流量统计:
Rixu Blog (日需博客)
日需博客,每日必需来踩踩哦..
posts - 108,comments - 54,trackbacks - 0
Fedora 15上做双主模型的集群
拓扑如下所示:
 
-----------     ----------
|   HA1    |____|  HA2   |
|__________|    |________|
HA1:
IP:192.168.1.78/24
HA2:
IP:192.168.1.151/24
VIP:192.168.1.110

一、配置网络属性
HA1:
#ifconfit eth0 192.168.1.78/24
#route add default gw 192.168.1.1
#hostname node1.luowei.com
HA2:
#ifconfig eth0 192.168.1.151/24
#route add default gw 192.168.1.1
#hostname node2.luowei.com
 
二、配置主机名及两个之间不实用密码能相互通信
#vim /etc/hosts 添加如下内容
192.168.1.78 node1.luowei.com node1
192.168.1.151 node2.luowei.com node2
同样在HA2上也添加这些内容
#ping node2|node1能解析出来就OK啦
分别在两个HA上生成一对密钥,如下所示
[root@node1 ~]# ssh-keygen -t rsa  //生成公钥和密钥
Generating public/private rsa key pair.
Enter file in which to save the key (/root/.ssh/id_rsa): 
Created directory '/root/.ssh'.
Enter passphrase (empty for no passphrase): 
Enter same passphrase again: 
Your identification has been saved in /root/.ssh/id_rsa.
Your public key has been saved in /root/.ssh/id_rsa.pub.
The key fingerprint is:
59:71:5d:4d:4c:6d:71:b1:ec:04:17:26:49:cb:27:a1 root@node1.luowei.com
The key's randomart image is:
+--[ RSA 2048]----+
|          . o*.@@|
|           oo.X B|
|          .E + * |
|         o    =  |
|        S      . |
|                 |
|                 |
|                 |
|                 |  //这个图案就是所谓的指纹信息吧,呵呵,redhat上没有
+-----------------+
[root@node1 ~]# ssh-copy-id -i .ssh/id_rsa.pub root@node2    //拷贝公钥到对方机
器上
The authenticity of host 'node2 (192.168.1.151)' can't be established.
RSA key fingerprint is 77:b6:c6:09:51:f9:f4:70:c1:35:81:47:a5:19:f4:d2.
Are you sure you want to continue connecting (yes/no)? yes
Warning: Permanently added 'node2,192.168.1.151' (RSA) to the list of known
hosts.
root@node2's password:     //输入对方机器的密码
Now try logging into the machine, with "ssh 'root@node2'", and check in:
  ~/.ssh/authorized_keys
to make sure we haven't added extra keys that you weren't expecting.
在HA2上做同上的操作,我就不具体演示了!
 
三、配置yum源,我使用的是163做的镜像源
http://mirrors.163.com/
这上面有对应fedora的yum源配置使用的说明,我就不做详细阐述了
如果你没有DNS解析域名,还要在/etc/hosts文件中手动添加解析奥,我的如下:
66.35.62.166   mirrors.fedoraproject.org
213.129.242.84   mirrors.rpmfusion.org
123.58.173.106  mirrors.163.com
这些对应的域名和IP关系大家都会,就是使用ping可以解析出,不解释!
 
四、安装集群软件
两个节点上都要做的
#yum install corosync pacemaker -y //由于是网络镜像,会比较慢,耐心等会吧!
