Before physical deployment standby with RMAN incremental backups.

Similar Questions

• Roll forward standby with RMAN incremental backup when used DG broker

  Hello
  
  I'm about to use the process defined in the note: steps to roll forward a database ensures the RMAN incremental backup using during primaries and Eve are in the file [836986.1 ID] ASM system. I'm on 11.1.0.7.
  
  However, it does not mention the DataGuard broker.
  Did someone played this process when you use the broker.
  The broker he manages properly?
  Does it matter if the broker is upwards or downwards during the physical recovery of standby or replacement of control file?
  Should the broker configuration be re-created?
  
  Thank you.

  user11993454 wrote:
  Hello

  I'm about to use the process defined in the note: steps to roll forward a database ensures the RMAN incremental backup using during primaries and Eve are in the file [836986.1 ID] ASM system. I'm on 11.1.0.7.

  However, it does not mention the DataGuard broker.
  Did someone played this process when you use the broker.
  The broker he manages properly?
  Does it matter if the broker is upwards or downwards during the physical recovery of standby or replacement of control file?
  Should the broker configuration be re-created?

  Thank you.

  (1) disable the broker
  (2) perform the Roll forard SNA.
  (3) select the broker

• Roll forward standby database with the incremental backup, when a data file is deleted in primary education

  Hello

  I'm nologging operations + deleting some files in the primary and you want to roll forward the day before using the incremental backup Yvert.

  I do in particular, as the files are dropped?

  I got to meet ( Doc ID 1531031.1 ) which explains how to roll forward when a data file is added.

  If I follow the same steps, to make the move to restore the data file newly added, will it work in my case?

  Can someone please clarify?

  Thank you

  San

  I was wondering if reocover noredo is performed before restored controlfile, oracle will apply the incremental backup error-free files, and in this case, what would be the status of the data file in the control file.

  Why do you consider to retrieve the day before first and then in the restaurant of the controlfile will lead to problems. Please read my first post on this thread - I had clearly mentioned that you would not face problems if you go with the method of deployment.

  Here is a demo for you with force logging is disabled. For the first time the day before resuming and restored then the controlfile ensures:

  Primary: oraprim

  Standby: orastb

  Tablespace DataFile of MYTS is removed on primary:

  SYS @ oraprim > select force_logging in the database of v$.

  FORCE_LOGGING

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

  NO.

  Currently the tablespace is to have 2 data files.

  SYS @ oraprim > select file_name in dba_data_files where nom_tablespace = 'MYTS;

  FILE_NAME

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

  /U01/app/Oracle/oradata/oraprim/myts01.dbf

  /U01/app/Oracle/oradata/oraprim/myts02.dbf

  In standby mode, the tablespace is to have 2 data files:

  SYS @ orastb > select name from v$ datafile where ts #= 6;

  NAME

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

  /U01/app/Oracle/oradata/orastb/myts01.dbf

  /U01/app/Oracle/oradata/orastb/myts02.dbf

  Postponement of the day before on the primary log shipping

  SYS @ oraprim > alter system set log_archive_dest_state_3 = delay;

  Modified system.

  Dropped 1 MYTS datafile on the primary.

  SYS @ oraprim > alter tablespace myts drop datafile ' / u01/app/oracle/oradata/oraprim/myts02.dbf';

  Tablespace altered.

  Removed some archives to create a space.

  [oracle@ora12c-1 2016_01_05] $ rm - rf * 31 *.

