JIL operations normally take a few seconds to complete, but they are now taking a couple of minutes to complete. Warnings like below show up in the application server logs
[03/09/2026 10:05:25] CAUAJM_W_10998 Performance warning: Below query execution time has exceeded the DBLibThresholdTime configuration value. Elapsed time [108] seconds:
[03/09/2026 10:05:25] CAUAJM_I_18030 Event Server: <ORACLE> Completed Stored Procedure: <BEGIN :RetVal := ujo_chk_cond_pkg.ujo_chk_cond ('ALL',1, :B_job, :B_cond_job, :B_cond_type, :Done); END;>
The use case also involves constant creation of temporary jobs (with auto_delete:1) as part of business flows, so jobs are created, run and deleted, round the clock.
This stored procedure is invoked by JIL at the end of a session to warn about jobs that define a condition having a dependency on jobs that no longer exist due to deletion.
There were simultaneous JIL sessions to do the adhoc job creation and so the application server query is delaying because all these sessions were trying to obtain a full table lock on the ujo_cond_job table in order to perform dependency checks.
For the purposes of the JIL sessions that are being invoked via an adhoc script, it suffices to modify the JIL command to include the parameters "-V none" to skip the dependency check performed at the end of every JIL session. Since the calls to JIL are for the most part automated, it is not necessary to be notified of jobs with dependencies on the jobs created and deleted on the fly.
Once the adhoc JIL creation script was modified to achieve the above, performance immediately improved, no database performance warning anymore either