物理备库出现gap，主库为5节点RAC，数据库版本10.2.0.5

登录物理备库

SQL> select * from v$archive_gap;

   THREAD# LOW_SEQUENCE# HIGH_SEQUENCE#

---------- ------------- --------------         1        122309         122325         3          3517           3525发现gap比较多，以前出现一个archive gap的时候，我们通常是在RAC主库使用rman命令copy，即将ASM格式的archivelog改为普通格式的，然后传到备库register并recover这次gap太多，用这个方法太慢了，更换一个在主库一个节点上rman执行backup archivelog from sequence 122309 until sequence 122325 thread 1;backup  archivelog from sequence 3517 until sequence 3525 thread 3;然后将备份集传到备库先使用catalog start with将这两个备份集注册到控制文件，然后使用restore恢复出archivelogRMAN> restore archivelog from sequence 122309 until sequence 122325 thread 1;

Starting restore at 13-OCT-11

using channel ORA_DISK_1

channel ORA_DISK_1: starting archive log restore to default destination

channel ORA_DISK_1: restoring archive logarchive log thread=1 sequence=122309channel ORA_DISK_1: restoring archive logarchive log thread=1 sequence=122310channel ORA_DISK_1: restoring archive logarchive log thread=1 sequence=122311channel ORA_DISK_1: restoring archive logarchive log thread=1 sequence=122312channel ORA_DISK_1: restoring archive logarchive log thread=1 sequence=122313channel ORA_DISK_1: restoring archive logarchive log thread=1 sequence=122314channel ORA_DISK_1: restoring archive logarchive log thread=1 sequence=122315channel ORA_DISK_1: restoring archive logarchive log thread=1 sequence=122316channel ORA_DISK_1: restoring archive logarchive log thread=1 sequence=122317channel ORA_DISK_1: restoring archive logarchive log thread=1 sequence=122318channel ORA_DISK_1: restoring archive logarchive log thread=1 sequence=122319channel ORA_DISK_1: restoring archive logarchive log thread=1 sequence=122320channel ORA_DISK_1: restoring archive logarchive log thread=1 sequence=122321channel ORA_DISK_1: restoring archive logarchive log thread=1 sequence=122322channel ORA_DISK_1: restoring archive logarchive log thread=1 sequence=122323channel ORA_DISK_1: restoring archive logarchive log thread=1 sequence=122324channel ORA_DISK_1: restoring archive logarchive log thread=1 sequence=122325channel ORA_DISK_1: reading from backup piece /data/backup/j9movv4f_1_1channel ORA_DISK_1: restored backup piece 1piece handle=/data/backup/j9movv4f_1_1 tag=TAG20111013T082342channel ORA_DISK_1: restore complete, elapsed time: 00:00:56Finished restore at 13-OCT-11

此时alertlog相应记录为

Thu Oct 13 10:34:38 CST 2011Archivelog restore complete. Elapsed time: 0:00:19 Archivelog restore complete. Elapsed time: 0:00:21 Archivelog restore complete. Elapsed time: 0:00:21 Archivelog restore complete. Elapsed time: 0:00:21 Archivelog restore complete. Elapsed time: 0:00:21 Archivelog restore complete. Elapsed time: 0:00:22 Archivelog restore complete. Elapsed time: 0:00:22 Archivelog restore complete. Elapsed time: 0:00:22 Thu Oct 13 10:35:00 CST 2011Archivelog restore complete. Elapsed time: 0:00:20 Archivelog restore complete. Elapsed time: 0:00:22 Archivelog restore complete. Elapsed time: 0:00:22 Archivelog restore complete. Elapsed time: 0:00:22 Archivelog restore complete. Elapsed time: 0:00:22 Archivelog restore complete. Elapsed time: 0:00:21 Archivelog restore complete. Elapsed time: 0:00:21 Archivelog restore complete. Elapsed time: 0:00:21 Archivelog restore complete. Elapsed time: 0:00:02 此时v$archive_gap已经没有了记录SQL> select * from v$archive_gap;

no rows selected

但是recover进程不会自动恢复这些log，需要先cancel再开启SQL> recover managed standby database cancel;Media recovery complete.SQL> recover managed standby database using current logfile disconnect from session;Media recovery complete.此时备库恢复正常

来自 “ ITPUB博客 ” ，链接：http://blog.itpub.net/15480802/viewspace-709084/，如需转载，请注明出处，否则将追究法律责任。