18.4、Restoring Datafiles and Archived Redo Logs
如果介质失败损坏一个或多个数据文件，恢复损坏的文件前必须还原数据文件。如果还原的位置不是原始位置，必须在控制文件显示数据文件的新位置。
在裸设备上还原数据文件的过程和在文件系统上是相同的。
模拟1、在testtbs01中创建表，插入数据，然后切换日志，备份归档日志，testtbs01的某个数据文件被损坏，对应归档日志删除，还原和恢复
（1）在testtbs01中创建表并插入数据
SQL>
create table boss_new_test
( id number(5) constraint id_pk primary key,
lastname varchar2(10) constraint ln_nn not null,
firstname varchar2(10) not null unique,
userid varchar2(8) constraint ui_uq unique,
startdate date default sysdate,
user_type varchar2(4) constraint ut_ck check (user_type in(‘in‘,‘out‘))
)
initrans 1 maxtrans 255
pctfree 20 pctused 50
storage(initial 64m next 8m pctincrease 0)
tablespace testtbs01;
SQL> insert into boss_new_test values(1,‘aaa‘,‘a‘,‘00001‘,sysdate,‘in‘);
SQL> commit;
（2）创建存储过程插入数据
（3）查看检查点
SQL> set linesize 120
SQL> col "文件名" for a40
SQL>
select
ts.name "表空间名"
, df.file# "文件号"
, df.checkpoint_change# "检查点"
, df.name "文件名"
from v$tablespace ts,v$datafile df
where ts.ts#=df.ts# and ts.name=‘TESTTBS01‘
order by df.file#;
表空间名 文件号 检查点 文件名
------------------------------ ---------- ---------- ----------------------------------------
TESTTBS01 6 566111 /oracle/oradata/boss/testtbs01_01.dbf
SQL> select dbid,name,log_mode,checkpoint_change# from v$database;
DBID NAME LOG_MODE CHECKPOINT_CHANGE#
---------- --------- ------------ ------------------
1375601832 BOSS ARCHIVELOG 566111
SQL> select file#,name,status,CHECKPOINT_CHANGE#,recover from v$datafile_header where file#=6;
FILE# NAME STATUS CHECKPOINT_CHANGE# REC
---------- ---------------------------------------- ------- ------------------ ---
6 /oracle/oradata/boss/testtbs01_01.dbf ONLINE 566111 NO
SQL> select group#,members,sequence#,archived,status,first_change# from v$log;
GROUP# MEMBERS SEQUENCE# ARC STATUS FIRST_CHANGE#
---------- ---------- ---------- --- ---------------- -------------
1 1 5 YES INACTIVE 534657
2 1 6 NO CURRENT 563694
3 1 4 YES INACTIVE 514244
（4）进行日志切换，数据文件头scn和数据库scn并未变化
SQL> alter system switch logfile;
SQL> select group#,members,sequence#,archived,status,first_change# from v$log;
GROUP# MEMBERS SEQUENCE# ARC STATUS FIRST_CHANGE#
---------- ---------- ---------- --- ---------------- -------------
1 1 5 YES INACTIVE 534657
2 1 6 YES ACTIVE 563694
3 1 7 NO CURRENT 569439
SQL> select file#,name,status,CHECKPOINT_CHANGE#,recover from v$datafile_header where file#=6;
FILE# NAME STATUS CHECKPOINT_CHANGE# REC
---------- ---------------------------------------- ------- ------------------ ---
6 /oracle/oradata/boss/testtbs01_01.dbf ONLINE 566111 NO
SQL> select dbid,name,log_mode,checkpoint_change# from v$database;
DBID NAME LOG_MODE CHECKPOINT_CHANGE#
---------- ---------------------------------------- ------------ ------------------
1375601832 BOSS ARCHIVELOG 566111
SQL> alter system checkpoint;
SQL> select group#,members,sequence#,archived,status,first_change# from v$log;
GROUP# MEMBERS SEQUENCE# ARC STATUS FIRST_CHANGE#
---------- ---------- ---------- --- ---------------- -------------
1 1 5 YES INACTIVE 534657
2 1 6 YES INACTIVE 563694
3 1 7 NO CURRENT 569439
（5）备份新的归档日志
# cp -rf o1_mf_1_6_9rx90qq2_.arc /oradata/bossbak/test_ar_rec
（6）删除testtbs01的某个数据文件，删除o1_mf_1_6_9rx90qq2_.arc归档
# rm -rf testtbs01_01.dbf
# pwd
/oracle/flash_recovery_area/BOSS/archivelog/2014_06_04
