《ORACLE11GR2在REDHAT63环境中搭建RAC集群及简单维护.docx》由会员分享，可在线阅读，更多相关《ORACLE11GR2在REDHAT63环境中搭建RAC集群及简单维护.docx（66页珍藏版）》请在三一办公上搜索。
1、1、 系统环境1.1、 硬件环境服务器：vmware 10，新建个虚拟机主机名分别为rac1和rac2，每个虚拟机分配40G磁盘空间，添加两个网络适配器。其中第二个适配器网络连接调整为自定义，且两个节点保持一致。Widonws本机ip：192.168.6.11.2、 软件环境数据库：oracle11.2.0.4 database-x86-64GRID：oracle11.2.0.4_grid-x86-64 操作系统：rhel-server-6.3-x86_64采用最小化安装1.3、 网络环境Ip地址规划分配为IP 名称子网掩码IP 地址Rac1-public255.255.255.0192.16
2、8.6.11Rac2-public255.255.255.0192.168.6.12Rac1-private255.255.255.02.2.2.2Rac2-private255.255.255.02.2.2.3Rac1-vip255.255.255.0192.168.6.13Rac2-vip255.255.255.0192.168.6.14SCAN255.255.255.0192.168.6.151.4、 共享磁盘分区计划创建四个共享磁盘sdb、sdc、sdd、sde，每个磁盘计划分三个分区2、 前期环境准备1.2.1 配置静态地址vi /etc/sysconfig/network-scri
3、pts/ifcfg-eth0修改ip地址。每个虚拟机eth0网卡为public，DEVICE=eth0BOOTPROTO=staticHWADDR=00:0C:29:D1:4E:A6NM_CONTROLLED=yesONBOOT=yesTYPE=Ethernet*UUID=e59cb6a0-deb0-4164-a2b0-8b4dcc0cb027IPADDR=192.168.6.11NETMASK=255.255.255.0GATEWAY=192.168.6.1vi /etc/sysconfig/network-scripts/ifcfg-eth1修改ip地址。每个虚拟机eth1网卡位priva
4、teDEVICE=eth1BOOTPROTO=staticHWADDR=00:0C:29:D1:4E:A6NM_CONTROLLED=yesONBOOT=yesTYPE=Ethernet*UUID=e59cb6a0-deb0-4164-a2b0-8b4dcc0cb027IPADDR=2.2.2.2NETMASK=255.255.255.0修改完成后执行service network restart2.2 在rac1和rac2上分别关闭防火墙service iptables stop -停止防火墙chkconfig iptables off -禁用防火墙2.3 在rac1和rac2上分别修改主机
5、名vi /etc/sysconfig/network 重启生效，一个rac1，另一个rac2HOSTNAME=rac12.4 在rac1和rac2 上分别改hosts vi /etc/hosts 添加对应的ip信息#public192.168.6.180 rac1192.168.6.181 rac2#private2.2.2.1 rac1-priv2.2.2.2 rac2-priv#virtual192.168.6.182 rac1-vip192.168.6.183 rac2-vip#scan192.168.6.184 cluster-scan2.5 在rac1和rac2上分别执行配置内核参数
6、vi /etc/sysctl.conf加入以下内容fs.aio-max-nr = 1048576fs.file-max = 6815744kernel.shmall = 2147483648kernel.shmmax = 68719476736kernel.shmmni = 4096kernel.sem = 250 32000 100 128net.ipv4.ip_local_port_range = 9000 65500net.core.rmem_default = 262144net.core.rmem_max = 4194304net.core.wmem_default = 262144
7、net.core.wmem_max = 1048586使修改参数立即生效：sysctl -p2.6 在rac1和rac2上分别执行修改limitsvi /etc/security/limits.conf加入以下信息grid soft nproc 2047grid hard nproc 16384grid soft nofile 1024grid hard nofile 65536oracle soft nproc 2047oracle hard nproc 16384oracle soft nofile 1024oracle hard nofile 655362.7 在rac1和rac2上分别
8、修改/etc/pam.d/loginvi /etc/pam.d/login加入以下信息session required /lib/security/pam_limits.so session required pam_limits.so2.8 在rac1和rac2上分别执行修改/etc/profilevi /etc/profile加入以下信息if $USER = oracle | $USER = grid ; thenif $SHELL = /bin/ksh ; thenulimit -p 16384ulimit -n 65536elseulimit -u 16384 -n 65536fium
9、ask 022fi2.9 在rac1和rac2 上分别执行禁用 selinuxvi /etc/selinux/config修改 SELINUX值SELINUX=disabled2.10 在rac1和rac2上分别执行停止 ntp 服务service ntpd stopchkconfig ntpd offmv /etc/ntp.conf /etc/ntp.conf.bak2.11 在rac1和rac2上分别处理/dev/shm 共享内存不足的处理df -h 查看tmpfs分区是否大于1G，如果过小需增加。vi /etc/fstab 默认的：tmpfs /dev/shm tmpfs default
