Job Queues using Scheduler

Hi,
Can you please let us know whether we could create any job queues using Oracle Scheduler so that one job queue can be run with one license and another using a different license?
Thanks,
Sandeep.

Hi Sandy,
I am not sure what you mean but when you are on 11g you can schedule jobs to run in remote systems/databases. So I can imagine that the remote databases can be any edition, as long as you can connect to it. Maybe you should check the license information docs <http://download.oracle.com/docs/cd/E11882_01/license.112/e10594/toc.htm>
I hope this helps,
Ronald
http://ronr.blogspot.com

Similar Messages

Chaining job dependencies using Scheduler

While I see that one can assign Resource Groups to Job Classes for jobs running concurrently, is there a way to assure that one job task will not start up unless dependent jobs have finished?
What I want if for job_B to start only after job_A has finished. I do not want job_A to have to call job_B. I want there to be a tightness so that job_B will follow job_A either immediately or in a reasonable amount of time, but job_B will not fire if job_A is still running.
I can schedule job_A to run at the top of the hour. I can schedule job_B to run at 15 after the hour every 6th hour. I want to make sure however, that job_B will wait until job_A finishes in the event job_A takes longer than 14 minutes.
Does this functionality exist, or do I have to put a flag in some extra 'jobs_running' table for job_B to check, and then have it trigger job_B if it has checked, but could not run. It would be really nice if this were already built into the underlying DBA_ tables for Scheduler.
Please advise.
Thanks,
Scott Rappoport

Hi Sandy,
I am not sure what you mean but when you are on 11g you can schedule jobs to run in remote systems/databases. So I can imagine that the remote databases can be any edition, as long as you can connect to it. Maybe you should check the license information docs <http://download.oracle.com/docs/cd/E11882_01/license.112/e10594/toc.htm>
I hope this helps,
Ronald
http://ronr.blogspot.com

Job queues are not executed as scheduled

When I upgraded from 9i to 10g, I also migrated my procedures into packages. These procedures were created to be executed on a job queue. I noticed that since I've migrated them into packages, the job queue's stopped executing - it doesn't even look like it tried.
I've pasted a sample of one record taken from 10/23/2008 8:39:00AM:
LAST_DATE: 10/22/2008 8:15:03AM
NEXT_DATE: 10/23/2008 8:15:00AM
BROKEN: N
WHAT: DZSP.JOBS_UTIL.UPLOAD_PLC_2MAXIMO;
INTERVAL: TRUNC(SYSDATE+1)+8.25/24
FAILURES: 0
Your assistance is greatly appreciated!

Thanks for your reply.
Yes, I am using the DBMS_JOBS package. I was not aware that there was another scheduler - is there a difference or preference?
The setting for JOB_QUEUE_PROCESSES is set to 10, DEFAULT=NO.
Hope that helps.

I've used Project Web App on Sharepoint 2013. After clear job queue about force check-in then I've got this error. This made me can't access some pages in PWA anymore e.g. Project center, Resoure center, PWA setting, etc.

I've used Project Web App on Sharepoint 2013. After clear job queue about force check-in then I've got this error. This made me can't access some pages in PWA anymore e.g. Project center, Resoure center, PWA setting, etc. Who have ever found this problem,
pls. help.
The pop up error msg. is "Sorry, you don't have access to this page"

Hi,
According to your description, after you cleared job queue about force check-in then you can't access some pages in PWA.
Maybe you need to check in the pages which has been checked out.
If you have administrative rights, it is possible to override the check-out via the View All Site Content page:
Site Actions->View All Site Content->Pages->Hover the item you want to check in, and from the context-menu (arrow-down next to the filename), choose "Discard Check Out"
Besides, to troubleshooting the error “Sorry, you don't have access to this page”, refer to the following articles:
http://sharepoint.rackspace.com/sharepoint-2013-troubleshooting-an-access-issue-with-a-custom-master-page
http://technet.microsoft.com/en-us/library/ff758656(v=office.15).aspx
In addition, as this issue is related to project server, I suggest you create a new thread on project server, more experts will assist you:
https://social.technet.microsoft.com/Forums/projectserver/en-US/home?category=project
Best Regards,
Lisa Chen
Lisa Chen
TechNet Community Support

