恢复测试数据库时遇到了日志组不让删的情况。
因为这个测试数据库是从RAC克隆而来，每个节点都有独立的REDO LOG FILE，变成单实例后，
只有节点1的REDO LOG FILE再用，那就把节点2的删掉吧，再删节点2的REDO LOG FILE时，遇到了ORA-01567错误。
SQL> select b.thread#,a.group#,a.member,bytes/1024/1024,b.members from v$logfile a,v$log b where a.group#=b.group#
THREAD# GROUP# MEMBER BYTES/1024/1024 MEMBERS
---------- ----------
-------------------------------------------------------------------------------- --------------- ----------
1 1
/u01/app/oracle/fast_recovery_area/ORCL/onlinelog/o1_mf_1_kk0nkxxb_.log 500 1
1 3
/u01/app/oracle/fast_recovery_area/ORCL/onlinelog/o1_mf_3_kk0nkyjm_.log 500 1
2 13
/u01/app/oracle/fast_recovery_area/ORCL/onlinelog/o1_mf_13_kk0nkzw0_.log 500 1
2 16
/u01/app/oracle/fast_recovery_area/ORCL/onlinelog/o1_mf_16_kk0nl2xt_.log 500 1
SQL> select group#,thread#,sequence#,members,status from v$log;
GROUP# THREAD# SEQUENCE# MEMBERS STATUS
---------- ---------- ---------- ---------- ----------------
1 1 7 1 INACTIVE
3 1 8 1 CURRENT
13 2 1 1 INACTIVE
16 2 0 1 UNUSED
尝试直接删除THREAD=2对应的日志文件组13,16
SQL> alter database drop logfile group 13;
alter database drop logfile group 13
*
ERROR at line 1:
ORA-01567: dropping log 1 would leave less than 2 log files for instance orcl (thread 2)
ORA-00312: online log 13 thread 2: '
/u01/app/oracle/fast_recovery_area/ORCL/onlinelog/o1_mf_13_kk0nkzw0_.log'
在节点2的REDO LOG FILE就剩2个的时候，就不让删了，这是REDO的机制，每个节点不能少于2个日志组，如果少于2个日志无法切换了。
解决方法很简单，把节点2给踢出去就行了。
SQL> alter database disable THREAD 2;
Database altered.
SQL> alter database drop logfile group 13;
Database altered.
SQL> alter database drop logfile group 16;
Database altered.
然后还得去操作系统上把这些删除掉的REDO LOG FILE删掉。
?