Overview

This document describes a method for quickly recovering from a specific data file issue or loss in a TAC environment by temporarily building a SINGLE node using a backup. However, it is not recommended to use the temporarily built SINGLE environment for production purposes.

Test Environment
OS Version: Rocky Linux 8.10 (Green Obsidian)
Tibero Version: Tibero7 (DB7.2.4) Build 303667
AS-IS Tibero Configuration: 2Node TAC (Raw-Device)
TO-BE Tibero Configuration: SINGLE

Method

Preparation

Temporary TO-BE Server (SINGLE Build)

Precautions When Building the TO-BE Server
Set the Hostname and OS Profile to be the same as the AS-IS server.
Compress the Tibero engine on the AS-IS server and copy it to the TO-BE server.
Decompress the engine. ( ※ Do not create the DB. )
Redefine the control and data file paths in the tip file. (raw->filesystem)
Register tbcm.

-- Decompress TO-BE engine
[t7_1]tibero7@tibero_raw:/home/tibero7/tibero_engine> tar -xvf tibero7.tar.gz
tibero7/
tibero7/nls/
tibero7/nls/zoneinfo/
tibero7/nls/zoneinfo/timezone_be.dat
tibero7/nls/zoneinfo/timezone_le.dat
tibero7/client/
.. omitted

-- Redefine control and data file paths (raw->filesystem)
CONTROL_FILES="/home/tibero7/tibero_data/t7_1/control/c1.ctl","/home/tibero7/tibero_redo/t7_1/control/c2.ctl"
DB_CREATE_FILE_DEST="/home/tibero7/tibero_data/t7_1/data"
.. omitted

-- CM registration
[t7_1]tibero7@tibero_raw:/home/tibero7> cmrctl show
Resource List of Node cm1
=====================================================================
  CLUSTER     TYPE        NAME       STATUS           DETAIL
 ------------------------
     COMMON  network         t7_int       UP (private) 192.168.29.105/8651
     COMMON  cluster         t7_cls       UP inc: t7_int, pub: N/A
     t7_cls     file       t7_cls:0       UP /home/tibero7/tibero_engine/tbcm.log
     t7_cls  service             t7     DOWN Database, Active Cluster (auto-restart: OFF)
     t7_cls       db           t7_1     DOWN t7, /home/tibero7/tibero_engine/tibero7, failed retry cnt: 0
=====================================================================

Execution Scenario

1. Assume Specific Data File Corruption

SQL> col file_name for a30
SQL> col tablespace_name for a20
SQL> select file_name, tablespace_name, sum(bytes/1024/1024) as MB 
      from dba_data_files 
      group by file_name, tablespace_name 
      order by 1,2;
      

FILE_NAME                      TABLESPACE_NAME              MB
------------------------------ -------------------- ----------
/dev/raw/raw17                 SYSTEM                      500
/dev/raw/raw18                 USR                         500
/dev/raw/raw19                 UNDO0                       500
/dev/raw/raw20                 UNDO1                       500
/dev/raw/raw24                 SYSSUB                      500
/dev/raw/raw25                 TEST                        500 -- Assume data file corruption
/dev/raw/raw26                 TEST                        500 -- Assume data file corruption
/dev/raw/raw27                 BACKUP                      500

8 rows selected.

2. TAC (Raw-Device Configuration) Online Backup

 -- db backup
alter database begin backup;

!dd if=/dev/raw/raw17 bs=128M iflag=direct oflag=direct of=/home/tibero7/system001.dtf bs=128M status=progress
!dd if=/dev/raw/raw18 bs=128M iflag=direct oflag=direct of=/home/tibero7/usr001.dtf bs=128M status=progress
!dd if=/dev/raw/raw19 bs=128M iflag=direct oflag=direct of=/home/tibero7/undo001.dtf bs=128M status=progress
!dd if=/dev/raw/raw20 bs=128M iflag=direct oflag=direct of=/home/tibero7/undo002.dtf bs=128M status=progress
!dd if=/dev/raw/raw22 bs=128M iflag=direct oflag=direct of=/home/tibero7/data001.dtf bs=128M status=progress
!dd if=/dev/raw/raw23 bs=128M iflag=direct oflag=direct of=/home/tibero7/data002.dtf bs=128M status=progress
!dd if=/dev/raw/raw24 bs=128M iflag=direct oflag=direct of=/home/tibero7/syssub001.dtf bs=128M status=progress
!dd if=/dev/raw/raw25 bs=128M iflag=direct oflag=direct of=/home/tibero7/test001.dtf bs=128M status=progress
!dd if=/dev/raw/raw26 bs=128M iflag=direct oflag=direct of=/home/tibero7/test002.dtf bs=128M status=progress
!dd if=/dev/raw/raw27 bs=128M iflag=direct oflag=direct of=/home/tibero7/backup001.dtf bs=128M status=progress

alter database end backup;

alter system switch logfile;
alter system switch logfile;
alter system switch logfile;
alter system switch logfile;
alter system switch logfile;

alter system checkpoint;

