Oracle教程FG435-块级恢复

本文档风哥主要介绍Oracle块级恢复相关知识，包括块级恢复的概念、块级恢复方法、块级恢复使用、块级恢复配置、块级恢复监控、块级恢复故障处理等内容，由风哥教程参考Oracle官方文档Troubleshooting内容编写，适合DBA人员在学习和测试中使用，如果要应用于生产环境则需要自行确认。

Part01-基础概念与理论知识

1.1 块级恢复的概念

Oracle块级恢复（Block Media Recovery）是使用RMAN恢复单个或多个损坏数据块的技术。块级恢复只恢复损坏的块，而不需要恢复整个数据文件，大大减少了恢复时间。块级恢复可以在数据库运行时进行，不影响其他数据块的访问。块级恢复是处理数据块损坏的首选方法。更多视频教程www.fgedu.net.cn

Oracle块级恢复的特点：

快速恢复：只恢复损坏的块，恢复速度快
在线恢复：数据库可以保持运行状态
精确恢复：只恢复损坏的块，不影响其他数据
减少停机：减少数据库停机时间
灵活恢复：可以选择恢复特定的块

1.2 块级恢复方法

Oracle块级恢复方法：

RMAN BLOCKRECOVER：使用RMAN恢复损坏块
恢复损坏列表：恢复V$DATABASE_BLOCK_CORRUPTION中的所有损坏块
恢复指定块：恢复指定的数据文件和块号
恢复表空间块：恢复表空间中的损坏块
恢复数据文件块：恢复数据文件中的损坏块

# RMAN BLOCKRECOVER命令说明

BLOCKRECOVER DATAFILE file# BLOCK block#
恢复指定数据文件的指定块
示例：BLOCKRECOVER DATAFILE 4 BLOCK 1024;

BLOCKRECOVER DATAFILE file# BLOCK block# TO block#
恢复指定数据文件的块范围
示例：BLOCKRECOVER DATAFILE 4 BLOCK 1024 TO 1030;

BLOCKRECOVER CORRUPTION LIST
恢复V$DATABASE_BLOCK_CORRUPTION中的所有损坏块
示例：BLOCKRECOVER CORRUPTION LIST;

BLOCKRECOVER TABLESPACE tablespace
恢复表空间中的损坏块
示例：BLOCKRECOVER TABLESPACE users;

BLOCKRECOVER DATAFILE file#
恢复数据文件中的损坏块
示例：BLOCKRECOVER DATAFILE 4;

# 块级恢复前提条件
– 必须有可用的RMAN备份
– 备份必须包含损坏的块
– 归档日志必须可用
– 数据库必须在归档模式下

1.3 块级恢复使用

Oracle块级恢复使用：

损坏检测：检测数据块损坏
损坏定位：定位损坏块的位置
备份验证：验证备份是否包含损坏块
块级恢复：执行块级恢复
恢复验证：验证恢复结果

风哥提示：块级恢复是处理数据块损坏的首选方法，建议定期检测数据块损坏，及时进行块级恢复。

Part02-生产环境规划与建议

2.1 块级恢复规划

Oracle块级恢复规划要点：

# 块级恢复规划
– 备份策略：确保有可用的RMAN备份
– 归档模式：数据库必须在归档模式下
– 监控机制：建立数据块损坏监控
– 恢复流程：建立块级恢复流程

# 备份策略规划
– 定期备份：定期进行RMAN备份
– 备份验证：验证备份的可用性
– 备份保留：保留足够的备份
– 归档日志：保留足够的归档日志

# 监控机制规划
– 定期检测：定期检测数据块损坏
– 告警机制：建立损坏告警机制
– 记录保存：保存损坏记录
– 分析报告：定期生成分析报告

# 恢复流程规划
– 损坏检测：检测数据块损坏
– 损坏定位：定位损坏块位置
– 备份验证：验证备份可用性
– 块级恢复：执行块级恢复
– 恢复验证：验证恢复结果

# 恢复时间规划
– 响应时间：快速响应损坏告警
– 恢复时间：快速完成块级恢复
– 验证时间：验证恢复结果
– 记录时间：记录恢复过程

2.2 块级恢复场景

Oracle块级恢复场景：

物理损坏：物理损坏导致的数据块损坏
逻辑损坏：逻辑损坏导致的数据块损坏
存储故障：存储故障导致的数据块损坏
介质损坏：介质损坏导致的数据块损坏
系统崩溃：系统崩溃导致的数据块损坏

