如何使用dd备份数据文件头部和ASM磁盘头部

Posted 2023-04-12

参考技术A 在文件系统下用dd备份文件头部做测试
SQL> select name from v$datafile;

NAME
--------------------------------------------------------------------------------
/oradata/upgr/system01.dbf
/oradata/upgr/sysaux01.dbf
/oradata/upgr/undotbs01.dbf
/oradata/upgr/users01.dbf
SQL>

SQL> set lines 156 pages 100
SQL> col dd_backup_cmd for a156
SQL> col dd_restore_cmd for a156
SQL> select 'dd if='||name||' of=/backup/'||substr(name,instr(name,'/',-1)+1)||'.FH bs='||block_size||' count=2' dd_backup_cmd from v$datafile;
DD_BACKUP_CMD

--------------------------------------------------------------------------------
dd if=/oradata/upgr/system01.dbf of=/backup/system01.dbf.FH bs=8192 count=2
dd if=/oradata/upgr/sysaux01.dbf of=/backup/sysaux01.dbf.FH bs=8192 count=2
dd if=/oradata/upgr/undotbs01.dbf of=/backup/undotbs01.dbf.FH bs=8192 count=2
dd if=/oradata/upgr/users01.dbf of=/backup/users01.dbf.FH bs=8192 count=2

SQL> select 'dd of='||name||' if=/backup/'||substr(name,instr(name,'/',-1)+1)||'.FH bs='||block_size||' count=2 conv=notrunc' dd_restore_cmd from v$datafile;
DD_RESTORE_CMD

------------------------------------------------------------------------------------------
dd of=/oradata/upgr/system01.dbf if=/backup/system01.dbf.FH bs=8192 count=2 conv=notrunc
dd of=/oradata/upgr/sysaux01.dbf if=/backup/sysaux01.dbf.FH bs=8192 count=2 conv=notrunc
dd of=/oradata/upgr/undotbs01.dbf if=/backup/undotbs01.dbf.FH bs=8192 count=2 conv=notrunc
dd of=/oradata/upgr/users01.dbf if=/backup/users01.dbf.FH bs=8192 count=2 conv=notrunc

SQL>

[oracle@db1 ~]$ dd if=/oradata/upgr/system01.dbf of=/backup/system01.dbf.FH bs=8192 count=2

2+0 records in
2+0 records out
16384 bytes (16 kB) copied, 0.000412654 s, 39.7 MB/s

破坏数据文件头部

[oracle@db1 ~]$ dd if=/dev/zero of=/oradata/upgr/system01.dbf bs=8192 count=2 conv=notrunc
2+0 records in
2+0 records out
16384 bytes (16 kB) copied, 0.000434433 s, 37.7 MB/s

[oracle@db1 ~]$ sqlplus / as sysdba

SQL*Plus: Release 11.2.0.4.0 Production on Thu Jun 16 15:23:33 2016
Copyright (c) 1982, 2013, Oracle. All rights reserved.
Connected to an idle instance.
SQL> startup
ORACLE instance started.

Total System Global Area 313159680 bytes
Fixed Size 2252824 bytes
Variable Size 167776232 bytes
Database Buffers 138412032 bytes
Redo Buffers 4718592 bytes
Database mounted.
ORA-01157: cannot identify/lock data file 1 - see DBWR trace file
ORA-01110: data file 1: '/oradata/upgr/system01.dbf'

SQL> shutdown abort
ORACLE instance shut down.
SQL> exit
Disconnected from Oracle Database 11g Enterprise Edition Release 11.2.0.4.0 - 64bit Production
With the Partitioning, Data Mining and Real Application Testing options

恢复数据文件头部

[oracle@db1 ~]$ dd of=/oradata/upgr/system01.dbf if=/backup/system01.dbf.FH bs=8192 count=2 conv=notrunc
2+0 records in
2+0 records out
16384 bytes (16 kB) copied, 0.000574576 s, 28.5 MB/s

[oracle@db1 ~]$ sqlplus / as sysdba

SQL*Plus: Release 11.2.0.4.0 Production on Thu Jun 16 15:24:32 2016
Copyright (c) 1982, 2013, Oracle. All rights reserved.
Connected to an idle instance.
SQL> startup
ORACLE instance started.

