Update V$RMAN_BACKUP_JOB_DETAILS view

When attempting  a 11.2 upgrade is run in message that says there is a backup job running and gives the option to continue.  I did continue but I would prefer getting rid of the condition that causes the message.   Basically the V$RMAN_BACKUP_JOB_DETAILS
view is showing a status of 'RUNNING'.  There is some dbms_???? that will make changes to the underlying table to end that, but I've forgotten.  Anyone recall what that dbms_??? is?
xoxomosAsked:
Who is Participating?
 
xoxomosAuthor Commented:
http://docs.oracle.com/cd/B28359_01/backup.111/b28270/rcmtroub.htm#BRADV89865

Contains actions relevant to the RMAN job as well as error messages generated by RMAN, the database server, and the media vendor. RMAN error messages have an RMAN-xxxxx prefix. Normal action descriptions do not have a prefix.
Note that you can execute the following PL/SQL to remove all entries from V$RMAN_STATUS:
SYS.DBMS_BACKUP_RESTORE.resetCfileSection(28);
The preceding function removes all job-related entries. No rows will be visible until new backup jobs are shown in V$RMAN_BACKUP_JOB_DETAILS.
0
 
slightwv (䄆 Netminder) Commented:
>>There is some dbms_???? that will make changes to the underlying table to end that, but I've forgotten.

Make packages change status' and various objects/columns/rows.  As for a generic package that changes anything you want, sorry, never heard of it and doubt it exists.

>>run in message that says there is a backup job running and gives the option to continue

If it says a backup is running, there is likely a backup running.

Maybe you should stop the backup?  Maybe turn off the schedule long enough to perform the upgrade.
0
 
xoxomosAuthor Commented:
The status of 'running', I believe is due to perhaps the database being taken down because the 'running' is from several days ago.  A couple 'completed' show up after if I look in the history.
I found a blog earlier this month where the author also ran into this situation while upgrading and posted the solution, dbms_???, but I did not think to save. :-(
0
Ultimate Tool Kit for Technology Solution Provider

Broken down into practical pointers and step-by-step instructions, the IT Service Excellence Tool Kit delivers expert advice for technology solution providers. Get your free copy now.

 
slightwv (䄆 Netminder) Commented:
Can you bounce the database?

See if the session might still be running and kill it.

There is some information here that might help you with that:
https://community.oracle.com/message/10708623

It looks at command_id in v$rman_backup_job_details then gets the sid from v$rman_status with the command_id.

Then uses sid in v$ssession to get the serial# to issue the kill command.
0
 
xoxomosAuthor Commented:
Thanks, yes, i saw that and I may use it if needed, but the solution I saw before
involved basically using some dbms_ that basically truncated the table underlying that
 v$rman_backup_job_details  view.
0
 
xoxomosAuthor Commented:
I do believe a bounce would resolve the problem though, but since there will be only a four hour window i'd like to find that solution again just in case bounce did not resolve.
0
 
Geert GOracle dbaCommented:
you could also query the table v$backup_async_io
select trunc(bytes/total_bytes*100, 2) percent_done, filename, status
from v$backup_async_io where status = 'IN PROGRESS';

Open in new window


if you see the percent_done changing when repeatedly querying then a backup is running
in v$session you can find which process is doing the backing up

avoiding a backup at the time for upgrade is your best option
0
 
slightwv (䄆 Netminder) Commented:
>>I saw before involved basically using some dbms_ that basically truncated the table underlying that   v$rman_backup_job_details  view.

I do not doubt that something exists that would purge the backup logs but it could also purge the entries from the RMAN catalog.

If you can find the reference again, I would take a look at it for you but I would be very careful it.

You can always reach out to Oracle Support and ask how to stop that specific job.  They are the only ones that would be able to guarantee that any system tweak would be 100% safe.
0
 
xoxomosAuthor Commented:
Nopes, bouncing the database does not work.  When I start dbua after bouncing i'm still getting that message.  :-(
0
 
slightwv (䄆 Netminder) Commented:
If tracking down the SID and trying to kill the session doesn't work, I am sort of out of ideas.

I would then suggest working with Oracle Support.
0
 
Geert GOracle dbaCommented:
edited ... removed comment
0
 
xoxomosAuthor Commented:
Finally figured I should look in my console logs.

SQL> select start_time, status from v$rman_backup_job_details where status = 'RUNNING';

START_TIM STATUS
--------- -----------------------
25-JAN-14 RUNNING
01-FEB-14 RUNNING

SQL>
SQL>
SQL> execute sys.dbms_backup_restore.resetCfileSection(28);

PL/SQL procedure successfully completed.

SQL> select start_time, status from v$rman_backup_job_details where status = 'RUNNING';

no rows selected
0
 
slightwv (䄆 Netminder) Commented:
Sorry we couldn't be more help.  Feel free to accept your last post as the answer.
0
 
xoxomosAuthor Commented:
Apology refused :-)    :-)
 I should have thought to look at my console logs days ago instead of browser history :-(
 I will accept the last part though because in this case there is actually no session actually running so none of those solutions would apply in this case.  Somehow a bit just did not get set the correct way somewhere for some at present unknown reason.
I should
0
 
xoxomosAuthor Commented:
In this case there is actually no session actually happening, no process to be cancelled so none of those solutions would apply in this case.
0
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

All Courses

From novice to tech pro — start learning today.