  [oracle@ora12c-1 2016_01_05] $ ls - lrt

  13696 total

  -rw - r - 1 oracle oinstall 10534400 5 January 18:46 o1_mf_1_302_c8qjl3t7_.arc

  -rw - r - 1 oracle oinstall 2714624 5 January 18:47 o1_mf_1_303_c8qjmhpq_.arc

  -rw - r - 1 oracle oinstall 526336 5 January 18:49 o1_mf_1_304_c8qjp7sb_.arc

  -rw - r - 1 oracle oinstall 23552 5 January 18:49 o1_mf_1_305_c8qjpsmh_.arc

  -rw - r - 1 oracle oinstall 53760 5 January 18:50 o1_mf_1_306_c8qjsfqo_.arc

  -rw - r - 1 oracle oinstall 14336 Jan 5 18:51 o1_mf_1_307_c8qjt9rh_.arc

  -rw - r - 1 oracle oinstall 1024 5 January 18:53 o1_mf_1_309_c8qjxt4z_.arc

  -rw - r - 1 oracle oinstall 110592 5 January 18:53 o1_mf_1_308_c8qjxt34_.arc

  [oracle@ora12c-1 2016_01_05] $

  Current main MYTS data files:

  SYS @ oraprim > select file_name in dba_data_files where nom_tablespace = 'MYTS;

  FILE_NAME

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

  /U01/app/Oracle/oradata/oraprim/myts01.dbf

  Current data of MYTS standby files:

  SYS @ orastb > select name from v$ datafile where ts #= 6;

  NAME

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

  /U01/app/Oracle/oradata/orastb/myts01.dbf

  /U01/app/Oracle/oradata/orastb/myts02.dbf

  Gap is created:

  SYS @ orastb > select the process, status, sequence # v$ managed_standby;

  STATUS OF PROCESS SEQUENCE #.

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

  ARCH. CLOSING 319

  ARCH. CLOSING 311

  CONNECTED ARCH 0

  ARCH. CLOSING 310

  MRP0 WAIT_FOR_GAP 312

  RFS IDLE 0

  RFS IDLE 0

  RFS IDLE 0

  RFS IDLE 320

  9 selected lines.

  Backup incremental RMAN is taken elementary school.

  RMAN > incremental backup of the format of database of SNA 2686263 ' / u02/bkp/%d_inc_%U.bak';

  From backup 5 January 16

  using the control file of the target instead of recovery catalog database

  the DISC 2 channel configuration is ignored

  the DISC 3 channel configuration is ignored

  configuration for DISK 4 channel is ignored

  allocated channel: ORA_DISK_1

  channel ORA_DISK_1: SID = 41 type device = DISK

  channel ORA_DISK_1: starting full datafile from backup set

  channel ORA_DISK_1: specifying datafile (s) in the backup set

  Enter a number of file datafile = 00001 name=/u01/app/oracle/oradata/oraprim/system01.dbf

  Enter a number of file datafile = name=/u01/app/oracle/oradata/oraprim/sysaux01.dbf 00003

  Enter a number of file datafile = name=/u01/app/oracle/oradata/oraprim/undotbs01.dbf 00004

  Enter a number of file datafile = name=/u01/app/oracle/oradata/oraprim/users01.dbf 00006

  Enter a number of file datafile = name=/u01/app/oracle/oradata/oraprim/myts01.dbf 00057

  channel ORA_DISK_1: starting total, 1-January 5, 16

  channel ORA_DISK_1: finished piece 1-January 5, 16

  piece handle=/u02/bkp/ORAPRIM_inc_42qqkmaq_1_1.bak tag = TAG20160105T190016 comment = NONE

  channel ORA_DISK_1: complete set of backups, time: 00:00:02

  Backup finished on 5 January 16

  Saved controlfile on primary:

  RMAN > backup current controlfile to Eve format ' / u02/bkp/ctl.ctl';

  Cancel recovery in standby mode:

  SYS @ orastb > alter database recover managed standby database cancel;

  Database altered.

  Recover the day before by using the above backup items

  RMAN > recover database noredo;

