今天突然应用程序报错，无法连接数据库，以连接上的也无法做任何操作，但是没有一会，又恢复正常。宕机，查看alert.log看最后出现Thread 1 cannot allocate new log, sequence 101.

分析觉得应该是dbwr写的太慢,但是redo切换的太频繁,日志量比较大造成的.

查看redo

select *  from v$log;

GROUP#    THREAD#  SEQUENCE#      BYTES    MEMBERS ARCHIVED STATUS           FIRST_CHANGE# FIRST_TIMEITPUB个人空间6j&Yn5I
~'e+GGm
---------- ---------- ---------- ---------- ---------- -------- ---------------- ------------- -----------ITPUB个人空间c~W8C.@WSxa
         1          1        103  100288000          1 NO       INACTIVE              12308155 2009-4-3 16
c&VB]_(R~N0         2          1        104  100288000          1 NO       INACTIVE              12308158 2009-4-3 16
Ukp%U }do1U{q0         3          1        105  100288000          1 NO       CURRENT               12308162 2009-4-3 16

准备把redo增加到500m

1.增加两个重做日志组

alter database add logfile group 4 '/opt/oracle/oradata/orcl/redo04.log' size 500m;

alter database add logfile group 5 ' /opt/oracle/oradata/orcl/redo05.log' size 500m;

2.切换日志组

alter system switch logfile;

alter system switch logfile;

3.删除3个日志组

alter database drop logfile group1;

alter database drop logfile group2;

alter database drop logfile group3;

4.重新建日志组

alter database add logfile group 1 ' /opt/oracle/oradata/orcl/redo01.log' size 500m;

alter database add logfile group 2 ' /opt/oracle/oradata/orcl/redo02.log' size 500m;

alter database add logfile group 3 ' /opt/oracle/oradata/orcl/redo03.log' size 500m;

5.切换日志组

alter system switch logfile;

alter system switch logfile;

alter system switch logfile;

6删除日志组

alter database logfile group 4

alter database logfile group 5

性能还得在继续观察一段时间.关注