DSXchange

Hi all,

for the 3rd time this month we encouter a strange problem:

Using the Oracle connector stage in 8.7FP1 to wite in Bulk mode,
column names are occasionally truncated resulting in the "The input schema column CASE_S is not used and the connector is configured to stop when unused columns are detected in the input schema"

The column name defined in the Columns tab is CASE_SERVICE_DOMAIN (as it is on db-level...) So it seems to be truncated to CASE_S when generating the SQL

We obviously do not want to set the Drop unmatched columns, since this will result in data errors.

We haven't been able to catch the error in Manual mode yet. No specific before or after operations are done, oracle data cache is disabled

Rerunning the job works ok, but the initial failures are impacting our onvernight loads...

any clues (PMR will be logged at IBM also) ?

William

We are now experiencing similar behaviour in a lookup stage:

Caught ORCHESTRATE exception in parallel process: APT_ParseError: Expected identifier, got: DA

The actual column defined in de lookup is DAY...
Seems te be truncated to DA somehow...

William

You seem to be firmly in IBM's hands at this point... no clue why it would be doing anything of the sort.

APAR JR44271 did the job.

William