本文是原创文章。转载请注明出处：http://blog.csdn.net/msdnchina/article/details/44336789
近期遇到一个案例，是将一个表空间offline之后，alert日志中不断的报错，最初该报错是在11.2.0.2被发现的。后来我在11.2.0.4的linux x86-64 bit下试验了一下，也发现了同样的现象。
例如以下是我的实验步骤：
[oracle@rhel63single ~]$ sqlplus / as sysdba
SQL*Plus: Release 11.2.0.4.0 Production on Fri Mar 13 09:11:24 2015
Copyright (c) 1982, 2013, Oracle. All rights reserved.
Connected to:
Oracle Database 11g Enterprise Edition Release 11.2.0.4.0 - 64bit Production
With the Partitioning, OLAP, Data Mining and Real Application Testing options
SQL> select name from v$dbfile;
NAME
--------------------------------------------------------------------------------
/u01/app/oracle/oradata/test/users01.dbf
/u01/app/oracle/oradata/test/undotbs01.dbf
/u01/app/oracle/oradata/test/sysaux01.dbf
/u01/app/oracle/oradata/test/system01.dbf
/u01/app/oracle/oradata/test/ten01.dbf
/u01/app/oracle/oradata/test/tb_test_01.dbf
/u01/app/oracle/oradata/test/ts1.dbf
/u01/app/oracle/oradata/test/ts2.dbf
/u01/app/oracle/oradata/test/test01.dbf
/u01/app/oracle/oradata/test/test_uni_sz_2m_01.dbf
/u01/app/oracle/oradata/test/test_uni_sz_1m_01.dbf
11 rows selected.
SQL> create tablespace testtbs datafile '/u01/app/oracle/oradata/test/test.dbf' size 10M;
Tablespace created.
SQL> create table testtab tablespace testtbs as select * from dual;
Table created.
SQL> select * from testtab;
D
-
X
SQL> alter database datafile '/u01/app/oracle/oradata/test/test.dbf' offline;
Database altered.
SQL> select * from testtab;
select * from testtab
*
ERROR at line 1:
ORA-00376: file 12 cannot be read at this time
ORA-01110: data file 12: '/u01/app/oracle/oradata/test/test.dbf'
SQL> ! date
Fri Mar 13 09:17:12 CST 2015
SQL>
从alert日志中看到：
Fri Mar 13 09:15:38 2015 create tablespace test datafile '/u01/app/oracle/oradata/test/test.dbf' size 10M ORA-1543 signalled during: create tablespace test datafile '/u01/app/oracle/oradata/test/test.dbf' size 10M... Fri Mar 13 09:15:57 2015 create tablespace testtbs datafile '/u01/app/oracle/oradata/test/test.dbf' size 10M Completed: create tablespace testtbs datafile '/u01/app/oracle/oradata/test/test.dbf' size 10M Fri Mar 13 09:16:46 2015 alter database datafile '/u01/app/oracle/oradata/test/test.dbf' offline Completed: alter database datafile '/u01/app/oracle/oradata/test/test.dbf' offline Fri Mar 13 09:17:01 2015 Checker run found 1 new persistent data failures Fri Mar 13 09:20:28 2015 Errors in file /u01/app/oracle/diag/rdbms/test/test/trace/test_m000_3186.trc: ORA-01135: file 12 accessed for DML/query is offline ORA-01110: data file 12: '/u01/app/oracle/oradata/test/test.dbf' Fri Mar 13 09:30:29 2015 Errors in file /u01/app/oracle/diag/rdbms/test/test/trace/test_m000_3217.trc: ORA-01135: file 12 accessed for DML/query is offline ORA-01110: data file 12: '/u01/app/oracle/oradata/test/test.dbf' Fri Mar 13 09:40:30 2015 Errors in file /u01/app/oracle/diag/rdbms/test/test/trace/test_m000_3250.trc: ORA-01135: file 12 accessed for DML/query is offline ORA-01110: data file 12: '/u01/app/oracle/oradata/test/test.dbf'
查看/u01/app/oracle/diag/rdbms/test/test/trace/test_m000_3250.trc文件的内容例如以下：
[oracle@rhel63single trace]$ cat /u01/app/oracle/diag/rdbms/test/test/trace/test_m000_3250.trc Trace file /u01/app/oracle/diag/rdbms/test/test/trace/test_m000_3250.trc Oracle Database 11g Enterprise Edition Release 11.2.0.4.0 - 64bit Production With the Partitioning, OLAP, Data Mining and Real Application Testing options ORACLE_HOME = /u02/app/oracle/product/11.2.0.4/db_1 System name: Linux Node name: rhel63single Release: 2.6.32-279.el6.x86_64 Version: #1 SMP Wed Jun 13 18:24:36 EDT 2012 Machine: x86_64 VM name: VMWare Version: 6 Instance name: test Redo