# 块级恢复场景

1. 物理损坏
– 存储介质故障
– 磁盘坏块
– IO错误
– 数据文件损坏

2. 逻辑损坏
– Oracle内部错误
– 数据不一致
– 索引损坏
– 段损坏

3. 存储故障
– 存储系统故障
– 存储控制器故障
– 存储网络故障
– 存储配置错误

4. 介质损坏
– 磁盘介质损坏
– SSD介质损坏
– 磁带介质损坏
– 光盘介质损坏

5. 系统崩溃
– 操作系统崩溃
– 系统断电
– 系统重启
– 系统故障

2.3 块级恢复最佳实践

Oracle块级恢复最佳实践：

定期备份：定期进行RMAN备份
定期检测：定期检测数据块损坏
快速响应：快速响应损坏告警
验证恢复：验证恢复结果
记录过程：记录恢复过程

生产环境建议：块级恢复需要建立完善的恢复流程，建议定期备份、定期检测、快速响应。学习交流加群风哥微信: itpux-com

Part03-生产环境项目实施方案

3.1 块级恢复配置

3.1.1 配置块级恢复检测

# 1. 使用RMAN验证数据文件
RMAN> validate datafile 4;

Starting validate at 2026-03-31 10:00:00
using target database control file instead of recovery catalog
allocated channel: ORA_DISK_1
channel ORA_DISK_1: SID=150 device type=DISK
channel ORA_DISK_1: starting validation of datafile
channel ORA_DISK_1: specifying datafile(s) for validation
input datafile file number=00004 name=/oracle/app/oracle/oradata/FGEDUDB/users01.dbf
channel ORA_DISK_1: validation complete, elapsed time: 00:00:10
List of Datafiles
=================
File Status Marked Corrupt Empty Blocks Blocks Examined High SCN
—- —— ————– ———— ————— ———-
4 OK 0 2559 12800 12345678
File Name: /oracle/app/oracle/oradata/FGEDUDB/users01.dbf
Block Type Blocks Failing Blocks Processed
———- ————– —————-
Data 0 8192
Index 0 2048
Other 0 2559
Finished validate at 2026-03-31 10:00:10

# 2. 使用RMAN验证数据库
RMAN> validate database;

Starting validate at 2026-03-31 10:00:00
using target database control file instead of recovery catalog
allocated channel: ORA_DISK_1
channel ORA_DISK_1: SID=150 device type=DISK
channel ORA_DISK_1: starting validation of datafile
channel ORA_DISK_1: specifying datafile(s) for validation
input datafile file number=00001 name=/oracle/app/oracle/oradata/FGEDUDB/system01.dbf
input datafile file number=00002 name=/oracle/app/oracle/oradata/FGEDUDB/sysaux01.dbf
input datafile file number=00003 name=/oracle/app/oracle/oradata/FGEDUDB/undotbs01.dbf
input datafile file number=00004 name=/oracle/app/oracle/oradata/FGEDUDB/users01.dbf
channel ORA_DISK_1: validation complete, elapsed time: 00:00:30
List of Datafiles
=================
File Status Marked Corrupt Empty Blocks Blocks Examined High SCN
—- —— ————– ———— ————— ———-
1 OK 0 10240 51200 12345678
2 OK 0 8192 40960 12345678
3 OK 0 2048 25600 12345678
4 OK 0 2559 12800 12345678
Finished validate at 2026-03-31 10:00:30

# 3. 查看损坏块信息
SQL> select * from v$database_block_corruption;

FILE# BLOCK# BLOCKS CORRUPTION_CHANGE# CORRUPTION_TYPE
———- ———- ———- —————— —————
4 1024 1 0.00 ALL ZERO

# 4. 使用DBV验证数据文件
$ dbv file=/oracle/app/oracle/oradata/FGEDUDB/users01.dbf

DBVERIFY: Release 19.0.0.0.0 – Production on Tue Mar 31 10:00:00 2026

DBVERIFY – Verification starting : FILE = /oracle/app/oracle/oradata/FGEDUDB/users01.dbf

Page 1024 is marked corrupt
Corrupt block relative dba: 0x01000400, file 4, block 1024
Completely zero block found during dbv:

DBVERIFY – Verification complete

Total Pages Examined : 12800
Total Pages Processed (Data) : 8192
Total Pages Failing (Data) : 0
Total Pages Processed (Index): 2048
Total Pages Failing (Index): 0
Total Pages Processed (Other): 2559
Total Pages Processed (Seg) : 0
Total Pages Failing (Seg) : 0
Total Pages Empty : 0
Total Pages Marked Corrupt : 1
Total Pages Influx : 0
Total Pages Encrypted : 0
Highest block SCN : 12345678 (0.12345678)