  From pick up to 5 January 16

  the DISC 2 channel configuration is ignored

  the DISC 3 channel configuration is ignored

  configuration for DISK 4 channel is ignored

  allocated channel: ORA_DISK_1

  channel ORA_DISK_1: SID = 26 type of device = DISK

  channel ORA_DISK_1: from additional data file from the restore backup set

  channel ORA_DISK_1: specifying datafile (s) to restore from backup set

  destination for the restoration of the data file 00001: /u01/app/oracle/oradata/orastb/system01.dbf

  destination for the restoration of the data file 00003: /u01/app/oracle/oradata/orastb/sysaux01.dbf

  destination for the restoration of the data file 00004: /u01/app/oracle/oradata/orastb/undotbs01.dbf

  destination for the restoration of the data file 00006: /u01/app/oracle/oradata/orastb/users01.dbf

  destination for the restoration of the data file 00057: /u01/app/oracle/oradata/orastb/myts01.dbf

  channel ORA_DISK_1: backup /u02/bkp/ORAPRIM_inc_3uqqkma0_1_1.bak piece reading

  channel ORA_DISK_1: room handle=/u02/bkp/ORAPRIM_inc_3uqqkma0_1_1.bak tag = TAG20160105T190016

  channel ORA_DISK_1: restored the backup part 1

  channel ORA_DISK_1: restore complete, duration: 00:00:01

  Finished recover to 5 January 16

  Restored the controlfile and mounted the day before:

  RMAN > shutdown immediate

  dismounted database

  Instance Oracle to close

  RMAN > startup nomount

  connected to the database target (not started)

  Oracle instance started

  Total System Global Area 939495424 bytes

  Bytes of size 2295080 fixed

  348130008 variable size bytes

  583008256 of database buffers bytes

  Redo buffers 6062080 bytes

  RMAN > restore controlfile eve of ' / u02/ctl.ctl ';

  From 5 January 16 restore

  allocated channel: ORA_DISK_1

  channel ORA_DISK_1: SID = 20 type of device = DISK

  channel ORA_DISK_1: restore the control file

  channel ORA_DISK_1: restore complete, duration: 00:00:01

  output file name=/u01/app/oracle/oradata/orastb/control01.ctl

  output file name=/u01/app/oracle/fast_recovery_area/orastb/control02.ctl

  Finished restore at 5 January 16

  RMAN > change the editing of the database;

  Statement processed

  output channel: ORA_DISK_1

  Now the data file does not exist on the standby mode:

  SYS @ orastb > alter database recover managed standby database disconnect;

  Database altered.

  SYS @ orastb > select the process, status, sequence # v$ managed_standby;

  STATUS OF PROCESS SEQUENCE #.

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

  CONNECTED ARCH 0

  CONNECTED ARCH 0

  CONNECTED ARCH 0

  ARCH. CLOSING 329

  RFS IDLE 0

  RFS IDLE 330

  RFS IDLE 0

  MRP0 APPLYING_LOG 330

  8 selected lines.

  SYS @ orastb > select name from v$ datafile where ts #= 6;

  NAME

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

  /U01/app/Oracle/oradata/orastb/myts01.dbf

  Hope that gives you a clear picture. You can use this to roll forward day before using the SNA roll forward Eve physical database using RMAN incremental backup | Shivananda Rao

  -Jonathan Rolland

• Problem creating physical standby with RMAN

  Hi guru

  Perhaps this impact that all face you when creating physical standby DB. I'm trying to create a physical standby database Using RMAN double order a server

  (pri machine) machine on standby.

  Here are the steps I followed to create the above:

  Step 1:-activate forced logging

  SQL > ALTER DATABASE FORCE LOGGING;

  Step 2:-set up a standby Redo Log

  SQL > ALTER DATABASE ADD STANDBY LOGFILE GROUP 10

  2 > ('/ oracle/dbs/log1c.rdo','/oracle/dbs/log2c.rdo') SIZE 500 M;

  Step 3:-set the parameters of primary database initialization

  SQL > create pfile='?/dbs/pfileorcl.ora' of spfile;

  Edit the pfile to add the sleep settings, shown here:

  db_unique_name = "orcl".