安装完成之后就是配置了,注意配置的时候选择的端口和地址不能跟已存在的集群冲突,所
以我就做了一下简单的设置
#export ais_port=4000
#export ais_mcast=226.94.1.1
接下来就是配置corosync了:
#cd /etc/corosync/
#cp corosync.conf.example corosync.conf
#vim !$ 把配置改成如下
# Please read the corosync.conf.5 manual page
compatibility: whitetank
totem {
 version: 2
 secauth: on
 threads: 0
 interface {
  ringnumber: 0
  bindnetaddr: 192.168.1.0   //指定集群所在的网段的网络号
  mcastaddr: 226.94.1.1  //组播地址
  mcastport: 4000  //端口号
  ttl: 1
 }
}
logging {
 fileline: off
 to_stderr: no
 to_logfile: yes
 to_syslog: no
 logfile: /var/log/cluster/corosync.log
 debug: off
 timestamp: on
 logger_subsys {
  subsys: AMF
  debug: off
 }
}
amf {
 mode: disabled
}
####以下是添加的内容
service {
 ver: 1    //定义pacemaker的版本,fedora上使用版本1,而在redhat上可以
使用0
 name: pacemaker 
}
aisexec {
        user:   root
        group:  root
}
其中注释的内容为所修改的内容
配置完成之后,拷贝一个到另一个节点上
#scp -p  /etc/corosync/corosync.conf node2:/etc/corosync/
确保没有错误的情况下,可以在HA1上启动了,启动之后还要进行一些列的检测
#/etc/init.d/corosync start
添加认证密钥
#corosync-keygen  //这个要是新机器的话,时间会长一点,要有点耐性等待!
#scp -p authkeys corosync.conf node2:/etc/corosync/
配置完成之后,现在HA1上启动corosync:
#server corosync start
Starting corosync (via systemctl):                         [  OK  ] oK,corosync
服务启动成功!
接下来就是检测集群是否正确启动并且已经可以和其他节点建立集群关系了:
查看corosync引擎是否正常启动:
[root@node1 ~]# grep -e "Corosync Cluster Engine" -e "configuration file"
/var/log/messages
Sep 18 23:09:44 node1 smartd[786]: Opened configuration file /etc/smartd.conf
Sep 19 13:41:03 node1 smartd[801]: Opened configuration file /etc/smartd.conf
Sep 19 20:44:55 node1 smartd[680]: Opened configuration file /etc/smartd.conf
[root@node1 ~]# grep -e "Corosync Cluster Engine" -e "configuration file"
/var/log/cluster/corosync.log 
Sep 18 17:12:06 corosync [MAIN  ] Corosync Cluster Engine ('1.4.1'): started and
ready to provide service.
Sep 18 17:12:06 corosync [MAIN  ] Successfully read main configuration file
'/etc/corosync/corosync.conf'.
Sep 18 17:12:06 corosync [MAIN  ] Corosync Cluster Engine exiting with status 8
at main.c:1702.
Sep 18 17:16:11 corosync [MAIN  ] Corosync Cluster Engine ('1.4.1'): started and
ready to provide service.
查看初始化成员节点通知是否正常发出:
[root@node1 ~]#  grep TOTEM /var/log/cluster/corosync.log
检查启动过程中是否有错误产生:
[root@node2 ~]# grep ERROR: /var/log/cluster/corosync.log | grep -v
unpack_resources
查看pacemaker是否正常启动:
[root@node1 ~]# grep pcmk_startup /var/log/cluster/corosync.log 
Sep 19 13:48:48 corosync [pcmk  ] info: pcmk_startup: CRM: Initialized
Sep 19 13:48:48 corosync [pcmk  ] Logging: Initialized pcmk_startup
Sep 19 13:48:48 corosync [pcmk  ] info: pcmk_startup: Maximum core file size is:
4294967295
Sep 19 13:48:48 corosync [pcmk  ] info: pcmk_startup: Service: 9
Sep 19 13:48:48 corosync [pcmk  ] info: pcmk_startup: Local hostname: node1.luo
检查完毕,接下来就可以启动另一个节点了,最好在同一个节点上启动所有的其他的集群节
点:
[root@node1 ~]# ssh node2 -- '/etc/init.d/corosync start'
Starting corosync (via systemctl):  [  OK  ]
启动成功了!
接下来就是启动pacemaker了!