How to find out the user from the Jobs queue in Report server

Hello All!
I have a doubt about finding out the user from the scheduled jobs queue. Say I go ahead and schedule a report on Reports Server how can I find out the user name. When I view the jobs using showjobs I could see that the DBMS_JOBS table has a column under "Job Owner". But it invariantly shows it is "rwuser". So is there a way to find out which user has scheduled which job?
Regards
Shobha

hi,
The below tables will give only the name .
USER_ADDRS
USER_ADDR
USER_ADDRP
USR02
i think you need email address .
you can use this Tcode : su01d
and give the user name and excute it
i hope it will help you.
Ram
Edited by: Ram velanati on Jun 30, 2008 6:57 PM

How to stop a running job in 10g Scheduler?

The following is a duplicate post. I posted the following to the general database forum before seeing that otn has a new scheduler forum:
I am not able to find in the Admin Guide a method to stop a currently running instance of a job in the 10g scheduler.
In 9i, I run the following script calling DBMS_JOB.broken and DBMS_JOB.remove to shut down currently running jobs:
DECLARE
jobid NUMBER;
CURSOR c1
IS
SELECT job
FROM dba_jobs
WHERE priv_user = 'ME';
BEGIN
OPEN c1;
LOOP
FETCH c1
INTO jobid;
EXIT WHEN c1%NOTFOUND;
DBMS_JOB.broken (jobid, TRUE);
COMMIT;
DBMS_JOB.remove (jobid);
COMMIT;
END LOOP;
CLOSE c1;
END;
How may I create similar code to shut down currently running jobs using DBMS_SCHEDULER in 10g? According to the Admin Guide, disabling jobs with the force option will still allow the job to finish.
How can I terminate a running job in 10g?

You can stop a currently running job using the STOP_JOB api.
STOP_JOB Procedure
This procedure stops currently running jobs or all jobs in a job class. Any instance of the job will be stopped. After stopping the job, the state of a one-time job will be set to SUCCEEDED whereas the state of a repeating job will be set to SCHEDULED or COMPLETED depending on whether the next run of the job is scheduled.
Syntax
DBMS_SCHEDULER.STOP_JOB (
job_name IN VARCHAR2
force IN BOOLEAN DEFAULT FALSE);
Parameters
Table 83-44 STOP_JOB Procedure Parameters
Parameter Description
job_name
The name of the job or job class. Can be a comma-delimited list. For a job class, the SYS schema should be specified.
If the name of a job class is specified, the jobs that belong to that job class are stopped. The job class is not affected by this call.
force
If force is set to FALSE, the Scheduler tries to gracefully stop the job using an interrupt mechanism. This method gives control back to the slave process, which can update the status of the job in the job queue to stopped. If this fails, an error is returned.
If force is set to TRUE, the Scheduler will immediately terminate the job slave. Oracle recommends that STOP_JOB with force set to TRUE be used only after a STOP_JOB with force set to FALSE has failed.
Use of the force option requires the MANAGE SCHEDULER system privilege.
Setting force to TRUE is not supported for jobs of type executable.
Usage Notes
STOP_JOB without the force option requires that you be the owner of the job or have ALTER privileges on that job. You can also stop a job if you have the CREATE ANY JOB or MANAGE SCHEDULER privilege.
STOP_JOB with the force option requires that have the MANAGE SCHEDULER privilege.

