DSXchange

Yes, I have seen this type of problem before - it happens when calling a multi-instance job with the same instanceid that is running at that time. The first time I saw this it was in a similarly confusing scenario as yours - sometimes it would work and sometimes it wouldn't and it was difficult to f...

Sreenivasulu, when you create hashed files you have the option of creating them in the project directory or at some path. It has always been recommended to use the path option for hashed files for a number of reasons, the most important being (in my opinion): (a) your DS project directory remains se...

singhald, the message means exactly what it reported; you are doing a reference lookup but the key(s) that you used didn't return just one row but several. Since datastage uses only one value it had a to choose just one of those returned rows - in this case the last returned one. You need to change ...

Nirpendra,

please post your solution so that others might profit should they have the same problem.

Hashed files can be migrated. Without more information from you no more detail can be supplied. One version to another? One platform to another? One project to another? With or without data? Back to a previous version? Combination of some of the above? There are also several posts that searching for...

Nripendra, I am trying to help you figure out your problem. It seems that your settings for a fixed length sequential file aren't working, since you posted the question here. Doesn't it seem logical that you start off the debugging process by getting the job so that it does write to a file, and then...

So now it has been narrowed down to the sequential stage and it's settings. If you modify this to write a default type file (comma-separated, quoted strings, non-fixed length, no other attributes set and you output this to your /tmp directory; what does "wc {yourfilename}" show?

The old PX documentation is wrong, the NullToValue() won't work there. You can use Handle_Null() with the same syntax.

Nripendra,

I don't understand what you mean by the dataset giving you the same output - does that mean that the dataset is empty? Is it created? Does your Director Monitor or Designer link view show that rows are actually being written to the output link (3rd time I've asked)?

The script called in the filter stage does "things" and calls Cosort, which will consume the input and generate an output file upon input stream completion, not sending any data rows to the sequential file, this is probably triggering the sequential file error.

Nripendra, we are trying to narrow down the possible causes, but it does need to work both ways. You have just responded that the job is running fine, but it isn't. Does the Monitor or Designer at runtime show the same number of rows going out of the transformer as go in? Change your target stage to...

Senthil,

can you run your job in production but point it to the original development database and see if the error might be due to a database version or configuration issue?

Hello Mimi, The usual method of using the database mechanism of transaction to define what might be rolled back on an abort cannot be used in your case, since the jobs that needs to be rolled back might already be finished when your sequence gets an error. The is no builtin mechanism in DataStage th...

Ray,

the thread police aren't going to catch us. We're on a mission from God

That, or how about "It's 106 miles to Chicago, we've got a full tank of gas, half a pack of cigarettes, it's dark and we're wearing sunglasses. We are ready to do some serious DataStage consultation."

Hello Kaps, since you have CoSORT you should use it - it really does improve sort performance. On the other hand, I think your method of using a filter in this case isn't the optimal approach, at least initially. I would recommend starting off with writing the data to the sequential file and then us...