SQL> alter database backup controlfile to trace as '/home/tibero7/reset.sql' resetlogs;

Database altered.

3. Copy Backup and All Node Archive Files to TO-BE Server

-- SCP backup to TO-BE server
[t7_1]tibero7@tibero_raw:/home/tibero7> scp ./*.dtf *.sql root@192.168.29.105:/home/tibero7
The authenticity of host '192.168.29.105 (192.168.29.105)' can't be established.
ECDSA key fingerprint is SHA256:X1dZuXKK781n1qhXT7zPo/tAnYUL4AxmGN9ywX3sy/w.
Are you sure you want to continue connecting (yes/no/[fingerprint])? yes
Warning: Permanently added '192.168.29.105' (ECDSA) to the list of known hosts.
root@192.168.29.105's password:
backup001.dtf                                               100%  500MB 176.7MB/s   00:02
data001.d  500MB 166.4MB/s   00:03
data002.dtf                                                 100%  500MB 169.5MB/s   00:02
syssub001.dtf                                               100%  500MB 166.7MB/s   00:02
system001.dtf                                               100%  500MB 159.5MB/s   00:03
test001.dtf                                                 100% 1000MB 124.5MB/s   00:08
test002.dtf                                                 100%  500MB 123.7MB/s   00.dtf                                                 100%  500MB 117.3MB/s   00:04
undo002.dtf                                                 100%  500MB 116.8MB/s   00:04
usr001.dtf                                                  100%  500MB 119.4MB/s   00:04
reset.sql                                                   100% 4403     2.5MB/s   00:00

-- SCP archive files from node 1 to TO-BE server
[t7_1]tibero7@tibero_raw:/home/tibero7/tibero_arch/t7/t7_1/arch> scp ./*.arc root@192.168.29.105:/home/tibero7/tibero_arch/t7/t7_1/arch/
root@192.168.29.105's password:

log-t0-r0-s12.arc                                      100%  512   484.1KB/s   00:00
log-t0-r0-s13.arc                                      100%   40KB  19.8MB/s   00:00
log-t0-r0-s14.arc                                      100%  320KB  39.5MB/s   00:00
log-t0-r0-s15.arc                                      100% 6389KB  58.8MB/s   00:00
log-t0-r0-s16.arc                                      100%  164KB  54.6MB/s   00:00
log-t0-r0-s17.arc                                      100%  100MB 163.2MB/s   00:00
log-t0-r0-s18.arc                                      100%   89MB 176.5MB/s   00:00
log-t0-r0-s19.arc                                      100%  512   216.7KB/s   00:00
log-t0-r0-s20.arc                                      100% 1024     1.2MB/s   00:00
log-t0-r0-s21.arc                                      100%  512   745.3KB/s   00:00
log-t0-r0-s22.arc                                      100%  512   748.9KB/s   00:00


-- SCP archive files from node 2 to TO-BE server
[t7_2]tibero7@tibero_raw:/home/tibero7/tibero_arch/t7/t7_2/arch> scp ./*.arc root@192.168.29.105:/home/tibero7/tibero_arch/t7/t7_1/arch/
root@192.168.29.105's password:

log-t1-r0-s7.arc                                       100%  512   563.4KB/s   00:00
log-t1-r0-s8.arc                                       100%  512     1.7MB/s   00:00
log-t1-r0-s9.arc                                       100%   30KB  53.4MB/s   00:00
log-t1-r0-s10.arc                                      100%  179KB  72.2MB/s   00:00
log-t1-r0-s11.arc                                      100% 3072     5.9MB/s   00:00
log-t1-r0-s12.arc                                      100%   95MB 159.5MB/s   00:00
log-t1-r0-s13.arc                                      100%  512   449.6KB/s   00:00
log-t1-r0-s14.arc                                      100%  512   954.0KB/s   00:00
log-t1-r0-s15.arc                                      100%  512     1.5MB/s   00:00
log-t1-r0-s16.arc                                      100%  512     1.5MB/s   00:00

