月归档：十月 2014

ORA-27069: skgfdisp: 尝试在文件范围外执行 I/O

发表于 2014 年 10 月 29 日 由 惜分飞

接到网友技术支持请求,win 2003 ntfs格式文件系统,Oracle 8.1.7版本,主机重启后,数据库无法正常启动,offline datafile 15,数据库open成功,但是datafile 无法正常online,报错为:ORA-27069: skgfdisp: attempt to do I/O beyond the range of the file,请求协助处理

SQL> recover datafile 'D:\ORACLE\ORADATA\ORCL\ZSF_DATA.DBF';
ORA-00283: 恢复会话因错误而取消
ORA-01115: 从文件 15 读取块时出现 IO 错误 (块 # 1030071)
ORA-01110: 数据文件 15: 'D:\ORACLE\ORADATA\ORCL\ZSF_DATA.DBF'
ORA-27069: skgfdisp: 尝试在文件范围外执行 I/O
OSD-04026: 无效的参数经过. (OS 1030071)

使用bbed,成功online datafile 15

Tue Oct 28 16:30:35 2014
ALTER DATABASE RECOVER  datafile 15  
Tue Oct 28 16:30:35 2014
Media Recovery Datafile: 15
Media Recovery Start
Media Recovery Log 
Recovery of Online Redo Log: Thread 1 Group 1 Seq 245110 Reading mem 0
  Mem# 0 errs 0: D:\ORACLE\ORADATA\ORCL\REDO03.LOG
Media Recovery failed with error 1115
ORA-283 signalled during: ALTER DATABASE RECOVER  datafile 15  ...
Tue Oct 28 16:32:53 2014
Shutting down instance (abort)
License high water mark = 6
Instance terminated by USER, pid = 1548
Starting up ORACLE RDBMS Version: 8.1.7.0.0.
System parameters with non-default values:
  processes                = 600
  shared_pool_size         = 52428800
  large_pool_size          = 20971520
  java_pool_size           = 20971520
  control_files            = D:\oracle\oradata\ORCL\control01.ctl, D:\oracle\oradata\ORCL\control02.ctl
  db_block_buffers         = 19200
  db_block_size            = 8192
  compatible               = 8.1.0
  log_buffer               = 32768
  log_checkpoint_interval  = 10000
  log_checkpoint_timeout   = 1800
  db_files                 = 1024
  db_file_multiblock_read_count= 8
  max_enabled_roles        = 30
  remote_login_passwordfile= EXCLUSIVE
  global_names             = TRUE
  distributed_transactions = 500
  instance_name            = ORCL
  service_names            = ORCL
  mts_dispatchers          = (PROTOCOL=TCP)(PRE=oracle.aurora.server.SGiopServer)
  open_links               = 4
  sort_area_size           = 65536
  sort_area_retained_size  = 65536
  db_name                  = ORCL
  open_cursors             = 500
  ifile                    = D:\oracle\admin\ORCL\pfile\init.ora
  os_authent_prefix        = 
  job_queue_processes      = 4
  job_queue_interval       = 10
  parallel_max_servers     = 5
  background_dump_dest     = D:\oracle\admin\ORCL\bdump
  user_dump_dest           = D:\oracle\admin\ORCL\udump
  max_dump_file_size       = 10240
  oracle_trace_collection_name= 
PMON started with pid=2
DBW0 started with pid=3
LGWR started with pid=4
CKPT started with pid=5
SMON started with pid=6
RECO started with pid=7
SNP0 started with pid=8
SNP1 started with pid=9
SNP2 started with pid=10
SNP3 started with pid=11
Tue Oct 28 16:33:01 2014
starting up 1 shared server(s) ...
starting up 1 dispatcher(s) for network address '(ADDRESS=(PARTIAL=YES)(PROTOCOL=TCP))'...
Tue Oct 28 16:33:02 2014
ALTER DATABASE   MOUNT
Tue Oct 28 16:33:06 2014
Successful mount of redo thread 1, with mount id 1389958722.
Tue Oct 28 16:33:06 2014
Database mounted in Exclusive Mode.
Completed: ALTER DATABASE   MOUNT
Tue Oct 28 16:33:49 2014
ALTER DATABASE RECOVER  database until cancel  
Tue Oct 28 16:33:49 2014
Media Recovery Start
Media Recovery Log 
kcrrga: Warning.  Log sequence in archive filename wrapped
to fix length as indicated by %S in LOG_ARCHIVE_FORMAT.
Old log archive with same name might be overwritten.
ORA-279 signalled during: ALTER DATABASE RECOVER  database until cancel  ...
Tue Oct 28 16:34:03 2014
ALTER DATABASE RECOVER    LOGFILE 'D:\ORACLE\ORADATA\ORCL\REDO02.LOG'  
Tue Oct 28 16:34:03 2014
Media Recovery Log D:\ORACLE\ORADATA\ORCL\REDO02.LOG
Incomplete recovery applied all redo ever generated.
Recovery completed through change %s139866389
Media Recovery Complete
Completed: ALTER DATABASE RECOVER    LOGFILE 'D:\ORACLE\ORADA
Tue Oct 28 16:34:29 2014
alter database datafile 15 online

