问题描述：

rac扩展表空间时，漏了+号，会默认创建在$ORACLE_HOME/dbs下面
例如alter tablespace USER add datafile ‘data’ size 30g;
这会导致当远端节点，需要写入数据时，会报无法找到数据文件的错误

解决办法：

第一种情况：

数据文件无数据，直接删除（不推荐）
确认数据文件里是否有数据

  1. select t.file_name,t1.owner,t1.segment_name,t1.segment_type,t1.tablespace_name from dba_data_files t,dba_extents t1 where t.file_id=t1.file_id and file_name='/u01/app/oracle/product/11.2.0/db_1/dbs/data'；

然后删除

  1. alter tablespace test drop datafile '/u01/app/oracle/product/11.2.0/db_1/dbs/data';

最后物理删除rm

第二种情况：

迁移数据文件到asm

  1. SQL> select file_id,file_name from dba_data_files where file_name not like '+DATA%';
  3.    FILE_ID        FILE_NAME
  4.   ----------     --------------------------------------------------------------------------------
  5.          6       +FRA/xiamen/datafile/test1215.260.999093473
  6.          8        /u01/app/oracle/product/11.2.0/db_1/dbs/data
  8. SQL> alter database datafile 8 offline;
  10. Database altered.
  12. RMAN> copy datafile 8 to '+data';
  14. Starting backup at 01-FEB-19
  15. using channel ORA_DISK_1
  16. channel ORA_DISK_1: starting datafile copy
  17. input datafile file number=00008 name=/u01/app/oracle/product/11.2.0/db_1/dbs/data
  18. output file name=+DATA/xiamen/datafile/test1215.289.999096869 tag=TAG20190201T145428 RECID=6 STAMP=999096869
  19. channel ORA_DISK_1: datafile copy complete, elapsed time: 00:00:03
  20. Finished backup at 01-FEB-19
  22. RMAN> switch datafile 8 to copy;
  24. datafile 8 switched to datafile copy "+DATA/xiamen/datafile/test1215.289.999096869"
  26. RMAN> recover datafile 8;
  28. Starting recover at 01-FEB-19
  29. using channel ORA_DISK_1
  31. starting media recovery
  32. media recovery complete, elapsed time: 00:00:01
  34. Finished recover at 01-FEB-19
  36. RMAN> sql 'alter database datafile 8 online';
  38. sql statement: alter database datafile 8 online

最后删除原有的dbs下面的数据文件