数据库管理-第九十六期 19c OCM之路-第六堂（20230801）

最新推荐文章于 2024-02-12 10:44:58 发布

胖头鱼的鱼缸（尹海文）

最新推荐文章于 2024-02-12 10:44:58 发布

阅读量338

点赞数 1

分类专栏： Oracle 文章标签：数据库

本文链接：https://blog.csdn.net/yhw1809/article/details/132041759

版权

Oracle 专栏收录该内容

169 篇文章 34 订阅

订阅专栏

第九十六期 19c OCM之路-第六堂（20230801）

首先，我已经报考了9月1日于上海举行的19c OCM升级考试，希望能一步到位！
终于来到了最后一个SkillSet: Configuring Real Application Clusters配置RAC集群，这里和11g从头开始，安装GI，DB到创建数据库不一样，从12c OCM考试开始，是安装好了GI和DB软件的，就部署难度而言是有所下降的。但是就RAC本身配置来说，却增加了不少内容。

1 环境

总共有10个已经完成处理的用于ASM的磁盘，分别为DISK01-10（asmca中可直接看到），根据题目对其进行操作使用。其中DISK01-04已经创建为+DGDATA磁盘组。
GI_HOME=/u01/app/19.0.0/grid
ORACLE_HOME=/u01/app/oracle/product/19.0.0/dbhome_1

2 考题

Enabling Grid Infrastructure 开启GI自动启动
Administering ASM disk groups 管理ASM磁盘组
1)create asm diskgroup +DGACFS using DISK05 and DISK06 with external redundancy.
2)create asm diskgroup +DGFRA using DISK07 and DISK08 with external redundancy.
3)Create ADVM volume name ACFS_VOL in +DGACFS disk group with size 200M.
4)Create a ACFS file system on ACFS_VOL and mount it at: /u01/app/oracle/acfs_share
5)Add one more OCR image in +DGACFS and +DGFRA disk group
6)Copy /home/oracle/scripts/images.zip to /u01/app/oracle/acfs_share, and create a snapshot name SNAP_001 for ACFS
Managing Grid Infrastructure 管理GI
1)Export the OLR on the host03 to the directory /home/grid/olrdump
2)Configure OCR backup location to /home/grid/ocrbackup
3)Configure the cluster policy management method by referring the followings contents from /home/oracle/scripts/policysetfile.txt. Perform the neceassary operation to make the configuration be implemented, the active day policy.
4)Configure eth1 and eth2 both for privte and asm
Creating a RAC database 创建RAC数据库
1)The datbase name is RACDB.
2)General purpose.
3)Policy management in racdbpool server pool.
4)Do not configure EM.
5)The password for sys and system is oracle.
6)The datafiles are stored in +DGDATA disk group.
7)Fast recovery area is +DGFRA disk group and the size is 4GB.
8)Do not install sample schema
9)The instance total memory is 800MB.
Additional configuration 附加配置
1)Create a service named SRV1 in racdbpool and run SRV1 on host03 when start it.
2)Create a service named SRV2 in racdbpool, which need to be started to support transaction level failover.
3)Convert RAC to RAC ONENODE
4)Run the SQL scripts, how to make proper action to lower the contention of insert activties issued by millions of concurrent user from two nodes.
5)Convert RAC ONENODE backup to normal RAC mode
6)Configure database to reduce duplicate blocks parallel query image.
7)Configure the database backup the controlfile automatically, regardless of any change on the physical structure of the database.

3 操作

host03,host04上操作

sudo /u01/app/19.0.0/grid/bin/crsctl enable crs

题目1)-4)使用ASMCA进行操作

5)
sudo /u01/app/19.0.0/grid/bin/ocrconfig -add +DGACFS
sudo /u01/app/19.0.0/grid/bin/ocrconfig -add +DGFRA

6)
sudo /u01/app/19.0.0/grid/bin/ocrdump -local /home/grid/olrdump

3.GI配置

1)
cp /home/oracle/scripts/images.zip /u01/app/oracle/acfs_shars
sudo /sbin/acfsutil snap create SNAP_001 /u01/app/oracle/acfs_share

2)
mkdir /home/grid/ocrbackup
sudo /u01/app/19.0.0/grid/ocrconfig -backuploc /home/grid/ocrbackup