10、s 0 0改成：tmpfs /dev/shm tmpfs defaults,size=1024m 0 0mount -o remount /dev/shm2.12 在rac1和rac2上分别检查软件是否全部安装 rpm -qa | grep binutils-rpm -qa | grep compat-libstdc+-rpm -qa | grep elfutils-libelf-rpm -qa | grep elfutils-libelf-devel-rpm -qa | grep glibc-rpm -qa | grep glibc-common-rpm -qa | grep glibc-d
11、evel-rpm -qa | grep gcc-rpm -qa | grep gcc-c+-rpm -qa | grep libaio-rpm -qa | grep libaio-devel-rpm -qa | grep libgcc-rpm -qa | grep libstdc+-rpm -qa | grep libstdc+-devel-rpm -qa | grep make-rpm -qa | grep sysstat-rpm -qa | grep unixODBC-rpm -qa | grep unixODBC-devel-2.13 在rac1和rac2 2分别将未安装的包通过yum安
12、装mkdir /yummount /dev/cdrom /yumvi /etc/yum.repos.d/chenbin.repo添加以下内容 rhel-chenbinname=Red Hat Enterprise Linux $releasever - $basearch - Debugbaseurl=file:/yum/enabled=1gpgcheck=1gpgkey=file:/etc/pki/rpm-gpg/RPM-GPG-KEY-redhat-releaseyum list -查看可用包yum -y install binutils* compat-* elfutils-libelf
13、* gcc-* gcc-* kernel-* ksh-* libaio-* libgcc-* libgomp-* libstdc+-* make-* numactl-devel-* sysstat-* unixODBC-* pdksh*2.14 在oradb1和rac2上分别关闭不需要的服务chkconfig autofs offchkconfig acpid offchkconfig sendmail offchkconfig cups-config-daemon offchkconfig cpus offchkconfig xfs offchkconfig lm_sensors offch
14、kconfig gpm offchkconfig openibd offchkconfig pcmcia offchkconfig cpuspeed offchkconfig nfslock offchkconfig ip6tables offchkconfig rpcidmapd offchkconfig apmd offchkconfig sendmail offchkconfig arptables_jf offchkconifg microcode_ctl offchkconfig rpcgssd offchkconfig ntpd off3、 添加组和用户233.1 在rac1和ra
15、c2上分别添加oracle和grid用户和组groupadd -g 501 oinstallgroupadd -g 502 dbagroupadd -g 503 opergroupadd -g 504 asmadmingroupadd -g 505 asmopergroupadd -g 506 asmdbauseradd -g oinstall -G dba,asmdba,oper oracleuseradd -g oinstall -G asmadmin,asmdba,asmoper,oper,dba grid3.2 在rac1和rac2上分别为oracle和grid用户设密码passwd
16、oraclepasswd grid3.3 在rac1和rac2上分别创建目录grid 和 oracle 用户的mkdir -p /u01/app/oraclemkdir -p /u01/app/gridmkdir -p /u01/app/11.2.0/gridchown -R grid:oinstall /u01/app/gridchown -R grid:oinstall /u01/app/11.2.0chown -R oracle:oinstall /u01/app/oraclechmod -R 775 /u01mkdir -p /u01/app/oraInventorychown -R
17、grid:oinstall /u01/app/oraInventorychmod -R 775 /u01/app/oraInventory3.4 在rac1和rac2上分别修改oracle用户的.bash_profile文件vi /home/oracle/.bash_profileexport ORACLE_SID=rac1export ORACLE_BASE=/u01/app/oracleexport ORACLE_HOME=$ORACLE_BASE/product/11.2.0/dbhome_1export LD_LIBRARY_PATH=$ORACLE_HOME/libexport NL
18、S_DATE_FORMAT=yyyy-mm-dd HH24:MI:SSexport TMP=/tmpexport TMPDIR=$TMPexport PATH=$PATH:$ORACLE_HOME/bin3.5 在rac1和rac2上分别修改grid用户的.bash_profile文件vi /home/grid/.bash_profileexport ORACLE_SID=+ASM1export ORACLE_BASE=/u01/app/gridexport ORACLE_HOME=/u01/app/11.2.0/gridexport LD_LIBRARY_PATH=$ORACLE_HOME/
19、libexport NLS_DATE_FORMAT=yyyy-mm-dd HH24:MI:SSexport PATH=$ORACLE_HOME/bin:$PATH注意另外一台数据库实例名须做相应修改：Oracle：export ORACLE_SID=rac2grid：export ORACLE_SID=+ASM23.6 建立 ssh 等效性（可不配置）Root用户等效性可不用配置，配置了仅管理服务器方便，gird用户和oracle的ssh等效性由安装程序配置，mkdir /.ssh chmod 755 /.ssh ssh-keygen -t rsassh-keygen -t dsacat /.