Tue Oct 28 16:34:29 2014
Completed: alter database datafile 15 online
Tue Oct 28 16:34:36 2014
alter database open resetlogs

RESETLOGS is being done without consistancy checks. This may result
in a corrupted database. The database should be recreated.
RESETLOGS after incomplete recovery UNTIL CHANGE 139866389
Tue Oct 28 16:34:38 2014
Thread 1 opened at log sequence 1
  Current log# 2 seq# 1 mem# 0: D:\ORACLE\ORADATA\ORCL\REDO02.LOG
Successful open of redo thread 1.
Tue Oct 28 16:34:38 2014
SMON: enabling cache recovery
Tue Oct 28 16:34:38 2014
Dictionary check beginning
Dictionary check complete
Tue Oct 28 16:34:39 2014
SMON: enabling tx recovery
Tue Oct 28 16:34:44 2014
Completed: alter database open resetlogs

数据库datafile 15 online成功后,客户操作业务继续发生ORA-600[ktsxs_add2]错误

Tue Oct 28 17:07:42 2014
Errors in file D:\oracle\admin\ORCL\udump\ORA02340.TRC:
ORA-00600: 内部错误代码，自变量: [ktsxs_add2], [14], [15], [42534], [5732], [5733], [], []

Tue Oct 28 17:07:53 2014
Errors in file D:\oracle\admin\ORCL\udump\ORA02340.TRC:
ORA-00600: 内部错误代码，自变量: [ktsxs_add2], [14], [15], [42534], [5732], [5733], [], []

Tue Oct 28 17:08:03 2014
Errors in file D:\oracle\admin\ORCL\udump\ORA02340.TRC:
ORA-00600: 内部错误代码，自变量: [ktsxs_add2], [14], [15], [42534], [5732], [5733], [], []

Tue Oct 28 17:08:16 2014
Errors in file D:\oracle\admin\ORCL\udump\ORA02340.TRC:
ORA-00600: 内部错误代码，自变量: [ktsxs_add2], [14], [15], [42534], [5732], [5733], [], []

Tue Oct 28 17:08:23 2014
Errors in file D:\oracle\admin\ORCL\udump\ORA02308.TRC:
ORA-00600: 内部错误代码，自变量: [ktsxs_add2], [14], [15], [42534], [5732], [5733], [], []

Tue Oct 28 17:08:31 2014
Errors in file D:\oracle\admin\ORCL\udump\ORA02340.TRC:
ORA-00600: 内部错误代码，自变量: [ktsxs_add2], [14], [15], [42534], [5732], [5733], [], []

Tue Oct 28 17:08:38 2014
Errors in file D:\oracle\admin\ORCL\udump\ORA02308.TRC:
ORA-00600: 内部错误代码，自变量: [ktsxs_add2], [14], [15], [42534], [5732], [5733], [], []

通过分析相关日志发现是insert插入表报错,很好理解,该库的datafile 15已经超过了系统的限制,现在继续插入数据,因此报错,查询可能异常对象

SQL> col segment_name for a20
SQL> SELECT distinct OWNER, SEGMENT_NAME, SEGMENT_TYPE, A.PARTITION_NAME
  2    FROM DBA_EXTENTS A
  3   WHERE FILE_ID = 15
  4     AND 1030071 <= BLOCK_ID;

OWNER                          SEGMENT_NAME         SEGMENT_TYPE
------------------------------ -------------------- ------------------
PARTITION_NAME
------------------------------
ZSF                            DETAIL               TABLE


