ORACLE 11G RAC rebuild em problem

A problem occurred in the previous period, that is, restarting a RAC node suddenly reported that the recovery area is insufficient, check that there should be no problem, has been modified to 500G, there should be more than 200 g of the remaining space, how do not, and deleted a lot of files, And with Rman in the Crosscheck synchronization control files, or not. Later check, found that the node restarted after forgetting the original modified parameters, directly into the initial parameters, no wonder has been reported that the space is not enough, the initial parameters only then a point!

After the parameter is reset, the node is up, but the problem comes too! Em into not go, check the EM state, decided to restart the next em,emctl STOP dbconsole no problem, but Emctl START dbconsole There is a problem, has been failed.

View EM logs also did not see what the problem, decided to rebuild EM.

Use command: Emca-deconfig dbcontrol db-repos drop-cluster to delete em, successively entered the database name, service name, port number, cluster name, sys password, sysman password and other information, Dbcontrol deleted, But when the database was deleted there was no movement, waiting for a large half a day and no movement had to stop.

Then embarrassed, I do not know that the EM calculation is not deleted, but the reconstruction of the place, the hint that the database already exists, should be not deleted, at least not deleted clean.

A command to delete a database is checked on the web, executed separately on the two nodes of the RAC, <oracle_home>/sysman/admin/emdrep/bin/repmanager

Later checked, decided that since is not deleted clean, then delete thoroughly point.

First check whether the directory has a residual related directory, to the <oracle_home>/oc4j/j2ee/directory to see if there are oc4j_dbconsole_

Delete emca-deconfig Dbcontrol Db-repos drop again, this time does not hang up, but still error, it looks like the listener did not start up, and feel the password input is also a problem, but also can not be rebuilt, still reported that the database existing errors. There is no problem with viewing the monitor, and it is true that EM itself has problems.

Find the information again, see a paragraph: "Reconfigure Dbconsole, you need to enter the DBSNMP password, but any password will show errors, need to be modified!", it feels like the source of all the problems!

Each rebuild is changed beforehand Dbsnmp password: sql>alter user dbsnmp identified by xxx;

Then manually delete some objects in the database that create EM:

sql> DROP public synonym Setemviewusercontext;
The synonym has been deleted.
sql> DROP public synonym mgmt_target_blackouts;
The synonym has been deleted.
sql> drop role Mgmt_user;
The role has been deleted.
sql> DROP USER Sysman CASCADE;
The user has been deleted.
sql> DROP USER Mgmt_view CASCADE;
The user has been deleted.

Delete Sysman When there are some problems, delete the times wrong, said can not delete the current connected users, using the V$session view, found that there is really three Sysman connection, may be rebuilt at the same time, in order to prevent accidents, the first Sysman users locked, in killing three processes, You can delete it after a while!

This should remove the feeling, using the command emca-config Dbcontrol db-repos create-cluster reconstruction of EM, but the result is ferocious, actually said my statement has a problem, use-help checked again, did not feel any problem ah. Is there any problem?

When using DBCA to open the configuration database products, the original EM option becomes optional, then no problem, since the command line said I have errors, I use the graphical interface head, all the way to fill the past, static structure!

But the results again told me not to think too much, at the end of the creation of the database, the error job_queue_processes generated should be greater than or equal to 1, which is how to look at the parameters, found that the parameter is actually 0. Modify the resulting: Alteer system set job_queue_processes=5 scope=both sid= ' * ';

After the modification, for the sake of security and re-use the command removed again: emca-deconfig Dbcontrol db-repos drop-cluster.

This time deleted quickly, again using DBCA Select Create EM, about half an hour later, EM correctly created successfully.

After entering the EM interface, it is found that the last backup job created through EM has not been executed in these days, and the job is gone, and sure enough, it needs to be reconfigured!

ORACLE 11G RAC rebuild em problem