Broken Jobs and

the Job Queue [ID 103349.1]

***Checked for

relevance on 11-Jul-2012***

NOTE:

=====

The discussed

functionality is still available in Oracle 10g and Oracle 11g.

However, with

Oracle 10gR1 the DBMS_SCHEDULER package was introduced with many new views.

It is advised to

use the new Oracle scheduler instead of old style Oracle jobs.

=====

The job queue is

used to schedule the execution of a stored procedure at a

specified timed

interval.  This tool is the only way the

database can be forced

to perform a

action more than once.

If you are

receiving multiple ORA-12012 errors you should check for any broken jobs.

Error:

ORA 12012

Text:

error on auto execute of job

-------

------------------------------------------------------------------------

Cause:

Some kind of error was caught while doing an automatic execute of a job.

Action: Look at the accompanying errors for

details on why the execute failed.

This can be done

by executing the following SQL statement:

SELECT JOB

FROM DBA_JOBS

WHERE BROKEN = 'Y';

-or-

SELECT FAILURES, JOB

FROM DBA_JOBS;

If this statement

returns a job with the number of failures set to 16 the job

has been broken.

Also, note the

following:

a) A job does not have to have failures = 16

to be broken. (e.g. if you break

the job manually, this does not set the #

of failures to 16)

b) A job can be broken and failures can be

greater than 16.  (e.g. if the job

automatically breaks after 16 failures and

the job is manually run and fails

again, the job will remain broken and

failures will be incremented to 17).

Once a job is

broken it no longer will be executed by the database.  If a job

returns an error

while Oracle is attempting to execute it, Oracle tries to

execute it again. The first attempt is made

after one minute, the second

attempt after two

minutes, the third after four minutes, and so on, with the

interval doubling

between each attempt. If the job fails 16 times, Oracle

automatically marks

the job as broken and no longer tries to execute it.

However, between

attempts, you have the opportunity to correct the problem

that is preventing

the job from running. This will not disturb the retry

cycle, and Oracle

will eventually attempt to run the job again.

There are two ways

to unbreak a broken job:

1. EXECUTE DBMS_JOB.RUN(JOB NUMBER);

Example:

EXECUTE DBMS_JOB.RUN(10);

This statement will force job 10 to

execute immediately.  If this is

successful the job will complete and reset

failures to 0.  This would

then flag the job as being unbroken.

2. EXECUTE DBMS_JOB.BROKEN

(JOB

=> ,

BROKEN => )

Example:

EXECUTE DBMS_JOB.BROKEN(

JOB

=> 10,

BROKEN => FALSE);

The execution of this statement would

unbreak job 10.

If the execution

of either one of these procedures returns an error saying that

the specified job

can not be found have the client log in as the owner of the

job and then

attempt the statement again.  The owner

of the job is the same as

the person who

submitted the job.

Finally, you

should always check the alert.log for any accompanying error

messages along

with looking in the background_dump_dest directory for any

generated SNP

trace files.