ZSF                            DETAIL1              INDEX


ZSF                            DETAIL2              INDEX



OWNER                          SEGMENT_NAME         SEGMENT_TYPE
------------------------------ -------------------- ------------------
PARTITION_NAME
------------------------------
ZSF                            DETAIL3              INDEX


ZSF                            DETAIL4              INDEX


ZSF                            FK_RECI_ORD          INDEX



OWNER                          SEGMENT_NAME         SEGMENT_TYPE
------------------------------ -------------------- ------------------
PARTITION_NAME
------------------------------
ZSF                            PREPAY1              INDEX


ZSF                            RECEDETAIL1          INDEX

创建新表空间

Create tablespace zsf_new datafile  'D:\ORACLE\ORADATA\ORCL\ZSF_DATA_new01.dbf' size 4096m;
alter tablespace zsf_new add datafile 'D:\ORACLE\ORADATA\ORCL\ZSF_DATA_new02.dbf' 
size 128m autoextend on next 128M maxsize 4096m;

迁移异常对象到新表空间

alter table ZSF.DETAIL move tablespace ZSF_new;
alter index ZSF.DETAIL1 rebuild tablespace ZSF_new;
alter index ZSF.DETAIL2 rebuild tablespace ZSF_new;
alter index ZSF.DETAIL3 rebuild tablespace ZSF_new;
alter index ZSF.DETAIL4 rebuild tablespace ZSF_new;
alter index ZSF.FK_RECI_ORD rebuild tablespace ZSF_new;
alter index ZSF.PREPAY1 rebuild tablespace ZSF_new;
alter index ZSF.RECEDETAIL1 rebuild tablespace ZSF_new;

然后对于datafile 15所在表空间增加新文件,因为已经迁移了异常对象,然后resize datafile 15小于8G,关闭自扩展,至此该数据库恢复完成

发表在 非常规恢复 | 标签为 , , , , , , | 评论关闭

11g中 connect by 语句执行计划改变

发表于 2014 年 10 月 27 日 由 惜分飞

从10.2.0.3升级到11.2.0.4的朋友，如果细心会发现，以下sql在11.2.0.4中执行效率变低(该sql主要是获取连接用户获取权限信息)
select privilege#,level from sysauth$ connect by grantee#=prior privilege# and privilege#>0 start with grantee#=:1 and privilege#>0
如果你接触是Oracle版本比较多,而且还比较细心,你可能会进一步发现在11.2.0.2中该条sql是:select /*+ connect_by_filtering */ privilege#, level from sysauth$ connect by grantee#=prior privilege# and privilege#>0 start with grantee#=:1 and privilege#>0 也就是说使用了/*+ connect_by_filtering */提示.我这里通过简单测试说明问题.

在11.2.0.4环境中

14:16:19 SQL> set autot trace exp stat
14:16:20 SQL> set time on
14:16:20 SQL> set timing on
14:16:20 SQL> var a1 number;
14:16:20 SQL> exec :a1:=6;

PL/SQL 过程已成功完成。

已用时间:  00: 00: 00.00
14:16:20 SQL> select privilege#,level from sysauth$ connect by grantee#=prior 
14:16:20 SQL> privilege# and privilege#>0 start with grantee#=:a1  and privilege#>0
14:16:22 SQL> /

已用时间:  00: 00: 00.01

执行计划
----------------------------------------------------------
Plan hash value: 2624122540

------------------------------------------------------------------------------------------------------
| Id  | Operation                               | Name       | Rows  | Bytes | Cost (%CPU)| Time     |
------------------------------------------------------------------------------------------------------
|   0 | SELECT STATEMENT                        |            |     7 |   182 |     3  (34)| 00:00:01 |
|*  1 |  CONNECT BY NO FILTERING WITH START-WITH|            |       |       |            |          |
|   2 |   INDEX FAST FULL SCAN                  | I_SYSAUTH1 |   618 |  4944 |     2   (0)| 00:00:01 |
------------------------------------------------------------------------------------------------------

Predicate Information (identified by operation id):
---------------------------------------------------

   1 - access("GRANTEE#"=PRIOR "PRIVILEGE#")
       filter("PRIVILEGE#">0 AND "GRANTEE#"=TO_NUMBER(:A1) AND "PRIVILEGE#">0)


统计信息
----------------------------------------------------------
          1  recursive calls
          0  db block gets
          7  consistent gets
          0  physical reads
          0  redo size
        599  bytes sent via SQL*Net to client
        520  bytes received via SQL*Net from client
          2  SQL*Net roundtrips to/from client
          2  sorts (memory)
          0  sorts (disk)
          1  rows processed

这里可以看出来这里使用的执行计划使用了CONNECT BY NO FILTERING WITH START-WITH,逻辑读为7.

