Re: Oracle job scheduler problem

Found solution to problem.

Need to issue commit after dbms_job.summit;

"Steve" <no_at_anon> wrote in message
news:1017144474.15583.0.nnrp-01.c2d92da9_at_news.demon.co.uk...
> In a nutshell jobs are not running until I adjust number of
> job_queue_processes using alter system.
>
> Details:
>
> I have the following parameters set:
>
> job_queue_processes = 3
> job_queue_interval = 60
>
> Start the database, then in a SQL*Plus session I issue a job as follows:
>
> sql> var jn number;
> sql> execute dbms_job.submit(:jn, 'DO_STATS;', sysdate + 2 / 1440,
'sysdate
> + 2 / 1440');
>
> I won't actually have do_stats running every two minutes, this is just for
> test purposes at the moment.
>
> Selecting from dba_jobs I get
>
> JOB LAST_DATE LAST_SEC THIS_DATE THIS_SEC NEXT_DATE NEXT_SEC FAILURES
WHAT
> INTERVAL
> ---- --------- -------- --------- -------- --------- -------- ---------- -

--


> ----------------- --------------------

>   41                                       26-MAR-02 11:59:59

> DO_STATS;         sysdate + 2 / 1440

>

> Current Date Time is

>

> TO_CHAR(SYSDATE,'DD-

> --------------------

> 26-mar-2002 11:58:33

>

> I wait for a few minutes:

>

> TO_CHAR(SYSDATE,'DD-

> --------------------

> 26-mar-2002 12:01:59

>

> Check dba_jobs again

>

>   JOB LAST_DATE LAST_SEC THIS_DATE THIS_SEC NEXT_DATE NEXT_SEC   FAILURES

> WHAT                 INTERVAL

> ----- --------- -------- --------- -------- --------- -------- ---------- 

--


> ------------------ --------------------

>    41                                       26-MAR-02 11:59:59

> DO_STATS;         sysdate + 2 / 1440

>

> Job has not run.

>

> Log in as System and issue

>

> sql> alter system set job_queue_processes = 4

>

> Query dba_jobs again

>

>   JOB LAST_DATE LAST_SEC THIS_DATE THIS_SEC NEXT_DATE NEXT_SEC   FAILURES

> WHAT                 INTERVAL

> ----- --------- -------- --------- -------- --------- -------- ---------- 

--


> ------------------ --------------------

>    41 26-MAR-02 12:07:18                    26-MAR-02 12:09:18          0

> DO_STATS;         sysdate + 2 / 1440

>

> Job has run and continues to do so every two minutes.

>

> Why did none of the initial 3 SNP processes pick the job up, why did I

have


> to create another SNP process?

>

> Regards

>

>