Hello all.
Running Oracle 11.2.0.2 on Linux (64 bit)
I've a problem with Data Guard, and can't see where I've gone wrong.
PRIMARY = bvlive
PHYSICAL STANDBY = bvstby
Replicated bvstby from bvlive, set up the DG broker, and all's well. But wait, a minute or two later, I can see that
the Transport and Apply Lag start to increase.
Tried creating a table on the primary, it doesn't get to the standby.
Checked the SCN on both databases and they're different.
So, redo isn't being applied. However, if I issue a:
ALTER SYSTEM ARCHIVE LOG CURRENT;
from the Primary, the standby gets updated - only for the lag to return.
Looking at the alert log for the Standby - I can see that it gets stuck at:
Media Recovery Waiting for thread 1 sequence 100 (in transit)
but before this, there's this clue:
RFS[2]: No standby redo logfiles available for thread 1
but there are no error messages to support this.
and also:
SELECT SEQUENCE#,FIRST_TIME, NEXT_TIME, APPLIED FROM V$ARCHIVED_LOG ORDER BY SEQUENCE#;
shows that everything has been applied:
SEQUENCE# FIRST_TIM NEXT_TIME APPLIED
---------- --------- --------- ---------
87 13-SEP-12 13-SEP-12 YES
88 13-SEP-12 13-SEP-12 YES
89 13-SEP-12 13-SEP-12 YES
90 13-SEP-12 13-SEP-12 YES
91 13-SEP-12 13-SEP-12 YES
92 13-SEP-12 13-SEP-12 YES
93 13-SEP-12 13-SEP-12 YES
94 13-SEP-12 13-SEP-12 YES
95 13-SEP-12 13-SEP-12 YES
96 13-SEP-12 13-SEP-12 YES
97 13-SEP-12 13-SEP-12 YES
98 13-SEP-12 13-SEP-12 YES
99 13-SEP-12 13-SEP-12 YES
Yet the lag just grows and grows.
Here are the parameters of interest:
Primary:
<font color="red">
*.db_file_name_convert='bvstby','bvlive'
*.db_name='bvlive'
*.db_unique_name='bvlive'
*.fal_server='bvlive'
*.log_archive_config='dg_config=(bvlive,bvstby)'
*.log_archive_dest_1='location=USE_DB_RECOVERY_FILE_DEST valid_for=(all_logfiles,all_roles ) db_unique_name=bvlive'
*.log_archive_dest_2='service=bvstby async valid_for=(online_logfiles,primary_role) db_unique_name=bvstby'
*.log_archive_dest_state_1='ENABLE'
*.log_archive_dest_state_2='ENABLE'
*.log_file_name_convert='bvstby','bvlive'
*.standby_file_management='AUTO'
</font>
Standby:
<font color="blue">
*.db_file_name_convert='bvlive','bvstby'
*.db_name='bvlive'
*.db_unique_name='bvstby'
*.fal_client='bvstby'
*.fal_server='bvlive'
*.log_archive_config='dg_config=(bvlive,bvstby)'
*.log_archive_dest_1='location=USE_DB_RECOVERY_FILE_DEST','valid_for=(ALL_LOGFILES, ALL_ROLES) db_unique_name=bvstby'
*.log_archive_dest_state_1='ENABLE'
*.log_archive_dest_state_2='ENABLE'
*.log_archive_dest_state_3='ENABLE'
*.log_file_name_convert='bvlive','bvstby'
*.standby_file_management='AUTO'
</font>
I'm guessing it's something very obvious, but I've looked around and not seen anything which can shed light on this. The absence of an error message in the alert log doesn't help.
Has anyone encountered this before? Ideas/comments gratefully recieved.
Thanks,
Ray