Total System Global Area 313159680 bytes
Fixed Size 2252824 bytes
Variable Size 167776232 bytes
Database Buffers 138412032 bytes
Redo Buffers 4718592 bytes
Database mounted.
Database opened.
SQL> exit
Disconnected from Oracle Database 11g Enterprise Edition Release 11.2.0.4.0 - 64bit Production
With the Partitioning, Data Mining and Real Application Testing options
[oracle@db1 ~]$

在ASM情况下用dd备份文件头部做测试
SQL> select ' dd '||
2 ' if=/backup/'||e.group_kffxp||'_'||e.disk_kffxp||'_'||a.file_number||'.FH'||
3 ' of='||d.path||
4 ' bs='||c.allocation_unit_size||
5 ' seek='||e.au_kffxp||
6 ' count=1 conv=notrunc' dd_restore_cmd
7 from
8 v$asm_alias a,
9 v$asm_file b,
10 v$asm_diskgroup c,
11 v$asm_disk d,
12 x$kffxp e
13 where a.file_number=e.number_kffxp
14 and a.group_number=e.group_kffxp
15 and b.group_number=a.group_number
16 and d.group_number=c.group_number
17 and e.group_kffxp=d.group_number
18 and a.file_number=b.file_number
19 and d.disk_number=e.disk_kffxp
20 and a.system_created='Y'
21 and b.type='DATAFILE'
22 and e.xnum_kffxp=0
23 order by a.file_number
24 /

DD_RESTORE_CMD

--------------------------------------------------------------------------------------------
dd if=/backup/1_3_283.FH of=/dev/oracleasm/disks/ASMDISK7 bs=1048576 seek=21 count=1 conv=notrunc
dd if=/backup/1_4_285.FH of=/dev/oracleasm/disks/ASMDISK8 bs=1048576 seek=262 count=1 conv=notrunc
dd if=/backup/1_2_287.FH of=/dev/oracleasm/disks/ASMDISK6 bs=1048576 seek=266 count=1 conv=notrunc
dd if=/backup/1_1_288.FH of=/dev/oracleasm/disks/ASMDISK5 bs=1048576 seek=299 count=1 conv=notrunc
dd if=/backup/1_3_291.FH of=/dev/oracleasm/disks/ASMDISK7 bs=1048576 seek=163 count=1 conv=notrunc

SQL>
SQL> select ' dd'||
2 ' if='||d.path||
3 ' of=/backup/'||e.group_kffxp||'_'||e.disk_kffxp||'_'||a.file_number||'.FH'||
4 ' bs='||c.allocation_unit_size||
5 ' skip='||e.au_kffxp||
6 ' count=1' dd_backup_cmd
7 from
8 v$asm_alias a,
9 v$asm_file b,
10 v$asm_diskgroup c,
11 v$asm_disk d,
12 x$kffxp e
13 where a.file_number=e.number_kffxp
14 and a.group_number=e.group_kffxp
15 and b.group_number=a.group_number
16 and d.group_number=c.group_number
17 and e.group_kffxp=d.group_number
18 and a.file_number=b.file_number
19 and d.disk_number=e.disk_kffxp
20 and a.system_created='Y'
21 and b.type='DATAFILE'
22 and e.xnum_kffxp=0
23 order by a.file_number
24 /

