The easiest and 100% sure solution is to re-install the software after making sure your projects have been exported and backed up at the Unix level.

I don't think so. DIRECT path is great for loading data, it just sucks with the indexes afterwards. I guess you're driving at boosting load performance. DIRECT path is a three-stage load, meaning indexes and constraints are disabled after locking the partition (or table), data is loaded, constraint ...

It's using DIRECT path loading option into named partitions under the covers. So, a locally managed index only contains the data within each given partition. Since each partition will load independent of the others, locally managed indexes are optimal. If your data only hits 1 of 96 partitions, only...

Great! Please re-post a Cygwin compliant version with appropriate commentary! Thanks!

Okay, but are you naming the row to update by primary key, or are your updates wildcard? Meaning, one row from your job will affect thousands of rows in the target?

Cat them together using unix cat and then use command line FTP. Your whole problem is that you are using an FTP stage that is meant to remotely read data files when you just need to transfer files. Use the command line version of FTP and all your problems will go away. Your missing clue that you're ...

Any triggers doing some behind the scenes work?

Well if the columns don't match in files, how does concatenation work? Are you mixing record types? Even so, just have the FTP reach each file as a single column rather than using column definitions. If delimited, just use a delimiter value that won't be found in your data and declare a single colum...

If you don't care for performance, why not use a link collector and funnel the rows from both FTP stages into a single output file?

Reasons jobs "stall" at the end of the job: 1. Full job logs automatically purging 2. After-stage or after-job routines going out and doing something 3. After-stage or after-job routines in two jobs locking same object (database table, sequential file, etc) 4. Commit is issued, database doing it's t...