[root@node1 corosync]# /etc/init.d/pacemaker  start
Starting pacemaker (via systemctl):                        [  OK  ]
ok,同样启动成功
# ps axf //查看进程
 1724 ?        R      5:59 /usr/lib/heartbeat/stonithd
 1725 ?        R      5:59 /usr/lib/heartbeat/cib
 1726 ?        S      0:00 /usr/lib/heartbeat/lrmd
 1727 ?        R      5:59 /usr/lib/heartbeat/attrd
 1728 ?        S      0:00 /usr/lib/heartbeat/pengine
 1729 ?        R      5:59 /usr/lib/heartbeat/crmd
可以看出已经有进程了
当然这个时候有个关键性的设置,就是关闭防火墙,如果你没有关闭防火墙功能,下面将会
给你带来很大的麻烦,我开始就是没有关闭防火墙,后来看日志才知道,所以你做的时候可
以把防火墙先关闭了,但是在真正应用之中,还是要开启防火墙功能
#setup 然后在里面选择Firewall configure 然后disabled就行了
接下来使用crm的内部命令进行查看
#crm_mon 或crm status
Online: [ node2.luowei.com node1.luowei.com ]可以看出,集群的节点都启动了

一切准备停当,接下来就是双主集群的配置了!
五、安装apache服务和集群文件系统-GFS2
为了方便验证,我就安装一个apache服务用于测试:
#yum install httpd -y
在HA1上的添加测试页面:
#echo "<h1>node1.luowei.com<h1>" >/var/www/html/index.html
在HA2上的添加测试页面:
#echo "<h1>node2.luowei.com<h1>" >/var/www/html/index.html
然后把两个节点上的/etc/httpd/conf/httpd.conf的配置文件,保持一下的内容是开启的,
如果有注释的,请去掉注释
<Location /server-status>
    SetHandler server-status
    Order deny,allow
    Deny from all
    Allow from 127.0.0.1
</Location>
保证httpd服务不会随着开机自动启动
#chkconfig httpd off
#crm configure property stonith-enabled=false  //关闭stonith设备
#crm configure property no-quorum-policy=ignore //关闭两节点之间的选举
#crm configure
为httpd添加资源
# crm configure primitive WebSite ocf:heartbeat:apache params
configfile=/etc/httpd/conf/httpd.conf op monitor interval=1min
# crm configure primitive ClusterIP ocf:heartbeat:IPaddr2 params
ip=192.168.1.110 cidr_netmask=32 op monitor interval=30s  //添加一个虚拟IP
[root@node1 ~]# crm status
============
Last updated: Mon Sep 19 23:44:05 2011
Stack: openais
Current DC: node2.luowei.com - partition with quorum
Version: 1.1.5-1.fc15-01e86afaaa6d4a8c4836f68df80ababd6ca3902f
2 Nodes configured, 2 expected votes
2 Resources configured.
============
Online: [ node2.luowei.com node1.luowei.com ]
 ClusterIP (ocf::heartbeat:IPaddr2): Started node1.luowei.com
 WebSite (ocf::heartbeat:apache): Started node2.luowei.com
可以看到两个资源不在同一个节点上,所以需要做一下的设置:
#crm configure colocation website-with-ip INFINITY: WebSite ClusterIP  //做一个
位置约束
然后再使用crm status 查看资源已经都流转到同一个节点上了,如下所示
Online: [ node2.luowei.com node1.luowei.com ]
 ClusterIP (ocf::heartbeat:IPaddr2): Started node1.luowei.com
 WebSite (ocf::heartbeat:apache): Started node1.luowei.com
还要控制资源的启动停止顺序
#crm configure order apache-after-ip mandatory: ClusterIP WebSite  //定义ip的资
源要在apache的服务启动之前启动
指定优先的Location
#crm configure location prefer-pcmk-l WebSite 50: node1.luowei.com
#crm configure show  //查看一下自己的配置如下
[root@node1 ~]# crm configure show
node node1.luowei.com
node node2.luowei.com
primitive ClusterIP ocf:heartbeat:IPaddr2 \
 params ip="192.168.1.110" cidr_netmask="32" \
 op monitor interval="30s"
primitive WebSite ocf:heartbeat:apache \
 params configfile="/etc/httpd/conf/httpd.conf" \
 op monitor interval="1min"
location prefer-pcmk-l WebSite 50: node1.luowei.com
colocation website-with-ip inf: WebSite ClusterIP
order apache-after-ip inf: ClusterIP WebSite
property $id="cib-bootstrap-options" \
 dc-version="1.1.5-1.fc15-01e86afaaa6d4a8c4836f68df80ababd6ca3902f" \
 cluster-infrastructure="openais" \
 expected-quorum-votes="2" \
 stonith-enabled="false" \
 no-quorum-policy="ignore"
rsc_defaults $id="rsc-options" \
 resource-stickiness="100"
如上图所示,资源已经启动了,所以接下来就可以往下做了!