20、ssh/id_rsa.pub /.ssh/authorized_keyscat /.ssh/id_dsa.pub /.ssh/authorized_keys在rac1上执行ssh rac2 cat /.ssh/id_rsa.pub /.ssh/authorized_keys ssh rac2 cat /.ssh/id_dsa.pub /.ssh/authorized_keys在rac2上执行ssh rac1 cat /.ssh/id_rsa.pub /.ssh/authorized_keys ssh rac1 cat /.ssh/id_dsa.pub /.ssh/authorized_keys
21、在两台服务器上验证是否可以正常访问对方，执行时不再提示输入密码，则表示SSH 对等性配置成功ssh rac1 datessh rac2 datessh rac1-priv datessh rac2-priv date4、 添加共享磁盘（VM）44.1 在本机上创建共享磁盘生产环境使用共享存储，直接略过在windows的cmd中进入VMware Workstation安装目录，执行命令创建磁盘：注意创建共享磁盘的路径及文件大小，例如我的VM装在c盘，cd c:Program Files (x86)VMwareVMware Workstationvmware-vdiskmanager.exe -c
22、 -s 5g -a lsilogic -t 2 F:shdiskvot.vmdkVmware-vdiskmanager.exe -c -s 2g -a lsilogic -t 2 F:shdiskfra.vmdkvmware-vdiskmanager.exe -c -s 5g -a lsilogic -t 2 F:shdiskdata.vmdkvmware-vdiskmanager.exe -c -s 5g -a lsilogic -t 2 F:shdiskdata1.vmdk4.2 将共享磁盘挂到两台虚拟机关闭两台虚拟机并退出VM，分别用记事本打开虚拟机安装位置中虚拟机名字. vmx，添加以
23、下内容disk.EnableUUID=TRUEdisk.locking = FALSEscsi1.shared = TRUEdiskLib.dataCacheMaxSize = 0diskLib.dataCacheMaxReadAheadSize = 0diskLib.dataCacheMinReadAheadSize = 0diskLib.dataCachePageSize= 4096diskLib.maxUnsyncedWrites = 0 scsi1.present = TRUEscsi1.virtualDev = lsilogicscsil.sharedBus = VIRTUALscs
24、i1:0.present = TRUEscsi1:0.mode = independent-persistentscsi1:0.fileName = F:shdiskvot.vmdkscsi1:0.deviceType = diskscsi1:0.redo = scsi1:1.present = TRUEscsi1:1.mode = independent-persistentscsi1:1.fileName = F:shdiskfra.vmdkscsi1:1.deviceType = diskscsi1:1.redo = scsi1:2.present = TRUEscsi1:2.mode
25、= independent-persistentscsi1:2.fileName = F:shdiskdata.vmdkscsi1:2.deviceType = diskscsi1:2.redo = scsi1:3.present = TRUEscsi1:3.mode = independent-persistentscsi1:3.fileName = F:shdiskdata1.vmdkscsi1:3.deviceType = diskscsi1:3.redo = 4.3 两个虚拟机开机，格式化硬盘，确认是否共享成功。在rac1上执行fdsik -l -确认看以正常看到sdb,sbc,sdd
26、,sde按前期规划创建分区，分区完成后在rac2 中执行fdisk -l 查询是否同步4.4 在rac1和rac2上分别将裸设备文件和分区设备文件进行绑定vi /etc/udev/rules.d/60-raw.rules 根据磁盘情况修改下面信息加入到文件中ACTION=add, KERNEL=sdb1, RUN+=/bin/raw /dev/raw/raw1 %NACTION=add, KERNEL=sdb2, RUN+=/bin/raw /dev/raw/raw2 %NACTION=add, KERNEL=sdb3, RUN+=/bin/raw /dev/raw/raw3 %NACTION
27、=add, KERNEL=sdc1, RUN+=/bin/raw /dev/raw/raw4 %NACTION=add, KERNEL=sdc2, RUN+=/bin/raw /dev/raw/raw5 %NACTION=add, KERNEL=sdc3, RUN+=/bin/raw /dev/raw/raw6 %NACTION=add, KERNEL=sdd1, RUN+=/bin/raw /dev/raw/raw7 %NACTION=add, KERNEL=sdd2, RUN+=/bin/raw /dev/raw/raw8 %NACTION=add, KERNEL=sdd3, RUN+=/
