DSXchange

The hashed file cache is for ALL cached hashed files, not for each.

First try this and report the answer. Change directory to where DataStage is installed (the DSEngine directory). ls -al .uvconfig If there is truly no such file, and you have not corrupted the uvconfig file, then you need to regenerate the .uvconfig file using a uvregen command (or bin/uv -admin -re...

Welcome aboard. :D

If you are seeking employment, please post in the Talent forum. Otherwise, you will be best served to ask the vendor - no-one here is likely to have a comprehensive list.

It's not that at all. 999MB is the maximum that DataStage can malloc() for hashed file cache.

Put parallelism into the DataStage job. That's why I asked about Enterprise Edition. However it's also possible in server jobs - you can have multiple connections to the database through the same stage, each selecting a subset of the rows (simple example, odds and evens on a two-way split, or day of...

I disagree. First, none of my six machines had different environment variable settings. Second, APT_NO_JOBMON is a "double negative" - disabled means the job monitoring is enabled. The APT_MONITOR... environment variables affect only when monitoring happens, not whether monitoring happens. In any ca...

Once the hashed file is created, unless you need to change its shape, you don't need to create it afresh every run; clear file is very fast and will suffice. Conversely, if you do create the hashed file, clear file is a waste of time.

999MB is the upper limit for hashed file cache. This is shared by all hashed files for which the caching option is set. Read cache and write cache are separate (each maxing out at 999MB).

Not enough information. What kind of database? How are you extracting? How are you "introducing parallelism"? (We don't want to advise anything you've already tried.) Do you have Enterprise Edition? Remember that, as far as the database server is concerned, a running DataStage job is just another cl...

Get your DBA to trace inbound and outbound activity on the DB2 server. Also keep an eye on locks - though what kind of a lock persists for 10 hours is a question!

Create a script on the DataStage server. If the file names conform to a pattern you can use MGET.

Don't use the FTP stage; it tries to transfer a line at a time, and the concept of "line" in XML documents is rather vague.

"Incomplete record read from file" suggests that at least one record in the file was short. Perhaps you have an extra newline at the end, because the job read and wrote the same number of records.

I had this very issue in a training class this week. Six identically configured clients (allegedly) - four could get row counts, two could not. Connecting to the same server (though to different projects). As far as I could determine (hurriedly) there was nothing in the projects to cause the differe...

Just look at the help in the client. What's supposed to happen is what you expected. That it's not is a worry. A quick fix might be to include $DB2INSTANCE as a job parameter, with a default value of $UNSET so that, in the job, it is not set unless explicitly overridden. In this case the property in...

Try changing the project default. Open Director, choose Options from the Tools menu.