ITPub博客

Here, I would like to keep a note for the installation of Oracle RAC DB.Last month, I failed in the installation for two times. First of all, I got an error about virtual network due to the wrong configuration of Network on VIP address.I can ping the VIPs before installation,this is totally uncorrect configuration. After disable the specified network card it worked fine.The second failure is about private network.The first node rebooted unfortunately when I run the script(root.sh) on the second

Today when I installed a new Oracle RAC DB, an error occured which is about "The specified nodes are not clusterable".The cluster nodes are in a domain network, and the public/private/virtual name are all with the domain name.Before I run the SHELL script .runInstaller.sh, I only run the security authentication commands to identify these issues:ssh node1.domain.com datessh node2.domain.com datessh node1-priv.domain.com datessh node2-priv.domain.com dateBut I didn't run the command like

Don't use package DBMS_STATS to do that, using FND_STATS is a good command.Also,don't use the ANALYZE command.Eg.Use the following command to gather schema statistics: exec fnd_stats.gather_schema_statistics('PO') exec fnd_stats.gather_schema_statistics('ALL')Use the following command for gathering statistics on a temporary table(ie: temporary tablename = temp1 in schema INV): exec fnd_stats.gather_table_stats('INV','temp1');[@more@]

My last blog article was posted one year ago here,it's coincidence that I logged in my blog today.It is a whole year which flied quickly,here I have to say really SORRY to the visitors who came to my blog with hope and left with nothing new.In the past one year,I spent more time enjoying my life with my family,reserching some new technologies on oracle.Some got,some lost......Tomorrow is nice and I'm ready to go,here I made a promise that I'd like to bring lots of new and valueable articles on d

I'v been back from USA for more than a week so far and everything is on the right track,like time difference,food,culture and environment etc.This trip is really impressive and absolutely unforgettable,and I'll keep it in mind as a great fortune forever.The American life is amazing at some points.For example,most of them own a big big house,whereas we just own a small small apartment;basically they drive a SUV car but a few of us drive a uncomfortable car;it is leisure time for them to work beca

Due to work and study,sorry for updating my web blog so late!In these several months there were great changes about me.First,I quited my job from Aspire for the consideration of working career.Second,I realized the importantance of English and made great progress on it.Third,It's my honor of being a member of a great foreign company which is expanding my career greatly.Finally,After succeeding in applying for passport and US visa,I'm about to go abord,because I'm going to the US for training nex

1、put the following file(login.sql) into $ORACLE_HOME；2、set environment virable：SQLPATH=$ORACLE_HOME3、the content of file(login.sql) as follows：define _editor=viset serveroutput on size 1000000set trimspool onset long 5000set linesize 100set pagesize 9999column plan_plus_exp format a80column global_name new_value gnameset termout offdefine gname=idlecolumn global_name new_value gnameselect lower(user) || '@' || substr( global_name, 1,decode( dot, 0, length(global_name), dot-1) ) global_nam

RAC：Oracle rac enables multiple independent servers that are linked by an interconnect to share access to database,providing ha,scalability,and redundancy during failure.RAC provide the benefit of the system failure,such as node failures and instance crashes;DG：dg ensures ha,data protection,and disaster recovery for enterprise data,it provide the benefit of the site failure.Flashback Database ：it provide fast recovery from logical data corruption and user errors.RMAN ：it is an utility that s

RPO(Recovery Point Objective) is focus on the loss of data;RTO(Recovery Time Objective) is focus on the loss of service.Some solutions for disaster recovery：1.Disk array copy technology(EMC SRDF,IBM PPRC,HP BusinessCopy,HDS TrueCopy);2.Storage volume copy technology(Veritas VVR);3.Datatabase copy technology(DSG RealSync,Quest SharePlex);4.Application copy technology.You should pay attention to three key considerations:TCO(Total Cost of the Ownership)ROI(Return On Investment)BCM(Business Continu

IBM最近提出了一个smarter planet的plan，旨在着手改善世界，使得人类赖以生存的星球更加智能。试想一下，如果我们能够通过改善电网就能节约50％左右的电能，如果我们能够改善交通就能节省途中时间、减少事故等，如果我们能够提升食品质量就能挽救无辜的生命等等。这不是如果，目前人类已经在打造这些如果成为fact。事实上，现在我们可以借助科技的力量来change这个星球。随着IT技术的发展，我们可以利用新的技术和新的理念来智能化这个世界上的任何事和物，使得它们变得更好，更有效，更加服务于我们这个星球。目前，该smarter planet还没

最近较闲，着手准备一些技能培训，其中涉及到SQL Tuning，如下是我总结出的10条SQL语句优化精髓：1、使用绑定变量；2、产生的redo和undo量越少越好；3、避免排序操作；4、避免复杂嵌套的语句；5、使用索引并正确利用上索引；6、控制好事务的粒度；7、走最优的执行计划；8、编写最佳的等价语句；9、可不可以使用上物化视图特性和分区特性；10、正确处理锁等待事件；这里每一条都可展开一个topic来讲解。先分享给大家，有时间再一一详细的阐述！[@more@]

