[20180423]flashback tablespace与snapshot standby.txt

--//缺省建立表空间是打开flashback on,如果某个表空间flashback off,在dg启动snapshot standby时注意,可能"回不来",
--//通过测试说明问题.

1.环境:
SCOTT@book> @ ver1
PORT_STRING                    VERSION        BANNER
------------------------------ -------------- --------------------------------------------------------------------------------
x86_64/Linux 2.4.xx            11.2.0.4.0     Oracle Database 11g Enterprise Edition Release 11.2.0.4.0 - 64bit Production

2.设置备库tea表空间关闭flashback.

SCOTT@book> alter tablespace tea flashback off;
Tablespace altered.

SCOTT@book> select * from v$tablespace;
         TS# NAME      INC BIG FLA ENC
------------ --------- --- --- --- ---
           0 SYSTEM    YES NO  YES
           1 SYSAUX    YES NO  YES
           2 UNDOTBS1  YES NO  YES
           4 USERS     YES NO  YES
           3 TEMP      NO  NO  YES
           6 EXAMPLE   YES NO  YES
           7 TEA       YES NO  NO
7 rows selected.

--//tea表空间FLASHBACK_ON设置为NO.注意这些信息应该记录在控制文件,你可以发现备库还是On .
--//备库:
SYS@bookdg> select flashback_on from v$database;
FLASHBACK_ON
------------------
NO

SYS@bookdg> select * from v$tablespace;
       TS# NAME                 INC BIG FLA ENC
---------- -------------------- --- --- --- ---
         0 SYSTEM               YES NO  YES
         1 SYSAUX               YES NO  YES
         2 UNDOTBS1             YES NO  YES
         4 USERS                YES NO  YES
         3 TEMP                 NO  NO  YES
         6 EXAMPLE              YES NO  YES
         7 TEA                  YES NO  YES
7 rows selected.
--//备库还是yes,tea表空间.

SYS@bookdg> alter tablespace tea flashback off;
alter tablespace tea flashback off
                     *
ERROR at line 1:
ORA-16000: database open for read-only access
--//无法在open read only修改.

SYS@bookdg> startup mount
ORACLE instance started.
Total System Global Area  634732544 bytes
Fixed Size                  2255792 bytes
Variable Size             197133392 bytes
Database Buffers          427819008 bytes
Redo Buffers                7524352 bytes
Database mounted.
SYS@bookdg> alter tablespace tea flashback off;
Tablespace altered.

SYS@bookdg> select * from v$tablespace;
       TS# NAME                 INC BIG FLA ENC
---------- -------------------- --- --- --- ---
         0 SYSTEM               YES NO  YES
         1 SYSAUX               YES NO  YES
         2 UNDOTBS1             YES NO  YES
         4 USERS                YES NO  YES
         3 TEMP                 NO  NO  YES
         6 EXAMPLE              YES NO  YES
         7 TEA                  YES NO  NO
7 rows selected.
--//OK,现在成功.

3.备库打开snapshot standby:
--//参考http://blog.itpub.net/267265/viewspace-2134547/
--//实际上就是保证存储点,只要闪回区足够,许多dml操作没有问题,在转换physical standby时,返回原来的存储点;

--//备库:
SYS@bookdg> alter database convert to snapshot standby;
Database altered.

SYS@bookdg> select database_role from v$database;
DATABASE_ROLE
----------------
SNAPSHOT STANDBY

SYS@bookdg> alter database open ;
Database altered.

SCOTT@bookdg> create table tt1 tablespace tea as select * from dba_objects;
Table created.

--//在主库也产生一些日志对于tea表空间:
SCOTT@book> create table empx tablespace tea as select * from emp;
Table created.

--//现在转换为physical standby
--//备库:
SYS@bookdg> shutdown immediate
Database closed.
Database dismounted.
ORACLE instance shut down.

SYS@bookdg> startup mount
ORACLE instance started.
Total System Global Area  634732544 bytes
Fixed Size                  2255792 bytes
Variable Size             197133392 bytes
Database Buffers          427819008 bytes
Redo Buffers                7524352 bytes
Database mounted.

SYS@bookdg> alter database convert to physical standby ;
alter database convert to physical standby
*
ERROR at line 1:
ORA-38753: Cannot flashback data file 6; no flashback log data.
ORA-01110: data file 6: '/mnt/ramdisk/book/tea01.dbf'

--//可以发现备库的数据文件6无法转换,这个应该引起足够重视,在备库转换为snapshot standby时,注意检查表空间是否flashvback是否都是on的状态.
$ oerr ora 38753
38753, 00000, "Cannot flashback data file %s; no flashback log data."
// *Cause:  An attempt to perform a FLASHBACK DATABASE failed because the file
//          does not have enough flashback log data to cover the time to
//          flash back.  Either the file did not have flashback generation
//          enabled for it, or had flashback generation turned off for it
//          some time during the time span of the flashback.
// *Action: The file cannot be flashed back.  The file must be taken offline
//          or the tablespace dropped before continuing with the FLASHBACK
//          DATABASE command.