# 5. 查看损坏块所属对象
SQL> select
segment_type,
owner,
segment_name
from dba_extents
where file_id = 4
and 1024 between block_id and block_id + blocks – 1;

SEGMENT_TYPE OWNER SEGMENT_NAME
————– ——- ————
TABLE SCOTT EMP

3.1.2 配置块级恢复执行

# 1. 恢复单个损坏块
RMAN> blockrecover datafile 4 block 1024;

Starting blockrecover at 2026-03-31 10:10:00
using target database control file instead of recovery catalog
allocated channel: ORA_DISK_1
channel ORA_DISK_1: SID=150 device type=DISK
channel ORA_DISK_1: restoring block(s)
channel ORA_DISK_1: specifying block(s) to restore from backup set
restoring blocks of datafile 00004
channel ORA_DISK_1: reading from backup piece /backup/FGEDUDB/backup_12345.bak
channel ORA_DISK_1: restored block(s) from backup piece 1
channel ORA_DISK_1: block restore complete, elapsed time: 00:00:10
starting media recovery
media recovery complete, elapsed time: 00:00:05
Finished blockrecover at 2026-03-31 10:10:15

# 2. 恢复多个损坏块
RMAN> blockrecover datafile 4 block 1024, 1025, 1026;

Starting blockrecover at 2026-03-31 10:10:00
using target database control file instead of recovery catalog
allocated channel: ORA_DISK_1
channel ORA_DISK_1: SID=150 device type=DISK
channel ORA_DISK_1: restoring block(s)
channel ORA_DISK_1: specifying block(s) to restore from backup set
restoring blocks of datafile 00004
channel ORA_DISK_1: reading from backup piece /backup/FGEDUDB/backup_12345.bak
channel ORA_DISK_1: restored block(s) from backup piece 1
channel ORA_DISK_1: block restore complete, elapsed time: 00:00:15
starting media recovery
media recovery complete, elapsed time: 00:00:10
Finished blockrecover at 2026-03-31 10:10:25

# 3. 恢复块范围
RMAN> blockrecover datafile 4 block 1024 to 1030;

Starting blockrecover at 2026-03-31 10:10:00
using target database control file instead of recovery catalog
allocated channel: ORA_DISK_1
channel ORA_DISK_1: SID=150 device type=DISK
channel ORA_DISK_1: restoring block(s)
channel ORA_DISK_1: specifying block(s) to restore from backup set
restoring blocks of datafile 00004
channel ORA_DISK_1: reading from backup piece /backup/FGEDUDB/backup_12345.bak
channel ORA_DISK_1: restored block(s) from backup piece 1
channel ORA_DISK_1: block restore complete, elapsed time: 00:00:20
starting media recovery
media recovery complete, elapsed time: 00:00:15
Finished blockrecover at 2026-03-31 10:10:35

# 4. 恢复损坏列表中的所有块
RMAN> blockrecover corruption list;

# 5. 恢复表空间中的损坏块
RMAN> blockrecover tablespace users;

# 6. 恢复数据文件中的损坏块
RMAN> blockrecover datafile 4;

# 7. 从指定备份恢复块
RMAN> blockrecover datafile 4 block 1024 from backupset 12345;

# 8. 验证恢复结果
SQL> select * from v$database_block_corruption;

no rows selected

SQL> select count(*) from scott.emp;

COUNT(*)
———-
14

3.2 块级恢复监控

3.2.1 配置块级恢复监控脚本

# 1. 创建块级恢复监控脚本
$ vi /home/oracle/scripts/monitor_block_corruption.sh

#!/bin/bash
# from:www.itpux.com.qq113257174.wx:itpux-com
# web: http://www.fgedu.net.cn
# from:www.itpux.com.qq113257174.wx:itpux-com
# web: http://www.fgedu.net.cn
# from:www.itpux.com.qq113257174.wx:itpux-com
# web: http://www.fgedu.net.cn
# 块级恢复监控脚本

export ORACLE_HOME=/oracle/app/oracle/product/19c/dbhome_1
export ORACLE_SID=FGEDUDB
export PATH=$ORACLE_HOME/bin:$PATH

LOG_DIR=”/home/oracle/scripts/block_recovery_logs”
mkdir -p $LOG_DIR

echo “$(date): Starting block corruption monitoring…” > $LOG_DIR/block_monitor.log