# rm -rf o1_mf_1_6_9rx90qq2_.arc
（7）
SQL> alter system flush buffer_cache;
SQL> select * from boss_new_test;
ID LASTNAME FIRSTNAME USERID STARTDATE USER
---------- ---------- ---------- -------- ------------ ----
2 aa a 00002 04-JUN-14 in
3 bb b 00003 04-JUN-14 in
4 cc c 00004 04-JUN-14 in
5 dd d 00005 04-JUN-14 in
6 ff f 00006 04-JUN-14 in
SQL> create table boss_new_test1
2 ( id number(5) constraint id_pk primary key,
3 lastname varchar2(10) constraint ln_nn not null,
4 firstname varchar2(10) not null unique,
5 userid varchar2(8) constraint ui_uq unique,
6 startdate date default sysdate,
7 user_type varchar2(4) constraint ut_ck check (user_type in(‘in‘,‘out‘))
8 )
9 initrans 1 maxtrans 255
10 pctfree 20 pctused 50
11 storage(initial 64m next 8m pctincrease 0)
12 tablespace testtbs01;
user_type varchar2(4) constraint ut_ck check (user_type in(‘in‘,‘out‘))
*
ERROR at line 7:
ORA-01565: error in identifying file ‘/oracle/oradata/boss/testtbs01_01.dbf‘
ORA-27037: unable to obtain file status
Linux Error: 2: No such file or directory
Additional information: 3
SQL> shutdown immediate;
SQL> startup open;
Database mounted.
ORA-01157: cannot identify/lock data file 6 - see DBWR trace file
ORA-01110: data file 6: ‘/oracle/oradata/boss/testtbs01_01.dbf‘
# cd $ORACLE_BASE/admin/boss/bdump
（8）还原和恢复数据文件
SQL> set linesize 150
# mkdir -p /oracle/oradata/boss/others
# cp -rf /oradata/bossbak/testtbs01_01.dbf /oracle/oradata/boss/others
SQL> select
2 ts.name "表空间名"
3 , df.file# "文件号"
4 , df.checkpoint_change# "检查点"
5 , df.name "文件名"
6 , df.status "在线状态"
7 , rf.error "恢复原因"
8 , rf.change# "系统变更号"
9 , rf.time
10 from v$tablespace ts,v$datafile df,v$recover_file rf
11 where ts.ts#=df.ts# and df.file#=rf.file#
12 order by df.file#;
表空间名 文件号 检查点 文件名 在线状 恢复原因 系统变更号 TIME
------------------------------ ---------- ---------- ---------------------------------------- ------- ------------------ ---------- ------------
TESTTBS01 6 570729 /oracle/oradata/boss/testtbs01_01.dbf ONLINE FILE NOT FOUND 0
##离线数据文件
SQL> alter database datafile ‘/oracle/oradata/boss/testtbs01_01.dbf‘ offline;
##还原数据文件
SQL> recover datafile ‘/oracle/oradata/boss/testtbs01_01.dbf‘;
ORA-00283: recovery session canceled due to errors
ORA-01110: data file 6: ‘/oracle/oradata/boss/testtbs01_01.dbf‘
ORA-01157: cannot identify/lock data file 6 - see DBWR trace file
ORA-01110: data file 6: ‘/oracle/oradata/boss/testtbs01_01.dbf‘
SQL> alter database rename file ‘/oracle/oradata/boss/testtbs01_01.dbf‘ to ‘/oracle/oradata/boss/others/testtbs01_01.dbf‘;
alter database rename file ‘/oracle/oradata/boss/testtbs01_01.dbf‘ to ‘/oracle/oradata/boss/others/testtbs01_01.dbf‘
*
ERROR at line 1:
ORA-01511: error in renaming log/data files
ORA-01141: error renaming data file 6 - new file ‘/oracle/oradata/boss/others/testtbs01_01.dbf‘ not found
ORA-01110: data file 6: ‘/oracle/oradata/boss/testtbs01_01.dbf‘
ORA-27041: unable to open file
Linux Error: 13: Permission denied
Additional information: 2
# chown -R oracle:oinstall /oracle
# chown -R oracle:oinstall /oradata
SQL> alter database rename file ‘/oracle/oradata/boss/testtbs01_01.dbf‘ to ‘/oracle/oradata/boss/others/testtbs01_01.dbf‘;
##重做日志恢复
SQL> recover datafile 6;
Media recovery complete.