要离开工作快4年的卓望(www.aspire-tech.com)了，要离开这座熟悉的深港基地大楼了。很感谢aspire让我成长，很感谢aspire的人和事。。。。。。欢迎大家推荐数据库的相关职位哈，最好base在深圳。thanks.[@more@]

HP Oracle Database Machine Hardware(all hardware components in one standard (42U) rack)8–HP Proliant DL360 G5 database servers, with• 2 quad-core Intel Xeon Processor E5430 (2.66GHz)• 32GB memory• 1–HP InfiniBand Dual Port HCA• 4–146GB SAS 10K hard disk drives4–24-port InfiniBand switches14–HP Exadata Storage Server Hardware—each is an HP ProLiant DL180 G5, with• 2 quad-core Intel Xeon Processor E5430 (2.66GHz)• 8GB memory• 1–HP InfiniBand Dual Po

在RAC的Oracle Net配置中有三种模式，这里记录一下：1、Connect-time Failover方式；只有一个service，会尝试第一个node，然后尝试另外的node可以failover，不过balance需手工当一个node发生故障时，该node上的事务丢失，会话丢失，自动重连。如下的例子：imall =(DESCRIPTION=(ADDRESS_LIST=(LOAD_BALANCE=OFF)(FAILOVER=ON)(ADDRESS=(PROTOCOL=TCP)(HOST=imall1)(PORT=1521))(ADDRESS=(PROTOCOL=TCP)(HOST=imall2)(PORT=1521)))(CONNECT_DATA=(SERVICE_NAME=imall)))2、Load Balancing方式；[@more@]只有一个service，Oracle Net会尝试其

虽说10月底了，深圳的气候仍然像夏天一样炎热，只不过现在下班后那天是黑朦朦的。最近一直都很忙，应该说这大半年一直都很忙，忙中有乐忙中偷闲。首先祝贺自己通过了Oracle的OCM大师级别认证，让自己能更加深入学习和传播Oracle数据库。给二家软件公司培训Oracle的一系列课程也耗去了我的许多空闲时间，不过总体感觉下来效果还不错的，接下来还会有一个系列的培训课程应该是比较轻松的了。给一家医院做数据库的故障恢复就像是医生给病人看病一样，有时隔行还真的隔着大山。替一个老板负责他公司的系统优化工作耗去了我多个晚上的休息时间，

1、4种安装模式说明：新的数据库和实例emrep、repository、oms、agent；2、采用已有数据库安装时需提供hostname、port、service/SID、sys的password、两个表空间MGMT_TABLESPACE和MGMT_ECM_DEPOT_TS，repository对应schema是sysman、安装前drop掉，dispatchers参数不能设置、session_cached_cursors》＝200、aq_tm_process>=1、执行两个脚本dbmspool.sql和prvtpool.plb；3、oms端的secure通过在安装时require secure communications选项来设置，agent端的secure可通过emctl secure agent命令来设置；4、agent10g/sysman/admin/script/*

1、Stream的源数据库需为archived模式；2、参数修改：global_names置为true；compatible>=10.2；job_queue_processes=4；aq_tm_processes=4;open_links=4;3、两边设置stream的管理员用户strmadm，角色有connect、resource、dba、create any job等；4、两边的db_domain不需设置；5、global_name的修改update global_name set global_name='dg1';commit;再重启实例即可；6、配置tnsnames.ora使得两边可以互相访问；7、创建dblink使得两边可以互相访问，db link的名称为global_name的值；8、若创建schema需保证源库上schema对应的表空间在

1、Report：分析repository并报告各种问题a)目标数据库的物理结构：report schemab)需要备份的数据文件：report need backup incremental 3report need backup days 3 tablespace ts_testreport need backup redundancy 2 databasereport need backup recovery window of 6 daysc)陈旧备份：report obsolete redundancy 2d)不可恢复数据文件(nologging时)：report unrecoverable[@more@]2、List：分析repository并罗列备份集和映像副本a)数据文件备份集：list backup of tablespace ts_test;b)控制文件备份集：list backup of controlfile

1、 设置Flash recovery area：两个动态参数(db_recovery_file_dest_size和db_recovery_file_dest)2、 设置db_flashback_retention_target3、 设置flashback database为on：Shutdown immediate;Startup mount exclusive;Alter database flashback on;Alter database open;4、 闪回操作：Shutdown immediate;Startup mount;flashback database to timestamp(to_date('2008-08-07 08:16:07','yyyy-mm-dd hh24:mi:ss'))alter database open resetlogs;5、 查询：V$flashback_database_logfileV$flashback_database_logV$flashback_datab

首先是三个参数的引入trc_level:跟踪的级别，可取值off，user，admin和supporttrc_directory:跟踪产生的trace文件所在目录trc_file:跟踪产生的trace文件具体的命令使用格式如下：[@more@]LSNRCTL> help set trc_fileset|show trc_{ } [<value>]: set|show trace parameters of current listenerLSNRCTL> help set trc_directoryset|show trc_{ } [<value>]: set|show trace parameters of current listenerLSNRCTL> help set trc_levelset|show trc_{ } [<value>]: set|show trace parameters of current list