Redo Log File

a.日志文件概要
1.Record all changes made to data
2.provide a recovery mechanism
3.can be organized into groups
4.at least two groups required
b.概念
1.同组的日志文件是完全一致的
2.LGWR
3.必须至少具有两组日志工作组合
4.log sequence numbers(lsn)
5.multiplexed redo log files 日志文件多路复用
6.log switch,日志的切换会触发检查点
7.checkpoint operation also occurs
the writint of dirty block from buffer cache to disk(DBWN)
c.写redo的场景
1.事务提交的时候
2.每3秒写一次
3.当redo log buffer已经三分之一满的时候
4.当redo log buffer中存在1M的改变记录的时候
5.当dbwn些数据到磁盘中之前
d.常用操作
1.切换日志文件
alter system switch log;
2.查看日志文件状态,日志文件历史信息
select group#,thread#,sequence#,status from v$log;
select * from v$log_history
select * from v$logfile(查看日志组和成员的关系,以及状态)
3.强制写checkpoint
fast_start_mttr_target=600,600秒写一次磁盘
alter system set fast_start_mttr_target=600 scope=spfile;
alter system checkpoint
4.增加日志工作组
alter database add logfile group 4(filename1,filename2) size xM;
alter database add logfile member 'filename' to group1
5.查看日志文件路径
select * from v$logfile
6.删除日志文件
alter database drop logfile member '文件名'
select member from v$logfile
不能对组中的最后一个成员进行删除，如果要删除，只能进行删除组的操作
alter database drop logfile group (select group# from v$log where status <> current)
7.重命名日志文件
1>shutdown immediate
2>copy logfile from old position to new position
3>startup mount
4>alter database rename file oldfile to newfile
5>alter database open;
8.清空日志文件
alter database clear logfile group group#(数据会丢失,不建议使用)
alter database clear unarchived logfile group group#(阻止日志文件归档,不建议使用)
9.omf 关于日志文件的参数
db_create_online_log_dest_1..n
e.achived redo log file
1.查看archived mode(ARCn)
select archive from v$instance;
select log_mode from v$database;