4. Recreate Single Node Control File

ㄴ Remove corrupted data files

ㄴ UNDO tablespace must be included regardless of node

-- Modify control file path
CREATE CONTROLFILE REUSE DATABASE "t7"
LOGFILE
-- Thread #0
GROUP 0 (
        '/home/tibero7/tibero_redo/t7_1/redo/redo001.redo',
        '/home/tibero7/tibero_redo/t7_1/redo/redo002.redo'
         ,
GROUP 1 (
        '/home/tibero7/tibero_redo/t7_1/redo/redo011.redo',
        '/home/tibero7/tibero_redo/t7_1/redo/redo012.redo'
          ) SIZE 100M,
GROUP 2 (
        '/home/tibero7/tibero_redo/t7_1/redo/redo021.redo',
        '/home/tibero7/tibero_redo/t7_1/redo/redo022.redo'
          ) SIZE 100M
      
RESETLOGS
DATAFILE
        '/home/tibero7/tibero_data/t7_1/data/system001.dtf',
        '/home/tibero7/tibero_data/t7_1/data/undo001.dtf',
        '/home/tibero7/tibero_data/t7_1/data/usr001.dtf',
        '/home/tibero7/tibero_data/t7_1/data/syssub001.dtf',
        '/home/tibero7/tibero_data/t7_1/data/undo002.dtf',
        --'/home/tibero7/tibero_data/t7.dtf', -- Remove or comment out this data file
        --'/home/tibero7/tibero_data/t7_1/data/test002.dtf', -- Remove or comment out this data file
        '/home/tibero7/tibero_data/t7_1/data/backup001.dtf'

..omitted

-- After nomount startup, recreate control file
SQL> @reset.sql

Control File created.

-- Add redo group for thread 1 and activate thread 1
 SQL> ALTER DATABASE ADD LOGFILE THREAD 1 GROUP 10 (
      '/home/tibero7/tibero_redo/t7_1/redo/redo031.redo',
      '/home/tibero7/tibero_redo/t7_1/redo/redo032.redo'
          ) SIZE 100M reuse ;
 SQL> ALTER DATABASE ADD LOGFILE THREAD 1 GROUP 11 (
      '/home/tibero7/tibero_redo/t7_1/redo/redo041.redo',
      '/home/tibero7/tibero_redo/t7_1/redo/redo042.redo'
          ) SIZE 100M reuse ;
 SQL> ALTER DATABASE ADD LOGFILE THREAD 1 GROUP 12 (
      '/home/tibero7/tibero_redo/t7_1/redo/redo051.redo',
      '/home/tibero7/tibero_redo/t7_1/redo/redo052.redo'
          ) SIZE 100M reuse ;

 SQL> alter database enable public thread 1;
 
 Database altered.

5. Perform Recovery

※ Precautions When Adding Parameters
These are hidden parameters, so be sure to consult with a Tibero engineer before use.
_TX_RECOVERY_SUSPEND=Y
ㄴ On startup, forcibly starts up by skipping recovery for previous tx.
_CF_DD_CHECK=N
ㄴ If a DF exists in DD but not in CF, all CF-DD corrections are skipped.

-- Perform recovery
SQL> alter database recover automatic;
TBR-1147: Need log file (thread 0 seq 23) to recover from TSN 204335.
Recovery done until 2025/11/28 10:30:12 TSN 204330.
Recovery is incomplete and resetlogs is required.

-- Add parameters
_TX_RECOVERY_SUSPEND=Y
_CF_DD_CHECK=N

-- Startup with resetlogs
[t7_1]tibero7@tibero_raw:/home/tibero7/tibero_engine> tbboot resetlogs

Listener port = 2629

Tibero 7

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

-- Add temp file
SQL> ALTER TABLESPACE TEMP ADD TEMPFILE '/home/tibero7/tibero_data/t7_1/data/temp001.dtf' SIZE 500M REUSE AUTOEXTEND OFF ;

6. Use tbexport/tbimport to extract and recover the necessary data from all normal areas except the damaged TEST TableSpace area.

Related to

Search

Welcome to Tibero GTS!

Incomplete Recovery After Building a Single Node Using a TAC Backup

Overview

Method

Preparation

Execution Scenario

업무 외 시간 안내

Search

Welcome to Tibero GTS!

Overview

Method

Preparation

Execution Scenario