rac 数据库打patch set之后upgrade database.-CSDN博客

Linux ORACLE RAC 10201升级10203
一.准备工作
1.有效备份
作好对ORACLE软件和数据库的物理备份.
2.检查无效对象。
3.确保SYSTEM的空闲空间在50M以上。
4. SHARED_POOL_SIZE 和 JAVA_POOL_SIZE 至少为150Mb。
5.关闭数据库，关闭oracle进程。
6.备份相关配置文件
二.备份TSTZ
下载并解压Patch 5601428，应用该Patch（应用该Patch不需要关闭数据库）参考399832.1。
$unzip p5601428_10201_GENERIC.zip 5601428
$cd 5601428$opatch apply
$sqlplus ‘/as sysdba’SQ
L>@?/rdbms/admin/utltzuv2.sql
SQL>host
$cd $ORACLE_HOME /oracore/zoneinfo
$more tzuv2ext_*.log
$more tzuv2ext_*.bad
SQL>select * from sys.sys_tzuv2_temptab;

三.进行升级
1.升级CRS
$ cd Disk1
$ ./runInstaller
--next
--选择crs主目录
--next
--check summary
--install
--when prompt login as root,execute $ORACLE_HOME/root.sh on both nodes
2. 升级ORACLE 软件
$ cd Disk1
$ ./runInstaller
--选择oracle主目录
--next
--check summary
--install
--when prompt login as root,execute $ORACLE_HOME/root.sh on both nodes
在两节点上执行在$ORACLE_HOME/install下root102.sh
3.启动监听
$srvctl start listener -n jktz14.升级数据库
$sqlplus “/as sysdba”St
artup nomount
ALTER SYSTEM SET CLUSTER_DATABASE=FALSE SCOPE=spfile;
Shutdown immediate
Startup upgrade
Spool update.log
@?/rdbms/admin/catupgrd.sql
Spool off
ALTER SYSTEM SET CLUSTER_DATABASE=TRUE SCOPE=spfile;
cd $ORACLE_HOME/install
./changePerm.sh5.
检查无效对象情况，并对无效对象进行编译。
6.升级后检查
#su – oracle-
- Checking the Status of the Oracle Clusterware Installation
$crsctl query crs softwareversion – lists-
-chech crs status
ocrcheck
./app/oracle/product/10.2.0/crs/bin/olsnodes
crsctl query crs votedisk
crsctl query crs activeversion
$ crsctl check crs
$crs_stat -t
四.重启数据库，并进行应用测试。
五.回退方案1.卸载升级包
(1)把初始化参数CLUSTER_DATABASE为false.关闭数据库和ORACLE所有进程
(2) 备份下面两个文件
ORACLE_HOME/rdbms/admin/catrelod.sqlORACLE_HOME/network/admin/tnsnames.ora
(3)
$ cd $ORACLE_HOME
$ sqlplus /NOLOG
SQL> CONNECT SYS/SYS_password AS SYSDBA
SQL> STARTUP DOWNGRADE
SQL> SPOOL downgrade.out
SQL> @?/rdbms/admin/catdwgrd.sql
SQL> SPOOL OFF
SQL> SHUTDOWN IMMEDIATE
SQL> EXIT
(4)把前面备份文件拷到原目录。
$ sqlplus /NOLOG
SQL> CONNECT SYS/SYS_password AS SYSDBA
SQL> STARTUP DOWNGRADE
SQL> SPOOL catrelod.out
SQL> @?/rdbms/admin/catrelod.sql
SQL> SPOOL OFF
SQL> SHUTDOWN IMMEDIATE
(5)还原数据库文件，编译无效对象，并把CLUSTER_DATABASE参数设置为true.2.启用原有软件备份和原有备份数据库。

hpux下oracle RAC从10.2.0.3升级到10.2.0.4全过程（四）：升级补丁

首先升级clusterware补丁，在图形界面中运行./runInstaller进行安装，注意选择正确的路径，比如clusterware软件的路径在“/home/db/oracle/crs"下，那么就必须指定这个路径，否则得不偿失。

具体过程略过，在补丁安装完毕之后，系统提示需要分别在两个节点下进行如下步骤：

首先在副节点上按照提示的步骤进行执行（主次顺序无所谓）：
[storm02@/home/db/oracle/crs/bin]#./crsctl stop crs
Stopping resources. This could take several minutes.
Successfully stopped CRS resources.
Stopping CSSD.
Shutting down CSS daemon.
Shutdown request successfully issued.

这时可以看到副节点的vip地址被主节点所接管：

接着执行root102.sh脚本：
[storm02@/home/db/oracle/crs/bin]#cd ..
[storm02@/home/db/oracle/crs]#cd install
[storm02@/home/db/oracle/crs/install]#ls
cluster.ini         install.incl        preupdate.sh        rootdelete.sh       templocal
cmdllroot.sh        make.log            root102.sh          rootdeletenode.sh
envVars.properties paramfile.crs       rootaddnode.sbs     rootinstall
files10203.log      patch102            rootconfig          rootlocaladd
install.excl        prepatch10203       rootdeinstall.sh    rootupgrade
[storm02@/home/db/oracle/crs/install]#sh root102.sh
Creating pre-patch directory for saving pre-patch clusterware files
Completed patching clusterware files to /home/db/oracle/crs
Relinking some shared libraries.
Relinking of patched files is complete.
WARNING: directory '/home/db/oracle' is not owned by root
WARNING: directory '/home' is not owned by root
Preparing to recopy patched init and RC scripts.
Recopying init and RC scripts.
Startup will be queued to init within 30 seconds.
Starting up the CRS daemons.
Waiting for the patched CRS daemons to start.
This may take a while on some systems.
.
10204 patch successfully applied.
clscfg: EXISTING configuration version 3 detected.
clscfg: version 3 is 10G Release 2.
Successfully accumulated necessary OCR keys.
Using ports: CSS=49895 CRS=49896 EVMC=49898 and EVMR=49897.
node :
node 2: storm02 storm02_hb storm02
Creating OCR keys for user 'root', privgrp 'sys'..
Operation successful.
clscfg -upgrade completed successfully

黑体字部分显示副节点上clusterware的补丁已经成功安装。
补丁打完之后，重新启动副节点上crs进程，这个时候会把数据库一起给启动起来，所以切记一定要关闭数据库以及监听进程，防止在打database补丁的时候出错。

副节点上补丁打完之后，再按照相同的步骤给主节点打补丁，最后运行root102.sh脚本。

在这之后，分别在两个节点上进行clusterware版本的验证：
[storm01@/home/db/oracle/crs/bin]#./crsctl query crs softwareversion
CRS software version on node [storm01] is [10.2.0.4.0]
[storm02@/home/db/oracle/crs/bin]#./crsctl query crs softwareversion
CRS software version on node [storm02] is [10.2.0.4.0]
确认无错之后，紧接着进行数据库补丁的升级，其过程不赘述，但一定要保证监听和数据库是关闭状态。