10.2.0.3环境中

14:32:57 SQL> set lines 150
14:33:00 SQL> set autot trace exp stat
14:33:01 SQL> set time on
14:33:01 SQL> set timing on
14:33:01 SQL> var a1 number;
14:33:01 SQL> exec :a1:=6;

PL/SQL procedure successfully completed.

Elapsed: 00:00:00.00
14:33:01 SQL> select privilege#,level from sysauth$ connect by grantee#=prior 
14:33:01 SQL> privilege# and privilege#>0 start with grantee#=:a1  and privilege#>0  ;

Elapsed: 00:00:00.00

Execution Plan
----------------------------------------------------------
Plan hash value: 2620769641

----------------------------------------------------------------------------------------
| Id  | Operation                 | Name       | Rows  | Bytes | Cost (%CPU)| Time     |
----------------------------------------------------------------------------------------
|   0 | SELECT STATEMENT          |            |     3 |    24 |     2   (0)| 00:00:01 |
|*  1 |  CONNECT BY WITH FILTERING|            |       |       |            |          |
|*  2 |   INDEX RANGE SCAN        | I_SYSAUTH1 |     3 |    24 |     2   (0)| 00:00:01 |
|   3 |   NESTED LOOPS            |            |       |       |            |          |
|   4 |    CONNECT BY PUMP        |            |       |       |            |          |
|*  5 |    INDEX RANGE SCAN       | I_SYSAUTH1 |     3 |    24 |     2   (0)| 00:00:01 |
----------------------------------------------------------------------------------------

Predicate Information (identified by operation id):
---------------------------------------------------

   1 - access("GRANTEE#"=PRIOR "PRIVILEGE#")
       filter("PRIVILEGE#">0)
   2 - access("GRANTEE#"=TO_NUMBER(:A1) AND "PRIVILEGE#">0)
   5 - access("GRANTEE#"=PRIOR "PRIVILEGE#" AND "PRIVILEGE#">0)


Statistics
----------------------------------------------------------
          0  recursive calls
          0  db block gets
          4  consistent gets
          0  physical reads
          0  redo size
        583  bytes sent via SQL*Net to client
        492  bytes received via SQL*Net from client
          2  SQL*Net roundtrips to/from client
          3  sorts (memory)
          0  sorts (disk)
          1  rows processed

这里执行计划使用的为CONNECT BY WITH FILTERING,而且逻辑读为4,对于这个sql来说,使用CONNECT BY WITH FILTERING执行效率更高.

这里可以很明显的看到:connect by查询的执行计划从10g的CONNECT BY WITH FILTERING变为了11g中的CONNECT BY NO FILTERING WITH SW (UNIQUE)，从而使得执行计划发生改变。但是Oracle一般有个特性,就是当引入新特性之时,一般都会伴随隐含参数或者event来屏蔽新特性.这里也例外,我们可以通过”_optimizer_connect_by_elim_dups” = false和”_connect_by_use_union_all” = “old_plan_mode”来屏蔽11g中关于connect by执行计划的改变,使得执行计划恢复到10G的CONNECT BY WITH FILTERING方式

14:30:45 SQL> alter session set "_optimizer_connect_by_elim_dups" = false;

会话已更改。

已用时间:  00: 00: 00.00
14:30:46 SQL> alter session set "_connect_by_use_union_all" = "old_plan_mode";

会话已更改。

已用时间:  00: 00: 00.00
14:30:46 SQL> set autot trace exp stat
14:30:46 SQL> set time on
14:30:46 SQL> set timing on
14:30:46 SQL> var a1 number;
14:30:46 SQL> exec :a1:=6;

PL/SQL 过程已成功完成。

已用时间:  00: 00: 00.00
14:30:46 SQL> select privilege#,level from sysauth$ connect by grantee#=prior 
14:30:46 SQL> privilege# and privilege#>0 start with grantee#=:a1  and privilege#>0 ;