thread mounted by this instance: 1 Oracle process number: 22 Unix process pid: 3250, image: oracle@rhel63single (M000) *** 2015-03-13 09:40:30.195 *** SESSION ID:(130.7) 2015-03-13 09:40:30.195 *** CLIENT ID:() 2015-03-13 09:40:30.195 *** SERVICE NAME:(SYS$BACKGROUND) 2015-03-13 09:40:30.195 *** MODULE NAME:(MMON_SLAVE) 2015-03-13 09:40:30.195 *** ACTION NAME:(Monitor Tablespace Thresholds) 2015-03-13 09:40:30.195 DDE rules only execution for: ORA 1110 ----- START Event Driven Actions Dump ---- ---- END Event Driven Actions Dump ---- ----- START DDE Actions Dump ----- Executing SYNC actions ----- START DDE Action: 'DB_STRUCTURE_INTEGRITY_CHECK' (Async) ----- Successfully dispatched ----- END DDE Action: 'DB_STRUCTURE_INTEGRITY_CHECK' (SUCCESS, 0 csec) ----- Executing ASYNC actions ----- END DDE Actions Dump (total 0 csec) ----- ORA-01135: file 12 accessed for DML/query is offline ORA-01110: data file 12: '/u01/app/oracle/oradata/test/test.dbf' [oracle@rhel63single trace]$
后来在mos中搜索了一下：
Getting Trace Files With DB_STRUCTURE_INTEGRITY_CHECK And An ORA-01110. (文档 ID 1252614.1)
该文章与本现象相符。
该文章给出原因是：
These files are generated by the Health Check Monitor and specifically, the DB Structure Integrity Check.
http://docs.oracle.com/cd/B28359_01/server.111/b28310/diag007.htm#ADMIN11269
这个hcm（health check monitor）是从11.1就有的。
例如以下来自：Checker Run Found xx New Persistent Data Failures. (文档 ID 1391202.1)
By default the database runs the Health Check on a regular basis.During the course of these health check executions if any failures are detected then these kind of messages are logged to the alert log. Health Monitor checks (also known as checkers, health checks, or checks) examine various layers and components of the database. Health checks detect file corruptions, physical and logical block corruptions, undo and redo corruptions, data dictionary corruptions, and more. The health checks generate reports of their findings and, in many cases, recommendations for resolving problems.
拓展一下：与health check相关的视图（事实上是synonym）有：
V$HM_CHECK V$HM_CHECK_PARAM V$HM_FINDING V$HM_INFO V$HM_RECOMMENDATION V$HM_RUN
以上视图的查询结果能够用到客户db的巡检中。
解决方式见下：
Solution
The ORA-01110 is not seen in the alert log, but if you set an Error Stack for it, you would see it. So, we need to set an Error Stack for the ORA-01110 to get the name of the offline data file.
1) Set the following event and check the alert log for occurrence of the ORA-01110.
SQL> connect / as sysdba
SQL> alter system set events '1110 trace name errorstack level 3';
2) If the trace files are still being generated, then once a file is created, the ORA-01110 should now be seen
in the alert log. When you capture it, disable the tracing:
SQL> connect / as sysdba
SQL> alter system set events '1110 trace name errorstack off';
3. The ORA-01110 in the alert log seen, should give the name of the data file that's offline.
4. Please switch that data file/ table space status to online, and that should avoid the traces.
5. If the table space/ data file is set to offline for a reason, then the traces are expected and should be of no
concern.
Note that you can also use below query to find offline data files. The point from above steps is to get the data file triggering the error:
SQL> connect / as sysdba
SQL> column file_name format a40;
SQL> column tablespace_name format a12;
SQL> select file_name,file_id,tablespace_name,online_status from dba_data_files order by
tablespace_name, file_id;
Even when the temp file is offline, ORA-01110 is seen in alert log or trace file. Run the sql to check temp files and online or drop it.
SQL>select file_name, tablespace_name, status from dba_temp_files order by file_name;