Simple dataguard on SE

Question

Hi.
I have set up a simple dataguard solution for one of our veritas cluster systems. We are running 10.2.0.4 on Rhel 4 x86-64.
Logs are shipped using rsync to standby node. PL/SQL and DBMS_JOBS ensures scheduled log switching and crontab jobs ensures both log shipping from primary node aswell as log apply on standby node.
In the alert log on the standby node, I can see that logs are applied correctly.
We uses db_file_name_convert and log_file_name_convert on the standby node.
Now, when I want to test a failover, I issue a alter system archive log current on primary node, shuts it down, and ships over all logs to standby node.
I do an "recover standby database auto" on the standby node and it stops when reaching the last known archive log.
So, how do I fail over.
I have tried the following:
Specify log: {<RET>=suggested | filename | AUTO | CANCEL}
CANCEL
Media recovery cancelled.
SQL> alter database commit to switchover to primary;
alter database commit to switchover to primary
ERROR at line 1:
ORA-16139: media recovery required
and
SQL> alter database open resetlogs;
alter database open resetlogs
ERROR at line 1:
ORA-01666: control file is for a standby database
Anyone who have an idea?
Thanks.
Kjell Ove

Answer 1

This implementation, in SE, is also known as the poor man's dataguard. You must make all steps manually, don't expect the switch over to perform the way it has been defined for a EE environment. You must transfer all the available archivelog files to the standby site, then stop the recovery process, activate the standby and open it just as a regular database:
ALTER DATABASE RECOVER [FROM 'pathname'] STANDBY DATABASE;
SQL> ALTER DATABASE RECOVER CANCEL;
sql> Alter database activate standby database;
sql> shutdown immediate;
sql> startup mount;
sql> alter database open read write;
~ Madrid
http://hrivera99.blogspot.com