DD_BACKUP_CMD
---------------------------------------------------------------------------------------------------------------

dd if=/dev/oracleasm/disks/ASMDISK7 of=/backup/1_3_283.FH bs=1048576 skip=21 count=1
dd if=/dev/oracleasm/disks/ASMDISK8 of=/backup/1_4_285.FH bs=1048576 skip=262 count=1
dd if=/dev/oracleasm/disks/ASMDISK6 of=/backup/1_2_287.FH bs=1048576 skip=266 count=1
dd if=/dev/oracleasm/disks/ASMDISK5 of=/backup/1_1_288.FH bs=1048576 skip=299 count=1
dd if=/dev/oracleasm/disks/ASMDISK7 of=/backup/1_3_291.FH bs=1048576 skip=163 count=1

SQL>
备份ASM中数据文件头部
[oracle@ohs1 ~]$ dd if=/dev/oracleasm/disks/ASMDISK7 of=/backup/1_3_283.FH bs=1048576 skip=21 count=1

1+0 records in
1+0 records out
1048576 bytes (1.0 MB) copied, 0.00664794 seconds, 158 MB/s

破坏ASM中数据文件头部

[oracle@ohs1 ~]$ dd if=/dev/zero of=/dev/oracleasm/disks/ASMDISK7 seek=21 count=1 conv=notrunc bs=1M
1+0 records in
1+0 records out
1048576 bytes (1.0 MB) copied, 0.000517497 seconds, 2.0 GB/s

[oracle@ohs1 ~]$ sqlplus / as sysdba

SQL*Plus: Release 11.2.0.3.0 Production on Thu Jun 16 14:58:53 2016
Copyright (c) 1982, 2011, Oracle. All rights reserved.
Connected to an idle instance.
SQL> startup
ORACLE instance started.

Total System Global Area 318046208 bytes
Fixed Size 1344680 bytes
Variable Size 188746584 bytes
Database Buffers 121634816 bytes
Redo Buffers 6320128 bytes
Database mounted.
ORA-01122: database file 1 failed verification check
ORA-01110: data file 1: '+DATA_PGOLD/racdb/system01.dbf'
ORA-01210: data file header is media corrupt

SQL> shut abort
ORACLE instance shut down.
SQL> exit
Disconnected from Oracle Database 11g Enterprise Edition Release 11.2.0.3.0 - Production
With the Partitioning, Real Application Clusters, Automatic Storage Management, OLAP,

Data Mining and Real Application Testing options
恢复ASM中数据文件头部

[oracle@ohs1 ~]$ dd if=/backup/1_3_283.FH of=/dev/oracleasm/disks/ASMDISK7 bs=1048576 seek=21 count=1 conv=notrunc

1+0 records in
1+0 records out
1048576 bytes (1.0 MB) copied, 0.00153561 seconds, 683 MB/s

[oracle@ohs1 ~]$ sqlplus / as sysdba
SQL*Plus: Release 11.2.0.3.0 Production on Thu Jun 16 15:00:08 2016
Copyright (c) 1982, 2011, Oracle. All rights reserved.
Connected to an idle instance.
SQL> startup
ORACLE instance started.
Total System Global Area 318046208 bytes
Fixed Size 1344680 bytes
Variable Size 188746584 bytes
Database Buffers 121634816 bytes
Redo Buffers 6320128 bytes
Database mounted.
Database opened.
SQL> select open_mode from v$database;
OPEN_MODE
--------------------
READ WRITE
SQL>

生成备份数据文件头部所需要的dd脚本(数据库用户)

set lines 156 pages 100
col dd_backup_cmd for a156
col dd_restore_cmd for a156
select 'dd if='||name||' of=/backup/'||substr(name,instr(name,'/',-1)+1)||'.FH bs='||block_size||' count=2' dd_backup_cmd from v$datafile;
select 'dd of='||name||' if=/backup/'||substr(name,instr(name,'/',-1)+1)||'.FH bs='||block_size||' count=2 conv=notrunc' dd_restore_cmd from v$datafile;

