DSXchange

kennyapril wrote:My data stage server occupies more space when ever I run the jobs which results in disk space errors always

why is this happening?


can anyone provide a solution for this situation?

OMG they've killed Kenny!

This is where the title "IBM Information Champion (InfoSphere)" comes in handy!

Did you look at asbnode_install.log file? It looks like a problem registering (into XMETA?). This is one where you really should involve your official support provider. The Java stack traces aren't much use to common mortals who don't have access to documentation about what is - or is supposed to be...

What you have done is proved that the slowness is not within DataStage (or at least that, whatever it is, it's obscuring any performance issue in DataStage). The problem is at source. Is your SAP implemented on SQL Server? How did you determine which columns to index?

Pick up the rejects in the Sequential File stage with a reject link, reparse it (Column Import stage perhaps), generate the default value for the missing field (Column Generator stage perhaps) and funnel the two sets of rows back together.

Change your configuration file to specify resource and scratch disk on other file systems where there is more space.

Read the file using a single VarChar field and parse it in a downstream Transformer stage, wherein you can handle the missing column appropriately (supply a default value?).

Do you have the associated COBOL file definition (FD) file?

If so import from that. If not, import the table definition for the Sequential File.

Try again, with the arguments to the Convert() function in the correct order.


This was my error. I have edited the original post so that the function is now correct.

venkatdev wrote:U can find the max(REF_DT) befor joing based on the key column ..

U may not wish to do so.

The second person personal pronoun in English is spelled "you". U is one of our posters (it's a fairly common Burmese name).

I always try to create jobs that don't abort. I prefer to retain control over what's happening.

Here are the matching pattern-match tokens. 1N - numeric 1N:@VM:" " - optional numeric 1A - alphabetic 1A:@VM:" " - optional alphabetic Testing for alphanumeric is more complex using pattern matching, but you can Convert() alphanumeric characters to "" and work with wha...

Once the domain name, user name and password have been entered, you will get a drop-down list of DataStage server/project combinations managed by that domain. Choose your DataStage project from that list.

I've done three of them, all without problems.