4.恢复:
--//简单一点,主库tea表空间设置为read only;
--//主库:
SCOTT@book> alter tablespace tea read only;
Tablespace altered.

$ scp /mnt/ramdisk/book/tea01.dbf oracle@192.168.100.40:/mnt/ramdisk/book/
tea01.dbf                            100%   40MB  40.0MB/s   00:01

--//备库:
SYS@bookdg> alter database convert to physical standby ;
alter database convert to physical standby
*
ERROR at line 1:
ORA-38753: Cannot flashback data file 6; no flashback log data.
ORA-01110: data file 6: '/mnt/ramdisk/book/tea01.dbf'
--//不行:

SYS@bookdg> alter tablespace tea flashback on;
Tablespace altered.

SYS@bookdg> alter database convert to physical standby ;
alter database convert to physical standby
*
ERROR at line 1:
ORA-19926: Database cannot be converted at this time

--//根本不能这样转换.

SYS@bookdg> alter tablespace tea offline;
alter tablespace tea offline
*
ERROR at line 1:
ORA-01109: database not open

SYS@bookdg> alter database datafile 6 offline;
Database altered.

--//关闭备库重来.
SYS@bookdg> startup mount
ORACLE instance started.
Total System Global Area  634732544 bytes
Fixed Size                  2255792 bytes
Variable Size             197133392 bytes
Database Buffers          427819008 bytes
Redo Buffers                7524352 bytes
Database mounted.

SYS@bookdg> alter database convert to physical standby ;
Database altered.

SYS@bookdg> shutdown immediate
ORA-01507: database not mounted
ORACLE instance shut down.

SYS@bookdg> startup mount
ORACLE instance started.
Total System Global Area  634732544 bytes
Fixed Size                  2255792 bytes
Variable Size             197133392 bytes
Database Buffers          427819008 bytes
Redo Buffers                7524352 bytes
Database mounted.

SYS@bookdg> alter database datafile 6 online;
Database altered.

SYS@bookdg> alter database recover managed standby database using current logfile disconnect ;
Database altered.

SYS@bookdg> @ &r/dg/dg
PROCESS       PID STATUS       CLIENT_P GROUP# THREAD#  SEQUENCE#     BLOCK#     BLOCKS DELAY_MINS
--------- ------- ------------ -------- ------ ------- ---------- ---------- ---------- ----------
ARCH         2394 CONNECTED    ARCH     N/A          0          0          0          0          0
ARCH         2396 CONNECTED    ARCH     N/A          0          0          0          0          0
ARCH         2398 CONNECTED    ARCH     N/A          0          0          0          0          0
RFS          2403 IDLE         UNKNOWN  N/A          0          0          0          0          0
RFS          2405 IDLE         LGWR     2            1        789         50          1          0
ARCH         2400 CLOSING      ARCH     4            1        788          1        183          0
MRP0         2407 APPLYING_LOG N/A      N/A          1        789         50     102400          0
7 rows selected.
--//日志开始传输并应用.

SYS@bookdg> alter database recover managed standby database cancel ;
Database altered.

SYS@bookdg> alter database open read only;
Database altered.

SYS@bookdg> alter database recover managed standby database using current logfile disconnect ;
Database altered.

SYS@bookdg> select count(*) from scott.empx;
  COUNT(*)
----------
        14

SYS@bookdg> select count(*) from scott.tt1;
select count(*) from scott.tt1
                           *
ERROR at line 1:
ORA-00942: table or view does not exist
--//测试时建立的表不存在.
        
--//主库执行:
SCOTT@book> alter tablespace tea read write;
Tablespace altered.

SCOTT@book> delete from empx where rownum=1;
1 row deleted.

SCOTT@book> commit ;
Commit complete.

--//在备库检查,发现少一条记录.
SYS@bookdg> select count(*) from scott.empx;
  COUNT(*)
----------
        13

总结:
--//这个在以后工作中注意,在转换snapshot standby,注意表空间flaashback是否在on状态.

[20180423]flashback tablespace与snapshot standby.txt的更多相关文章

  1. 聊聊Oracle 11g的Snapshot Standby Database(上)

    Oracle 11g是Data Guard的重要里程碑版本.在11g中,Active DataGuard.Advanced Compression等特性大大丰富了Data Guard的功能和在实践领域 ...

