ORA-00257 archiver error 的处理方法

有时impdp操作时会hang住，也不报任何错误，可能的原因是archive log过多了
有时会直接抛出ORA-00527的错误, 解决方法如下

1. 用sys用户登录
  sqlplus sys/pwd@tns as sysdba
2. 看看archiv log所在位置
SQL> show parameter log_archive_dest;
NAME                                     TYPE        VALUE
------------------------------------ ----------- ------------------------------
log_archive_dest                   string
log_archive_dest_1               string
log_archive_dest_10             string
3. 一般VALUE为空时，可以用archive log list;检查一下归档目录和log sequence
SQL> archive log list;
Database log mode                 Archive Mode
Automatic archival                   Enabled
Archive destination                  USE_DB_RECOVERY_FILE_DEST
Oldest online log sequence     360
Next log sequence to archive  360
Current log sequence              362
4. 检查flash recovery area的使用情况,可以看见archivelog已经很大了，达到96.62
SQL> select * from V$FLASH_RECOVERY_AREA_USAGE;
FILE_TYPE    PERCENT_SPACE_USED PERCENT_SPACE_RECLAIMABLE NUMBER_OF_FILES
------------ ------------------ ------------------------- ---------------
CONTROLFILE                 .13                        0               1
ONLINELOG                  2.93                        0               3
ARCHIVELOG                96.62                       0              141
BACKUPPIECE                   0                         0               0
IMAGECOPY                      0                         0               0
FLASHBACKLOG                0                         0               0
5. 计算flash recovery area已经占用的空间
SQL> select sum(percent_space_used)*3/100 from v$flash_recovery_area_usage;
SUM(PERCENT_SPACE_USED)*3/100
-----------------------------
                       2.9904
                     
6. 找到recovery目录, show parameter recover
SQL> show parameter recover;
NAME                                 TYPE        VALUE
------------------------------------ ----------- ------------------------------
db_recovery_file_dest                string           /u01/app/oracle/flash_recovery_area
db_recovery_file_dest_size        big integer   5G
recovery_parallelism                   integer         0
                     
7 上述结果告诉我们，归档位置用的是默认值，放在flash_recovery_area下(db_recovery_file_dest目录=/u01/app/oracle/flash_recovery_area)
[root@sha3 10.2.0]# echo $ORACLE_BASE
/u01/app/oracle
[root@sha3 10.2.0]# cd $ORACLE_BASE/flash_recovery_area/tt/archivelog
转移或清除对应的归档日志, 删除一些不用的日期目录的文件，注意保留最后几个文件（比如360以后的）
---------------------------------------------------------------------------------------
注意:
在删除归档日志后，必须用RMAN维护控制文件，否则空间显示仍然不释放。
---------------------------------------------------------------------------------------
8. rman target sys/pwd
[root@sha3 oracle]# rman target sys/pass
Recovery Manager: Release 10.2.0.4.0 - Production on Tue Jan 20 01:41:26 2009
Copyright (c) 1982, 2007, .  All rights reserved.
connected to target database: tt (DBID=4147983671)
 
9. 检查一些无用的archivelog
RMAN> crosscheck archivelog all;
10. 删除过期的归档
RMAN> delete expired archivelog all;
delete archivelog until time 'sysdate-1' ; 删除截止到前一天的所有archivelog
11. 再次查询，发现使用率正常，已经降到23.03
SQL> select * from V$FLASH_RECOVERY_AREA_USAGE;
FILE_TYPE    PERCENT_SPACE_USED PERCENT_SPACE_RECLAIMABLE NUMBER_OF_FILES
------------ ------------------ ------------------------- ---------------
CONTROLFILE                 .13                         0               1
ONLINELOG                  2.93                         0               3
ARCHIVELOG                23.03                         0              36
BACKUPPIECE                   0                         0               0
IMAGECOPY                     0                         0               0
FLASHBACKLOG                  0                         0               0
其它有用的Command:
----------------------------------
如果archive log模式下不能正常startup,则先恢复成noarchive log,startup成功后，再shutdown;
shutdown immediate;
startup mount;
alter database noarchivelog;
alter database open;
shutdown immediate;
再次startup以archive log模式
shutdown immediate;
startup mount;
show parameter log_archive_dest;
alter database archivelog;
archive log list;               
alter database open;
如果还不行，则删除一些archlog log
SQL> select group#,sequence# from v$log;
    GROUP# SEQUENCE#
---------- ----------
         1         62
         3         64
         2         63
原来是日志组一的一个日志不能归档
SQL> alter database clear unarchived logfile group 1;
alter database open;
最后，也可以指定位置Arch Log, 请按照如下配置
select name from v$datafile;
alter system set log_archive_dest='/opt/app/oracle/oradata/usagedb/arch' scope=spfile
或者修改大小
SQL> alter system set db_recovery_file_dest_size=3G scope=both;

转自