ITPub博客

由于存储光纤交换机重新做了配置，并保存了配置信息，导致所有连接此交换机的系统全部报错：Linux：Jun 13 17:51:01 bhcx kernel: SCSI error : <1 0 1 1> return code = 0x10000Jun 13 17:51:01 bhcx kernel: end_request: I/O error, dev sdg, sector 98990423Jun 13 17:51:01 bhcx kernel: SCSI error : <1 0 1 1> return code = 0x10000Jun 13 17:51:01 bhcx kernel: end_request: I/O error, dev sdg, sector 98990431Jun 13 17:51:39 bhcx kernel: SCSI error : <1 0 1 1> return code = 0x10000Jun 13 17:51:39

今天在做数据库例检时，alert.log文件中出现如下错误：Thu Jun 9 09:00:27 2011Errors in file /u01/app/oracle/admin/xxxx/bdump/xxxx2_m001_577740.trc:ORA-00600: internal error code, arguments: [kcbz_check_objd_typ_3], [0], [0], [1], [], [], [], []查了一下，有如下BUG：NBBugFixedDescription725104911.2.0.2, 12.1.0.0Corruption in bitmap index introduced when using transportable tablespacesV10020002 V10020003 V10020004 V10020005 V11010006 V11010007 V11020001 BITMAP CORR/IND ERROR OERI TRANSPORT WRONGRES +

没装PHP导致rpm -Uvh php-5.1.6-23.2.el5_3.x86_64.rpm php-cli-5.1.6-23.2.el5_3.x86_64.rpm php-common-5.1.6-23.2.el5_3.x86_64.rpm gmp-4.1.4-10.el5.x86_64.rpmapachectl stopapachectl start 问题解决[@more@]

2011-05-31 10:06:11.360: [ CRSEVT][12232]32CAAMonitorHandler :: 0:Action Script for resource 'ora.xxxx.xxxx2.inst' stdout redirection failed for `/u01/app/oracle/crs/crs/log/startAXUlaa.stdout` : No such file or directory查了一下，只有一个类似的BUGBug 4598992 "Action script for resource 'ora.xxx.vip' stdout redirection" errors in crsd.logFixed:This issue is fixed in11.2 (Future Release) 10.2.0.4 (Server Patch Set) 而这台服务器的版本是10.0.2.1，升级真的很重要[@more@]

crsd.log中出现2011-05-31 09:51:31.372: [ OCRSRV][3605]th_select_handler: Failed to retrieve procctx from ht. constr = [357257968] retval lht [-27] Signal CV.2011-05-31 09:51:37.213: [ OCRSRV][3605]th_select_handler: Failed to retrieve procctx from ht. constr = [387231440] retval lht [-27] Signal CV.2011-05-31 09:52:12.070: [ OCRSRV][3605]th_select_handler: Failed to retrieve procctx from ht. constr = [357257968] retval lht [-27] Signal CV.2011-05-31 09:52:12.986: [ OCRSRV][3605]th_select_han

有一个维护的生产环境需要每日自动将rman备份结果集及archive log转移至别的磁盘，如果使用cp，需要先将目的删除，然后再从源拷贝至目的。浪费很多IOrsync可以增量的进行备份，并且，如果源删除了文件，也可以在同步的过程中自动在目的端进行删除http://hpux.connect.org.uk/hppd/hpux/Networking/Admin/rsync-3.0.8/需要注意的是，rsync需要gettext、libiconv、popt的支持，在上面链接的Run-time dependencies有对应的链接[@more@]

第一次自己发现的并且被Oracle确认的BUG，Bug 12412766庆祝一下。呵呵[@more@]

遭遇Bug 4511371建立了一个DBMonitor的库，11.2.0.2的，想通过这个库访问被管理的库 10.2.0.1，触发了这个BUGSQL> select t.dbid, t.instance_number, t.snap_id2 from sys.wrm$_snapshot@dbl_c001 t3 where to_char(t.begin_interval_time, 'DD HH24') = '28 07';DBID INSTANCE_NUMBER SNAP_ID---------- --------------- ----------2795737348 1 31628SQL> select *2 from table(dbms_workload_repository.awr_report_html@dbl_c001(2795737348,3 1,4 31628 - 1,5 31628));select **第 1 行出现错误:ORA-04052: 在查找远程对象 SY

