DSXchange

Try to recompile the job - this often will suffice. But your root problem shouldn't have occurred and the problem might not be limited to this job. Search this forum for DS.TOOLS and reindexing.

Naveen, since the sequential file has no pointer forward, you can never know if your next READ is going to reach the end-of-file. Depending upon what you want to do at the EOF you have several approaches. As John mentioned, only a fixed-length record file will let you know the number of lines - and ...

Kim - thanks for adding to the thread, I had a vague recollection of your posting the tool but couldn't remember the details. I have had enough of trying to program exits from web servers - it just never seems to work as simply as the docs lead you to believe.

Banactp,

my first thought is with the connector to the nodes - what would cause at attempt by another to connect to the same node? And my response is "you might have done something {non-kosher} with the partitioning in the stages leading up to the transformer". Did you?

saadmirza, as DataStage has a command line tool to invoke, control and report on jobs, all you would need to do is invoke your web server package exit functionality. This is different depending upon your platform and package (apache, MS, etc.) as well as configuration, but easily implemented once yo...

HSBC, I went through debugging hello trying to solve the same issue a couple of months ago. The end conclusion is that the DSAttachJob() will effectively hang for 30 minutes if you attempt to attach to your own job. It is a bug (even though one isn't allowed to attach to oneself per the handbook) th...

snasimmr,

since you are using different keys, you need different links from the hash file to the transformer stage. Draw a second reference link from the hash file stage to the Xform; add the information (again) to the hash file stage and then link one key to each lookup.

Snassimr, you mean you actually believed your DBA? If you tell your aggregator that the data is sorted then it has to be sorted. It assumes that when a group-change in your aggregation column occurs it is because it can do a group computation and doesn't have to store anything. Once a row is out of ...

Sure, no problem at all. The only issues you would get is if you read from and write to the same hash file at the same time, and those are solvable as well.

Ray,

I remember it happening once or twice to me at POCs as well and never really found a cause. Karma? Sunspots? Aliens? Who knows...

Martin,

sounds like you are on the right track. Have you tried it? Does it work?

Ray/Others, do you know how this problem comes about - something to do with the user groups? I have seen the problem before (years ago) and understand how to fix it but would like to know what causes it. Doesn't the developer role normally get granted sql privileges? I looked at our file and the use...

aditya,

the method you described is the normal way of filtering data. Perhaps you have a syntax error or similar error in your constraint. Most likely your "attribute_value" is specified incorrectly, you can check that and perhaps post your exact constraint to this thread.

-Arnd.

edward_m,

the message is telling you that your definition of columns specified n columns, but your data only contains n-1 columns (in the 16th row of data, column Field010).

Look at your source data with an editor and see what is missing on line 16.

Leo_t_nice, I put together some BASIC code to gather all the -l information for a search of the whole system into a hash file, then searched the TMP folders and removed valid dataset data files. Found over 50Gb of junked storage today! But my method is kludgy and ungainly and I'm still hoping for a ...