场景一：定义web资源实现web集群的高可用！
本次环境：
Centos6系统 2台，各安装了http服务。
Pacemaker corosync服务已经安装
1、 查看当前环境
crm(live)configure# show node erick.com attributesstandby=off node erick1.com attributesstandby=off property cib-bootstrap-options: dc-version=1.1.14-8.el6-70404b0 cluster-infrastructure="classicopenais (with plugin)" expected-quorum-votes=2 stonith-enabled=false no-quorum-policy=stop last-lrm-refresh=1467844686
1、 定义web的vip资源
crm(live)configure# primitive webipocf:heartbeat:IPaddr params ip=192.168.223.110 crm(live)configure# verify crm(live)configure# commit crm(live)configure# show node erick.com attributes standby=off node erick1.com attributesstandby=off primitive webip IPaddr paramsip=192.168.223.110 property cib-bootstrap-options: dc-version=1.1.14-8.el6-70404b0 cluster-infrastructure="classicopenais (with plugin)" expected-quorum-votes=2 stonith-enabled=false no-quorum-policy=stop last-lrm-refresh=1467844686
在节点上查看是否vip已经生成
[[email protected] ~]# ip a 1: lo: <LOOPBACK,UP,LOWER_UP>mtu 16436 qdisc noqueue state UNKNOWN link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo inet6 ::1/128 scope host valid_lft forever preferred_lft forever 2: eth0:<BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen1000 link/ether 00:0c:29:8c:43:f0 brd ff:ff:ff:ff:ff:ff inet 192.168.223.131/24 brd 192.168.223.255 scope global eth0 inet 192.168.223.110/24brd 192.168.223.255 scope global secondary eth0 inet6 fe80::20c:29ff:fe8c:43f0/64 scope link valid_lft forever preferred_lft forever [[email protected]~]#
定义web资源
crm(live)configure# primitivewebserver lsb:httpd crm(live)configure# verify crm(live)configure# comm crm(live)configure# show node erick.com attributesstandby=off node erick1.com attributesstandby=off primitive webip IPaddr paramsip=192.168.223.110 primitive webserver lsb:httpd property cib-bootstrap-options: dc-version=1.1.14-8.el6-70404b0 cluster-infrastructure="classicopenais (with plugin)" expected-quorum-votes=2 stonith-enabled=false no-quorum-policy=stop last-lrm-refresh=1467844686 crm(live)configure#
现在来利用pacemaker启动服务
crm(live)# resource start webip crm(live)# resource start webserver crm(live)# crm(live)# crm(live)# crm(live)# status Last updated: Thu Jul 7 21:16:03 2016 Last change: Thu Jul 7 21:16:00 2016 by root via cibadmin onerick.com Stack: classic openais (with plugin) Current DC: erick.com (version1.1.14-8.el6-70404b0) - partition with quorum 2 nodes and 2 resources configured,2 expected votes Online: [ erick.com erick1.com ] Full list of resources: webip (ocf::heartbeat:IPaddr): Started erick.com webserver (lsb:httpd): Started erick1.com crm(live)#
现在可以看到资源非常均衡的分别在两个节点上起动了。这样有个问题，会导致web无法访问。现在我我们要实现把两个资源定义在一起，使其能够在一个节点启动
crm(live)configure# colocationwebserver_before_webip inf: webserver webip crm(live)configure# verify crm(live)configure# commit crm(live)configure# show node erick.com attributesstandby=off node erick1.com attributesstandby=off primitive webip IPaddr paramsip=192.168.223.110 metatarget-role=Started primitive webserver lsb:httpd metatarget-role=Started colocation webserver_before_webipinf: webserver webip property cib-bootstrap-options: dc-version=1.1.14-8.el6-70404b0 cluster-infrastructure="classicopenais (with plugin)" expected-quorum-votes=2 stonith-enabled=false no-quorum-policy=stop last-lrm-refresh=1467844686 crm(live)configure# up crm(live)# status Last updated: Thu Jul 7 21:20:50 2016 Last change: Thu Jul 7 21:20:37 2016 by root via cibadmin onerick.com Stack: classic openais (with plugin) Current DC: erick.com (version1.1.14-8.el6-70404b0) - partition with quorum 2 nodes and 2 resources configured,2 expected votes Online: [ erick.com erick1.com ] Full list of resources: webip (ocf::heartbeat:IPaddr): Started erick.com webserver (lsb:httpd): Started erick.com
可以看到两个资源都在erick节点了，那么我门使用浏览器测试下：
Ok，访问成功！