Select Statement Takes 5 Days in Job Queue Manager

Here's our problem...
In one of our packages, there is a D-SQL select statement. Every time we schedule to run that package using job queque manager, it gets stuck. The SQL Tracing shows it's scanning the blocks, but it just won't finish. If we run the same sql statement (generated from the output) at the same time, it runs the result in a fly. Now if the package is run through sql+, it runs through just fine. So there must be a bug with running it in Job Queue.
Anyone suggestions?
TIA!!!
Helen

Here's our problem...
In one of our packages, there is a D-SQL select statement. Every time we schedule to run that package using job queque manager, it gets stuck. The SQL Tracing shows it's scanning the blocks, but it just won't finish. If we run the same sql statement (generated from the output) at the same time, it runs the result in a fly. Now if the package is run through sql+, it runs through just fine. So there must be a bug with running it in Job Queue.
Anyone suggestions?
TIA!!!
Helen

Error when attempting to maintain cube by submitting task to Job Queue

Hi,
I've been working with Cubes/Dimensions on AWM 10.2.0.3 for the last little while and have been maintaining cubes by selecting the Run maintenance task immediately in this session. This has been working reasonable well for me.
I am now looking at how this can be put into production and trying to use the Submit the task to the Job Queue selection. When I choose this option and select the Run immediately radio item I have no problems. However when I choose to run this in the future I receive the following error:
Action BUILDDATABASE failed on object OLAPBI.DUBDEV_OLAPBI
oracle.AWXML.AWException: Action BUILDDATABASE failed on object OLAPBI.DUBDEV_OLAPBI
at oracle.AWAction.BuildDatabase.Execute(BuildDatabase.java:737)
at oracle.olap.awm.wizard.awbuild.BuildWizardHelper.runBuild(BuildWizardHelper.java:275)
at oracle.olap.awm.navigator.node.modeler.cube.ModelerCubeNode.actionPerformed(ModelerCubeNode.java:480)
at javax.swing.AbstractButton.fireActionPerformed(Unknown Source)
at javax.swing.AbstractButton$ForwardActionEvents.actionPerformed(Unknown Source)
at javax.swing.DefaultButtonModel.fireActionPerformed(Unknown Source)
at javax.swing.DefaultButtonModel.setPressed(Unknown Source)
at javax.swing.AbstractButton.doClick(Unknown Source)
at javax.swing.plaf.basic.BasicMenuItemUI.doClick(Unknown Source)
at javax.swing.plaf.basic.BasicMenuItemUI$MouseInputHandler.mouseReleased(Unknown Source)
at java.awt.Component.processMouseEvent(Unknown Source)
at java.awt.Component.processEvent(Unknown Source)
at java.awt.Container.processEvent(Unknown Source)
at java.awt.Component.dispatchEventImpl(Unknown Source)
at java.awt.Container.dispatchEventImpl(Unknown Source)
at java.awt.Component.dispatchEvent(Unknown Source)
at java.awt.LightweightDispatcher.retargetMouseEvent(Unknown Source)
at java.awt.LightweightDispatcher.processMouseEvent(Unknown Source)
at java.awt.LightweightDispatcher.dispatchEvent(Unknown Source)
at java.awt.Container.dispatchEventImpl(Unknown Source)
at java.awt.Window.dispatchEventImpl(Unknown Source)
at java.awt.Component.dispatchEvent(Unknown Source)
at java.awt.EventQueue.dispatchEvent(Unknown Source)
at java.awt.EventDispatchThread.pumpOneEventForHierarchy(Unknown Source)
at java.awt.EventDispatchThread.pumpEventsForHierarchy(Unknown Source)
at java.awt.EventDispatchThread.pumpEvents(Unknown Source)
at java.awt.EventDispatchThread.pumpEvents(Unknown Source)
at java.awt.EventDispatchThread.run(Unknown Source)
Caused by: oracle.AWXML.AWException: ***Error Occured in BUILD_DRIVER:
at oracle.AWAction.BuildDatabase.Execute(BuildDatabase.java:731)
... 27 more
I am only setting the time to be a couple of minutes in the future in order to verify that it is working but I receive this error for the GLOBAL_AW schema as well as my own AW Schema.
Any ideas??
Thanks in advance,
LM

