If a database is cloned or recreated on another machine with the same name, the new database will not be discovered on the GRID.
There is a reason behind this .
Database called test already discovered on GRID – database on ABC machine.
Create or clone a database with the same name test on another machine XYZ.
Try to discover this new database on the GRID using the manual option or the option to monitor via grid control while creating the database.
The new database test will not be discovered on the GRID and cannot be seen in the list of targets on the GRID .
That is because a databse with the same name the test database has also has been discovered and runs from another box.
Only when a manual install was attempted and the name of the instance was issued as test.domain instead of just test that it worked .
As you can see below – the repository ( sysman in emrep database ) has a table called mgmt_targets which maintains these details and does not allow duplicates( so a little tweaking is necessary )
The agent tried to upload test automatically when installed on the new box and start it up- and it failed to do so as this entry exists in the table.
Currently both databases are disovered – run this command in the emrep ( repository database) as sysman.
SQL> select target_name, from mgmt_targets where target_name like ‘%TEST%’;