DSXchange

Which database? The easiest is to use a CALL DSExecute(...) API to run a system command, such as sqlplus, isql, sqlcmd, dbaccess, etc. Just formulate the DML or use an existing script and invoke it using the command line program. Please don't do this to avoid using an ODBC/OCI stage for loading data...

Please describe the job design, stages, etc. What if it's not stuck reading the file, but writing to the target? Is there a before/after stage/job routine call that's the true problem? Viewing the data on the Sequential file proves that the definitions are valid.

Sure, switch the job at the end. The next time the jobs come back into development for enhancement or modification, you'll switch them again, and then back again. Kind of tedious, and open for error.

Sucks for development! If the current environment setting is not to your liking, it's more difficult to change when working on a job. It kind of renders that View Data option a little frustrating, because you have to keep pasting in overiding values.

Ahhh, the DB2 plugin. It works differently from other stages like Oracle and ODBC. If the datatype is DATE, it automatically uses ICONV inside the stage. The only way to prevent this is to switch the metadata to CHAR, which relies on the NLS setting of the database for the format, or TIMESTAMP, whic...

First, it's not Universe, it's the DS Engine. To enter the command line, you must have the dsenv file loaded into your user environment. Then, you need to run dssh from the DS Engine directory. The commands I posted earlier change to the appropriate directory, load the the dsenv file into your envir...

Not a genius, just been around awhile.

Hashed files can handle multiple processes reading/writing, as long as they are not accessing the same row. Avoid using the locking mechanisms as it is a bad technique.

For a Server job, yes this should work. Independent streams execute as much as possible until a shared passive stage is used to link the streams. If there's no reference stage, then the independent streams should execute to completion. The issue is this job cannot be re-ran without re-running all st...

Two or more processes cannot WRITE to a sequential file simultaneously. Consider using a parameter in the file name.

The "best" way to read small files is to simply do this: CALL DSExecute("NT", "type /here/there/yourfile.txt", ScreenOutput, ReturnCode) The variable ScreenOutput will be a dynamic array of the file contents, each line in the screen output will be a row in the array. If you're on Unix, just switch N...

If you're pulling in micro-batches all day long, you may get into some issues coordinating data across tables, meaning if inserts/updates are not packaged as a single transaction you may get into situations where some rows in a table are not synchronized with their children at th time you snapshot t...

Wow! Waiting for the right moment for your first post?

I've never seen this problem with 7.5.2 on AIX and Solaris. But, my prompts are usually the same as the parameter name so I haven't exactly been trying to test this.

You're all wrong except DSGuru2B. It's the GUI that can't display the full contents of a column. The job is fine.

The value is already in the environment. Have your script access it directly.