  LOG_ARCHIVE_CONFIG = 'DG_CONFIG = (ORCL, orcldr)'

  LOG_ARCHIVE_DEST_2 = "SERVICE = orcldr LGWR ASYNC VALID_FOR =(ONLINE_LOGFILES,PRIMARY_ROLE) DB_UNIQUE_NAME = orcldr'"

  * .fal_server = orcldr

  * .fal_client = orcl

  * .standby_file_management = auto

  Step 4:-activate archiving

  SQL > SHUTDOWN IMMEDIATE;

  SQL > STARTUP MOUNT;

  SQL > ALTER DATABASE ARCHIVELOG;

  SQL > ALTER DATABASE OPEN;

  Step 5:-tnsnames for eve of the configuration database

  This should be done on the primary database to edit tnsnames.ora or using the command NetCA and create by the name orcldr

  orcldr =

  (DESCRIPTION =

  (ADDRESS = (PROTOCOL = TCP) (HOST = Eve-svr)(PORT = 1521))

  (CONNECT_DATA =

  (SERVER = DEDICATED)

  (SERVICE_NAME = orcldr)

  )

  )

  Step 6:-backup of database and copy it to standby mode

  This backup script must be run on primary and copied to sleep on the same mount point, run these scripts with RMAN, all files in /rman_backup need to be copied in /rman_backup on the standby server.

  $rman target.

  RMAN > run

  {

  allocate channel c1 type disk;

  allocate channel c2 type disk;

  backup of database format ' / rman_backup/%U';

  ARCHIVELOG backup all the format ' / rman_backup %U ';

  backup current controlfile to Eve format ' / rman_backup/%U';

  }

  Step 7:-availability of measures database

  Installation of Oracle software

  This should be the same version and patch group with exactly same ORACLE_HOME mount point.

  Commissioning of the listener

  Create and start a listener on the database of sleep by using NetCA, or create listener.ora in $ORACLE_HOME/network/admin

  LISTENER =

  (DESCRIPTION =

  (ADDRESS = (PROTOCOL = TCP) (HOST = Eve-svr)(PORT = 1521))

  )

  Step 8:-set of backup database initialization parameters

  Copy the pfile file created in the database primary it renames in initorcldr.ora and change these settings:

  db_unique_name = "orcldr".

  LOG_ARCHIVE_CONFIG = 'DG_CONFIG = (orcldr, ORCL)'

  LOG_ARCHIVE_DEST_2 = "SERVICE = orcl LGWR ASYNC VALID_FOR =(ONLINE_LOGFILES,PRIMARY_ROLE) DB_UNIQUE_NAME = orcl'"

  * .fal_server = orcl

  * .fal_client = orcldr

  *. LOG_ARCHIVE_DEST_STATE_2 = 'ENABLE '.

  Step 9:-installation tnsnames for the primary database

  This should be done on the pending change tnsnames.ora or using the command NetCA database and create it in the name orcl

  ORCL =

  (DESCRIPTION =

  (ADDRESS = (PROTOCOL = TCP) (HOST = primary-svr)(PORT = 1521))

  (CONNECT_DATA =

  (SERVER = DEDICATED)

  (SERVICE_NAME = orcl)

  )

  )

  Step 10:-copy a password file

  A password file must be created on the primary and copied to the standby site. The sys password must be the same on both sites. Copy orapworcl in $ORACLE_HOME/dbs and rename the file in orapworcldr

  Step 11:-create for Oracle mount points

  Mount point must be created on the standby server with the same convention used in primary education, which includes the location for controlfiles, redo logs, data files, destination of the newspaper Archives and alert logs.

  $ mkdir Pei/u01/app/oracle/oradata/orcl /.

  $ mkdir Pei/u01/app/oracle/flash_recovery_area /.

  $ mkdir Pei/u01/app/oracle/admin/orcl/adump

  $ mkdir Pei/u01/app/oracle/admin/orcl/bdump

  $ mkdir Pei/u01/app/oracle/admin/orcl/cdump

  $ mkdir Pei/u01/app/oracle/admin/orcl/udump

  Step 12:-use RMAN to restore the backup and standby configuration

  Connecting to RMAN and run the following command to create the database pending, this must be performed on the standby server after the backup copy and the tnsnames parameter.