可以在浏览其中输入http://192.168.1.110可以访问web服务了!
 
六、安装DRBD软件包
DRBD实现节点之间的数据同步的,实现备份功能。
1.# yum install drbd-pacemaker drbd-udev -y
2.安装完drbd之后,首先要在两个节点上做一个单独的磁盘分区来存放数据
这里我用一块新的磁盘(/dev/sdb)进行实验,划分磁盘分区如下所示:
#fdisk /dev/sdb
[root@node1 ~]# fdisk /dev/sda1
Device contains neither a valid DOS partition table, nor Sun, SGI or OSF
disklabel
Building a new DOS disklabel with disk identifier 0xcaf34d49.
Changes will remain in memory only, until you decide to write them.
After that, of course, the previous content won't be recoverable.
Warning: invalid flag 0x0000 of partition table 4 will be corrected by w(rite)
Command (m for help): p
Disk /dev/sda1: 524 MB, 524288000 bytes
255 heads, 63 sectors/track, 63 cylinders, total 1024000 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0xcaf34d49
     Device Boot      Start         End      Blocks   Id  System
Command (m for help): p
Disk /dev/sda1: 524 MB, 524288000 bytes
255 heads, 63 sectors/track, 63 cylinders, total 1024000 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0xcaf34d49
     Device Boot      Start         End      Blocks   Id  System
Command (m for help): q
# partprobe /dev/sdb
# pvcreate /dev/sdb1
# vgcreate VolGroupb /dev/sdb1
# lvcreate -n drbd-demo -L 1G VolGroupb
[root@node1 ~]# lvs
  LV        VG        Attr   LSize  Origin Snap%  Move Log Copy%  Convert
  lv_root   VolGroup  -wi-ao 17.56g                                      
  lv_swap   VolGroup  -wi-ao  1.94g                                      
  drbd-demo VolGroupb -wi-a-  1.00g  
ok!HA1上的逻辑卷做好了,这个过程也要在HA2上来一遍,我就不多展示了!
3.准备完成,接下来就是配置DRBD了!
#vim /etc/drbd.conf
include "drbd.d/global_common.conf";
include "drbd.d/*.res";
global {
        usage-count yes;
}
common {
        protocol C;
}
resource wwwdata {
        meta-disk internal;
        device  /dev/drbd1;
        syncer {
                verify-alg sha1;
        }
        net {
        allow-two-primaries;
        }
        on node1.luowei.com {
                disk    /dev/mapper/VolGroupb-drbd--demo;
                address 192.168.1.78:7789;    //定义HA1节点的
        }
        on node2.luowei.com {
                disk    /dev/mapper/VolGroupb-drbd--demo;
                address 192.168.1.151:7789;   //定义HA2节点的
        }
}
4.接下来就是初始化并加载DRBD了
# drbdadm create-md wwwdata
New drbd meta data block successfully created.
初始化成功!