Hi LM,
I have not seen this specific issue before but there are a few things you can check:
1) Did you exit AWM before the scheduled job started? You must exit AWM before the jobs starts as the job will try to attach the AW in read-write mode which is the default mode used by AWM. Therefore, if you are still in AWM when the job starts the job will fail. Usually you will get a more helpful error message that says something like "unable to attach AW".
2) Try scheduling the job one hour in the future. Then start Enterprise Manager and look at the job queue. You should find a job with the prefix AW$ which will be your job. Look at the details for the job and make sure the program to be executed looks correct, ie it actually contains a program and a name of a cube/dimension. I created a job once, that for some reason did not contain a PL/SQL command block. No idea why.
3) If the PL/SQL block is present try running it from the command line and see what happens, login in to SQLPlus as the same user as you use for AWM.
Hope this helps
Keith Laker
Oracle EMEA Consulting
BI Blog: http://oraclebi.blogspot.com/
DM Blog: http://oracledmt.blogspot.com/
BI on Oracle: http://www.oracle.com/bi/
BI on OTN: http://www.oracle.com/technology/products/bi/
BI Samples: http://www.oracle.com/technology/products/bi/samples/

Simple Queue-Queue Failing with : kwqjaq: Only job queue slave can run prop

I know this is a simple problem, but I can't get by it.
I have two DB servers and I just want to queue a message on a local queue and have it propagated to a remote queue.
Here are my steps:
I do this on both databases:
grant dba to johnb identified by johnb ;
connect johnb/johnb
create type my_type as object ( a number, b varchar2(2000), c varchar2(2000)) ;
--- multi consumer queue
Begin
/* This queue table should already exist */
dbms_aqadm.create_queue_table( queue_table => 'MY_QUEUE_TAB',
queue_payload_type => 'MY_TYPE',
multiple_consumers => TRUE,
comment => 'Test MultiConsumer OAQ',
compatible => '8.1.0' );
DBMS_AQADM.CREATE_QUEUE (
queue_name => 'MY_QUEUE'
, queue_table => 'MY_QUEUE_TAB' );
dbms_aqadm.start_queue('MY_QUEUE') ;
End;
-- on "Local DB" I do the following:
DECLARE
subscriber sys.aq$_agent;
BEGIN
subscriber := sys.aq$_agent( 'subscriber1' , 'JOHNB.MY_QUEUE@otmdmz2' , null);
DBMS_AQADM.ADD_SUBSCRIBER(queue_name => 'JOHNB.MY_QUEUE', subscriber => subscriber);
END;
-- scheduling propagation
BEGIN
DBMS_AQADM.SCHEDULE_PROPAGATION( queue_name => 'JOHNB.MY_QUEUE' , destination => 'otmdmz2' , destination_queue => 'JOHNB.MY_QUEUE');
END;
BEGIN
DBMS_AQADM.ALTER_PROPAGATION_SCHEDULE( queue_name => 'JOHNB.MY_QUEUE' , destination => 'otmdmz2' , destination_queue => 'JOHNB.MY_QUEUE' , latency => 1 );
END;
-- I queue with this
-- enqueue local
DECLARE
enqueue_options DBMS_AQ.enqueue_options_t;
enqueue_properties DBMS_AQ.message_properties_t;
enqueue_handle RAW(16) ;
enqueue_message JOHNB.MY_TYPE;
BEGIN
enqueue_message := JOHNB.MY_TYPE(1, 'an aq test','sending to remote db' ) ;
dbms_aq.enqueue(queue_name => 'JOHNB.MY_QUEUE',
enqueue_options => enqueue_options,
message_properties => enqueue_properties,
payload => enqueue_message,
msgid => enqueue_handle) ;
commit ;
end;
No messages move to remote DB.
I can see the message in the local queue table.
Both databases have the same init.ora parameters : job_queue_processes=1000, aq_tm_processes=1.
If I attempt to run the job I get the message in the subject line in a trace file:
kwqjaq: Only job queue slave can run propagation job, exiting
Though I am sure simple configuration issue, I cannot find a relevant note.
Thanks,
John

