Dataguard Broker - Swithover failed

Question

I have a two node RAC(n1,n2) with one physical standby(n3)
I have enabled dataguard broker and did a swithover with one instance up in RAC(n1 up)
Swithover and switchback have worked fine.
But when trying to do a swithover from the second RAC(n2 up) node it failed and now all the db's are in standby mode.
When tyring to bring up the RAC instances, alert log shows as standby instances.
Physical Standby Database mounted.
Completed: ALTER DATABASE MOUNT
ALTER DATABASE OPEN
From sql:
SQL> select switchover_status from v$database;
SWITCHOVER_STATUS
SESSIONS ACTIVE
On standby:
SQL> select switchover_status from v$database;
SWITCHOVER_STATUS
TO PRIMARY
Has anyone come accross this situation? How to bring it back to the Primary instance mode?

Answer 1

On-Standby
1. alter database commit to switchover to physical standby with session shutdown;
2. shutdown immediate;
3. startup nomount;
4. alter database mount standby database;
5. alter system set log_archive_dest_state_2=defer;
6. recover managed standby database disconnect;
On-Primary
1. alter database commit to switchover to primary;
2. alter system set log_archive_dest_stat_2=enable;
3. shutdown immediate;
4. startup;