Oracle通过SCN做增量备份修复DG

　　DG由于网络原因或者bug原因经常不同步，有时隔得时间久了，就会丢失归档日志，或者长时间的归档恢复较慢，有一种可以基于scn的方式来恢复DG库，使用基于scn的增量备份来恢复standby库可以节省大量时间，或者进行dg异常的重新同步。

　　1，查询主库和备库的scn，redo号等信息，选取合适的scn号进行主库的备份

--查询DG端最低的scn，以下查询结果应该相同
 select to_char(current_scn) scn from v$database;
 select min(fhscn) from x$kcvfh;

 select distinct to_char(checkpoint_change#) from v$datafile_header order by 1; 

--以归档日志查询相关scn
 select THREAD#,min(sequence#) from v$archived_log where applied='NO' group by THREAD#;  
 select THREAD#,max(sequence#) from v$archived_log where applied='NO' group by THREAD#; 

 SELECT thread#,SEQUENCE#,to_char(FIRST_CHANGE#) fc,to_char(NEXT_CHANGE#) nc FROM v$archived_log WHERE SEQUENCE# = 32995 ORDER BY 1,2;

　　2，主库端基于scn的增量备份

RMAN> BACKUP INCREMENTAL FROM SCN <SCN from previous step> 
DATABASE FORMAT '/tmp/ForStandby_%U' tag 'FORSTANDBY'
--or
  backup device type disk incremental from scn 11125946510 database format'/home/oracle/db_incre%U.bbk'; 
--or
run {
allocate channel t1 type 'sbt_tape' parms 'ENV=(NB_ORA_CLIENT=bossb1)';
allocate channel t2 type 'sbt_tape' parms 'ENV=(NB_ORA_CLIENT=bossb1)';
allocate channel t3 type 'sbt_tape' parms 'ENV=(NB_ORA_CLIENT=bossb1)';
allocate channel t4 type 'sbt_tape' parms 'ENV=(NB_ORA_CLIENT=bossb1)';
allocate channel t5 type 'sbt_tape' parms 'ENV=(NB_ORA_CLIENT=bossb1)';
allocate channel t6 type 'sbt_tape' parms 'ENV=(NB_ORA_CLIENT=bossb1)';
allocate channel t7 type 'sbt_tape' parms 'ENV=(NB_ORA_CLIENT=bossb1)';
allocate channel t8 type 'sbt_tape' parms 'ENV=(NB_ORA_CLIENT=bossb1)';
backup incremental from scn 12844051683640 database format 'forstandby_%u' tag 'forstandby';
backup current controlfile for standby format 'forstandbyctrl.bck' tag 'forstandby_cnt'; 
  release channel t1;
  release channel t2;
  release channel t3;
  release channel t4;
  release channel t5;
  release channel t6;
  release channel t7;
  release channel t8;
}

 ALTER DATABASE CREATE standby controlfile AS '/home/oracle/standby.ctl';

　　3，传输文件到dg端，rman catalog start with

　　4，恢复控制文件，恢复数据库

RMAN>restore controlfile from'/data/oracle/backup/restore/standby.ctl';    
--mount database
RMAN> RECOVER DATABASE NOREDO;

　　5，启动同步进程，检查同步状态等。