3)policysetfile.txt原文件为：
SERVER_POOL_NAMES=Free bigpool ora.racdbpool
POLICY
NAME=day
DESCRIPTION=The day policy
SERVERPOOL
NAME=ora.racdbpool
IMPORTANCE=10
MAX_SIZE=5
MIN_SIZE=1
SERVER_CATEGORY=ora.hub.category
SERVERPOOL
NAME=bigpool
IMPORTANCE=0
MAX_SIZE=0
MIN_SIZE=0
SERVER_CATEGORY=big
POLICY
NAME=night
DESCRIPTION=The night policy
SERVERPOOL
NAME=ora.racdbpool
IMPORTANCE=10
MAX_SIZE=5
MIN_SIZE=1
SERVER_CATEGORY=ora.hub.category
SERVERPOOL
NAME=bigpool
IMPORTANCE=5
MAX_SIZE=1
MIN_SIZE=1
SERVER_CATEGORY=big

sudo cp /home/oracle/scripts/policysetfile.txt /home/grid/
sudo chown grid:oinstall /home/grid/policysetfile.txt
#需要对文件进行缩进处理
SERVER_POOL_NAMES=Free bigpool ora.racdbpool
POLICY
  NAME=day
  DESCRIPTION=The day policy
  SERVERPOOL
    NAME=ora.racdbpool
    IMPORTANCE=10
    MAX_SIZE=5
    MIN_SIZE=1
    SERVER_CATEGORY=ora.hub.category
  SERVERPOOL
    NAME=bigpool
    IMPORTANCE=0
    MAX_SIZE=0
    MIN_SIZE=0
    SERVER_CATEGORY=big
POLICY
  NAME=night
  DESCRIPTION=The night policy
  SERVERPOOL
    NAME=ora.racdbpool
    IMPORTANCE=10
    MAX_SIZE=5
    MIN_SIZE=1
    SERVER_CATEGORY=ora.hub.category
  SERVERPOOL
    NAME=bigpool
    IMPORTANCE=5
    MAX_SIZE=1
    MIN_SIZE=1
    SERVER_CATEGORY=big
 
crsctl add category big
crsctl modify policyset -file /home/grid/policysetfile.txt
crsctl modify policyset -attr "LAST_ACTIVATED_POLICY=day"

4)
oifcfg setif -global eth1/10.0.0.0:cluster_interconnect,asm
oifcfg setif -global eth2/10.0.0.0:cluster_interconnect,asm

sudo /u01/app/12.1.0/grid/bin/crsctl stop cluster -all
sudo /u01/app/12.1.0/grid/bin/crsctl start cluster -all

文档位置：
Oracle Database 19c
Clusterware Administration and Deployment Guide
$\qquad$ -> E Oracle Clusterware Control (CRSCTL) Utility Reference
$\qquad\qquad$ -> Oracle RAC Environment CRSCTL Commands
$\qquad\qquad\qquad$ -> crsctl modify policyset

使用DBCA创建数据库
附加配置

1)
srvctl add service -d racdb -s SRV1 -g racdbpool -c singleton
srvctl start service -db racdb -service SRV1

2)
srvctl add service -db racdb -service SRV2 -serverpool racdbpool -failovertype transaction -commit_outcom true -retention 86400 -notification true
srvctl start service -db racdb -service SRV2

3)
srvctl stop instance -db racdb -node host04 -force
srvctl convert database -db racdb -dbtype RACONENODE
srvctl config database -d racdb

4)脚本内容：
#create table t1 (id number primary key,name varchar2(10),createdate date) partition by range(createdate) interval(numtoyminterval(1,'MONTH')) (partition SAL1 values less than (to_date('2007-01-01','yyyy-mm-dd')));
#create sequence seq_t1;

alter sequence seq_t1 cache 100000;
select constraint_name,constraint_type,index_name from dba_constraints where table_name='T1';
alter table t1 drop constraint <constraint_name>;
drop index <constraint_name>;
create unique index idx_t1 on t1(id) parallel 4 global partition by hash(id) partitions 4;
alter table t1 add constraint pd_t1 primery key(id) using index idx_t1;

5)
srvctl convert database -db racdb -dbtype RAC

6)sqlplus / as sysdba
alter system set parallel_force_local=true;

7)rman target/
configure controlfile autobackup on;
configure controlfile autoback format for device type disk to '+DGDATA';
configure snapshot controlfile name to '+DGDATA/snapcf_RACDB.f';