由于我们是2台机器，现在我们若一台服务器宕机后，默认是的仲裁是服务关闭，所以我们要修改下策略，改成ignore
crm(live)configure# propertyno-quorum-policy=ignore crm(live)configure# verify crm(live)configure# commit crm(live)configure# crm(live)configure# show node erick.com attributesstandby=off node erick1.com attributesstandby=off primitive webip IPaddr paramsip=192.168.223.110 metatarget-role=Started primitive webserver lsb:httpd metatarget-role=Started colocation webserver_before_webipinf: webserver webip property cib-bootstrap-options: dc-version=1.1.14-8.el6-70404b0 cluster-infrastructure="classicopenais (with plugin)" expected-quorum-votes=2 stonith-enabled=false no-quorum-policy=ignore last-lrm-refresh=1467844686
现在，我们关闭一台虚拟机的服务，（模仿其宕机）查看资源转移情况
[[email protected] ~]# vim/etc/httpd/conf/httpd.conf [[email protected] ~]# /etc/init.d/corosyncstop Signaling Corosync Cluster Engine(corosync) to terminate: [ OK ] Waiting for corosync services tounload:. [ OK ] [[email protected]~]#
我们到erick1节点查看情况
crm(live)# status Last updated: Thu Jul 7 07:06:59 2016 Last change: Thu Jul 7 21:24:24 2016 by root via cibadmin onerick.com Stack: classic openais (with plugin) Current DC: erick1.com (version1.1.14-8.el6-70404b0) - partition WITHOUT quorum 2 nodes and 2 resources configured,2 expected votes Online: [ erick1.com ] OFFLINE: [ erick.com ] Full list of resources: webip (ocf::heartbeat:IPaddr): Started erick1.com webserver (lsb:httpd): Startederick1.com crm(live)#
通过web访问下
可以看到资源已经转移到erick1上来了。下面我们把erick节点的服务启动起来。查看资源会不会转移
[[email protected] ~]# /etc/init.d/corosync start Starting Corosync Cluster Engine (corosync): [ OK ] crm(live)# status Last updated: Thu Jul 7 07:09:03 2016 Last change: Thu Jul 7 21:24:24 2016 by root via cibadmin onerick.com Stack: classic openais (with plugin) Current DC: erick1.com (version1.1.14-8.el6-70404b0) - partition with quorum 2 nodes and 2 resources configured,2 expected votes Online: [ erick.com erick1.com ] Full list of resources: webip (ocf::heartbeat:IPaddr): Started erick1.com webserver (lsb:httpd): Started erick1.com
资源并没有转移，说明不陪权重是不会转移的。
情景二：
我们以上只能做到当机器宕机时资源实现转移，当httpd服务挂掉时，并不会实现转移。下面我们来实现监控资源
crm(live)configure# monitorwebserver 20s:15s crm(live)configure# verify crm(live)configure# commit crm(live)configure# show node erick.com attributesstandby=off node erick1.com attributesstandby=off primitive webip IPaddr paramsip=192.168.223.110 metatarget-role=Started primitive webserver lsb:httpd meta target-role=Started op monitor interval=20stimeout=15s colocation webserver_before_webipinf: webserver webip property cib-bootstrap-options: dc-version=1.1.14-8.el6-70404b0 cluster-infrastructure="classicopenais (with plugin)" expected-quorum-votes=2 stonith-enabled=false no-quorum-policy=ignore last-lrm-refresh=1467844686
若此时我们httpd服务强行杀掉后，还会自动启动，只有当自己无法启动时才会启动。下面我们把erick1节点httpd的端口改为22这样http服务就无法启动了（过程略），手动kill掉httpd进程（过程略），下面查看资源分布情况
Full list of resources: webip (ocf::heartbeat:IPaddr): Started erick.com webserver (lsb:httpd): Startederick.com Failed Actions: * webserver_start_0 on erick1.com‘unknown error‘ (1): call=90, status=complete, exitreason=‘none‘, last-rc-change=‘Thu Jul 707:19:27 2016‘, queued=0ms, exec=87ms crm(live)# resource cleanupwebserver ##清除失败的资源 Cleaning up webserver on erick.com,removing fail-count-webserver Cleaning up webserver on erick1.com,removing fail-count-webserver Waiting for 2 replies from theCRMd.. OK crm(live)# crm(live)# status Last updated: Thu Jul 7 21:40:15 2016 Last change: Thu Jul 7 07:20:45 2016 by hacluster via crmd onerick1.com Stack: classic openais (with plugin) Current DC: erick1.com (version1.1.14-8.el6-70404b0) - partition with quorum 2 nodes and 2 resources configured,2 expected votes Online: [ erick.com erick1.com ] Full list of resources: webip (ocf::heartbeat:IPaddr): Started erick.com webserver (lsb:httpd): Startederick.com
通过web验证下
本次实验完成!！
时间： 2024-10-06 05:27:31