【正文】 ost1vg 00f6f1569990a1ef 00f6f1569990a12c [host2][root][/]varyong host1vg；varyong host2vg[host2][root][/]lsfsName Options 10485760 rw no 15728640 rw no[host2][root][/]no po routerevalidate=1Setting routerevalidate to 1Setting routerevalidate to 1 in nextboot file確認(rèn)：[host2][root][/]no a|grep routerevalidate routerevalidate = 1 Ipkts Ierrs Collen0 2481098 0 1500 2481098 0 1500 142470 0 1500 link3 22 0 1500 1500 1335969 0 0lo0 1335968 0 Network Address 164734 0 0en0 164734 0 0en1 host1_l2_boot1 142476 0 1500 20 0 0en2 host1_l2_boot2 4 0 0en3 0 0 1335969 Network Address 1500 link4 12 0 0en1 12 0 0en2 20 0 0en2 6 0 0en3 6 0 0lo0 515177 0 0lo0515177 0 Ipkts Ierrs Collen0 1013619 0 link4 12 0 0en1 host2_l2_boot1 141876 0 link5 1500 515199 0 0 stop_host1 host2stop_host2banner stop host2banner end host2exit 0確認(rèn)host1: cat /etc/hosts/dev/tty0host2:cat/dev/tty0216。 兩邊用chdev l hdiskpower0 a pv=yes 先將兩邊的盤符認(rèn)出來，這樣之后系統(tǒng)才能自動(dòng)掃到磁盤... hdisk5 None .但壞處脈絡(luò)不清晰，在配置和日后使用時(shí)不能養(yǎng)成良好習(xí)慣，必然造成兩邊的經(jīng)常不一致，使得停機(jī)整理VG這樣各節(jié)點(diǎn)同步的事情重復(fù)勞動(dòng)，并且很容易疏忽和遺漏。 smitty hacmp Extended ConfigurationExtended Topology ConfigurationConfigure HACMP NodesAdd a Node to the HACMP Cluster同理可以添加第二個(gè)節(jié)點(diǎn)向這些網(wǎng)絡(luò)添加boot地址網(wǎng)絡(luò)接口：smitty hacmp Extended Configuration Extended Topology ConfigurationConfigure HACMP Communication Interfaces/DevicesAdd Communication Interfaces/DevicesAdd Predefined Communication Interfaces and Devices Communication Interfaces 選擇之前建立 的net_ether_01增加2個(gè)boot地址：1. Device /dev/tty0host2 smitty hacmpSystem Management (CSPOC)StorageVolume GroupsManage Concurrent Access Volume Groups for MultiNode Disk HeartbeatCreate a new Volume Group and Logical Volume for MultiNode Disk Heartbeatsmit hacmpExtended ConfigurationExtended Topology ConfigurationShow HACMP Topology Network net_diskhbmulti_01 host1_l2_boot1 host1_l2_boot2 Network net_rs232_01 Network net_diskhb_01 host2_l1_boot2 host2_l1_boot1 Network net_ether_02 host1_l1_boot1 host1_l1_boot2 Network net_ether_02 NODE host2: Network net_ether_01 host2_tty0_01 可以看到已符合規(guī)劃要求，可繼續(xù)了 . [/usr/sbin/cluster/app/start_host1]* Stop Script Application Monitor Name(s)同樣增加 host2_app* Server N。 [host1_app]*Start ScriptConfigure HACMP Applications Network net_ether_02 host2_l1_boot1 host2_l1_boot2 host1_tty0_01NODE host1: host2_2 /dev/mndhb_lv_01 Network net_ether_01NODE host1: 2. tty0 Device Pathsmitty hacmp Extended Configuration Extended Topology ConfigurationConfigure HACMP NetworksAdd a Network to the HACMP Clusterrs232添加心跳設(shè)備接口：smitty hacmp Extended Configuration Extended Topology ConfigurationConfigure HACMP Communication Interfaces/DevicesAdd Communication Interfaces/DevicesAdd Predefined Communication Interfaces and Devices Communication Devices 選擇之前建立的net_rs232_01同樣完成net_ether_02網(wǎng)絡(luò)的創(chuàng)建。 此選項(xiàng)決定了HACMP的IP切換方式，但值得一提的是只有“boot1/boot”、“boot2/standby”、“svc/service”三個(gè)IP分別為三個(gè)不同網(wǎng)段時(shí)必須選用IP Aliases方式。smitty hacmp Extended Configuration Extended Topology ConfigurationConfigure HACMP NetworksAdd a Network to the HACMP Clusterether其中Enable IP Address Takeover via IP Aliases 注：此處的Node Name需要手動(dòng)輸入，為機(jī)器主機(jī)名。 而且完全用標(biāo)準(zhǔn)配置則復(fù)雜的做不了，簡(jiǎn)單的卻可能做錯(cuò)，不做推薦。 None hdisk5[host2][root][/]lspv 磁盤心跳1.同樣反向檢測(cè)一下。編寫完成后記得拷貝到另一節(jié)點(diǎn)： stop_host2 mkdir p /usr/sbin/cluster/app/log[host1][root][/usr/sbin/cluster/app]lsstart_host1 0 515199 0 2 0en3 0 6 host2_l1_boot2 link3 0 0 12 1500 0 141876 0 1013619 1500 0[host2][root][/]netstat iName 0 515177 16896 127 0 515177 16896 link1 1500 link5 1500 0 5 1500 0 141859 1500 0 141859 0 1013585 0en0 0 63684 link2 Opkts Oerrs Mtu 0 0 0lo0 0 1335969 0 1335968 1500 1500 0 22 host1_l1_boot2 0en1 0 10 1500 0 2481124 1500 0 2481124 Opkts Oerrs Mtu 0 1335968 16896 ::1%1 0 0 0en3 0 4 0en3 0 20 0en1 0 10 0en1 0 164719 0en0 0 164719 1500 按照規(guī)劃，2臺(tái)機(jī)器修改IP地址 ，smitty tcpip,最終為[host1][root][/]netstat inName no /ora10run no /dev/ora11runlv [host2][root][/]lspv。 00f6f1569990a12c host1vgvaryoffvg host2vg host1vg host1vgvaryoffvg N/A[host1][root][/] vi /etc/filesystems將log = /dev/loglv02的改為log =/dev/host1_loglv確認(rèn):[host1][root][/]mount /ora11runc 1 1 1 /ora11runchost1_loglv 2 40 N/A umount vg上所有fs如 umount /ora11runc2） 修改loglv名稱[host1][root][/] chlv n host1_loglv loglv02 0516712 chlv: The chlv succeeded, however chfs must now be 1 1 closed/syncd 40 LV STATEhost1vg （host2vg也要修改）1） 察看[host1][root][/]varyonvg host1vg[host1][root][/]lsvg l host1vg jfs2 /ora11runc no /dev/ora11runclv /ora11run Auto Accounting.../dev/ora11runlv VFS Nodename 由于后面需要修改loglv，必須建立文件系統(tǒng)才會(huì)有l(wèi)oglv，所以需要先建立在host1vg 上的 /ora11runc和host2vg上的/ora11run的JFS2文件系統(tǒng)，其他文件系統(tǒng)可在實(shí)施中的配置中2邊同時(shí)添加。 active 00f6f1569990a12c nonehdisk3 rootvg activeh