Overview

This guide explains how to perform the backup/recovery manager tbrmgr backup/recovery provided by Tibero using the SYSBACKUP account, which is an account with all permissions related to backup and recovery for Backup/Recovery purposes.

Method

Check SYSBACKUP's roles and privileges

1. Check Privileges

[tac1] SQL> select * from dba_sys_privs where grantee='SYSBACKUP';

GRANTEE              PRIVILEGE                                ADMIN_OPTION
-------------------- ---------------------------------------- ------------
SYSBACKUP            ALTER SYSTEM                             NO
SYSBACKUP            CREATE SESSION                           NO
SYSBACKUP            ALTER SESSION                            NO
SYSBACKUP            CREATE TABLESPACE                        NO
SYSBACKUP            ALTER TABLESPACE                         NO
SYSBACKUP            DROP TABLESPACE                          NO
SYSBACKUP            ALTER DATABASE                           NO
SYSBACKUP            ALTER RMGR                               NO

8 rows selected.

2. Check Roles

[tac1] SQL> select * from dba_role_privs where grantee='SYSBACKUP';

GRANTEE              GRANTED_ROLE         ADMIN_OPTION DEFAULT_ROLE
-------------------- -------------------- ------------ ------------
SYSBACKUP            SELECT_CATALOG_ROLE  NO           YES

1 row selected.

3. Check Privileges granted to Roles

[tac1] SQL> SELECT GRANTEE, PRIVILEGE FROM DBA_SYS_PRIVS WHERE GRANTEE IN ('SELECT_CATALOG_ROLE');

GRANTEE              PRIVILEGE                               
-------------------- ----------------------------------------
SELECT_CATALOG_ROLE  SELECT ANY DICTIONARY

1 row selected.

Backup & Recovery using Full Backup

1. Create sample schema and insert sample data

--Create sample Tablespace

SQL>create tablespace test01 datafile '+DS0/tac/test01.dtf' size 100M autoextend off;

--Create sample schema

SQL>create user test01 identified by test01 default tablespace test01;
SQL>grant connect, resource to test01;

--Insert sample data

[tac1] SQL>conn test01/test01
[tac1] SQL>create table t (col number);

[tac1] SQL>BEGIN
  FOR I IN 1..1000
  LOOP
    INSERT INTO t
    SELECT LEVEL
    FROM dual
    CONNECT BY LEVEL <= 10;
  END LOOP;


  COMMIT;
END;
/

--Verify data

[tac1] SQL> select count(*) from t;

COUNT(*)
----------
10000

1 row selected.

2. Perform tbrmgr Full Backup

--Check tablespace and datafile

[tac1] SQL>col DF_NAME format a40
[tac1] SQL>col TS_NAME format a15 

[tac1] SQL>SELECT a.name as DF_NAME, b.name as TS_NAME, a.FILE#, b.TS# 
FROM v$datafile a 
JOIN v$tablespace b 
ON a.TS#=b.TS# ;


DF_NAME                                  TS_NAME              FILE#        TS#
---------------------------------------- --------------- ---------- ----------
+DS0/tac/system01/system01.dtf           SYSTEM                   0          0
+DS0/tac/system01/undo0_01.dtf           UNDO0                    1          1
+DS0/tac/system01/usr01.dtf              USR                      2          3
+DS0/tac/system01/tpr01.dtf              SYSSUB                   3          4
+DS0/tac/system02/undo1_01.dtf           UNDO1                    4          5
+DS0/tac/test01.dtf                      TEST01                   5          6

6 rows selected.

--Perform tbrmgr full backup

$ tbrmgr backup --userid sysbackup/tibero@tac1 -o /home/tibero/backup -w

==============================================================================
=   Recovery Manager(RMGR) starts                                            =
=                                                                            =
=   TmaxTibero Corporation Copyright (c) 2020-. All rights reserved.         =
==============================================================================
archive log check succeeded
RMGR '-o' option used
: backing up to /home/tibero/backup

==============================================================================
    RMGR - Backup (FULL)
==============================================================================
Initializing the backup progress, it may take few minutes...

BACKUP  (set_id:  1, ts_id:  0, df_id:  0)
100.0% |===============================|   38400/38400   blks    0.00s
Synchronizing...
BACKUP  (set_id:  1, ts_id:  1, df_id:  1)
100.0% |===============================|   12800/12800   blks    0.00s
Synchronizing...
BACKUP  (set_id:  1, ts_id:  3, df_id:  2)
100.0% |===============================|   12800/12800   blks    0.00s
Synchronizing...
BACKUP  (set_id:  1, ts_id:  4, df_id:  3)
100.0% |===============================|   12800/12800   blks    0.00s
Synchronizing...
BACKUP  (set_id:  1, ts_id:  5, df_id:  4)
100.0% |===============================|   38400/38400   blks    0.00s
Synchronizing...
BACKUP  (set_id:  1, ts_id:  6, df_id:  5)
100.0% |===============================|   12800/12800   blks    0.00s
Synchronizing...


