Hello list, I have a strange problem when trying to create a job using the DBMS_SCHEDULER package: I verified that the job name does not exists using the following statements: (All the following statements have been run using the same Oracle user: MY_USER) select count(*) from all_scheduler_jobs where upper(job_name) = 'MY_JOBNAME'; ==> returns 0 select count(*) from all_scheduler_running_jobs where upper(job_name) = 'MY_JOBNAME'; ==> returns 0 select count(*) from all_jobs where upper(what) like '%MY_JOBNAME%' ==> returns 0 select count(*) from all_scheduler_chains where upper(chain_name) = 'MY_JOBNAME'; ==> returns 0 select count(*) from all_scheduler_groups where upper(group_name) = 'MY_JOBNAME'; ==> returns 0 select count(*) from all_objects where upper(object_name) = 'MY_JOBNAME'; ==> returns 0 execute DBMS_SCHEDULER.drop_job(job_name => 'MY_JOBNAME', force => true); ==> ORA-27475: "MY_USER.MY_JOBNAME" must be a job But when I try to create a job with that name using: execute DBMS_SCHEDULER.create_job(job_name => 'MY_JOBNAME', job_type => 'PLSQL_BLOCK', job_action => 'begin ... end;', enabled => TRUE); I get the following error: ORA-27477: "MY_USER.MY_JOBNAME" already exists If I use a different job name (e.g. MY_JOBNAMEX) I can create and drop the job without problems. Does anyone have an idea on where else I could look for an object with that name that prevents creating the job? I am using Oracle 11.2.0.3 on CentOS Thanks in advance Thomas -- //www.freelists.org/webpage/oracle-l