SQL> alter database datafile ‘/oracle/oradata/boss/others/testtbs01_01.dbf‘ online;
SQL> alter database open;
SQL> select * from boss_new_test;
ID LASTNAME FIRSTNAME USERID STARTDATE USER
---------- ---------- ---------- -------- ------------ ----
2 aa a 00002 04-JUN-14 in
3 bb b 00003 04-JUN-14 in
4 cc c 00004 04-JUN-14 in
5 dd d 00005 04-JUN-14 in
6 ff f 00006 04-JUN-14 in
二次测试：
（1）导出表boss_new_test
$ exp \"sys/a as sysdba\" file=/oradata/bossbak/boss_new_test_20140604.dmp log=/oradata/bossbak/expboss_new_test.log tables=boss_new_test
（2）删除表boss_new_test
SQL> truncate table boss_new_test;
SQL> drop table boss_new_test;
SQL> purge table boss_new_test;
（3）SQL> drop tablespace testtbs01 including contents and datafiles;
(4)
SQL> select group#,members,sequence#,archived,status,first_change# from v$log;
GROUP# MEMBERS SEQUENCE# ARC STATUS FIRST_CHANGE#
---------- ---------- ---------- --- ---------------- -------------
1 1 11 YES INACTIVE 572850
2 1 12 YES INACTIVE 572852
3 1 13 NO CURRENT 572854
SQL> alter system switch logfile;
SQL> alter system switch logfile;
SQL> alter system switch logfile;
alter system switch logfile;
alter system switch logfile;
alter system switch logfile;
（5）备份新的归档日志
$ cp -rf *.arc /oradata/bossbak/test_ar_rec
$ ll
总用量 1724
-rw-r----- 1 oracle oinstall 1735168 6月 4 15:05 o1_mf_1_13_9rxk4vq9_.arc
-rw-r----- 1 oracle oinstall 1024 6月 4 15:05 o1_mf_1_14_9rxk4x7j_.arc
-rw-r----- 1 oracle oinstall 5632 6月 4 15:05 o1_mf_1_15_9rxk52tx_.arc
-rw-r----- 1 oracle oinstall 1024 6月 4 15:05 o1_mf_1_16_9rxk52wv_.arc
-rw-r----- 1 oracle oinstall 1024 6月 4 15:05 o1_mf_1_17_9rxk57y7_.arc
-rw-r----- 1 oracle oinstall 1024 6月 4 15:05 o1_mf_1_18_9rxk5818_.arc
（6）删除testtbs01的某个数据文件，删除所有归档
$ rm -rf testtbs01_01.dbf
$ cd /oracle/flash_recovery_area/BOSS/archivelog/2014_06_04
$ rm -rf *
（7）
SQL> create table boss_new_test1
2 ( id number(5) constraint id_pk primary key,
3 lastname varchar2(10) constraint ln_nn not null,
4 firstname varchar2(10) not null unique,
5 userid varchar2(8) constraint ui_uq unique,
6 startdate date default sysdate,
7 user_type varchar2(4) constraint ut_ck check (user_type in(‘in‘,‘out‘))
8 )
9 initrans 1 maxtrans 255
10 pctfree 20 pctused 50
11 storage(initial 64m next 8m pctincrease 0)
12 tablespace testtbs01;
user_type varchar2(4) constraint ut_ck check (user_type in(‘in‘,‘out‘))
*
ERROR at line 7:
ORA-01565: error in identifying file ‘/oracle/oradata/boss/testtbs01_01.dbf‘
ORA-27037: unable to obtain file status
Linux Error: 2: No such file or directory
Additional information: 3
SQL> select file#,name,status,CHECKPOINT_CHANGE#,recover from v$datafile_header where file#=6;
FILE# NAME STATUS CHECKPOINT_CHANGE# REC
---------- ---------------------------------------- ------- ------------------ ---
6 /oracle/oradata/boss/testtbs01_01.dbf ONLINE 575477 NO
SQL> alter system checkpoint;
SQL> select file#,name,status,CHECKPOINT_CHANGE#,recover from v$datafile_header where file#=6;
FILE# NAME STATUS CHECKPOINT_CHANGE# REC
---------- ---------------------------------------- ------- ------------------ ---
6 /oracle/oradata/boss/testtbs01_01.dbf ONLINE 575795 NO
SQL> shutdown immediate;
SQL> startup open;
Database mounted.
