DG学习笔记(5)_Standby Redo Log

目录

1 Standby Redo Logs 概念

2 RFS 进程不写Standby Redo Log 的情形

3创建Standby Redo Log

4 Standby Redo Log带来的优势

1 Standby Redo Logs 概念

Standby redo logs are required for physical standby databases running in Maximum protection or maximum available mode and are recommanded for maximum performance.

Logical standby database do not support standby redo logs, they can be allocated on a logical standby database, but they will not be used.

(最大保护模式和最大可用性模式中是必须的.而在最大性能中不是必须的,在逻辑Standby中不需要)

Standby redo logs must be archived before the data can be applied to the standby database, the standby archival operation occurs automatically.

Standby redo logs configuration must be identical to the primary database online log configuration.(same number and size). If you have different sized online log files, the RFS process will automatically switch to the same size standby redo log as needed.

(Standby Redo Log的大小设置应与Primary端的Redo Log 设置一样)

2 RFS 进程不写Standby Redo Log 的情形

The RFS process will not write the standby redo logs and will write an archive log if:

(1) There are no standby redo logs

(2) It cannot find the same size standby redo log as the incoming online log file.

(3) All of the standby redo logs of the correct size are not yet archived.

3 创建Standby Redo Log

SQL>ALTER DATABASE ADD STANDBY LOGFILE (’/oracle/dbs/log1c.rdo’, ’/oracle/dbs/log2c.rdo’)  SIZE 500K;

SQL> ALTER DATABASE ADD STANDBY LOGFILE MEMBER ’/oracle/dbs/log2b.rdo’  TO GROUP 2;

Oracle Corporation recommends that you create standby redo logs on the primary database so that switching roles is easier and without additional DBA intervention.

一般情况下,两边的参数基本一致,方便SwitchOver.

4        Standby Redo Log带来的优势

Standby redo logs form. a separate pool of log file groups and provide the following advantages over archived online redo logs:

Because standby redo logs are preallocated files, they avoid the operating system overhead of file system metadata updates common with sequential files.

Standby redo logs can reside on raw devices, thus providing I/O write performance equal to that of the primary database.

Standby redo logs can be multiplexed using multiple members, improving reliability over archived redo logs.

the easiest way to determine if the current standby log configuration is satisfactory is to examine the contents of the RFS process trace file and the database alert log. If messages indicate that the RFS process frequently has to wait for a group because archiving has not completed, add more standby groups.

(确定Standby Redo Log 是否满足要求的判断依据)