Another case of the documentation being less than precise - it happens.
Without personally having investigated, that's my understanding as well bearing in mind the following observation/explanation in the same article which seems to apply for all except 11gR2:
Although you typically use the JOB_QUEUE_PROCESSES initialization parameter to limit
the number job queue processes for DBMS_JOB and you use the Scheduler attribute
max_job_slave_processes to limit the number of job slave processes for the Scheduler,
the Scheduler is affected by the JOB_QUEUE_PROCESSES parameter.
The maximum number of job slave processes for Scheduler is determined by the lesser
of the values of JOB_QUEUE_PROCESSES and max_job_slave_processes. For example:
If JOB_QUEUE_PROCESSES is set to 10 and max_job_slave_processes is set to 20,
the job coordinator will start no more than 10 job slave processes to be shared
between DBMS_JOB and the Scheduler.
If JOB_QUEUE_PROCESSES is 20 and max_job_slave_processes is 10, the coordinator
will start up to 20 job slave processes. The Scheduler can use only 10 of these,
but DBMS_JOB can use all 20.
If JOB_QUEUE_PROCESSES is 0, DBMS_JOB is disabled, and the maximum number of
job slave processes for Scheduler is controlled by the max_job_slave_processes
Scheduler attribute.

Job queue / Intelligent Agent / dbsnmp

Hi everybody,
we are testing Oracle8 on a Linux SuSE6.0 system with SMP.
Installing the software was ok by following the instructions on
the SuSE-website and doing some own work.
First to answer/solve a lot of questions/problems towards the
dbsnmp_start:
We also had the problems that were described before; error
message 0315 was posted in the logfiles and the dbsnmp died
immediately.
I solved the problem by installing the TCL_NEW-package of SuSE5.3
which is of version 8.0p1-11. After setting the permission to
root:dba on dbsnmp*-files the dbsnmpd starts working.
Discovering the database from our NT-machine works fine, also,
the agnet works (he can ping the db), but transmitting jobs
fails.
The error message after sending a job is "Error accessing job
queue". The job was submitted (so it seems to be) by the linux
machine but the scheduling fails.
Has anybody a solution?
Dietmar
null

Hi (again)
We had success. After spending some more time on it and a lot of
brain-(cpu)-time, we have solved the problem.
Steps I have taken:
- upgrading from 8.0.5 to 8.0.5.1 by patching (as mentioned the
mail before)
- applying the lnx_patch, too (making a link to "make" for the
command "gmake" used in patch-shellscript)
- applying the patch patch2.tgz from the download section
- copying the two libraries out of the patch2 (libclient.a and
libclntsh.so.1.0 to /lib and to ORACLE_HOME/lib
- changing permissions and ownership to oracle:dba of those
libraries
Thanks to S C Maturi for his hint on patching. It was the
basement for our success.
Dietmar
Dietmar Stein (guest) wrote:
: Thanks for the advice, but the problem remains the same. I will
: add some information I forgot in the first message:
: - using standard edition 8.0.5, upgraded to 8.0.5.1.0
: - not applied the lnx_patch (did the work on myself before)
: Another question: does the problem remain with Enterprise
Edition
: (any experiences?)?
: Dietmar
: S C Maturi (guest) wrote:
: : Dietmar,
: : If you are using the 8.0.5 Standard Edition -
: : is this happening after applying the 8.0.5.1 patch ?
: : Maturi
: : Dietmar Stein (guest) wrote:
: : : Hi everybody,
: : : we are testing Oracle8 on a Linux SuSE6.0 system with SMP.
: : : Installing the software was ok by following the
instructions
: on
: : : the SuSE-website and doing some own work.
: : : First to answer/solve a lot of questions/problems towards
the
: : : dbsnmp_start:
: : : We also had the problems that were described before; error
: : : message 0315 was posted in the logfiles and the dbsnmp died
: : : immediately.
: : : I solved the problem by installing the TCL_NEW-package of
: : SuSE5.3
: : : which is of version 8.0p1-11. After setting the permission
to
: : : root:dba on dbsnmp*-files the dbsnmpd starts working.
: : : Discovering the database from our NT-machine works fine,
: also,
: : : the agnet works (he can ping the db), but transmitting jobs
: : : fails.
: : : The error message after sending a job is "Error accessing
job
: : : queue". The job was submitted (so it seems to be) by the
: linux
: : : machine but the scheduling fails.
: : : Has anybody a solution?
: : : Dietmar
null