已用时间:  00: 00: 00.01

执行计划
----------------------------------------------------------
Plan hash value: 2620769641

----------------------------------------------------------------------------------------
| Id  | Operation                 | Name       | Rows  | Bytes | Cost (%CPU)| Time     |
----------------------------------------------------------------------------------------
|   0 | SELECT STATEMENT          |            |     2 |    16 |     2   (0)| 00:00:01 |
|*  1 |  CONNECT BY WITH FILTERING|            |       |       |            |          |
|*  2 |   INDEX RANGE SCAN        | I_SYSAUTH1 |     2 |    16 |     2   (0)| 00:00:01 |
|   3 |   NESTED LOOPS            |            |       |       |            |          |
|   4 |    CONNECT BY PUMP        |            |       |       |            |          |
|*  5 |    INDEX RANGE SCAN       | I_SYSAUTH1 |     2 |    16 |     2   (0)| 00:00:01 |
----------------------------------------------------------------------------------------

Predicate Information (identified by operation id):
---------------------------------------------------

   1 - access("GRANTEE#"=PRIOR "PRIVILEGE#")
       filter("PRIVILEGE#">0)
   2 - access("GRANTEE#"=TO_NUMBER(:A1) AND "PRIVILEGE#">0)
   5 - access("GRANTEE#"=PRIOR "PRIVILEGE#" AND "PRIVILEGE#">0)


统计信息
----------------------------------------------------------
          1  recursive calls
          0  db block gets
          4  consistent gets
          0  physical reads
          0  redo size
        599  bytes sent via SQL*Net to client
        520  bytes received via SQL*Net from client
          2  SQL*Net roundtrips to/from client
          3  sorts (memory)
          0  sorts (disk)
          1  rows processed

11.2.0.2中也许是考虑到connect by 不够成熟,因此使用了hint /*+ connect_by_filtering */ 来固定执行计划

14:22:09 SQL> select /*+ connect_by_filtering */ privilege#,level from sysauth$ connect by grantee#=prior
14:22:09 SQL>  privilege# and privilege#>0 start with grantee#=:a1  and privilege#>0 
14:22:10 SQL> /

已用时间:  00: 00: 00.00

执行计划
----------------------------------------------------------
Plan hash value: 2620769641

----------------------------------------------------------------------------------------
| Id  | Operation                 | Name       | Rows  | Bytes | Cost (%CPU)| Time     |
----------------------------------------------------------------------------------------
|   0 | SELECT STATEMENT          |            |     7 |   182 |     8  (25)| 00:00:01 |
|*  1 |  CONNECT BY WITH FILTERING|            |       |       |            |          |
|*  2 |   INDEX RANGE SCAN        | I_SYSAUTH1 |     2 |    16 |     2   (0)| 00:00:01 |
|   3 |   NESTED LOOPS            |            |     5 |   105 |     4   (0)| 00:00:01 |
|   4 |    CONNECT BY PUMP        |            |       |       |            |          |
|*  5 |    INDEX RANGE SCAN       | I_SYSAUTH1 |     2 |    16 |     1   (0)| 00:00:01 |
----------------------------------------------------------------------------------------

Predicate Information (identified by operation id):
---------------------------------------------------

   1 - access("GRANTEE#"=PRIOR "PRIVILEGE#")
       filter("PRIVILEGE#">0)
   2 - access("GRANTEE#"=TO_NUMBER(:A1) AND "PRIVILEGE#">0)
   5 - access("GRANTEE#"="connect$_by$_pump$_002"."prior privilege# " AND
              "PRIVILEGE#">0)


统计信息
----------------------------------------------------------
          1  recursive calls
          0  db block gets
          4  consistent gets
          0  physical reads
          0  redo size
        599  bytes sent via SQL*Net to client
        520  bytes received via SQL*Net from client
          2  SQL*Net roundtrips to/from client
          3  sorts (memory)
          0  sorts (disk)
          1  rows processed

CONNECT BY NO FILTERING WITH SW (UNIQUE)和CONNECT BY WITH FILTERING,没有明显的优劣之分,只有在特定的情况下,进行了实际的测试,选择合适自己的sql的执行计划

发表在 Oracle性能优化 | 标签为 | 评论关闭

数据库启动报ORA-00704 ORA-39714错误解决

发表于 2014 年 10 月 25 日 由 惜分飞

数据库启动失败,报ORA-00704、ORA-39714错误

[oracle@www.xifenfei.com ~]$ sqlplus / as sysdba

SQL*Plus: Release 12.1.0.1.0 Production on Thu Aug 7 08:15:35 2014

Copyright (c) 1982, 2013, Oracle.  All rights reserved.


