Basic transformer in Parallel Job

[mary]

Hi All,

We are using basic transformer in our parallel job to call a server routine.

Our job flow is like this
Sequential file --->BasicTransformer----API db2 stage

The sequential file is having one reject link.
In Basic transformer the null check and special character check is done. The special character check in the last column is done using the server routine called in the transformer.
In DB2 API stage we have a before sql to delete the record from the table with a condition. in generic sql an insertion is happening.

The problem we are facing is, when we run the job sometimes it is inserting the records and sometimes not.

Whenever it is inserting the below log we are getting
"ABAJBAMZN_NODE_ITEM(Trns_Format).ABAJBAMZN_NODE_ITEM_BDA#0.Trns_Format: DSD.StageRun Active stage finishing.
1211786 rows read from Input_records
1211786 rows written to Output_records
127.830 CPU seconds used, 4937.000 seconds elapsed."


whenever it is not inserting the below log we are getting
"ABAJBAMZN_NODE_ITEM(Trns_Format).ABAJBAMZN_NODE_ITEM_BDA#0.Trns_Format: DSD.StageRun Active stage finishing.
1211786 rows read from Input_records
0 rows written to Output_records
127.840 CPU seconds used, 5162.000 seconds elapsed.
"

Thanks In advance

[ray.wurlod]

What is different (possibly in the data) between when the inserts are successful and when they are not? Are there any other warnings (for example database not available) also logged?

[mary]

In both scenario the job is successfully finished and no warning related to Database.

We ran the same jobs in different environment. There the job got aborted .But we can see at backend the records are getting processed in the table(First records are getting deleted and then inserted)

[mary]

Another issue we faced is if the job aborts once and if we try to run the sequencer again, the job is giving the erro unable to run the job error at job level.

If we compile the routine and then run , the job is running fine.