Running 2 dependent jobs on different schedules

Dear all,
I have the following problem:
Job1 (a chain running external jobs) runs one per day at 08:00
Job2 (a chain running external jobs) is dependent upon job1 completing and runs cyclically every 10 minutes.
I have attempted to use scheduler events but this only runs the first instanceof job2. The next instance sits there in a not started state waiting for the next event trigger (which will not happen until the next day).
I don't think I can combine the jobs and schedules and creating new procedures to accomodate this is an unwanted overhead.
I have created a workable solution using conditional sql syntax in the start rule condition of the job2 chain using data held in the all_scheduler_jobs table but I don't really think the logs were created for such a use. The conditional sql used in the rule of step 1 of job2 is:
(SELECT COUNT(*) FROM all_scheduler_jobs WHERE owner = <job_owner>'' AND job_name = ''<job_name>'' AND state = ''COMPLETED'' AND TRUNC((SELECT MAX(last_start_date) FROM all_scheduler_jobs WHERE owner = '<job_owner>'' AND job_name = '<job_name>'' AND state = 'COMPLETED')) = TRUNC(systimestamp) ) = 1
Does anyone have any alternative solutions for sucha a schedule ? Any help would be appreciated.
Thanks

Hi,
If I am understanding you correctly job2 should run ecvery 10 minutes and additionally run when job1 completes .
I would recommend splitting this up into 2 jobs - job2a and job2b . Have job2a run every 10 minutes and job2b run when job1 completes.
Here is sample code you can use to have job2b run when job1 completes (replace the job action and type with your chain)
Hope this helps,
Ravi.
-- create a table for output
create table job_output (log_date timestamp with time zone,
output varchar2(4000));
-- add an event queue subscriber for this user's messages
exec dbms_scheduler.add_event_queue_subscriber('myagent')
-- create the first job and have it raise an event whenever it completes
-- (succeeds, fails or stops)
begin
dbms_scheduler.create_job
( 'first_job', job_action =>
'insert into job_output values(systimestamp, ''first job runs'');',
job_type => 'plsql_block',
enabled => false, repeat_interval=>'freq=secondly;interval=30' ) ;
dbms_scheduler.set_attribute ( 'first_job' , 'max_runs' , 2);
dbms_scheduler.set_attribute
( 'first_job' , 'raise_events' , dbms_scheduler.job_run_completed);
end;
-- create a simple second job that runs after the first has completed
begin
dbms_scheduler.create_job('second_job',
job_type=>'plsql_block',
job_action=>
'insert into job_output values(systimestamp, ''second job runs'');',
event_condition =>
'tab.user_data.object_name = ''FIRST_JOB''',
queue_spec =>'sys.scheduler$_event_queue,myagent',
enabled=>true);
end;
-- enable the first job so it starts running
exec dbms_scheduler.enable('first_job')
-- wait until the first job has run twice
exec dbms_lock.sleep(60)
select * from job_output;

Destination disabled. []: [CrystalEnterprise.Ftp]. Please note the name of the job server used for your request and contact your system administrator to make sure the specified destination is enabled. (FWB 00031)

Hi
In BO 4.0 SP 9 when a administrator tries to schedule a report via CMC there is no error
But when a user schedules a report and the destination is FTP location -> Use default settings he gets following error
Destination disabled. []: [CrystalEnterprise.Ftp]. Please note the name of the job server used for your request and contact your system administrator to make sure the specified destination is enabled. (FWB 00031)
There is only one Job Server and the destinations are enabled in it
There is no Job server for Crystal Reports Job Server
Do i need to create it and how.