28、bin/raw /dev/raw/raw9 %NKERNEL=raw0-9, OWNER=grid, GROUP=asmadmin, MODE=0660此处、KERNEL=”raw0-9” 10以必须单独指定，否则无效、MODE=”0666”和GROUP=”oinstall”的设置，在后面的check会报waring应改为:MODE=”0660 GROUP=”asmadmin”执行：start_udev命令，此处建议重启虚拟机执行 raw -qa查看绑定状态:确保两个节点看以看到相同的内容执行ls -l /dev/raw/查看 裸设备归属权限:5、 安装Grid55.1 在rac1和rac2上
29、分别安装cvuqdisk包在两个 Oracle RAC 节点上安装操作系统程序包 cvuqdisk。cvuqdisk RPM 包含在 Oracle Grid Infrastructure 安装介质上的 rpm 目录中。rpm -ivh cvuqdisk-1.0.7-1.rpm5.2 检查 CRS 的安装环境只需要在其中一个节点的gird用户执行.执行需提前设置好grid、oracle用户ssh等效性，本次不检查./runcluvfy.sh stage -pre crsinst -n rac1,rac2 -verbose未检测通过的显示为failed，根据情况修复一下即可：resolv.conf
30、不一致的DNS问题可以忽略5.3 安装Grid Infrastructure软件5.3.1 开始安装export DISPLAY=192.168.6.18:0.0 ./runInstaller5.3.2 语言支持中文5.3.3 配置Scan name与host文件一致此处的SCAN_NAME与/etc/hosts 文件中的scan地址保持一致5.3.4 Grid用户等效性配置点击setup出现提示安装成功。Next5.3.5 创建OCR磁盘组设置asm密码 5.3.6 选择grid安装位置Package这个只要确保安装了ksh即可。 Task resolve此项是因为没有使用dns解析scan
31、，可忽略5.3.7 开始安装5.3.8 执行脚本在每个服务器上以root身份执行图中的2个脚本，两个节点都执行完成后点OK。/u01/app/oraInventory/orainstRoot.sh/u01/app/grid/11.2.0/root.sh出现以下字样表示在该节点安装成功Configure Oracle Grid Infrastructure for a Cluster . succeeded5.4 确认Grid安装5.4.1 CRS状态crs_stat -t -v除GSD外其他全部为onlinecrsctl stat res -t5.4.2 voting disk状态crsctl
32、 query css votedisk# STATE File Universal Id File Name Disk group- - - - -1. ONLINE 7b8903f49cc84fa8bf06d199bdf5dfe3 (ORCL:DISK01) CRSDG5.4.3 检查Oracle集群注册表(OCR)ocrcheckStatus of Oracle Cluster Registry is as follows :Version : 3Total space (kbytes) : 262120Used space (kbytes) : 2264Available space (
33、kbytes) : 259856ID : 1510360228Device/File Name : +CRSDGDevice/File integrity check succeededDevice/File not configuredDevice/File not configuredDevice/File not configuredDevice/File not configuredCluster registry integrity check succeededLogical corruption check bypassed due to non-privileged user5
34、.4.4 检查CRS状态crsctl check crsCRS-4638: Oracle High Availability Services is onlineCRS-4537: Cluster Ready Services is onlineCRS-4529: Cluster Synchronization Services is onlineCRS-4533: Event Manager is online6、 ASM磁盘组配置66.1 检查监听状态gridrac01 $ lsnrctl status(DESCRIPTION=(ADDRESS=(PROTOCOL=ipc)(KEY=LIS
35、TENER)(DESCRIPTION=(ADDRESS=(PROTOCOL=tcp)(HOST=172.18.3.211)(PORT=1521) (DESCRIPTION=(ADDRESS=(PROTOCOL=tcp)(HOST=172.18.3.213)(PORT=1521)Services Summary.Service +ASM has 1 instance(s).Instance +ASM1, status READY, has 1 handler(s) for this service.The command completed successfully6.2 创建DATA磁盘组以g
36、rid用户执行asmca命令，创建DATADG和FRADG两个磁盘组。6.3 创建FLA磁盘组：FLASHDG创建成功，退出ASMCA。验证：crs_stat -t -v或 crsctl stat res -t7、 安装数据库软件77.1 切换到oracle用户export DISPLAY=192.168.6.18:0.0 ./runInstaller 7.2 两个节点安装7.3 oracle用户等效性输入oracle用户密码。点击setup7.4 选择路径7.5 执行root脚本在两个节点上，分别以root身份执行上述脚本，然后点击OK。数据库软件安装完成，点击close退出。7.6 测试数