ORA-01157: cannot identify/lock data file 6 - see DBWR trace file
ORA-01110: data file 6: ‘/oracle/oradata/boss/testtbs01_01.dbf‘
（8）
$ cp -rf /oradata/bossbak/testtbs01_01.dbf /oracle/oradata/boss/
SQL> select file#,name,status,CHECKPOINT_CHANGE#,recover from v$datafile_header where file#=6;
FILE# NAME STATUS CHECKPOINT_CHANGE# REC
---------- ---------------------------------------- ------- ------------------ ---
6 /oracle/oradata/boss/testtbs01_01.dbf OFFLINE 575015 YES
SQL> select dbid,name,log_mode,checkpoint_change# from v$database;
DBID NAME LOG_MODE CHECKPOINT_CHANGE#
---------- ---------------------------------------- ------------ ------------------
1375601832 BOSS ARCHIVELOG 576285
SQL> select
2 ts.name "表空间名"
3 , df.file# "文件号"
4 , df.checkpoint_change# "检查点"
5 , df.name "文件名"
6 from v$tablespace ts,v$datafile df
7 where ts.ts#=df.ts# and ts.name=‘TESTTBS01‘
8 order by df.file#;
表空间名 文件号 检查点 文件名
------------------------------ ---------- ---------- ----------------------------------------
TESTTBS01 6 576285 /oracle/oradata/boss/testtbs01_01.dbf
##多次日志切换，只能从归档日志恢复
SQL> recover datafile ‘/oracle/oradata/boss/testtbs01_01.dbf‘;
ORA-00279: change 575015 generated at 06/04/2014 14:55:59 needed for thread 1
ORA-00289: suggestion : /oracle/flash_recovery_area/BOSS/archivelog/2014_06_04/o1_mf_1_13_%u_.arc
ORA-00280: change 575015 for thread 1 is in sequence #13
Specify log: {<RET>=suggested | filename | AUTO | CANCEL}
##拷贝所需归档日志，输入auto
$ cp -rf /oradata/bossbak/test_ar_rec/*.arc /oracle/flash_recovery_area/BOSS/archivelog/2014_06_04
SQL> recover datafile ‘/oracle/oradata/boss/testtbs01_01.dbf‘;
ORA-00279: change 575015 generated at 06/04/2014 14:55:59 needed for thread 1
ORA-00289: suggestion : /oracle/flash_recovery_area/BOSS/archivelog/2014_06_04/o1_mf_1_13_%u_.arc
ORA-00280: change 575015 for thread 1 is in sequence #13
Specify log: {<RET>=suggested | filename | AUTO | CANCEL}
auto
ORA-00279: change 575465 generated at 06/04/2014 15:02:51 needed for thread 1
ORA-00289: suggestion : /oracle/flash_recovery_area/BOSS/archivelog/2014_06_04/o1_mf_1_14_%u_.arc
ORA-00280: change 575465 for thread 1 is in sequence #14
ORA-00278: log file ‘/oracle/flash_recovery_area/BOSS/archivelog/2014_06_04/o1_mf_1_13_9rxk4vq9_.arc‘ no longer needed
for this recovery
ORA-00279: change 575467 generated at 06/04/2014 15:02:53 needed for thread 1
ORA-00289: suggestion : /oracle/flash_recovery_area/BOSS/archivelog/2014_06_04/o1_mf_1_15_%u_.arc
ORA-00280: change 575467 for thread 1 is in sequence #15
ORA-00278: log file ‘/oracle/flash_recovery_area/BOSS/archivelog/2014_06_04/o1_mf_1_14_9rxk4x7j_.arc‘ no longer needed
for this recovery
ORA-00279: change 575470 generated at 06/04/2014 15:02:58 needed for thread 1
ORA-00289: suggestion : /oracle/flash_recovery_area/BOSS/archivelog/2014_06_04/o1_mf_1_16_%u_.arc
ORA-00280: change 575470 for thread 1 is in sequence #16
ORA-00278: log file ‘/oracle/flash_recovery_area/BOSS/archivelog/2014_06_04/o1_mf_1_15_9rxk52tx_.arc‘ no longer needed
for this recovery
Log applied.
Media recovery complete.
SQL> alter database datafile ‘/oracle/oradata/boss/testtbs01_01.dbf‘ online;
SQL> alter database open;
SQL> select * from boss_new_test;
ID LASTNAME FIRSTNAME USERID STARTDATE USER
---------- ---------- ---------- -------- ------------ ----
1 aaa a 00001 04-JUN-14 in
2 bbb b 00002 04-JUN-14 in
3 ccc c 00003 04-JUN-14 in
4 ddd d 00004 04-JUN-14 in
5 eee e 00005 04-JUN-14 in
6 fff f 00006 04-JUN-14 in
总结：数据文件恢复成功后，数据库scn、数据文件scn是相同的，它们和数据文件头scn不一定相同，但是数据文件头scn>=数据库scn
Performing User-Managed Database-18.4、Restoring Datafiles and Archived Redo Logs,布布扣,bubuko.com
Performing User-Managed Database-18.4、Restoring Datafiles and Archived Redo Logs