5.接下来查看DRBD的模块载入内核并检测是不是都正常
[root@node1 ~]# modprobe drbd
[root@node1 ~]# drbdadm up wwwdata
[root@node1 ~]# cat /proc/drbd 
version: 8.3.9 (api:88/proto:86-95)
srcversion: CF228D42875CF3A43F2945A 
 1: cs:WFConnection ro:Secondary/Unknown ds:Inconsistent/DUnknown C r----s
    ns:0 nr:0 dw:0 dr:0 al:0 bm:0 lo:0 pe:0 ua:0 ap:0 ep:1 wo:f oos:1048508
可以看出已经出现了Secondary了,下面在第二个节点上使用上面同样的方法进行模块载入
并检测,此处省略.....
6.然后在任意一个节点上查看,现在两个都已经是Secondary了,所以一切正常
[root@node1 ~]# drbd-overview 
  1:wwwdata  Connected Secondary/Secondary Inconsistent/Inconsistent C r-----
7.现在我们把HA1设置为主节点
[root@node1 ~]# drbdadm -- --overwrite-data-of-peer primary wwwdata
然后使用如下命令可以实时监视这整个数据从主节点想备用节点上拷贝数据的过程:
[root@node1 ~]# watch -n 1 'drbd-overview' 
  1:wwwdata  SyncSource Primary/Secondary UpToDate/Inconsistent C r----- 
 [==>.................] sync'ed:  0.8% (1042492/1048508)K
  1:wwwdata  Connected Primary/Secondary UpToDate/UpToDate C r-----完成了数据的
同步,现在HA1处于Primary状态,它允许写入了,可以在上面创建文件系统并把一些数据放
进去了。
8.向DRBD中添加数据:
[root@node1 ~]# mkfs.ext4 /dev/drbd1  //格式化分区
[root@node1 ~]# mount /dev/drbd1 /mnt/   //挂载分区
[root@node1 ~]# echo "<h2>drbd test page</h2>" >/mnt/index.html
[root@node1 ~]# umount /mnt/   //卸载分区
9.在集群中配置DRBD
[root@node1 ~]# crm 
crm(live)# cib new drbd
crm(drbd)# configure
crm(drbd)configure# primitive WebData ocf:linbit:drbd  params
drbd_resource=wwwdata op monitor interval=60s
crm(drbd)configure# ms WebDataClone WebData meta master-max=1 master-node-max=1
clone-max=2 clone-node-max=1 notify=true 
crm(drbd)configure#commit
[root@node1 ~]# crm status
============
Last updated: Tue Sep 20 22:08:10 2011
Stack: openais
Current DC: node1.luowei.com - partition with quorum
Version: 1.1.5-1.fc15-01e86afaaa6d4a8c4836f68df80ababd6ca3902f
2 Nodes configured, 2 expected votes
3 Resources configured.
============
Online: [ node2.luowei.com node1.luowei.com ]
 ClusterIP (ocf::heartbeat:IPaddr2): Started node1.luowei.com
 WebSite (ocf::heartbeat:apache): Started node1.luowei.com
 Master/Slave Set: WebDataClone [WebData]
     Masters: [ node2.luowei.com ]
     Slaves: [ node1.luowei.com ]
有上面的输出信息可以看出,资源启动正常,但是我们注意到drbd的主节点在HA2上,为了
统一到同一个节点上,还需要进一步约束资源
[root@node1 ~]# crm 
crm(live)# configure 
crm(live)configure# primitive WebFS ocf:heartbeat:Filesystem params
device="/dev/drbd/by-res/wwwdata" directory="/var/www/html" fstype="ext4"
crm(live)configure# colocation fs_ondrbd inf: WebFS WebDataClone:Master
crm(live)configure# order WebFS-after-WebData inf: WebDataClone:promote
WebFS:start
crm(live)configure# colocation WebSite-with-WebFS inf: WebSite WebFS 
crm(live)configure# order WebSite-after-WebFS inf: WebFS WebSite 
crm(live)configure# commit
再次查看,如下内容:
[root@node1 ~]# crm status
============
Last updated: Tue Sep 20 22:38:16 2011
Stack: openais
Current DC: node1.luowei.com - partition with quorum
Version: 1.1.5-1.fc15-01e86afaaa6d4a8c4836f68df80ababd6ca3902f
2 Nodes configured, 2 expected votes
4 Resources configured.
============
Online: [ node2.luowei.com node1.luowei.com ]
 ClusterIP (ocf::heartbeat:IPaddr2): Started node2.luowei.com
 Master/Slave Set: WebDataClone [WebData]
     Masters: [ node2.luowei.com ]
     Slaves: [ node1.luowei.com ]
 WebFS (ocf::heartbeat:Filesystem): Started node2.luowei.com
我们可以看出,资源都在同一个节点上
 
七、接下来就是在上面的基础之上做双主模式的集群了:
1、安装集群文件系统
#yum install gfs2-utils gfs2-cluster gfs-pcmk  //两个节点上都要进行安装的
2、添加DLM服务
[root@node1 ~]# crm 
crm(live)# configure 
crm(live)configure# primitive dlm ocf:pacemaker:controld op monitor
interval=120s
crm(live)configure# clone dlm-clone dlm meta interleave=true
crm(live)configure# commit
3、创建gfs-control这个集群资源:
[root@node1 ~]# clear
[root@node1 ~]# crm 
crm(live)# configure 
crm(live)configure# primitive gfs-control ocf:pacemaker:controld params
daemon=gfs_controld.pcmk args="-g 0" op monitor interval=120s
crm(live)configure# clone gfs-clone gfs-control meta interleave=true
crm(live)configure# colocation gfs-with-dlm INFINITY: gfs-clone dlm-clone 
crm(live)configure# order start-gfs-after-dlm mandatory: dlm-clone gfs-clone 
crm(live)configure# commit
然后查看一下我们的配置如下所示:
#crm configure show
node node1.luowei.com
node node2.luowei.com
primitive ClusterIP ocf:heartbeat:IPaddr2 \
 params ip="192.168.1.110" cidr_netmask="32" \
 op monitor interval="30s"
primitive WebData ocf:linbit:drbd \
 params drbd_resource="wwwdata" \
 op monitor interval="60s"
primitive WebFS ocf:heartbeat:Filesystem \
 params device="/dev/drbd/by-res/wwwdata" directory="/var/www/html"
fstype="ext4"
primitive WebSite ocf:heartbeat:apache \
 params configfile="/etc/httpd/conf/httpd.conf" \
 op monitor interval="1min"
primitive dlm ocf:pacemaker:controld \
 op monitor interval="120s"
primitive gfs-control ocf:pacemaker:controld \
 params daemon="gfs_controld.pcmk" args="-g 0" \
 op monitor interval="120s"
ms WebDataClone WebData \
 meta master-max="1" master-node-max="1" clone-max="2" clone-node-max="1"
notify="true"
clone dlm-clone dlm \
 meta interleave="true"
clone gfs-clone gfs-control \
 meta interleave="true"
location prefer-pcmk-l WebSite 50: node1.luowei.com
colocation WebSite-with-WebFS inf: WebSite WebFS
colocation fs_ondrbd inf: WebFS WebDataClone:Master
colocation gfs-with-dlm inf: gfs-clone dlm-clone
colocation website-with-ip inf: WebSite ClusterIP
order WebFS-after-WebData inf: WebDataClone:promote WebFS:start
order WebSite-after-WebFS inf: WebFS WebSite
order apache-after-ip inf: ClusterIP WebSite
order start-gfs-after-dlm inf: dlm-clone gfs-clone
property $id="cib-bootstrap-options" \
 dc-version="1.1.5-1.fc15-01e86afaaa6d4a8c4836f68df80ababd6ca3902f" \
 cluster-infrastructure="openais" \
 expected-quorum-votes="2" \
 stonith-enabled="false" \
 no-quorum-policy="ignore"
rsc_defaults $id="rsc-options" \
 resource-stickiness="100"
查看集群输出的信息:
[root@node1 ~]# crm_mon
============
Last updated: Tue Sep 20 23:18:22 2011
Stack: openais
Current DC: node1.luowei.com - partition with quorum
Version: 1.1.5-1.fc15-01e86afaaa6d4a8c4836f68df80ababd6ca3902f
2 Nodes configured, 2 expected votes
6 Resources configured.
============
Online: [ node2.luowei.com node1.luowei.com ]
ClusterIP (ocf::heartbeat:IPaddr2): Started node1.luowei.com
 Master/Slave Set: WebDataClone [WebData]
     Masters: [ node2.luowei.com ]
     Slaves: [ node1.luowei.com ]
WebSite  (ocf::heartbeat:apache): Started node1.luowei.com
Clone Set: dlm-clone
 Started: [node2.luowei.com node1.luowei.com]
Clone Set: gfs-clone
 Startde: [node2.luowei.com node1.luowei.com]
WebFS   (ocf::heartbeat:Filesystem):    Started node1.luowei.com
4、创建GFS2文件系统
[root@node1 ~]# crm_resource --resource WebFS --set-parameter target-role --meta
--parameter-value Stopped 
这个时候使用crm status可以看到apache 和WebFS两个资源都已经停止。
5、创建并迁移数据到GFS2分区
在两个节点上都执行以下命令:
[root@node2 ~]# mkfs.gfs2 -p lock_dlm -j 2 -t pcmk:web /dev/drbd1
This will destroy any data on /dev/drbd1.
It appears to contain: Linux rev 1.0 ext4 filesystem data, UUID=19976683-c802-
479c-854d-e786617be523 (extents) (large files) (huge files)
Are you sure you want to proceed? [y/n] y
6、然后迁移数据到这个新的文件系统并且为集群重新配置GFS2
[root@node1 ~]# crm 
crm(live)# configure 
crm(live)configure# primitive WebFS ocf:heartbeat:Filesystem params
device="/dev/drbd/by-res/wwwdata" directory="/var/www/html" fstype="gfs2"
crm(live)configure# colocation WebSite-with-WebFS inf: WebSite WebFS
crm(live)configure# colocation fs_on_debd inf: WebFS WebDataClone:Master
crm(live)configure# order WebFS-after-WebData inf: WebDataClone:promote
WebFS:start
crm(live)configure# order WebSite-after-WebFS inf: WebFS WebSite
crm(live)configure# colocation WebFS-with-gfs-control INFINITY: WebFS gfs-clone
crm(live)configure# order start-WebFS-after-gfs-control mandatory: gfs-clone
WebFS
crm(live)configure# commit
7、重新配置pacemaker为Active/Active
[root@node1 ~]# crm 
crm(live)# configure clone WebIP ClusterIP meta globally-unique="true" clone-
max="2" clone-node-max="2"
crm(live)# configure primitive ClusterIP ocf:heartbeat:IPaddr2 params
ip="192.168.1.110" cidr_netmask="32" clusterip_hash="sourceip" op monitor
interval="30s" //设置ClusterIP的参数
crm(live)# configure clone WebFSClone WebFS
crm(live)# configure clone WebSiteClone WebSite
同时把CIB文件中的master-max改为2
资源配置完成,
主从到主主集群架构就这样配置完成!

Logo
作者:Gezidan
出处:http://www.rixu.net    
本文版权归作者和博客园共有,欢迎转载,但未经作者同意必须保留此段声明,且在文章页面明显位置给出原文连接,否则保留追究法律责任的权利。
本文转载自 
http://roqi410.blog.51cto.com/2186161/669877
posted on 2011-09-23 09:59 日需博客 阅读(2308) 评论(1)  编辑 收藏 引用 所属分类: 技术文章转载

FeedBack:
# re: Fedora 15上做主从、双主模型的集群
2011-09-23 19:30 | cheap lace front wigs
这个很有用,收藏一下,做大型服务不错  回复  更多评论
  

只有注册用户登录后才能发表评论。
网站导航: 博客园   IT新闻   BlogJava   知识库   博问   管理