  $ export ORACLE_SID = orcldr

  $ rman target sys/tiger@orcl auxiliary sys/tiger@orcldr

  RMAN > target double standby database dorecover;

  Here, I'm getting problem such as:

  While he was trying to connect to databases traget and auxiliary, I noticed that when RMAN is connected across the two databases it shows

  ORCL (DBID xyz)

  ORCL (unmounted)

  Here, in my opinion the second database must be the name of the standby database or auxiiary db name is Eve db which is orcldr in my case

  After this issue, I am faced with the second issues such as:

  in command

  target double standby database dorecover;

  After you change the command to

  RMAN > duplicate target database "sleep."

  RMAN-05520 database name mismatch

  Can u please let me know where I can be wrong

  Finally I have the problem after all this googling and thinking the same thing.

  Thanks to you all

• Set - up physical Standby with RMAN

  Hello
  
  I took a database, including the archivelogs with RMAN. After 5-6 hours after the end of the backup, I created the primary as on DB standby control file:
  
  rman target $ /.
  RMAN > BACKUP CURRENT CONTROLFILE to STANDBY;
  
  
  Can I use the FULL existing backup to create a standby DB when the standby control file was created several hours later the backup is complete, or do I need to create a new FULL backup?
  
  
  Concerning
  Aingle

  Yes,
  My experience on Unix is sooner or later, in the absence of a tape library, you find yourself using NFS.
  Be aware that this should be the buffer of NFS 3 + and with a reading and writing of 32767octets.

  Apart from this, the procedure seems OK.

  -----------
  Sybrand Bakker
  Senior Oracle DBA

• Resume failure RMAN incremental backup to disk 0 cold

  Linux x 86-64 10.2.0.4
  
  I take an incremental backup of 0 to disk in a database that is mounted. I'm backup the database on an NFS volume. I lost the NFS mounting. Is it possible to take a backup where it stopped at? He complains "ORA-27038: created file already exists" if I restart it. RMAN can easily return to a restore. I don't have time to redo everything all again. The database is greater than 10 TB. The target is not upward. Is there something else I could do?

  If you have lost the NFS mounting, has not lost what had been saved to this mount? So, don't you create a fresh backup of all the files again?

  How are you "restart" backup? If you issue a fresh BACKUP DATABASE, it must create a new backup, with new backupsets - I hope the backuppiece (i.e. the name of the file) is unique and/or the file with the same name is not existing on the target.

  Hemant K Collette

• RMAN incremental backup

  Hi all

  I have a question, I am on RHEL-6 / Oracle 11 GR 2.

  After taking the backup of level 0 (RMAN), if I plan tale level1 backup then how oracle knows that they are the backup piece is considered a level 1 backup.

  I mean, that's how oracle knows that these are the blocks that had to be backeup.

  Kind regards.

  Please refer to the https://docs.oracle.com/cd/B19306_01/backup.102/b14192/bkup004.htm documentation

  Article: 4.4.1 Alogrithm incremental backup

  -Jonathan Rolland

• Case of level 1 RMAN incremental backup failure...

  Hello

  I have a question. If the incremental level 1 RMAN backup fails, the next level 1 backup attempt will have backup all changes since the last backup of level 0 OR level 1 successful?

  Best regards

  Hello

  I have a question. If the incremental level 1 RMAN backup fails, the next level 1 backup attempt will have backup all changes since the last backup of level 0 OR level 1 successful?

  Check the dictionary BACKUP_SET V$ to confirm if the last backup has been affected by the data dictionary. If incremental backup failed is picked up then the last incremental backup will be required in the case of the next LEVEL 1 DIFFERENTIAL INCREMENTIELLE backup.

  But to be on the safe side, it is recommended that you take a LEVEL 01 CUMILATIVE INCREMENTAL backup of level 0 up to this day. For the recovery process can go through even if the previous backup is not valid.

  Thank you &

  Best regards

• RMAN incremental backup continue to make backups "complete"?

  Hello, I am new to RMAN. I used only the full backup by RMAN, and it was OK. Now, due to lack of disk space, I decided to make incremental backups.