Please check if you have proper rights to schedule to FTP. You can create a new job server, whenever you schedule it, there are multiple job servers, it will handle based on the load. But it is not mandatory, depends on the load.

Email multiple spools from background job - SM36 using Spool List recipient

I create a background job using Spool List Recipient to email me the reports. The program that I scheduled to run creates 2 spools. When the spools are sent via email, only the last spool is sent.
Please advise on how I can get the 2 spools sent via email.

i would not use the Spool List recipient from Job and use the function RSPO_SPOOLJOB_TO_OFFICE in your Program instead.
tell me if it helps.
Regards,
Laurent

Job with Multiple Schedule of same time & only one is executed, Why

Hi ALL,
I've Created a package to notify report in email as excel, which is using a path to render report and save it as excel.
This package was need to be schedule every first day of month and on every Friday even at morning 7 AM.
So logically these are two events.
But if we see on 1 May 2015 we have Friday and plus first day of month
So technically this should fail.
either for IO operation for render &
storing data over same location with same file name.
To replicate this i created two three schedule
Below is the job history, job is invoked by only one schedule
Question-
1) Why one one schedule execute?
2) why "sched2" only , it could be "sched3" only or "DentalOperationStackAndRank-Sched1" Only?
Another this i result into error if different job executing same package at same schedule.
FYI- package is deployed at database level
HS

Hi HS,
As described in this
article, more than one job can run on the same schedule, and more than one schedule can apply to the same job.
And I make a test about your scenario, when configuring a SQL Server job with three schedules that reference same dates and times, my job behaves as yours that only one schedule is executed, but the job is invoked by schedule2 or schedule 3 in my environment.
It is a normal behavior of SQL Server job in my opinion. For more details about such scenario, you can track the job following this
blog.
However, when I configure three jobs that run on the same schedule, everything works well and the three jobs all run successfully.
I recommend you configure three different jobs that execute same package at same schedule .
Thanks,
Lydia Zhang
Lydia Zhang
TechNet Community Support

Can I create a table in a procedure submitted to job queue?

I have created a package (with AUTHID CURRENT_USER) where some of the procedures create temporary tables to facilitate processing. These procedures run just fine when executed directly from within an anonymous block at the SQL*PLUS level. However, when I submit the procedures to the job queue (via DBMS_JOB.SUBMIT), the job is submitted successfully but fails when run. Investigating the Alert Log shows an error of insufficient privilege on the CREATE TABLE command in the procedure.
QUESTION:
Can I create a table from a procedure running in the Job Queue? If so, then how to get it to work? Does the job run in a different environment that needs Create Table privileges set to my schema?
Thanks for any info you can provide.
John

FYI: Found the problem. In the Administrator's Guide (of course not in the supplied packages documentation about DBMS_JOB) I found:
"How Jobs Execute
SNP background processes execute jobs. To execute a job, the process creates a session to run the job.
When an SNP process runs a job, the job is run in the same environment in which it was submitted and with the owner's default privileges.
When you force a job to run using the procedure DBMS_JOB.RUN, the job is run by your user process. When your user process runs a job, it is run with your default privileges only. Privileges granted to you through roles are unavailable."
And of course we had set up our users to get all privileges through Roles, so CREATE TABLE wasn't one of my DEFAULT PRIVILEGES!
It sure would be nice if Oracle documentation could get its act together and provide ALL information about a topic in a logical place. The effort to find the information about privileges occurred after it took me 1/2 hour to figure out why my submissions were failing - I didn't have the ';' included in the quoted string for the procedure to be called - which I only figured out after looking at the code for DBMS_JOB where it made the note to make sure you include the ';'. Wouldn't it be good to have that MINOR DETAIL mentioned in the description of DBMS_JOB.SUBMIT?????

Job Queues using Scheduler

Similar Messages

Maybe you are looking for