Standby database not applying archivelogs, manually apply after registering

Question

Hi
I have a small problem with physical standby db.
Standby db was created and it was working fine, was applying logs with no problem. One day I had to switch it to read only mode, it stayed in this mode for quiet a while. Then there was a problem with space for archive logs. I fixed this, copied missing logs, registered them and these were applied.
And then database stopped on the sequence which was automatically registered by primary database.
v$managed_standby:
PROCESS STATUS CLIENT_P SEQUENCE# DELAY_MINS BLOCK#
ARCH CLOSING ARCH 19667 0 126977
ARCH CLOSING ARCH 19668 0 133121
MRP0 WAIT_FOR_LOG N/A 19600 0 0
As you can see there is WAIT_FOR_LOG sequence 19600 which should be applied, which is in the directory, to which oracle user has rights to read etc, etc.
Only way to force database to apply this log is to manually register it, but I have to add "or replace", because file is already registered.
/path/dbsid1_19600_668777138.log 2 19600 YES NO NO A 28-OCT-09 28-OCT-09
alter database register or replace physical logfile '/path/dbsid1_19600_668777138.log';
After this I have:
/path/dbsid_19600_668777138.log 2 19600 YES NO NO A 28-OCT-09 28-OCT-09
/path/dbsid_19600_668777138.log 0 19600 YES YES NO A 29-OCT-09 28-OCT-09
Registering this file causes applying at once.
In pfile on primary:
log_archive_dest_2 string SERVICE=DRSTDB2 ARCH DELAY=2880
Pay attention - destination 2.
Question is obvious - why logs are not applied automatically?
Why logs ARE applying manually but the DEST_ID is set to 0?
There was no major structure change in primary, besides - logs are applied after all. Bouncing the database doesn't give me any good, switching to readonly and back to recovery mode neither.
Please can you help? I can build this standby again but this is not a solution.
Any additional info on request.
Regards
Jarek Jozwik
Edited by: user11281267 on 30-Oct-2009 06:15

Answer 1

You haven't given any usefuly information on your problem.
show parameter log_archive_dest
show parameter fal
show parameter dg
What errors are in your alert logs ?
What command are you using to recover ?
Check the contents of v$archived_log
# run on primary to detect failures :-
select destination, status, fail_date, valid_now
from v$archive_dest
where status != 'VALID' or VALID_NOW != 'YES';
# run on standby to get exact position of rollforward :-
select thread#, to_char(snapshot_time,'dd-mon-yyyy:hh24:mi'),
to_char(applied_time,'dd-mon-yyyy:hh24:mi'),
to_char(newest_time,'dd-mon-yyyy:hh24:mi') from V$STANDBY_APPLY_SNAPSHOT;
Are you using dataguard broker ?