  I put 'to recovey retention policy window of 3 days; then made a new backup (by MS) job, looks like:

  $rman_script = ' backup database cumulative device type disk tag '% TAG' extra strength 1;

  backup disc tag '% TAG' type device archivelog all not saved delete all entries;

  But, always visible, size of the backup is the same all the time as a full backup? Where is my mistake?

  This is a backup log:

  RMAN > database backup incremental level 1 cumulative device type disk tag "BACKUP_DB10_00002_070413021501".

  From backup 4 July 13

  allocated channel: ORA_DISK_1

  channel ORA_DISK_1: sid = 1043 devtype = DISK

  channel ORA_DISK_1: additional start-up compressed 1 datafile backupset

  channel ORA_DISK_1: specifying datafile (s) in backupset

  input datafile fno = name=/database/app/oracle/oradata/db/tablespace1.dbf 00007

  input datafile fno = name=/database/app/oracle/oradata/db/tablespace2.dbf 00002

  ...

  channel ORA_DISK_1: starting total, 1 to 4 July 13

  channel ORA_DISK_1: finished piece 1 to 4 July 13

  piece handle=/backup/oracle/flash_recovery_area/db/backupset/2013_07_04/o1_mf_nnnd1_BACKUP_DB10_00002_0_8x9d45f5_.bkp tag = BACKUP_DB10_00002_070413021501 comment = NONE

  channel ORA_DISK_1: complete set of backups, time: 00:41:15

  Backup finished on 4 July 13

  > database backup incremental level 1 cumulative device type disk tag "BACKUP_DB10_00002_070413021501".

  There is a Note of support Oracle ' size of level 1 or level 1 cumulative backup greater than level 0 backups [ID 1534691.1].

  which applies if you use the DUNGEON and change the tag.

  I note that you change the tag every day, but do not use the clause to KEEP this note should not apply.

  Why not explicitly run EXTRA STRENGTH 0 backup once and then monitor subsequent LEVEL 1 backups?

  Hemant K Collette

• With RMAN for backup, but not with the archive logs

  We run a 10.2.0.4 database and the rest 11.2.0.3. All databases are on RHEL 5.7.
  
  Our backups are via RMAN, with the data files, controlfiles and spfile saved to disk, then each day, it is copied off site to a backup server.
  
  My question is around logs archiving and RMAN.
  Our current situation is:
  -We do not save the logs to archive via RMAN. for example: backup of the list of the archivelog all; gives "specification does not match backups in the repository.
  -We set the archive_lag_target so archiving logs is produced every 30 minutes.
  -a cron job running on a remote server runs periodically to synchronize data (rsync server) the directory of archives with a remote control.
  -daily archive logs on the server host are bzip. Purge of these newspapers is scheduled for more than 3 days rman 'window of recovery '.
  
  I'm interested in feedback as to if our strategy around archiving logs is correct or not.
  
  Should we be backup logs archiving via RMAN? Are there potential problems with our approach?
  We just kept our backups in this way the DBA in the past want to see newspapers archived as opposed to it while a backupset.
  
  Curious to know what others are doing.

  Hello

  Welcome to OTN!

  Archivelogs with RMAN backup is a better option. When the archives are supported by RMAN, RMAN would be aware of the backup items which are required at the time of the recovery archives. You don't have to do it manually.

  On the disk space, you can delete the archives once they were saved by RMAN.

  RMAN > delete any backup archivelog entry;

  This supports the archivelogs and then remove the archives from the disk.

  If you have a dataguard configured, you can set the deletion policy 'applied on Eve', which removes the archives as and when they are applied to the database ensures there by the management of disk space again.

  RMAN > configure archivelog political of deletion to applied pending;

• Merger of the RMAN incremental backups

  Hi all
  
  I have a confusion.
  
  Sunday I L0 scheduled backup (backup as compressed).
  Then next 6 days (MON - sat) I incremental backup Level1 (differential).
  
  So let's assume that if I need to restore the Thursday, I need Sunday L0 and L1 of (MON, TUE, WED).
  
  Is it possible to merge L0 and L1 (until WEDNESDAY).
  
  Please specify.
  
  Thanks in advance.
  