SQL> create table t as select * from dba_objects;表已创建。SQL> create index t_idx on t(object_id);索引已创建。SQL> exec dbms_stats.gather_table_stats(user,'T',null,null);PL/SQL 过程已成功完成。SQL> var b1 number;SQL> var b2 number;SQL> exec :b1:=null;PL/SQL 过程已成功完成。SQL> exec :b2:=null;PL/SQL 过程已成功完成。先看看在10g下的表现，10.2.0.4以下SQL_ID 66gnjuyzgnhv3, child number 0-------------------------------------select * from t where t.object_id >= nvl(:b1,0) and

今天在检查刚接手的DB的AWR报告时发现一个SQL Gets很大，于是进行tunning，看了一眼user_tables，发现这个表的last_analyzed居然是2007年的，很奇怪，这个表应该是在不断的进行修改，并且，数据库是Oracle 10g，自动收集统计信息的任务也没有关闭，为什么没有更新统计信息呢？查看dba_tab_modification表，也没有发现任何有关于这个表的记录，奇了不管他，直接收集统计信息，这下发现问题了[@more@]SQL> exec dbms_stats.gather_table_stats(user,'XXX',method_opt => 'for all indexed columns',cascade => true,degree => 2

新接手一个RAC环境，版本x86_64 Oracle 10.2.0.1，crsd.log频繁的出现警告，内容如下：2011-04-18 11:42:13.820: [ OCRSRV][1273047392]th_select_handler: Failed to retrieve procctx from ht. constr = [-1734320976] retval lht [-27] Signal CV.2011-04-18 11:42:15.076: [ CRSEVT][1545783648]0CAAMonitorHandler :: 0:Action Script for resource 'ora.bhcx_rac1.LISTENER_BHCX_RAC1.lsnr' stdout redirection failed for `/u01/app/oracle/product/10.2.0/crs/crs/log/startv1uXRJ.stdout` : No such file or directory2011-04-18

刚接手一个使用exp/imp进行备份的系统，查看备份日志时发现如下错误EXP-00079: Data in table "BILLMKTDETAIL" is protected. Conventional path may only be exporting partial table.. . exporting table BILLMKTDETAIL 21793 rows exportedEXP-00091: Exporting questionable statistics.EXP-00091: Exporting questionable statistics.[@more@]对于EXP-00079，Metalink Doc ID 277606.1的解决方案有三种：1. Run a DIRECT path export (Oracle8i and above), while connected as a user who has the

Active database duplicate，可以在没有备份的情况下复制一个比较新的数据库副本，这样，可以节省对数据库的备份时间。具体操作：[@more@]目标数据库实例准备C:Userssundog>oradim -new -sid TEST实例已创建。C:Userssundog>orapwd file=D:oracleproduct11.2.0dbhome_1databasePWDtest.ora password=sys entries=10创建目标数据库目录结构adminoradatafast_recovery_area创建目标数据库初始化参数文件C:Userssundog>sqlplus sys/sys@orcl as sysdbaSQL*Plus: Release 11.2.0.2.0 Production on 星期四 4月 7 14:29:27 2011Copyri

注册xmlschma时报错 SQL> begin 2 DBMS_XMLSCHEMA.registerSchema('UBL-CommonAggregateComponents-2.0.xsd', bfilename('XMLSCHEMA','UBL-CommonAggregateComponents-2.0.xsd')); 3 end; 4 / begin * 第 1 行出现错误: ORA-31083: 创建 SQL 类型 "NCTEST"."TransportHandlingUnitTy6750_T" 时出错 ORA-04031: 无法分配 7411440 字节的共享内存 ("shared pool","TransportHandlingUnitTy6750_T","type object de","KOKT object") ORA-06512: 在 &q

今天一个朋友碰到一点问题，目标环境是AIX RAC 10.2.0.4，alert日志里出现Thu Jan 6 09:04:13 2011ALTER SYSTEM SET service_names='orcl' SCOPE=MEMORY SID='orcl1';Thu Jan 6 09:04:13 2011Immediate Kill Session#: 252, Serial#: 25Immediate Kill Session: sess: 70000020821ec30 OS pid: 201078Thu Jan 6 09:04:13 2011Process OS id : 201078 alive after killErrors in file Thu Jan 6 09:04:13 2011ALTER SYSTEM SET service_names='' SCOPE=MEMORY SID='orcl1';Immediate Kill Session#: 253, Serial#: 1332Immediate Kill