生成备份ASM数据文件头部所需要的dd脚本(在GI用户下ASM实例)

set lines 156 pages 100
col dd_backup_cmd for a156
col dd_restore_cmd for a156
select ' dd'||
' if='||d.path||
' of=/backup/'||e.group_kffxp||'_'||e.disk_kffxp||'_'||a.file_number||'.FH'||
' bs='||c.allocation_unit_size||
' skip='||e.au_kffxp||
' count=1' dd_backup_cmd
from
v$asm_alias a,
v$asm_file b,
v$asm_diskgroup c,
v$asm_disk d,
x$kffxp e
where a.file_number=e.number_kffxp
and a.group_number=e.group_kffxp
and b.group_number=a.group_number
and d.group_number=c.group_number
and e.group_kffxp=d.group_number
and a.file_number=b.file_number
and d.disk_number=e.disk_kffxp
and a.system_created='Y'
and b.type='DATAFILE'
and e.xnum_kffxp=0
order by a.file_number
/

select ' dd '||
' if=/backup/'||e.group_kffxp||'_'||e.disk_kffxp||'_'||a.file_number||'.FH'||
' of='||d.path||
' bs='||c.allocation_unit_size||
' seek='||e.au_kffxp||
' count=1 conv=notrunc' dd_restore_cmd
from
v$asm_alias a,
v$asm_file b,
v$asm_diskgroup c,
v$asm_disk d,
x$kffxp e
where a.file_number=e.number_kffxp
and a.group_number=e.group_kffxp
and b.group_number=a.group_number
and d.group_number=c.group_number
and e.group_kffxp=d.group_number
and a.file_number=b.file_number
and d.disk_number=e.disk_kffxp
and a.system_created='Y'
and b.type='DATAFILE'
and e.xnum_kffxp=0
order by a.file_number
/

生成备份ASM磁盘头部所需要的dd脚本(在GI用户下ASM实例)

set lines 150
set pages 1000
select
'dd if='||path||
' of=/backup/'||substr(path,instr(path,'/',-1)+1)||'_'||group_number||'_'||disk_number||'.DH bs=1M count=100'
from v$asm_disk asm_header_backup
where header_status='MEMBER'
order by group_number,disk_number;

select 'dd if=/backup/'||substr(path,instr(path,'/',-1)+1)||'_'||group_number||'_'||disk_number||'.DH'||
' of='||path||' bs=1M count=100 conv=notrunc'
from v$asm_disk asm_header_restore
where header_status='MEMBER'
order by group_number,disk_number;

生成asmcmd cp命令(数据库用户下执行)
select 'asmcmd cp '||name||' /backup/'||substr(name,instr(name,'/',-1)+1) asmcmd_cp from v$controlfile
union all
select 'asmcmd cp '||name||' /backup/'||substr(name,instr(name,'/',-1)+1) from v$datafile
union all
select 'asmcmd cp '||member||' /backup/'||thread#||'_'||a.group#||'_'||substr(member,instr(member,'/',-1)+1)
from v$log a, v$logfile b where a.group# = b.group#;
ASMCMD_CP
--------------------------------------------------------------------------------

asmcmd cp +DATA_PGOLD/racdb/control01.ctl /backup/control01.ctl
asmcmd cp +DATA_PGOLD/racdb/control02.ctl /backup/control02.ctl
asmcmd cp +DATA_PGOLD/racdb/system01.dbf /backup/system01.dbf
asmcmd cp +DATA_PGOLD/racdb/sysaux01.dbf /backup/sysaux01.dbf
asmcmd cp +DATA_PGOLD/racdb/undotbs01.dbf /backup/undotbs01.dbf
asmcmd cp +DATA_PGOLD/racdb/users01.dbf /backup/users01.dbf
asmcmd cp +DATA_PGOLD/racdb/undotbs02.dbf /backup/undotbs02.dbf
asmcmd cp +DATA_PGOLD/racdb/redo02.log /backup/1_2_redo02.log
asmcmd cp +DATA_PGOLD/racdb/redo01.log /backup/1_1_redo01.log
asmcmd cp +DATA_PGOLD/racdb/redo03.log /backup/2_3_redo03.log
asmcmd cp +DATA_PGOLD/racdb/redo04.log /backup/2_4_redo04.log
11 rows selected.
SQL>

