First, is at least one instance available?

[oracle]$ srvctl status database -d <db_unique_name>

ORA-01034: ORACLE not available

ORA-27101: shared memory realm does not exist

Linux Error: 2: No such file or directory

Your database is not up, check the status of database.

srvctl status database -d <dbname>

user10674559 wrote:

 

srvctl status database -d orcl

Instance orcl1 is running on node host01

Instance orcl2 is running on node host02

Instance orcl3 is running on node host03

Connect to any instance as sysdba and show the results of:

[oracle]$ export ORACLE_SID=orcl1

[oracle]$ sqlplus /nolog

SQL> connect / as sysdba

SQL> show parameters area

SQL> select name, round(space_limit/1048576),round(space_used/1048576), round((round(space_used/1048576)*100)/(round(space_limit/1048576))) pct_used

from  v$RECOVERY_FILE_DEST;

SQL>SELECT * FROM   V$RECOVERY_AREA_USAGE;

Regards,

Juan M

user10674559 wrote:

 

Thanks Juan M and rchem for update. Here is what you asked:

 

export ORACLE_SID=orcl1

sqlplus /nolog

 

 

SQL*Plus: Release 11.2.0.3.0 Production on Fri Jul 1 08:28:02 2016

 

Copyright (c) 1982, 2011, Oracle.  All rights reserved.

 

SQL> connect / as sysdba

Connected.

SQL> show parameters area

 

 

NAME                                 TYPE        VALUE

------------------------------------ ----------- ------------------------------

bitmap_merge_area_size               integer     1048576

create_bitmap_area_size              integer     8388608

hash_area_size                       integer     131072

sort_area_retained_size              integer     0

sort_area_size                       integer     65536

workarea_size_policy                 string      AUTO

 

SQL>  select name, round(space_limit/1048576),round(space_used/1048576), round((round(space_used/1048576)*100)/(round(space_limit/1048576))) pct_used

from  v$RECOVERY_FILE_DEST;  2

 

NAME

--------------------------------------------------------------------------------

ROUND(SPACE_LIMIT/1048576) ROUND(SPACE_USED/1048576)   PCT_USED

-------------------------- ------------------------- ----------

+FRA

                      2000                      1980         99

 

SQL>SELECT * FROM   V$RECOVERY_AREA_USAGE;

FILE_TYPE            PERCENT_SPACE_USED PERCENT_SPACE_RECLAIMABLE NUMBER_OF_FILES

-------------------- ------------------ ------------------------- ---------------

CONTROL FILE                         .9                         0               1

REDO LOG                           15.3                         0               6

ARCHIVED LOG                       19.7                        .2              13

BACKUP PIECE                         63                         0               6

IMAGE COPY                            0                         0               0

FLASHBACK LOG                         0                         0               0

FOREIGN ARCHIVED LOG                  0                         0               0

 

 

7 rows selected.

Hi,

If you have space in +FRA asm disk group, try increasing the FRA like:

SQL>ALTER SYSTEM SET db_recovery_file_dest_size=10G SCOPE=BOTH SID='*';

Then, connect to RMAN and purge some archived redo logs (because you said "RAC to practice").

RMAN> DELETE NOPROMPT ARCHIVELOG ALL COMPLETED BEFORE 'TRUNC(SYSDATE)';

Then, work in how to backup your database and to administer the FRA space usage.

(edited entry):

Also your current backups of database use 63% of your FRA, you can drop the backups if not needed or make the db_recovery_file_dest_size bigger.

Regards,

Juan M

user10674559 wrote:

 

so what are other ways to free up space in FRA other than RMAN?

There is none.

You can configure your RMAN environment in order to help you to identify what files are candidates to delete and free space.

This could be done at automatic way managed by Oracle. BUT you have to make a proper configuration.

I advise to spend time reading carefully the following documentation:

Maintaining RMAN Backups and Repository Records - 11g Release 2 (11.2)

Cheers,

Juan M

I think you need to stop the database, start it in mount, resize the db_recovery_file_dest_size parameter and alter database open.. later you should backup your archlogs like backup as compressed backupset archivelog all delete input format '/tmp/to_disk/%d_%U.bck' ;

and the last step would be to delete archivelog all backed up 1 times to device type disk completed before 'SYSDATE-1' (I have 30 days of archlogs in my FRA, the DB is not producing a lot of archlogs)