基于GFS的LVS配置实践.docx

基于GFS的LVS配置实践配置GSF的内容如下：1。安装过程中安装 ““Cluster ClusterStorage2。配置 /etc/hosts文件：0 srv2 srv23。配置/etc/cluster/cluster.conf文件如下：?xml version=1.0 ?cluster config_version=3 name=eyou_clusterfence_daemon post_fail_delay=0 post_join_delay=3/clusternodesclusternode name=srv1 nodeid=1 votes=1fence//clusternodeclusternode name=srv2 nodeid=2 votes=1fence//clusternode/clusternodescman expected_votes=1 two_node=1/fencedevices/rmfailoverdomains/resources//rm/cluster注意：这个文件和/etc/hosts文件相关。以上步骤在两台机器上都运行一下。在一台机器上运行第4步：4。配置存储（硬件工程师），需要在fdisk -l里面看到两个存储上的盘（sdb,sdc)pvcreate /dev/sdbpvcreate /dev/sdcvgcreate VolGrp00 /dev/sdb /dev/sdclvcreate -L 100G LV0 /dev/VolGrp00mkfs.gfs2 -p lock_dlm -t eyou_cluster:gfs1 -j 2 /dev/VolGrp00/LV0＃此命令和Cluter.conf文件相关。在两台机器上同时运行第5步：5。启动Cluster和GFS。chkconfig cman onchkconfig clvmd onchkconfig gfs2 on/etc/init.d/cman start/etc/init.d/clvmd start/etc/init.d/gfs2 start6。mount /dev/VolGrp00/LV0 /ShareData_17。测试：在两台机器全部运行6，在其中一台机器上建立一个文件，如果在两外一台机器上能马上看到就正常。LVS配置过程如下：以下配置在两台机器上全部运行：1。配置/etc/sysconfig/ha/lvs.cf文件如下：serial_no = 110primary = 2service = lvsbackup_active = 1backup = 0heartbeat = 1heartbeat_port = 539keepalive = 6deadtime = 18network = directdebug_level = NONEmonitor_links = 0virtual httpd {active = 1address = 4 eth0:1vip_nmask = port = 80send = GET / HTTP/1.0\r\n\r\nexpect = HTTPuse_regex = 0load_monitor = nonescheduler = wlcprotocol = tcptimeout = 6reentry = 15quiesce_server = 0server srv1 {address = 0active = 1weight = 1}server srv2 {address = 2active = 1weight = 1}}virtual smtp {active = 1address = 4 eth0:1vip_nmask = port = 25use_regex = 0load_monitor = nonescheduler = wlcprotocol = tcptimeout = 6reentry = 15quiesce_server = 0server srv1 {address = 0active = 1weight = 1}server srv2 {address = 2active = 1weight = 1}}virtual pop3 {active = 1address = 4 eth0:1vip_nmask = port = 110use_regex = 0load_monitor = nonescheduler = wlcprotocol = tcptimeout = 6reentry = 15quiesce_server = 0server srv1 {address = 0active = 1weight = 1}server srv2 {address = 2ac