  2. Oracle 11g中的snapshot standby特性

    在Oracle 11g中,data guard最吸引人的,除了active data guard的实时查询特性(即可以以只读方式打开物理standby数据库的同时MRP进程能继续做recover),快 ...

  3. 聊聊Oracle 11g的Snapshot Standby Database(下)

    3.Snapshot Standby行为研究 下面我们分析一下Snapshot Standby的工作性质和行为性质.我们在主库方向研究当前状态. --主库日志情况 SQL> select gro ...

  4. Snapshot Standby

    INTRODUCTION Snapshot standby database是ORACLE 11g的新特性.允许Physical standby短时间的使用read write模式. Snapshot ...

  5. snapshot standby database

    快照备库接收和归档主库发送来的redo,但是不会应用:切换成physical standby之后会自动开启redo apply.快照standby不可以参加主备切换:在最大保护性模式下,如果只有一个备 ...

  6. 转 Oracle Transportable TableSpace(TTS) 传输表空间 说明

    ############1   迁移数据库的集中方法 三.相关技术 迁移方式 优势 不足1 Export and import • 对数据库版本,以及系统平台没有要求 • 不支持并发,速度慢• 停机时 ...

  7. V4 Reduce Transportable Tablespace Downtime using Cross Platform Incremental Backup (Doc ID 2471245.1)

    V4 Reduce Transportable Tablespace Downtime using Cross Platform Incremental Backup (Doc ID 2471245. ...

  8. Oracle 11.2.0.4.0 Dataguard部署和日常维护(6)-Dataguard Snapshot篇

    1. 检查当前主备库同步状态 on primary select ads.dest_id,max(sequence#) "Current Sequence", max(log_se ...

  9. [Oracle] 临时将Physical Standby激活

    Oracle 10g/11g下如何将物理Standby库临时激活用于测试 在实际运营环境中, 我们经常碰到类似这样的需求: 譬如想不影响现网业务评估DB补丁在现网环境中运行的时间, 或者是想在做DB切 ...

随机推荐

  1. static加载顺序简介

    1.先执行父类的静态代码块和静态变量初始化,并且静态代码块和静态变量的执行顺序只跟代码中出现的顺序有关. 2.执行子类的静态代码块和静态变量初始化. 3.执行父类的实例变量初始化 4.执行父类的构造函 ...

  2. [Node.js与数据库]node-mysql 模块介绍

    [Node.js与数据库]node-mysql 模块介绍   转载至:https://itbilu.com/nodejs/npm/NyPG8LhlW.html#multiple-statement-q ...

  3. 用DDD模拟案例分析

    之前我写了几篇关于DDD的介绍和一些小例子说明,我想这对于介绍DDD还是有些模糊,甚至还不知道怎么用DDD来分析设计.昨天和园友讨论也发现没有例子很难说明,所以今天我模拟了一个案例,同时这个案例也是真 ...

  4. 21天打造分布式爬虫-urllib库(一)

    1.1.urlopen函数的用法 #encoding:utf-8 from urllib import request res = request.urlopen("https://www. ...

  5. leetcode — remove-nth-node-from-end-of-list

    /** * Source : https://oj.leetcode.com/problems/remove-nth-node-from-end-of-list/ * * Created by lve ...

  6. leetcode — zigzag-conversion

    /** * Source : https://oj.leetcode.com/problems/zigzag-conversion/ * * Created by lverpeng on 2017/6 ...

  7. 了解Job和JobDeatil ,JobDataMap (三)

    一:定义 Job:实现任务逻辑的接口. JobDeatil:JobDeatil为Job提供了许多设置属性,以及JobDataMap成员变量属性,他用来储存特定的Job实例状态信息,调度器需要使用Job ...

  8. DNS Wildcard(DNS泛域名)

    在DNS中,泛域名(wildcard Resource Record)可以被认为是一种合成RR的机制,借助于它,DNS服务器可以响应本来不存在的域名的请求,它的设计初衷是用来把所有邮件都转发到一个邮件 ...

  9. Python 浅谈编程规范和软件开发目录规范的重要性

    最近参加了一个比赛,然后看到队友编程的代码,我觉得真的是觉得注释和命名规范的重要性了,因为几乎每个字符都要咨询他,用老师的话来说,这就是命名不规范的后续反应.所以此时的我意识到写一篇关于注释程序的重要 ...

  10. vuex状态管理之学习笔记

    概述及使用场景 Vuex 是一个主要应用在中大型单页应用的类似于 Flux 的数据管理架构.它主要帮我们更好地组织代码,以及把应用内的的状态保持在可维护.可理解的状态. 但如果是简单的应用 ,就没有必 ...