  Kind regards
  Bikram

  As long as you use differential incremental backups, you need all of them between the last level 0 backup and recovery at the time point. It is exactly the concept of cumulative incremental backups to change this behavior. In the case of cumulative backups you would only need the full backup and the incremental backup from Wednesday.

  Werner

• RMAN incremental backup levels

  Hi experts,
  
  It is a 10g database operates in archivelog mode.
  
  It's the backup plan I currently use:
  
  backup database archivelog; (runs once a day)
  
  
  database backup incremental level 1; This runs every 3 hours
  
  
  All of the incremental backups that I do are level 1.
  
  * I will be able to restore, if I have no extra 0 level? **
  
  Thank you, John

  user629010 wrote:
  Thanks for the replies.

  So I should make a level 0 backup immediately after the full backup (backup of database ;) is that correct?

  Not its correct level 0 is in fact the entire backup only. This is the point which includes everything and hereonly, you start your next batch of incremental backups. You can throw the last backups made if you have taken a level 0 backup, because it covers everything and is a full backup.

  HTH
  Aman...

• NOLOGGING with DEC and RMAN incremental backup

  DB Version: 10.2.0.4
  OS : AIX
  
  We use RMAN for our backup scheduled via cron.
  Retention Policy : REDUNDANCY 1
  Type             : Incremental 

  On the 31 of every month in most of our patterns, we recreate a lot of tables using DEC.
  
  Due to the size, the CREATE TABLE instructions take a while. Because we run on the Standard edition, we cannot use the PARALLELISM.
  NOLOGGING is the only option to improve performance. But we are concerned the roll forward not being generated for these tables and the potential loss of data in case of crash.
  
  Currently we take a Level0 backup every Wednesday and Sunday night and LEVEL1 backup on the rest of the days.
  
  I know that all will be saved on Level0. But what happens if 31 falls on a day which occurs the only backup of level 1. We will be safe only until the next Level0 arrives.
  
  No work around for this?

  HI T.Boyd,

  NOLOGGING is the only option to improve performance.

  So no recovery will be generated but the blocks in the data files will change.

  But what happens if 31 falls on a day which occurs the only backup of level 1.

  Only blocks changed since the last level 0 (or level 1 if you use differential differential) will be saved which includes blocks modified by the ETG.

  No work around for this?

  No need for a work around. You can check at any time whether a certain activity nologging will compromise a restoration with < report="" unrecoverable;=""> .
  Kind regards
  Tycho

• Oracle 12 c with RMAN database backup

  Hi All-

  I started back to the top of one of my PDB using utility RMAN following the instructions on the link below.

  http://www.Oracle.com/WebFolder/technetwork/tutorials/OBE/DB/12C/R1/PDB/pdb_pitr/pdb_pitr.html

  --Connect to Oracle RMAN.
  rman> Connect target /
  --Set the Oracle RMAN backup device type and backup file location.
  rman> configure channel device type disk format '/stage/db_backup_files/cdb1/%U';
  --Turn on automatic backup of control files.
  rman> configure controlfile autobackup on;
  --Back up the database and archive logs.
  rman>backup database plus archivelog;
  rman> exit
  

  Back to the top has been done successfully, but when I started with the recovery process he gave me error reporting collection.

  Run {set until what SNA = 2263440; restore database connectable to pdb2; retrieve auxiliary destination of the pluggable database pdb2 = "/ stage/db_backup_files/cdb1 '; alter connectable database to pdb2 ;} open resetlogs}

  This is the track to RMAN

  Recovery Manager: release 12.1.0.2.0 - Production on sat 22 Aug 12:22:44 2015

  Copyright (c) 1982, 2014, Oracle and/or its affiliates.  All rights reserved.

