DSXchange

You are being really parsimonious with the information you're giving.

So, why do YOU think the job is hanging? Is anything happening within the logic that might cause a lock? Is there any database access within the shared container?

saraswati wrote:Can anyone help me on this?

This is bit urgent.

It's not usual for me to be monitoring DSXchange at 4am. If somebody wanted to pay me handsomely for doing so, that might be another matter.

Timeout IS regarded as "failure" in the Wait For File activity. All you need are success and failure triggers. No other trigger will be honoured.

It's more likely, given what's in the error message, that the root cause is not the DataStage services at all. Read the error message carefully. Cannot connect on port 9080 means that you can not connect to the services tier - so that in all probability your WebSphere Application Server (server1) is...

By this question can we infer that you have both UNIX-based servers and Windows-based servers, and want to be able to initiate jobs on the one from the other? If not, what DO you mean?

Generating how? What are the specifications?

The IBM training materials do provide a good level of detail.

The Copy stage is not optimized out when the job is compiled - you will always see it in the generated OSH. However, the Copy stage may be optimized out when the score is composed. You can verify whether this is the case by dumping the score into the job log.

There is more than one process, if your job includes Transformer stages. You may have killed the job process but left the Transformer stage processes (DSD.StageRun) as zombies, still locking the job.

Make sure you eliminate trailing newline characters. They will be returned as field marks (@FM). Add some diagnostics to verify that the commands' outputs are being returned correctly to DataStage.

If it doesn't work, you are not managing your metadata properly. Always load table definitions into jobs and don't edit them there.

Read the warnings carefully, extract every bit of information you can from them, and change your job design so that they don't occur.

Yes, I had to learn that too, when parallel jobs came in.

That would be fairly easy. Do you want to write the entire record to this file, or just the non-printable characters?

What does this job look like when you deconstruct the container? (Show us.)

That question has been answered at least a dozen times on DSXchange.Search for "will partition despite".