-1

I have seen some recommendations in the Oracle Docs about what to do if the FRA is full.

For example I have this config also: CONFIGURE ARCHIVELOG DELETION POLICY TO BACKED UP 1 TIMES TO 'SBT_TAPE'; The Databases have Dataguard and RAC. Oracle says: 1. Make more disk space - I cannot

BACKUP RECOVERY AREA - I am interrested in this option. What does it that is not the same as the standard full backup where FRA also gets cleaned? Is it faster? I see it deletes the flashback logs, maybe I need them?

If you use some other alternatives, I would be much appreciated.

Thanks in advance.

Improve this question
0

You need to design a good frecuency of your backups using rman.

If you FRA is full then you need backup database more frequently to free archivelogs unnecessary.

You could reduce the depth of your flashback logs too.

Are you store your rman backuppieces within FRA ?

If you have dataguard you could to use the slave database to backup and use the primary site only for flashback.


export NLS_DATE_FORMAT='yyyy-mm-dd hh24:mi'
$ORACLE_HOME/bin/rman << eof
CONNECT TARGET /
$retention
CONFIGURE BACKUP OPTIMIZATION OFF; # default
CONFIGURE DEFAULT DEVICE TYPE TO DISK;
CONFIGURE CONTROLFILE AUTOBACKUP ON;
CONFIGURE CONTROLFILE AUTOBACKUP FORMAT FOR DEVICE TYPE DISK TO '+DISKGROUP_FRA/%F';
CONFIGURE CONTROLFILE AUTOBACKUP FORMAT FOR DEVICE TYPE SBT_TAPE TO '%F'; 
CONFIGURE DEVICE TYPE DISK PARALLELISM 4 BACKUP TYPE TO BACKUPSET; 
'SBT_LIBRARY=oracle.disksbt,ENV=(BACKUP_DIR=$DEST_AST_RMAN_BACKUP)';
CONFIGURE CHANNEL 1 DEVICE TYPE DISK CONNECT 'connString';
CONFIGURE CHANNEL 2 DEVICE TYPE DISK CONNECT 'connString'
CONFIGURE CHANNEL 3 DEVICE TYPE DISK CONNECT 'connString'
CONFIGURE CHANNEL 4 DEVICE TYPE DISK CONNECT 'connString'
CONFIGURE DATAFILE BACKUP COPIES FOR DEVICE TYPE DISK TO 1; # default
CONFIGURE ARCHIVELOG BACKUP COPIES FOR DEVICE TYPE DISK TO 1; # default
CONFIGURE MAXSETSIZE TO UNLIMITED; # default
CONFIGURE ARCHIVELOG DELETION POLICY TO NONE; # default
CONFIGURE SNAPSHOT CONTROLFILE NAME TO '+DISKGROUP_FRA/snapshot_controlfile.f';
show all;
DELETE NOPROMPT EXPIRED BACKUP;
CROSSCHECK ARCHIVELOG ALL;
DELETE NOPROMPT EXPIRED ARCHIVELOG ALL;
BACKUP AS BACKUPSET DATABASE INCLUDE CURRENT CONTROLFILE tag=TARSYS_${fecha}  PLUS ARCHIVELOG DELETE INPUT tag=TARSYS_${fecha};
CROSSCHECK BACKUP;
DELETE NOPROMPT OBSOLETE;
DELETE NOPROMPT EXPIRED BACKUP;
CROSSCHECK ARCHIVELOG ALL;
DELETE NOPROMPT EXPIRED ARCHIVELOG ALL;

In my case $retention can be configured in my script 


backup_dest="$DEST_DISK_EXP_BACKUP"
if test "$RMAN_RETENTION_DAYS" = "null" -o "X$RMAN_RETENTION_DAYS" = "X"
   then
     retention="CONFIGURE RETENTION POLICY TO REDUNDANCY "$RMAN_RETENTION_REDUNDANCY" ;"
else
     retention="CONFIGURE RETENTION POLICY TO RECOVERY WINDOW OF "$RMAN_RETENTION_DAYS" DAYS;"
fi

Improve this answer
11
  • Hi, Thanks for the fast reply. Backups are designed by best practice and the database is backuped frequently. The Flahsback retention policy is as default. I am using backuppieces on FRA , yes. My question was, what are the best actions when the FRA is full and what kind of problems those actions can do and how do they help. Like a quick help when it happens. – ultimo_frogman Apr 26 '20 at 7:47
  • Could you backup with compress option? (I think rembember that you need enterprise, I'm not sure) – acano Apr 26 '20 at 8:09
  • You could to use - CONFIGURE CHANNEL 5 DEVICE TYPE sbt parms 'SBT_LIBRARY=oracle.disksbt,ENV=(BACKUP_DIR=/NAS)'; and then use the command BACKUP RECOVERY AREA or copy your FRA directly , but I never used this technique. – acano Apr 26 '20 at 8:13
  • When I have space problems in FRA I normally reduce my RETENTION policy and I backup more frecuencly my database – acano Apr 26 '20 at 8:14
  • 1
    Thanks. I cannot use compress. When I issue BACKUP RECOVERY AREA it just copies the FRA to the Backup Tape. This does not free up space from FRA. With retention policy you mean: ALTER RETENTION POLICY TO RECOVERY WINDOW OF 1 DAYS. And after smaller retention policy it will delete older files there ? – ultimo_frogman Apr 26 '20 at 10:12
0

You should verify that the dataguard database is current. If it fell behind, then Oracle will want to keep the flashback logs to bring the dataguard standby current. Another option is that if you have a guaranteed restore point, Oracle will maintain Flashback archive logs so that you can return to the guaranteed save point.

LIST RESTORE POINT ALL;

Monitoring Recovery Progress

Improve this answer

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.