37、据库安装完成oraclerac1$ sqlplus / as sysdba SQL*Plus: Release 11.2.0.4.0 Production on Wed Oct 1 22:42:56 2014 Copyright (c) 1982, 2011, Oracle. All rights reserved. Connected to an idle instance. SQL8、 使用DBCA创建数据库88.1 使用oracle用户登录，执行dbca 8.2 启用em8.3 选择asm磁盘组8.4 选择FRA磁盘组：确定是否启用快速恢复区、归档日志设置最大连接数8.5 注意选合适的字
38、符集点击exit完成安装9、 集群数据库维护99.1 所有Oracle实例状态srvctl status database -d oradbInstance rac1is running on node node1Instance rac2 is running on node node29.2 单个Oracle实例状态srvctl status instance -d oradb -i rac1Instance rac1is running on node node19.3 节点应用程序状态srvctl status nodeappsVIP node1-vip is enabledVIP n
39、ode1-vip is running on node: node1VIP node2-vip is enabledVIP node2-vip is running on node: node2Network is enabledNetwork is running on node: node1Network is running on node: node2GSD is disabledGSD is not running on node: node1GSD is not running on node: node2ONS is enabledONS daemon is running on
40、 node: node1ONS daemon is running on node: node29.4 节点应用程序配置srvctl config nodeappsNetwork exists: 1/192.168.0.0/255.255.0.0/eth0, type staticVIP exists: /node1-vip/192.168.1.151/192.168.0.0/255.255.0.0/eth0, hosting node node1VIP exists: /node2-vip/192.168.1.152/192.168.0.0/255.255.0.0/eth0, hosting
41、 node node2GSD existsONS exists: Local port 6100, remote port 6200, EM port 20169.5 数据库配置srvctl config database -d oradb -aDatabase unique name: oradbDatabase name: oradbOracle home: /u01/app/oracle/product/11.2.0/dbhome_1Oracle user: oracleSpfile: +DATADG/oradb/spfileoradb.oraDomain:Start options:
42、openStop options: immediateDatabase role: PRIMARYManagement policy: AUTOMATICServer pools: oradbDatabase instances: rac1,rac2Disk Groups: DATADG,FRADGMount point paths:Services:Type: RACDatabase is enabledDatabase is administrator managed9.6 ASM状态srvctl status asmASM is running on node2,node19.7 ASM
43、配置srvctl config asm -aASM home: /u01/app/11.2.0/gridASM listener: LISTENERASM is enabled.9.8 TNS监听器状态srvctl status listenerListener LISTENER is enabledListener LISTENER is running on node(s): node2,node19.9 TNS监听器配置srvctl config listener -aName: LISTENERNetwork: 1, Owner: gridHome: /u01/app/11.2.0/grid on node(s) node2,node1End points: TCP:15219.10 SCAN状态srvctl status scanSCAN VIP scan1 is enabledSCAN VIP scan1 is running on node node2SCAN VIP scan2 is enabledSCAN VIP scan2 is running on node node1SCAN VIP scan3 is enabledSCAN VIP