Connected to:
Oracle Database 12c Enterprise Edition Release 12.1.0.1.0 - 64bit Production
With the Partitioning, OLAP, Advanced Analytics and Real Application Testing options

SQL> startup
ORACLE instance started.

Total System Global Area  663945216 bytes
Fixed Size                  2291808 bytes
Variable Size             369100704 bytes
Database Buffers          289406976 bytes
Redo Buffers                3145728 bytes
Database mounted.
ORA-01092: ORACLE instance terminated. Disconnection forced
ORA-00704: bootstrap process failure
ORA-39714: upgrade script utlmmig.sql failed
Process ID: 11592
Session ID: 1 Serial number: 5


SQL> startup upgrade
SP2-0642: SQL*Plus internal error state 2133, context 3114:0:0
Unsafe to proceed
ORA-03114: not connected to ORACLE


SQL> exit
Disconnected from Oracle Database 12c Enterprise Edition Release 12.1.0.1.0 - 64bit Production
With the Partitioning, OLAP, Advanced Analytics and Real Application Testing options

alert日志报错

Thu Aug 07 07:42:25 2014
SMON: enabling cache recovery
Thu Aug 07 07:42:25 2014
Errors in file /u01/app/oracle/diag/rdbms/orcl/ORCL/trace/ORCL_ora_11592.trc:
ORA-39714: upgrade script utlmmig.sql failed
Thu Aug 07 07:42:25 2014
Errors in file /u01/app/oracle/diag/rdbms/orcl/ORCL/trace/ORCL_ora_11592.trc:
ORA-00704: bootstrap process failure
ORA-39714: upgrade script utlmmig.sql failed
Thu Aug 07 07:42:25 2014
Errors in file /u01/app/oracle/diag/rdbms/orcl/ORCL/trace/ORCL_ora_11592.trc:
ORA-00704: bootstrap process failure
ORA-39714: upgrade script utlmmig.sql failed
Thu Aug 07 07:42:25 2014
Error 704 happened during db open, shutting down database
USER (ospid: 11592): terminating the instance due to error 704

通过分析utlmmig.sql脚本知道,数据库在升级bootstrap$之前会先在props$表中插入BOOTSTRAP_UPGRADE_ERROR相关记录,数据库在启动之时会检测该值,如果发现该值存在,数据库只能以upgrade模式启动,清理掉相关记录,数据库即可正常启动

[oracle@www.xifenfei.com ~]$ sqlplus / as sysdba

SQL*Plus: Release 12.1.0.1.0 Production on Thu Aug 7 07:42:44 2014

Copyright (c) 1982, 2013, Oracle.  All rights reserved.

Connected to an idle instance.

SQL> startup upgrade
ORACLE instance started.

Total System Global Area  663945216 bytes
Fixed Size                  2291808 bytes
Variable Size             369100704 bytes
Database Buffers          289406976 bytes
Redo Buffers                3145728 bytes
Database mounted.
Database opened.
SQL>   delete from props$ where name = 'BOOTSTRAP_UPGRADE_ERROR';

1 row deleted.

SQL>   delete from props$ where name = 'LOGMNR_BOOTSTRAP_UPGRADE_ERROR';

0 rows deleted.

SQL>   commit;

Commit complete.

SQL> 
SQL> shutdown immediate
Database closed.
Database dismounted.
ORACLE instance shut down.
SQL> startup
ORACLE instance started.

Total System Global Area  663945216 bytes
Fixed Size                  2291808 bytes
Variable Size             369100704 bytes
Database Buffers          289406976 bytes
Redo Buffers                3145728 bytes
Database mounted.
Database opened.
SQL>

数据库虽然正常启动成功,但是由于bootstrap$对象升级失败,后续还是有很大风险,建议分析报错原因,解决原因然后继续升级bootstrap$基表

发表在 ORA-xxxxx | 标签为 , , , | 评论关闭