# 使用RMAN验证数据库
rman target / <> $LOG_DIR/block_monitor.log
run {
allocate channel ch1 type disk;
validate database;
release channel ch1;
}
exit
EOF

# 检查损坏块
sqlplus -s / as sysdba <> $LOG_DIR/block_monitor.log
set heading on
set feedback on
column corruption_type format a20

select file#, block#, blocks, corruption_type
from v\$database_block_corruption
order by file#, block#;

exit
EOF

# 检查是否有损坏
CORRUPT_COUNT=$(sqlplus -s / as sysdba < exit
EOF
)

if [ “$CORRUPT_COUNT” -gt 0 ]; then
echo “$(date): CORRUPTION DETECTED: $CORRUPT_COUNT corrupt blocks found” >> $LOG_DIR/block_monitor.log

# 发送告警邮件
mail -s “Block Corruption Alert: $CORRUPT_COUNT corrupt blocks found” admin@fgedu.net.cn < $LOG_DIR/block_monitor.log # 执行块级恢复 echo "$(date): Starting block recovery..." >> $LOG_DIR/block_monitor.log

rman target / <> $LOG_DIR/block_monitor.log
blockrecover corruption list;
exit
EOF

echo “$(date): Block recovery completed.” >> $LOG_DIR/block_monitor.log
else
echo “$(date): No corruption detected.” >> $LOG_DIR/block_monitor.log
fi

echo “$(date): Block corruption monitoring completed.” >> $LOG_DIR/block_monitor.log

# 2. 设置脚本权限
$ chmod +x /home/oracle/scripts/monitor_block_corruption.sh

# 3. 测试脚本
$ /home/oracle/scripts/monitor_block_corruption.sh

# 4. 设置定期监控
$ crontab -e

# 每天凌晨2点监控块损坏
0 2 * * * /home/oracle/scripts/monitor_block_corruption.sh >> /home/oracle/scripts/block_monitor.log 2>&1

3.2.2 配置块级恢复报告

# 1. 创建块级恢复报告脚本
$ vi /home/oracle/scripts/block_recovery_report.sh

export ORACLE_HOME=/oracle/app/oracle/product/19c/dbhome_1
export ORACLE_SID=FGEDUDB
export PATH=$ORACLE_HOME/bin:$PATH

LOG_DIR=”/home/oracle/scripts/block_recovery_logs”
REPORT_FILE=”$LOG_DIR/block_recovery_report.txt”

echo “Block Recovery Report” > $REPORT_FILE
echo “Generated: $(date)” >> $REPORT_FILE
echo “========================================” >> $REPORT_FILE

# 获取损坏块信息
sqlplus -s / as sysdba <> $REPORT_FILE
set heading on
set feedback on
set linesize 200
column file_name format a50
column corruption_type format a20
column owner format a10
column segment_name format a30

prompt
prompt Corrupt Blocks Summary:
prompt ———————-

select
d.file_name,
c.file#,
c.block#,
c.blocks,
c.corruption_type
from v\$database_block_corruption c, dba_data_files d
where c.file# = d.file_id
order by c.file#, c.block#;

prompt
prompt Corrupt Blocks Details:
prompt ———————–

select
e.owner,
e.segment_name,
e.segment_type,
c.file#,
c.block#,
c.corruption_type
from v\$database_block_corruption c, dba_extents e
where c.file# = e.file_id
and c.block# between e.block_id and e.block_id + e.blocks – 1
order by c.file#, c.block#;

exit
EOF

# 发送报告
mail -s “Block Recovery Report” admin@fgedu.net.cn < $REPORT_FILE # 2. 设置脚本权限 $ chmod +x /home/oracle/scripts/block_recovery_report.sh # 3. 测试脚本 $ /home/oracle/scripts/block_recovery_report.sh # 4. 查看报告 $ cat /home/oracle/scripts/block_recovery_logs/block_recovery_report.txt Block Recovery Report Generated: Tue Mar 31 10:00:00 CST 2026 ======================================== Corrupt Blocks Summary: ---------------------- FILE_NAME FILE# BLOCK# BLOCKS CORRUPTION_TYPE -------------------------------------------------- ---------- ---------- ---------- -------------------- /oracle/app/oracle/oradata/FGEDUDB/users01.dbf 4 1024 1 ALL ZERO Corrupt Blocks Details: ----------------------- OWNER SEGMENT_NAME SEGMENT_TYPE FILE# BLOCK# CORRUPTION_TYPE ---------- ------------------------------ ---------------- ---------- ---------- -------------------- SCOTT EMP TABLE 4 1024 ALL ZERO