       
Switching an online logfile...
Backing up archive logfiles...
Archivelog log-tac1-t0-r0-s8.arc (thread: 0, set_id: 1, low_tsn: 198541, next_tsn: 202169) backed up
Archivelog log-tac1-t1-r0-s6.arc (thread: 1, set_id: 1, low_tsn: 198541, next_tsn: 202169) backed up

Archive logfiles backup succeeded

Backing up the control file...
Control file backup succeeded

Database backup succeeded

RMGR backup ends

3. Remove tablespace to simulate failure

--In TAS, datafile deletion is not possible while DB is running, so shut down DB

$ cmrctl stop db --name tac1 --option immediate
$ cmrctl stop db --name tac2 --option immediate

--Connect to TAS and delete test01.dtf data file

[15:01]ASCMD +DS0/tac> ls
c1.ctl
c2.ctl
system01/
system02/
archive/
.passwd
test01.dtf


Number of files found: 7


[15:01]ASCMD +DS0/tac> rm test01.dtf
Removing file: +DS0/tac/test01.dtf

4. Perform tbrmgr Recovery

--Start DB and confirm failure

$ cmrctl start db --name tac1
BOOT SUCCESS! (MODE : MOUNT)

Check sys.log
********************************************************
* Critical Warning : Raise svmode failed. The reason is
*   TBR-1024 :  Database needs media recovery: open failed(+DS0/tac/test01.dtf).
*   Current server mode is MOUNT.
********************************************************

--Perform tbrmgr recovery
$ tbrmgr recover --userid sysbackup/tibero@tac1 -o /home/tibero/backup -w

==============================================================================
=   Recovery Manager(RMGR) starts                                            =
=                                                                            =
=   TmaxTibero Corporation Copyright (c) 2020-. All rights reserved.         =
==============================================================================
RMGR '-o' option used
: restoring from /home/tibero/backup

==============================================================================
    RMGR - recovery (WITH ARCHIVELOG)
==============================================================================
CM information
===========================================================
CM NAME             : cm1
CM UI PORT          : 18629
RESOURCE FILE PATH  : /cm/cmresource
CM MODE             : GUARD ON, FENCE OFF, ROOT ON
LOG LEVEL           : 2
CM BLOCK SIZE       : 512
===========================================================
Trying to shutdown other instances
=============================== SUCCESS! ===============================
Succeeded to request at the local node to stop the service.
Please use "cmrctl show service" command to verify the result.
=======================================================================
Shutdown succeeded
Shutting down the instance...

tbdown failed. proc info file is deleted.
Hint: Please check if the tbsvr instance was already stopped.
info file is deleted.
unlink failed.: No such file or directory

Control file #0 (+DS0/tac/c1.ctl) is accessible
Control file #1 (+DS0/tac/c2.ctl) is accessible

All control files are accessible. No need to restore the backup control file.

Booting up the instance...

Change core dump dir to /db/tibero7/bin/prof.
Listener port = 8629

Tibero 7  

TmaxTibero Corporation Copyright (c) 2020-. All rights reserved.
Tibero instance started up (MOUNT mode).
Initializing the restore progress, it may take few minutes...

RMGR begins restoring backup files.
Full backup set_id: 1
Last incremental backup set_id: 1

RESTORE (set_id:  1, ts_id:  0, df_id:  0) - FULL BACKUP
100.0% |===============================|   38400/38400   blks    2.01s
Synchronizing...
RESTORE (set_id:  1, ts_id:  1, df_id:  1) - FULL BACKUP
100.0% |===============================|   12800/12800   blks    0.00s
Synchronizing...
RESTORE (set_id:  1, ts_id:  3, df_id:  2) - FULL BACKUP
100.0% |===============================|   12800/12800   blks    0.00s
Synchronizing...
RESTORE (set_id:  1, ts_id:  4, df_id:  3) - FULL BACKUP
100.0% |===============================|   12800/12800   blks    0.00s
Synchronizing...
RESTORE (set_id:  1, ts_id:  5, df_id:  4) - FULL BACKUP
100.0% |===============================|   38400/38400   blks    1.04s
Synchronizing...
RESTORE (set_id:  1, ts_id:  6, df_id:  5) - FULL BACKUP
100.0% |===============================|   12800/12800   blks    0.00s
Synchronizing...
Settling...



Restoring logfiles...
Archivelog log-tac1-t0-r0-s8.arc (thread: 0, set_id: 1, low_tsn: 198541, next_tsn: 202169) restored
Archivelog log-tac1-t1-r0-s6.arc (thread: 1, set_id: 1, low_tsn: 198541, next_tsn: 202169) restored

Archive logfiles restoration succeeded

Database restoration succeeded

Given recovery point: COMPLETE RECOVERY
Recoverable point using only backup archivelogs: change 202168

recoverSQL: ALTER DATABASE RECOVER AUTOMATIC  
Database automatic recovery succeeded

Shutting down the instance...

Tibero instance terminated (NORMAL mode).              

Booting up the instance...

Change core dump dir to /db/tibero7/bin/prof.
Listener port = 8629

Tibero 7  

TmaxTibero Corporation Copyright (c) 2020-. All rights reserved.
Tibero instance started up (NORMAL mode).
RMGR recovery ends

5. Verify Data

--Verify data

[tac1] SQL> conn test01/test01
Connected to Tibero.

[tac1] SQL> select count(*) from t;    


COUNT(*)
----------
10000

1 row selected

Note

This is the method to delete backup sets using tbrmgr.

--Delete backup sets using tbrmgr

$ tbrmgr delete -b 1 --userid sysbackup/tibero@tac1

==============================================================================
=   Recovery Manager(RMGR) starts                                            =
=                                                                            =
=   TmaxTibero Corporation Copyright (c) 2020-. All rights reserved.         =
==============================================================================
archive log check succeeded
RMGR '-o' option not used
: deleting from the path actually backed up previously

==============================================================================
    RMGR - delete
==============================================================================
1 out of 1 backup_sets erased
RMGR delete ends

Related to

Search

Welcome to Tibero GTS!

Performing tbrmgr backup/recovery (Using SYSBACKUP)

Overview

Method

Check SYSBACKUP's roles and privileges

Backup & Recovery using Full Backup

업무 외 시간 안내

Search

Welcome to Tibero GTS!

Overview

Method

Check SYSBACKUP's roles and privileges

Backup & Recovery using Full Backup