  RMAN > Connect target /.

  connected to target database: RELIABLE (DBID = 1028507357)

  RMAN > run {}

  2 > ALTER DATABASE connectable to MultiPlay NARROW;

  3 > set up what SNA = 14899894;

  4 > restore database pluggable MultiPlay;

  5 > restore database pluggable MultiPlay;

  6 > alter database connectable to MultiPlay open resetlogs;

  {7 >}

  using the control file of the target instead of recovery catalog database

  Statement processed

  executing command: SET until clause

  From restoration to 22 August 15

  allocated channel: ORA_DISK_1

  channel ORA_DISK_1: SID = 22 type of device = DISK

  channel ORA_DISK_1: from the restore backup set data file

  channel ORA_DISK_1: specifying datafile (s) to restore from backup set

  channel ORA_DISK_1: restore datafile 00035 to E:\APP\ORADATA\RELIABLE\MULTIPLAY\SYSTEM01. DBF

  channel ORA_DISK_1: restore datafile 00036 to E:\APP\ORADATA\RELIABLE\MULTIPLAY\SYSAUX01. DBF

  channel ORA_DISK_1: restore datafile 00037 to E:\APP\ORADATA\RELIABLE\MULTIPLAY\MULTIPLAY_USERS01. DBF

  channel ORA_DISK_1: backup E:\APP\RECOVERY_AREA\RELIABLE\40A433F530464F3BBF2E7C0B1553ECE8\BACKUPSET\2015_08_22\O1_MF_NNNDF_TAG20150 piece reading

  822T115328_BXJ5C182_. BKP

  channel ORA_DISK_1: piece handle = E:\APP\RECOVERY_AREA\RELIABLE\40A433F530464F3BBF2E7C0B1553ECE8\BACKUPSET\2015_08_22\O1_MF_NNNDF_TAG20150822T115328_BX

  J5C182_. Tag BKP = TAG20150822T115328

  channel ORA_DISK_1: restored the backup part 1

  channel ORA_DISK_1: restore complete, duration: 00:01:05

  Restoration finished in 22 August 15

  From pick up to August 22, 15

  using channel ORA_DISK_1

  RMAN-05026: WARNING: following tablespaces set applies to the Point in time specified assuming

  List of tablespaces should have UNDO segments

  SYSTEM tablespace

  Tablespace UNDOTBS1

  Creation of automatic instance with SID = "zbnt."

  RMAN-00571: ===========================================================

  RMAN-00569: = ERROR MESSAGE STACK FOLLOWS =.

  RMAN-00571: ===========================================================

  RMAN-03002: failure of the command recover at 22/08/2015 12:24:56

  ORA-19852: error creating services for the instance auxiliary zbnt (error 0)

  ORA-27302: failure occurred:

  ORA-27303: additional information: could not start the instance

  Thank you

  Siradji

  Just a wild guess

  You do not use this script in a command as Administrator window

  Therefore, you won't be able to create any service.

  I also have a question: your only troubleshooting activity consisted of hitting CTRL + C and CTRL-V, is not it?

  It's so boring to most of the people here do not even TRY to solve their own problems.

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

  Sybrand Bakker

  Senior Oracle DBA

• RMAN incremental backup issue

  Hi all
  
  I'm read on Oracle11g release2 doc RMAN.
  
  In Chapter 9:
  
  "For example, you can implement a three-tier backup plan as well as a level 0 backup is performed every month, a cumulative level 1 is taken from the week, and a premium of level 1 is taken daily. In this strategy, you will never have to apply more than one day to complete recovery redo. »
  
  First I got an error when I try to make a level 0 backup

  RMAN> BACKUP INCREMENTAL LEVEL 0 database;
   
  Starting backup at 27-APR-11
  using channel ORA_DISK_1
  using channel ORA_DISK_2
  using channel ORA_DISK_3
  using channel ORA_DISK_4
  RMAN-00571: ===========================================================
  RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS ===============
  RMAN-00571: ===========================================================
  RMAN-03002: failure of backup command at 04/27/2011 20:42:19
  RMAN-20021: database not set
  RMAN-06031: could not translate database keyword

  and if the success of step, how I would go about to implement this?
  I can do this via dbms_scheduler? or I need to do this via cron?
  
  -Thank you very much for the entry

  Your DB must be in use archivelogmode use hot backups. Is your db in archivelog mode?