然后执行以上命令即可

生成rman copy命令(数据库用户下执行)
select 'copy current controlfile to ''/tmp/'||substr(name,instr(name,'/',-1)+1)||''';' rman_copy from v$controlfile
union all
select 'copy datafile '||file#||' to ''/backup/'||substr(name,instr(name,'/',-1)+1)||''';' from v$datafile;
RMAN_COPY
--------------------------------------------------------------------------------

copy current controlfile to '/tmp/control01.ctl';
copy current controlfile to '/tmp/control02.ctl';
copy datafile 1 to '/backup/system01.dbf';
copy datafile 2 to '/backup/sysaux01.dbf';
copy datafile 3 to '/backup/undotbs01.dbf';
copy datafile 4 to '/backup/users01.dbf';
copy datafile 5 to '/backup/undotbs02.dbf';

7 rows selected.
SQL>

然后使用rman这些以上命令

Joomla - 如何使用用户在模板参数中提供的信息从头部取消设置 JS 文件

【中文标题】Joomla - 如何使用用户在模板参数中提供的信息从头部取消设置 JS 文件

【英文标题】：Joomla - how to Unset JS file from head using info given by user in template parameters

【发布时间】：2015-09-04 22:46:41

【问题描述】：

我已经使用下面的常用方法成功地从 Joomla 模板 HTML 头手动取消设置 JS 和 CSS 文件。

unset($this->_scripts['/media/jui/js/bootstrap.min.js']);
unset($this->_stylesheets['/media/jui/js/bootstrap.min.css']);

我的问题是这样的。是否可以将文件和路径信息存储到变量中，然后在上述示例中使用变量代替路径/文件？如果是这样，有什么好的方法可以做到这一点？

原因？我想让我的模板的用户能够在模板参数中输入多个用逗号分隔的文件/路径，将这些不同的文件/路径存储到一个数组中，然后循环遍历该数组（从 HTML 头中取消设置/删除每个文件/路径模板）。

目的？有时需要从 html 头部和页面底部删除 js，以减少页面加载时间、页面过渡闪烁/白色闪烁等。在这种情况下，我不想使用插件。如果可能的话，我宁愿给我的用户一些细粒度的控制。当然，我会在模板参数中留出一个位置，供用户将他们从标题中删除的任何 js 添加到底部。

我正在使用的代码？

// grab the js file paths user enters into template parameters and add them to an array i can loop through

$unsetJSFiles = $this->params->get( 'customFilesUnsetJS' );
$unsetJSFilesArray = explode(',', $unsetJSFiles);

// loop through each js file path user entered and unset each one from joomla html head  

foreach ($unsetJSFilesArray as $valueJS) 
    $doc = JFactory::getDocument();
    unset($doc->_scripts[$valueJS]);
    

它似乎不起作用。我已经成功地测试了这些值是否通过 print_r、echo 等输出了我正在寻找的内容。但是循环通过不起作用。这是我在 *** 上提出的第一个真正的问题。研究了很多线程并用谷歌搜索，但我找不到答案。非常感谢您的任何建议！

【参考方案1】：

尝试在循环外（循环上方）声明$doc！

你应该用这个取消它并声明根！

unset($this->_scripts[JURI::root(true) . '/media/jui/js/jquery.min.js']);

所以试试（未测试！！）：

// grab the js file paths user enters into template parameters and add them to an array i can loop through

$doc = JFactory::getDocument();
$unsetJSFiles = $this->params->get( 'customFilesUnsetJS' );
$unsetJSFilesArray = explode(',', $unsetJSFiles);

// loop through each js file path user entered and unset each one from joomla html head  

foreach ($unsetJSFilesArray as $valueJS)         
    unset($this->_scripts[JURI::root(true) . $valueJS]);