Dataset stage: Input-file data set does not exist

[vish08]

I have couple of jobs.
Job 1: reads data from Oracle and writes to DB2 & Dataset
Job2: read that dataset and update Oracle tabel

These jobs runs one after another every hour. through scheduler


Problem: After running for 4 hours, job2 is failing with error"
/home/dsadm/Ascential/DataStage/DataSets/eng/Ord: Data set initialization for "/home/dsadm/Ascential/DataStage/DataSets/eng/Ord": This input-file data set does not exist: /home/dsadm/Ascential/DataStage/DataSets/eng/Ord.

Any help appreciated

[ArndW]

It is difficult, if not impossible, to help with the data given. Your job reads a dataset located at "home/dsadm/Ascential/DataStage/DataSets/eng/Ord" which does not exist. That would lead one to assume that something went wrong with Job 1; or that Job 1 wrote to a dataset, but one located elsewhere on your system.

[vish08]

The jobs are running fine for 4 to 6 hours automatically through schduler. But after that time, job 2 is failing.

Read Oracle tabel and write to DB2. At the same time update a field in Source(Oracle tabel) after succcesful write into DB2.
So JOB1:
Read Oracle tabel and write to DB2 and Dataset
Job2:
read that Dataset and update a field in Oracle tabel.

[chulett]

All you really did is repeat yourself, which is not all that helpful.

How exactly does this 'scheduler' run 'one after another'? Without more specifics I'm guessing that at some point one or the other starts taking more than an hour and you get some kind of 'overlap' where they run out of sequence.

[etldwh_techie2050]

As mentioned chulett & ArndW...
Can you check if the job1 has produced the desired Dataset at the location "home/dsadm/Ascential/DataStage/DataSets/eng/Ord" prompted by error?

Investigate why job1 could not create Dataset resulting sequencer & job2 failure...

[Sainath.Srinivasan]

Check your job log for run timings and check whether there is any overlap.

Also check the datetime in the dataset files.

[sureshreddy2009]

Hi
Check first job1 has finished successfully,if it has not finished successfully, then dataset will give error for rest of the jobs, even though it already created for previous runs.if the job is successfull then see whether that is created at some other lcoation, at last go to firstjob manually change the dataset name as sample.ds something like that and run one by one and see the results.

[vish08]

As chulett suggested, I added a script in Job2 to delay for a minute.
So far from last 13 hrs, it is running fine.

[sjordery]

For anyone else that gets this message and finds this thread via a search, the cause of this error message for me was running a job that was writing to datasetA whilst concurrently trying to read datasetA in another job!